summaryrefslogtreecommitdiffstatshomepage
path: root/src/lib
diff options
context:
space:
mode:
Diffstat (limited to 'src/lib')
-rw-r--r--src/lib/expat/ascii.h85
-rw-r--r--src/lib/expat/asciitab.h36
-rw-r--r--src/lib/expat/expat.h1013
-rw-r--r--src/lib/expat/expat_external.h115
-rw-r--r--src/lib/expat/iasciitab.h37
-rw-r--r--src/lib/expat/intconfig.h19
-rw-r--r--src/lib/expat/internal.h73
-rw-r--r--src/lib/expat/latin1tab.h36
-rw-r--r--src/lib/expat/nametab.h150
-rw-r--r--src/lib/expat/utf8tab.h37
-rw-r--r--src/lib/expat/xmlparse.c6258
-rw-r--r--src/lib/expat/xmlrole.c1320
-rw-r--r--src/lib/expat/xmlrole.h114
-rw-r--r--src/lib/expat/xmltok.c1629
-rw-r--r--src/lib/expat/xmltok.h316
-rw-r--r--src/lib/expat/xmltok_impl.c1779
-rw-r--r--src/lib/expat/xmltok_impl.h46
-rw-r--r--src/lib/expat/xmltok_ns.c106
-rw-r--r--src/lib/lib.mak84
-rw-r--r--src/lib/util/astring.c596
-rw-r--r--src/lib/util/astring.h228
-rw-r--r--src/lib/util/avcomp.c840
-rw-r--r--src/lib/util/avcomp.h60
-rw-r--r--src/lib/util/aviio.c2744
-rw-r--r--src/lib/util/aviio.h123
-rw-r--r--src/lib/util/bitmap.c263
-rw-r--r--src/lib/util/bitmap.h169
-rw-r--r--src/lib/util/cdrom.c617
-rw-r--r--src/lib/util/cdrom.h139
-rw-r--r--src/lib/util/chd.c2860
-rw-r--r--src/lib/util/chd.h325
-rw-r--r--src/lib/util/corefile.c726
-rw-r--r--src/lib/util/corefile.h116
-rw-r--r--src/lib/util/corestr.c120
-rw-r--r--src/lib/util/corestr.h60
-rw-r--r--src/lib/util/coreutil.c102
-rw-r--r--src/lib/util/coreutil.h44
-rw-r--r--src/lib/util/harddisk.c170
-rw-r--r--src/lib/util/harddisk.h51
-rw-r--r--src/lib/util/huffman.c934
-rw-r--r--src/lib/util/huffman.h62
-rw-r--r--src/lib/util/jedparse.c435
-rw-r--r--src/lib/util/jedparse.h89
-rw-r--r--src/lib/util/md5.c239
-rw-r--r--src/lib/util/md5.h42
-rw-r--r--src/lib/util/options.c1133
-rw-r--r--src/lib/util/options.h216
-rw-r--r--src/lib/util/palette.c651
-rw-r--r--src/lib/util/palette.h259
-rw-r--r--src/lib/util/png.c1143
-rw-r--r--src/lib/util/png.h137
-rw-r--r--src/lib/util/pool.c614
-rw-r--r--src/lib/util/pool.h117
-rw-r--r--src/lib/util/sha1.c387
-rw-r--r--src/lib/util/sha1.h61
-rw-r--r--src/lib/util/unicode.c346
-rw-r--r--src/lib/util/unicode.h92
-rw-r--r--src/lib/util/unzip.c616
-rw-r--r--src/lib/util/unzip.h143
-rw-r--r--src/lib/util/xmlfile.c857
-rw-r--r--src/lib/util/xmlfile.h160
-rw-r--r--src/lib/zlib/Makefile154
-rw-r--r--src/lib/zlib/Makefile.in154
-rw-r--r--src/lib/zlib/adler32.c149
-rw-r--r--src/lib/zlib/compress.c79
-rw-r--r--src/lib/zlib/crc32.c423
-rw-r--r--src/lib/zlib/crc32.h441
-rw-r--r--src/lib/zlib/deflate.c1736
-rw-r--r--src/lib/zlib/deflate.h331
-rw-r--r--src/lib/zlib/example.c565
-rw-r--r--src/lib/zlib/gzio.c1026
-rw-r--r--src/lib/zlib/infback.c623
-rw-r--r--src/lib/zlib/inffast.c318
-rw-r--r--src/lib/zlib/inffast.h11
-rw-r--r--src/lib/zlib/inffixed.h94
-rw-r--r--src/lib/zlib/inflate.c1368
-rw-r--r--src/lib/zlib/inflate.h115
-rw-r--r--src/lib/zlib/inftrees.c329
-rw-r--r--src/lib/zlib/inftrees.h55
-rw-r--r--src/lib/zlib/minigzip.c322
-rw-r--r--src/lib/zlib/readme125
-rw-r--r--src/lib/zlib/trees.c1219
-rw-r--r--src/lib/zlib/trees.h128
-rw-r--r--src/lib/zlib/uncompr.c61
-rw-r--r--src/lib/zlib/zconf.h332
-rw-r--r--src/lib/zlib/zconf.in.h332
-rw-r--r--src/lib/zlib/zlib.h1357
-rw-r--r--src/lib/zlib/zutil.c318
-rw-r--r--src/lib/zlib/zutil.h269
89 files changed, 44773 insertions, 0 deletions
diff --git a/src/lib/expat/ascii.h b/src/lib/expat/ascii.h
new file mode 100644
index 00000000000..337e5bb7ea0
--- /dev/null
+++ b/src/lib/expat/ascii.h
@@ -0,0 +1,85 @@
+/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+ See the file COPYING for copying permission.
+*/
+
+#define ASCII_A 0x41
+#define ASCII_B 0x42
+#define ASCII_C 0x43
+#define ASCII_D 0x44
+#define ASCII_E 0x45
+#define ASCII_F 0x46
+#define ASCII_G 0x47
+#define ASCII_H 0x48
+#define ASCII_I 0x49
+#define ASCII_J 0x4A
+#define ASCII_K 0x4B
+#define ASCII_L 0x4C
+#define ASCII_M 0x4D
+#define ASCII_N 0x4E
+#define ASCII_O 0x4F
+#define ASCII_P 0x50
+#define ASCII_Q 0x51
+#define ASCII_R 0x52
+#define ASCII_S 0x53
+#define ASCII_T 0x54
+#define ASCII_U 0x55
+#define ASCII_V 0x56
+#define ASCII_W 0x57
+#define ASCII_X 0x58
+#define ASCII_Y 0x59
+#define ASCII_Z 0x5A
+
+#define ASCII_a 0x61
+#define ASCII_b 0x62
+#define ASCII_c 0x63
+#define ASCII_d 0x64
+#define ASCII_e 0x65
+#define ASCII_f 0x66
+#define ASCII_g 0x67
+#define ASCII_h 0x68
+#define ASCII_i 0x69
+#define ASCII_j 0x6A
+#define ASCII_k 0x6B
+#define ASCII_l 0x6C
+#define ASCII_m 0x6D
+#define ASCII_n 0x6E
+#define ASCII_o 0x6F
+#define ASCII_p 0x70
+#define ASCII_q 0x71
+#define ASCII_r 0x72
+#define ASCII_s 0x73
+#define ASCII_t 0x74
+#define ASCII_u 0x75
+#define ASCII_v 0x76
+#define ASCII_w 0x77
+#define ASCII_x 0x78
+#define ASCII_y 0x79
+#define ASCII_z 0x7A
+
+#define ASCII_0 0x30
+#define ASCII_1 0x31
+#define ASCII_2 0x32
+#define ASCII_3 0x33
+#define ASCII_4 0x34
+#define ASCII_5 0x35
+#define ASCII_6 0x36
+#define ASCII_7 0x37
+#define ASCII_8 0x38
+#define ASCII_9 0x39
+
+#define ASCII_TAB 0x09
+#define ASCII_SPACE 0x20
+#define ASCII_EXCL 0x21
+#define ASCII_QUOT 0x22
+#define ASCII_AMP 0x26
+#define ASCII_APOS 0x27
+#define ASCII_MINUS 0x2D
+#define ASCII_PERIOD 0x2E
+#define ASCII_COLON 0x3A
+#define ASCII_SEMI 0x3B
+#define ASCII_LT 0x3C
+#define ASCII_EQUALS 0x3D
+#define ASCII_GT 0x3E
+#define ASCII_LSQB 0x5B
+#define ASCII_RSQB 0x5D
+#define ASCII_UNDERSCORE 0x5F
diff --git a/src/lib/expat/asciitab.h b/src/lib/expat/asciitab.h
new file mode 100644
index 00000000000..79a15c28ca1
--- /dev/null
+++ b/src/lib/expat/asciitab.h
@@ -0,0 +1,36 @@
+/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+ See the file COPYING for copying permission.
+*/
+
+/* 0x00 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x04 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x08 */ BT_NONXML, BT_S, BT_LF, BT_NONXML,
+/* 0x0C */ BT_NONXML, BT_CR, BT_NONXML, BT_NONXML,
+/* 0x10 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x14 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x18 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x1C */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x20 */ BT_S, BT_EXCL, BT_QUOT, BT_NUM,
+/* 0x24 */ BT_OTHER, BT_PERCNT, BT_AMP, BT_APOS,
+/* 0x28 */ BT_LPAR, BT_RPAR, BT_AST, BT_PLUS,
+/* 0x2C */ BT_COMMA, BT_MINUS, BT_NAME, BT_SOL,
+/* 0x30 */ BT_DIGIT, BT_DIGIT, BT_DIGIT, BT_DIGIT,
+/* 0x34 */ BT_DIGIT, BT_DIGIT, BT_DIGIT, BT_DIGIT,
+/* 0x38 */ BT_DIGIT, BT_DIGIT, BT_COLON, BT_SEMI,
+/* 0x3C */ BT_LT, BT_EQUALS, BT_GT, BT_QUEST,
+/* 0x40 */ BT_OTHER, BT_HEX, BT_HEX, BT_HEX,
+/* 0x44 */ BT_HEX, BT_HEX, BT_HEX, BT_NMSTRT,
+/* 0x48 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x4C */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x50 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x54 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x58 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_LSQB,
+/* 0x5C */ BT_OTHER, BT_RSQB, BT_OTHER, BT_NMSTRT,
+/* 0x60 */ BT_OTHER, BT_HEX, BT_HEX, BT_HEX,
+/* 0x64 */ BT_HEX, BT_HEX, BT_HEX, BT_NMSTRT,
+/* 0x68 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x6C */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x70 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x74 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x78 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_OTHER,
+/* 0x7C */ BT_VERBAR, BT_OTHER, BT_OTHER, BT_OTHER,
diff --git a/src/lib/expat/expat.h b/src/lib/expat/expat.h
new file mode 100644
index 00000000000..00ddbcee4ad
--- /dev/null
+++ b/src/lib/expat/expat.h
@@ -0,0 +1,1013 @@
+/* Copyright (c) 1998, 1999, 2000 Thai Open Source Software Center Ltd
+ See the file COPYING for copying permission.
+*/
+
+#ifndef Expat_INCLUDED
+#define Expat_INCLUDED 1
+
+#ifdef __VMS
+/* 0 1 2 3 0 1 2 3
+ 1234567890123456789012345678901 1234567890123456789012345678901 */
+#define XML_SetProcessingInstructionHandler XML_SetProcessingInstrHandler
+#define XML_SetUnparsedEntityDeclHandler XML_SetUnparsedEntDeclHandler
+#define XML_SetStartNamespaceDeclHandler XML_SetStartNamespcDeclHandler
+#define XML_SetExternalEntityRefHandlerArg XML_SetExternalEntRefHandlerArg
+#endif
+
+#include <stdlib.h>
+#include "expat_external.h"
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+struct XML_ParserStruct;
+typedef struct XML_ParserStruct *XML_Parser;
+
+/* Should this be defined using stdbool.h when C99 is available? */
+typedef unsigned char XML_Bool;
+#define XML_TRUE ((XML_Bool) 1)
+#define XML_FALSE ((XML_Bool) 0)
+
+/* The XML_Status enum gives the possible return values for several
+ API functions. The preprocessor #defines are included so this
+ stanza can be added to code that still needs to support older
+ versions of Expat 1.95.x:
+
+ #ifndef XML_STATUS_OK
+ #define XML_STATUS_OK 1
+ #define XML_STATUS_ERROR 0
+ #endif
+
+ Otherwise, the #define hackery is quite ugly and would have been
+ dropped.
+*/
+enum XML_Status {
+ XML_STATUS_ERROR = 0,
+#define XML_STATUS_ERROR XML_STATUS_ERROR
+ XML_STATUS_OK = 1,
+#define XML_STATUS_OK XML_STATUS_OK
+ XML_STATUS_SUSPENDED = 2
+#define XML_STATUS_SUSPENDED XML_STATUS_SUSPENDED
+};
+
+enum XML_Error {
+ XML_ERROR_NONE,
+ XML_ERROR_NO_MEMORY,
+ XML_ERROR_SYNTAX,
+ XML_ERROR_NO_ELEMENTS,
+ XML_ERROR_INVALID_TOKEN,
+ XML_ERROR_UNCLOSED_TOKEN,
+ XML_ERROR_PARTIAL_CHAR,
+ XML_ERROR_TAG_MISMATCH,
+ XML_ERROR_DUPLICATE_ATTRIBUTE,
+ XML_ERROR_JUNK_AFTER_DOC_ELEMENT,
+ XML_ERROR_PARAM_ENTITY_REF,
+ XML_ERROR_UNDEFINED_ENTITY,
+ XML_ERROR_RECURSIVE_ENTITY_REF,
+ XML_ERROR_ASYNC_ENTITY,
+ XML_ERROR_BAD_CHAR_REF,
+ XML_ERROR_BINARY_ENTITY_REF,
+ XML_ERROR_ATTRIBUTE_EXTERNAL_ENTITY_REF,
+ XML_ERROR_MISPLACED_XML_PI,
+ XML_ERROR_UNKNOWN_ENCODING,
+ XML_ERROR_INCORRECT_ENCODING,
+ XML_ERROR_UNCLOSED_CDATA_SECTION,
+ XML_ERROR_EXTERNAL_ENTITY_HANDLING,
+ XML_ERROR_NOT_STANDALONE,
+ XML_ERROR_UNEXPECTED_STATE,
+ XML_ERROR_ENTITY_DECLARED_IN_PE,
+ XML_ERROR_FEATURE_REQUIRES_XML_DTD,
+ XML_ERROR_CANT_CHANGE_FEATURE_ONCE_PARSING,
+ /* Added in 1.95.7. */
+ XML_ERROR_UNBOUND_PREFIX,
+ /* Added in 1.95.8. */
+ XML_ERROR_UNDECLARING_PREFIX,
+ XML_ERROR_INCOMPLETE_PE,
+ XML_ERROR_XML_DECL,
+ XML_ERROR_TEXT_DECL,
+ XML_ERROR_PUBLICID,
+ XML_ERROR_SUSPENDED,
+ XML_ERROR_NOT_SUSPENDED,
+ XML_ERROR_ABORTED,
+ XML_ERROR_FINISHED,
+ XML_ERROR_SUSPEND_PE,
+ /* Added in 2.0. */
+ XML_ERROR_RESERVED_PREFIX_XML,
+ XML_ERROR_RESERVED_PREFIX_XMLNS,
+ XML_ERROR_RESERVED_NAMESPACE_URI
+};
+
+enum XML_Content_Type {
+ XML_CTYPE_EMPTY = 1,
+ XML_CTYPE_ANY,
+ XML_CTYPE_MIXED,
+ XML_CTYPE_NAME,
+ XML_CTYPE_CHOICE,
+ XML_CTYPE_SEQ
+};
+
+enum XML_Content_Quant {
+ XML_CQUANT_NONE,
+ XML_CQUANT_OPT,
+ XML_CQUANT_REP,
+ XML_CQUANT_PLUS
+};
+
+/* If type == XML_CTYPE_EMPTY or XML_CTYPE_ANY, then quant will be
+ XML_CQUANT_NONE, and the other fields will be zero or NULL.
+ If type == XML_CTYPE_MIXED, then quant will be NONE or REP and
+ numchildren will contain number of elements that may be mixed in
+ and children point to an array of XML_Content cells that will be
+ all of XML_CTYPE_NAME type with no quantification.
+
+ If type == XML_CTYPE_NAME, then the name points to the name, and
+ the numchildren field will be zero and children will be NULL. The
+ quant fields indicates any quantifiers placed on the name.
+
+ CHOICE and SEQ will have name NULL, the number of children in
+ numchildren and children will point, recursively, to an array
+ of XML_Content cells.
+
+ The EMPTY, ANY, and MIXED types will only occur at top level.
+*/
+
+typedef struct XML_cp XML_Content;
+
+struct XML_cp {
+ enum XML_Content_Type type;
+ enum XML_Content_Quant quant;
+ XML_Char * name;
+ unsigned int numchildren;
+ XML_Content * children;
+};
+
+
+/* This is called for an element declaration. See above for
+ description of the model argument. It's the caller's responsibility
+ to free model when finished with it.
+*/
+typedef void (XMLCALL *XML_ElementDeclHandler) (void *userData,
+ const XML_Char *name,
+ XML_Content *model);
+
+XMLPARSEAPI(void)
+XML_SetElementDeclHandler(XML_Parser parser,
+ XML_ElementDeclHandler eldecl);
+
+/* The Attlist declaration handler is called for *each* attribute. So
+ a single Attlist declaration with multiple attributes declared will
+ generate multiple calls to this handler. The "default" parameter
+ may be NULL in the case of the "#IMPLIED" or "#REQUIRED"
+ keyword. The "isrequired" parameter will be true and the default
+ value will be NULL in the case of "#REQUIRED". If "isrequired" is
+ true and default is non-NULL, then this is a "#FIXED" default.
+*/
+typedef void (XMLCALL *XML_AttlistDeclHandler) (
+ void *userData,
+ const XML_Char *elname,
+ const XML_Char *attname,
+ const XML_Char *att_type,
+ const XML_Char *dflt,
+ int isrequired);
+
+XMLPARSEAPI(void)
+XML_SetAttlistDeclHandler(XML_Parser parser,
+ XML_AttlistDeclHandler attdecl);
+
+/* The XML declaration handler is called for *both* XML declarations
+ and text declarations. The way to distinguish is that the version
+ parameter will be NULL for text declarations. The encoding
+ parameter may be NULL for XML declarations. The standalone
+ parameter will be -1, 0, or 1 indicating respectively that there
+ was no standalone parameter in the declaration, that it was given
+ as no, or that it was given as yes.
+*/
+typedef void (XMLCALL *XML_XmlDeclHandler) (void *userData,
+ const XML_Char *version,
+ const XML_Char *encoding,
+ int standalone);
+
+XMLPARSEAPI(void)
+XML_SetXmlDeclHandler(XML_Parser parser,
+ XML_XmlDeclHandler xmldecl);
+
+
+typedef struct {
+ void *(*malloc_fcn)(size_t size);
+ void *(*realloc_fcn)(void *ptr, size_t size);
+ void (*free_fcn)(void *ptr);
+} XML_Memory_Handling_Suite;
+
+/* Constructs a new parser; encoding is the encoding specified by the
+ external protocol or NULL if there is none specified.
+*/
+XMLPARSEAPI(XML_Parser)
+XML_ParserCreate(const XML_Char *encoding);
+
+/* Constructs a new parser and namespace processor. Element type
+ names and attribute names that belong to a namespace will be
+ expanded; unprefixed attribute names are never expanded; unprefixed
+ element type names are expanded only if there is a default
+ namespace. The expanded name is the concatenation of the namespace
+ URI, the namespace separator character, and the local part of the
+ name. If the namespace separator is '\0' then the namespace URI
+ and the local part will be concatenated without any separator.
+ It is a programming error to use the separator '\0' with namespace
+ triplets (see XML_SetReturnNSTriplet).
+*/
+XMLPARSEAPI(XML_Parser)
+XML_ParserCreateNS(const XML_Char *encoding, XML_Char namespaceSeparator);
+
+
+/* Constructs a new parser using the memory management suite referred to
+ by memsuite. If memsuite is NULL, then use the standard library memory
+ suite. If namespaceSeparator is non-NULL it creates a parser with
+ namespace processing as described above. The character pointed at
+ will serve as the namespace separator.
+
+ All further memory operations used for the created parser will come from
+ the given suite.
+*/
+XMLPARSEAPI(XML_Parser)
+XML_ParserCreate_MM(const XML_Char *encoding,
+ const XML_Memory_Handling_Suite *memsuite,
+ const XML_Char *namespaceSeparator);
+
+/* Prepare a parser object to be re-used. This is particularly
+ valuable when memory allocation overhead is disproportionatly high,
+ such as when a large number of small documnents need to be parsed.
+ All handlers are cleared from the parser, except for the
+ unknownEncodingHandler. The parser's external state is re-initialized
+ except for the values of ns and ns_triplets.
+
+ Added in Expat 1.95.3.
+*/
+XMLPARSEAPI(XML_Bool)
+XML_ParserReset(XML_Parser parser, const XML_Char *encoding);
+
+/* atts is array of name/value pairs, terminated by 0;
+ names and values are 0 terminated.
+*/
+typedef void (XMLCALL *XML_StartElementHandler) (void *userData,
+ const XML_Char *name,
+ const XML_Char **atts);
+
+typedef void (XMLCALL *XML_EndElementHandler) (void *userData,
+ const XML_Char *name);
+
+
+/* s is not 0 terminated. */
+typedef void (XMLCALL *XML_CharacterDataHandler) (void *userData,
+ const XML_Char *s,
+ int len);
+
+/* target and data are 0 terminated */
+typedef void (XMLCALL *XML_ProcessingInstructionHandler) (
+ void *userData,
+ const XML_Char *target,
+ const XML_Char *data);
+
+/* data is 0 terminated */
+typedef void (XMLCALL *XML_CommentHandler) (void *userData,
+ const XML_Char *data);
+
+typedef void (XMLCALL *XML_StartCdataSectionHandler) (void *userData);
+typedef void (XMLCALL *XML_EndCdataSectionHandler) (void *userData);
+
+/* This is called for any characters in the XML document for which
+ there is no applicable handler. This includes both characters that
+ are part of markup which is of a kind that is not reported
+ (comments, markup declarations), or characters that are part of a
+ construct which could be reported but for which no handler has been
+ supplied. The characters are passed exactly as they were in the XML
+ document except that they will be encoded in UTF-8 or UTF-16.
+ Line boundaries are not normalized. Note that a byte order mark
+ character is not passed to the default handler. There are no
+ guarantees about how characters are divided between calls to the
+ default handler: for example, a comment might be split between
+ multiple calls.
+*/
+typedef void (XMLCALL *XML_DefaultHandler) (void *userData,
+ const XML_Char *s,
+ int len);
+
+/* This is called for the start of the DOCTYPE declaration, before
+ any DTD or internal subset is parsed.
+*/
+typedef void (XMLCALL *XML_StartDoctypeDeclHandler) (
+ void *userData,
+ const XML_Char *doctypeName,
+ const XML_Char *sysid,
+ const XML_Char *pubid,
+ int has_internal_subset);
+
+/* This is called for the start of the DOCTYPE declaration when the
+ closing > is encountered, but after processing any external
+ subset.
+*/
+typedef void (XMLCALL *XML_EndDoctypeDeclHandler)(void *userData);
+
+/* This is called for entity declarations. The is_parameter_entity
+ argument will be non-zero if the entity is a parameter entity, zero
+ otherwise.
+
+ For internal entities (<!ENTITY foo "bar">), value will
+ be non-NULL and systemId, publicID, and notationName will be NULL.
+ The value string is NOT nul-terminated; the length is provided in
+ the value_length argument. Since it is legal to have zero-length
+ values, do not use this argument to test for internal entities.
+
+ For external entities, value will be NULL and systemId will be
+ non-NULL. The publicId argument will be NULL unless a public
+ identifier was provided. The notationName argument will have a
+ non-NULL value only for unparsed entity declarations.
+
+ Note that is_parameter_entity can't be changed to XML_Bool, since
+ that would break binary compatibility.
+*/
+typedef void (XMLCALL *XML_EntityDeclHandler) (
+ void *userData,
+ const XML_Char *entityName,
+ int is_parameter_entity,
+ const XML_Char *value,
+ int value_length,
+ const XML_Char *base,
+ const XML_Char *systemId,
+ const XML_Char *publicId,
+ const XML_Char *notationName);
+
+XMLPARSEAPI(void)
+XML_SetEntityDeclHandler(XML_Parser parser,
+ XML_EntityDeclHandler handler);
+
+/* OBSOLETE -- OBSOLETE -- OBSOLETE
+ This handler has been superceded by the EntityDeclHandler above.
+ It is provided here for backward compatibility.
+
+ This is called for a declaration of an unparsed (NDATA) entity.
+ The base argument is whatever was set by XML_SetBase. The
+ entityName, systemId and notationName arguments will never be
+ NULL. The other arguments may be.
+*/
+typedef void (XMLCALL *XML_UnparsedEntityDeclHandler) (
+ void *userData,
+ const XML_Char *entityName,
+ const XML_Char *base,
+ const XML_Char *systemId,
+ const XML_Char *publicId,
+ const XML_Char *notationName);
+
+/* This is called for a declaration of notation. The base argument is
+ whatever was set by XML_SetBase. The notationName will never be
+ NULL. The other arguments can be.
+*/
+typedef void (XMLCALL *XML_NotationDeclHandler) (
+ void *userData,
+ const XML_Char *notationName,
+ const XML_Char *base,
+ const XML_Char *systemId,
+ const XML_Char *publicId);
+
+/* When namespace processing is enabled, these are called once for
+ each namespace declaration. The call to the start and end element
+ handlers occur between the calls to the start and end namespace
+ declaration handlers. For an xmlns attribute, prefix will be
+ NULL. For an xmlns="" attribute, uri will be NULL.
+*/
+typedef void (XMLCALL *XML_StartNamespaceDeclHandler) (
+ void *userData,
+ const XML_Char *prefix,
+ const XML_Char *uri);
+
+typedef void (XMLCALL *XML_EndNamespaceDeclHandler) (
+ void *userData,
+ const XML_Char *prefix);
+
+/* This is called if the document is not standalone, that is, it has an
+ external subset or a reference to a parameter entity, but does not
+ have standalone="yes". If this handler returns XML_STATUS_ERROR,
+ then processing will not continue, and the parser will return a
+ XML_ERROR_NOT_STANDALONE error.
+ If parameter entity parsing is enabled, then in addition to the
+ conditions above this handler will only be called if the referenced
+ entity was actually read.
+*/
+typedef int (XMLCALL *XML_NotStandaloneHandler) (void *userData);
+
+/* This is called for a reference to an external parsed general
+ entity. The referenced entity is not automatically parsed. The
+ application can parse it immediately or later using
+ XML_ExternalEntityParserCreate.
+
+ The parser argument is the parser parsing the entity containing the
+ reference; it can be passed as the parser argument to
+ XML_ExternalEntityParserCreate. The systemId argument is the
+ system identifier as specified in the entity declaration; it will
+ not be NULL.
+
+ The base argument is the system identifier that should be used as
+ the base for resolving systemId if systemId was relative; this is
+ set by XML_SetBase; it may be NULL.
+
+ The publicId argument is the public identifier as specified in the
+ entity declaration, or NULL if none was specified; the whitespace
+ in the public identifier will have been normalized as required by
+ the XML spec.
+
+ The context argument specifies the parsing context in the format
+ expected by the context argument to XML_ExternalEntityParserCreate;
+ context is valid only until the handler returns, so if the
+ referenced entity is to be parsed later, it must be copied.
+ context is NULL only when the entity is a parameter entity.
+
+ The handler should return XML_STATUS_ERROR if processing should not
+ continue because of a fatal error in the handling of the external
+ entity. In this case the calling parser will return an
+ XML_ERROR_EXTERNAL_ENTITY_HANDLING error.
+
+ Note that unlike other handlers the first argument is the parser,
+ not userData.
+*/
+typedef int (XMLCALL *XML_ExternalEntityRefHandler) (
+ XML_Parser parser,
+ const XML_Char *context,
+ const XML_Char *base,
+ const XML_Char *systemId,
+ const XML_Char *publicId);
+
+/* This is called in two situations:
+ 1) An entity reference is encountered for which no declaration
+ has been read *and* this is not an error.
+ 2) An internal entity reference is read, but not expanded, because
+ XML_SetDefaultHandler has been called.
+ Note: skipped parameter entities in declarations and skipped general
+ entities in attribute values cannot be reported, because
+ the event would be out of sync with the reporting of the
+ declarations or attribute values
+*/
+typedef void (XMLCALL *XML_SkippedEntityHandler) (
+ void *userData,
+ const XML_Char *entityName,
+ int is_parameter_entity);
+
+/* This structure is filled in by the XML_UnknownEncodingHandler to
+ provide information to the parser about encodings that are unknown
+ to the parser.
+
+ The map[b] member gives information about byte sequences whose
+ first byte is b.
+
+ If map[b] is c where c is >= 0, then b by itself encodes the
+ Unicode scalar value c.
+
+ If map[b] is -1, then the byte sequence is malformed.
+
+ If map[b] is -n, where n >= 2, then b is the first byte of an
+ n-byte sequence that encodes a single Unicode scalar value.
+
+ The data member will be passed as the first argument to the convert
+ function.
+
+ The convert function is used to convert multibyte sequences; s will
+ point to a n-byte sequence where map[(unsigned char)*s] == -n. The
+ convert function must return the Unicode scalar value represented
+ by this byte sequence or -1 if the byte sequence is malformed.
+
+ The convert function may be NULL if the encoding is a single-byte
+ encoding, that is if map[b] >= -1 for all bytes b.
+
+ When the parser is finished with the encoding, then if release is
+ not NULL, it will call release passing it the data member; once
+ release has been called, the convert function will not be called
+ again.
+
+ Expat places certain restrictions on the encodings that are supported
+ using this mechanism.
+
+ 1. Every ASCII character that can appear in a well-formed XML document,
+ other than the characters
+
+ $@\^`{}~
+
+ must be represented by a single byte, and that byte must be the
+ same byte that represents that character in ASCII.
+
+ 2. No character may require more than 4 bytes to encode.
+
+ 3. All characters encoded must have Unicode scalar values <=
+ 0xFFFF, (i.e., characters that would be encoded by surrogates in
+ UTF-16 are not allowed). Note that this restriction doesn't
+ apply to the built-in support for UTF-8 and UTF-16.
+
+ 4. No Unicode character may be encoded by more than one distinct
+ sequence of bytes.
+*/
+typedef struct {
+ int map[256];
+ void *data;
+ int (XMLCALL *convert)(void *data, const char *s);
+ void (XMLCALL *release)(void *data);
+} XML_Encoding;
+
+/* This is called for an encoding that is unknown to the parser.
+
+ The encodingHandlerData argument is that which was passed as the
+ second argument to XML_SetUnknownEncodingHandler.
+
+ The name argument gives the name of the encoding as specified in
+ the encoding declaration.
+
+ If the callback can provide information about the encoding, it must
+ fill in the XML_Encoding structure, and return XML_STATUS_OK.
+ Otherwise it must return XML_STATUS_ERROR.
+
+ If info does not describe a suitable encoding, then the parser will
+ return an XML_UNKNOWN_ENCODING error.
+*/
+typedef int (XMLCALL *XML_UnknownEncodingHandler) (
+ void *encodingHandlerData,
+ const XML_Char *name,
+ XML_Encoding *info);
+
+XMLPARSEAPI(void)
+XML_SetElementHandler(XML_Parser parser,
+ XML_StartElementHandler start,
+ XML_EndElementHandler end);
+
+XMLPARSEAPI(void)
+XML_SetStartElementHandler(XML_Parser parser,
+ XML_StartElementHandler handler);
+
+XMLPARSEAPI(void)
+XML_SetEndElementHandler(XML_Parser parser,
+ XML_EndElementHandler handler);
+
+XMLPARSEAPI(void)
+XML_SetCharacterDataHandler(XML_Parser parser,
+ XML_CharacterDataHandler handler);
+
+XMLPARSEAPI(void)
+XML_SetProcessingInstructionHandler(XML_Parser parser,
+ XML_ProcessingInstructionHandler handler);
+XMLPARSEAPI(void)
+XML_SetCommentHandler(XML_Parser parser,
+ XML_CommentHandler handler);
+
+XMLPARSEAPI(void)
+XML_SetCdataSectionHandler(XML_Parser parser,
+ XML_StartCdataSectionHandler start,
+ XML_EndCdataSectionHandler end);
+
+XMLPARSEAPI(void)
+XML_SetStartCdataSectionHandler(XML_Parser parser,
+ XML_StartCdataSectionHandler start);
+
+XMLPARSEAPI(void)
+XML_SetEndCdataSectionHandler(XML_Parser parser,
+ XML_EndCdataSectionHandler end);
+
+/* This sets the default handler and also inhibits expansion of
+ internal entities. These entity references will be passed to the
+ default handler, or to the skipped entity handler, if one is set.
+*/
+XMLPARSEAPI(void)
+XML_SetDefaultHandler(XML_Parser parser,
+ XML_DefaultHandler handler);
+
+/* This sets the default handler but does not inhibit expansion of
+ internal entities. The entity reference will not be passed to the
+ default handler.
+*/
+XMLPARSEAPI(void)
+XML_SetDefaultHandlerExpand(XML_Parser parser,
+ XML_DefaultHandler handler);
+
+XMLPARSEAPI(void)
+XML_SetDoctypeDeclHandler(XML_Parser parser,
+ XML_StartDoctypeDeclHandler start,
+ XML_EndDoctypeDeclHandler end);
+
+XMLPARSEAPI(void)
+XML_SetStartDoctypeDeclHandler(XML_Parser parser,
+ XML_StartDoctypeDeclHandler start);
+
+XMLPARSEAPI(void)
+XML_SetEndDoctypeDeclHandler(XML_Parser parser,
+ XML_EndDoctypeDeclHandler end);
+
+XMLPARSEAPI(void)
+XML_SetUnparsedEntityDeclHandler(XML_Parser parser,
+ XML_UnparsedEntityDeclHandler handler);
+
+XMLPARSEAPI(void)
+XML_SetNotationDeclHandler(XML_Parser parser,
+ XML_NotationDeclHandler handler);
+
+XMLPARSEAPI(void)
+XML_SetNamespaceDeclHandler(XML_Parser parser,
+ XML_StartNamespaceDeclHandler start,
+ XML_EndNamespaceDeclHandler end);
+
+XMLPARSEAPI(void)
+XML_SetStartNamespaceDeclHandler(XML_Parser parser,
+ XML_StartNamespaceDeclHandler start);
+
+XMLPARSEAPI(void)
+XML_SetEndNamespaceDeclHandler(XML_Parser parser,
+ XML_EndNamespaceDeclHandler end);
+
+XMLPARSEAPI(void)
+XML_SetNotStandaloneHandler(XML_Parser parser,
+ XML_NotStandaloneHandler handler);
+
+XMLPARSEAPI(void)
+XML_SetExternalEntityRefHandler(XML_Parser parser,
+ XML_ExternalEntityRefHandler handler);
+
+/* If a non-NULL value for arg is specified here, then it will be
+ passed as the first argument to the external entity ref handler
+ instead of the parser object.
+*/
+XMLPARSEAPI(void)
+XML_SetExternalEntityRefHandlerArg(XML_Parser parser,
+ void *arg);
+
+XMLPARSEAPI(void)
+XML_SetSkippedEntityHandler(XML_Parser parser,
+ XML_SkippedEntityHandler handler);
+
+XMLPARSEAPI(void)
+XML_SetUnknownEncodingHandler(XML_Parser parser,
+ XML_UnknownEncodingHandler handler,
+ void *encodingHandlerData);
+
+/* This can be called within a handler for a start element, end
+ element, processing instruction or character data. It causes the
+ corresponding markup to be passed to the default handler.
+*/
+XMLPARSEAPI(void)
+XML_DefaultCurrent(XML_Parser parser);
+
+/* If do_nst is non-zero, and namespace processing is in effect, and
+ a name has a prefix (i.e. an explicit namespace qualifier) then
+ that name is returned as a triplet in a single string separated by
+ the separator character specified when the parser was created: URI
+ + sep + local_name + sep + prefix.
+
+ If do_nst is zero, then namespace information is returned in the
+ default manner (URI + sep + local_name) whether or not the name
+ has a prefix.
+
+ Note: Calling XML_SetReturnNSTriplet after XML_Parse or
+ XML_ParseBuffer has no effect.
+*/
+
+XMLPARSEAPI(void)
+XML_SetReturnNSTriplet(XML_Parser parser, int do_nst);
+
+/* This value is passed as the userData argument to callbacks. */
+XMLPARSEAPI(void)
+XML_SetUserData(XML_Parser parser, void *userData);
+
+/* Returns the last value set by XML_SetUserData or NULL. */
+#define XML_GetUserData(parser) (*(void **)(parser))
+
+/* This is equivalent to supplying an encoding argument to
+ XML_ParserCreate. On success XML_SetEncoding returns non-zero,
+ zero otherwise.
+ Note: Calling XML_SetEncoding after XML_Parse or XML_ParseBuffer
+ has no effect and returns XML_STATUS_ERROR.
+*/
+XMLPARSEAPI(enum XML_Status)
+XML_SetEncoding(XML_Parser parser, const XML_Char *encoding);
+
+/* If this function is called, then the parser will be passed as the
+ first argument to callbacks instead of userData. The userData will
+ still be accessible using XML_GetUserData.
+*/
+XMLPARSEAPI(void)
+XML_UseParserAsHandlerArg(XML_Parser parser);
+
+/* If useDTD == XML_TRUE is passed to this function, then the parser
+ will assume that there is an external subset, even if none is
+ specified in the document. In such a case the parser will call the
+ externalEntityRefHandler with a value of NULL for the systemId
+ argument (the publicId and context arguments will be NULL as well).
+ Note: For the purpose of checking WFC: Entity Declared, passing
+ useDTD == XML_TRUE will make the parser behave as if the document
+ had a DTD with an external subset.
+ Note: If this function is called, then this must be done before
+ the first call to XML_Parse or XML_ParseBuffer, since it will
+ have no effect after that. Returns
+ XML_ERROR_CANT_CHANGE_FEATURE_ONCE_PARSING.
+ Note: If the document does not have a DOCTYPE declaration at all,
+ then startDoctypeDeclHandler and endDoctypeDeclHandler will not
+ be called, despite an external subset being parsed.
+ Note: If XML_DTD is not defined when Expat is compiled, returns
+ XML_ERROR_FEATURE_REQUIRES_XML_DTD.
+*/
+XMLPARSEAPI(enum XML_Error)
+XML_UseForeignDTD(XML_Parser parser, XML_Bool useDTD);
+
+
+/* Sets the base to be used for resolving relative URIs in system
+ identifiers in declarations. Resolving relative identifiers is
+ left to the application: this value will be passed through as the
+ base argument to the XML_ExternalEntityRefHandler,
+ XML_NotationDeclHandler and XML_UnparsedEntityDeclHandler. The base
+ argument will be copied. Returns XML_STATUS_ERROR if out of memory,
+ XML_STATUS_OK otherwise.
+*/
+XMLPARSEAPI(enum XML_Status)
+XML_SetBase(XML_Parser parser, const XML_Char *base);
+
+XMLPARSEAPI(const XML_Char *)
+XML_GetBase(XML_Parser parser);
+
+/* Returns the number of the attribute/value pairs passed in last call
+ to the XML_StartElementHandler that were specified in the start-tag
+ rather than defaulted. Each attribute/value pair counts as 2; thus
+ this correspondds to an index into the atts array passed to the
+ XML_StartElementHandler.
+*/
+XMLPARSEAPI(int)
+XML_GetSpecifiedAttributeCount(XML_Parser parser);
+
+/* Returns the index of the ID attribute passed in the last call to
+ XML_StartElementHandler, or -1 if there is no ID attribute. Each
+ attribute/value pair counts as 2; thus this correspondds to an
+ index into the atts array passed to the XML_StartElementHandler.
+*/
+XMLPARSEAPI(int)
+XML_GetIdAttributeIndex(XML_Parser parser);
+
+/* Parses some input. Returns XML_STATUS_ERROR if a fatal error is
+ detected. The last call to XML_Parse must have isFinal true; len
+ may be zero for this call (or any other).
+
+ Though the return values for these functions has always been
+ described as a Boolean value, the implementation, at least for the
+ 1.95.x series, has always returned exactly one of the XML_Status
+ values.
+*/
+XMLPARSEAPI(enum XML_Status)
+XML_Parse(XML_Parser parser, const char *s, int len, int isFinal);
+
+XMLPARSEAPI(void *)
+XML_GetBuffer(XML_Parser parser, int len);
+
+XMLPARSEAPI(enum XML_Status)
+XML_ParseBuffer(XML_Parser parser, int len, int isFinal);
+
+/* Stops parsing, causing XML_Parse() or XML_ParseBuffer() to return.
+ Must be called from within a call-back handler, except when aborting
+ (resumable = 0) an already suspended parser. Some call-backs may
+ still follow because they would otherwise get lost. Examples:
+ - endElementHandler() for empty elements when stopped in
+ startElementHandler(),
+ - endNameSpaceDeclHandler() when stopped in endElementHandler(),
+ and possibly others.
+
+ Can be called from most handlers, including DTD related call-backs,
+ except when parsing an external parameter entity and resumable != 0.
+ Returns XML_STATUS_OK when successful, XML_STATUS_ERROR otherwise.
+ Possible error codes:
+ - XML_ERROR_SUSPENDED: when suspending an already suspended parser.
+ - XML_ERROR_FINISHED: when the parser has already finished.
+ - XML_ERROR_SUSPEND_PE: when suspending while parsing an external PE.
+
+ When resumable != 0 (true) then parsing is suspended, that is,
+ XML_Parse() and XML_ParseBuffer() return XML_STATUS_SUSPENDED.
+ Otherwise, parsing is aborted, that is, XML_Parse() and XML_ParseBuffer()
+ return XML_STATUS_ERROR with error code XML_ERROR_ABORTED.
+
+ *Note*:
+ This will be applied to the current parser instance only, that is, if
+ there is a parent parser then it will continue parsing when the
+ externalEntityRefHandler() returns. It is up to the implementation of
+ the externalEntityRefHandler() to call XML_StopParser() on the parent
+ parser (recursively), if one wants to stop parsing altogether.
+
+ When suspended, parsing can be resumed by calling XML_ResumeParser().
+*/
+XMLPARSEAPI(enum XML_Status)
+XML_StopParser(XML_Parser parser, XML_Bool resumable);
+
+/* Resumes parsing after it has been suspended with XML_StopParser().
+ Must not be called from within a handler call-back. Returns same
+ status codes as XML_Parse() or XML_ParseBuffer().
+ Additional error code XML_ERROR_NOT_SUSPENDED possible.
+
+ *Note*:
+ This must be called on the most deeply nested child parser instance
+ first, and on its parent parser only after the child parser has finished,
+ to be applied recursively until the document entity's parser is restarted.
+ That is, the parent parser will not resume by itself and it is up to the
+ application to call XML_ResumeParser() on it at the appropriate moment.
+*/
+XMLPARSEAPI(enum XML_Status)
+XML_ResumeParser(XML_Parser parser);
+
+enum XML_Parsing {
+ XML_INITIALIZED,
+ XML_PARSING,
+ XML_FINISHED,
+ XML_SUSPENDED
+};
+
+typedef struct {
+ enum XML_Parsing parsing;
+ XML_Bool finalBuffer;
+} XML_ParsingStatus;
+
+/* Returns status of parser with respect to being initialized, parsing,
+ finished, or suspended and processing the final buffer.
+ XXX XML_Parse() and XML_ParseBuffer() should return XML_ParsingStatus,
+ XXX with XML_FINISHED_OK or XML_FINISHED_ERROR replacing XML_FINISHED
+*/
+XMLPARSEAPI(void)
+XML_GetParsingStatus(XML_Parser parser, XML_ParsingStatus *status);
+
+/* Creates an XML_Parser object that can parse an external general
+ entity; context is a '\0'-terminated string specifying the parse
+ context; encoding is a '\0'-terminated string giving the name of
+ the externally specified encoding, or NULL if there is no
+ externally specified encoding. The context string consists of a
+ sequence of tokens separated by formfeeds (\f); a token consisting
+ of a name specifies that the general entity of the name is open; a
+ token of the form prefix=uri specifies the namespace for a
+ particular prefix; a token of the form =uri specifies the default
+ namespace. This can be called at any point after the first call to
+ an ExternalEntityRefHandler so longer as the parser has not yet
+ been freed. The new parser is completely independent and may
+ safely be used in a separate thread. The handlers and userData are
+ initialized from the parser argument. Returns NULL if out of memory.
+ Otherwise returns a new XML_Parser object.
+*/
+XMLPARSEAPI(XML_Parser)
+XML_ExternalEntityParserCreate(XML_Parser parser,
+ const XML_Char *context,
+ const XML_Char *encoding);
+
+enum XML_ParamEntityParsing {
+ XML_PARAM_ENTITY_PARSING_NEVER,
+ XML_PARAM_ENTITY_PARSING_UNLESS_STANDALONE,
+ XML_PARAM_ENTITY_PARSING_ALWAYS
+};
+
+/* Controls parsing of parameter entities (including the external DTD
+ subset). If parsing of parameter entities is enabled, then
+ references to external parameter entities (including the external
+ DTD subset) will be passed to the handler set with
+ XML_SetExternalEntityRefHandler. The context passed will be 0.
+
+ Unlike external general entities, external parameter entities can
+ only be parsed synchronously. If the external parameter entity is
+ to be parsed, it must be parsed during the call to the external
+ entity ref handler: the complete sequence of
+ XML_ExternalEntityParserCreate, XML_Parse/XML_ParseBuffer and
+ XML_ParserFree calls must be made during this call. After
+ XML_ExternalEntityParserCreate has been called to create the parser
+ for the external parameter entity (context must be 0 for this
+ call), it is illegal to make any calls on the old parser until
+ XML_ParserFree has been called on the newly created parser.
+ If the library has been compiled without support for parameter
+ entity parsing (ie without XML_DTD being defined), then
+ XML_SetParamEntityParsing will return 0 if parsing of parameter
+ entities is requested; otherwise it will return non-zero.
+ Note: If XML_SetParamEntityParsing is called after XML_Parse or
+ XML_ParseBuffer, then it has no effect and will always return 0.
+*/
+XMLPARSEAPI(int)
+XML_SetParamEntityParsing(XML_Parser parser,
+ enum XML_ParamEntityParsing parsing);
+
+/* If XML_Parse or XML_ParseBuffer have returned XML_STATUS_ERROR, then
+ XML_GetErrorCode returns information about the error.
+*/
+XMLPARSEAPI(enum XML_Error)
+XML_GetErrorCode(XML_Parser parser);
+
+/* These functions return information about the current parse
+ location. They may be called from any callback called to report
+ some parse event; in this case the location is the location of the
+ first of the sequence of characters that generated the event. When
+ called from callbacks generated by declarations in the document
+ prologue, the location identified isn't as neatly defined, but will
+ be within the relevant markup. When called outside of the callback
+ functions, the position indicated will be just past the last parse
+ event (regardless of whether there was an associated callback).
+
+ They may also be called after returning from a call to XML_Parse
+ or XML_ParseBuffer. If the return value is XML_STATUS_ERROR then
+ the location is the location of the character at which the error
+ was detected; otherwise the location is the location of the last
+ parse event, as described above.
+*/
+XMLPARSEAPI(XML_Size) XML_GetCurrentLineNumber(XML_Parser parser);
+XMLPARSEAPI(XML_Size) XML_GetCurrentColumnNumber(XML_Parser parser);
+XMLPARSEAPI(XML_Index) XML_GetCurrentByteIndex(XML_Parser parser);
+
+/* Return the number of bytes in the current event.
+ Returns 0 if the event is in an internal entity.
+*/
+XMLPARSEAPI(int)
+XML_GetCurrentByteCount(XML_Parser parser);
+
+/* If XML_CONTEXT_BYTES is defined, returns the input buffer, sets
+ the integer pointed to by offset to the offset within this buffer
+ of the current parse position, and sets the integer pointed to by size
+ to the size of this buffer (the number of input bytes). Otherwise
+ returns a NULL pointer. Also returns a NULL pointer if a parse isn't
+ active.
+
+ NOTE: The character pointer returned should not be used outside
+ the handler that makes the call.
+*/
+XMLPARSEAPI(const char *)
+XML_GetInputContext(XML_Parser parser,
+ int *offset,
+ int *size);
+
+/* For backwards compatibility with previous versions. */
+#define XML_GetErrorLineNumber XML_GetCurrentLineNumber
+#define XML_GetErrorColumnNumber XML_GetCurrentColumnNumber
+#define XML_GetErrorByteIndex XML_GetCurrentByteIndex
+
+/* Frees the content model passed to the element declaration handler */
+XMLPARSEAPI(void)
+XML_FreeContentModel(XML_Parser parser, XML_Content *model);
+
+/* Exposing the memory handling functions used in Expat */
+XMLPARSEAPI(void *)
+XML_MemMalloc(XML_Parser parser, size_t size);
+
+XMLPARSEAPI(void *)
+XML_MemRealloc(XML_Parser parser, void *ptr, size_t size);
+
+XMLPARSEAPI(void)
+XML_MemFree(XML_Parser parser, void *ptr);
+
+/* Frees memory used by the parser. */
+XMLPARSEAPI(void)
+XML_ParserFree(XML_Parser parser);
+
+/* Returns a string describing the error. */
+XMLPARSEAPI(const XML_LChar *)
+XML_ErrorString(enum XML_Error code);
+
+/* Return a string containing the version number of this expat */
+XMLPARSEAPI(const XML_LChar *)
+XML_ExpatVersion(void);
+
+typedef struct {
+ int major;
+ int minor;
+ int micro;
+} XML_Expat_Version;
+
+/* Return an XML_Expat_Version structure containing numeric version
+ number information for this version of expat.
+*/
+XMLPARSEAPI(XML_Expat_Version)
+XML_ExpatVersionInfo(void);
+
+/* Added in Expat 1.95.5. */
+enum XML_FeatureEnum {
+ XML_FEATURE_END = 0,
+ XML_FEATURE_UNICODE,
+ XML_FEATURE_UNICODE_WCHAR_T,
+ XML_FEATURE_DTD,
+ XML_FEATURE_CONTEXT_BYTES,
+ XML_FEATURE_MIN_SIZE,
+ XML_FEATURE_SIZEOF_XML_CHAR,
+ XML_FEATURE_SIZEOF_XML_LCHAR,
+ XML_FEATURE_NS
+ /* Additional features must be added to the end of this enum. */
+};
+
+typedef struct {
+ enum XML_FeatureEnum feature;
+ const XML_LChar *name;
+ long int value;
+} XML_Feature;
+
+XMLPARSEAPI(const XML_Feature *)
+XML_GetFeatureList(void);
+
+
+/* Expat follows the GNU/Linux convention of odd number minor version for
+ beta/development releases and even number minor version for stable
+ releases. Micro is bumped with each release, and set to 0 with each
+ change to major or minor version.
+*/
+#define XML_MAJOR_VERSION 2
+#define XML_MINOR_VERSION 0
+#define XML_MICRO_VERSION 0
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* not Expat_INCLUDED */
diff --git a/src/lib/expat/expat_external.h b/src/lib/expat/expat_external.h
new file mode 100644
index 00000000000..99c5b922cbf
--- /dev/null
+++ b/src/lib/expat/expat_external.h
@@ -0,0 +1,115 @@
+/* Copyright (c) 1998, 1999, 2000 Thai Open Source Software Center Ltd
+ See the file COPYING for copying permission.
+*/
+
+#ifndef Expat_External_INCLUDED
+#define Expat_External_INCLUDED 1
+
+/* External API definitions */
+
+#if defined(_MSC_EXTENSIONS) && !defined(__BEOS__) && !defined(__CYGWIN__)
+#define XML_USE_MSC_EXTENSIONS 1
+#endif
+
+/* Expat tries very hard to make the API boundary very specifically
+ defined. There are two macros defined to control this boundary;
+ each of these can be defined before including this header to
+ achieve some different behavior, but doing so it not recommended or
+ tested frequently.
+
+ XMLCALL - The calling convention to use for all calls across the
+ "library boundary." This will default to cdecl, and
+ try really hard to tell the compiler that's what we
+ want.
+
+ XMLIMPORT - Whatever magic is needed to note that a function is
+ to be imported from a dynamically loaded library
+ (.dll, .so, or .sl, depending on your platform).
+
+ The XMLCALL macro was added in Expat 1.95.7. The only one which is
+ expected to be directly useful in client code is XMLCALL.
+
+ Note that on at least some Unix versions, the Expat library must be
+ compiled with the cdecl calling convention as the default since
+ system headers may assume the cdecl convention.
+*/
+#ifndef XMLCALL
+#if defined(XML_USE_MSC_EXTENSIONS)
+#define XMLCALL __cdecl
+#elif defined(__GNUC__) && defined(__i386)
+#define XMLCALL __attribute__((cdecl))
+#else
+/* For any platform which uses this definition and supports more than
+ one calling convention, we need to extend this definition to
+ declare the convention used on that platform, if it's possible to
+ do so.
+
+ If this is the case for your platform, please file a bug report
+ with information on how to identify your platform via the C
+ pre-processor and how to specify the same calling convention as the
+ platform's malloc() implementation.
+*/
+#define XMLCALL
+#endif
+#endif /* not defined XMLCALL */
+
+
+#if !defined(XML_STATIC) && !defined(XMLIMPORT)
+#ifndef XML_BUILDING_EXPAT
+/* using Expat from an application */
+
+#ifdef XML_USE_MSC_EXTENSIONS
+#define XMLIMPORT __declspec(dllimport)
+#endif
+
+#endif
+#endif /* not defined XML_STATIC */
+
+
+/* If we didn't define it above, define it away: */
+#ifndef XMLIMPORT
+#define XMLIMPORT
+#endif
+
+
+#define XMLPARSEAPI(type) XMLIMPORT type XMLCALL
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef XML_UNICODE_WCHAR_T
+#define XML_UNICODE
+#endif
+
+#ifdef XML_UNICODE /* Information is UTF-16 encoded. */
+#ifdef XML_UNICODE_WCHAR_T
+typedef wchar_t XML_Char;
+typedef wchar_t XML_LChar;
+#else
+typedef unsigned short XML_Char;
+typedef char XML_LChar;
+#endif /* XML_UNICODE_WCHAR_T */
+#else /* Information is UTF-8 encoded. */
+typedef char XML_Char;
+typedef char XML_LChar;
+#endif /* XML_UNICODE */
+
+#ifdef XML_LARGE_SIZE /* Use large integers for file/stream positions. */
+#if defined(XML_USE_MSC_EXTENSIONS) && _MSC_VER < 1400
+typedef __int64 XML_Index;
+typedef unsigned __int64 XML_Size;
+#else
+typedef long long XML_Index;
+typedef unsigned long long XML_Size;
+#endif
+#else
+typedef long XML_Index;
+typedef unsigned long XML_Size;
+#endif /* XML_LARGE_SIZE */
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* not Expat_External_INCLUDED */
diff --git a/src/lib/expat/iasciitab.h b/src/lib/expat/iasciitab.h
new file mode 100644
index 00000000000..24a1d5ccc9a
--- /dev/null
+++ b/src/lib/expat/iasciitab.h
@@ -0,0 +1,37 @@
+/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+ See the file COPYING for copying permission.
+*/
+
+/* Like asciitab.h, except that 0xD has code BT_S rather than BT_CR */
+/* 0x00 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x04 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x08 */ BT_NONXML, BT_S, BT_LF, BT_NONXML,
+/* 0x0C */ BT_NONXML, BT_S, BT_NONXML, BT_NONXML,
+/* 0x10 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x14 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x18 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x1C */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x20 */ BT_S, BT_EXCL, BT_QUOT, BT_NUM,
+/* 0x24 */ BT_OTHER, BT_PERCNT, BT_AMP, BT_APOS,
+/* 0x28 */ BT_LPAR, BT_RPAR, BT_AST, BT_PLUS,
+/* 0x2C */ BT_COMMA, BT_MINUS, BT_NAME, BT_SOL,
+/* 0x30 */ BT_DIGIT, BT_DIGIT, BT_DIGIT, BT_DIGIT,
+/* 0x34 */ BT_DIGIT, BT_DIGIT, BT_DIGIT, BT_DIGIT,
+/* 0x38 */ BT_DIGIT, BT_DIGIT, BT_COLON, BT_SEMI,
+/* 0x3C */ BT_LT, BT_EQUALS, BT_GT, BT_QUEST,
+/* 0x40 */ BT_OTHER, BT_HEX, BT_HEX, BT_HEX,
+/* 0x44 */ BT_HEX, BT_HEX, BT_HEX, BT_NMSTRT,
+/* 0x48 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x4C */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x50 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x54 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x58 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_LSQB,
+/* 0x5C */ BT_OTHER, BT_RSQB, BT_OTHER, BT_NMSTRT,
+/* 0x60 */ BT_OTHER, BT_HEX, BT_HEX, BT_HEX,
+/* 0x64 */ BT_HEX, BT_HEX, BT_HEX, BT_NMSTRT,
+/* 0x68 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x6C */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x70 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x74 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x78 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_OTHER,
+/* 0x7C */ BT_VERBAR, BT_OTHER, BT_OTHER, BT_OTHER,
diff --git a/src/lib/expat/intconfig.h b/src/lib/expat/intconfig.h
new file mode 100644
index 00000000000..88092e6a006
--- /dev/null
+++ b/src/lib/expat/intconfig.h
@@ -0,0 +1,19 @@
+#ifndef INTCONFIG_H
+#define INTCONFIG_H
+
+#include <stdlib.h>
+#include <string.h>
+
+#define XML_NS 1
+#define XML_DTD 1
+#define XML_CONTEXT_BYTES 1024
+
+#ifdef USE_LSB
+#define BYTEORDER 1234
+#else
+#define BYTEORDER 4321
+#endif
+
+#define HAVE_MEMMOVE
+
+#endif
diff --git a/src/lib/expat/internal.h b/src/lib/expat/internal.h
new file mode 100644
index 00000000000..ff056c659cd
--- /dev/null
+++ b/src/lib/expat/internal.h
@@ -0,0 +1,73 @@
+/* internal.h
+
+ Internal definitions used by Expat. This is not needed to compile
+ client code.
+
+ The following calling convention macros are defined for frequently
+ called functions:
+
+ FASTCALL - Used for those internal functions that have a simple
+ body and a low number of arguments and local variables.
+
+ PTRCALL - Used for functions called though function pointers.
+
+ PTRFASTCALL - Like PTRCALL, but for low number of arguments.
+
+ inline - Used for selected internal functions for which inlining
+ may improve performance on some platforms.
+
+ Note: Use of these macros is based on judgement, not hard rules,
+ and therefore subject to change.
+*/
+
+#if defined(__GNUC__) && defined(__i386__)
+/* We'll use this version by default only where we know it helps.
+
+ regparm() generates warnings on Solaris boxes. See SF bug #692878.
+
+ Instability reported with egcs on a RedHat Linux 7.3.
+ Let's comment out:
+ #define FASTCALL __attribute__((stdcall, regparm(3)))
+ and let's try this:
+*/
+#define FASTCALL __attribute__((regparm(3)))
+#define PTRFASTCALL __attribute__((regparm(3)))
+#endif
+
+/* Using __fastcall seems to have an unexpected negative effect under
+ MS VC++, especially for function pointers, so we won't use it for
+ now on that platform. It may be reconsidered for a future release
+ if it can be made more effective.
+ Likely reason: __fastcall on Windows is like stdcall, therefore
+ the compiler cannot perform stack optimizations for call clusters.
+*/
+
+/* Make sure all of these are defined if they aren't already. */
+
+#ifndef FASTCALL
+#define FASTCALL
+#endif
+
+#ifndef PTRCALL
+#define PTRCALL
+#endif
+
+#ifndef PTRFASTCALL
+#define PTRFASTCALL
+#endif
+
+#ifndef XML_MIN_SIZE
+#if !defined(__cplusplus) && !defined(inline)
+#ifdef __GNUC__
+#define inline __inline
+#endif /* __GNUC__ */
+#endif
+#endif /* XML_MIN_SIZE */
+
+#ifdef __cplusplus
+#define inline inline
+#else
+#ifndef inline
+#define inline
+#endif
+#endif
diff --git a/src/lib/expat/latin1tab.h b/src/lib/expat/latin1tab.h
new file mode 100644
index 00000000000..53c25d76b26
--- /dev/null
+++ b/src/lib/expat/latin1tab.h
@@ -0,0 +1,36 @@
+/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+ See the file COPYING for copying permission.
+*/
+
+/* 0x80 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0x84 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0x88 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0x8C */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0x90 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0x94 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0x98 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0x9C */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0xA0 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0xA4 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0xA8 */ BT_OTHER, BT_OTHER, BT_NMSTRT, BT_OTHER,
+/* 0xAC */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0xB0 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0xB4 */ BT_OTHER, BT_NMSTRT, BT_OTHER, BT_NAME,
+/* 0xB8 */ BT_OTHER, BT_OTHER, BT_NMSTRT, BT_OTHER,
+/* 0xBC */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0xC0 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xC4 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xC8 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xCC */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xD0 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xD4 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_OTHER,
+/* 0xD8 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xDC */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xE0 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xE4 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xE8 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xEC */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xF0 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xF4 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_OTHER,
+/* 0xF8 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xFC */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
diff --git a/src/lib/expat/nametab.h b/src/lib/expat/nametab.h
new file mode 100644
index 00000000000..b05e62c77a6
--- /dev/null
+++ b/src/lib/expat/nametab.h
@@ -0,0 +1,150 @@
+static const unsigned namingBitmap[] = {
+0x00000000, 0x00000000, 0x00000000, 0x00000000,
+0x00000000, 0x00000000, 0x00000000, 0x00000000,
+0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF,
+0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF,
+0x00000000, 0x04000000, 0x87FFFFFE, 0x07FFFFFE,
+0x00000000, 0x00000000, 0xFF7FFFFF, 0xFF7FFFFF,
+0xFFFFFFFF, 0x7FF3FFFF, 0xFFFFFDFE, 0x7FFFFFFF,
+0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFE00F, 0xFC31FFFF,
+0x00FFFFFF, 0x00000000, 0xFFFF0000, 0xFFFFFFFF,
+0xFFFFFFFF, 0xF80001FF, 0x00000003, 0x00000000,
+0x00000000, 0x00000000, 0x00000000, 0x00000000,
+0xFFFFD740, 0xFFFFFFFB, 0x547F7FFF, 0x000FFFFD,
+0xFFFFDFFE, 0xFFFFFFFF, 0xDFFEFFFF, 0xFFFFFFFF,
+0xFFFF0003, 0xFFFFFFFF, 0xFFFF199F, 0x033FCFFF,
+0x00000000, 0xFFFE0000, 0x027FFFFF, 0xFFFFFFFE,
+0x0000007F, 0x00000000, 0xFFFF0000, 0x000707FF,
+0x00000000, 0x07FFFFFE, 0x000007FE, 0xFFFE0000,
+0xFFFFFFFF, 0x7CFFFFFF, 0x002F7FFF, 0x00000060,
+0xFFFFFFE0, 0x23FFFFFF, 0xFF000000, 0x00000003,
+0xFFF99FE0, 0x03C5FDFF, 0xB0000000, 0x00030003,
+0xFFF987E0, 0x036DFDFF, 0x5E000000, 0x001C0000,
+0xFFFBAFE0, 0x23EDFDFF, 0x00000000, 0x00000001,
+0xFFF99FE0, 0x23CDFDFF, 0xB0000000, 0x00000003,
+0xD63DC7E0, 0x03BFC718, 0x00000000, 0x00000000,
+0xFFFDDFE0, 0x03EFFDFF, 0x00000000, 0x00000003,
+0xFFFDDFE0, 0x03EFFDFF, 0x40000000, 0x00000003,
+0xFFFDDFE0, 0x03FFFDFF, 0x00000000, 0x00000003,
+0x00000000, 0x00000000, 0x00000000, 0x00000000,
+0xFFFFFFFE, 0x000D7FFF, 0x0000003F, 0x00000000,
+0xFEF02596, 0x200D6CAE, 0x0000001F, 0x00000000,
+0x00000000, 0x00000000, 0xFFFFFEFF, 0x000003FF,
+0x00000000, 0x00000000, 0x00000000, 0x00000000,
+0x00000000, 0x00000000, 0x00000000, 0x00000000,
+0x00000000, 0xFFFFFFFF, 0xFFFF003F, 0x007FFFFF,
+0x0007DAED, 0x50000000, 0x82315001, 0x002C62AB,
+0x40000000, 0xF580C900, 0x00000007, 0x02010800,
+0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF,
+0x0FFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0x03FFFFFF,
+0x3F3FFFFF, 0xFFFFFFFF, 0xAAFF3F3F, 0x3FFFFFFF,
+0xFFFFFFFF, 0x5FDFFFFF, 0x0FCF1FDC, 0x1FDC1FFF,
+0x00000000, 0x00004C40, 0x00000000, 0x00000000,
+0x00000007, 0x00000000, 0x00000000, 0x00000000,
+0x00000080, 0x000003FE, 0xFFFFFFFE, 0xFFFFFFFF,
+0x001FFFFF, 0xFFFFFFFE, 0xFFFFFFFF, 0x07FFFFFF,
+0xFFFFFFE0, 0x00001FFF, 0x00000000, 0x00000000,
+0x00000000, 0x00000000, 0x00000000, 0x00000000,
+0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF,
+0xFFFFFFFF, 0x0000003F, 0x00000000, 0x00000000,
+0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF,
+0xFFFFFFFF, 0x0000000F, 0x00000000, 0x00000000,
+0x00000000, 0x07FF6000, 0x87FFFFFE, 0x07FFFFFE,
+0x00000000, 0x00800000, 0xFF7FFFFF, 0xFF7FFFFF,
+0x00FFFFFF, 0x00000000, 0xFFFF0000, 0xFFFFFFFF,
+0xFFFFFFFF, 0xF80001FF, 0x00030003, 0x00000000,
+0xFFFFFFFF, 0xFFFFFFFF, 0x0000003F, 0x00000003,
+0xFFFFD7C0, 0xFFFFFFFB, 0x547F7FFF, 0x000FFFFD,
+0xFFFFDFFE, 0xFFFFFFFF, 0xDFFEFFFF, 0xFFFFFFFF,
+0xFFFF007B, 0xFFFFFFFF, 0xFFFF199F, 0x033FCFFF,
+0x00000000, 0xFFFE0000, 0x027FFFFF, 0xFFFFFFFE,
+0xFFFE007F, 0xBBFFFFFB, 0xFFFF0016, 0x000707FF,
+0x00000000, 0x07FFFFFE, 0x0007FFFF, 0xFFFF03FF,
+0xFFFFFFFF, 0x7CFFFFFF, 0xFFEF7FFF, 0x03FF3DFF,
+0xFFFFFFEE, 0xF3FFFFFF, 0xFF1E3FFF, 0x0000FFCF,
+0xFFF99FEE, 0xD3C5FDFF, 0xB080399F, 0x0003FFCF,
+0xFFF987E4, 0xD36DFDFF, 0x5E003987, 0x001FFFC0,
+0xFFFBAFEE, 0xF3EDFDFF, 0x00003BBF, 0x0000FFC1,
+0xFFF99FEE, 0xF3CDFDFF, 0xB0C0398F, 0x0000FFC3,
+0xD63DC7EC, 0xC3BFC718, 0x00803DC7, 0x0000FF80,
+0xFFFDDFEE, 0xC3EFFDFF, 0x00603DDF, 0x0000FFC3,
+0xFFFDDFEC, 0xC3EFFDFF, 0x40603DDF, 0x0000FFC3,
+0xFFFDDFEC, 0xC3FFFDFF, 0x00803DCF, 0x0000FFC3,
+0x00000000, 0x00000000, 0x00000000, 0x00000000,
+0xFFFFFFFE, 0x07FF7FFF, 0x03FF7FFF, 0x00000000,
+0xFEF02596, 0x3BFF6CAE, 0x03FF3F5F, 0x00000000,
+0x03000000, 0xC2A003FF, 0xFFFFFEFF, 0xFFFE03FF,
+0xFEBF0FDF, 0x02FE3FFF, 0x00000000, 0x00000000,
+0x00000000, 0x00000000, 0x00000000, 0x00000000,
+0x00000000, 0x00000000, 0x1FFF0000, 0x00000002,
+0x000000A0, 0x003EFFFE, 0xFFFFFFFE, 0xFFFFFFFF,
+0x661FFFFF, 0xFFFFFFFE, 0xFFFFFFFF, 0x77FFFFFF,
+};
+static const unsigned char nmstrtPages[] = {
+0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x00,
+0x00, 0x09, 0x0A, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
+0x10, 0x11, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x12, 0x13,
+0x00, 0x14, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x15, 0x16, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x17,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x18,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+};
+static const unsigned char namePages[] = {
+0x19, 0x03, 0x1A, 0x1B, 0x1C, 0x1D, 0x1E, 0x00,
+0x00, 0x1F, 0x20, 0x21, 0x22, 0x23, 0x24, 0x25,
+0x10, 0x11, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x12, 0x13,
+0x26, 0x14, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x27, 0x16, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x17,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x18,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+};
diff --git a/src/lib/expat/utf8tab.h b/src/lib/expat/utf8tab.h
new file mode 100644
index 00000000000..7bb3e77603f
--- /dev/null
+++ b/src/lib/expat/utf8tab.h
@@ -0,0 +1,37 @@
+/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+ See the file COPYING for copying permission.
+*/
+
+
+/* 0x80 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0x84 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0x88 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0x8C */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0x90 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0x94 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0x98 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0x9C */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0xA0 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0xA4 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0xA8 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0xAC */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0xB0 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0xB4 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0xB8 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0xBC */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0xC0 */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2,
+/* 0xC4 */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2,
+/* 0xC8 */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2,
+/* 0xCC */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2,
+/* 0xD0 */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2,
+/* 0xD4 */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2,
+/* 0xD8 */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2,
+/* 0xDC */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2,
+/* 0xE0 */ BT_LEAD3, BT_LEAD3, BT_LEAD3, BT_LEAD3,
+/* 0xE4 */ BT_LEAD3, BT_LEAD3, BT_LEAD3, BT_LEAD3,
+/* 0xE8 */ BT_LEAD3, BT_LEAD3, BT_LEAD3, BT_LEAD3,
+/* 0xEC */ BT_LEAD3, BT_LEAD3, BT_LEAD3, BT_LEAD3,
+/* 0xF0 */ BT_LEAD4, BT_LEAD4, BT_LEAD4, BT_LEAD4,
+/* 0xF4 */ BT_LEAD4, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0xF8 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0xFC */ BT_NONXML, BT_NONXML, BT_MALFORM, BT_MALFORM,
diff --git a/src/lib/expat/xmlparse.c b/src/lib/expat/xmlparse.c
new file mode 100644
index 00000000000..ed9344ca8ff
--- /dev/null
+++ b/src/lib/expat/xmlparse.c
@@ -0,0 +1,6258 @@
+/* Copyright (c) 1998, 1999, 2000 Thai Open Source Software Center Ltd
+ See the file COPYING for copying permission.
+*/
+
+#include <stddef.h>
+#include <string.h> /* memset(), memcpy() */
+#include <assert.h>
+
+#define XML_BUILDING_EXPAT 1
+
+#include "intconfig.h"
+
+#include "expat.h"
+
+#ifdef XML_UNICODE
+#define XML_ENCODE_MAX XML_UTF16_ENCODE_MAX
+#define XmlConvert XmlUtf16Convert
+#define XmlGetInternalEncoding XmlGetUtf16InternalEncoding
+#define XmlGetInternalEncodingNS XmlGetUtf16InternalEncodingNS
+#define XmlEncode XmlUtf16Encode
+#define MUST_CONVERT(enc, s) (!(enc)->isUtf16 || (((unsigned long)s) & 1))
+typedef unsigned short ICHAR;
+#else
+#define XML_ENCODE_MAX XML_UTF8_ENCODE_MAX
+#define XmlConvert XmlUtf8Convert
+#define XmlGetInternalEncoding XmlGetUtf8InternalEncoding
+#define XmlGetInternalEncodingNS XmlGetUtf8InternalEncodingNS
+#define XmlEncode XmlUtf8Encode
+#define MUST_CONVERT(enc, s) (!(enc)->isUtf8)
+typedef char ICHAR;
+#endif
+
+
+#ifndef XML_NS
+
+#define XmlInitEncodingNS XmlInitEncoding
+#define XmlInitUnknownEncodingNS XmlInitUnknownEncoding
+#undef XmlGetInternalEncodingNS
+#define XmlGetInternalEncodingNS XmlGetInternalEncoding
+#define XmlParseXmlDeclNS XmlParseXmlDecl
+
+#endif
+
+#ifdef XML_UNICODE
+
+#ifdef XML_UNICODE_WCHAR_T
+#define XML_T(x) (const wchar_t)x
+#define XML_L(x) L ## x
+#else
+#define XML_T(x) (const unsigned short)x
+#define XML_L(x) x
+#endif
+
+#else
+
+#define XML_T(x) x
+#define XML_L(x) x
+
+#endif
+
+/* Round up n to be a multiple of sz, where sz is a power of 2. */
+#define ROUND_UP(n, sz) (((n) + ((sz) - 1)) & ~((sz) - 1))
+
+/* Handle the case where memmove() doesn't exist. */
+#ifndef HAVE_MEMMOVE
+#ifdef HAVE_BCOPY
+#define memmove(d,s,l) bcopy((s),(d),(l))
+#else
+#error memmove does not exist on this platform, nor is a substitute available
+#endif /* HAVE_BCOPY */
+#endif /* HAVE_MEMMOVE */
+
+#include "internal.h"
+#include "xmltok.h"
+#include "xmlrole.h"
+
+typedef const XML_Char *KEY;
+
+typedef struct {
+ KEY name;
+} NAMED;
+
+typedef struct {
+ NAMED **v;
+ unsigned char power;
+ size_t size;
+ size_t used;
+ const XML_Memory_Handling_Suite *mem;
+} HASH_TABLE;
+
+/* Basic character hash algorithm, taken from Python's string hash:
+ h = h * 1000003 ^ character, the constant being a prime number.
+
+*/
+#ifdef XML_UNICODE
+#define CHAR_HASH(h, c) \
+ (((h) * 0xF4243) ^ (unsigned short)(c))
+#else
+#define CHAR_HASH(h, c) \
+ (((h) * 0xF4243) ^ (unsigned char)(c))
+#endif
+
+/* For probing (after a collision) we need a step size relative prime
+ to the hash table size, which is a power of 2. We use double-hashing,
+ since we can calculate a second hash value cheaply by taking those bits
+ of the first hash value that were discarded (masked out) when the table
+ index was calculated: index = hash & mask, where mask = table->size - 1.
+ We limit the maximum step size to table->size / 4 (mask >> 2) and make
+ it odd, since odd numbers are always relative prime to a power of 2.
+*/
+#define SECOND_HASH(hash, mask, power) \
+ ((((hash) & ~(mask)) >> ((power) - 1)) & ((mask) >> 2))
+#define PROBE_STEP(hash, mask, power) \
+ ((unsigned char)((SECOND_HASH(hash, mask, power)) | 1))
+
+typedef struct {
+ NAMED **p;
+ NAMED **end;
+} HASH_TABLE_ITER;
+
+#define INIT_TAG_BUF_SIZE 32 /* must be a multiple of sizeof(XML_Char) */
+#define INIT_DATA_BUF_SIZE 1024
+#define INIT_ATTS_SIZE 16
+#define INIT_ATTS_VERSION 0xFFFFFFFF
+#define INIT_BLOCK_SIZE 1024
+#define INIT_BUFFER_SIZE 1024
+
+#define EXPAND_SPARE 24
+
+typedef struct binding {
+ struct prefix *prefix;
+ struct binding *nextTagBinding;
+ struct binding *prevPrefixBinding;
+ const struct attribute_id *attId;
+ XML_Char *uri;
+ int uriLen;
+ int uriAlloc;
+} BINDING;
+
+typedef struct prefix {
+ const XML_Char *name;
+ BINDING *binding;
+} PREFIX;
+
+typedef struct {
+ const XML_Char *str;
+ const XML_Char *localPart;
+ const XML_Char *prefix;
+ int strLen;
+ int uriLen;
+ int prefixLen;
+} TAG_NAME;
+
+/* TAG represents an open element.
+ The name of the element is stored in both the document and API
+ encodings. The memory buffer 'buf' is a separately-allocated
+ memory area which stores the name. During the XML_Parse()/
+ XMLParseBuffer() when the element is open, the memory for the 'raw'
+ version of the name (in the document encoding) is shared with the
+ document buffer. If the element is open across calls to
+ XML_Parse()/XML_ParseBuffer(), the buffer is re-allocated to
+ contain the 'raw' name as well.
+
+ A parser re-uses these structures, maintaining a list of allocated
+ TAG objects in a free list.
+*/
+typedef struct tag {
+ struct tag *parent; /* parent of this element */
+ const char *rawName; /* tagName in the original encoding */
+ int rawNameLength;
+ TAG_NAME name; /* tagName in the API encoding */
+ char *buf; /* buffer for name components */
+ char *bufEnd; /* end of the buffer */
+ BINDING *bindings;
+} TAG;
+
+typedef struct {
+ const XML_Char *name;
+ const XML_Char *textPtr;
+ int textLen; /* length in XML_Chars */
+ int processed; /* # of processed bytes - when suspended */
+ const XML_Char *systemId;
+ const XML_Char *base;
+ const XML_Char *publicId;
+ const XML_Char *notation;
+ XML_Bool open;
+ XML_Bool is_param;
+ XML_Bool is_internal; /* true if declared in internal subset outside PE */
+} ENTITY;
+
+typedef struct {
+ enum XML_Content_Type type;
+ enum XML_Content_Quant quant;
+ const XML_Char * name;
+ int firstchild;
+ int lastchild;
+ int childcnt;
+ int nextsib;
+} CONTENT_SCAFFOLD;
+
+#define INIT_SCAFFOLD_ELEMENTS 32
+
+typedef struct block {
+ struct block *next;
+ int size;
+ XML_Char s[1];
+} BLOCK;
+
+typedef struct {
+ BLOCK *blocks;
+ BLOCK *freeBlocks;
+ const XML_Char *end;
+ XML_Char *ptr;
+ XML_Char *start;
+ const XML_Memory_Handling_Suite *mem;
+} STRING_POOL;
+
+/* The XML_Char before the name is used to determine whether
+ an attribute has been specified. */
+typedef struct attribute_id {
+ XML_Char *name;
+ PREFIX *prefix;
+ XML_Bool maybeTokenized;
+ XML_Bool xmlns;
+} ATTRIBUTE_ID;
+
+typedef struct {
+ const ATTRIBUTE_ID *id;
+ XML_Bool isCdata;
+ const XML_Char *value;
+} DEFAULT_ATTRIBUTE;
+
+typedef struct {
+ unsigned long version;
+ unsigned long hash;
+ const XML_Char *uriName;
+} NS_ATT;
+
+typedef struct {
+ const XML_Char *name;
+ PREFIX *prefix;
+ const ATTRIBUTE_ID *idAtt;
+ int nDefaultAtts;
+ int allocDefaultAtts;
+ DEFAULT_ATTRIBUTE *defaultAtts;
+} ELEMENT_TYPE;
+
+typedef struct {
+ HASH_TABLE generalEntities;
+ HASH_TABLE elementTypes;
+ HASH_TABLE attributeIds;
+ HASH_TABLE prefixes;
+ STRING_POOL pool;
+ STRING_POOL entityValuePool;
+ /* false once a parameter entity reference has been skipped */
+ XML_Bool keepProcessing;
+ /* true once an internal or external PE reference has been encountered;
+ this includes the reference to an external subset */
+ XML_Bool hasParamEntityRefs;
+ XML_Bool standalone;
+#ifdef XML_DTD
+ /* indicates if external PE has been read */
+ XML_Bool paramEntityRead;
+ HASH_TABLE paramEntities;
+#endif /* XML_DTD */
+ PREFIX defaultPrefix;
+ /* === scaffolding for building content model === */
+ XML_Bool in_eldecl;
+ CONTENT_SCAFFOLD *scaffold;
+ unsigned contentStringLen;
+ unsigned scaffSize;
+ unsigned scaffCount;
+ int scaffLevel;
+ int *scaffIndex;
+} DTD;
+
+typedef struct open_internal_entity {
+ const char *internalEventPtr;
+ const char *internalEventEndPtr;
+ struct open_internal_entity *next;
+ ENTITY *entity;
+ int startTagLevel;
+ XML_Bool betweenDecl; /* WFC: PE Between Declarations */
+} OPEN_INTERNAL_ENTITY;
+
+typedef enum XML_Error PTRCALL Processor(XML_Parser parser,
+ const char *start,
+ const char *end,
+ const char **endPtr);
+
+static Processor prologProcessor;
+static Processor prologInitProcessor;
+static Processor contentProcessor;
+static Processor cdataSectionProcessor;
+#ifdef XML_DTD
+static Processor ignoreSectionProcessor;
+static Processor externalParEntProcessor;
+static Processor externalParEntInitProcessor;
+static Processor entityValueProcessor;
+static Processor entityValueInitProcessor;
+#endif /* XML_DTD */
+static Processor epilogProcessor;
+static Processor errorProcessor;
+static Processor externalEntityInitProcessor;
+static Processor externalEntityInitProcessor2;
+static Processor externalEntityInitProcessor3;
+static Processor externalEntityContentProcessor;
+static Processor internalEntityProcessor;
+
+static enum XML_Error
+handleUnknownEncoding(XML_Parser parser, const XML_Char *encodingName);
+static enum XML_Error
+processXmlDecl(XML_Parser parser, int isGeneralTextEntity,
+ const char *s, const char *next);
+static enum XML_Error
+initializeEncoding(XML_Parser parser);
+static enum XML_Error
+doProlog(XML_Parser parser, const ENCODING *enc, const char *s,
+ const char *end, int tok, const char *next, const char **nextPtr,
+ XML_Bool haveMore);
+static enum XML_Error
+processInternalEntity(XML_Parser parser, ENTITY *entity,
+ XML_Bool betweenDecl);
+static enum XML_Error
+doContent(XML_Parser parser, int startTagLevel, const ENCODING *enc,
+ const char *start, const char *end, const char **endPtr,
+ XML_Bool haveMore);
+static enum XML_Error
+doCdataSection(XML_Parser parser, const ENCODING *, const char **startPtr,
+ const char *end, const char **nextPtr, XML_Bool haveMore);
+#ifdef XML_DTD
+static enum XML_Error
+doIgnoreSection(XML_Parser parser, const ENCODING *, const char **startPtr,
+ const char *end, const char **nextPtr, XML_Bool haveMore);
+#endif /* XML_DTD */
+
+static enum XML_Error
+storeAtts(XML_Parser parser, const ENCODING *, const char *s,
+ TAG_NAME *tagNamePtr, BINDING **bindingsPtr);
+static enum XML_Error
+addBinding(XML_Parser parser, PREFIX *prefix, const ATTRIBUTE_ID *attId,
+ const XML_Char *uri, BINDING **bindingsPtr);
+static int
+defineAttribute(ELEMENT_TYPE *type, ATTRIBUTE_ID *, XML_Bool isCdata,
+ XML_Bool isId, const XML_Char *dfltValue, XML_Parser parser);
+static enum XML_Error
+storeAttributeValue(XML_Parser parser, const ENCODING *, XML_Bool isCdata,
+ const char *, const char *, STRING_POOL *);
+static enum XML_Error
+appendAttributeValue(XML_Parser parser, const ENCODING *, XML_Bool isCdata,
+ const char *, const char *, STRING_POOL *);
+static ATTRIBUTE_ID *
+getAttributeId(XML_Parser parser, const ENCODING *enc, const char *start,
+ const char *end);
+static int
+setElementTypePrefix(XML_Parser parser, ELEMENT_TYPE *);
+static enum XML_Error
+storeEntityValue(XML_Parser parser, const ENCODING *enc, const char *start,
+ const char *end);
+static int
+reportProcessingInstruction(XML_Parser parser, const ENCODING *enc,
+ const char *start, const char *end);
+static int
+reportComment(XML_Parser parser, const ENCODING *enc, const char *start,
+ const char *end);
+static void
+reportDefault(XML_Parser parser, const ENCODING *enc, const char *start,
+ const char *end);
+
+static const XML_Char * getContext(XML_Parser parser);
+static XML_Bool
+setContext(XML_Parser parser, const XML_Char *context);
+
+static void FASTCALL normalizePublicId(XML_Char *s);
+
+static DTD * dtdCreate(const XML_Memory_Handling_Suite *ms);
+/* do not call if parentParser != NULL */
+static void dtdReset(DTD *p, const XML_Memory_Handling_Suite *ms);
+static void
+dtdDestroy(DTD *p, XML_Bool isDocEntity, const XML_Memory_Handling_Suite *ms);
+static int
+dtdCopy(DTD *newDtd, const DTD *oldDtd, const XML_Memory_Handling_Suite *ms);
+static int
+copyEntityTable(HASH_TABLE *, STRING_POOL *, const HASH_TABLE *);
+
+static NAMED *
+lookup(HASH_TABLE *table, KEY name, size_t createSize);
+static void FASTCALL
+hashTableInit(HASH_TABLE *, const XML_Memory_Handling_Suite *ms);
+static void FASTCALL hashTableClear(HASH_TABLE *);
+static void FASTCALL hashTableDestroy(HASH_TABLE *);
+static void FASTCALL
+hashTableIterInit(HASH_TABLE_ITER *, const HASH_TABLE *);
+static NAMED * FASTCALL hashTableIterNext(HASH_TABLE_ITER *);
+
+static void FASTCALL
+poolInit(STRING_POOL *, const XML_Memory_Handling_Suite *ms);
+static void FASTCALL poolClear(STRING_POOL *);
+static void FASTCALL poolDestroy(STRING_POOL *);
+static XML_Char *
+poolAppend(STRING_POOL *pool, const ENCODING *enc,
+ const char *ptr, const char *end);
+static XML_Char *
+poolStoreString(STRING_POOL *pool, const ENCODING *enc,
+ const char *ptr, const char *end);
+static XML_Bool FASTCALL poolGrow(STRING_POOL *pool);
+static const XML_Char * FASTCALL
+poolCopyString(STRING_POOL *pool, const XML_Char *s);
+static const XML_Char *
+poolCopyStringN(STRING_POOL *pool, const XML_Char *s, int n);
+static const XML_Char * FASTCALL
+poolAppendString(STRING_POOL *pool, const XML_Char *s);
+
+static int FASTCALL nextScaffoldPart(XML_Parser parser);
+static XML_Content * build_model(XML_Parser parser);
+static ELEMENT_TYPE *
+getElementType(XML_Parser parser, const ENCODING *enc,
+ const char *ptr, const char *end);
+
+static XML_Parser
+parserCreate(const XML_Char *encodingName,
+ const XML_Memory_Handling_Suite *memsuite,
+ const XML_Char *nameSep,
+ DTD *dtd);
+static void
+parserInit(XML_Parser parser, const XML_Char *encodingName);
+
+#define poolStart(pool) ((pool)->start)
+#define poolEnd(pool) ((pool)->ptr)
+#define poolLength(pool) ((pool)->ptr - (pool)->start)
+#define poolChop(pool) ((void)--(pool->ptr))
+#define poolLastChar(pool) (((pool)->ptr)[-1])
+#define poolDiscard(pool) ((pool)->ptr = (pool)->start)
+#define poolFinish(pool) ((pool)->start = (pool)->ptr)
+#define poolAppendChar(pool, c) \
+ (((pool)->ptr == (pool)->end && !poolGrow(pool)) \
+ ? 0 \
+ : ((*((pool)->ptr)++ = c), 1))
+
+struct XML_ParserStruct {
+ /* The first member must be userData so that the XML_GetUserData
+ macro works. */
+ void *m_userData;
+ void *m_handlerArg;
+ char *m_buffer;
+ const XML_Memory_Handling_Suite m_mem;
+ /* first character to be parsed */
+ const char *m_bufferPtr;
+ /* past last character to be parsed */
+ char *m_bufferEnd;
+ /* allocated end of buffer */
+ const char *m_bufferLim;
+ XML_Index m_parseEndByteIndex;
+ const char *m_parseEndPtr;
+ XML_Char *m_dataBuf;
+ XML_Char *m_dataBufEnd;
+ XML_StartElementHandler m_startElementHandler;
+ XML_EndElementHandler m_endElementHandler;
+ XML_CharacterDataHandler m_characterDataHandler;
+ XML_ProcessingInstructionHandler m_processingInstructionHandler;
+ XML_CommentHandler m_commentHandler;
+ XML_StartCdataSectionHandler m_startCdataSectionHandler;
+ XML_EndCdataSectionHandler m_endCdataSectionHandler;
+ XML_DefaultHandler m_defaultHandler;
+ XML_StartDoctypeDeclHandler m_startDoctypeDeclHandler;
+ XML_EndDoctypeDeclHandler m_endDoctypeDeclHandler;
+ XML_UnparsedEntityDeclHandler m_unparsedEntityDeclHandler;
+ XML_NotationDeclHandler m_notationDeclHandler;
+ XML_StartNamespaceDeclHandler m_startNamespaceDeclHandler;
+ XML_EndNamespaceDeclHandler m_endNamespaceDeclHandler;
+ XML_NotStandaloneHandler m_notStandaloneHandler;
+ XML_ExternalEntityRefHandler m_externalEntityRefHandler;
+ XML_Parser m_externalEntityRefHandlerArg;
+ XML_SkippedEntityHandler m_skippedEntityHandler;
+ XML_UnknownEncodingHandler m_unknownEncodingHandler;
+ XML_ElementDeclHandler m_elementDeclHandler;
+ XML_AttlistDeclHandler m_attlistDeclHandler;
+ XML_EntityDeclHandler m_entityDeclHandler;
+ XML_XmlDeclHandler m_xmlDeclHandler;
+ const ENCODING *m_encoding;
+ INIT_ENCODING m_initEncoding;
+ const ENCODING *m_internalEncoding;
+ const XML_Char *m_protocolEncodingName;
+ XML_Bool m_ns;
+ XML_Bool m_ns_triplets;
+ void *m_unknownEncodingMem;
+ void *m_unknownEncodingData;
+ void *m_unknownEncodingHandlerData;
+ void (XMLCALL *m_unknownEncodingRelease)(void *);
+ PROLOG_STATE m_prologState;
+ Processor *m_processor;
+ enum XML_Error m_errorCode;
+ const char *m_eventPtr;
+ const char *m_eventEndPtr;
+ const char *m_positionPtr;
+ OPEN_INTERNAL_ENTITY *m_openInternalEntities;
+ OPEN_INTERNAL_ENTITY *m_freeInternalEntities;
+ XML_Bool m_defaultExpandInternalEntities;
+ int m_tagLevel;
+ ENTITY *m_declEntity;
+ const XML_Char *m_doctypeName;
+ const XML_Char *m_doctypeSysid;
+ const XML_Char *m_doctypePubid;
+ const XML_Char *m_declAttributeType;
+ const XML_Char *m_declNotationName;
+ const XML_Char *m_declNotationPublicId;
+ ELEMENT_TYPE *m_declElementType;
+ ATTRIBUTE_ID *m_declAttributeId;
+ XML_Bool m_declAttributeIsCdata;
+ XML_Bool m_declAttributeIsId;
+ DTD *m_dtd;
+ const XML_Char *m_curBase;
+ TAG *m_tagStack;
+ TAG *m_freeTagList;
+ BINDING *m_inheritedBindings;
+ BINDING *m_freeBindingList;
+ int m_attsSize;
+ int m_nSpecifiedAtts;
+ int m_idAttIndex;
+ ATTRIBUTE *m_atts;
+ NS_ATT *m_nsAtts;
+ unsigned long m_nsAttsVersion;
+ unsigned char m_nsAttsPower;
+ POSITION m_position;
+ STRING_POOL m_tempPool;
+ STRING_POOL m_temp2Pool;
+ char *m_groupConnector;
+ unsigned int m_groupSize;
+ XML_Char m_namespaceSeparator;
+ XML_Parser m_parentParser;
+ XML_ParsingStatus m_parsingStatus;
+#ifdef XML_DTD
+ XML_Bool m_isParamEntity;
+ XML_Bool m_useForeignDTD;
+ enum XML_ParamEntityParsing m_paramEntityParsing;
+#endif
+};
+
+#define MALLOC(s) (parser->m_mem.malloc_fcn((s)))
+#define REALLOC(p,s) (parser->m_mem.realloc_fcn((p),(s)))
+#define FREE(p) (parser->m_mem.free_fcn((p)))
+
+#define userData (parser->m_userData)
+#define handlerArg (parser->m_handlerArg)
+#define startElementHandler (parser->m_startElementHandler)
+#define endElementHandler (parser->m_endElementHandler)
+#define characterDataHandler (parser->m_characterDataHandler)
+#define processingInstructionHandler \
+ (parser->m_processingInstructionHandler)
+#define commentHandler (parser->m_commentHandler)
+#define startCdataSectionHandler \
+ (parser->m_startCdataSectionHandler)
+#define endCdataSectionHandler (parser->m_endCdataSectionHandler)
+#define defaultHandler (parser->m_defaultHandler)
+#define startDoctypeDeclHandler (parser->m_startDoctypeDeclHandler)
+#define endDoctypeDeclHandler (parser->m_endDoctypeDeclHandler)
+#define unparsedEntityDeclHandler \
+ (parser->m_unparsedEntityDeclHandler)
+#define notationDeclHandler (parser->m_notationDeclHandler)
+#define startNamespaceDeclHandler \
+ (parser->m_startNamespaceDeclHandler)
+#define endNamespaceDeclHandler (parser->m_endNamespaceDeclHandler)
+#define notStandaloneHandler (parser->m_notStandaloneHandler)
+#define externalEntityRefHandler \
+ (parser->m_externalEntityRefHandler)
+#define externalEntityRefHandlerArg \
+ (parser->m_externalEntityRefHandlerArg)
+#define internalEntityRefHandler \
+ (parser->m_internalEntityRefHandler)
+#define skippedEntityHandler (parser->m_skippedEntityHandler)
+#define unknownEncodingHandler (parser->m_unknownEncodingHandler)
+#define elementDeclHandler (parser->m_elementDeclHandler)
+#define attlistDeclHandler (parser->m_attlistDeclHandler)
+#define entityDeclHandler (parser->m_entityDeclHandler)
+#define xmlDeclHandler (parser->m_xmlDeclHandler)
+#define encoding (parser->m_encoding)
+#define initEncoding (parser->m_initEncoding)
+#define internalEncoding (parser->m_internalEncoding)
+#define unknownEncodingMem (parser->m_unknownEncodingMem)
+#define unknownEncodingData (parser->m_unknownEncodingData)
+#define unknownEncodingHandlerData \
+ (parser->m_unknownEncodingHandlerData)
+#define unknownEncodingRelease (parser->m_unknownEncodingRelease)
+#define protocolEncodingName (parser->m_protocolEncodingName)
+#define ns (parser->m_ns)
+#define ns_triplets (parser->m_ns_triplets)
+#define prologState (parser->m_prologState)
+#define processor (parser->m_processor)
+#define errorCode (parser->m_errorCode)
+#define eventPtr (parser->m_eventPtr)
+#define eventEndPtr (parser->m_eventEndPtr)
+#define positionPtr (parser->m_positionPtr)
+#define position (parser->m_position)
+#define openInternalEntities (parser->m_openInternalEntities)
+#define freeInternalEntities (parser->m_freeInternalEntities)
+#define defaultExpandInternalEntities \
+ (parser->m_defaultExpandInternalEntities)
+#define tagLevel (parser->m_tagLevel)
+#define buffer (parser->m_buffer)
+#define bufferPtr (parser->m_bufferPtr)
+#define bufferEnd (parser->m_bufferEnd)
+#define parseEndByteIndex (parser->m_parseEndByteIndex)
+#define parseEndPtr (parser->m_parseEndPtr)
+#define bufferLim (parser->m_bufferLim)
+#define dataBuf (parser->m_dataBuf)
+#define dataBufEnd (parser->m_dataBufEnd)
+#define _dtd (parser->m_dtd)
+#define curBase (parser->m_curBase)
+#define declEntity (parser->m_declEntity)
+#define doctypeName (parser->m_doctypeName)
+#define doctypeSysid (parser->m_doctypeSysid)
+#define doctypePubid (parser->m_doctypePubid)
+#define declAttributeType (parser->m_declAttributeType)
+#define declNotationName (parser->m_declNotationName)
+#define declNotationPublicId (parser->m_declNotationPublicId)
+#define declElementType (parser->m_declElementType)
+#define declAttributeId (parser->m_declAttributeId)
+#define declAttributeIsCdata (parser->m_declAttributeIsCdata)
+#define declAttributeIsId (parser->m_declAttributeIsId)
+#define freeTagList (parser->m_freeTagList)
+#define freeBindingList (parser->m_freeBindingList)
+#define inheritedBindings (parser->m_inheritedBindings)
+#define tagStack (parser->m_tagStack)
+#define atts (parser->m_atts)
+#define attsSize (parser->m_attsSize)
+#define nSpecifiedAtts (parser->m_nSpecifiedAtts)
+#define idAttIndex (parser->m_idAttIndex)
+#define nsAtts (parser->m_nsAtts)
+#define nsAttsVersion (parser->m_nsAttsVersion)
+#define nsAttsPower (parser->m_nsAttsPower)
+#define tempPool (parser->m_tempPool)
+#define temp2Pool (parser->m_temp2Pool)
+#define groupConnector (parser->m_groupConnector)
+#define groupSize (parser->m_groupSize)
+#define namespaceSeparator (parser->m_namespaceSeparator)
+#define parentParser (parser->m_parentParser)
+#define ps_parsing (parser->m_parsingStatus.parsing)
+#define ps_finalBuffer (parser->m_parsingStatus.finalBuffer)
+#ifdef XML_DTD
+#define isParamEntity (parser->m_isParamEntity)
+#define useForeignDTD (parser->m_useForeignDTD)
+#define paramEntityParsing (parser->m_paramEntityParsing)
+#endif /* XML_DTD */
+
+XML_Parser XMLCALL
+XML_ParserCreate(const XML_Char *encodingName)
+{
+ return XML_ParserCreate_MM(encodingName, NULL, NULL);
+}
+
+XML_Parser XMLCALL
+XML_ParserCreateNS(const XML_Char *encodingName, XML_Char nsSep)
+{
+ XML_Char tmp[2];
+ *tmp = nsSep;
+ return XML_ParserCreate_MM(encodingName, NULL, tmp);
+}
+
+static const XML_Char implicitContext[] = {
+ 'x', 'm', 'l', '=', 'h', 't', 't', 'p', ':', '/', '/',
+ 'w', 'w', 'w', '.', 'w', '3', '.', 'o', 'r', 'g', '/',
+ 'X', 'M', 'L', '/', '1', '9', '9', '8', '/',
+ 'n', 'a', 'm', 'e', 's', 'p', 'a', 'c', 'e', '\0'
+};
+
+XML_Parser XMLCALL
+XML_ParserCreate_MM(const XML_Char *encodingName,
+ const XML_Memory_Handling_Suite *memsuite,
+ const XML_Char *nameSep)
+{
+ XML_Parser parser = parserCreate(encodingName, memsuite, nameSep, NULL);
+ if (parser != NULL && ns) {
+ /* implicit context only set for root parser, since child
+ parsers (i.e. external entity parsers) will inherit it
+ */
+ if (!setContext(parser, implicitContext)) {
+ XML_ParserFree(parser);
+ return NULL;
+ }
+ }
+ return parser;
+}
+
+static XML_Parser
+parserCreate(const XML_Char *encodingName,
+ const XML_Memory_Handling_Suite *memsuite,
+ const XML_Char *nameSep,
+ DTD *dtd)
+{
+ XML_Parser parser;
+
+ if (memsuite) {
+ XML_Memory_Handling_Suite *mtemp;
+ parser = (XML_Parser)
+ memsuite->malloc_fcn(sizeof(struct XML_ParserStruct));
+ if (parser != NULL) {
+ mtemp = (XML_Memory_Handling_Suite *)&(parser->m_mem);
+ mtemp->malloc_fcn = memsuite->malloc_fcn;
+ mtemp->realloc_fcn = memsuite->realloc_fcn;
+ mtemp->free_fcn = memsuite->free_fcn;
+ }
+ }
+ else {
+ XML_Memory_Handling_Suite *mtemp;
+ parser = (XML_Parser)malloc(sizeof(struct XML_ParserStruct));
+ if (parser != NULL) {
+ mtemp = (XML_Memory_Handling_Suite *)&(parser->m_mem);
+ mtemp->malloc_fcn = malloc;
+ mtemp->realloc_fcn = realloc;
+ mtemp->free_fcn = free;
+ }
+ }
+
+ if (!parser)
+ return parser;
+
+ buffer = NULL;
+ bufferLim = NULL;
+
+ attsSize = INIT_ATTS_SIZE;
+ atts = (ATTRIBUTE *)MALLOC(attsSize * sizeof(ATTRIBUTE));
+ if (atts == NULL) {
+ FREE(parser);
+ return NULL;
+ }
+ dataBuf = (XML_Char *)MALLOC(INIT_DATA_BUF_SIZE * sizeof(XML_Char));
+ if (dataBuf == NULL) {
+ FREE(atts);
+ FREE(parser);
+ return NULL;
+ }
+ dataBufEnd = dataBuf + INIT_DATA_BUF_SIZE;
+
+ if (dtd)
+ _dtd = dtd;
+ else {
+ _dtd = dtdCreate(&parser->m_mem);
+ if (_dtd == NULL) {
+ FREE(dataBuf);
+ FREE(atts);
+ FREE(parser);
+ return NULL;
+ }
+ }
+
+ freeBindingList = NULL;
+ freeTagList = NULL;
+ freeInternalEntities = NULL;
+
+ groupSize = 0;
+ groupConnector = NULL;
+
+ unknownEncodingHandler = NULL;
+ unknownEncodingHandlerData = NULL;
+
+ namespaceSeparator = '!';
+ ns = XML_FALSE;
+ ns_triplets = XML_FALSE;
+
+ nsAtts = NULL;
+ nsAttsVersion = 0;
+ nsAttsPower = 0;
+
+ poolInit(&tempPool, &(parser->m_mem));
+ poolInit(&temp2Pool, &(parser->m_mem));
+ parserInit(parser, encodingName);
+
+ if (encodingName && !protocolEncodingName) {
+ XML_ParserFree(parser);
+ return NULL;
+ }
+
+ if (nameSep) {
+ ns = XML_TRUE;
+ internalEncoding = XmlGetInternalEncodingNS();
+ namespaceSeparator = *nameSep;
+ }
+ else {
+ internalEncoding = XmlGetInternalEncoding();
+ }
+
+ return parser;
+}
+
+static void
+parserInit(XML_Parser parser, const XML_Char *encodingName)
+{
+ processor = prologInitProcessor;
+ XmlPrologStateInit(&prologState);
+ protocolEncodingName = (encodingName != NULL
+ ? poolCopyString(&tempPool, encodingName)
+ : NULL);
+ curBase = NULL;
+ XmlInitEncoding(&initEncoding, &encoding, 0);
+ userData = NULL;
+ handlerArg = NULL;
+ startElementHandler = NULL;
+ endElementHandler = NULL;
+ characterDataHandler = NULL;
+ processingInstructionHandler = NULL;
+ commentHandler = NULL;
+ startCdataSectionHandler = NULL;
+ endCdataSectionHandler = NULL;
+ defaultHandler = NULL;
+ startDoctypeDeclHandler = NULL;
+ endDoctypeDeclHandler = NULL;
+ unparsedEntityDeclHandler = NULL;
+ notationDeclHandler = NULL;
+ startNamespaceDeclHandler = NULL;
+ endNamespaceDeclHandler = NULL;
+ notStandaloneHandler = NULL;
+ externalEntityRefHandler = NULL;
+ externalEntityRefHandlerArg = parser;
+ skippedEntityHandler = NULL;
+ elementDeclHandler = NULL;
+ attlistDeclHandler = NULL;
+ entityDeclHandler = NULL;
+ xmlDeclHandler = NULL;
+ bufferPtr = buffer;
+ bufferEnd = buffer;
+ parseEndByteIndex = 0;
+ parseEndPtr = NULL;
+ declElementType = NULL;
+ declAttributeId = NULL;
+ declEntity = NULL;
+ doctypeName = NULL;
+ doctypeSysid = NULL;
+ doctypePubid = NULL;
+ declAttributeType = NULL;
+ declNotationName = NULL;
+ declNotationPublicId = NULL;
+ declAttributeIsCdata = XML_FALSE;
+ declAttributeIsId = XML_FALSE;
+ memset(&position, 0, sizeof(POSITION));
+ errorCode = XML_ERROR_NONE;
+ eventPtr = NULL;
+ eventEndPtr = NULL;
+ positionPtr = NULL;
+ openInternalEntities = NULL;
+ defaultExpandInternalEntities = XML_TRUE;
+ tagLevel = 0;
+ tagStack = NULL;
+ inheritedBindings = NULL;
+ nSpecifiedAtts = 0;
+ unknownEncodingMem = NULL;
+ unknownEncodingRelease = NULL;
+ unknownEncodingData = NULL;
+ parentParser = NULL;
+ ps_parsing = XML_INITIALIZED;
+#ifdef XML_DTD
+ isParamEntity = XML_FALSE;
+ useForeignDTD = XML_FALSE;
+ paramEntityParsing = XML_PARAM_ENTITY_PARSING_NEVER;
+#endif
+}
+
+/* moves list of bindings to freeBindingList */
+static void FASTCALL
+moveToFreeBindingList(XML_Parser parser, BINDING *bindings)
+{
+ while (bindings) {
+ BINDING *b = bindings;
+ bindings = bindings->nextTagBinding;
+ b->nextTagBinding = freeBindingList;
+ freeBindingList = b;
+ }
+}
+
+XML_Bool XMLCALL
+XML_ParserReset(XML_Parser parser, const XML_Char *encodingName)
+{
+ TAG *tStk;
+ OPEN_INTERNAL_ENTITY *openEntityList;
+ if (parentParser)
+ return XML_FALSE;
+ /* move tagStack to freeTagList */
+ tStk = tagStack;
+ while (tStk) {
+ TAG *tag = tStk;
+ tStk = tStk->parent;
+ tag->parent = freeTagList;
+ moveToFreeBindingList(parser, tag->bindings);
+ tag->bindings = NULL;
+ freeTagList = tag;
+ }
+ /* move openInternalEntities to freeInternalEntities */
+ openEntityList = openInternalEntities;
+ while (openEntityList) {
+ OPEN_INTERNAL_ENTITY *openEntity = openEntityList;
+ openEntityList = openEntity->next;
+ openEntity->next = freeInternalEntities;
+ freeInternalEntities = openEntity;
+ }
+ moveToFreeBindingList(parser, inheritedBindings);
+ FREE(unknownEncodingMem);
+ if (unknownEncodingRelease)
+ unknownEncodingRelease(unknownEncodingData);
+ poolClear(&tempPool);
+ poolClear(&temp2Pool);
+ parserInit(parser, encodingName);
+ dtdReset(_dtd, &parser->m_mem);
+ return setContext(parser, implicitContext);
+}
+
+enum XML_Status XMLCALL
+XML_SetEncoding(XML_Parser parser, const XML_Char *encodingName)
+{
+ /* Block after XML_Parse()/XML_ParseBuffer() has been called.
+ XXX There's no way for the caller to determine which of the
+ XXX possible error cases caused the XML_STATUS_ERROR return.
+ */
+ if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED)
+ return XML_STATUS_ERROR;
+ if (encodingName == NULL)
+ protocolEncodingName = NULL;
+ else {
+ protocolEncodingName = poolCopyString(&tempPool, encodingName);
+ if (!protocolEncodingName)
+ return XML_STATUS_ERROR;
+ }
+ return XML_STATUS_OK;
+}
+
+XML_Parser XMLCALL
+XML_ExternalEntityParserCreate(XML_Parser oldParser,
+ const XML_Char *context,
+ const XML_Char *encodingName)
+{
+ XML_Parser parser = oldParser;
+ DTD *newDtd = NULL;
+ DTD *oldDtd = _dtd;
+ XML_StartElementHandler oldStartElementHandler = startElementHandler;
+ XML_EndElementHandler oldEndElementHandler = endElementHandler;
+ XML_CharacterDataHandler oldCharacterDataHandler = characterDataHandler;
+ XML_ProcessingInstructionHandler oldProcessingInstructionHandler
+ = processingInstructionHandler;
+ XML_CommentHandler oldCommentHandler = commentHandler;
+ XML_StartCdataSectionHandler oldStartCdataSectionHandler
+ = startCdataSectionHandler;
+ XML_EndCdataSectionHandler oldEndCdataSectionHandler
+ = endCdataSectionHandler;
+ XML_DefaultHandler oldDefaultHandler = defaultHandler;
+ XML_UnparsedEntityDeclHandler oldUnparsedEntityDeclHandler
+ = unparsedEntityDeclHandler;
+ XML_NotationDeclHandler oldNotationDeclHandler = notationDeclHandler;
+ XML_StartNamespaceDeclHandler oldStartNamespaceDeclHandler
+ = startNamespaceDeclHandler;
+ XML_EndNamespaceDeclHandler oldEndNamespaceDeclHandler
+ = endNamespaceDeclHandler;
+ XML_NotStandaloneHandler oldNotStandaloneHandler = notStandaloneHandler;
+ XML_ExternalEntityRefHandler oldExternalEntityRefHandler
+ = externalEntityRefHandler;
+ XML_SkippedEntityHandler oldSkippedEntityHandler = skippedEntityHandler;
+ XML_UnknownEncodingHandler oldUnknownEncodingHandler
+ = unknownEncodingHandler;
+ XML_ElementDeclHandler oldElementDeclHandler = elementDeclHandler;
+ XML_AttlistDeclHandler oldAttlistDeclHandler = attlistDeclHandler;
+ XML_EntityDeclHandler oldEntityDeclHandler = entityDeclHandler;
+ XML_XmlDeclHandler oldXmlDeclHandler = xmlDeclHandler;
+ ELEMENT_TYPE * oldDeclElementType = declElementType;
+
+ void *oldUserData = userData;
+ void *oldHandlerArg = handlerArg;
+ XML_Bool oldDefaultExpandInternalEntities = defaultExpandInternalEntities;
+ XML_Parser oldExternalEntityRefHandlerArg = externalEntityRefHandlerArg;
+#ifdef XML_DTD
+ enum XML_ParamEntityParsing oldParamEntityParsing = paramEntityParsing;
+ int oldInEntityValue = prologState.inEntityValue;
+#endif
+ XML_Bool oldns_triplets = ns_triplets;
+
+#ifdef XML_DTD
+ if (!context)
+ newDtd = oldDtd;
+#endif /* XML_DTD */
+
+ /* Note that the magical uses of the pre-processor to make field
+ access look more like C++ require that `parser' be overwritten
+ here. This makes this function more painful to follow than it
+ would be otherwise.
+ */
+ if (ns) {
+ XML_Char tmp[2];
+ *tmp = namespaceSeparator;
+ parser = parserCreate(encodingName, &parser->m_mem, tmp, newDtd);
+ }
+ else {
+ parser = parserCreate(encodingName, &parser->m_mem, NULL, newDtd);
+ }
+
+ if (!parser)
+ return NULL;
+
+ startElementHandler = oldStartElementHandler;
+ endElementHandler = oldEndElementHandler;
+ characterDataHandler = oldCharacterDataHandler;
+ processingInstructionHandler = oldProcessingInstructionHandler;
+ commentHandler = oldCommentHandler;
+ startCdataSectionHandler = oldStartCdataSectionHandler;
+ endCdataSectionHandler = oldEndCdataSectionHandler;
+ defaultHandler = oldDefaultHandler;
+ unparsedEntityDeclHandler = oldUnparsedEntityDeclHandler;
+ notationDeclHandler = oldNotationDeclHandler;
+ startNamespaceDeclHandler = oldStartNamespaceDeclHandler;
+ endNamespaceDeclHandler = oldEndNamespaceDeclHandler;
+ notStandaloneHandler = oldNotStandaloneHandler;
+ externalEntityRefHandler = oldExternalEntityRefHandler;
+ skippedEntityHandler = oldSkippedEntityHandler;
+ unknownEncodingHandler = oldUnknownEncodingHandler;
+ elementDeclHandler = oldElementDeclHandler;
+ attlistDeclHandler = oldAttlistDeclHandler;
+ entityDeclHandler = oldEntityDeclHandler;
+ xmlDeclHandler = oldXmlDeclHandler;
+ declElementType = oldDeclElementType;
+ userData = oldUserData;
+ if (oldUserData == oldHandlerArg)
+ handlerArg = userData;
+ else
+ handlerArg = parser;
+ if (oldExternalEntityRefHandlerArg != oldParser)
+ externalEntityRefHandlerArg = oldExternalEntityRefHandlerArg;
+ defaultExpandInternalEntities = oldDefaultExpandInternalEntities;
+ ns_triplets = oldns_triplets;
+ parentParser = oldParser;
+#ifdef XML_DTD
+ paramEntityParsing = oldParamEntityParsing;
+ prologState.inEntityValue = oldInEntityValue;
+ if (context) {
+#endif /* XML_DTD */
+ if (!dtdCopy(_dtd, oldDtd, &parser->m_mem)
+ || !setContext(parser, context)) {
+ XML_ParserFree(parser);
+ return NULL;
+ }
+ processor = externalEntityInitProcessor;
+#ifdef XML_DTD
+ }
+ else {
+ /* The DTD instance referenced by _dtd is shared between the document's
+ root parser and external PE parsers, therefore one does not need to
+ call setContext. In addition, one also *must* not call setContext,
+ because this would overwrite existing prefix->binding pointers in
+ _dtd with ones that get destroyed with the external PE parser.
+ This would leave those prefixes with dangling pointers.
+ */
+ isParamEntity = XML_TRUE;
+ XmlPrologStateInitExternalEntity(&prologState);
+ processor = externalParEntInitProcessor;
+ }
+#endif /* XML_DTD */
+ return parser;
+}
+
+static void FASTCALL
+destroyBindings(BINDING *bindings, XML_Parser parser)
+{
+ for (;;) {
+ BINDING *b = bindings;
+ if (!b)
+ break;
+ bindings = b->nextTagBinding;
+ FREE(b->uri);
+ FREE(b);
+ }
+}
+
+void XMLCALL
+XML_ParserFree(XML_Parser parser)
+{
+ TAG *tagList;
+ OPEN_INTERNAL_ENTITY *entityList;
+ if (parser == NULL)
+ return;
+ /* free tagStack and freeTagList */
+ tagList = tagStack;
+ for (;;) {
+ TAG *p;
+ if (tagList == NULL) {
+ if (freeTagList == NULL)
+ break;
+ tagList = freeTagList;
+ freeTagList = NULL;
+ }
+ p = tagList;
+ tagList = tagList->parent;
+ FREE(p->buf);
+ destroyBindings(p->bindings, parser);
+ FREE(p);
+ }
+ /* free openInternalEntities and freeInternalEntities */
+ entityList = openInternalEntities;
+ for (;;) {
+ OPEN_INTERNAL_ENTITY *openEntity;
+ if (entityList == NULL) {
+ if (freeInternalEntities == NULL)
+ break;
+ entityList = freeInternalEntities;
+ freeInternalEntities = NULL;
+ }
+ openEntity = entityList;
+ entityList = entityList->next;
+ FREE(openEntity);
+ }
+
+ destroyBindings(freeBindingList, parser);
+ destroyBindings(inheritedBindings, parser);
+ poolDestroy(&tempPool);
+ poolDestroy(&temp2Pool);
+#ifdef XML_DTD
+ /* external parameter entity parsers share the DTD structure
+ parser->m_dtd with the root parser, so we must not destroy it
+ */
+ if (!isParamEntity && _dtd)
+#else
+ if (_dtd)
+#endif /* XML_DTD */
+ dtdDestroy(_dtd, (XML_Bool)!parentParser, &parser->m_mem);
+ FREE((void *)atts);
+ FREE(groupConnector);
+ FREE(buffer);
+ FREE(dataBuf);
+ FREE(nsAtts);
+ FREE(unknownEncodingMem);
+ if (unknownEncodingRelease)
+ unknownEncodingRelease(unknownEncodingData);
+ FREE(parser);
+}
+
+void XMLCALL
+XML_UseParserAsHandlerArg(XML_Parser parser)
+{
+ handlerArg = parser;
+}
+
+enum XML_Error XMLCALL
+XML_UseForeignDTD(XML_Parser parser, XML_Bool useDTD)
+{
+#ifdef XML_DTD
+ /* block after XML_Parse()/XML_ParseBuffer() has been called */
+ if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED)
+ return XML_ERROR_CANT_CHANGE_FEATURE_ONCE_PARSING;
+ useForeignDTD = useDTD;
+ return XML_ERROR_NONE;
+#else
+ return XML_ERROR_FEATURE_REQUIRES_XML_DTD;
+#endif
+}
+
+void XMLCALL
+XML_SetReturnNSTriplet(XML_Parser parser, int do_nst)
+{
+ /* block after XML_Parse()/XML_ParseBuffer() has been called */
+ if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED)
+ return;
+ ns_triplets = do_nst ? XML_TRUE : XML_FALSE;
+}
+
+void XMLCALL
+XML_SetUserData(XML_Parser parser, void *p)
+{
+ if (handlerArg == userData)
+ handlerArg = userData = p;
+ else
+ userData = p;
+}
+
+enum XML_Status XMLCALL
+XML_SetBase(XML_Parser parser, const XML_Char *p)
+{
+ if (p) {
+ p = poolCopyString(&_dtd->pool, p);
+ if (!p)
+ return XML_STATUS_ERROR;
+ curBase = p;
+ }
+ else
+ curBase = NULL;
+ return XML_STATUS_OK;
+}
+
+const XML_Char * XMLCALL
+XML_GetBase(XML_Parser parser)
+{
+ return curBase;
+}
+
+int XMLCALL
+XML_GetSpecifiedAttributeCount(XML_Parser parser)
+{
+ return nSpecifiedAtts;
+}
+
+int XMLCALL
+XML_GetIdAttributeIndex(XML_Parser parser)
+{
+ return idAttIndex;
+}
+
+void XMLCALL
+XML_SetElementHandler(XML_Parser parser,
+ XML_StartElementHandler start,
+ XML_EndElementHandler end)
+{
+ startElementHandler = start;
+ endElementHandler = end;
+}
+
+void XMLCALL
+XML_SetStartElementHandler(XML_Parser parser,
+ XML_StartElementHandler start) {
+ startElementHandler = start;
+}
+
+void XMLCALL
+XML_SetEndElementHandler(XML_Parser parser,
+ XML_EndElementHandler end) {
+ endElementHandler = end;
+}
+
+void XMLCALL
+XML_SetCharacterDataHandler(XML_Parser parser,
+ XML_CharacterDataHandler handler)
+{
+ characterDataHandler = handler;
+}
+
+void XMLCALL
+XML_SetProcessingInstructionHandler(XML_Parser parser,
+ XML_ProcessingInstructionHandler handler)
+{
+ processingInstructionHandler = handler;
+}
+
+void XMLCALL
+XML_SetCommentHandler(XML_Parser parser,
+ XML_CommentHandler handler)
+{
+ commentHandler = handler;
+}
+
+void XMLCALL
+XML_SetCdataSectionHandler(XML_Parser parser,
+ XML_StartCdataSectionHandler start,
+ XML_EndCdataSectionHandler end)
+{
+ startCdataSectionHandler = start;
+ endCdataSectionHandler = end;
+}
+
+void XMLCALL
+XML_SetStartCdataSectionHandler(XML_Parser parser,
+ XML_StartCdataSectionHandler start) {
+ startCdataSectionHandler = start;
+}
+
+void XMLCALL
+XML_SetEndCdataSectionHandler(XML_Parser parser,
+ XML_EndCdataSectionHandler end) {
+ endCdataSectionHandler = end;
+}
+
+void XMLCALL
+XML_SetDefaultHandler(XML_Parser parser,
+ XML_DefaultHandler handler)
+{
+ defaultHandler = handler;
+ defaultExpandInternalEntities = XML_FALSE;
+}
+
+void XMLCALL
+XML_SetDefaultHandlerExpand(XML_Parser parser,
+ XML_DefaultHandler handler)
+{
+ defaultHandler = handler;
+ defaultExpandInternalEntities = XML_TRUE;
+}
+
+void XMLCALL
+XML_SetDoctypeDeclHandler(XML_Parser parser,
+ XML_StartDoctypeDeclHandler start,
+ XML_EndDoctypeDeclHandler end)
+{
+ startDoctypeDeclHandler = start;
+ endDoctypeDeclHandler = end;
+}
+
+void XMLCALL
+XML_SetStartDoctypeDeclHandler(XML_Parser parser,
+ XML_StartDoctypeDeclHandler start) {
+ startDoctypeDeclHandler = start;
+}
+
+void XMLCALL
+XML_SetEndDoctypeDeclHandler(XML_Parser parser,
+ XML_EndDoctypeDeclHandler end) {
+ endDoctypeDeclHandler = end;
+}
+
+void XMLCALL
+XML_SetUnparsedEntityDeclHandler(XML_Parser parser,
+ XML_UnparsedEntityDeclHandler handler)
+{
+ unparsedEntityDeclHandler = handler;
+}
+
+void XMLCALL
+XML_SetNotationDeclHandler(XML_Parser parser,
+ XML_NotationDeclHandler handler)
+{
+ notationDeclHandler = handler;
+}
+
+void XMLCALL
+XML_SetNamespaceDeclHandler(XML_Parser parser,
+ XML_StartNamespaceDeclHandler start,
+ XML_EndNamespaceDeclHandler end)
+{
+ startNamespaceDeclHandler = start;
+ endNamespaceDeclHandler = end;
+}
+
+void XMLCALL
+XML_SetStartNamespaceDeclHandler(XML_Parser parser,
+ XML_StartNamespaceDeclHandler start) {
+ startNamespaceDeclHandler = start;
+}
+
+void XMLCALL
+XML_SetEndNamespaceDeclHandler(XML_Parser parser,
+ XML_EndNamespaceDeclHandler end) {
+ endNamespaceDeclHandler = end;
+}
+
+void XMLCALL
+XML_SetNotStandaloneHandler(XML_Parser parser,
+ XML_NotStandaloneHandler handler)
+{
+ notStandaloneHandler = handler;
+}
+
+void XMLCALL
+XML_SetExternalEntityRefHandler(XML_Parser parser,
+ XML_ExternalEntityRefHandler handler)
+{
+ externalEntityRefHandler = handler;
+}
+
+void XMLCALL
+XML_SetExternalEntityRefHandlerArg(XML_Parser parser, void *arg)
+{
+ if (arg)
+ externalEntityRefHandlerArg = (XML_Parser)arg;
+ else
+ externalEntityRefHandlerArg = parser;
+}
+
+void XMLCALL
+XML_SetSkippedEntityHandler(XML_Parser parser,
+ XML_SkippedEntityHandler handler)
+{
+ skippedEntityHandler = handler;
+}
+
+void XMLCALL
+XML_SetUnknownEncodingHandler(XML_Parser parser,
+ XML_UnknownEncodingHandler handler,
+ void *data)
+{
+ unknownEncodingHandler = handler;
+ unknownEncodingHandlerData = data;
+}
+
+void XMLCALL
+XML_SetElementDeclHandler(XML_Parser parser,
+ XML_ElementDeclHandler eldecl)
+{
+ elementDeclHandler = eldecl;
+}
+
+void XMLCALL
+XML_SetAttlistDeclHandler(XML_Parser parser,
+ XML_AttlistDeclHandler attdecl)
+{
+ attlistDeclHandler = attdecl;
+}
+
+void XMLCALL
+XML_SetEntityDeclHandler(XML_Parser parser,
+ XML_EntityDeclHandler handler)
+{
+ entityDeclHandler = handler;
+}
+
+void XMLCALL
+XML_SetXmlDeclHandler(XML_Parser parser,
+ XML_XmlDeclHandler handler) {
+ xmlDeclHandler = handler;
+}
+
+int XMLCALL
+XML_SetParamEntityParsing(XML_Parser parser,
+ enum XML_ParamEntityParsing peParsing)
+{
+ /* block after XML_Parse()/XML_ParseBuffer() has been called */
+ if (ps_parsing == XML_PARSING || ps_parsing == XML_SUSPENDED)
+ return 0;
+#ifdef XML_DTD
+ paramEntityParsing = peParsing;
+ return 1;
+#else
+ return peParsing == XML_PARAM_ENTITY_PARSING_NEVER;
+#endif
+}
+
+enum XML_Status XMLCALL
+XML_Parse(XML_Parser parser, const char *s, int len, int isFinal)
+{
+ switch (ps_parsing) {
+ case XML_SUSPENDED:
+ errorCode = XML_ERROR_SUSPENDED;
+ return XML_STATUS_ERROR;
+ case XML_FINISHED:
+ errorCode = XML_ERROR_FINISHED;
+ return XML_STATUS_ERROR;
+ default:
+ ps_parsing = XML_PARSING;
+ }
+
+ if (len == 0) {
+ ps_finalBuffer = (XML_Bool)isFinal;
+ if (!isFinal)
+ return XML_STATUS_OK;
+ positionPtr = bufferPtr;
+ parseEndPtr = bufferEnd;
+
+ /* If data are left over from last buffer, and we now know that these
+ data are the final chunk of input, then we have to check them again
+ to detect errors based on that fact.
+ */
+ errorCode = processor(parser, bufferPtr, parseEndPtr, &bufferPtr);
+
+ if (errorCode == XML_ERROR_NONE) {
+ switch (ps_parsing) {
+ case XML_SUSPENDED:
+ XmlUpdatePosition(encoding, positionPtr, bufferPtr, &position);
+ positionPtr = bufferPtr;
+ return XML_STATUS_SUSPENDED;
+ case XML_INITIALIZED:
+ case XML_PARSING:
+ ps_parsing = XML_FINISHED;
+ /* fall through */
+ default:
+ return XML_STATUS_OK;
+ }
+ }
+ eventEndPtr = eventPtr;
+ processor = errorProcessor;
+ return XML_STATUS_ERROR;
+ }
+#ifndef XML_CONTEXT_BYTES
+ else if (bufferPtr == bufferEnd) {
+ const char *end;
+ int nLeftOver;
+ enum XML_Error result;
+ parseEndByteIndex += len;
+ positionPtr = s;
+ ps_finalBuffer = (XML_Bool)isFinal;
+
+ errorCode = processor(parser, s, parseEndPtr = s + len, &end);
+
+ if (errorCode != XML_ERROR_NONE) {
+ eventEndPtr = eventPtr;
+ processor = errorProcessor;
+ return XML_STATUS_ERROR;
+ }
+ else {
+ switch (ps_parsing) {
+ case XML_SUSPENDED:
+ result = XML_STATUS_SUSPENDED;
+ break;
+ case XML_INITIALIZED:
+ case XML_PARSING:
+ result = XML_STATUS_OK;
+ if (isFinal) {
+ ps_parsing = XML_FINISHED;
+ return result;
+ }
+ }
+ }
+
+ XmlUpdatePosition(encoding, positionPtr, end, &position);
+ nLeftOver = s + len - end;
+ if (nLeftOver) {
+ if (buffer == NULL || nLeftOver > bufferLim - buffer) {
+ /* FIXME avoid integer overflow */
+ char *temp;
+ temp = (buffer == NULL
+ ? (char *)MALLOC(len * 2)
+ : (char *)REALLOC(buffer, len * 2));
+ if (temp == NULL) {
+ errorCode = XML_ERROR_NO_MEMORY;
+ return XML_STATUS_ERROR;
+ }
+ buffer = temp;
+ if (!buffer) {
+ errorCode = XML_ERROR_NO_MEMORY;
+ eventPtr = eventEndPtr = NULL;
+ processor = errorProcessor;
+ return XML_STATUS_ERROR;
+ }
+ bufferLim = buffer + len * 2;
+ }
+ memcpy(buffer, end, nLeftOver);
+ }
+ bufferPtr = buffer;
+ bufferEnd = buffer + nLeftOver;
+ positionPtr = bufferPtr;
+ parseEndPtr = bufferEnd;
+ eventPtr = bufferPtr;
+ eventEndPtr = bufferPtr;
+ return result;
+ }
+#endif /* not defined XML_CONTEXT_BYTES */
+ else {
+ void *buff = XML_GetBuffer(parser, len);
+ if (buff == NULL)
+ return XML_STATUS_ERROR;
+ else {
+ memcpy(buff, s, len);
+ return XML_ParseBuffer(parser, len, isFinal);
+ }
+ }
+}
+
+enum XML_Status XMLCALL
+XML_ParseBuffer(XML_Parser parser, int len, int isFinal)
+{
+ const char *start;
+ enum XML_Status result = XML_STATUS_OK;
+
+ switch (ps_parsing) {
+ case XML_SUSPENDED:
+ errorCode = XML_ERROR_SUSPENDED;
+ return XML_STATUS_ERROR;
+ case XML_FINISHED:
+ errorCode = XML_ERROR_FINISHED;
+ return XML_STATUS_ERROR;
+ default:
+ ps_parsing = XML_PARSING;
+ }
+
+ start = bufferPtr;
+ positionPtr = start;
+ bufferEnd += len;
+ parseEndPtr = bufferEnd;
+ parseEndByteIndex += len;
+ ps_finalBuffer = (XML_Bool)isFinal;
+
+ errorCode = processor(parser, start, parseEndPtr, &bufferPtr);
+
+ if (errorCode != XML_ERROR_NONE) {
+ eventEndPtr = eventPtr;
+ processor = errorProcessor;
+ return XML_STATUS_ERROR;
+ }
+ else {
+ switch (ps_parsing) {
+ case XML_SUSPENDED:
+ result = XML_STATUS_SUSPENDED;
+ break;
+ case XML_INITIALIZED:
+ case XML_PARSING:
+ if (isFinal) {
+ ps_parsing = XML_FINISHED;
+ return result;
+ }
+ default: ; /* should not happen */
+ }
+ }
+
+ XmlUpdatePosition(encoding, positionPtr, bufferPtr, &position);
+ positionPtr = bufferPtr;
+ return result;
+}
+
+void * XMLCALL
+XML_GetBuffer(XML_Parser parser, int len)
+{
+ switch (ps_parsing) {
+ case XML_SUSPENDED:
+ errorCode = XML_ERROR_SUSPENDED;
+ return NULL;
+ case XML_FINISHED:
+ errorCode = XML_ERROR_FINISHED;
+ return NULL;
+ default: ;
+ }
+
+ if (len > bufferLim - bufferEnd) {
+ /* FIXME avoid integer overflow */
+ int neededSize = len + (int)(bufferEnd - bufferPtr);
+#ifdef XML_CONTEXT_BYTES
+ int keep = (int)(bufferPtr - buffer);
+
+ if (keep > XML_CONTEXT_BYTES)
+ keep = XML_CONTEXT_BYTES;
+ neededSize += keep;
+#endif /* defined XML_CONTEXT_BYTES */
+ if (neededSize <= bufferLim - buffer) {
+#ifdef XML_CONTEXT_BYTES
+ if (keep < bufferPtr - buffer) {
+ int offset = (int)(bufferPtr - buffer) - keep;
+ memmove(buffer, &buffer[offset], bufferEnd - bufferPtr + keep);
+ bufferEnd -= offset;
+ bufferPtr -= offset;
+ }
+#else
+ memmove(buffer, bufferPtr, bufferEnd - bufferPtr);
+ bufferEnd = buffer + (bufferEnd - bufferPtr);
+ bufferPtr = buffer;
+#endif /* not defined XML_CONTEXT_BYTES */
+ }
+ else {
+ char *newBuf;
+ int bufferSize = (int)(bufferLim - bufferPtr);
+ if (bufferSize == 0)
+ bufferSize = INIT_BUFFER_SIZE;
+ do {
+ bufferSize *= 2;
+ } while (bufferSize < neededSize);
+ newBuf = (char *)MALLOC(bufferSize);
+ if (newBuf == 0) {
+ errorCode = XML_ERROR_NO_MEMORY;
+ return NULL;
+ }
+ bufferLim = newBuf + bufferSize;
+#ifdef XML_CONTEXT_BYTES
+ if (bufferPtr) {
+ int keep = (int)(bufferPtr - buffer);
+ if (keep > XML_CONTEXT_BYTES)
+ keep = XML_CONTEXT_BYTES;
+ memcpy(newBuf, &bufferPtr[-keep], bufferEnd - bufferPtr + keep);
+ FREE(buffer);
+ buffer = newBuf;
+ bufferEnd = buffer + (bufferEnd - bufferPtr) + keep;
+ bufferPtr = buffer + keep;
+ }
+ else {
+ bufferEnd = newBuf + (bufferEnd - bufferPtr);
+ bufferPtr = buffer = newBuf;
+ }
+#else
+ if (bufferPtr) {
+ memcpy(newBuf, bufferPtr, bufferEnd - bufferPtr);
+ FREE(buffer);
+ }
+ bufferEnd = newBuf + (bufferEnd - bufferPtr);
+ bufferPtr = buffer = newBuf;
+#endif /* not defined XML_CONTEXT_BYTES */
+ }
+ }
+ return bufferEnd;
+}
+
+enum XML_Status XMLCALL
+XML_StopParser(XML_Parser parser, XML_Bool resumable)
+{
+ switch (ps_parsing) {
+ case XML_SUSPENDED:
+ if (resumable) {
+ errorCode = XML_ERROR_SUSPENDED;
+ return XML_STATUS_ERROR;
+ }
+ ps_parsing = XML_FINISHED;
+ break;
+ case XML_FINISHED:
+ errorCode = XML_ERROR_FINISHED;
+ return XML_STATUS_ERROR;
+ default:
+ if (resumable) {
+#ifdef XML_DTD
+ if (isParamEntity) {
+ errorCode = XML_ERROR_SUSPEND_PE;
+ return XML_STATUS_ERROR;
+ }
+#endif
+ ps_parsing = XML_SUSPENDED;
+ }
+ else
+ ps_parsing = XML_FINISHED;
+ }
+ return XML_STATUS_OK;
+}
+
+enum XML_Status XMLCALL
+XML_ResumeParser(XML_Parser parser)
+{
+ enum XML_Status result = XML_STATUS_OK;
+
+ if (ps_parsing != XML_SUSPENDED) {
+ errorCode = XML_ERROR_NOT_SUSPENDED;
+ return XML_STATUS_ERROR;
+ }
+ ps_parsing = XML_PARSING;
+
+ errorCode = processor(parser, bufferPtr, parseEndPtr, &bufferPtr);
+
+ if (errorCode != XML_ERROR_NONE) {
+ eventEndPtr = eventPtr;
+ processor = errorProcessor;
+ return XML_STATUS_ERROR;
+ }
+ else {
+ switch (ps_parsing) {
+ case XML_SUSPENDED:
+ result = XML_STATUS_SUSPENDED;
+ break;
+ case XML_INITIALIZED:
+ case XML_PARSING:
+ if (ps_finalBuffer) {
+ ps_parsing = XML_FINISHED;
+ return result;
+ }
+ default: ;
+ }
+ }
+
+ XmlUpdatePosition(encoding, positionPtr, bufferPtr, &position);
+ positionPtr = bufferPtr;
+ return result;
+}
+
+void XMLCALL
+XML_GetParsingStatus(XML_Parser parser, XML_ParsingStatus *status)
+{
+ assert(status != NULL);
+ *status = parser->m_parsingStatus;
+}
+
+enum XML_Error XMLCALL
+XML_GetErrorCode(XML_Parser parser)
+{
+ return errorCode;
+}
+
+XML_Index XMLCALL
+XML_GetCurrentByteIndex(XML_Parser parser)
+{
+ if (eventPtr)
+ return parseEndByteIndex - (parseEndPtr - eventPtr);
+ return -1;
+}
+
+int XMLCALL
+XML_GetCurrentByteCount(XML_Parser parser)
+{
+ if (eventEndPtr && eventPtr)
+ return (int)(eventEndPtr - eventPtr);
+ return 0;
+}
+
+const char * XMLCALL
+XML_GetInputContext(XML_Parser parser, int *offset, int *size)
+{
+#ifdef XML_CONTEXT_BYTES
+ if (eventPtr && buffer) {
+ *offset = (int)(eventPtr - buffer);
+ *size = (int)(bufferEnd - buffer);
+ return buffer;
+ }
+#endif /* defined XML_CONTEXT_BYTES */
+ return (char *) 0;
+}
+
+XML_Size XMLCALL
+XML_GetCurrentLineNumber(XML_Parser parser)
+{
+ if (eventPtr && eventPtr >= positionPtr) {
+ XmlUpdatePosition(encoding, positionPtr, eventPtr, &position);
+ positionPtr = eventPtr;
+ }
+ return position.lineNumber + 1;
+}
+
+XML_Size XMLCALL
+XML_GetCurrentColumnNumber(XML_Parser parser)
+{
+ if (eventPtr && eventPtr >= positionPtr) {
+ XmlUpdatePosition(encoding, positionPtr, eventPtr, &position);
+ positionPtr = eventPtr;
+ }
+ return position.columnNumber;
+}
+
+void XMLCALL
+XML_FreeContentModel(XML_Parser parser, XML_Content *model)
+{
+ FREE(model);
+}
+
+void * XMLCALL
+XML_MemMalloc(XML_Parser parser, size_t size)
+{
+ return MALLOC(size);
+}
+
+void * XMLCALL
+XML_MemRealloc(XML_Parser parser, void *ptr, size_t size)
+{
+ return REALLOC(ptr, size);
+}
+
+void XMLCALL
+XML_MemFree(XML_Parser parser, void *ptr)
+{
+ FREE(ptr);
+}
+
+void XMLCALL
+XML_DefaultCurrent(XML_Parser parser)
+{
+ if (defaultHandler) {
+ if (openInternalEntities)
+ reportDefault(parser,
+ internalEncoding,
+ openInternalEntities->internalEventPtr,
+ openInternalEntities->internalEventEndPtr);
+ else
+ reportDefault(parser, encoding, eventPtr, eventEndPtr);
+ }
+}
+
+const XML_LChar * XMLCALL
+XML_ErrorString(enum XML_Error code)
+{
+ static const XML_LChar* const message[] = {
+ 0,
+ XML_L("out of memory"),
+ XML_L("syntax error"),
+ XML_L("no element found"),
+ XML_L("not well-formed (invalid token)"),
+ XML_L("unclosed token"),
+ XML_L("partial character"),
+ XML_L("mismatched tag"),
+ XML_L("duplicate attribute"),
+ XML_L("junk after document element"),
+ XML_L("illegal parameter entity reference"),
+ XML_L("undefined entity"),
+ XML_L("recursive entity reference"),
+ XML_L("asynchronous entity"),
+ XML_L("reference to invalid character number"),
+ XML_L("reference to binary entity"),
+ XML_L("reference to external entity in attribute"),
+ XML_L("XML or text declaration not at start of entity"),
+ XML_L("unknown encoding"),
+ XML_L("encoding specified in XML declaration is incorrect"),
+ XML_L("unclosed CDATA section"),
+ XML_L("error in processing external entity reference"),
+ XML_L("document is not standalone"),
+ XML_L("unexpected parser state - please send a bug report"),
+ XML_L("entity declared in parameter entity"),
+ XML_L("requested feature requires XML_DTD support in Expat"),
+ XML_L("cannot change setting once parsing has begun"),
+ XML_L("unbound prefix"),
+ XML_L("must not undeclare prefix"),
+ XML_L("incomplete markup in parameter entity"),
+ XML_L("XML declaration not well-formed"),
+ XML_L("text declaration not well-formed"),
+ XML_L("illegal character(s) in public id"),
+ XML_L("parser suspended"),
+ XML_L("parser not suspended"),
+ XML_L("parsing aborted"),
+ XML_L("parsing finished"),
+ XML_L("cannot suspend in external parameter entity"),
+ XML_L("reserved prefix (xml) must not be undeclared or bound to another namespace name"),
+ XML_L("reserved prefix (xmlns) must not be declared or undeclared"),
+ XML_L("prefix must not be bound to one of the reserved namespace names")
+ };
+ if (code > 0 && code < sizeof(message)/sizeof(message[0]))
+ return message[code];
+ return NULL;
+}
+
+const XML_LChar * XMLCALL
+XML_ExpatVersion(void) {
+
+ /* V1 is used to string-ize the version number. However, it would
+ string-ize the actual version macro *names* unless we get them
+ substituted before being passed to V1. CPP is defined to expand
+ a macro, then rescan for more expansions. Thus, we use V2 to expand
+ the version macros, then CPP will expand the resulting V1() macro
+ with the correct numerals. */
+ /* ### I'm assuming cpp is portable in this respect... */
+
+#define V1(a,b,c) XML_L(#a)XML_L(".")XML_L(#b)XML_L(".")XML_L(#c)
+#define V2(a,b,c) XML_L("expat_")V1(a,b,c)
+
+ return V2(XML_MAJOR_VERSION, XML_MINOR_VERSION, XML_MICRO_VERSION);
+
+#undef V1
+#undef V2
+}
+
+XML_Expat_Version XMLCALL
+XML_ExpatVersionInfo(void)
+{
+ XML_Expat_Version version;
+
+ version.major = XML_MAJOR_VERSION;
+ version.minor = XML_MINOR_VERSION;
+ version.micro = XML_MICRO_VERSION;
+
+ return version;
+}
+
+const XML_Feature * XMLCALL
+XML_GetFeatureList(void)
+{
+ static const XML_Feature features[] = {
+ {XML_FEATURE_SIZEOF_XML_CHAR, XML_L("sizeof(XML_Char)"),
+ sizeof(XML_Char)},
+ {XML_FEATURE_SIZEOF_XML_LCHAR, XML_L("sizeof(XML_LChar)"),
+ sizeof(XML_LChar)},
+#ifdef XML_UNICODE
+ {XML_FEATURE_UNICODE, XML_L("XML_UNICODE"), 0},
+#endif
+#ifdef XML_UNICODE_WCHAR_T
+ {XML_FEATURE_UNICODE_WCHAR_T, XML_L("XML_UNICODE_WCHAR_T"), 0},
+#endif
+#ifdef XML_DTD
+ {XML_FEATURE_DTD, XML_L("XML_DTD"), 0},
+#endif
+#ifdef XML_CONTEXT_BYTES
+ {XML_FEATURE_CONTEXT_BYTES, XML_L("XML_CONTEXT_BYTES"),
+ XML_CONTEXT_BYTES},
+#endif
+#ifdef XML_MIN_SIZE
+ {XML_FEATURE_MIN_SIZE, XML_L("XML_MIN_SIZE"), 0},
+#endif
+#ifdef XML_NS
+ {XML_FEATURE_NS, XML_L("XML_NS"), 0},
+#endif
+ {XML_FEATURE_END, NULL, 0}
+ };
+
+ return features;
+}
+
+/* Initially tag->rawName always points into the parse buffer;
+ for those TAG instances opened while the current parse buffer was
+ processed, and not yet closed, we need to store tag->rawName in a more
+ permanent location, since the parse buffer is about to be discarded.
+*/
+static XML_Bool
+storeRawNames(XML_Parser parser)
+{
+ TAG *tag = tagStack;
+ while (tag) {
+ int bufSize;
+ int nameLen = sizeof(XML_Char) * (tag->name.strLen + 1);
+ char *rawNameBuf = tag->buf + nameLen;
+ /* Stop if already stored. Since tagStack is a stack, we can stop
+ at the first entry that has already been copied; everything
+ below it in the stack is already been accounted for in a
+ previous call to this function.
+ */
+ if (tag->rawName == rawNameBuf)
+ break;
+ /* For re-use purposes we need to ensure that the
+ size of tag->buf is a multiple of sizeof(XML_Char).
+ */
+ bufSize = nameLen + ROUND_UP(tag->rawNameLength, sizeof(XML_Char));
+ if (bufSize > tag->bufEnd - tag->buf) {
+ char *temp = (char *)REALLOC(tag->buf, bufSize);
+ if (temp == NULL)
+ return XML_FALSE;
+ /* if tag->name.str points to tag->buf (only when namespace
+ processing is off) then we have to update it
+ */
+ if (tag->name.str == (XML_Char *)tag->buf)
+ tag->name.str = (XML_Char *)temp;
+ /* if tag->name.localPart is set (when namespace processing is on)
+ then update it as well, since it will always point into tag->buf
+ */
+ if (tag->name.localPart)
+ tag->name.localPart = (XML_Char *)temp + (tag->name.localPart -
+ (XML_Char *)tag->buf);
+ tag->buf = temp;
+ tag->bufEnd = temp + bufSize;
+ rawNameBuf = temp + nameLen;
+ }
+ memcpy(rawNameBuf, tag->rawName, tag->rawNameLength);
+ tag->rawName = rawNameBuf;
+ tag = tag->parent;
+ }
+ return XML_TRUE;
+}
+
+static enum XML_Error PTRCALL
+contentProcessor(XML_Parser parser,
+ const char *start,
+ const char *end,
+ const char **endPtr)
+{
+ enum XML_Error result = doContent(parser, 0, encoding, start, end,
+ endPtr, (XML_Bool)!ps_finalBuffer);
+ if (result == XML_ERROR_NONE) {
+ if (!storeRawNames(parser))
+ return XML_ERROR_NO_MEMORY;
+ }
+ return result;
+}
+
+static enum XML_Error PTRCALL
+externalEntityInitProcessor(XML_Parser parser,
+ const char *start,
+ const char *end,
+ const char **endPtr)
+{
+ enum XML_Error result = initializeEncoding(parser);
+ if (result != XML_ERROR_NONE)
+ return result;
+ processor = externalEntityInitProcessor2;
+ return externalEntityInitProcessor2(parser, start, end, endPtr);
+}
+
+static enum XML_Error PTRCALL
+externalEntityInitProcessor2(XML_Parser parser,
+ const char *start,
+ const char *end,
+ const char **endPtr)
+{
+ const char *next = start; /* XmlContentTok doesn't always set the last arg */
+ int tok = XmlContentTok(encoding, start, end, &next);
+ switch (tok) {
+ case XML_TOK_BOM:
+ /* If we are at the end of the buffer, this would cause the next stage,
+ i.e. externalEntityInitProcessor3, to pass control directly to
+ doContent (by detecting XML_TOK_NONE) without processing any xml text
+ declaration - causing the error XML_ERROR_MISPLACED_XML_PI in doContent.
+ */
+ if (next == end && !ps_finalBuffer) {
+ *endPtr = next;
+ return XML_ERROR_NONE;
+ }
+ start = next;
+ break;
+ case XML_TOK_PARTIAL:
+ if (!ps_finalBuffer) {
+ *endPtr = start;
+ return XML_ERROR_NONE;
+ }
+ eventPtr = start;
+ return XML_ERROR_UNCLOSED_TOKEN;
+ case XML_TOK_PARTIAL_CHAR:
+ if (!ps_finalBuffer) {
+ *endPtr = start;
+ return XML_ERROR_NONE;
+ }
+ eventPtr = start;
+ return XML_ERROR_PARTIAL_CHAR;
+ }
+ processor = externalEntityInitProcessor3;
+ return externalEntityInitProcessor3(parser, start, end, endPtr);
+}
+
+static enum XML_Error PTRCALL
+externalEntityInitProcessor3(XML_Parser parser,
+ const char *start,
+ const char *end,
+ const char **endPtr)
+{
+ int tok;
+ const char *next = start; /* XmlContentTok doesn't always set the last arg */
+ eventPtr = start;
+ tok = XmlContentTok(encoding, start, end, &next);
+ eventEndPtr = next;
+
+ switch (tok) {
+ case XML_TOK_XML_DECL:
+ {
+ enum XML_Error result;
+ result = processXmlDecl(parser, 1, start, next);
+ if (result != XML_ERROR_NONE)
+ return result;
+ switch (ps_parsing) {
+ case XML_SUSPENDED:
+ *endPtr = next;
+ return XML_ERROR_NONE;
+ case XML_FINISHED:
+ return XML_ERROR_ABORTED;
+ default:
+ start = next;
+ }
+ }
+ break;
+ case XML_TOK_PARTIAL:
+ if (!ps_finalBuffer) {
+ *endPtr = start;
+ return XML_ERROR_NONE;
+ }
+ return XML_ERROR_UNCLOSED_TOKEN;
+ case XML_TOK_PARTIAL_CHAR:
+ if (!ps_finalBuffer) {
+ *endPtr = start;
+ return XML_ERROR_NONE;
+ }
+ return XML_ERROR_PARTIAL_CHAR;
+ }
+ processor = externalEntityContentProcessor;
+ tagLevel = 1;
+ return externalEntityContentProcessor(parser, start, end, endPtr);
+}
+
+static enum XML_Error PTRCALL
+externalEntityContentProcessor(XML_Parser parser,
+ const char *start,
+ const char *end,
+ const char **endPtr)
+{
+ enum XML_Error result = doContent(parser, 1, encoding, start, end,
+ endPtr, (XML_Bool)!ps_finalBuffer);
+ if (result == XML_ERROR_NONE) {
+ if (!storeRawNames(parser))
+ return XML_ERROR_NO_MEMORY;
+ }
+ return result;
+}
+
+static enum XML_Error
+doContent(XML_Parser parser,
+ int startTagLevel,
+ const ENCODING *enc,
+ const char *s,
+ const char *end,
+ const char **nextPtr,
+ XML_Bool haveMore)
+{
+ /* save one level of indirection */
+ DTD * const dtd = _dtd;
+
+ const char **eventPP;
+ const char **eventEndPP;
+ if (enc == encoding) {
+ eventPP = &eventPtr;
+ eventEndPP = &eventEndPtr;
+ }
+ else {
+ eventPP = &(openInternalEntities->internalEventPtr);
+ eventEndPP = &(openInternalEntities->internalEventEndPtr);
+ }
+ *eventPP = s;
+
+ for (;;) {
+ const char *next = s; /* XmlContentTok doesn't always set the last arg */
+ int tok = XmlContentTok(enc, s, end, &next);
+ *eventEndPP = next;
+ switch (tok) {
+ case XML_TOK_TRAILING_CR:
+ if (haveMore) {
+ *nextPtr = s;
+ return XML_ERROR_NONE;
+ }
+ *eventEndPP = end;
+ if (characterDataHandler) {
+ XML_Char c = 0xA;
+ characterDataHandler(handlerArg, &c, 1);
+ }
+ else if (defaultHandler)
+ reportDefault(parser, enc, s, end);
+ /* We are at the end of the final buffer, should we check for
+ XML_SUSPENDED, XML_FINISHED?
+ */
+ if (startTagLevel == 0)
+ return XML_ERROR_NO_ELEMENTS;
+ if (tagLevel != startTagLevel)
+ return XML_ERROR_ASYNC_ENTITY;
+ *nextPtr = end;
+ return XML_ERROR_NONE;
+ case XML_TOK_NONE:
+ if (haveMore) {
+ *nextPtr = s;
+ return XML_ERROR_NONE;
+ }
+ if (startTagLevel > 0) {
+ if (tagLevel != startTagLevel)
+ return XML_ERROR_ASYNC_ENTITY;
+ *nextPtr = s;
+ return XML_ERROR_NONE;
+ }
+ return XML_ERROR_NO_ELEMENTS;
+ case XML_TOK_INVALID:
+ *eventPP = next;
+ return XML_ERROR_INVALID_TOKEN;
+ case XML_TOK_PARTIAL:
+ if (haveMore) {
+ *nextPtr = s;
+ return XML_ERROR_NONE;
+ }
+ return XML_ERROR_UNCLOSED_TOKEN;
+ case XML_TOK_PARTIAL_CHAR:
+ if (haveMore) {
+ *nextPtr = s;
+ return XML_ERROR_NONE;
+ }
+ return XML_ERROR_PARTIAL_CHAR;
+ case XML_TOK_ENTITY_REF:
+ {
+ const XML_Char *name;
+ ENTITY *entity;
+ XML_Char ch = (XML_Char) XmlPredefinedEntityName(enc,
+ s + enc->minBytesPerChar,
+ next - enc->minBytesPerChar);
+ if (ch) {
+ if (characterDataHandler)
+ characterDataHandler(handlerArg, &ch, 1);
+ else if (defaultHandler)
+ reportDefault(parser, enc, s, next);
+ break;
+ }
+ name = poolStoreString(&dtd->pool, enc,
+ s + enc->minBytesPerChar,
+ next - enc->minBytesPerChar);
+ if (!name)
+ return XML_ERROR_NO_MEMORY;
+ entity = (ENTITY *)lookup(&dtd->generalEntities, name, 0);
+ poolDiscard(&dtd->pool);
+ /* First, determine if a check for an existing declaration is needed;
+ if yes, check that the entity exists, and that it is internal,
+ otherwise call the skipped entity or default handler.
+ */
+ if (!dtd->hasParamEntityRefs || dtd->standalone) {
+ if (!entity)
+ return XML_ERROR_UNDEFINED_ENTITY;
+ else if (!entity->is_internal)
+ return XML_ERROR_ENTITY_DECLARED_IN_PE;
+ }
+ else if (!entity) {
+ if (skippedEntityHandler)
+ skippedEntityHandler(handlerArg, name, 0);
+ else if (defaultHandler)
+ reportDefault(parser, enc, s, next);
+ break;
+ }
+ if (entity->open)
+ return XML_ERROR_RECURSIVE_ENTITY_REF;
+ if (entity->notation)
+ return XML_ERROR_BINARY_ENTITY_REF;
+ if (entity->textPtr) {
+ enum XML_Error result;
+ if (!defaultExpandInternalEntities) {
+ if (skippedEntityHandler)
+ skippedEntityHandler(handlerArg, entity->name, 0);
+ else if (defaultHandler)
+ reportDefault(parser, enc, s, next);
+ break;
+ }
+ result = processInternalEntity(parser, entity, XML_FALSE);
+ if (result != XML_ERROR_NONE)
+ return result;
+ }
+ else if (externalEntityRefHandler) {
+ const XML_Char *context;
+ entity->open = XML_TRUE;
+ context = getContext(parser);
+ entity->open = XML_FALSE;
+ if (!context)
+ return XML_ERROR_NO_MEMORY;
+ if (!externalEntityRefHandler(externalEntityRefHandlerArg,
+ context,
+ entity->base,
+ entity->systemId,
+ entity->publicId))
+ return XML_ERROR_EXTERNAL_ENTITY_HANDLING;
+ poolDiscard(&tempPool);
+ }
+ else if (defaultHandler)
+ reportDefault(parser, enc, s, next);
+ break;
+ }
+ case XML_TOK_START_TAG_NO_ATTS:
+ /* fall through */
+ case XML_TOK_START_TAG_WITH_ATTS:
+ {
+ TAG *tag;
+ enum XML_Error result;
+ XML_Char *toPtr;
+ if (freeTagList) {
+ tag = freeTagList;
+ freeTagList = freeTagList->parent;
+ }
+ else {
+ tag = (TAG *)MALLOC(sizeof(TAG));
+ if (!tag)
+ return XML_ERROR_NO_MEMORY;
+ tag->buf = (char *)MALLOC(INIT_TAG_BUF_SIZE);
+ if (!tag->buf) {
+ FREE(tag);
+ return XML_ERROR_NO_MEMORY;
+ }
+ tag->bufEnd = tag->buf + INIT_TAG_BUF_SIZE;
+ }
+ tag->bindings = NULL;
+ tag->parent = tagStack;
+ tagStack = tag;
+ tag->name.localPart = NULL;
+ tag->name.prefix = NULL;
+ tag->rawName = s + enc->minBytesPerChar;
+ tag->rawNameLength = XmlNameLength(enc, tag->rawName);
+ ++tagLevel;
+ {
+ const char *rawNameEnd = tag->rawName + tag->rawNameLength;
+ const char *fromPtr = tag->rawName;
+ toPtr = (XML_Char *)tag->buf;
+ for (;;) {
+ int bufSize;
+ int convLen;
+ XmlConvert(enc,
+ &fromPtr, rawNameEnd,
+ (ICHAR **)&toPtr, (ICHAR *)tag->bufEnd - 1);
+ convLen = (int)(toPtr - (XML_Char *)tag->buf);
+ if (fromPtr == rawNameEnd) {
+ tag->name.strLen = convLen;
+ break;
+ }
+ bufSize = (int)(tag->bufEnd - tag->buf) << 1;
+ {
+ char *temp = (char *)REALLOC(tag->buf, bufSize);
+ if (temp == NULL)
+ return XML_ERROR_NO_MEMORY;
+ tag->buf = temp;
+ tag->bufEnd = temp + bufSize;
+ toPtr = (XML_Char *)temp + convLen;
+ }
+ }
+ }
+ tag->name.str = (XML_Char *)tag->buf;
+ *toPtr = XML_T('\0');
+ result = storeAtts(parser, enc, s, &(tag->name), &(tag->bindings));
+ if (result)
+ return result;
+ if (startElementHandler)
+ startElementHandler(handlerArg, tag->name.str,
+ (const XML_Char **)atts);
+ else if (defaultHandler)
+ reportDefault(parser, enc, s, next);
+ poolClear(&tempPool);
+ break;
+ }
+ case XML_TOK_EMPTY_ELEMENT_NO_ATTS:
+ /* fall through */
+ case XML_TOK_EMPTY_ELEMENT_WITH_ATTS:
+ {
+ const char *rawName = s + enc->minBytesPerChar;
+ enum XML_Error result;
+ BINDING *bindings = NULL;
+ XML_Bool noElmHandlers = XML_TRUE;
+ TAG_NAME name;
+ name.str = poolStoreString(&tempPool, enc, rawName,
+ rawName + XmlNameLength(enc, rawName));
+ if (!name.str)
+ return XML_ERROR_NO_MEMORY;
+ poolFinish(&tempPool);
+ result = storeAtts(parser, enc, s, &name, &bindings);
+ if (result)
+ return result;
+ poolFinish(&tempPool);
+ if (startElementHandler) {
+ startElementHandler(handlerArg, name.str, (const XML_Char **)atts);
+ noElmHandlers = XML_FALSE;
+ }
+ if (endElementHandler) {
+ if (startElementHandler)
+ *eventPP = *eventEndPP;
+ endElementHandler(handlerArg, name.str);
+ noElmHandlers = XML_FALSE;
+ }
+ if (noElmHandlers && defaultHandler)
+ reportDefault(parser, enc, s, next);
+ poolClear(&tempPool);
+ while (bindings) {
+ BINDING *b = bindings;
+ if (endNamespaceDeclHandler)
+ endNamespaceDeclHandler(handlerArg, b->prefix->name);
+ bindings = bindings->nextTagBinding;
+ b->nextTagBinding = freeBindingList;
+ freeBindingList = b;
+ b->prefix->binding = b->prevPrefixBinding;
+ }
+ }
+ if (tagLevel == 0)
+ return epilogProcessor(parser, next, end, nextPtr);
+ break;
+ case XML_TOK_END_TAG:
+ if (tagLevel == startTagLevel)
+ return XML_ERROR_ASYNC_ENTITY;
+ else {
+ int len;
+ const char *rawName;
+ TAG *tag = tagStack;
+ tagStack = tag->parent;
+ tag->parent = freeTagList;
+ freeTagList = tag;
+ rawName = s + enc->minBytesPerChar*2;
+ len = XmlNameLength(enc, rawName);
+ if (len != tag->rawNameLength
+ || memcmp(tag->rawName, rawName, len) != 0) {
+ *eventPP = rawName;
+ return XML_ERROR_TAG_MISMATCH;
+ }
+ --tagLevel;
+ if (endElementHandler) {
+ const XML_Char *localPart;
+ const XML_Char *prefix;
+ XML_Char *uri;
+ localPart = tag->name.localPart;
+ if (ns && localPart) {
+ /* localPart and prefix may have been overwritten in
+ tag->name.str, since this points to the binding->uri
+ buffer which gets re-used; so we have to add them again
+ */
+ uri = (XML_Char *)tag->name.str + tag->name.uriLen;
+ /* don't need to check for space - already done in storeAtts() */
+ while (*localPart) *uri++ = *localPart++;
+ prefix = (XML_Char *)tag->name.prefix;
+ if (ns_triplets && prefix) {
+ *uri++ = namespaceSeparator;
+ while (*prefix) *uri++ = *prefix++;
+ }
+ *uri = XML_T('\0');
+ }
+ endElementHandler(handlerArg, tag->name.str);
+ }
+ else if (defaultHandler)
+ reportDefault(parser, enc, s, next);
+ while (tag->bindings) {
+ BINDING *b = tag->bindings;
+ if (endNamespaceDeclHandler)
+ endNamespaceDeclHandler(handlerArg, b->prefix->name);
+ tag->bindings = tag->bindings->nextTagBinding;
+ b->nextTagBinding = freeBindingList;
+ freeBindingList = b;
+ b->prefix->binding = b->prevPrefixBinding;
+ }
+ if (tagLevel == 0)
+ return epilogProcessor(parser, next, end, nextPtr);
+ }
+ break;
+ case XML_TOK_CHAR_REF:
+ {
+ int n = XmlCharRefNumber(enc, s);
+ if (n < 0)
+ return XML_ERROR_BAD_CHAR_REF;
+ if (characterDataHandler) {
+ XML_Char buf[XML_ENCODE_MAX];
+ characterDataHandler(handlerArg, buf, XmlEncode(n, (ICHAR *)buf));
+ }
+ else if (defaultHandler)
+ reportDefault(parser, enc, s, next);
+ }
+ break;
+ case XML_TOK_XML_DECL:
+ return XML_ERROR_MISPLACED_XML_PI;
+ case XML_TOK_DATA_NEWLINE:
+ if (characterDataHandler) {
+ XML_Char c = 0xA;
+ characterDataHandler(handlerArg, &c, 1);
+ }
+ else if (defaultHandler)
+ reportDefault(parser, enc, s, next);
+ break;
+ case XML_TOK_CDATA_SECT_OPEN:
+ {
+ enum XML_Error result;
+ if (startCdataSectionHandler)
+ startCdataSectionHandler(handlerArg);
+#if 0
+ /* Suppose you doing a transformation on a document that involves
+ changing only the character data. You set up a defaultHandler
+ and a characterDataHandler. The defaultHandler simply copies
+ characters through. The characterDataHandler does the
+ transformation and writes the characters out escaping them as
+ necessary. This case will fail to work if we leave out the
+ following two lines (because & and < inside CDATA sections will
+ be incorrectly escaped).
+
+ However, now we have a start/endCdataSectionHandler, so it seems
+ easier to let the user deal with this.
+ */
+ else if (characterDataHandler)
+ characterDataHandler(handlerArg, dataBuf, 0);
+#endif
+ else if (defaultHandler)
+ reportDefault(parser, enc, s, next);
+ result = doCdataSection(parser, enc, &next, end, nextPtr, haveMore);
+ if (result != XML_ERROR_NONE)
+ return result;
+ else if (!next) {
+ processor = cdataSectionProcessor;
+ return result;
+ }
+ }
+ break;
+ case XML_TOK_TRAILING_RSQB:
+ if (haveMore) {
+ *nextPtr = s;
+ return XML_ERROR_NONE;
+ }
+ if (characterDataHandler) {
+ if (MUST_CONVERT(enc, s)) {
+ ICHAR *dataPtr = (ICHAR *)dataBuf;
+ XmlConvert(enc, &s, end, &dataPtr, (ICHAR *)dataBufEnd);
+ characterDataHandler(handlerArg, dataBuf,
+ (int)(dataPtr - (ICHAR *)dataBuf));
+ }
+ else
+ characterDataHandler(handlerArg,
+ (XML_Char *)s,
+ (int)((XML_Char *)end - (XML_Char *)s));
+ }
+ else if (defaultHandler)
+ reportDefault(parser, enc, s, end);
+ /* We are at the end of the final buffer, should we check for
+ XML_SUSPENDED, XML_FINISHED?
+ */
+ if (startTagLevel == 0) {
+ *eventPP = end;
+ return XML_ERROR_NO_ELEMENTS;
+ }
+ if (tagLevel != startTagLevel) {
+ *eventPP = end;
+ return XML_ERROR_ASYNC_ENTITY;
+ }
+ *nextPtr = end;
+ return XML_ERROR_NONE;
+ case XML_TOK_DATA_CHARS:
+ if (characterDataHandler) {
+ if (MUST_CONVERT(enc, s)) {
+ for (;;) {
+ ICHAR *dataPtr = (ICHAR *)dataBuf;
+ XmlConvert(enc, &s, next, &dataPtr, (ICHAR *)dataBufEnd);
+ *eventEndPP = s;
+ characterDataHandler(handlerArg, dataBuf,
+ (int)(dataPtr - (ICHAR *)dataBuf));
+ if (s == next)
+ break;
+ *eventPP = s;
+ }
+ }
+ else
+ characterDataHandler(handlerArg,
+ (XML_Char *)s,
+ (int)((XML_Char *)next - (XML_Char *)s));
+ }
+ else if (defaultHandler)
+ reportDefault(parser, enc, s, next);
+ break;
+ case XML_TOK_PI:
+ if (!reportProcessingInstruction(parser, enc, s, next))
+ return XML_ERROR_NO_MEMORY;
+ break;
+ case XML_TOK_COMMENT:
+ if (!reportComment(parser, enc, s, next))
+ return XML_ERROR_NO_MEMORY;
+ break;
+ default:
+ if (defaultHandler)
+ reportDefault(parser, enc, s, next);
+ break;
+ }
+ *eventPP = s = next;
+ switch (ps_parsing) {
+ case XML_SUSPENDED:
+ *nextPtr = next;
+ return XML_ERROR_NONE;
+ case XML_FINISHED:
+ return XML_ERROR_ABORTED;
+ default: ;
+ }
+ }
+ /* not reached */
+}
+
+/* Precondition: all arguments must be non-NULL;
+ Purpose:
+ - normalize attributes
+ - check attributes for well-formedness
+ - generate namespace aware attribute names (URI, prefix)
+ - build list of attributes for startElementHandler
+ - default attributes
+ - process namespace declarations (check and report them)
+ - generate namespace aware element name (URI, prefix)
+*/
+static enum XML_Error
+storeAtts(XML_Parser parser, const ENCODING *enc,
+ const char *attStr, TAG_NAME *tagNamePtr,
+ BINDING **bindingsPtr)
+{
+ DTD * const dtd = _dtd; /* save one level of indirection */
+ ELEMENT_TYPE *elementType;
+ int nDefaultAtts;
+ const XML_Char **appAtts; /* the attribute list for the application */
+ int attIndex = 0;
+ int prefixLen;
+ int i;
+ int n;
+ XML_Char *uri;
+ int nPrefixes = 0;
+ BINDING *binding;
+ const XML_Char *localPart;
+
+ /* lookup the element type name */
+ elementType = (ELEMENT_TYPE *)lookup(&dtd->elementTypes, tagNamePtr->str,0);
+ if (!elementType) {
+ const XML_Char *name = poolCopyString(&dtd->pool, tagNamePtr->str);
+ if (!name)
+ return XML_ERROR_NO_MEMORY;
+ elementType = (ELEMENT_TYPE *)lookup(&dtd->elementTypes, name,
+ sizeof(ELEMENT_TYPE));
+ if (!elementType)
+ return XML_ERROR_NO_MEMORY;
+ if (ns && !setElementTypePrefix(parser, elementType))
+ return XML_ERROR_NO_MEMORY;
+ }
+ nDefaultAtts = elementType->nDefaultAtts;
+
+ /* get the attributes from the tokenizer */
+ n = XmlGetAttributes(enc, attStr, attsSize, atts);
+ if (n + nDefaultAtts > attsSize) {
+ int oldAttsSize = attsSize;
+ ATTRIBUTE *temp;
+ attsSize = n + nDefaultAtts + INIT_ATTS_SIZE;
+ temp = (ATTRIBUTE *)REALLOC((void *)atts, attsSize * sizeof(ATTRIBUTE));
+ if (temp == NULL)
+ return XML_ERROR_NO_MEMORY;
+ atts = temp;
+ if (n > oldAttsSize)
+ XmlGetAttributes(enc, attStr, n, atts);
+ }
+
+ appAtts = (const XML_Char **)atts;
+ for (i = 0; i < n; i++) {
+ /* add the name and value to the attribute list */
+ ATTRIBUTE_ID *attId = getAttributeId(parser, enc, atts[i].name,
+ atts[i].name
+ + XmlNameLength(enc, atts[i].name));
+ if (!attId)
+ return XML_ERROR_NO_MEMORY;
+ /* Detect duplicate attributes by their QNames. This does not work when
+ namespace processing is turned on and different prefixes for the same
+ namespace are used. For this case we have a check further down.
+ */
+ if ((attId->name)[-1]) {
+ if (enc == encoding)
+ eventPtr = atts[i].name;
+ return XML_ERROR_DUPLICATE_ATTRIBUTE;
+ }
+ (attId->name)[-1] = 1;
+ appAtts[attIndex++] = attId->name;
+ if (!atts[i].normalized) {
+ enum XML_Error result;
+ XML_Bool isCdata = XML_TRUE;
+
+ /* figure out whether declared as other than CDATA */
+ if (attId->maybeTokenized) {
+ int j;
+ for (j = 0; j < nDefaultAtts; j++) {
+ if (attId == elementType->defaultAtts[j].id) {
+ isCdata = elementType->defaultAtts[j].isCdata;
+ break;
+ }
+ }
+ }
+
+ /* normalize the attribute value */
+ result = storeAttributeValue(parser, enc, isCdata,
+ atts[i].valuePtr, atts[i].valueEnd,
+ &tempPool);
+ if (result)
+ return result;
+ appAtts[attIndex] = poolStart(&tempPool);
+ poolFinish(&tempPool);
+ }
+ else {
+ /* the value did not need normalizing */
+ appAtts[attIndex] = poolStoreString(&tempPool, enc, atts[i].valuePtr,
+ atts[i].valueEnd);
+ if (appAtts[attIndex] == 0)
+ return XML_ERROR_NO_MEMORY;
+ poolFinish(&tempPool);
+ }
+ /* handle prefixed attribute names */
+ if (attId->prefix) {
+ if (attId->xmlns) {
+ /* deal with namespace declarations here */
+ enum XML_Error result = addBinding(parser, attId->prefix, attId,
+ appAtts[attIndex], bindingsPtr);
+ if (result)
+ return result;
+ --attIndex;
+ }
+ else {
+ /* deal with other prefixed names later */
+ attIndex++;
+ nPrefixes++;
+ (attId->name)[-1] = 2;
+ }
+ }
+ else
+ attIndex++;
+ }
+
+ /* set-up for XML_GetSpecifiedAttributeCount and XML_GetIdAttributeIndex */
+ nSpecifiedAtts = attIndex;
+ if (elementType->idAtt && (elementType->idAtt->name)[-1]) {
+ for (i = 0; i < attIndex; i += 2)
+ if (appAtts[i] == elementType->idAtt->name) {
+ idAttIndex = i;
+ break;
+ }
+ }
+ else
+ idAttIndex = -1;
+
+ /* do attribute defaulting */
+ for (i = 0; i < nDefaultAtts; i++) {
+ const DEFAULT_ATTRIBUTE *da = elementType->defaultAtts + i;
+ if (!(da->id->name)[-1] && da->value) {
+ if (da->id->prefix) {
+ if (da->id->xmlns) {
+ enum XML_Error result = addBinding(parser, da->id->prefix, da->id,
+ da->value, bindingsPtr);
+ if (result)
+ return result;
+ }
+ else {
+ (da->id->name)[-1] = 2;
+ nPrefixes++;
+ appAtts[attIndex++] = da->id->name;
+ appAtts[attIndex++] = da->value;
+ }
+ }
+ else {
+ (da->id->name)[-1] = 1;
+ appAtts[attIndex++] = da->id->name;
+ appAtts[attIndex++] = da->value;
+ }
+ }
+ }
+ appAtts[attIndex] = 0;
+
+ /* expand prefixed attribute names, check for duplicates,
+ and clear flags that say whether attributes were specified */
+ i = 0;
+ if (nPrefixes) {
+ int j; /* hash table index */
+ unsigned long version = nsAttsVersion;
+ int nsAttsSize = (int)1 << nsAttsPower;
+ /* size of hash table must be at least 2 * (# of prefixed attributes) */
+ if ((nPrefixes << 1) >> nsAttsPower) { /* true for nsAttsPower = 0 */
+ NS_ATT *temp;
+ /* hash table size must also be a power of 2 and >= 8 */
+ while (nPrefixes >> nsAttsPower++);
+ if (nsAttsPower < 3)
+ nsAttsPower = 3;
+ nsAttsSize = (int)1 << nsAttsPower;
+ temp = (NS_ATT *)REALLOC(nsAtts, nsAttsSize * sizeof(NS_ATT));
+ if (!temp)
+ return XML_ERROR_NO_MEMORY;
+ nsAtts = temp;
+ version = 0; /* force re-initialization of nsAtts hash table */
+ }
+ /* using a version flag saves us from initializing nsAtts every time */
+ if (!version) { /* initialize version flags when version wraps around */
+ version = INIT_ATTS_VERSION;
+ for (j = nsAttsSize; j != 0; )
+ nsAtts[--j].version = version;
+ }
+ nsAttsVersion = --version;
+
+ /* expand prefixed names and check for duplicates */
+ for (; i < attIndex; i += 2) {
+ const XML_Char *s = appAtts[i];
+ if (s[-1] == 2) { /* prefixed */
+ ATTRIBUTE_ID *id;
+ const BINDING *b;
+ unsigned long uriHash = 0;
+ ((XML_Char *)s)[-1] = 0; /* clear flag */
+ id = (ATTRIBUTE_ID *)lookup(&dtd->attributeIds, s, 0);
+ b = id->prefix->binding;
+ if (!b)
+ return XML_ERROR_UNBOUND_PREFIX;
+
+ /* as we expand the name we also calculate its hash value */
+ for (j = 0; j < b->uriLen; j++) {
+ const XML_Char c = b->uri[j];
+ if (!poolAppendChar(&tempPool, c))
+ return XML_ERROR_NO_MEMORY;
+ uriHash = CHAR_HASH(uriHash, c);
+ }
+ while (*s++ != XML_T(':'))
+ ;
+ do { /* copies null terminator */
+ const XML_Char c = *s;
+ if (!poolAppendChar(&tempPool, *s))
+ return XML_ERROR_NO_MEMORY;
+ uriHash = CHAR_HASH(uriHash, c);
+ } while (*s++);
+
+ { /* Check hash table for duplicate of expanded name (uriName).
+ Derived from code in lookup(HASH_TABLE *table, ...).
+ */
+ unsigned char step = 0;
+ unsigned long mask = nsAttsSize - 1;
+ j = uriHash & mask; /* index into hash table */
+ while (nsAtts[j].version == version) {
+ /* for speed we compare stored hash values first */
+ if (uriHash == nsAtts[j].hash) {
+ const XML_Char *s1 = poolStart(&tempPool);
+ const XML_Char *s2 = nsAtts[j].uriName;
+ /* s1 is null terminated, but not s2 */
+ for (; *s1 == *s2 && *s1 != 0; s1++, s2++);
+ if (*s1 == 0)
+ return XML_ERROR_DUPLICATE_ATTRIBUTE;
+ }
+ if (!step)
+ step = PROBE_STEP(uriHash, mask, nsAttsPower);
+ j < step ? (j += nsAttsSize - step) : (j -= step);
+ }
+ }
+
+ if (ns_triplets) { /* append namespace separator and prefix */
+ tempPool.ptr[-1] = namespaceSeparator;
+ s = b->prefix->name;
+ do {
+ if (!poolAppendChar(&tempPool, *s))
+ return XML_ERROR_NO_MEMORY;
+ } while (*s++);
+ }
+
+ /* store expanded name in attribute list */
+ s = poolStart(&tempPool);
+ poolFinish(&tempPool);
+ appAtts[i] = s;
+
+ /* fill empty slot with new version, uriName and hash value */
+ nsAtts[j].version = version;
+ nsAtts[j].hash = uriHash;
+ nsAtts[j].uriName = s;
+
+ if (!--nPrefixes) {
+ i += 2;
+ break;
+ }
+ }
+ else /* not prefixed */
+ ((XML_Char *)s)[-1] = 0; /* clear flag */
+ }
+ }
+ /* clear flags for the remaining attributes */
+ for (; i < attIndex; i += 2)
+ ((XML_Char *)(appAtts[i]))[-1] = 0;
+ for (binding = *bindingsPtr; binding; binding = binding->nextTagBinding)
+ binding->attId->name[-1] = 0;
+
+ if (!ns)
+ return XML_ERROR_NONE;
+
+ /* expand the element type name */
+ if (elementType->prefix) {
+ binding = elementType->prefix->binding;
+ if (!binding)
+ return XML_ERROR_UNBOUND_PREFIX;
+ localPart = tagNamePtr->str;
+ while (*localPart++ != XML_T(':'))
+ ;
+ }
+ else if (dtd->defaultPrefix.binding) {
+ binding = dtd->defaultPrefix.binding;
+ localPart = tagNamePtr->str;
+ }
+ else
+ return XML_ERROR_NONE;
+ prefixLen = 0;
+ if (ns_triplets && binding->prefix->name) {
+ for (; binding->prefix->name[prefixLen++];)
+ ; /* prefixLen includes null terminator */
+ }
+ tagNamePtr->localPart = localPart;
+ tagNamePtr->uriLen = binding->uriLen;
+ tagNamePtr->prefix = binding->prefix->name;
+ tagNamePtr->prefixLen = prefixLen;
+ for (i = 0; localPart[i++];)
+ ; /* i includes null terminator */
+ n = i + binding->uriLen + prefixLen;
+ if (n > binding->uriAlloc) {
+ TAG *p;
+ uri = (XML_Char *)MALLOC((n + EXPAND_SPARE) * sizeof(XML_Char));
+ if (!uri)
+ return XML_ERROR_NO_MEMORY;
+ binding->uriAlloc = n + EXPAND_SPARE;
+ memcpy(uri, binding->uri, binding->uriLen * sizeof(XML_Char));
+ for (p = tagStack; p; p = p->parent)
+ if (p->name.str == binding->uri)
+ p->name.str = uri;
+ FREE(binding->uri);
+ binding->uri = uri;
+ }
+ /* if namespaceSeparator != '\0' then uri includes it already */
+ uri = binding->uri + binding->uriLen;
+ memcpy(uri, localPart, i * sizeof(XML_Char));
+ /* we always have a namespace separator between localPart and prefix */
+ if (prefixLen) {
+ uri += i - 1;
+ *uri = namespaceSeparator; /* replace null terminator */
+ memcpy(uri + 1, binding->prefix->name, prefixLen * sizeof(XML_Char));
+ }
+ tagNamePtr->str = binding->uri;
+ return XML_ERROR_NONE;
+}
+
+/* addBinding() overwrites the value of prefix->binding without checking.
+ Therefore one must keep track of the old value outside of addBinding().
+*/
+static enum XML_Error
+addBinding(XML_Parser parser, PREFIX *prefix, const ATTRIBUTE_ID *attId,
+ const XML_Char *uri, BINDING **bindingsPtr)
+{
+ static const XML_Char xmlNamespace[] = {
+ 'h', 't', 't', 'p', ':', '/', '/',
+ 'w', 'w', 'w', '.', 'w', '3', '.', 'o', 'r', 'g', '/',
+ 'X', 'M', 'L', '/', '1', '9', '9', '8', '/',
+ 'n', 'a', 'm', 'e', 's', 'p', 'a', 'c', 'e', '\0'
+ };
+ static const int xmlLen =
+ (int)sizeof(xmlNamespace)/sizeof(XML_Char) - 1;
+ static const XML_Char xmlnsNamespace[] = {
+ 'h', 't', 't', 'p', ':', '/', '/',
+ 'w', 'w', 'w', '.', 'w', '3', '.', 'o', 'r', 'g', '/',
+ '2', '0', '0', '0', '/', 'x', 'm', 'l', 'n', 's', '/', '\0'
+ };
+ static const int xmlnsLen =
+ (int)sizeof(xmlnsNamespace)/sizeof(XML_Char) - 1;
+
+ XML_Bool mustBeXML = XML_FALSE;
+ XML_Bool isXML = XML_TRUE;
+ XML_Bool isXMLNS = XML_TRUE;
+
+ BINDING *b;
+ int len;
+
+ /* empty URI is only valid for default namespace per XML NS 1.0 (not 1.1) */
+ if (*uri == XML_T('\0') && prefix->name)
+ return XML_ERROR_UNDECLARING_PREFIX;
+
+ if (prefix->name
+ && prefix->name[0] == XML_T('x')
+ && prefix->name[1] == XML_T('m')
+ && prefix->name[2] == XML_T('l')) {
+
+ /* Not allowed to bind xmlns */
+ if (prefix->name[3] == XML_T('n')
+ && prefix->name[4] == XML_T('s')
+ && prefix->name[5] == XML_T('\0'))
+ return XML_ERROR_RESERVED_PREFIX_XMLNS;
+
+ if (prefix->name[3] == XML_T('\0'))
+ mustBeXML = XML_TRUE;
+ }
+
+ for (len = 0; uri[len]; len++) {
+ if (isXML && (len > xmlLen || uri[len] != xmlNamespace[len]))
+ isXML = XML_FALSE;
+
+ if (!mustBeXML && isXMLNS
+ && (len > xmlnsLen || uri[len] != xmlnsNamespace[len]))
+ isXMLNS = XML_FALSE;
+ }
+ isXML = isXML && len == xmlLen;
+ isXMLNS = isXMLNS && len == xmlnsLen;
+
+ if (mustBeXML != isXML)
+ return mustBeXML ? XML_ERROR_RESERVED_PREFIX_XML
+ : XML_ERROR_RESERVED_NAMESPACE_URI;
+
+ if (isXMLNS)
+ return XML_ERROR_RESERVED_NAMESPACE_URI;
+
+ if (namespaceSeparator)
+ len++;
+ if (freeBindingList) {
+ b = freeBindingList;
+ if (len > b->uriAlloc) {
+ XML_Char *temp = (XML_Char *)REALLOC(b->uri,
+ sizeof(XML_Char) * (len + EXPAND_SPARE));
+ if (temp == NULL)
+ return XML_ERROR_NO_MEMORY;
+ b->uri = temp;
+ b->uriAlloc = len + EXPAND_SPARE;
+ }
+ freeBindingList = b->nextTagBinding;
+ }
+ else {
+ b = (BINDING *)MALLOC(sizeof(BINDING));
+ if (!b)
+ return XML_ERROR_NO_MEMORY;
+ b->uri = (XML_Char *)MALLOC(sizeof(XML_Char) * (len + EXPAND_SPARE));
+ if (!b->uri) {
+ FREE(b);
+ return XML_ERROR_NO_MEMORY;
+ }
+ b->uriAlloc = len + EXPAND_SPARE;
+ }
+ b->uriLen = len;
+ memcpy(b->uri, uri, len * sizeof(XML_Char));
+ if (namespaceSeparator)
+ b->uri[len - 1] = namespaceSeparator;
+ b->prefix = prefix;
+ b->attId = attId;
+ b->prevPrefixBinding = prefix->binding;
+ /* NULL binding when default namespace undeclared */
+ if (*uri == XML_T('\0') && prefix == &_dtd->defaultPrefix)
+ prefix->binding = NULL;
+ else
+ prefix->binding = b;
+ b->nextTagBinding = *bindingsPtr;
+ *bindingsPtr = b;
+ /* if attId == NULL then we are not starting a namespace scope */
+ if (attId && startNamespaceDeclHandler)
+ startNamespaceDeclHandler(handlerArg, prefix->name,
+ prefix->binding ? uri : 0);
+ return XML_ERROR_NONE;
+}
+
+/* The idea here is to avoid using stack for each CDATA section when
+ the whole file is parsed with one call.
+*/
+static enum XML_Error PTRCALL
+cdataSectionProcessor(XML_Parser parser,
+ const char *start,
+ const char *end,
+ const char **endPtr)
+{
+ enum XML_Error result = doCdataSection(parser, encoding, &start, end,
+ endPtr, (XML_Bool)!ps_finalBuffer);
+ if (result != XML_ERROR_NONE)
+ return result;
+ if (start) {
+ if (parentParser) { /* we are parsing an external entity */
+ processor = externalEntityContentProcessor;
+ return externalEntityContentProcessor(parser, start, end, endPtr);
+ }
+ else {
+ processor = contentProcessor;
+ return contentProcessor(parser, start, end, endPtr);
+ }
+ }
+ return result;
+}
+
+/* startPtr gets set to non-null if the section is closed, and to null if
+ the section is not yet closed.
+*/
+static enum XML_Error
+doCdataSection(XML_Parser parser,
+ const ENCODING *enc,
+ const char **startPtr,
+ const char *end,
+ const char **nextPtr,
+ XML_Bool haveMore)
+{
+ const char *s = *startPtr;
+ const char **eventPP;
+ const char **eventEndPP;
+ if (enc == encoding) {
+ eventPP = &eventPtr;
+ *eventPP = s;
+ eventEndPP = &eventEndPtr;
+ }
+ else {
+ eventPP = &(openInternalEntities->internalEventPtr);
+ eventEndPP = &(openInternalEntities->internalEventEndPtr);
+ }
+ *eventPP = s;
+ *startPtr = NULL;
+
+ for (;;) {
+ const char *next;
+ int tok = XmlCdataSectionTok(enc, s, end, &next);
+ *eventEndPP = next;
+ switch (tok) {
+ case XML_TOK_CDATA_SECT_CLOSE:
+ if (endCdataSectionHandler)
+ endCdataSectionHandler(handlerArg);
+#if 0
+ /* see comment under XML_TOK_CDATA_SECT_OPEN */
+ else if (characterDataHandler)
+ characterDataHandler(handlerArg, dataBuf, 0);
+#endif
+ else if (defaultHandler)
+ reportDefault(parser, enc, s, next);
+ *startPtr = next;
+ *nextPtr = next;
+ if (ps_parsing == XML_FINISHED)
+ return XML_ERROR_ABORTED;
+ else
+ return XML_ERROR_NONE;
+ case XML_TOK_DATA_NEWLINE:
+ if (characterDataHandler) {
+ XML_Char c = 0xA;
+ characterDataHandler(handlerArg, &c, 1);
+ }
+ else if (defaultHandler)
+ reportDefault(parser, enc, s, next);
+ break;
+ case XML_TOK_DATA_CHARS:
+ if (characterDataHandler) {
+ if (MUST_CONVERT(enc, s)) {
+ for (;;) {
+ ICHAR *dataPtr = (ICHAR *)dataBuf;
+ XmlConvert(enc, &s, next, &dataPtr, (ICHAR *)dataBufEnd);
+ *eventEndPP = next;
+ characterDataHandler(handlerArg, dataBuf,
+ (int)(dataPtr - (ICHAR *)dataBuf));
+ if (s == next)
+ break;
+ *eventPP = s;
+ }
+ }
+ else
+ characterDataHandler(handlerArg,
+ (XML_Char *)s,
+ (int)((XML_Char *)next - (XML_Char *)s));
+ }
+ else if (defaultHandler)
+ reportDefault(parser, enc, s, next);
+ break;
+ case XML_TOK_INVALID:
+ *eventPP = next;
+ return XML_ERROR_INVALID_TOKEN;
+ case XML_TOK_PARTIAL_CHAR:
+ if (haveMore) {
+ *nextPtr = s;
+ return XML_ERROR_NONE;
+ }
+ return XML_ERROR_PARTIAL_CHAR;
+ case XML_TOK_PARTIAL:
+ case XML_TOK_NONE:
+ if (haveMore) {
+ *nextPtr = s;
+ return XML_ERROR_NONE;
+ }
+ return XML_ERROR_UNCLOSED_CDATA_SECTION;
+ default:
+ *eventPP = next;
+ return XML_ERROR_UNEXPECTED_STATE;
+ }
+
+ *eventPP = s = next;
+ switch (ps_parsing) {
+ case XML_SUSPENDED:
+ *nextPtr = next;
+ return XML_ERROR_NONE;
+ case XML_FINISHED:
+ return XML_ERROR_ABORTED;
+ default: ;
+ }
+ }
+ /* not reached */
+}
+
+#ifdef XML_DTD
+
+/* The idea here is to avoid using stack for each IGNORE section when
+ the whole file is parsed with one call.
+*/
+static enum XML_Error PTRCALL
+ignoreSectionProcessor(XML_Parser parser,
+ const char *start,
+ const char *end,
+ const char **endPtr)
+{
+ enum XML_Error result = doIgnoreSection(parser, encoding, &start, end,
+ endPtr, (XML_Bool)!ps_finalBuffer);
+ if (result != XML_ERROR_NONE)
+ return result;
+ if (start) {
+ processor = prologProcessor;
+ return prologProcessor(parser, start, end, endPtr);
+ }
+ return result;
+}
+
+/* startPtr gets set to non-null is the section is closed, and to null
+ if the section is not yet closed.
+*/
+static enum XML_Error
+doIgnoreSection(XML_Parser parser,
+ const ENCODING *enc,
+ const char **startPtr,
+ const char *end,
+ const char **nextPtr,
+ XML_Bool haveMore)
+{
+ const char *next;
+ int tok;
+ const char *s = *startPtr;
+ const char **eventPP;
+ const char **eventEndPP;
+ if (enc == encoding) {
+ eventPP = &eventPtr;
+ *eventPP = s;
+ eventEndPP = &eventEndPtr;
+ }
+ else {
+ eventPP = &(openInternalEntities->internalEventPtr);
+ eventEndPP = &(openInternalEntities->internalEventEndPtr);
+ }
+ *eventPP = s;
+ *startPtr = NULL;
+ tok = XmlIgnoreSectionTok(enc, s, end, &next);
+ *eventEndPP = next;
+ switch (tok) {
+ case XML_TOK_IGNORE_SECT:
+ if (defaultHandler)
+ reportDefault(parser, enc, s, next);
+ *startPtr = next;
+ *nextPtr = next;
+ if (ps_parsing == XML_FINISHED)
+ return XML_ERROR_ABORTED;
+ else
+ return XML_ERROR_NONE;
+ case XML_TOK_INVALID:
+ *eventPP = next;
+ return XML_ERROR_INVALID_TOKEN;
+ case XML_TOK_PARTIAL_CHAR:
+ if (haveMore) {
+ *nextPtr = s;
+ return XML_ERROR_NONE;
+ }
+ return XML_ERROR_PARTIAL_CHAR;
+ case XML_TOK_PARTIAL:
+ case XML_TOK_NONE:
+ if (haveMore) {
+ *nextPtr = s;
+ return XML_ERROR_NONE;
+ }
+ return XML_ERROR_SYNTAX; /* XML_ERROR_UNCLOSED_IGNORE_SECTION */
+ default:
+ *eventPP = next;
+ return XML_ERROR_UNEXPECTED_STATE;
+ }
+ /* not reached */
+}
+
+#endif /* XML_DTD */
+
+static enum XML_Error
+initializeEncoding(XML_Parser parser)
+{
+ const char *s;
+#ifdef XML_UNICODE
+ char encodingBuf[128];
+ if (!protocolEncodingName)
+ s = NULL;
+ else {
+ int i;
+ for (i = 0; protocolEncodingName[i]; i++) {
+ if (i == sizeof(encodingBuf) - 1
+ || (protocolEncodingName[i] & ~0x7f) != 0) {
+ encodingBuf[0] = '\0';
+ break;
+ }
+ encodingBuf[i] = (char)protocolEncodingName[i];
+ }
+ encodingBuf[i] = '\0';
+ s = encodingBuf;
+ }
+#else
+ s = protocolEncodingName;
+#endif
+ if ((ns ? XmlInitEncodingNS : XmlInitEncoding)(&initEncoding, &encoding, s))
+ return XML_ERROR_NONE;
+ return handleUnknownEncoding(parser, protocolEncodingName);
+}
+
+static enum XML_Error
+processXmlDecl(XML_Parser parser, int isGeneralTextEntity,
+ const char *s, const char *next)
+{
+ const char *encodingName = NULL;
+ const XML_Char *storedEncName = NULL;
+ const ENCODING *newEncoding = NULL;
+ const char *version = NULL;
+ const char *versionend;
+ const XML_Char *storedversion = NULL;
+ int standalone = -1;
+ if (!(ns
+ ? XmlParseXmlDeclNS
+ : XmlParseXmlDecl)(isGeneralTextEntity,
+ encoding,
+ s,
+ next,
+ &eventPtr,
+ &version,
+ &versionend,
+ &encodingName,
+ &newEncoding,
+ &standalone)) {
+ if (isGeneralTextEntity)
+ return XML_ERROR_TEXT_DECL;
+ else
+ return XML_ERROR_XML_DECL;
+ }
+ if (!isGeneralTextEntity && standalone == 1) {
+ _dtd->standalone = XML_TRUE;
+#ifdef XML_DTD
+ if (paramEntityParsing == XML_PARAM_ENTITY_PARSING_UNLESS_STANDALONE)
+ paramEntityParsing = XML_PARAM_ENTITY_PARSING_NEVER;
+#endif /* XML_DTD */
+ }
+ if (xmlDeclHandler) {
+ if (encodingName != NULL) {
+ storedEncName = poolStoreString(&temp2Pool,
+ encoding,
+ encodingName,
+ encodingName
+ + XmlNameLength(encoding, encodingName));
+ if (!storedEncName)
+ return XML_ERROR_NO_MEMORY;
+ poolFinish(&temp2Pool);
+ }
+ if (version) {
+ storedversion = poolStoreString(&temp2Pool,
+ encoding,
+ version,
+ versionend - encoding->minBytesPerChar);
+ if (!storedversion)
+ return XML_ERROR_NO_MEMORY;
+ }
+ xmlDeclHandler(handlerArg, storedversion, storedEncName, standalone);
+ }
+ else if (defaultHandler)
+ reportDefault(parser, encoding, s, next);
+ if (protocolEncodingName == NULL) {
+ if (newEncoding) {
+ if (newEncoding->minBytesPerChar != encoding->minBytesPerChar) {
+ eventPtr = encodingName;
+ return XML_ERROR_INCORRECT_ENCODING;
+ }
+ encoding = newEncoding;
+ }
+ else if (encodingName) {
+ enum XML_Error result;
+ if (!storedEncName) {
+ storedEncName = poolStoreString(
+ &temp2Pool, encoding, encodingName,
+ encodingName + XmlNameLength(encoding, encodingName));
+ if (!storedEncName)
+ return XML_ERROR_NO_MEMORY;
+ }
+ result = handleUnknownEncoding(parser, storedEncName);
+ poolClear(&temp2Pool);
+ if (result == XML_ERROR_UNKNOWN_ENCODING)
+ eventPtr = encodingName;
+ return result;
+ }
+ }
+
+ if (storedEncName || storedversion)
+ poolClear(&temp2Pool);
+
+ return XML_ERROR_NONE;
+}
+
+static enum XML_Error
+handleUnknownEncoding(XML_Parser parser, const XML_Char *encodingName)
+{
+ if (unknownEncodingHandler) {
+ XML_Encoding info;
+ int i;
+ for (i = 0; i < 256; i++)
+ info.map[i] = -1;
+ info.convert = NULL;
+ info.data = NULL;
+ info.release = NULL;
+ if (unknownEncodingHandler(unknownEncodingHandlerData, encodingName,
+ &info)) {
+ ENCODING *enc;
+ unknownEncodingMem = MALLOC(XmlSizeOfUnknownEncoding());
+ if (!unknownEncodingMem) {
+ if (info.release)
+ info.release(info.data);
+ return XML_ERROR_NO_MEMORY;
+ }
+ enc = (ns
+ ? XmlInitUnknownEncodingNS
+ : XmlInitUnknownEncoding)(unknownEncodingMem,
+ info.map,
+ info.convert,
+ info.data);
+ if (enc) {
+ unknownEncodingData = info.data;
+ unknownEncodingRelease = info.release;
+ encoding = enc;
+ return XML_ERROR_NONE;
+ }
+ }
+ if (info.release != NULL)
+ info.release(info.data);
+ }
+ return XML_ERROR_UNKNOWN_ENCODING;
+}
+
+static enum XML_Error PTRCALL
+prologInitProcessor(XML_Parser parser,
+ const char *s,
+ const char *end,
+ const char **nextPtr)
+{
+ enum XML_Error result = initializeEncoding(parser);
+ if (result != XML_ERROR_NONE)
+ return result;
+ processor = prologProcessor;
+ return prologProcessor(parser, s, end, nextPtr);
+}
+
+#ifdef XML_DTD
+
+static enum XML_Error PTRCALL
+externalParEntInitProcessor(XML_Parser parser,
+ const char *s,
+ const char *end,
+ const char **nextPtr)
+{
+ enum XML_Error result = initializeEncoding(parser);
+ if (result != XML_ERROR_NONE)
+ return result;
+
+ /* we know now that XML_Parse(Buffer) has been called,
+ so we consider the external parameter entity read */
+ _dtd->paramEntityRead = XML_TRUE;
+
+ if (prologState.inEntityValue) {
+ processor = entityValueInitProcessor;
+ return entityValueInitProcessor(parser, s, end, nextPtr);
+ }
+ else {
+ processor = externalParEntProcessor;
+ return externalParEntProcessor(parser, s, end, nextPtr);
+ }
+}
+
+static enum XML_Error PTRCALL
+entityValueInitProcessor(XML_Parser parser,
+ const char *s,
+ const char *end,
+ const char **nextPtr)
+{
+ int tok;
+ const char *start = s;
+ const char *next = start;
+ eventPtr = start;
+
+ for (;;) {
+ tok = XmlPrologTok(encoding, start, end, &next);
+ eventEndPtr = next;
+ if (tok <= 0) {
+ if (!ps_finalBuffer && tok != XML_TOK_INVALID) {
+ *nextPtr = s;
+ return XML_ERROR_NONE;
+ }
+ switch (tok) {
+ case XML_TOK_INVALID:
+ return XML_ERROR_INVALID_TOKEN;
+ case XML_TOK_PARTIAL:
+ return XML_ERROR_UNCLOSED_TOKEN;
+ case XML_TOK_PARTIAL_CHAR:
+ return XML_ERROR_PARTIAL_CHAR;
+ case XML_TOK_NONE: /* start == end */
+ default:
+ break;
+ }
+ /* found end of entity value - can store it now */
+ return storeEntityValue(parser, encoding, s, end);
+ }
+ else if (tok == XML_TOK_XML_DECL) {
+ enum XML_Error result;
+ result = processXmlDecl(parser, 0, start, next);
+ if (result != XML_ERROR_NONE)
+ return result;
+ switch (ps_parsing) {
+ case XML_SUSPENDED:
+ *nextPtr = next;
+ return XML_ERROR_NONE;
+ case XML_FINISHED:
+ return XML_ERROR_ABORTED;
+ default:
+ *nextPtr = next;
+ }
+ /* stop scanning for text declaration - we found one */
+ processor = entityValueProcessor;
+ return entityValueProcessor(parser, next, end, nextPtr);
+ }
+ /* If we are at the end of the buffer, this would cause XmlPrologTok to
+ return XML_TOK_NONE on the next call, which would then cause the
+ function to exit with *nextPtr set to s - that is what we want for other
+ tokens, but not for the BOM - we would rather like to skip it;
+ then, when this routine is entered the next time, XmlPrologTok will
+ return XML_TOK_INVALID, since the BOM is still in the buffer
+ */
+ else if (tok == XML_TOK_BOM && next == end && !ps_finalBuffer) {
+ *nextPtr = next;
+ return XML_ERROR_NONE;
+ }
+ start = next;
+ eventPtr = start;
+ }
+}
+
+static enum XML_Error PTRCALL
+externalParEntProcessor(XML_Parser parser,
+ const char *s,
+ const char *end,
+ const char **nextPtr)
+{
+ const char *next = s;
+ int tok;
+
+ tok = XmlPrologTok(encoding, s, end, &next);
+ if (tok <= 0) {
+ if (!ps_finalBuffer && tok != XML_TOK_INVALID) {
+ *nextPtr = s;
+ return XML_ERROR_NONE;
+ }
+ switch (tok) {
+ case XML_TOK_INVALID:
+ return XML_ERROR_INVALID_TOKEN;
+ case XML_TOK_PARTIAL:
+ return XML_ERROR_UNCLOSED_TOKEN;
+ case XML_TOK_PARTIAL_CHAR:
+ return XML_ERROR_PARTIAL_CHAR;
+ case XML_TOK_NONE: /* start == end */
+ default:
+ break;
+ }
+ }
+ /* This would cause the next stage, i.e. doProlog to be passed XML_TOK_BOM.
+ However, when parsing an external subset, doProlog will not accept a BOM
+ as valid, and report a syntax error, so we have to skip the BOM
+ */
+ else if (tok == XML_TOK_BOM) {
+ s = next;
+ tok = XmlPrologTok(encoding, s, end, &next);
+ }
+
+ processor = prologProcessor;
+ return doProlog(parser, encoding, s, end, tok, next,
+ nextPtr, (XML_Bool)!ps_finalBuffer);
+}
+
+static enum XML_Error PTRCALL
+entityValueProcessor(XML_Parser parser,
+ const char *s,
+ const char *end,
+ const char **nextPtr)
+{
+ const char *start = s;
+ const char *next = s;
+ const ENCODING *enc = encoding;
+ int tok;
+
+ for (;;) {
+ tok = XmlPrologTok(enc, start, end, &next);
+ if (tok <= 0) {
+ if (!ps_finalBuffer && tok != XML_TOK_INVALID) {
+ *nextPtr = s;
+ return XML_ERROR_NONE;
+ }
+ switch (tok) {
+ case XML_TOK_INVALID:
+ return XML_ERROR_INVALID_TOKEN;
+ case XML_TOK_PARTIAL:
+ return XML_ERROR_UNCLOSED_TOKEN;
+ case XML_TOK_PARTIAL_CHAR:
+ return XML_ERROR_PARTIAL_CHAR;
+ case XML_TOK_NONE: /* start == end */
+ default:
+ break;
+ }
+ /* found end of entity value - can store it now */
+ return storeEntityValue(parser, enc, s, end);
+ }
+ start = next;
+ }
+}
+
+#endif /* XML_DTD */
+
+static enum XML_Error PTRCALL
+prologProcessor(XML_Parser parser,
+ const char *s,
+ const char *end,
+ const char **nextPtr)
+{
+ const char *next = s;
+ int tok = XmlPrologTok(encoding, s, end, &next);
+ return doProlog(parser, encoding, s, end, tok, next,
+ nextPtr, (XML_Bool)!ps_finalBuffer);
+}
+
+static enum XML_Error
+doProlog(XML_Parser parser,
+ const ENCODING *enc,
+ const char *s,
+ const char *end,
+ int tok,
+ const char *next,
+ const char **nextPtr,
+ XML_Bool haveMore)
+{
+#ifdef XML_DTD
+ static const XML_Char externalSubsetName[] = { '#' , '\0' };
+#endif /* XML_DTD */
+ static const XML_Char atypeCDATA[] = { 'C', 'D', 'A', 'T', 'A', '\0' };
+ static const XML_Char atypeID[] = { 'I', 'D', '\0' };
+ static const XML_Char atypeIDREF[] = { 'I', 'D', 'R', 'E', 'F', '\0' };
+ static const XML_Char atypeIDREFS[] = { 'I', 'D', 'R', 'E', 'F', 'S', '\0' };
+ static const XML_Char atypeENTITY[] = { 'E', 'N', 'T', 'I', 'T', 'Y', '\0' };
+ static const XML_Char atypeENTITIES[] =
+ { 'E', 'N', 'T', 'I', 'T', 'I', 'E', 'S', '\0' };
+ static const XML_Char atypeNMTOKEN[] = {
+ 'N', 'M', 'T', 'O', 'K', 'E', 'N', '\0' };
+ static const XML_Char atypeNMTOKENS[] = {
+ 'N', 'M', 'T', 'O', 'K', 'E', 'N', 'S', '\0' };
+ static const XML_Char notationPrefix[] = {
+ 'N', 'O', 'T', 'A', 'T', 'I', 'O', 'N', '(', '\0' };
+ static const XML_Char enumValueSep[] = { '|', '\0' };
+ static const XML_Char enumValueStart[] = { '(', '\0' };
+
+ /* save one level of indirection */
+ DTD * const dtd = _dtd;
+
+ const char **eventPP;
+ const char **eventEndPP;
+ enum XML_Content_Quant quant;
+
+ if (enc == encoding) {
+ eventPP = &eventPtr;
+ eventEndPP = &eventEndPtr;
+ }
+ else {
+ eventPP = &(openInternalEntities->internalEventPtr);
+ eventEndPP = &(openInternalEntities->internalEventEndPtr);
+ }
+
+ for (;;) {
+ int role;
+ XML_Bool handleDefault = XML_TRUE;
+ *eventPP = s;
+ *eventEndPP = next;
+ if (tok <= 0) {
+ if (haveMore && tok != XML_TOK_INVALID) {
+ *nextPtr = s;
+ return XML_ERROR_NONE;
+ }
+ switch (tok) {
+ case XML_TOK_INVALID:
+ *eventPP = next;
+ return XML_ERROR_INVALID_TOKEN;
+ case XML_TOK_PARTIAL:
+ return XML_ERROR_UNCLOSED_TOKEN;
+ case XML_TOK_PARTIAL_CHAR:
+ return XML_ERROR_PARTIAL_CHAR;
+ case XML_TOK_NONE:
+#ifdef XML_DTD
+ /* for internal PE NOT referenced between declarations */
+ if (enc != encoding && !openInternalEntities->betweenDecl) {
+ *nextPtr = s;
+ return XML_ERROR_NONE;
+ }
+ /* WFC: PE Between Declarations - must check that PE contains
+ complete markup, not only for external PEs, but also for
+ internal PEs if the reference occurs between declarations.
+ */
+ if (isParamEntity || enc != encoding) {
+ if (XmlTokenRole(&prologState, XML_TOK_NONE, end, end, enc)
+ == XML_ROLE_ERROR)
+ return XML_ERROR_INCOMPLETE_PE;
+ *nextPtr = s;
+ return XML_ERROR_NONE;
+ }
+#endif /* XML_DTD */
+ return XML_ERROR_NO_ELEMENTS;
+ default:
+ tok = -tok;
+ next = end;
+ break;
+ }
+ }
+ role = XmlTokenRole(&prologState, tok, s, next, enc);
+ switch (role) {
+ case XML_ROLE_XML_DECL:
+ {
+ enum XML_Error result = processXmlDecl(parser, 0, s, next);
+ if (result != XML_ERROR_NONE)
+ return result;
+ enc = encoding;
+ handleDefault = XML_FALSE;
+ }
+ break;
+ case XML_ROLE_DOCTYPE_NAME:
+ if (startDoctypeDeclHandler) {
+ doctypeName = poolStoreString(&tempPool, enc, s, next);
+ if (!doctypeName)
+ return XML_ERROR_NO_MEMORY;
+ poolFinish(&tempPool);
+ doctypePubid = NULL;
+ handleDefault = XML_FALSE;
+ }
+ doctypeSysid = NULL; /* always initialize to NULL */
+ break;
+ case XML_ROLE_DOCTYPE_INTERNAL_SUBSET:
+ if (startDoctypeDeclHandler) {
+ startDoctypeDeclHandler(handlerArg, doctypeName, doctypeSysid,
+ doctypePubid, 1);
+ doctypeName = NULL;
+ poolClear(&tempPool);
+ handleDefault = XML_FALSE;
+ }
+ break;
+#ifdef XML_DTD
+ case XML_ROLE_TEXT_DECL:
+ {
+ enum XML_Error result = processXmlDecl(parser, 1, s, next);
+ if (result != XML_ERROR_NONE)
+ return result;
+ enc = encoding;
+ handleDefault = XML_FALSE;
+ }
+ break;
+#endif /* XML_DTD */
+ case XML_ROLE_DOCTYPE_PUBLIC_ID:
+#ifdef XML_DTD
+ useForeignDTD = XML_FALSE;
+ declEntity = (ENTITY *)lookup(&dtd->paramEntities,
+ externalSubsetName,
+ sizeof(ENTITY));
+ if (!declEntity)
+ return XML_ERROR_NO_MEMORY;
+#endif /* XML_DTD */
+ dtd->hasParamEntityRefs = XML_TRUE;
+ if (startDoctypeDeclHandler) {
+ XML_Char *tem;
+ if (!XmlIsPublicId(enc, s, next, eventPP))
+ return XML_ERROR_PUBLICID;
+ tem = poolStoreString(&tempPool, enc,
+ s + enc->minBytesPerChar,
+ next - enc->minBytesPerChar);
+ if (!tem)
+ return XML_ERROR_NO_MEMORY;
+ normalizePublicId(tem);
+ doctypePubid = tem;
+ poolFinish(&tempPool);
+ handleDefault = XML_FALSE;
+ goto alreadyChecked;
+ }
+ /* fall through */
+ case XML_ROLE_ENTITY_PUBLIC_ID:
+ if (!XmlIsPublicId(enc, s, next, eventPP))
+ return XML_ERROR_PUBLICID;
+ alreadyChecked:
+ if (dtd->keepProcessing && declEntity) {
+ XML_Char *tem = poolStoreString(&dtd->pool,
+ enc,
+ s + enc->minBytesPerChar,
+ next - enc->minBytesPerChar);
+ if (!tem)
+ return XML_ERROR_NO_MEMORY;
+ normalizePublicId(tem);
+ declEntity->publicId = tem;
+ poolFinish(&dtd->pool);
+ if (entityDeclHandler)
+ handleDefault = XML_FALSE;
+ }
+ break;
+ case XML_ROLE_DOCTYPE_CLOSE:
+ if (doctypeName) {
+ startDoctypeDeclHandler(handlerArg, doctypeName,
+ doctypeSysid, doctypePubid, 0);
+ poolClear(&tempPool);
+ handleDefault = XML_FALSE;
+ }
+ /* doctypeSysid will be non-NULL in the case of a previous
+ XML_ROLE_DOCTYPE_SYSTEM_ID, even if startDoctypeDeclHandler
+ was not set, indicating an external subset
+ */
+#ifdef XML_DTD
+ if (doctypeSysid || useForeignDTD) {
+ XML_Bool hadParamEntityRefs = dtd->hasParamEntityRefs;
+ dtd->hasParamEntityRefs = XML_TRUE;
+ if (paramEntityParsing && externalEntityRefHandler) {
+ ENTITY *entity = (ENTITY *)lookup(&dtd->paramEntities,
+ externalSubsetName,
+ sizeof(ENTITY));
+ if (!entity)
+ return XML_ERROR_NO_MEMORY;
+ if (useForeignDTD)
+ entity->base = curBase;
+ dtd->paramEntityRead = XML_FALSE;
+ if (!externalEntityRefHandler(externalEntityRefHandlerArg,
+ 0,
+ entity->base,
+ entity->systemId,
+ entity->publicId))
+ return XML_ERROR_EXTERNAL_ENTITY_HANDLING;
+ if (dtd->paramEntityRead) {
+ if (!dtd->standalone &&
+ notStandaloneHandler &&
+ !notStandaloneHandler(handlerArg))
+ return XML_ERROR_NOT_STANDALONE;
+ }
+ /* if we didn't read the foreign DTD then this means that there
+ is no external subset and we must reset dtd->hasParamEntityRefs
+ */
+ else if (!doctypeSysid)
+ dtd->hasParamEntityRefs = hadParamEntityRefs;
+ /* end of DTD - no need to update dtd->keepProcessing */
+ }
+ useForeignDTD = XML_FALSE;
+ }
+#endif /* XML_DTD */
+ if (endDoctypeDeclHandler) {
+ endDoctypeDeclHandler(handlerArg);
+ handleDefault = XML_FALSE;
+ }
+ break;
+ case XML_ROLE_INSTANCE_START:
+#ifdef XML_DTD
+ /* if there is no DOCTYPE declaration then now is the
+ last chance to read the foreign DTD
+ */
+ if (useForeignDTD) {
+ XML_Bool hadParamEntityRefs = dtd->hasParamEntityRefs;
+ dtd->hasParamEntityRefs = XML_TRUE;
+ if (paramEntityParsing && externalEntityRefHandler) {
+ ENTITY *entity = (ENTITY *)lookup(&dtd->paramEntities,
+ externalSubsetName,
+ sizeof(ENTITY));
+ if (!entity)
+ return XML_ERROR_NO_MEMORY;
+ entity->base = curBase;
+ dtd->paramEntityRead = XML_FALSE;
+ if (!externalEntityRefHandler(externalEntityRefHandlerArg,
+ 0,
+ entity->base,
+ entity->systemId,
+ entity->publicId))
+ return XML_ERROR_EXTERNAL_ENTITY_HANDLING;
+ if (dtd->paramEntityRead) {
+ if (!dtd->standalone &&
+ notStandaloneHandler &&
+ !notStandaloneHandler(handlerArg))
+ return XML_ERROR_NOT_STANDALONE;
+ }
+ /* if we didn't read the foreign DTD then this means that there
+ is no external subset and we must reset dtd->hasParamEntityRefs
+ */
+ else
+ dtd->hasParamEntityRefs = hadParamEntityRefs;
+ /* end of DTD - no need to update dtd->keepProcessing */
+ }
+ }
+#endif /* XML_DTD */
+ processor = contentProcessor;
+ return contentProcessor(parser, s, end, nextPtr);
+ case XML_ROLE_ATTLIST_ELEMENT_NAME:
+ declElementType = getElementType(parser, enc, s, next);
+ if (!declElementType)
+ return XML_ERROR_NO_MEMORY;
+ goto checkAttListDeclHandler;
+ case XML_ROLE_ATTRIBUTE_NAME:
+ declAttributeId = getAttributeId(parser, enc, s, next);
+ if (!declAttributeId)
+ return XML_ERROR_NO_MEMORY;
+ declAttributeIsCdata = XML_FALSE;
+ declAttributeType = NULL;
+ declAttributeIsId = XML_FALSE;
+ goto checkAttListDeclHandler;
+ case XML_ROLE_ATTRIBUTE_TYPE_CDATA:
+ declAttributeIsCdata = XML_TRUE;
+ declAttributeType = atypeCDATA;
+ goto checkAttListDeclHandler;
+ case XML_ROLE_ATTRIBUTE_TYPE_ID:
+ declAttributeIsId = XML_TRUE;
+ declAttributeType = atypeID;
+ goto checkAttListDeclHandler;
+ case XML_ROLE_ATTRIBUTE_TYPE_IDREF:
+ declAttributeType = atypeIDREF;
+ goto checkAttListDeclHandler;
+ case XML_ROLE_ATTRIBUTE_TYPE_IDREFS:
+ declAttributeType = atypeIDREFS;
+ goto checkAttListDeclHandler;
+ case XML_ROLE_ATTRIBUTE_TYPE_ENTITY:
+ declAttributeType = atypeENTITY;
+ goto checkAttListDeclHandler;
+ case XML_ROLE_ATTRIBUTE_TYPE_ENTITIES:
+ declAttributeType = atypeENTITIES;
+ goto checkAttListDeclHandler;
+ case XML_ROLE_ATTRIBUTE_TYPE_NMTOKEN:
+ declAttributeType = atypeNMTOKEN;
+ goto checkAttListDeclHandler;
+ case XML_ROLE_ATTRIBUTE_TYPE_NMTOKENS:
+ declAttributeType = atypeNMTOKENS;
+ checkAttListDeclHandler:
+ if (dtd->keepProcessing && attlistDeclHandler)
+ handleDefault = XML_FALSE;
+ break;
+ case XML_ROLE_ATTRIBUTE_ENUM_VALUE:
+ case XML_ROLE_ATTRIBUTE_NOTATION_VALUE:
+ if (dtd->keepProcessing && attlistDeclHandler) {
+ const XML_Char *prefix;
+ if (declAttributeType) {
+ prefix = enumValueSep;
+ }
+ else {
+ prefix = (role == XML_ROLE_ATTRIBUTE_NOTATION_VALUE
+ ? notationPrefix
+ : enumValueStart);
+ }
+ if (!poolAppendString(&tempPool, prefix))
+ return XML_ERROR_NO_MEMORY;
+ if (!poolAppend(&tempPool, enc, s, next))
+ return XML_ERROR_NO_MEMORY;
+ declAttributeType = tempPool.start;
+ handleDefault = XML_FALSE;
+ }
+ break;
+ case XML_ROLE_IMPLIED_ATTRIBUTE_VALUE:
+ case XML_ROLE_REQUIRED_ATTRIBUTE_VALUE:
+ if (dtd->keepProcessing) {
+ if (!defineAttribute(declElementType, declAttributeId,
+ declAttributeIsCdata, declAttributeIsId,
+ 0, parser))
+ return XML_ERROR_NO_MEMORY;
+ if (attlistDeclHandler && declAttributeType) {
+ if (*declAttributeType == XML_T('(')
+ || (*declAttributeType == XML_T('N')
+ && declAttributeType[1] == XML_T('O'))) {
+ /* Enumerated or Notation type */
+ if (!poolAppendChar(&tempPool, XML_T(')'))
+ || !poolAppendChar(&tempPool, XML_T('\0')))
+ return XML_ERROR_NO_MEMORY;
+ declAttributeType = tempPool.start;
+ poolFinish(&tempPool);
+ }
+ *eventEndPP = s;
+ attlistDeclHandler(handlerArg, declElementType->name,
+ declAttributeId->name, declAttributeType,
+ 0, role == XML_ROLE_REQUIRED_ATTRIBUTE_VALUE);
+ poolClear(&tempPool);
+ handleDefault = XML_FALSE;
+ }
+ }
+ break;
+ case XML_ROLE_DEFAULT_ATTRIBUTE_VALUE:
+ case XML_ROLE_FIXED_ATTRIBUTE_VALUE:
+ if (dtd->keepProcessing) {
+ const XML_Char *attVal;
+ enum XML_Error result =
+ storeAttributeValue(parser, enc, declAttributeIsCdata,
+ s + enc->minBytesPerChar,
+ next - enc->minBytesPerChar,
+ &dtd->pool);
+ if (result)
+ return result;
+ attVal = poolStart(&dtd->pool);
+ poolFinish(&dtd->pool);
+ /* ID attributes aren't allowed to have a default */
+ if (!defineAttribute(declElementType, declAttributeId,
+ declAttributeIsCdata, XML_FALSE, attVal, parser))
+ return XML_ERROR_NO_MEMORY;
+ if (attlistDeclHandler && declAttributeType) {
+ if (*declAttributeType == XML_T('(')
+ || (*declAttributeType == XML_T('N')
+ && declAttributeType[1] == XML_T('O'))) {
+ /* Enumerated or Notation type */
+ if (!poolAppendChar(&tempPool, XML_T(')'))
+ || !poolAppendChar(&tempPool, XML_T('\0')))
+ return XML_ERROR_NO_MEMORY;
+ declAttributeType = tempPool.start;
+ poolFinish(&tempPool);
+ }
+ *eventEndPP = s;
+ attlistDeclHandler(handlerArg, declElementType->name,
+ declAttributeId->name, declAttributeType,
+ attVal,
+ role == XML_ROLE_FIXED_ATTRIBUTE_VALUE);
+ poolClear(&tempPool);
+ handleDefault = XML_FALSE;
+ }
+ }
+ break;
+ case XML_ROLE_ENTITY_VALUE:
+ if (dtd->keepProcessing) {
+ enum XML_Error result = storeEntityValue(parser, enc,
+ s + enc->minBytesPerChar,
+ next - enc->minBytesPerChar);
+ if (declEntity) {
+ declEntity->textPtr = poolStart(&dtd->entityValuePool);
+ declEntity->textLen = (int)(poolLength(&dtd->entityValuePool));
+ poolFinish(&dtd->entityValuePool);
+ if (entityDeclHandler) {
+ *eventEndPP = s;
+ entityDeclHandler(handlerArg,
+ declEntity->name,
+ declEntity->is_param,
+ declEntity->textPtr,
+ declEntity->textLen,
+ curBase, 0, 0, 0);
+ handleDefault = XML_FALSE;
+ }
+ }
+ else
+ poolDiscard(&dtd->entityValuePool);
+ if (result != XML_ERROR_NONE)
+ return result;
+ }
+ break;
+ case XML_ROLE_DOCTYPE_SYSTEM_ID:
+#ifdef XML_DTD
+ useForeignDTD = XML_FALSE;
+#endif /* XML_DTD */
+ dtd->hasParamEntityRefs = XML_TRUE;
+ if (startDoctypeDeclHandler) {
+ doctypeSysid = poolStoreString(&tempPool, enc,
+ s + enc->minBytesPerChar,
+ next - enc->minBytesPerChar);
+ if (doctypeSysid == NULL)
+ return XML_ERROR_NO_MEMORY;
+ poolFinish(&tempPool);
+ handleDefault = XML_FALSE;
+ }
+#ifdef XML_DTD
+ else
+ /* use externalSubsetName to make doctypeSysid non-NULL
+ for the case where no startDoctypeDeclHandler is set */
+ doctypeSysid = externalSubsetName;
+#endif /* XML_DTD */
+ if (!dtd->standalone
+#ifdef XML_DTD
+ && !paramEntityParsing
+#endif /* XML_DTD */
+ && notStandaloneHandler
+ && !notStandaloneHandler(handlerArg))
+ return XML_ERROR_NOT_STANDALONE;
+#ifndef XML_DTD
+ break;
+#else /* XML_DTD */
+ if (!declEntity) {
+ declEntity = (ENTITY *)lookup(&dtd->paramEntities,
+ externalSubsetName,
+ sizeof(ENTITY));
+ if (!declEntity)
+ return XML_ERROR_NO_MEMORY;
+ declEntity->publicId = NULL;
+ }
+ /* fall through */
+#endif /* XML_DTD */
+ case XML_ROLE_ENTITY_SYSTEM_ID:
+ if (dtd->keepProcessing && declEntity) {
+ declEntity->systemId = poolStoreString(&dtd->pool, enc,
+ s + enc->minBytesPerChar,
+ next - enc->minBytesPerChar);
+ if (!declEntity->systemId)
+ return XML_ERROR_NO_MEMORY;
+ declEntity->base = curBase;
+ poolFinish(&dtd->pool);
+ if (entityDeclHandler)
+ handleDefault = XML_FALSE;
+ }
+ break;
+ case XML_ROLE_ENTITY_COMPLETE:
+ if (dtd->keepProcessing && declEntity && entityDeclHandler) {
+ *eventEndPP = s;
+ entityDeclHandler(handlerArg,
+ declEntity->name,
+ declEntity->is_param,
+ 0,0,
+ declEntity->base,
+ declEntity->systemId,
+ declEntity->publicId,
+ 0);
+ handleDefault = XML_FALSE;
+ }
+ break;
+ case XML_ROLE_ENTITY_NOTATION_NAME:
+ if (dtd->keepProcessing && declEntity) {
+ declEntity->notation = poolStoreString(&dtd->pool, enc, s, next);
+ if (!declEntity->notation)
+ return XML_ERROR_NO_MEMORY;
+ poolFinish(&dtd->pool);
+ if (unparsedEntityDeclHandler) {
+ *eventEndPP = s;
+ unparsedEntityDeclHandler(handlerArg,
+ declEntity->name,
+ declEntity->base,
+ declEntity->systemId,
+ declEntity->publicId,
+ declEntity->notation);
+ handleDefault = XML_FALSE;
+ }
+ else if (entityDeclHandler) {
+ *eventEndPP = s;
+ entityDeclHandler(handlerArg,
+ declEntity->name,
+ 0,0,0,
+ declEntity->base,
+ declEntity->systemId,
+ declEntity->publicId,
+ declEntity->notation);
+ handleDefault = XML_FALSE;
+ }
+ }
+ break;
+ case XML_ROLE_GENERAL_ENTITY_NAME:
+ {
+ if (XmlPredefinedEntityName(enc, s, next)) {
+ declEntity = NULL;
+ break;
+ }
+ if (dtd->keepProcessing) {
+ const XML_Char *name = poolStoreString(&dtd->pool, enc, s, next);
+ if (!name)
+ return XML_ERROR_NO_MEMORY;
+ declEntity = (ENTITY *)lookup(&dtd->generalEntities, name,
+ sizeof(ENTITY));
+ if (!declEntity)
+ return XML_ERROR_NO_MEMORY;
+ if (declEntity->name != name) {
+ poolDiscard(&dtd->pool);
+ declEntity = NULL;
+ }
+ else {
+ poolFinish(&dtd->pool);
+ declEntity->publicId = NULL;
+ declEntity->is_param = XML_FALSE;
+ /* if we have a parent parser or are reading an internal parameter
+ entity, then the entity declaration is not considered "internal"
+ */
+ declEntity->is_internal = !(parentParser || openInternalEntities);
+ if (entityDeclHandler)
+ handleDefault = XML_FALSE;
+ }
+ }
+ else {
+ poolDiscard(&dtd->pool);
+ declEntity = NULL;
+ }
+ }
+ break;
+ case XML_ROLE_PARAM_ENTITY_NAME:
+#ifdef XML_DTD
+ if (dtd->keepProcessing) {
+ const XML_Char *name = poolStoreString(&dtd->pool, enc, s, next);
+ if (!name)
+ return XML_ERROR_NO_MEMORY;
+ declEntity = (ENTITY *)lookup(&dtd->paramEntities,
+ name, sizeof(ENTITY));
+ if (!declEntity)
+ return XML_ERROR_NO_MEMORY;
+ if (declEntity->name != name) {
+ poolDiscard(&dtd->pool);
+ declEntity = NULL;
+ }
+ else {
+ poolFinish(&dtd->pool);
+ declEntity->publicId = NULL;
+ declEntity->is_param = XML_TRUE;
+ /* if we have a parent parser or are reading an internal parameter
+ entity, then the entity declaration is not considered "internal"
+ */
+ declEntity->is_internal = !(parentParser || openInternalEntities);
+ if (entityDeclHandler)
+ handleDefault = XML_FALSE;
+ }
+ }
+ else {
+ poolDiscard(&dtd->pool);
+ declEntity = NULL;
+ }
+#else /* not XML_DTD */
+ declEntity = NULL;
+#endif /* XML_DTD */
+ break;
+ case XML_ROLE_NOTATION_NAME:
+ declNotationPublicId = NULL;
+ declNotationName = NULL;
+ if (notationDeclHandler) {
+ declNotationName = poolStoreString(&tempPool, enc, s, next);
+ if (!declNotationName)
+ return XML_ERROR_NO_MEMORY;
+ poolFinish(&tempPool);
+ handleDefault = XML_FALSE;
+ }
+ break;
+ case XML_ROLE_NOTATION_PUBLIC_ID:
+ if (!XmlIsPublicId(enc, s, next, eventPP))
+ return XML_ERROR_PUBLICID;
+ if (declNotationName) { /* means notationDeclHandler != NULL */
+ XML_Char *tem = poolStoreString(&tempPool,
+ enc,
+ s + enc->minBytesPerChar,
+ next - enc->minBytesPerChar);
+ if (!tem)
+ return XML_ERROR_NO_MEMORY;
+ normalizePublicId(tem);
+ declNotationPublicId = tem;
+ poolFinish(&tempPool);
+ handleDefault = XML_FALSE;
+ }
+ break;
+ case XML_ROLE_NOTATION_SYSTEM_ID:
+ if (declNotationName && notationDeclHandler) {
+ const XML_Char *systemId
+ = poolStoreString(&tempPool, enc,
+ s + enc->minBytesPerChar,
+ next - enc->minBytesPerChar);
+ if (!systemId)
+ return XML_ERROR_NO_MEMORY;
+ *eventEndPP = s;
+ notationDeclHandler(handlerArg,
+ declNotationName,
+ curBase,
+ systemId,
+ declNotationPublicId);
+ handleDefault = XML_FALSE;
+ }
+ poolClear(&tempPool);
+ break;
+ case XML_ROLE_NOTATION_NO_SYSTEM_ID:
+ if (declNotationPublicId && notationDeclHandler) {
+ *eventEndPP = s;
+ notationDeclHandler(handlerArg,
+ declNotationName,
+ curBase,
+ 0,
+ declNotationPublicId);
+ handleDefault = XML_FALSE;
+ }
+ poolClear(&tempPool);
+ break;
+ case XML_ROLE_ERROR:
+ switch (tok) {
+ case XML_TOK_PARAM_ENTITY_REF:
+ /* PE references in internal subset are
+ not allowed within declarations. */
+ return XML_ERROR_PARAM_ENTITY_REF;
+ case XML_TOK_XML_DECL:
+ return XML_ERROR_MISPLACED_XML_PI;
+ default:
+ return XML_ERROR_SYNTAX;
+ }
+#ifdef XML_DTD
+ case XML_ROLE_IGNORE_SECT:
+ {
+ enum XML_Error result;
+ if (defaultHandler)
+ reportDefault(parser, enc, s, next);
+ handleDefault = XML_FALSE;
+ result = doIgnoreSection(parser, enc, &next, end, nextPtr, haveMore);
+ if (result != XML_ERROR_NONE)
+ return result;
+ else if (!next) {
+ processor = ignoreSectionProcessor;
+ return result;
+ }
+ }
+ break;
+#endif /* XML_DTD */
+ case XML_ROLE_GROUP_OPEN:
+ if (prologState.level >= groupSize) {
+ if (groupSize) {
+ char *temp = (char *)REALLOC(groupConnector, groupSize *= 2);
+ if (temp == NULL)
+ return XML_ERROR_NO_MEMORY;
+ groupConnector = temp;
+ if (dtd->scaffIndex) {
+ int *temp = (int *)REALLOC(dtd->scaffIndex,
+ groupSize * sizeof(int));
+ if (temp == NULL)
+ return XML_ERROR_NO_MEMORY;
+ dtd->scaffIndex = temp;
+ }
+ }
+ else {
+ groupConnector = (char *)MALLOC(groupSize = 32);
+ if (!groupConnector)
+ return XML_ERROR_NO_MEMORY;
+ }
+ }
+ groupConnector[prologState.level] = 0;
+ if (dtd->in_eldecl) {
+ int myindex = nextScaffoldPart(parser);
+ if (myindex < 0)
+ return XML_ERROR_NO_MEMORY;
+ dtd->scaffIndex[dtd->scaffLevel] = myindex;
+ dtd->scaffLevel++;
+ dtd->scaffold[myindex].type = XML_CTYPE_SEQ;
+ if (elementDeclHandler)
+ handleDefault = XML_FALSE;
+ }
+ break;
+ case XML_ROLE_GROUP_SEQUENCE:
+ if (groupConnector[prologState.level] == '|')
+ return XML_ERROR_SYNTAX;
+ groupConnector[prologState.level] = ',';
+ if (dtd->in_eldecl && elementDeclHandler)
+ handleDefault = XML_FALSE;
+ break;
+ case XML_ROLE_GROUP_CHOICE:
+ if (groupConnector[prologState.level] == ',')
+ return XML_ERROR_SYNTAX;
+ if (dtd->in_eldecl
+ && !groupConnector[prologState.level]
+ && (dtd->scaffold[dtd->scaffIndex[dtd->scaffLevel - 1]].type
+ != XML_CTYPE_MIXED)
+ ) {
+ dtd->scaffold[dtd->scaffIndex[dtd->scaffLevel - 1]].type
+ = XML_CTYPE_CHOICE;
+ if (elementDeclHandler)
+ handleDefault = XML_FALSE;
+ }
+ groupConnector[prologState.level] = '|';
+ break;
+ case XML_ROLE_PARAM_ENTITY_REF:
+#ifdef XML_DTD
+ case XML_ROLE_INNER_PARAM_ENTITY_REF:
+ dtd->hasParamEntityRefs = XML_TRUE;
+ if (!paramEntityParsing)
+ dtd->keepProcessing = dtd->standalone;
+ else {
+ const XML_Char *name;
+ ENTITY *entity;
+ name = poolStoreString(&dtd->pool, enc,
+ s + enc->minBytesPerChar,
+ next - enc->minBytesPerChar);
+ if (!name)
+ return XML_ERROR_NO_MEMORY;
+ entity = (ENTITY *)lookup(&dtd->paramEntities, name, 0);
+ poolDiscard(&dtd->pool);
+ /* first, determine if a check for an existing declaration is needed;
+ if yes, check that the entity exists, and that it is internal,
+ otherwise call the skipped entity handler
+ */
+ if (prologState.documentEntity &&
+ (dtd->standalone
+ ? !openInternalEntities
+ : !dtd->hasParamEntityRefs)) {
+ if (!entity)
+ return XML_ERROR_UNDEFINED_ENTITY;
+ else if (!entity->is_internal)
+ return XML_ERROR_ENTITY_DECLARED_IN_PE;
+ }
+ else if (!entity) {
+ dtd->keepProcessing = dtd->standalone;
+ /* cannot report skipped entities in declarations */
+ if ((role == XML_ROLE_PARAM_ENTITY_REF) && skippedEntityHandler) {
+ skippedEntityHandler(handlerArg, name, 1);
+ handleDefault = XML_FALSE;
+ }
+ break;
+ }
+ if (entity->open)
+ return XML_ERROR_RECURSIVE_ENTITY_REF;
+ if (entity->textPtr) {
+ enum XML_Error result;
+ XML_Bool betweenDecl =
+ (role == XML_ROLE_PARAM_ENTITY_REF ? XML_TRUE : XML_FALSE);
+ result = processInternalEntity(parser, entity, betweenDecl);
+ if (result != XML_ERROR_NONE)
+ return result;
+ handleDefault = XML_FALSE;
+ break;
+ }
+ if (externalEntityRefHandler) {
+ dtd->paramEntityRead = XML_FALSE;
+ entity->open = XML_TRUE;
+ if (!externalEntityRefHandler(externalEntityRefHandlerArg,
+ 0,
+ entity->base,
+ entity->systemId,
+ entity->publicId)) {
+ entity->open = XML_FALSE;
+ return XML_ERROR_EXTERNAL_ENTITY_HANDLING;
+ }
+ entity->open = XML_FALSE;
+ handleDefault = XML_FALSE;
+ if (!dtd->paramEntityRead) {
+ dtd->keepProcessing = dtd->standalone;
+ break;
+ }
+ }
+ else {
+ dtd->keepProcessing = dtd->standalone;
+ break;
+ }
+ }
+#endif /* XML_DTD */
+ if (!dtd->standalone &&
+ notStandaloneHandler &&
+ !notStandaloneHandler(handlerArg))
+ return XML_ERROR_NOT_STANDALONE;
+ break;
+
+ /* Element declaration stuff */
+
+ case XML_ROLE_ELEMENT_NAME:
+ if (elementDeclHandler) {
+ declElementType = getElementType(parser, enc, s, next);
+ if (!declElementType)
+ return XML_ERROR_NO_MEMORY;
+ dtd->scaffLevel = 0;
+ dtd->scaffCount = 0;
+ dtd->in_eldecl = XML_TRUE;
+ handleDefault = XML_FALSE;
+ }
+ break;
+
+ case XML_ROLE_CONTENT_ANY:
+ case XML_ROLE_CONTENT_EMPTY:
+ if (dtd->in_eldecl) {
+ if (elementDeclHandler) {
+ XML_Content * content = (XML_Content *) MALLOC(sizeof(XML_Content));
+ if (!content)
+ return XML_ERROR_NO_MEMORY;
+ content->quant = XML_CQUANT_NONE;
+ content->name = NULL;
+ content->numchildren = 0;
+ content->children = NULL;
+ content->type = ((role == XML_ROLE_CONTENT_ANY) ?
+ XML_CTYPE_ANY :
+ XML_CTYPE_EMPTY);
+ *eventEndPP = s;
+ elementDeclHandler(handlerArg, declElementType->name, content);
+ handleDefault = XML_FALSE;
+ }
+ dtd->in_eldecl = XML_FALSE;
+ }
+ break;
+
+ case XML_ROLE_CONTENT_PCDATA:
+ if (dtd->in_eldecl) {
+ dtd->scaffold[dtd->scaffIndex[dtd->scaffLevel - 1]].type
+ = XML_CTYPE_MIXED;
+ if (elementDeclHandler)
+ handleDefault = XML_FALSE;
+ }
+ break;
+
+ case XML_ROLE_CONTENT_ELEMENT:
+ quant = XML_CQUANT_NONE;
+ goto elementContent;
+ case XML_ROLE_CONTENT_ELEMENT_OPT:
+ quant = XML_CQUANT_OPT;
+ goto elementContent;
+ case XML_ROLE_CONTENT_ELEMENT_REP:
+ quant = XML_CQUANT_REP;
+ goto elementContent;
+ case XML_ROLE_CONTENT_ELEMENT_PLUS:
+ quant = XML_CQUANT_PLUS;
+ elementContent:
+ if (dtd->in_eldecl) {
+ ELEMENT_TYPE *el;
+ const XML_Char *name;
+ int nameLen;
+ const char *nxt = (quant == XML_CQUANT_NONE
+ ? next
+ : next - enc->minBytesPerChar);
+ int myindex = nextScaffoldPart(parser);
+ if (myindex < 0)
+ return XML_ERROR_NO_MEMORY;
+ dtd->scaffold[myindex].type = XML_CTYPE_NAME;
+ dtd->scaffold[myindex].quant = quant;
+ el = getElementType(parser, enc, s, nxt);
+ if (!el)
+ return XML_ERROR_NO_MEMORY;
+ name = el->name;
+ dtd->scaffold[myindex].name = name;
+ nameLen = 0;
+ for (; name[nameLen++]; );
+ dtd->contentStringLen += nameLen;
+ if (elementDeclHandler)
+ handleDefault = XML_FALSE;
+ }
+ break;
+
+ case XML_ROLE_GROUP_CLOSE:
+ quant = XML_CQUANT_NONE;
+ goto closeGroup;
+ case XML_ROLE_GROUP_CLOSE_OPT:
+ quant = XML_CQUANT_OPT;
+ goto closeGroup;
+ case XML_ROLE_GROUP_CLOSE_REP:
+ quant = XML_CQUANT_REP;
+ goto closeGroup;
+ case XML_ROLE_GROUP_CLOSE_PLUS:
+ quant = XML_CQUANT_PLUS;
+ closeGroup:
+ if (dtd->in_eldecl) {
+ if (elementDeclHandler)
+ handleDefault = XML_FALSE;
+ dtd->scaffLevel--;
+ dtd->scaffold[dtd->scaffIndex[dtd->scaffLevel]].quant = quant;
+ if (dtd->scaffLevel == 0) {
+ if (!handleDefault) {
+ XML_Content *model = build_model(parser);
+ if (!model)
+ return XML_ERROR_NO_MEMORY;
+ *eventEndPP = s;
+ elementDeclHandler(handlerArg, declElementType->name, model);
+ }
+ dtd->in_eldecl = XML_FALSE;
+ dtd->contentStringLen = 0;
+ }
+ }
+ break;
+ /* End element declaration stuff */
+
+ case XML_ROLE_PI:
+ if (!reportProcessingInstruction(parser, enc, s, next))
+ return XML_ERROR_NO_MEMORY;
+ handleDefault = XML_FALSE;
+ break;
+ case XML_ROLE_COMMENT:
+ if (!reportComment(parser, enc, s, next))
+ return XML_ERROR_NO_MEMORY;
+ handleDefault = XML_FALSE;
+ break;
+ case XML_ROLE_NONE:
+ switch (tok) {
+ case XML_TOK_BOM:
+ handleDefault = XML_FALSE;
+ break;
+ }
+ break;
+ case XML_ROLE_DOCTYPE_NONE:
+ if (startDoctypeDeclHandler)
+ handleDefault = XML_FALSE;
+ break;
+ case XML_ROLE_ENTITY_NONE:
+ if (dtd->keepProcessing && entityDeclHandler)
+ handleDefault = XML_FALSE;
+ break;
+ case XML_ROLE_NOTATION_NONE:
+ if (notationDeclHandler)
+ handleDefault = XML_FALSE;
+ break;
+ case XML_ROLE_ATTLIST_NONE:
+ if (dtd->keepProcessing && attlistDeclHandler)
+ handleDefault = XML_FALSE;
+ break;
+ case XML_ROLE_ELEMENT_NONE:
+ if (elementDeclHandler)
+ handleDefault = XML_FALSE;
+ break;
+ } /* end of big switch */
+
+ if (handleDefault && defaultHandler)
+ reportDefault(parser, enc, s, next);
+
+ switch (ps_parsing) {
+ case XML_SUSPENDED:
+ *nextPtr = next;
+ return XML_ERROR_NONE;
+ case XML_FINISHED:
+ return XML_ERROR_ABORTED;
+ default:
+ s = next;
+ tok = XmlPrologTok(enc, s, end, &next);
+ }
+ }
+ /* not reached */
+}
+
+static enum XML_Error PTRCALL
+epilogProcessor(XML_Parser parser,
+ const char *s,
+ const char *end,
+ const char **nextPtr)
+{
+ processor = epilogProcessor;
+ eventPtr = s;
+ for (;;) {
+ const char *next = NULL;
+ int tok = XmlPrologTok(encoding, s, end, &next);
+ eventEndPtr = next;
+ switch (tok) {
+ /* report partial linebreak - it might be the last token */
+ case -XML_TOK_PROLOG_S:
+ if (defaultHandler) {
+ reportDefault(parser, encoding, s, next);
+ if (ps_parsing == XML_FINISHED)
+ return XML_ERROR_ABORTED;
+ }
+ *nextPtr = next;
+ return XML_ERROR_NONE;
+ case XML_TOK_NONE:
+ *nextPtr = s;
+ return XML_ERROR_NONE;
+ case XML_TOK_PROLOG_S:
+ if (defaultHandler)
+ reportDefault(parser, encoding, s, next);
+ break;
+ case XML_TOK_PI:
+ if (!reportProcessingInstruction(parser, encoding, s, next))
+ return XML_ERROR_NO_MEMORY;
+ break;
+ case XML_TOK_COMMENT:
+ if (!reportComment(parser, encoding, s, next))
+ return XML_ERROR_NO_MEMORY;
+ break;
+ case XML_TOK_INVALID:
+ eventPtr = next;
+ return XML_ERROR_INVALID_TOKEN;
+ case XML_TOK_PARTIAL:
+ if (!ps_finalBuffer) {
+ *nextPtr = s;
+ return XML_ERROR_NONE;
+ }
+ return XML_ERROR_UNCLOSED_TOKEN;
+ case XML_TOK_PARTIAL_CHAR:
+ if (!ps_finalBuffer) {
+ *nextPtr = s;
+ return XML_ERROR_NONE;
+ }
+ return XML_ERROR_PARTIAL_CHAR;
+ default:
+ return XML_ERROR_JUNK_AFTER_DOC_ELEMENT;
+ }
+ eventPtr = s = next;
+ switch (ps_parsing) {
+ case XML_SUSPENDED:
+ *nextPtr = next;
+ return XML_ERROR_NONE;
+ case XML_FINISHED:
+ return XML_ERROR_ABORTED;
+ default: ;
+ }
+ }
+}
+
+static enum XML_Error
+processInternalEntity(XML_Parser parser, ENTITY *entity,
+ XML_Bool betweenDecl)
+{
+ const char *textStart, *textEnd;
+ const char *next;
+ enum XML_Error result;
+ OPEN_INTERNAL_ENTITY *openEntity;
+
+ if (freeInternalEntities) {
+ openEntity = freeInternalEntities;
+ freeInternalEntities = openEntity->next;
+ }
+ else {
+ openEntity = (OPEN_INTERNAL_ENTITY *)MALLOC(sizeof(OPEN_INTERNAL_ENTITY));
+ if (!openEntity)
+ return XML_ERROR_NO_MEMORY;
+ }
+ entity->open = XML_TRUE;
+ entity->processed = 0;
+ openEntity->next = openInternalEntities;
+ openInternalEntities = openEntity;
+ openEntity->entity = entity;
+ openEntity->startTagLevel = tagLevel;
+ openEntity->betweenDecl = betweenDecl;
+ openEntity->internalEventPtr = NULL;
+ openEntity->internalEventEndPtr = NULL;
+ textStart = (char *)entity->textPtr;
+ textEnd = (char *)(entity->textPtr + entity->textLen);
+
+#ifdef XML_DTD
+ if (entity->is_param) {
+ int tok = XmlPrologTok(internalEncoding, textStart, textEnd, &next);
+ result = doProlog(parser, internalEncoding, textStart, textEnd, tok,
+ next, &next, XML_FALSE);
+ }
+ else
+#endif /* XML_DTD */
+ result = doContent(parser, tagLevel, internalEncoding, textStart,
+ textEnd, &next, XML_FALSE);
+
+ if (result == XML_ERROR_NONE) {
+ if (textEnd != next && ps_parsing == XML_SUSPENDED) {
+ entity->processed = (int)(next - textStart);
+ processor = internalEntityProcessor;
+ }
+ else {
+ entity->open = XML_FALSE;
+ openInternalEntities = openEntity->next;
+ /* put openEntity back in list of free instances */
+ openEntity->next = freeInternalEntities;
+ freeInternalEntities = openEntity;
+ }
+ }
+ return result;
+}
+
+static enum XML_Error PTRCALL
+internalEntityProcessor(XML_Parser parser,
+ const char *s,
+ const char *end,
+ const char **nextPtr)
+{
+ ENTITY *entity;
+ const char *textStart, *textEnd;
+ const char *next;
+ enum XML_Error result;
+ OPEN_INTERNAL_ENTITY *openEntity = openInternalEntities;
+ if (!openEntity)
+ return XML_ERROR_UNEXPECTED_STATE;
+
+ entity = openEntity->entity;
+ textStart = ((char *)entity->textPtr) + entity->processed;
+ textEnd = (char *)(entity->textPtr + entity->textLen);
+
+#ifdef XML_DTD
+ if (entity->is_param) {
+ int tok = XmlPrologTok(internalEncoding, textStart, textEnd, &next);
+ result = doProlog(parser, internalEncoding, textStart, textEnd, tok,
+ next, &next, XML_FALSE);
+ }
+ else
+#endif /* XML_DTD */
+ result = doContent(parser, openEntity->startTagLevel, internalEncoding,
+ textStart, textEnd, &next, XML_FALSE);
+
+ if (result != XML_ERROR_NONE)
+ return result;
+ else if (textEnd != next && ps_parsing == XML_SUSPENDED) {
+ entity->processed = (int)(next - (char *)entity->textPtr);
+ return result;
+ }
+ else {
+ entity->open = XML_FALSE;
+ openInternalEntities = openEntity->next;
+ /* put openEntity back in list of free instances */
+ openEntity->next = freeInternalEntities;
+ freeInternalEntities = openEntity;
+ }
+
+#ifdef XML_DTD
+ if (entity->is_param) {
+ int tok;
+ processor = prologProcessor;
+ tok = XmlPrologTok(encoding, s, end, &next);
+ return doProlog(parser, encoding, s, end, tok, next, nextPtr,
+ (XML_Bool)!ps_finalBuffer);
+ }
+ else
+#endif /* XML_DTD */
+ {
+ processor = contentProcessor;
+ /* see externalEntityContentProcessor vs contentProcessor */
+ return doContent(parser, parentParser ? 1 : 0, encoding, s, end,
+ nextPtr, (XML_Bool)!ps_finalBuffer);
+ }
+}
+
+static enum XML_Error PTRCALL
+errorProcessor(XML_Parser parser,
+ const char *s,
+ const char *end,
+ const char **nextPtr)
+{
+ return errorCode;
+}
+
+static enum XML_Error
+storeAttributeValue(XML_Parser parser, const ENCODING *enc, XML_Bool isCdata,
+ const char *ptr, const char *end,
+ STRING_POOL *pool)
+{
+ enum XML_Error result = appendAttributeValue(parser, enc, isCdata, ptr,
+ end, pool);
+ if (result)
+ return result;
+ if (!isCdata && poolLength(pool) && poolLastChar(pool) == 0x20)
+ poolChop(pool);
+ if (!poolAppendChar(pool, XML_T('\0')))
+ return XML_ERROR_NO_MEMORY;
+ return XML_ERROR_NONE;
+}
+
+static enum XML_Error
+appendAttributeValue(XML_Parser parser, const ENCODING *enc, XML_Bool isCdata,
+ const char *ptr, const char *end,
+ STRING_POOL *pool)
+{
+ DTD * const dtd = _dtd; /* save one level of indirection */
+ for (;;) {
+ const char *next;
+ int tok = XmlAttributeValueTok(enc, ptr, end, &next);
+ switch (tok) {
+ case XML_TOK_NONE:
+ return XML_ERROR_NONE;
+ case XML_TOK_INVALID:
+ if (enc == encoding)
+ eventPtr = next;
+ return XML_ERROR_INVALID_TOKEN;
+ case XML_TOK_PARTIAL:
+ if (enc == encoding)
+ eventPtr = ptr;
+ return XML_ERROR_INVALID_TOKEN;
+ case XML_TOK_CHAR_REF:
+ {
+ XML_Char buf[XML_ENCODE_MAX];
+ int i;
+ int n = XmlCharRefNumber(enc, ptr);
+ if (n < 0) {
+ if (enc == encoding)
+ eventPtr = ptr;
+ return XML_ERROR_BAD_CHAR_REF;
+ }
+ if (!isCdata
+ && n == 0x20 /* space */
+ && (poolLength(pool) == 0 || poolLastChar(pool) == 0x20))
+ break;
+ n = XmlEncode(n, (ICHAR *)buf);
+ if (!n) {
+ if (enc == encoding)
+ eventPtr = ptr;
+ return XML_ERROR_BAD_CHAR_REF;
+ }
+ for (i = 0; i < n; i++) {
+ if (!poolAppendChar(pool, buf[i]))
+ return XML_ERROR_NO_MEMORY;
+ }
+ }
+ break;
+ case XML_TOK_DATA_CHARS:
+ if (!poolAppend(pool, enc, ptr, next))
+ return XML_ERROR_NO_MEMORY;
+ break;
+ case XML_TOK_TRAILING_CR:
+ next = ptr + enc->minBytesPerChar;
+ /* fall through */
+ case XML_TOK_ATTRIBUTE_VALUE_S:
+ case XML_TOK_DATA_NEWLINE:
+ if (!isCdata && (poolLength(pool) == 0 || poolLastChar(pool) == 0x20))
+ break;
+ if (!poolAppendChar(pool, 0x20))
+ return XML_ERROR_NO_MEMORY;
+ break;
+ case XML_TOK_ENTITY_REF:
+ {
+ const XML_Char *name;
+ ENTITY *entity;
+ char checkEntityDecl;
+ XML_Char ch = (XML_Char) XmlPredefinedEntityName(enc,
+ ptr + enc->minBytesPerChar,
+ next - enc->minBytesPerChar);
+ if (ch) {
+ if (!poolAppendChar(pool, ch))
+ return XML_ERROR_NO_MEMORY;
+ break;
+ }
+ name = poolStoreString(&temp2Pool, enc,
+ ptr + enc->minBytesPerChar,
+ next - enc->minBytesPerChar);
+ if (!name)
+ return XML_ERROR_NO_MEMORY;
+ entity = (ENTITY *)lookup(&dtd->generalEntities, name, 0);
+ poolDiscard(&temp2Pool);
+ /* First, determine if a check for an existing declaration is needed;
+ if yes, check that the entity exists, and that it is internal.
+ */
+ if (pool == &dtd->pool) /* are we called from prolog? */
+ checkEntityDecl =
+#ifdef XML_DTD
+ prologState.documentEntity &&
+#endif /* XML_DTD */
+ (dtd->standalone
+ ? !openInternalEntities
+ : !dtd->hasParamEntityRefs);
+ else /* if (pool == &tempPool): we are called from content */
+ checkEntityDecl = !dtd->hasParamEntityRefs || dtd->standalone;
+ if (checkEntityDecl) {
+ if (!entity)
+ return XML_ERROR_UNDEFINED_ENTITY;
+ else if (!entity->is_internal)
+ return XML_ERROR_ENTITY_DECLARED_IN_PE;
+ }
+ else if (!entity) {
+ /* Cannot report skipped entity here - see comments on
+ skippedEntityHandler.
+ if (skippedEntityHandler)
+ skippedEntityHandler(handlerArg, name, 0);
+ */
+ /* Cannot call the default handler because this would be
+ out of sync with the call to the startElementHandler.
+ if ((pool == &tempPool) && defaultHandler)
+ reportDefault(parser, enc, ptr, next);
+ */
+ break;
+ }
+ if (entity->open) {
+ if (enc == encoding)
+ eventPtr = ptr;
+ return XML_ERROR_RECURSIVE_ENTITY_REF;
+ }
+ if (entity->notation) {
+ if (enc == encoding)
+ eventPtr = ptr;
+ return XML_ERROR_BINARY_ENTITY_REF;
+ }
+ if (!entity->textPtr) {
+ if (enc == encoding)
+ eventPtr = ptr;
+ return XML_ERROR_ATTRIBUTE_EXTERNAL_ENTITY_REF;
+ }
+ else {
+ enum XML_Error result;
+ const XML_Char *textEnd = entity->textPtr + entity->textLen;
+ entity->open = XML_TRUE;
+ result = appendAttributeValue(parser, internalEncoding, isCdata,
+ (char *)entity->textPtr,
+ (char *)textEnd, pool);
+ entity->open = XML_FALSE;
+ if (result)
+ return result;
+ }
+ }
+ break;
+ default:
+ if (enc == encoding)
+ eventPtr = ptr;
+ return XML_ERROR_UNEXPECTED_STATE;
+ }
+ ptr = next;
+ }
+ /* not reached */
+}
+
+static enum XML_Error
+storeEntityValue(XML_Parser parser,
+ const ENCODING *enc,
+ const char *entityTextPtr,
+ const char *entityTextEnd)
+{
+ DTD * const dtd = _dtd; /* save one level of indirection */
+ STRING_POOL *pool = &(dtd->entityValuePool);
+ enum XML_Error result = XML_ERROR_NONE;
+#ifdef XML_DTD
+ int oldInEntityValue = prologState.inEntityValue;
+ prologState.inEntityValue = 1;
+#endif /* XML_DTD */
+ /* never return Null for the value argument in EntityDeclHandler,
+ since this would indicate an external entity; therefore we
+ have to make sure that entityValuePool.start is not null */
+ if (!pool->blocks) {
+ if (!poolGrow(pool))
+ return XML_ERROR_NO_MEMORY;
+ }
+
+ for (;;) {
+ const char *next;
+ int tok = XmlEntityValueTok(enc, entityTextPtr, entityTextEnd, &next);
+ switch (tok) {
+ case XML_TOK_PARAM_ENTITY_REF:
+#ifdef XML_DTD
+ if (isParamEntity || enc != encoding) {
+ const XML_Char *name;
+ ENTITY *entity;
+ name = poolStoreString(&tempPool, enc,
+ entityTextPtr + enc->minBytesPerChar,
+ next - enc->minBytesPerChar);
+ if (!name) {
+ result = XML_ERROR_NO_MEMORY;
+ goto endEntityValue;
+ }
+ entity = (ENTITY *)lookup(&dtd->paramEntities, name, 0);
+ poolDiscard(&tempPool);
+ if (!entity) {
+ /* not a well-formedness error - see XML 1.0: WFC Entity Declared */
+ /* cannot report skipped entity here - see comments on
+ skippedEntityHandler
+ if (skippedEntityHandler)
+ skippedEntityHandler(handlerArg, name, 0);
+ */
+ dtd->keepProcessing = dtd->standalone;
+ goto endEntityValue;
+ }
+ if (entity->open) {
+ if (enc == encoding)
+ eventPtr = entityTextPtr;
+ result = XML_ERROR_RECURSIVE_ENTITY_REF;
+ goto endEntityValue;
+ }
+ if (entity->systemId) {
+ if (externalEntityRefHandler) {
+ dtd->paramEntityRead = XML_FALSE;
+ entity->open = XML_TRUE;
+ if (!externalEntityRefHandler(externalEntityRefHandlerArg,
+ 0,
+ entity->base,
+ entity->systemId,
+ entity->publicId)) {
+ entity->open = XML_FALSE;
+ result = XML_ERROR_EXTERNAL_ENTITY_HANDLING;
+ goto endEntityValue;
+ }
+ entity->open = XML_FALSE;
+ if (!dtd->paramEntityRead)
+ dtd->keepProcessing = dtd->standalone;
+ }
+ else
+ dtd->keepProcessing = dtd->standalone;
+ }
+ else {
+ entity->open = XML_TRUE;
+ result = storeEntityValue(parser,
+ internalEncoding,
+ (char *)entity->textPtr,
+ (char *)(entity->textPtr
+ + entity->textLen));
+ entity->open = XML_FALSE;
+ if (result)
+ goto endEntityValue;
+ }
+ break;
+ }
+#endif /* XML_DTD */
+ /* In the internal subset, PE references are not legal
+ within markup declarations, e.g entity values in this case. */
+ eventPtr = entityTextPtr;
+ result = XML_ERROR_PARAM_ENTITY_REF;
+ goto endEntityValue;
+ case XML_TOK_NONE:
+ result = XML_ERROR_NONE;
+ goto endEntityValue;
+ case XML_TOK_ENTITY_REF:
+ case XML_TOK_DATA_CHARS:
+ if (!poolAppend(pool, enc, entityTextPtr, next)) {
+ result = XML_ERROR_NO_MEMORY;
+ goto endEntityValue;
+ }
+ break;
+ case XML_TOK_TRAILING_CR:
+ next = entityTextPtr + enc->minBytesPerChar;
+ /* fall through */
+ case XML_TOK_DATA_NEWLINE:
+ if (pool->end == pool->ptr && !poolGrow(pool)) {
+ result = XML_ERROR_NO_MEMORY;
+ goto endEntityValue;
+ }
+ *(pool->ptr)++ = 0xA;
+ break;
+ case XML_TOK_CHAR_REF:
+ {
+ XML_Char buf[XML_ENCODE_MAX];
+ int i;
+ int n = XmlCharRefNumber(enc, entityTextPtr);
+ if (n < 0) {
+ if (enc == encoding)
+ eventPtr = entityTextPtr;
+ result = XML_ERROR_BAD_CHAR_REF;
+ goto endEntityValue;
+ }
+ n = XmlEncode(n, (ICHAR *)buf);
+ if (!n) {
+ if (enc == encoding)
+ eventPtr = entityTextPtr;
+ result = XML_ERROR_BAD_CHAR_REF;
+ goto endEntityValue;
+ }
+ for (i = 0; i < n; i++) {
+ if (pool->end == pool->ptr && !poolGrow(pool)) {
+ result = XML_ERROR_NO_MEMORY;
+ goto endEntityValue;
+ }
+ *(pool->ptr)++ = buf[i];
+ }
+ }
+ break;
+ case XML_TOK_PARTIAL:
+ if (enc == encoding)
+ eventPtr = entityTextPtr;
+ result = XML_ERROR_INVALID_TOKEN;
+ goto endEntityValue;
+ case XML_TOK_INVALID:
+ if (enc == encoding)
+ eventPtr = next;
+ result = XML_ERROR_INVALID_TOKEN;
+ goto endEntityValue;
+ default:
+ if (enc == encoding)
+ eventPtr = entityTextPtr;
+ result = XML_ERROR_UNEXPECTED_STATE;
+ goto endEntityValue;
+ }
+ entityTextPtr = next;
+ }
+endEntityValue:
+#ifdef XML_DTD
+ prologState.inEntityValue = oldInEntityValue;
+#endif /* XML_DTD */
+ return result;
+}
+
+static void FASTCALL
+normalizeLines(XML_Char *s)
+{
+ XML_Char *p;
+ for (;; s++) {
+ if (*s == XML_T('\0'))
+ return;
+ if (*s == 0xD)
+ break;
+ }
+ p = s;
+ do {
+ if (*s == 0xD) {
+ *p++ = 0xA;
+ if (*++s == 0xA)
+ s++;
+ }
+ else
+ *p++ = *s++;
+ } while (*s);
+ *p = XML_T('\0');
+}
+
+static int
+reportProcessingInstruction(XML_Parser parser, const ENCODING *enc,
+ const char *start, const char *end)
+{
+ const XML_Char *target;
+ XML_Char *data;
+ const char *tem;
+ if (!processingInstructionHandler) {
+ if (defaultHandler)
+ reportDefault(parser, enc, start, end);
+ return 1;
+ }
+ start += enc->minBytesPerChar * 2;
+ tem = start + XmlNameLength(enc, start);
+ target = poolStoreString(&tempPool, enc, start, tem);
+ if (!target)
+ return 0;
+ poolFinish(&tempPool);
+ data = poolStoreString(&tempPool, enc,
+ XmlSkipS(enc, tem),
+ end - enc->minBytesPerChar*2);
+ if (!data)
+ return 0;
+ normalizeLines(data);
+ processingInstructionHandler(handlerArg, target, data);
+ poolClear(&tempPool);
+ return 1;
+}
+
+static int
+reportComment(XML_Parser parser, const ENCODING *enc,
+ const char *start, const char *end)
+{
+ XML_Char *data;
+ if (!commentHandler) {
+ if (defaultHandler)
+ reportDefault(parser, enc, start, end);
+ return 1;
+ }
+ data = poolStoreString(&tempPool,
+ enc,
+ start + enc->minBytesPerChar * 4,
+ end - enc->minBytesPerChar * 3);
+ if (!data)
+ return 0;
+ normalizeLines(data);
+ commentHandler(handlerArg, data);
+ poolClear(&tempPool);
+ return 1;
+}
+
+static void
+reportDefault(XML_Parser parser, const ENCODING *enc,
+ const char *s, const char *end)
+{
+ if (MUST_CONVERT(enc, s)) {
+ const char **eventPP;
+ const char **eventEndPP;
+ if (enc == encoding) {
+ eventPP = &eventPtr;
+ eventEndPP = &eventEndPtr;
+ }
+ else {
+ eventPP = &(openInternalEntities->internalEventPtr);
+ eventEndPP = &(openInternalEntities->internalEventEndPtr);
+ }
+ do {
+ ICHAR *dataPtr = (ICHAR *)dataBuf;
+ XmlConvert(enc, &s, end, &dataPtr, (ICHAR *)dataBufEnd);
+ *eventEndPP = s;
+ defaultHandler(handlerArg, dataBuf, (int)(dataPtr - (ICHAR *)dataBuf));
+ *eventPP = s;
+ } while (s != end);
+ }
+ else
+ defaultHandler(handlerArg, (XML_Char *)s, (int)((XML_Char *)end - (XML_Char *)s));
+}
+
+
+static int
+defineAttribute(ELEMENT_TYPE *type, ATTRIBUTE_ID *attId, XML_Bool isCdata,
+ XML_Bool isId, const XML_Char *value, XML_Parser parser)
+{
+ DEFAULT_ATTRIBUTE *att;
+ if (value || isId) {
+ /* The handling of default attributes gets messed up if we have
+ a default which duplicates a non-default. */
+ int i;
+ for (i = 0; i < type->nDefaultAtts; i++)
+ if (attId == type->defaultAtts[i].id)
+ return 1;
+ if (isId && !type->idAtt && !attId->xmlns)
+ type->idAtt = attId;
+ }
+ if (type->nDefaultAtts == type->allocDefaultAtts) {
+ if (type->allocDefaultAtts == 0) {
+ type->allocDefaultAtts = 8;
+ type->defaultAtts = (DEFAULT_ATTRIBUTE *)MALLOC(type->allocDefaultAtts
+ * sizeof(DEFAULT_ATTRIBUTE));
+ if (!type->defaultAtts)
+ return 0;
+ }
+ else {
+ DEFAULT_ATTRIBUTE *temp;
+ int count = type->allocDefaultAtts * 2;
+ temp = (DEFAULT_ATTRIBUTE *)
+ REALLOC(type->defaultAtts, (count * sizeof(DEFAULT_ATTRIBUTE)));
+ if (temp == NULL)
+ return 0;
+ type->allocDefaultAtts = count;
+ type->defaultAtts = temp;
+ }
+ }
+ att = type->defaultAtts + type->nDefaultAtts;
+ att->id = attId;
+ att->value = value;
+ att->isCdata = isCdata;
+ if (!isCdata)
+ attId->maybeTokenized = XML_TRUE;
+ type->nDefaultAtts += 1;
+ return 1;
+}
+
+static int
+setElementTypePrefix(XML_Parser parser, ELEMENT_TYPE *elementType)
+{
+ DTD * const dtd = _dtd; /* save one level of indirection */
+ const XML_Char *name;
+ for (name = elementType->name; *name; name++) {
+ if (*name == XML_T(':')) {
+ PREFIX *prefix;
+ const XML_Char *s;
+ for (s = elementType->name; s != name; s++) {
+ if (!poolAppendChar(&dtd->pool, *s))
+ return 0;
+ }
+ if (!poolAppendChar(&dtd->pool, XML_T('\0')))
+ return 0;
+ prefix = (PREFIX *)lookup(&dtd->prefixes, poolStart(&dtd->pool),
+ sizeof(PREFIX));
+ if (!prefix)
+ return 0;
+ if (prefix->name == poolStart(&dtd->pool))
+ poolFinish(&dtd->pool);
+ else
+ poolDiscard(&dtd->pool);
+ elementType->prefix = prefix;
+
+ }
+ }
+ return 1;
+}
+
+static ATTRIBUTE_ID *
+getAttributeId(XML_Parser parser, const ENCODING *enc,
+ const char *start, const char *end)
+{
+ DTD * const dtd = _dtd; /* save one level of indirection */
+ ATTRIBUTE_ID *id;
+ const XML_Char *name;
+ if (!poolAppendChar(&dtd->pool, XML_T('\0')))
+ return NULL;
+ name = poolStoreString(&dtd->pool, enc, start, end);
+ if (!name)
+ return NULL;
+ /* skip quotation mark - its storage will be re-used (like in name[-1]) */
+ ++name;
+ id = (ATTRIBUTE_ID *)lookup(&dtd->attributeIds, name, sizeof(ATTRIBUTE_ID));
+ if (!id)
+ return NULL;
+ if (id->name != name)
+ poolDiscard(&dtd->pool);
+ else {
+ poolFinish(&dtd->pool);
+ if (!ns)
+ ;
+ else if (name[0] == XML_T('x')
+ && name[1] == XML_T('m')
+ && name[2] == XML_T('l')
+ && name[3] == XML_T('n')
+ && name[4] == XML_T('s')
+ && (name[5] == XML_T('\0') || name[5] == XML_T(':'))) {
+ if (name[5] == XML_T('\0'))
+ id->prefix = &dtd->defaultPrefix;
+ else
+ id->prefix = (PREFIX *)lookup(&dtd->prefixes, name + 6, sizeof(PREFIX));
+ id->xmlns = XML_TRUE;
+ }
+ else {
+ int i;
+ for (i = 0; name[i]; i++) {
+ /* attributes without prefix are *not* in the default namespace */
+ if (name[i] == XML_T(':')) {
+ int j;
+ for (j = 0; j < i; j++) {
+ if (!poolAppendChar(&dtd->pool, name[j]))
+ return NULL;
+ }
+ if (!poolAppendChar(&dtd->pool, XML_T('\0')))
+ return NULL;
+ id->prefix = (PREFIX *)lookup(&dtd->prefixes, poolStart(&dtd->pool),
+ sizeof(PREFIX));
+ if (id->prefix->name == poolStart(&dtd->pool))
+ poolFinish(&dtd->pool);
+ else
+ poolDiscard(&dtd->pool);
+ break;
+ }
+ }
+ }
+ }
+ return id;
+}
+
+#define CONTEXT_SEP XML_T('\f')
+
+static const XML_Char *
+getContext(XML_Parser parser)
+{
+ DTD * const dtd = _dtd; /* save one level of indirection */
+ HASH_TABLE_ITER iter;
+ XML_Bool needSep = XML_FALSE;
+
+ if (dtd->defaultPrefix.binding) {
+ int i;
+ int len;
+ if (!poolAppendChar(&tempPool, XML_T('=')))
+ return NULL;
+ len = dtd->defaultPrefix.binding->uriLen;
+ if (namespaceSeparator)
+ len--;
+ for (i = 0; i < len; i++)
+ if (!poolAppendChar(&tempPool, dtd->defaultPrefix.binding->uri[i]))
+ return NULL;
+ needSep = XML_TRUE;
+ }
+
+ hashTableIterInit(&iter, &(dtd->prefixes));
+ for (;;) {
+ int i;
+ int len;
+ const XML_Char *s;
+ PREFIX *prefix = (PREFIX *)hashTableIterNext(&iter);
+ if (!prefix)
+ break;
+ if (!prefix->binding)
+ continue;
+ if (needSep && !poolAppendChar(&tempPool, CONTEXT_SEP))
+ return NULL;
+ for (s = prefix->name; *s; s++)
+ if (!poolAppendChar(&tempPool, *s))
+ return NULL;
+ if (!poolAppendChar(&tempPool, XML_T('=')))
+ return NULL;
+ len = prefix->binding->uriLen;
+ if (namespaceSeparator)
+ len--;
+ for (i = 0; i < len; i++)
+ if (!poolAppendChar(&tempPool, prefix->binding->uri[i]))
+ return NULL;
+ needSep = XML_TRUE;
+ }
+
+
+ hashTableIterInit(&iter, &(dtd->generalEntities));
+ for (;;) {
+ const XML_Char *s;
+ ENTITY *e = (ENTITY *)hashTableIterNext(&iter);
+ if (!e)
+ break;
+ if (!e->open)
+ continue;
+ if (needSep && !poolAppendChar(&tempPool, CONTEXT_SEP))
+ return NULL;
+ for (s = e->name; *s; s++)
+ if (!poolAppendChar(&tempPool, *s))
+ return 0;
+ needSep = XML_TRUE;
+ }
+
+ if (!poolAppendChar(&tempPool, XML_T('\0')))
+ return NULL;
+ return tempPool.start;
+}
+
+static XML_Bool
+setContext(XML_Parser parser, const XML_Char *context)
+{
+ DTD * const dtd = _dtd; /* save one level of indirection */
+ const XML_Char *s = context;
+
+ while (*context != XML_T('\0')) {
+ if (*s == CONTEXT_SEP || *s == XML_T('\0')) {
+ ENTITY *e;
+ if (!poolAppendChar(&tempPool, XML_T('\0')))
+ return XML_FALSE;
+ e = (ENTITY *)lookup(&dtd->generalEntities, poolStart(&tempPool), 0);
+ if (e)
+ e->open = XML_TRUE;
+ if (*s != XML_T('\0'))
+ s++;
+ context = s;
+ poolDiscard(&tempPool);
+ }
+ else if (*s == XML_T('=')) {
+ PREFIX *prefix;
+ if (poolLength(&tempPool) == 0)
+ prefix = &dtd->defaultPrefix;
+ else {
+ if (!poolAppendChar(&tempPool, XML_T('\0')))
+ return XML_FALSE;
+ prefix = (PREFIX *)lookup(&dtd->prefixes, poolStart(&tempPool),
+ sizeof(PREFIX));
+ if (!prefix)
+ return XML_FALSE;
+ if (prefix->name == poolStart(&tempPool)) {
+ prefix->name = poolCopyString(&dtd->pool, prefix->name);
+ if (!prefix->name)
+ return XML_FALSE;
+ }
+ poolDiscard(&tempPool);
+ }
+ for (context = s + 1;
+ *context != CONTEXT_SEP && *context != XML_T('\0');
+ context++)
+ if (!poolAppendChar(&tempPool, *context))
+ return XML_FALSE;
+ if (!poolAppendChar(&tempPool, XML_T('\0')))
+ return XML_FALSE;
+ if (addBinding(parser, prefix, NULL, poolStart(&tempPool),
+ &inheritedBindings) != XML_ERROR_NONE)
+ return XML_FALSE;
+ poolDiscard(&tempPool);
+ if (*context != XML_T('\0'))
+ ++context;
+ s = context;
+ }
+ else {
+ if (!poolAppendChar(&tempPool, *s))
+ return XML_FALSE;
+ s++;
+ }
+ }
+ return XML_TRUE;
+}
+
+static void FASTCALL
+normalizePublicId(XML_Char *publicId)
+{
+ XML_Char *p = publicId;
+ XML_Char *s;
+ for (s = publicId; *s; s++) {
+ switch (*s) {
+ case 0x20:
+ case 0xD:
+ case 0xA:
+ if (p != publicId && p[-1] != 0x20)
+ *p++ = 0x20;
+ break;
+ default:
+ *p++ = *s;
+ }
+ }
+ if (p != publicId && p[-1] == 0x20)
+ --p;
+ *p = XML_T('\0');
+}
+
+static DTD *
+dtdCreate(const XML_Memory_Handling_Suite *ms)
+{
+ DTD *p = (DTD *)ms->malloc_fcn(sizeof(DTD));
+ if (p == NULL)
+ return p;
+ poolInit(&(p->pool), ms);
+ poolInit(&(p->entityValuePool), ms);
+ hashTableInit(&(p->generalEntities), ms);
+ hashTableInit(&(p->elementTypes), ms);
+ hashTableInit(&(p->attributeIds), ms);
+ hashTableInit(&(p->prefixes), ms);
+#ifdef XML_DTD
+ p->paramEntityRead = XML_FALSE;
+ hashTableInit(&(p->paramEntities), ms);
+#endif /* XML_DTD */
+ p->defaultPrefix.name = NULL;
+ p->defaultPrefix.binding = NULL;
+
+ p->in_eldecl = XML_FALSE;
+ p->scaffIndex = NULL;
+ p->scaffold = NULL;
+ p->scaffLevel = 0;
+ p->scaffSize = 0;
+ p->scaffCount = 0;
+ p->contentStringLen = 0;
+
+ p->keepProcessing = XML_TRUE;
+ p->hasParamEntityRefs = XML_FALSE;
+ p->standalone = XML_FALSE;
+ return p;
+}
+
+static void
+dtdReset(DTD *p, const XML_Memory_Handling_Suite *ms)
+{
+ HASH_TABLE_ITER iter;
+ hashTableIterInit(&iter, &(p->elementTypes));
+ for (;;) {
+ ELEMENT_TYPE *e = (ELEMENT_TYPE *)hashTableIterNext(&iter);
+ if (!e)
+ break;
+ if (e->allocDefaultAtts != 0)
+ ms->free_fcn(e->defaultAtts);
+ }
+ hashTableClear(&(p->generalEntities));
+#ifdef XML_DTD
+ p->paramEntityRead = XML_FALSE;
+ hashTableClear(&(p->paramEntities));
+#endif /* XML_DTD */
+ hashTableClear(&(p->elementTypes));
+ hashTableClear(&(p->attributeIds));
+ hashTableClear(&(p->prefixes));
+ poolClear(&(p->pool));
+ poolClear(&(p->entityValuePool));
+ p->defaultPrefix.name = NULL;
+ p->defaultPrefix.binding = NULL;
+
+ p->in_eldecl = XML_FALSE;
+
+ ms->free_fcn(p->scaffIndex);
+ p->scaffIndex = NULL;
+ ms->free_fcn(p->scaffold);
+ p->scaffold = NULL;
+
+ p->scaffLevel = 0;
+ p->scaffSize = 0;
+ p->scaffCount = 0;
+ p->contentStringLen = 0;
+
+ p->keepProcessing = XML_TRUE;
+ p->hasParamEntityRefs = XML_FALSE;
+ p->standalone = XML_FALSE;
+}
+
+static void
+dtdDestroy(DTD *p, XML_Bool isDocEntity, const XML_Memory_Handling_Suite *ms)
+{
+ HASH_TABLE_ITER iter;
+ hashTableIterInit(&iter, &(p->elementTypes));
+ for (;;) {
+ ELEMENT_TYPE *e = (ELEMENT_TYPE *)hashTableIterNext(&iter);
+ if (!e)
+ break;
+ if (e->allocDefaultAtts != 0)
+ ms->free_fcn(e->defaultAtts);
+ }
+ hashTableDestroy(&(p->generalEntities));
+#ifdef XML_DTD
+ hashTableDestroy(&(p->paramEntities));
+#endif /* XML_DTD */
+ hashTableDestroy(&(p->elementTypes));
+ hashTableDestroy(&(p->attributeIds));
+ hashTableDestroy(&(p->prefixes));
+ poolDestroy(&(p->pool));
+ poolDestroy(&(p->entityValuePool));
+ if (isDocEntity) {
+ ms->free_fcn(p->scaffIndex);
+ ms->free_fcn(p->scaffold);
+ }
+ ms->free_fcn(p);
+}
+
+/* Do a deep copy of the DTD. Return 0 for out of memory, non-zero otherwise.
+ The new DTD has already been initialized.
+*/
+static int
+dtdCopy(DTD *newDtd, const DTD *oldDtd, const XML_Memory_Handling_Suite *ms)
+{
+ HASH_TABLE_ITER iter;
+
+ /* Copy the prefix table. */
+
+ hashTableIterInit(&iter, &(oldDtd->prefixes));
+ for (;;) {
+ const XML_Char *name;
+ const PREFIX *oldP = (PREFIX *)hashTableIterNext(&iter);
+ if (!oldP)
+ break;
+ name = poolCopyString(&(newDtd->pool), oldP->name);
+ if (!name)
+ return 0;
+ if (!lookup(&(newDtd->prefixes), name, sizeof(PREFIX)))
+ return 0;
+ }
+
+ hashTableIterInit(&iter, &(oldDtd->attributeIds));
+
+ /* Copy the attribute id table. */
+
+ for (;;) {
+ ATTRIBUTE_ID *newA;
+ const XML_Char *name;
+ const ATTRIBUTE_ID *oldA = (ATTRIBUTE_ID *)hashTableIterNext(&iter);
+
+ if (!oldA)
+ break;
+ /* Remember to allocate the scratch byte before the name. */
+ if (!poolAppendChar(&(newDtd->pool), XML_T('\0')))
+ return 0;
+ name = poolCopyString(&(newDtd->pool), oldA->name);
+ if (!name)
+ return 0;
+ ++name;
+ newA = (ATTRIBUTE_ID *)lookup(&(newDtd->attributeIds), name,
+ sizeof(ATTRIBUTE_ID));
+ if (!newA)
+ return 0;
+ newA->maybeTokenized = oldA->maybeTokenized;
+ if (oldA->prefix) {
+ newA->xmlns = oldA->xmlns;
+ if (oldA->prefix == &oldDtd->defaultPrefix)
+ newA->prefix = &newDtd->defaultPrefix;
+ else
+ newA->prefix = (PREFIX *)lookup(&(newDtd->prefixes),
+ oldA->prefix->name, 0);
+ }
+ }
+
+ /* Copy the element type table. */
+
+ hashTableIterInit(&iter, &(oldDtd->elementTypes));
+
+ for (;;) {
+ int i;
+ ELEMENT_TYPE *newE;
+ const XML_Char *name;
+ const ELEMENT_TYPE *oldE = (ELEMENT_TYPE *)hashTableIterNext(&iter);
+ if (!oldE)
+ break;
+ name = poolCopyString(&(newDtd->pool), oldE->name);
+ if (!name)
+ return 0;
+ newE = (ELEMENT_TYPE *)lookup(&(newDtd->elementTypes), name,
+ sizeof(ELEMENT_TYPE));
+ if (!newE)
+ return 0;
+ if (oldE->nDefaultAtts) {
+ newE->defaultAtts = (DEFAULT_ATTRIBUTE *)
+ ms->malloc_fcn(oldE->nDefaultAtts * sizeof(DEFAULT_ATTRIBUTE));
+ if (!newE->defaultAtts) {
+ ms->free_fcn(newE);
+ return 0;
+ }
+ }
+ if (oldE->idAtt)
+ newE->idAtt = (ATTRIBUTE_ID *)
+ lookup(&(newDtd->attributeIds), oldE->idAtt->name, 0);
+ newE->allocDefaultAtts = newE->nDefaultAtts = oldE->nDefaultAtts;
+ if (oldE->prefix)
+ newE->prefix = (PREFIX *)lookup(&(newDtd->prefixes),
+ oldE->prefix->name, 0);
+ for (i = 0; i < newE->nDefaultAtts; i++) {
+ newE->defaultAtts[i].id = (ATTRIBUTE_ID *)
+ lookup(&(newDtd->attributeIds), oldE->defaultAtts[i].id->name, 0);
+ newE->defaultAtts[i].isCdata = oldE->defaultAtts[i].isCdata;
+ if (oldE->defaultAtts[i].value) {
+ newE->defaultAtts[i].value
+ = poolCopyString(&(newDtd->pool), oldE->defaultAtts[i].value);
+ if (!newE->defaultAtts[i].value)
+ return 0;
+ }
+ else
+ newE->defaultAtts[i].value = NULL;
+ }
+ }
+
+ /* Copy the entity tables. */
+ if (!copyEntityTable(&(newDtd->generalEntities),
+ &(newDtd->pool),
+ &(oldDtd->generalEntities)))
+ return 0;
+
+#ifdef XML_DTD
+ if (!copyEntityTable(&(newDtd->paramEntities),
+ &(newDtd->pool),
+ &(oldDtd->paramEntities)))
+ return 0;
+ newDtd->paramEntityRead = oldDtd->paramEntityRead;
+#endif /* XML_DTD */
+
+ newDtd->keepProcessing = oldDtd->keepProcessing;
+ newDtd->hasParamEntityRefs = oldDtd->hasParamEntityRefs;
+ newDtd->standalone = oldDtd->standalone;
+
+ /* Don't want deep copying for scaffolding */
+ newDtd->in_eldecl = oldDtd->in_eldecl;
+ newDtd->scaffold = oldDtd->scaffold;
+ newDtd->contentStringLen = oldDtd->contentStringLen;
+ newDtd->scaffSize = oldDtd->scaffSize;
+ newDtd->scaffLevel = oldDtd->scaffLevel;
+ newDtd->scaffIndex = oldDtd->scaffIndex;
+
+ return 1;
+} /* End dtdCopy */
+
+static int
+copyEntityTable(HASH_TABLE *newTable,
+ STRING_POOL *newPool,
+ const HASH_TABLE *oldTable)
+{
+ HASH_TABLE_ITER iter;
+ const XML_Char *cachedOldBase = NULL;
+ const XML_Char *cachedNewBase = NULL;
+
+ hashTableIterInit(&iter, oldTable);
+
+ for (;;) {
+ ENTITY *newE;
+ const XML_Char *name;
+ const ENTITY *oldE = (ENTITY *)hashTableIterNext(&iter);
+ if (!oldE)
+ break;
+ name = poolCopyString(newPool, oldE->name);
+ if (!name)
+ return 0;
+ newE = (ENTITY *)lookup(newTable, name, sizeof(ENTITY));
+ if (!newE)
+ return 0;
+ if (oldE->systemId) {
+ const XML_Char *tem = poolCopyString(newPool, oldE->systemId);
+ if (!tem)
+ return 0;
+ newE->systemId = tem;
+ if (oldE->base) {
+ if (oldE->base == cachedOldBase)
+ newE->base = cachedNewBase;
+ else {
+ cachedOldBase = oldE->base;
+ tem = poolCopyString(newPool, cachedOldBase);
+ if (!tem)
+ return 0;
+ cachedNewBase = newE->base = tem;
+ }
+ }
+ if (oldE->publicId) {
+ tem = poolCopyString(newPool, oldE->publicId);
+ if (!tem)
+ return 0;
+ newE->publicId = tem;
+ }
+ }
+ else {
+ const XML_Char *tem = poolCopyStringN(newPool, oldE->textPtr,
+ oldE->textLen);
+ if (!tem)
+ return 0;
+ newE->textPtr = tem;
+ newE->textLen = oldE->textLen;
+ }
+ if (oldE->notation) {
+ const XML_Char *tem = poolCopyString(newPool, oldE->notation);
+ if (!tem)
+ return 0;
+ newE->notation = tem;
+ }
+ newE->is_param = oldE->is_param;
+ newE->is_internal = oldE->is_internal;
+ }
+ return 1;
+}
+
+#define INIT_POWER 6
+
+static XML_Bool FASTCALL
+keyeq(KEY s1, KEY s2)
+{
+ for (; *s1 == *s2; s1++, s2++)
+ if (*s1 == 0)
+ return XML_TRUE;
+ return XML_FALSE;
+}
+
+static unsigned long FASTCALL
+hash(KEY s)
+{
+ unsigned long h = 0;
+ while (*s)
+ h = CHAR_HASH(h, *s++);
+ return h;
+}
+
+static NAMED *
+lookup(HASH_TABLE *table, KEY name, size_t createSize)
+{
+ size_t i;
+ if (table->size == 0) {
+ size_t tsize;
+ if (!createSize)
+ return NULL;
+ table->power = INIT_POWER;
+ /* table->size is a power of 2 */
+ table->size = (size_t)1 << INIT_POWER;
+ tsize = table->size * sizeof(NAMED *);
+ table->v = (NAMED **)table->mem->malloc_fcn(tsize);
+ if (!table->v) {
+ table->size = 0;
+ return NULL;
+ }
+ memset(table->v, 0, tsize);
+ i = hash(name) & ((unsigned long)table->size - 1);
+ }
+ else {
+ unsigned long h = hash(name);
+ unsigned long mask = (unsigned long)table->size - 1;
+ unsigned char step = 0;
+ i = h & mask;
+ while (table->v[i]) {
+ if (keyeq(name, table->v[i]->name))
+ return table->v[i];
+ if (!step)
+ step = PROBE_STEP(h, mask, table->power);
+ i < step ? (i += table->size - step) : (i -= step);
+ }
+ if (!createSize)
+ return NULL;
+
+ /* check for overflow (table is half full) */
+ if (table->used >> (table->power - 1)) {
+ unsigned char newPower = table->power + 1;
+ size_t newSize = (size_t)1 << newPower;
+ unsigned long newMask = (unsigned long)newSize - 1;
+ size_t tsize = newSize * sizeof(NAMED *);
+ NAMED **newV = (NAMED **)table->mem->malloc_fcn(tsize);
+ if (!newV)
+ return NULL;
+ memset(newV, 0, tsize);
+ for (i = 0; i < table->size; i++)
+ if (table->v[i]) {
+ unsigned long newHash = hash(table->v[i]->name);
+ size_t j = newHash & newMask;
+ step = 0;
+ while (newV[j]) {
+ if (!step)
+ step = PROBE_STEP(newHash, newMask, newPower);
+ j < step ? (j += newSize - step) : (j -= step);
+ }
+ newV[j] = table->v[i];
+ }
+ table->mem->free_fcn(table->v);
+ table->v = newV;
+ table->power = newPower;
+ table->size = newSize;
+ i = h & newMask;
+ step = 0;
+ while (table->v[i]) {
+ if (!step)
+ step = PROBE_STEP(h, newMask, newPower);
+ i < step ? (i += newSize - step) : (i -= step);
+ }
+ }
+ }
+ table->v[i] = (NAMED *)table->mem->malloc_fcn(createSize);
+ if (!table->v[i])
+ return NULL;
+ memset(table->v[i], 0, createSize);
+ table->v[i]->name = name;
+ (table->used)++;
+ return table->v[i];
+}
+
+static void FASTCALL
+hashTableClear(HASH_TABLE *table)
+{
+ size_t i;
+ for (i = 0; i < table->size; i++) {
+ table->mem->free_fcn(table->v[i]);
+ table->v[i] = NULL;
+ }
+ table->used = 0;
+}
+
+static void FASTCALL
+hashTableDestroy(HASH_TABLE *table)
+{
+ size_t i;
+ for (i = 0; i < table->size; i++)
+ table->mem->free_fcn(table->v[i]);
+ table->mem->free_fcn(table->v);
+}
+
+static void FASTCALL
+hashTableInit(HASH_TABLE *p, const XML_Memory_Handling_Suite *ms)
+{
+ p->power = 0;
+ p->size = 0;
+ p->used = 0;
+ p->v = NULL;
+ p->mem = ms;
+}
+
+static void FASTCALL
+hashTableIterInit(HASH_TABLE_ITER *iter, const HASH_TABLE *table)
+{
+ iter->p = table->v;
+ iter->end = iter->p + table->size;
+}
+
+static NAMED * FASTCALL
+hashTableIterNext(HASH_TABLE_ITER *iter)
+{
+ while (iter->p != iter->end) {
+ NAMED *tem = *(iter->p)++;
+ if (tem)
+ return tem;
+ }
+ return NULL;
+}
+
+static void FASTCALL
+poolInit(STRING_POOL *pool, const XML_Memory_Handling_Suite *ms)
+{
+ pool->blocks = NULL;
+ pool->freeBlocks = NULL;
+ pool->start = NULL;
+ pool->ptr = NULL;
+ pool->end = NULL;
+ pool->mem = ms;
+}
+
+static void FASTCALL
+poolClear(STRING_POOL *pool)
+{
+ if (!pool->freeBlocks)
+ pool->freeBlocks = pool->blocks;
+ else {
+ BLOCK *p = pool->blocks;
+ while (p) {
+ BLOCK *tem = p->next;
+ p->next = pool->freeBlocks;
+ pool->freeBlocks = p;
+ p = tem;
+ }
+ }
+ pool->blocks = NULL;
+ pool->start = NULL;
+ pool->ptr = NULL;
+ pool->end = NULL;
+}
+
+static void FASTCALL
+poolDestroy(STRING_POOL *pool)
+{
+ BLOCK *p = pool->blocks;
+ while (p) {
+ BLOCK *tem = p->next;
+ pool->mem->free_fcn(p);
+ p = tem;
+ }
+ p = pool->freeBlocks;
+ while (p) {
+ BLOCK *tem = p->next;
+ pool->mem->free_fcn(p);
+ p = tem;
+ }
+}
+
+static XML_Char *
+poolAppend(STRING_POOL *pool, const ENCODING *enc,
+ const char *ptr, const char *end)
+{
+ if (!pool->ptr && !poolGrow(pool))
+ return NULL;
+ for (;;) {
+ XmlConvert(enc, &ptr, end, (ICHAR **)&(pool->ptr), (ICHAR *)pool->end);
+ if (ptr == end)
+ break;
+ if (!poolGrow(pool))
+ return NULL;
+ }
+ return pool->start;
+}
+
+static const XML_Char * FASTCALL
+poolCopyString(STRING_POOL *pool, const XML_Char *s)
+{
+ do {
+ if (!poolAppendChar(pool, *s))
+ return NULL;
+ } while (*s++);
+ s = pool->start;
+ poolFinish(pool);
+ return s;
+}
+
+static const XML_Char *
+poolCopyStringN(STRING_POOL *pool, const XML_Char *s, int n)
+{
+ if (!pool->ptr && !poolGrow(pool))
+ return NULL;
+ for (; n > 0; --n, s++) {
+ if (!poolAppendChar(pool, *s))
+ return NULL;
+ }
+ s = pool->start;
+ poolFinish(pool);
+ return s;
+}
+
+static const XML_Char * FASTCALL
+poolAppendString(STRING_POOL *pool, const XML_Char *s)
+{
+ while (*s) {
+ if (!poolAppendChar(pool, *s))
+ return NULL;
+ s++;
+ }
+ return pool->start;
+}
+
+static XML_Char *
+poolStoreString(STRING_POOL *pool, const ENCODING *enc,
+ const char *ptr, const char *end)
+{
+ if (!poolAppend(pool, enc, ptr, end))
+ return NULL;
+ if (pool->ptr == pool->end && !poolGrow(pool))
+ return NULL;
+ *(pool->ptr)++ = 0;
+ return pool->start;
+}
+
+static XML_Bool FASTCALL
+poolGrow(STRING_POOL *pool)
+{
+ if (pool->freeBlocks) {
+ if (pool->start == 0) {
+ pool->blocks = pool->freeBlocks;
+ pool->freeBlocks = pool->freeBlocks->next;
+ pool->blocks->next = NULL;
+ pool->start = pool->blocks->s;
+ pool->end = pool->start + pool->blocks->size;
+ pool->ptr = pool->start;
+ return XML_TRUE;
+ }
+ if (pool->end - pool->start < pool->freeBlocks->size) {
+ BLOCK *tem = pool->freeBlocks->next;
+ pool->freeBlocks->next = pool->blocks;
+ pool->blocks = pool->freeBlocks;
+ pool->freeBlocks = tem;
+ memcpy(pool->blocks->s, pool->start,
+ (pool->end - pool->start) * sizeof(XML_Char));
+ pool->ptr = pool->blocks->s + (pool->ptr - pool->start);
+ pool->start = pool->blocks->s;
+ pool->end = pool->start + pool->blocks->size;
+ return XML_TRUE;
+ }
+ }
+ if (pool->blocks && pool->start == pool->blocks->s) {
+ int blockSize = (int)(pool->end - pool->start)*2;
+ pool->blocks = (BLOCK *)
+ pool->mem->realloc_fcn(pool->blocks,
+ (offsetof(BLOCK, s)
+ + blockSize * sizeof(XML_Char)));
+ if (pool->blocks == NULL)
+ return XML_FALSE;
+ pool->blocks->size = blockSize;
+ pool->ptr = pool->blocks->s + (pool->ptr - pool->start);
+ pool->start = pool->blocks->s;
+ pool->end = pool->start + blockSize;
+ }
+ else {
+ BLOCK *tem;
+ int blockSize = (int)(pool->end - pool->start);
+ if (blockSize < INIT_BLOCK_SIZE)
+ blockSize = INIT_BLOCK_SIZE;
+ else
+ blockSize *= 2;
+ tem = (BLOCK *)pool->mem->malloc_fcn(offsetof(BLOCK, s)
+ + blockSize * sizeof(XML_Char));
+ if (!tem)
+ return XML_FALSE;
+ tem->size = blockSize;
+ tem->next = pool->blocks;
+ pool->blocks = tem;
+ if (pool->ptr != pool->start)
+ memcpy(tem->s, pool->start,
+ (pool->ptr - pool->start) * sizeof(XML_Char));
+ pool->ptr = tem->s + (pool->ptr - pool->start);
+ pool->start = tem->s;
+ pool->end = tem->s + blockSize;
+ }
+ return XML_TRUE;
+}
+
+static int FASTCALL
+nextScaffoldPart(XML_Parser parser)
+{
+ DTD * const dtd = _dtd; /* save one level of indirection */
+ CONTENT_SCAFFOLD * me;
+ int next;
+
+ if (!dtd->scaffIndex) {
+ dtd->scaffIndex = (int *)MALLOC(groupSize * sizeof(int));
+ if (!dtd->scaffIndex)
+ return -1;
+ dtd->scaffIndex[0] = 0;
+ }
+
+ if (dtd->scaffCount >= dtd->scaffSize) {
+ CONTENT_SCAFFOLD *temp;
+ if (dtd->scaffold) {
+ temp = (CONTENT_SCAFFOLD *)
+ REALLOC(dtd->scaffold, dtd->scaffSize * 2 * sizeof(CONTENT_SCAFFOLD));
+ if (temp == NULL)
+ return -1;
+ dtd->scaffSize *= 2;
+ }
+ else {
+ temp = (CONTENT_SCAFFOLD *)MALLOC(INIT_SCAFFOLD_ELEMENTS
+ * sizeof(CONTENT_SCAFFOLD));
+ if (temp == NULL)
+ return -1;
+ dtd->scaffSize = INIT_SCAFFOLD_ELEMENTS;
+ }
+ dtd->scaffold = temp;
+ }
+ next = dtd->scaffCount++;
+ me = &dtd->scaffold[next];
+ if (dtd->scaffLevel) {
+ CONTENT_SCAFFOLD *parent = &dtd->scaffold[dtd->scaffIndex[dtd->scaffLevel-1]];
+ if (parent->lastchild) {
+ dtd->scaffold[parent->lastchild].nextsib = next;
+ }
+ if (!parent->childcnt)
+ parent->firstchild = next;
+ parent->lastchild = next;
+ parent->childcnt++;
+ }
+ me->firstchild = me->lastchild = me->childcnt = me->nextsib = 0;
+ return next;
+}
+
+static void
+build_node(XML_Parser parser,
+ int src_node,
+ XML_Content *dest,
+ XML_Content **contpos,
+ XML_Char **strpos)
+{
+ DTD * const dtd = _dtd; /* save one level of indirection */
+ dest->type = dtd->scaffold[src_node].type;
+ dest->quant = dtd->scaffold[src_node].quant;
+ if (dest->type == XML_CTYPE_NAME) {
+ const XML_Char *src;
+ dest->name = *strpos;
+ src = dtd->scaffold[src_node].name;
+ for (;;) {
+ *(*strpos)++ = *src;
+ if (!*src)
+ break;
+ src++;
+ }
+ dest->numchildren = 0;
+ dest->children = NULL;
+ }
+ else {
+ unsigned int i;
+ int cn;
+ dest->numchildren = dtd->scaffold[src_node].childcnt;
+ dest->children = *contpos;
+ *contpos += dest->numchildren;
+ for (i = 0, cn = dtd->scaffold[src_node].firstchild;
+ i < dest->numchildren;
+ i++, cn = dtd->scaffold[cn].nextsib) {
+ build_node(parser, cn, &(dest->children[i]), contpos, strpos);
+ }
+ dest->name = NULL;
+ }
+}
+
+static XML_Content *
+build_model (XML_Parser parser)
+{
+ DTD * const dtd = _dtd; /* save one level of indirection */
+ XML_Content *ret;
+ XML_Content *cpos;
+ XML_Char * str;
+ int allocsize = (dtd->scaffCount * sizeof(XML_Content)
+ + (dtd->contentStringLen * sizeof(XML_Char)));
+
+ ret = (XML_Content *)MALLOC(allocsize);
+ if (!ret)
+ return NULL;
+
+ str = (XML_Char *) (&ret[dtd->scaffCount]);
+ cpos = &ret[1];
+
+ build_node(parser, 0, ret, &cpos, &str);
+ return ret;
+}
+
+static ELEMENT_TYPE *
+getElementType(XML_Parser parser,
+ const ENCODING *enc,
+ const char *ptr,
+ const char *end)
+{
+ DTD * const dtd = _dtd; /* save one level of indirection */
+ const XML_Char *name = poolStoreString(&dtd->pool, enc, ptr, end);
+ ELEMENT_TYPE *ret;
+
+ if (!name)
+ return NULL;
+ ret = (ELEMENT_TYPE *) lookup(&dtd->elementTypes, name, sizeof(ELEMENT_TYPE));
+ if (!ret)
+ return NULL;
+ if (ret->name != name)
+ poolDiscard(&dtd->pool);
+ else {
+ poolFinish(&dtd->pool);
+ if (!setElementTypePrefix(parser, ret))
+ return NULL;
+ }
+ return ret;
+}
diff --git a/src/lib/expat/xmlrole.c b/src/lib/expat/xmlrole.c
new file mode 100644
index 00000000000..b0aff272432
--- /dev/null
+++ b/src/lib/expat/xmlrole.c
@@ -0,0 +1,1320 @@
+/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+ See the file COPYING for copying permission.
+*/
+
+#include <stddef.h>
+
+#include "intconfig.h"
+
+#include "expat_external.h"
+#include "internal.h"
+#include "xmlrole.h"
+#include "ascii.h"
+
+/* Doesn't check:
+
+ that ,| are not mixed in a model group
+ content of literals
+
+*/
+
+static const char KW_ANY[] = {
+ ASCII_A, ASCII_N, ASCII_Y, '\0' };
+static const char KW_ATTLIST[] = {
+ ASCII_A, ASCII_T, ASCII_T, ASCII_L, ASCII_I, ASCII_S, ASCII_T, '\0' };
+static const char KW_CDATA[] = {
+ ASCII_C, ASCII_D, ASCII_A, ASCII_T, ASCII_A, '\0' };
+static const char KW_DOCTYPE[] = {
+ ASCII_D, ASCII_O, ASCII_C, ASCII_T, ASCII_Y, ASCII_P, ASCII_E, '\0' };
+static const char KW_ELEMENT[] = {
+ ASCII_E, ASCII_L, ASCII_E, ASCII_M, ASCII_E, ASCII_N, ASCII_T, '\0' };
+static const char KW_EMPTY[] = {
+ ASCII_E, ASCII_M, ASCII_P, ASCII_T, ASCII_Y, '\0' };
+static const char KW_ENTITIES[] = {
+ ASCII_E, ASCII_N, ASCII_T, ASCII_I, ASCII_T, ASCII_I, ASCII_E, ASCII_S,
+ '\0' };
+static const char KW_ENTITY[] = {
+ ASCII_E, ASCII_N, ASCII_T, ASCII_I, ASCII_T, ASCII_Y, '\0' };
+static const char KW_FIXED[] = {
+ ASCII_F, ASCII_I, ASCII_X, ASCII_E, ASCII_D, '\0' };
+static const char KW_ID[] = {
+ ASCII_I, ASCII_D, '\0' };
+static const char KW_IDREF[] = {
+ ASCII_I, ASCII_D, ASCII_R, ASCII_E, ASCII_F, '\0' };
+static const char KW_IDREFS[] = {
+ ASCII_I, ASCII_D, ASCII_R, ASCII_E, ASCII_F, ASCII_S, '\0' };
+static const char KW_IGNORE[] = {
+ ASCII_I, ASCII_G, ASCII_N, ASCII_O, ASCII_R, ASCII_E, '\0' };
+static const char KW_IMPLIED[] = {
+ ASCII_I, ASCII_M, ASCII_P, ASCII_L, ASCII_I, ASCII_E, ASCII_D, '\0' };
+static const char KW_INCLUDE[] = {
+ ASCII_I, ASCII_N, ASCII_C, ASCII_L, ASCII_U, ASCII_D, ASCII_E, '\0' };
+static const char KW_NDATA[] = {
+ ASCII_N, ASCII_D, ASCII_A, ASCII_T, ASCII_A, '\0' };
+static const char KW_NMTOKEN[] = {
+ ASCII_N, ASCII_M, ASCII_T, ASCII_O, ASCII_K, ASCII_E, ASCII_N, '\0' };
+static const char KW_NMTOKENS[] = {
+ ASCII_N, ASCII_M, ASCII_T, ASCII_O, ASCII_K, ASCII_E, ASCII_N, ASCII_S,
+ '\0' };
+static const char KW_NOTATION[] =
+ { ASCII_N, ASCII_O, ASCII_T, ASCII_A, ASCII_T, ASCII_I, ASCII_O, ASCII_N,
+ '\0' };
+static const char KW_PCDATA[] = {
+ ASCII_P, ASCII_C, ASCII_D, ASCII_A, ASCII_T, ASCII_A, '\0' };
+static const char KW_PUBLIC[] = {
+ ASCII_P, ASCII_U, ASCII_B, ASCII_L, ASCII_I, ASCII_C, '\0' };
+static const char KW_REQUIRED[] = {
+ ASCII_R, ASCII_E, ASCII_Q, ASCII_U, ASCII_I, ASCII_R, ASCII_E, ASCII_D,
+ '\0' };
+static const char KW_SYSTEM[] = {
+ ASCII_S, ASCII_Y, ASCII_S, ASCII_T, ASCII_E, ASCII_M, '\0' };
+
+#ifndef MIN_BYTES_PER_CHAR
+#define MIN_BYTES_PER_CHAR(enc) ((enc)->minBytesPerChar)
+#endif
+
+#ifdef XML_DTD
+#define setTopLevel(state) \
+ ((state)->handler = ((state)->documentEntity \
+ ? internalSubset \
+ : externalSubset1))
+#else /* not XML_DTD */
+#define setTopLevel(state) ((state)->handler = internalSubset)
+#endif /* not XML_DTD */
+
+typedef int PTRCALL PROLOG_HANDLER(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc);
+
+static PROLOG_HANDLER
+ prolog0, prolog1, prolog2,
+ doctype0, doctype1, doctype2, doctype3, doctype4, doctype5,
+ internalSubset,
+ entity0, entity1, entity2, entity3, entity4, entity5, entity6,
+ entity7, entity8, entity9, entity10,
+ notation0, notation1, notation2, notation3, notation4,
+ attlist0, attlist1, attlist2, attlist3, attlist4, attlist5, attlist6,
+ attlist7, attlist8, attlist9,
+ element0, element1, element2, element3, element4, element5, element6,
+ element7,
+#ifdef XML_DTD
+ externalSubset0, externalSubset1,
+ condSect0, condSect1, condSect2,
+#endif /* XML_DTD */
+ declClose,
+ error;
+
+static int FASTCALL common(PROLOG_STATE *state, int tok);
+
+static int PTRCALL
+prolog0(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ state->handler = prolog1;
+ return XML_ROLE_NONE;
+ case XML_TOK_XML_DECL:
+ state->handler = prolog1;
+ return XML_ROLE_XML_DECL;
+ case XML_TOK_PI:
+ state->handler = prolog1;
+ return XML_ROLE_PI;
+ case XML_TOK_COMMENT:
+ state->handler = prolog1;
+ return XML_ROLE_COMMENT;
+ case XML_TOK_BOM:
+ return XML_ROLE_NONE;
+ case XML_TOK_DECL_OPEN:
+ if (!XmlNameMatchesAscii(enc,
+ ptr + 2 * MIN_BYTES_PER_CHAR(enc),
+ end,
+ KW_DOCTYPE))
+ break;
+ state->handler = doctype0;
+ return XML_ROLE_DOCTYPE_NONE;
+ case XML_TOK_INSTANCE_START:
+ state->handler = error;
+ return XML_ROLE_INSTANCE_START;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+prolog1(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_NONE;
+ case XML_TOK_PI:
+ return XML_ROLE_PI;
+ case XML_TOK_COMMENT:
+ return XML_ROLE_COMMENT;
+ case XML_TOK_BOM:
+ return XML_ROLE_NONE;
+ case XML_TOK_DECL_OPEN:
+ if (!XmlNameMatchesAscii(enc,
+ ptr + 2 * MIN_BYTES_PER_CHAR(enc),
+ end,
+ KW_DOCTYPE))
+ break;
+ state->handler = doctype0;
+ return XML_ROLE_DOCTYPE_NONE;
+ case XML_TOK_INSTANCE_START:
+ state->handler = error;
+ return XML_ROLE_INSTANCE_START;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+prolog2(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_NONE;
+ case XML_TOK_PI:
+ return XML_ROLE_PI;
+ case XML_TOK_COMMENT:
+ return XML_ROLE_COMMENT;
+ case XML_TOK_INSTANCE_START:
+ state->handler = error;
+ return XML_ROLE_INSTANCE_START;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+doctype0(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_DOCTYPE_NONE;
+ case XML_TOK_NAME:
+ case XML_TOK_PREFIXED_NAME:
+ state->handler = doctype1;
+ return XML_ROLE_DOCTYPE_NAME;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+doctype1(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_DOCTYPE_NONE;
+ case XML_TOK_OPEN_BRACKET:
+ state->handler = internalSubset;
+ return XML_ROLE_DOCTYPE_INTERNAL_SUBSET;
+ case XML_TOK_DECL_CLOSE:
+ state->handler = prolog2;
+ return XML_ROLE_DOCTYPE_CLOSE;
+ case XML_TOK_NAME:
+ if (XmlNameMatchesAscii(enc, ptr, end, KW_SYSTEM)) {
+ state->handler = doctype3;
+ return XML_ROLE_DOCTYPE_NONE;
+ }
+ if (XmlNameMatchesAscii(enc, ptr, end, KW_PUBLIC)) {
+ state->handler = doctype2;
+ return XML_ROLE_DOCTYPE_NONE;
+ }
+ break;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+doctype2(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_DOCTYPE_NONE;
+ case XML_TOK_LITERAL:
+ state->handler = doctype3;
+ return XML_ROLE_DOCTYPE_PUBLIC_ID;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+doctype3(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_DOCTYPE_NONE;
+ case XML_TOK_LITERAL:
+ state->handler = doctype4;
+ return XML_ROLE_DOCTYPE_SYSTEM_ID;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+doctype4(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_DOCTYPE_NONE;
+ case XML_TOK_OPEN_BRACKET:
+ state->handler = internalSubset;
+ return XML_ROLE_DOCTYPE_INTERNAL_SUBSET;
+ case XML_TOK_DECL_CLOSE:
+ state->handler = prolog2;
+ return XML_ROLE_DOCTYPE_CLOSE;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+doctype5(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_DOCTYPE_NONE;
+ case XML_TOK_DECL_CLOSE:
+ state->handler = prolog2;
+ return XML_ROLE_DOCTYPE_CLOSE;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+internalSubset(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_NONE;
+ case XML_TOK_DECL_OPEN:
+ if (XmlNameMatchesAscii(enc,
+ ptr + 2 * MIN_BYTES_PER_CHAR(enc),
+ end,
+ KW_ENTITY)) {
+ state->handler = entity0;
+ return XML_ROLE_ENTITY_NONE;
+ }
+ if (XmlNameMatchesAscii(enc,
+ ptr + 2 * MIN_BYTES_PER_CHAR(enc),
+ end,
+ KW_ATTLIST)) {
+ state->handler = attlist0;
+ return XML_ROLE_ATTLIST_NONE;
+ }
+ if (XmlNameMatchesAscii(enc,
+ ptr + 2 * MIN_BYTES_PER_CHAR(enc),
+ end,
+ KW_ELEMENT)) {
+ state->handler = element0;
+ return XML_ROLE_ELEMENT_NONE;
+ }
+ if (XmlNameMatchesAscii(enc,
+ ptr + 2 * MIN_BYTES_PER_CHAR(enc),
+ end,
+ KW_NOTATION)) {
+ state->handler = notation0;
+ return XML_ROLE_NOTATION_NONE;
+ }
+ break;
+ case XML_TOK_PI:
+ return XML_ROLE_PI;
+ case XML_TOK_COMMENT:
+ return XML_ROLE_COMMENT;
+ case XML_TOK_PARAM_ENTITY_REF:
+ return XML_ROLE_PARAM_ENTITY_REF;
+ case XML_TOK_CLOSE_BRACKET:
+ state->handler = doctype5;
+ return XML_ROLE_DOCTYPE_NONE;
+ case XML_TOK_NONE:
+ return XML_ROLE_NONE;
+ }
+ return common(state, tok);
+}
+
+#ifdef XML_DTD
+
+static int PTRCALL
+externalSubset0(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ state->handler = externalSubset1;
+ if (tok == XML_TOK_XML_DECL)
+ return XML_ROLE_TEXT_DECL;
+ return externalSubset1(state, tok, ptr, end, enc);
+}
+
+static int PTRCALL
+externalSubset1(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_COND_SECT_OPEN:
+ state->handler = condSect0;
+ return XML_ROLE_NONE;
+ case XML_TOK_COND_SECT_CLOSE:
+ if (state->includeLevel == 0)
+ break;
+ state->includeLevel -= 1;
+ return XML_ROLE_NONE;
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_NONE;
+ case XML_TOK_CLOSE_BRACKET:
+ break;
+ case XML_TOK_NONE:
+ if (state->includeLevel)
+ break;
+ return XML_ROLE_NONE;
+ default:
+ return internalSubset(state, tok, ptr, end, enc);
+ }
+ return common(state, tok);
+}
+
+#endif /* XML_DTD */
+
+static int PTRCALL
+entity0(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ENTITY_NONE;
+ case XML_TOK_PERCENT:
+ state->handler = entity1;
+ return XML_ROLE_ENTITY_NONE;
+ case XML_TOK_NAME:
+ state->handler = entity2;
+ return XML_ROLE_GENERAL_ENTITY_NAME;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+entity1(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ENTITY_NONE;
+ case XML_TOK_NAME:
+ state->handler = entity7;
+ return XML_ROLE_PARAM_ENTITY_NAME;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+entity2(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ENTITY_NONE;
+ case XML_TOK_NAME:
+ if (XmlNameMatchesAscii(enc, ptr, end, KW_SYSTEM)) {
+ state->handler = entity4;
+ return XML_ROLE_ENTITY_NONE;
+ }
+ if (XmlNameMatchesAscii(enc, ptr, end, KW_PUBLIC)) {
+ state->handler = entity3;
+ return XML_ROLE_ENTITY_NONE;
+ }
+ break;
+ case XML_TOK_LITERAL:
+ state->handler = declClose;
+ state->role_none = XML_ROLE_ENTITY_NONE;
+ return XML_ROLE_ENTITY_VALUE;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+entity3(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ENTITY_NONE;
+ case XML_TOK_LITERAL:
+ state->handler = entity4;
+ return XML_ROLE_ENTITY_PUBLIC_ID;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+entity4(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ENTITY_NONE;
+ case XML_TOK_LITERAL:
+ state->handler = entity5;
+ return XML_ROLE_ENTITY_SYSTEM_ID;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+entity5(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ENTITY_NONE;
+ case XML_TOK_DECL_CLOSE:
+ setTopLevel(state);
+ return XML_ROLE_ENTITY_COMPLETE;
+ case XML_TOK_NAME:
+ if (XmlNameMatchesAscii(enc, ptr, end, KW_NDATA)) {
+ state->handler = entity6;
+ return XML_ROLE_ENTITY_NONE;
+ }
+ break;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+entity6(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ENTITY_NONE;
+ case XML_TOK_NAME:
+ state->handler = declClose;
+ state->role_none = XML_ROLE_ENTITY_NONE;
+ return XML_ROLE_ENTITY_NOTATION_NAME;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+entity7(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ENTITY_NONE;
+ case XML_TOK_NAME:
+ if (XmlNameMatchesAscii(enc, ptr, end, KW_SYSTEM)) {
+ state->handler = entity9;
+ return XML_ROLE_ENTITY_NONE;
+ }
+ if (XmlNameMatchesAscii(enc, ptr, end, KW_PUBLIC)) {
+ state->handler = entity8;
+ return XML_ROLE_ENTITY_NONE;
+ }
+ break;
+ case XML_TOK_LITERAL:
+ state->handler = declClose;
+ state->role_none = XML_ROLE_ENTITY_NONE;
+ return XML_ROLE_ENTITY_VALUE;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+entity8(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ENTITY_NONE;
+ case XML_TOK_LITERAL:
+ state->handler = entity9;
+ return XML_ROLE_ENTITY_PUBLIC_ID;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+entity9(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ENTITY_NONE;
+ case XML_TOK_LITERAL:
+ state->handler = entity10;
+ return XML_ROLE_ENTITY_SYSTEM_ID;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+entity10(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ENTITY_NONE;
+ case XML_TOK_DECL_CLOSE:
+ setTopLevel(state);
+ return XML_ROLE_ENTITY_COMPLETE;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+notation0(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_NOTATION_NONE;
+ case XML_TOK_NAME:
+ state->handler = notation1;
+ return XML_ROLE_NOTATION_NAME;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+notation1(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_NOTATION_NONE;
+ case XML_TOK_NAME:
+ if (XmlNameMatchesAscii(enc, ptr, end, KW_SYSTEM)) {
+ state->handler = notation3;
+ return XML_ROLE_NOTATION_NONE;
+ }
+ if (XmlNameMatchesAscii(enc, ptr, end, KW_PUBLIC)) {
+ state->handler = notation2;
+ return XML_ROLE_NOTATION_NONE;
+ }
+ break;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+notation2(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_NOTATION_NONE;
+ case XML_TOK_LITERAL:
+ state->handler = notation4;
+ return XML_ROLE_NOTATION_PUBLIC_ID;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+notation3(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_NOTATION_NONE;
+ case XML_TOK_LITERAL:
+ state->handler = declClose;
+ state->role_none = XML_ROLE_NOTATION_NONE;
+ return XML_ROLE_NOTATION_SYSTEM_ID;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+notation4(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_NOTATION_NONE;
+ case XML_TOK_LITERAL:
+ state->handler = declClose;
+ state->role_none = XML_ROLE_NOTATION_NONE;
+ return XML_ROLE_NOTATION_SYSTEM_ID;
+ case XML_TOK_DECL_CLOSE:
+ setTopLevel(state);
+ return XML_ROLE_NOTATION_NO_SYSTEM_ID;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+attlist0(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ATTLIST_NONE;
+ case XML_TOK_NAME:
+ case XML_TOK_PREFIXED_NAME:
+ state->handler = attlist1;
+ return XML_ROLE_ATTLIST_ELEMENT_NAME;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+attlist1(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ATTLIST_NONE;
+ case XML_TOK_DECL_CLOSE:
+ setTopLevel(state);
+ return XML_ROLE_ATTLIST_NONE;
+ case XML_TOK_NAME:
+ case XML_TOK_PREFIXED_NAME:
+ state->handler = attlist2;
+ return XML_ROLE_ATTRIBUTE_NAME;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+attlist2(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ATTLIST_NONE;
+ case XML_TOK_NAME:
+ {
+ static const char * const types[] = {
+ KW_CDATA,
+ KW_ID,
+ KW_IDREF,
+ KW_IDREFS,
+ KW_ENTITY,
+ KW_ENTITIES,
+ KW_NMTOKEN,
+ KW_NMTOKENS,
+ };
+ int i;
+ for (i = 0; i < (int)(sizeof(types)/sizeof(types[0])); i++)
+ if (XmlNameMatchesAscii(enc, ptr, end, types[i])) {
+ state->handler = attlist8;
+ return XML_ROLE_ATTRIBUTE_TYPE_CDATA + i;
+ }
+ }
+ if (XmlNameMatchesAscii(enc, ptr, end, KW_NOTATION)) {
+ state->handler = attlist5;
+ return XML_ROLE_ATTLIST_NONE;
+ }
+ break;
+ case XML_TOK_OPEN_PAREN:
+ state->handler = attlist3;
+ return XML_ROLE_ATTLIST_NONE;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+attlist3(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ATTLIST_NONE;
+ case XML_TOK_NMTOKEN:
+ case XML_TOK_NAME:
+ case XML_TOK_PREFIXED_NAME:
+ state->handler = attlist4;
+ return XML_ROLE_ATTRIBUTE_ENUM_VALUE;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+attlist4(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ATTLIST_NONE;
+ case XML_TOK_CLOSE_PAREN:
+ state->handler = attlist8;
+ return XML_ROLE_ATTLIST_NONE;
+ case XML_TOK_OR:
+ state->handler = attlist3;
+ return XML_ROLE_ATTLIST_NONE;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+attlist5(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ATTLIST_NONE;
+ case XML_TOK_OPEN_PAREN:
+ state->handler = attlist6;
+ return XML_ROLE_ATTLIST_NONE;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+attlist6(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ATTLIST_NONE;
+ case XML_TOK_NAME:
+ state->handler = attlist7;
+ return XML_ROLE_ATTRIBUTE_NOTATION_VALUE;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+attlist7(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ATTLIST_NONE;
+ case XML_TOK_CLOSE_PAREN:
+ state->handler = attlist8;
+ return XML_ROLE_ATTLIST_NONE;
+ case XML_TOK_OR:
+ state->handler = attlist6;
+ return XML_ROLE_ATTLIST_NONE;
+ }
+ return common(state, tok);
+}
+
+/* default value */
+static int PTRCALL
+attlist8(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ATTLIST_NONE;
+ case XML_TOK_POUND_NAME:
+ if (XmlNameMatchesAscii(enc,
+ ptr + MIN_BYTES_PER_CHAR(enc),
+ end,
+ KW_IMPLIED)) {
+ state->handler = attlist1;
+ return XML_ROLE_IMPLIED_ATTRIBUTE_VALUE;
+ }
+ if (XmlNameMatchesAscii(enc,
+ ptr + MIN_BYTES_PER_CHAR(enc),
+ end,
+ KW_REQUIRED)) {
+ state->handler = attlist1;
+ return XML_ROLE_REQUIRED_ATTRIBUTE_VALUE;
+ }
+ if (XmlNameMatchesAscii(enc,
+ ptr + MIN_BYTES_PER_CHAR(enc),
+ end,
+ KW_FIXED)) {
+ state->handler = attlist9;
+ return XML_ROLE_ATTLIST_NONE;
+ }
+ break;
+ case XML_TOK_LITERAL:
+ state->handler = attlist1;
+ return XML_ROLE_DEFAULT_ATTRIBUTE_VALUE;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+attlist9(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ATTLIST_NONE;
+ case XML_TOK_LITERAL:
+ state->handler = attlist1;
+ return XML_ROLE_FIXED_ATTRIBUTE_VALUE;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+element0(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ELEMENT_NONE;
+ case XML_TOK_NAME:
+ case XML_TOK_PREFIXED_NAME:
+ state->handler = element1;
+ return XML_ROLE_ELEMENT_NAME;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+element1(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ELEMENT_NONE;
+ case XML_TOK_NAME:
+ if (XmlNameMatchesAscii(enc, ptr, end, KW_EMPTY)) {
+ state->handler = declClose;
+ state->role_none = XML_ROLE_ELEMENT_NONE;
+ return XML_ROLE_CONTENT_EMPTY;
+ }
+ if (XmlNameMatchesAscii(enc, ptr, end, KW_ANY)) {
+ state->handler = declClose;
+ state->role_none = XML_ROLE_ELEMENT_NONE;
+ return XML_ROLE_CONTENT_ANY;
+ }
+ break;
+ case XML_TOK_OPEN_PAREN:
+ state->handler = element2;
+ state->level = 1;
+ return XML_ROLE_GROUP_OPEN;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+element2(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ELEMENT_NONE;
+ case XML_TOK_POUND_NAME:
+ if (XmlNameMatchesAscii(enc,
+ ptr + MIN_BYTES_PER_CHAR(enc),
+ end,
+ KW_PCDATA)) {
+ state->handler = element3;
+ return XML_ROLE_CONTENT_PCDATA;
+ }
+ break;
+ case XML_TOK_OPEN_PAREN:
+ state->level = 2;
+ state->handler = element6;
+ return XML_ROLE_GROUP_OPEN;
+ case XML_TOK_NAME:
+ case XML_TOK_PREFIXED_NAME:
+ state->handler = element7;
+ return XML_ROLE_CONTENT_ELEMENT;
+ case XML_TOK_NAME_QUESTION:
+ state->handler = element7;
+ return XML_ROLE_CONTENT_ELEMENT_OPT;
+ case XML_TOK_NAME_ASTERISK:
+ state->handler = element7;
+ return XML_ROLE_CONTENT_ELEMENT_REP;
+ case XML_TOK_NAME_PLUS:
+ state->handler = element7;
+ return XML_ROLE_CONTENT_ELEMENT_PLUS;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+element3(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ELEMENT_NONE;
+ case XML_TOK_CLOSE_PAREN:
+ state->handler = declClose;
+ state->role_none = XML_ROLE_ELEMENT_NONE;
+ return XML_ROLE_GROUP_CLOSE;
+ case XML_TOK_CLOSE_PAREN_ASTERISK:
+ state->handler = declClose;
+ state->role_none = XML_ROLE_ELEMENT_NONE;
+ return XML_ROLE_GROUP_CLOSE_REP;
+ case XML_TOK_OR:
+ state->handler = element4;
+ return XML_ROLE_ELEMENT_NONE;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+element4(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ELEMENT_NONE;
+ case XML_TOK_NAME:
+ case XML_TOK_PREFIXED_NAME:
+ state->handler = element5;
+ return XML_ROLE_CONTENT_ELEMENT;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+element5(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ELEMENT_NONE;
+ case XML_TOK_CLOSE_PAREN_ASTERISK:
+ state->handler = declClose;
+ state->role_none = XML_ROLE_ELEMENT_NONE;
+ return XML_ROLE_GROUP_CLOSE_REP;
+ case XML_TOK_OR:
+ state->handler = element4;
+ return XML_ROLE_ELEMENT_NONE;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+element6(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ELEMENT_NONE;
+ case XML_TOK_OPEN_PAREN:
+ state->level += 1;
+ return XML_ROLE_GROUP_OPEN;
+ case XML_TOK_NAME:
+ case XML_TOK_PREFIXED_NAME:
+ state->handler = element7;
+ return XML_ROLE_CONTENT_ELEMENT;
+ case XML_TOK_NAME_QUESTION:
+ state->handler = element7;
+ return XML_ROLE_CONTENT_ELEMENT_OPT;
+ case XML_TOK_NAME_ASTERISK:
+ state->handler = element7;
+ return XML_ROLE_CONTENT_ELEMENT_REP;
+ case XML_TOK_NAME_PLUS:
+ state->handler = element7;
+ return XML_ROLE_CONTENT_ELEMENT_PLUS;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+element7(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_ELEMENT_NONE;
+ case XML_TOK_CLOSE_PAREN:
+ state->level -= 1;
+ if (state->level == 0) {
+ state->handler = declClose;
+ state->role_none = XML_ROLE_ELEMENT_NONE;
+ }
+ return XML_ROLE_GROUP_CLOSE;
+ case XML_TOK_CLOSE_PAREN_ASTERISK:
+ state->level -= 1;
+ if (state->level == 0) {
+ state->handler = declClose;
+ state->role_none = XML_ROLE_ELEMENT_NONE;
+ }
+ return XML_ROLE_GROUP_CLOSE_REP;
+ case XML_TOK_CLOSE_PAREN_QUESTION:
+ state->level -= 1;
+ if (state->level == 0) {
+ state->handler = declClose;
+ state->role_none = XML_ROLE_ELEMENT_NONE;
+ }
+ return XML_ROLE_GROUP_CLOSE_OPT;
+ case XML_TOK_CLOSE_PAREN_PLUS:
+ state->level -= 1;
+ if (state->level == 0) {
+ state->handler = declClose;
+ state->role_none = XML_ROLE_ELEMENT_NONE;
+ }
+ return XML_ROLE_GROUP_CLOSE_PLUS;
+ case XML_TOK_COMMA:
+ state->handler = element6;
+ return XML_ROLE_GROUP_SEQUENCE;
+ case XML_TOK_OR:
+ state->handler = element6;
+ return XML_ROLE_GROUP_CHOICE;
+ }
+ return common(state, tok);
+}
+
+#ifdef XML_DTD
+
+static int PTRCALL
+condSect0(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_NONE;
+ case XML_TOK_NAME:
+ if (XmlNameMatchesAscii(enc, ptr, end, KW_INCLUDE)) {
+ state->handler = condSect1;
+ return XML_ROLE_NONE;
+ }
+ if (XmlNameMatchesAscii(enc, ptr, end, KW_IGNORE)) {
+ state->handler = condSect2;
+ return XML_ROLE_NONE;
+ }
+ break;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+condSect1(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_NONE;
+ case XML_TOK_OPEN_BRACKET:
+ state->handler = externalSubset1;
+ state->includeLevel += 1;
+ return XML_ROLE_NONE;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+condSect2(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return XML_ROLE_NONE;
+ case XML_TOK_OPEN_BRACKET:
+ state->handler = externalSubset1;
+ return XML_ROLE_IGNORE_SECT;
+ }
+ return common(state, tok);
+}
+
+#endif /* XML_DTD */
+
+static int PTRCALL
+declClose(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ switch (tok) {
+ case XML_TOK_PROLOG_S:
+ return state->role_none;
+ case XML_TOK_DECL_CLOSE:
+ setTopLevel(state);
+ return state->role_none;
+ }
+ return common(state, tok);
+}
+
+static int PTRCALL
+error(PROLOG_STATE *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc)
+{
+ return XML_ROLE_NONE;
+}
+
+static int FASTCALL
+common(PROLOG_STATE *state, int tok)
+{
+#ifdef XML_DTD
+ if (!state->documentEntity && tok == XML_TOK_PARAM_ENTITY_REF)
+ return XML_ROLE_INNER_PARAM_ENTITY_REF;
+#endif
+ state->handler = error;
+ return XML_ROLE_ERROR;
+}
+
+void
+XmlPrologStateInit(PROLOG_STATE *state)
+{
+ state->handler = prolog0;
+#ifdef XML_DTD
+ state->documentEntity = 1;
+ state->includeLevel = 0;
+ state->inEntityValue = 0;
+#endif /* XML_DTD */
+}
+
+#ifdef XML_DTD
+
+void
+XmlPrologStateInitExternalEntity(PROLOG_STATE *state)
+{
+ state->handler = externalSubset0;
+ state->documentEntity = 0;
+ state->includeLevel = 0;
+}
+
+#endif /* XML_DTD */
diff --git a/src/lib/expat/xmlrole.h b/src/lib/expat/xmlrole.h
new file mode 100644
index 00000000000..4dd9f06f976
--- /dev/null
+++ b/src/lib/expat/xmlrole.h
@@ -0,0 +1,114 @@
+/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+ See the file COPYING for copying permission.
+*/
+
+#ifndef XmlRole_INCLUDED
+#define XmlRole_INCLUDED 1
+
+#ifdef __VMS
+/* 0 1 2 3 0 1 2 3
+ 1234567890123456789012345678901 1234567890123456789012345678901 */
+#define XmlPrologStateInitExternalEntity XmlPrologStateInitExternalEnt
+#endif
+
+#include "xmltok.h"
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+enum {
+ XML_ROLE_ERROR = -1,
+ XML_ROLE_NONE = 0,
+ XML_ROLE_XML_DECL,
+ XML_ROLE_INSTANCE_START,
+ XML_ROLE_DOCTYPE_NONE,
+ XML_ROLE_DOCTYPE_NAME,
+ XML_ROLE_DOCTYPE_SYSTEM_ID,
+ XML_ROLE_DOCTYPE_PUBLIC_ID,
+ XML_ROLE_DOCTYPE_INTERNAL_SUBSET,
+ XML_ROLE_DOCTYPE_CLOSE,
+ XML_ROLE_GENERAL_ENTITY_NAME,
+ XML_ROLE_PARAM_ENTITY_NAME,
+ XML_ROLE_ENTITY_NONE,
+ XML_ROLE_ENTITY_VALUE,
+ XML_ROLE_ENTITY_SYSTEM_ID,
+ XML_ROLE_ENTITY_PUBLIC_ID,
+ XML_ROLE_ENTITY_COMPLETE,
+ XML_ROLE_ENTITY_NOTATION_NAME,
+ XML_ROLE_NOTATION_NONE,
+ XML_ROLE_NOTATION_NAME,
+ XML_ROLE_NOTATION_SYSTEM_ID,
+ XML_ROLE_NOTATION_NO_SYSTEM_ID,
+ XML_ROLE_NOTATION_PUBLIC_ID,
+ XML_ROLE_ATTRIBUTE_NAME,
+ XML_ROLE_ATTRIBUTE_TYPE_CDATA,
+ XML_ROLE_ATTRIBUTE_TYPE_ID,
+ XML_ROLE_ATTRIBUTE_TYPE_IDREF,
+ XML_ROLE_ATTRIBUTE_TYPE_IDREFS,
+ XML_ROLE_ATTRIBUTE_TYPE_ENTITY,
+ XML_ROLE_ATTRIBUTE_TYPE_ENTITIES,
+ XML_ROLE_ATTRIBUTE_TYPE_NMTOKEN,
+ XML_ROLE_ATTRIBUTE_TYPE_NMTOKENS,
+ XML_ROLE_ATTRIBUTE_ENUM_VALUE,
+ XML_ROLE_ATTRIBUTE_NOTATION_VALUE,
+ XML_ROLE_ATTLIST_NONE,
+ XML_ROLE_ATTLIST_ELEMENT_NAME,
+ XML_ROLE_IMPLIED_ATTRIBUTE_VALUE,
+ XML_ROLE_REQUIRED_ATTRIBUTE_VALUE,
+ XML_ROLE_DEFAULT_ATTRIBUTE_VALUE,
+ XML_ROLE_FIXED_ATTRIBUTE_VALUE,
+ XML_ROLE_ELEMENT_NONE,
+ XML_ROLE_ELEMENT_NAME,
+ XML_ROLE_CONTENT_ANY,
+ XML_ROLE_CONTENT_EMPTY,
+ XML_ROLE_CONTENT_PCDATA,
+ XML_ROLE_GROUP_OPEN,
+ XML_ROLE_GROUP_CLOSE,
+ XML_ROLE_GROUP_CLOSE_REP,
+ XML_ROLE_GROUP_CLOSE_OPT,
+ XML_ROLE_GROUP_CLOSE_PLUS,
+ XML_ROLE_GROUP_CHOICE,
+ XML_ROLE_GROUP_SEQUENCE,
+ XML_ROLE_CONTENT_ELEMENT,
+ XML_ROLE_CONTENT_ELEMENT_REP,
+ XML_ROLE_CONTENT_ELEMENT_OPT,
+ XML_ROLE_CONTENT_ELEMENT_PLUS,
+ XML_ROLE_PI,
+ XML_ROLE_COMMENT,
+#ifdef XML_DTD
+ XML_ROLE_TEXT_DECL,
+ XML_ROLE_IGNORE_SECT,
+ XML_ROLE_INNER_PARAM_ENTITY_REF,
+#endif /* XML_DTD */
+ XML_ROLE_PARAM_ENTITY_REF
+};
+
+typedef struct prolog_state {
+ int (PTRCALL *handler) (struct prolog_state *state,
+ int tok,
+ const char *ptr,
+ const char *end,
+ const ENCODING *enc);
+ unsigned level;
+ int role_none;
+#ifdef XML_DTD
+ unsigned includeLevel;
+ int documentEntity;
+ int inEntityValue;
+#endif /* XML_DTD */
+} PROLOG_STATE;
+
+void XmlPrologStateInit(PROLOG_STATE *);
+#ifdef XML_DTD
+void XmlPrologStateInitExternalEntity(PROLOG_STATE *);
+#endif /* XML_DTD */
+
+#define XmlTokenRole(state, tok, ptr, end, enc) \
+ (((state)->handler)(state, tok, ptr, end, enc))
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* not XmlRole_INCLUDED */
diff --git a/src/lib/expat/xmltok.c b/src/lib/expat/xmltok.c
new file mode 100644
index 00000000000..3a7275adb25
--- /dev/null
+++ b/src/lib/expat/xmltok.c
@@ -0,0 +1,1629 @@
+/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+ See the file COPYING for copying permission.
+*/
+
+#include <stddef.h>
+
+#include "intconfig.h"
+
+#include "expat_external.h"
+#include "internal.h"
+#include "xmltok.h"
+#include "nametab.h"
+
+#ifdef XML_DTD
+#define IGNORE_SECTION_TOK_VTABLE , PREFIX(ignoreSectionTok)
+#else
+#define IGNORE_SECTION_TOK_VTABLE /* as nothing */
+#endif
+
+#define VTABLE1 \
+ { PREFIX(prologTok), PREFIX(contentTok), \
+ PREFIX(cdataSectionTok) IGNORE_SECTION_TOK_VTABLE }, \
+ { PREFIX(attributeValueTok), PREFIX(entityValueTok) }, \
+ PREFIX(sameName), \
+ PREFIX(nameMatchesAscii), \
+ PREFIX(nameLength), \
+ PREFIX(skipS), \
+ PREFIX(getAtts), \
+ PREFIX(charRefNumber), \
+ PREFIX(predefinedEntityName), \
+ PREFIX(updatePosition), \
+ PREFIX(isPublicId)
+
+#define VTABLE VTABLE1, PREFIX(toUtf8), PREFIX(toUtf16)
+
+#define UCS2_GET_NAMING(pages, hi, lo) \
+ (namingBitmap[(pages[hi] << 3) + ((lo) >> 5)] & (1 << ((lo) & 0x1F)))
+
+/* A 2 byte UTF-8 representation splits the characters 11 bits between
+ the bottom 5 and 6 bits of the bytes. We need 8 bits to index into
+ pages, 3 bits to add to that index and 5 bits to generate the mask.
+*/
+#define UTF8_GET_NAMING2(pages, byte) \
+ (namingBitmap[((pages)[(((byte)[0]) >> 2) & 7] << 3) \
+ + ((((byte)[0]) & 3) << 1) \
+ + ((((byte)[1]) >> 5) & 1)] \
+ & (1 << (((byte)[1]) & 0x1F)))
+
+/* A 3 byte UTF-8 representation splits the characters 16 bits between
+ the bottom 4, 6 and 6 bits of the bytes. We need 8 bits to index
+ into pages, 3 bits to add to that index and 5 bits to generate the
+ mask.
+*/
+#define UTF8_GET_NAMING3(pages, byte) \
+ (namingBitmap[((pages)[((((byte)[0]) & 0xF) << 4) \
+ + ((((byte)[1]) >> 2) & 0xF)] \
+ << 3) \
+ + ((((byte)[1]) & 3) << 1) \
+ + ((((byte)[2]) >> 5) & 1)] \
+ & (1 << (((byte)[2]) & 0x1F)))
+
+#define UTF8_GET_NAMING(pages, p, n) \
+ ((n) == 2 \
+ ? UTF8_GET_NAMING2(pages, (const unsigned char *)(p)) \
+ : ((n) == 3 \
+ ? UTF8_GET_NAMING3(pages, (const unsigned char *)(p)) \
+ : 0))
+
+/* Detection of invalid UTF-8 sequences is based on Table 3.1B
+ of Unicode 3.2: http://www.unicode.org/unicode/reports/tr28/
+ with the additional restriction of not allowing the Unicode
+ code points 0xFFFF and 0xFFFE (sequences EF,BF,BF and EF,BF,BE).
+ Implementation details:
+ (A & 0x80) == 0 means A < 0x80
+ and
+ (A & 0xC0) == 0xC0 means A > 0xBF
+*/
+
+#define UTF8_INVALID2(p) \
+ ((*p) < 0xC2 || ((p)[1] & 0x80) == 0 || ((p)[1] & 0xC0) == 0xC0)
+
+#define UTF8_INVALID3(p) \
+ (((p)[2] & 0x80) == 0 \
+ || \
+ ((*p) == 0xEF && (p)[1] == 0xBF \
+ ? \
+ (p)[2] > 0xBD \
+ : \
+ ((p)[2] & 0xC0) == 0xC0) \
+ || \
+ ((*p) == 0xE0 \
+ ? \
+ (p)[1] < 0xA0 || ((p)[1] & 0xC0) == 0xC0 \
+ : \
+ ((p)[1] & 0x80) == 0 \
+ || \
+ ((*p) == 0xED ? (p)[1] > 0x9F : ((p)[1] & 0xC0) == 0xC0)))
+
+#define UTF8_INVALID4(p) \
+ (((p)[3] & 0x80) == 0 || ((p)[3] & 0xC0) == 0xC0 \
+ || \
+ ((p)[2] & 0x80) == 0 || ((p)[2] & 0xC0) == 0xC0 \
+ || \
+ ((*p) == 0xF0 \
+ ? \
+ (p)[1] < 0x90 || ((p)[1] & 0xC0) == 0xC0 \
+ : \
+ ((p)[1] & 0x80) == 0 \
+ || \
+ ((*p) == 0xF4 ? (p)[1] > 0x8F : ((p)[1] & 0xC0) == 0xC0)))
+
+static int PTRFASTCALL
+isNever(const ENCODING *enc, const char *p)
+{
+ return 0;
+}
+
+static int PTRFASTCALL
+utf8_isName2(const ENCODING *enc, const char *p)
+{
+ return UTF8_GET_NAMING2(namePages, (const unsigned char *)p);
+}
+
+static int PTRFASTCALL
+utf8_isName3(const ENCODING *enc, const char *p)
+{
+ return UTF8_GET_NAMING3(namePages, (const unsigned char *)p);
+}
+
+#define utf8_isName4 isNever
+
+static int PTRFASTCALL
+utf8_isNmstrt2(const ENCODING *enc, const char *p)
+{
+ return UTF8_GET_NAMING2(nmstrtPages, (const unsigned char *)p);
+}
+
+static int PTRFASTCALL
+utf8_isNmstrt3(const ENCODING *enc, const char *p)
+{
+ return UTF8_GET_NAMING3(nmstrtPages, (const unsigned char *)p);
+}
+
+#define utf8_isNmstrt4 isNever
+
+static int PTRFASTCALL
+utf8_isInvalid2(const ENCODING *enc, const char *p)
+{
+ return UTF8_INVALID2((const unsigned char *)p);
+}
+
+static int PTRFASTCALL
+utf8_isInvalid3(const ENCODING *enc, const char *p)
+{
+ return UTF8_INVALID3((const unsigned char *)p);
+}
+
+static int PTRFASTCALL
+utf8_isInvalid4(const ENCODING *enc, const char *p)
+{
+ return UTF8_INVALID4((const unsigned char *)p);
+}
+
+struct normal_encoding {
+ ENCODING enc;
+ unsigned char type[256];
+#ifdef XML_MIN_SIZE
+ int (PTRFASTCALL *byteType)(const ENCODING *, const char *);
+ int (PTRFASTCALL *isNameMin)(const ENCODING *, const char *);
+ int (PTRFASTCALL *isNmstrtMin)(const ENCODING *, const char *);
+ int (PTRFASTCALL *byteToAscii)(const ENCODING *, const char *);
+ int (PTRCALL *charMatches)(const ENCODING *, const char *, int);
+#endif /* XML_MIN_SIZE */
+ int (PTRFASTCALL *isName2)(const ENCODING *, const char *);
+ int (PTRFASTCALL *isName3)(const ENCODING *, const char *);
+ int (PTRFASTCALL *isName4)(const ENCODING *, const char *);
+ int (PTRFASTCALL *isNmstrt2)(const ENCODING *, const char *);
+ int (PTRFASTCALL *isNmstrt3)(const ENCODING *, const char *);
+ int (PTRFASTCALL *isNmstrt4)(const ENCODING *, const char *);
+ int (PTRFASTCALL *isInvalid2)(const ENCODING *, const char *);
+ int (PTRFASTCALL *isInvalid3)(const ENCODING *, const char *);
+ int (PTRFASTCALL *isInvalid4)(const ENCODING *, const char *);
+};
+
+#define AS_NORMAL_ENCODING(enc) ((const struct normal_encoding *) (enc))
+
+#ifdef XML_MIN_SIZE
+
+#define STANDARD_VTABLE(E) \
+ E ## byteType, \
+ E ## isNameMin, \
+ E ## isNmstrtMin, \
+ E ## byteToAscii, \
+ E ## charMatches,
+
+#else
+
+#define STANDARD_VTABLE(E) /* as nothing */
+
+#endif
+
+#define NORMAL_VTABLE(E) \
+ E ## isName2, \
+ E ## isName3, \
+ E ## isName4, \
+ E ## isNmstrt2, \
+ E ## isNmstrt3, \
+ E ## isNmstrt4, \
+ E ## isInvalid2, \
+ E ## isInvalid3, \
+ E ## isInvalid4
+
+static int FASTCALL checkCharRefNumber(int);
+
+#include "xmltok_impl.h"
+#include "ascii.h"
+
+#ifdef XML_MIN_SIZE
+#define sb_isNameMin isNever
+#define sb_isNmstrtMin isNever
+#endif
+
+#ifdef XML_MIN_SIZE
+#define MINBPC(enc) ((enc)->minBytesPerChar)
+#else
+/* minimum bytes per character */
+#define MINBPC(enc) 1
+#endif
+
+#define SB_BYTE_TYPE(enc, p) \
+ (((struct normal_encoding *)(enc))->type[(unsigned char)*(p)])
+
+#ifdef XML_MIN_SIZE
+static int PTRFASTCALL
+sb_byteType(const ENCODING *enc, const char *p)
+{
+ return SB_BYTE_TYPE(enc, p);
+}
+#define BYTE_TYPE(enc, p) \
+ (AS_NORMAL_ENCODING(enc)->byteType(enc, p))
+#else
+#define BYTE_TYPE(enc, p) SB_BYTE_TYPE(enc, p)
+#endif
+
+#ifdef XML_MIN_SIZE
+#define BYTE_TO_ASCII(enc, p) \
+ (AS_NORMAL_ENCODING(enc)->byteToAscii(enc, p))
+static int PTRFASTCALL
+sb_byteToAscii(const ENCODING *enc, const char *p)
+{
+ return *p;
+}
+#else
+#define BYTE_TO_ASCII(enc, p) (*(p))
+#endif
+
+#define IS_NAME_CHAR(enc, p, n) \
+ (AS_NORMAL_ENCODING(enc)->isName ## n(enc, p))
+#define IS_NMSTRT_CHAR(enc, p, n) \
+ (AS_NORMAL_ENCODING(enc)->isNmstrt ## n(enc, p))
+#define IS_INVALID_CHAR(enc, p, n) \
+ (AS_NORMAL_ENCODING(enc)->isInvalid ## n(enc, p))
+
+#ifdef XML_MIN_SIZE
+#define IS_NAME_CHAR_MINBPC(enc, p) \
+ (AS_NORMAL_ENCODING(enc)->isNameMin(enc, p))
+#define IS_NMSTRT_CHAR_MINBPC(enc, p) \
+ (AS_NORMAL_ENCODING(enc)->isNmstrtMin(enc, p))
+#else
+#define IS_NAME_CHAR_MINBPC(enc, p) (0)
+#define IS_NMSTRT_CHAR_MINBPC(enc, p) (0)
+#endif
+
+#ifdef XML_MIN_SIZE
+#define CHAR_MATCHES(enc, p, c) \
+ (AS_NORMAL_ENCODING(enc)->charMatches(enc, p, c))
+static int PTRCALL
+sb_charMatches(const ENCODING *enc, const char *p, int c)
+{
+ return *p == c;
+}
+#else
+/* c is an ASCII character */
+#define CHAR_MATCHES(enc, p, c) (*(p) == c)
+#endif
+
+#define PREFIX(ident) normal_ ## ident
+#include "xmltok_impl.c"
+
+#undef MINBPC
+#undef BYTE_TYPE
+#undef BYTE_TO_ASCII
+#undef CHAR_MATCHES
+#undef IS_NAME_CHAR
+#undef IS_NAME_CHAR_MINBPC
+#undef IS_NMSTRT_CHAR
+#undef IS_NMSTRT_CHAR_MINBPC
+#undef IS_INVALID_CHAR
+
+enum { /* UTF8_cvalN is value of masked first byte of N byte sequence */
+ UTF8_cval1 = 0x00,
+ UTF8_cval2 = 0xc0,
+ UTF8_cval3 = 0xe0,
+ UTF8_cval4 = 0xf0
+};
+
+static void PTRCALL
+utf8_toUtf8(const ENCODING *enc,
+ const char **fromP, const char *fromLim,
+ char **toP, const char *toLim)
+{
+ char *to;
+ const char *from;
+ if (fromLim - *fromP > toLim - *toP) {
+ /* Avoid copying partial characters. */
+ for (fromLim = *fromP + (toLim - *toP); fromLim > *fromP; fromLim--)
+ if (((unsigned char)fromLim[-1] & 0xc0) != 0x80)
+ break;
+ }
+ for (to = *toP, from = *fromP; from != fromLim; from++, to++)
+ *to = *from;
+ *fromP = from;
+ *toP = to;
+}
+
+static void PTRCALL
+utf8_toUtf16(const ENCODING *enc,
+ const char **fromP, const char *fromLim,
+ unsigned short **toP, const unsigned short *toLim)
+{
+ unsigned short *to = *toP;
+ const char *from = *fromP;
+ while (from != fromLim && to != toLim) {
+ switch (((struct normal_encoding *)enc)->type[(unsigned char)*from]) {
+ case BT_LEAD2:
+ *to++ = (unsigned short)(((from[0] & 0x1f) << 6) | (from[1] & 0x3f));
+ from += 2;
+ break;
+ case BT_LEAD3:
+ *to++ = (unsigned short)(((from[0] & 0xf) << 12)
+ | ((from[1] & 0x3f) << 6) | (from[2] & 0x3f));
+ from += 3;
+ break;
+ case BT_LEAD4:
+ {
+ unsigned long n;
+ if (to + 1 == toLim)
+ goto after;
+ n = ((from[0] & 0x7) << 18) | ((from[1] & 0x3f) << 12)
+ | ((from[2] & 0x3f) << 6) | (from[3] & 0x3f);
+ n -= 0x10000;
+ to[0] = (unsigned short)((n >> 10) | 0xD800);
+ to[1] = (unsigned short)((n & 0x3FF) | 0xDC00);
+ to += 2;
+ from += 4;
+ }
+ break;
+ default:
+ *to++ = *from++;
+ break;
+ }
+ }
+after:
+ *fromP = from;
+ *toP = to;
+}
+
+#ifdef XML_NS
+static const struct normal_encoding utf8_encoding_ns = {
+ { VTABLE1, utf8_toUtf8, utf8_toUtf16, 1, 1, 0 },
+ {
+#include "asciitab.h"
+#include "utf8tab.h"
+ },
+ STANDARD_VTABLE(sb_) NORMAL_VTABLE(utf8_)
+};
+#endif
+
+static const struct normal_encoding utf8_encoding = {
+ { VTABLE1, utf8_toUtf8, utf8_toUtf16, 1, 1, 0 },
+ {
+#define BT_COLON BT_NMSTRT
+#include "asciitab.h"
+#undef BT_COLON
+#include "utf8tab.h"
+ },
+ STANDARD_VTABLE(sb_) NORMAL_VTABLE(utf8_)
+};
+
+#ifdef XML_NS
+
+static const struct normal_encoding internal_utf8_encoding_ns = {
+ { VTABLE1, utf8_toUtf8, utf8_toUtf16, 1, 1, 0 },
+ {
+#include "iasciitab.h"
+#include "utf8tab.h"
+ },
+ STANDARD_VTABLE(sb_) NORMAL_VTABLE(utf8_)
+};
+
+#endif
+
+static const struct normal_encoding internal_utf8_encoding = {
+ { VTABLE1, utf8_toUtf8, utf8_toUtf16, 1, 1, 0 },
+ {
+#define BT_COLON BT_NMSTRT
+#include "iasciitab.h"
+#undef BT_COLON
+#include "utf8tab.h"
+ },
+ STANDARD_VTABLE(sb_) NORMAL_VTABLE(utf8_)
+};
+
+static void PTRCALL
+latin1_toUtf8(const ENCODING *enc,
+ const char **fromP, const char *fromLim,
+ char **toP, const char *toLim)
+{
+ for (;;) {
+ unsigned char c;
+ if (*fromP == fromLim)
+ break;
+ c = (unsigned char)**fromP;
+ if (c & 0x80) {
+ if (toLim - *toP < 2)
+ break;
+ *(*toP)++ = (char)((c >> 6) | UTF8_cval2);
+ *(*toP)++ = (char)((c & 0x3f) | 0x80);
+ (*fromP)++;
+ }
+ else {
+ if (*toP == toLim)
+ break;
+ *(*toP)++ = *(*fromP)++;
+ }
+ }
+}
+
+static void PTRCALL
+latin1_toUtf16(const ENCODING *enc,
+ const char **fromP, const char *fromLim,
+ unsigned short **toP, const unsigned short *toLim)
+{
+ while (*fromP != fromLim && *toP != toLim)
+ *(*toP)++ = (unsigned char)*(*fromP)++;
+}
+
+#ifdef XML_NS
+
+static const struct normal_encoding latin1_encoding_ns = {
+ { VTABLE1, latin1_toUtf8, latin1_toUtf16, 1, 0, 0 },
+ {
+#include "asciitab.h"
+#include "latin1tab.h"
+ },
+ STANDARD_VTABLE(sb_)
+};
+
+#endif
+
+static const struct normal_encoding latin1_encoding = {
+ { VTABLE1, latin1_toUtf8, latin1_toUtf16, 1, 0, 0 },
+ {
+#define BT_COLON BT_NMSTRT
+#include "asciitab.h"
+#undef BT_COLON
+#include "latin1tab.h"
+ },
+ STANDARD_VTABLE(sb_)
+};
+
+static void PTRCALL
+ascii_toUtf8(const ENCODING *enc,
+ const char **fromP, const char *fromLim,
+ char **toP, const char *toLim)
+{
+ while (*fromP != fromLim && *toP != toLim)
+ *(*toP)++ = *(*fromP)++;
+}
+
+#ifdef XML_NS
+
+static const struct normal_encoding ascii_encoding_ns = {
+ { VTABLE1, ascii_toUtf8, latin1_toUtf16, 1, 1, 0 },
+ {
+#include "asciitab.h"
+/* BT_NONXML == 0 */
+ },
+ STANDARD_VTABLE(sb_)
+};
+
+#endif
+
+static const struct normal_encoding ascii_encoding = {
+ { VTABLE1, ascii_toUtf8, latin1_toUtf16, 1, 1, 0 },
+ {
+#define BT_COLON BT_NMSTRT
+#include "asciitab.h"
+#undef BT_COLON
+/* BT_NONXML == 0 */
+ },
+ STANDARD_VTABLE(sb_)
+};
+
+static int PTRFASTCALL
+unicode_byte_type(char hi, char lo)
+{
+ switch ((unsigned char)hi) {
+ case 0xD8: case 0xD9: case 0xDA: case 0xDB:
+ return BT_LEAD4;
+ case 0xDC: case 0xDD: case 0xDE: case 0xDF:
+ return BT_TRAIL;
+ case 0xFF:
+ switch ((unsigned char)lo) {
+ case 0xFF:
+ case 0xFE:
+ return BT_NONXML;
+ }
+ break;
+ }
+ return BT_NONASCII;
+}
+
+#define DEFINE_UTF16_TO_UTF8(E) \
+static void PTRCALL \
+E ## toUtf8(const ENCODING *enc, \
+ const char **fromP, const char *fromLim, \
+ char **toP, const char *toLim) \
+{ \
+ const char *from; \
+ for (from = *fromP; from != fromLim; from += 2) { \
+ int plane; \
+ unsigned char lo2; \
+ unsigned char lo = GET_LO(from); \
+ unsigned char hi = GET_HI(from); \
+ switch (hi) { \
+ case 0: \
+ if (lo < 0x80) { \
+ if (*toP == toLim) { \
+ *fromP = from; \
+ return; \
+ } \
+ *(*toP)++ = lo; \
+ break; \
+ } \
+ /* fall through */ \
+ case 0x1: case 0x2: case 0x3: \
+ case 0x4: case 0x5: case 0x6: case 0x7: \
+ if (toLim - *toP < 2) { \
+ *fromP = from; \
+ return; \
+ } \
+ *(*toP)++ = ((lo >> 6) | (hi << 2) | UTF8_cval2); \
+ *(*toP)++ = ((lo & 0x3f) | 0x80); \
+ break; \
+ default: \
+ if (toLim - *toP < 3) { \
+ *fromP = from; \
+ return; \
+ } \
+ /* 16 bits divided 4, 6, 6 amongst 3 bytes */ \
+ *(*toP)++ = ((hi >> 4) | UTF8_cval3); \
+ *(*toP)++ = (((hi & 0xf) << 2) | (lo >> 6) | 0x80); \
+ *(*toP)++ = ((lo & 0x3f) | 0x80); \
+ break; \
+ case 0xD8: case 0xD9: case 0xDA: case 0xDB: \
+ if (toLim - *toP < 4) { \
+ *fromP = from; \
+ return; \
+ } \
+ plane = (((hi & 0x3) << 2) | ((lo >> 6) & 0x3)) + 1; \
+ *(*toP)++ = ((plane >> 2) | UTF8_cval4); \
+ *(*toP)++ = (((lo >> 2) & 0xF) | ((plane & 0x3) << 4) | 0x80); \
+ from += 2; \
+ lo2 = GET_LO(from); \
+ *(*toP)++ = (((lo & 0x3) << 4) \
+ | ((GET_HI(from) & 0x3) << 2) \
+ | (lo2 >> 6) \
+ | 0x80); \
+ *(*toP)++ = ((lo2 & 0x3f) | 0x80); \
+ break; \
+ } \
+ } \
+ *fromP = from; \
+}
+
+#define DEFINE_UTF16_TO_UTF16(E) \
+static void PTRCALL \
+E ## toUtf16(const ENCODING *enc, \
+ const char **fromP, const char *fromLim, \
+ unsigned short **toP, const unsigned short *toLim) \
+{ \
+ /* Avoid copying first half only of surrogate */ \
+ if (fromLim - *fromP > ((toLim - *toP) << 1) \
+ && (GET_HI(fromLim - 2) & 0xF8) == 0xD8) \
+ fromLim -= 2; \
+ for (; *fromP != fromLim && *toP != toLim; *fromP += 2) \
+ *(*toP)++ = (GET_HI(*fromP) << 8) | GET_LO(*fromP); \
+}
+
+#define SET2(ptr, ch) \
+ (((ptr)[0] = ((ch) & 0xff)), ((ptr)[1] = ((ch) >> 8)))
+#define GET_LO(ptr) ((unsigned char)(ptr)[0])
+#define GET_HI(ptr) ((unsigned char)(ptr)[1])
+
+DEFINE_UTF16_TO_UTF8(little2_)
+DEFINE_UTF16_TO_UTF16(little2_)
+
+#undef SET2
+#undef GET_LO
+#undef GET_HI
+
+#define SET2(ptr, ch) \
+ (((ptr)[0] = ((ch) >> 8)), ((ptr)[1] = ((ch) & 0xFF)))
+#define GET_LO(ptr) ((unsigned char)(ptr)[1])
+#define GET_HI(ptr) ((unsigned char)(ptr)[0])
+
+DEFINE_UTF16_TO_UTF8(big2_)
+DEFINE_UTF16_TO_UTF16(big2_)
+
+#undef SET2
+#undef GET_LO
+#undef GET_HI
+
+#define LITTLE2_BYTE_TYPE(enc, p) \
+ ((p)[1] == 0 \
+ ? ((struct normal_encoding *)(enc))->type[(unsigned char)*(p)] \
+ : unicode_byte_type((p)[1], (p)[0]))
+#define LITTLE2_BYTE_TO_ASCII(enc, p) ((p)[1] == 0 ? (p)[0] : -1)
+#define LITTLE2_CHAR_MATCHES(enc, p, c) ((p)[1] == 0 && (p)[0] == c)
+#define LITTLE2_IS_NAME_CHAR_MINBPC(enc, p) \
+ UCS2_GET_NAMING(namePages, (unsigned char)p[1], (unsigned char)p[0])
+#define LITTLE2_IS_NMSTRT_CHAR_MINBPC(enc, p) \
+ UCS2_GET_NAMING(nmstrtPages, (unsigned char)p[1], (unsigned char)p[0])
+
+#ifdef XML_MIN_SIZE
+
+static int PTRFASTCALL
+little2_byteType(const ENCODING *enc, const char *p)
+{
+ return LITTLE2_BYTE_TYPE(enc, p);
+}
+
+static int PTRFASTCALL
+little2_byteToAscii(const ENCODING *enc, const char *p)
+{
+ return LITTLE2_BYTE_TO_ASCII(enc, p);
+}
+
+static int PTRCALL
+little2_charMatches(const ENCODING *enc, const char *p, int c)
+{
+ return LITTLE2_CHAR_MATCHES(enc, p, c);
+}
+
+static int PTRFASTCALL
+little2_isNameMin(const ENCODING *enc, const char *p)
+{
+ return LITTLE2_IS_NAME_CHAR_MINBPC(enc, p);
+}
+
+static int PTRFASTCALL
+little2_isNmstrtMin(const ENCODING *enc, const char *p)
+{
+ return LITTLE2_IS_NMSTRT_CHAR_MINBPC(enc, p);
+}
+
+#undef VTABLE
+#define VTABLE VTABLE1, little2_toUtf8, little2_toUtf16
+
+#else /* not XML_MIN_SIZE */
+
+#undef PREFIX
+#define PREFIX(ident) little2_ ## ident
+#define MINBPC(enc) 2
+/* CHAR_MATCHES is guaranteed to have MINBPC bytes available. */
+#define BYTE_TYPE(enc, p) LITTLE2_BYTE_TYPE(enc, p)
+#define BYTE_TO_ASCII(enc, p) LITTLE2_BYTE_TO_ASCII(enc, p)
+#define CHAR_MATCHES(enc, p, c) LITTLE2_CHAR_MATCHES(enc, p, c)
+#define IS_NAME_CHAR(enc, p, n) 0
+#define IS_NAME_CHAR_MINBPC(enc, p) LITTLE2_IS_NAME_CHAR_MINBPC(enc, p)
+#define IS_NMSTRT_CHAR(enc, p, n) (0)
+#define IS_NMSTRT_CHAR_MINBPC(enc, p) LITTLE2_IS_NMSTRT_CHAR_MINBPC(enc, p)
+
+#include "xmltok_impl.c"
+
+#undef MINBPC
+#undef BYTE_TYPE
+#undef BYTE_TO_ASCII
+#undef CHAR_MATCHES
+#undef IS_NAME_CHAR
+#undef IS_NAME_CHAR_MINBPC
+#undef IS_NMSTRT_CHAR
+#undef IS_NMSTRT_CHAR_MINBPC
+#undef IS_INVALID_CHAR
+
+#endif /* not XML_MIN_SIZE */
+
+#ifdef XML_NS
+
+static const struct normal_encoding little2_encoding_ns = {
+ { VTABLE, 2, 0,
+#if BYTEORDER == 1234
+ 1
+#else
+ 0
+#endif
+ },
+ {
+#include "asciitab.h"
+#include "latin1tab.h"
+ },
+ STANDARD_VTABLE(little2_)
+};
+
+#endif
+
+static const struct normal_encoding little2_encoding = {
+ { VTABLE, 2, 0,
+#if BYTEORDER == 1234
+ 1
+#else
+ 0
+#endif
+ },
+ {
+#define BT_COLON BT_NMSTRT
+#include "asciitab.h"
+#undef BT_COLON
+#include "latin1tab.h"
+ },
+ STANDARD_VTABLE(little2_)
+};
+
+#if BYTEORDER != 4321
+
+#ifdef XML_NS
+
+static const struct normal_encoding internal_little2_encoding_ns = {
+ { VTABLE, 2, 0, 1 },
+ {
+#include "iasciitab.h"
+#include "latin1tab.h"
+ },
+ STANDARD_VTABLE(little2_)
+};
+
+#endif
+
+static const struct normal_encoding internal_little2_encoding = {
+ { VTABLE, 2, 0, 1 },
+ {
+#define BT_COLON BT_NMSTRT
+#include "iasciitab.h"
+#undef BT_COLON
+#include "latin1tab.h"
+ },
+ STANDARD_VTABLE(little2_)
+};
+
+#endif
+
+
+#define BIG2_BYTE_TYPE(enc, p) \
+ ((p)[0] == 0 \
+ ? ((struct normal_encoding *)(enc))->type[(unsigned char)(p)[1]] \
+ : unicode_byte_type((p)[0], (p)[1]))
+#define BIG2_BYTE_TO_ASCII(enc, p) ((p)[0] == 0 ? (p)[1] : -1)
+#define BIG2_CHAR_MATCHES(enc, p, c) ((p)[0] == 0 && (p)[1] == c)
+#define BIG2_IS_NAME_CHAR_MINBPC(enc, p) \
+ UCS2_GET_NAMING(namePages, (unsigned char)p[0], (unsigned char)p[1])
+#define BIG2_IS_NMSTRT_CHAR_MINBPC(enc, p) \
+ UCS2_GET_NAMING(nmstrtPages, (unsigned char)p[0], (unsigned char)p[1])
+
+#ifdef XML_MIN_SIZE
+
+static int PTRFASTCALL
+big2_byteType(const ENCODING *enc, const char *p)
+{
+ return BIG2_BYTE_TYPE(enc, p);
+}
+
+static int PTRFASTCALL
+big2_byteToAscii(const ENCODING *enc, const char *p)
+{
+ return BIG2_BYTE_TO_ASCII(enc, p);
+}
+
+static int PTRCALL
+big2_charMatches(const ENCODING *enc, const char *p, int c)
+{
+ return BIG2_CHAR_MATCHES(enc, p, c);
+}
+
+static int PTRFASTCALL
+big2_isNameMin(const ENCODING *enc, const char *p)
+{
+ return BIG2_IS_NAME_CHAR_MINBPC(enc, p);
+}
+
+static int PTRFASTCALL
+big2_isNmstrtMin(const ENCODING *enc, const char *p)
+{
+ return BIG2_IS_NMSTRT_CHAR_MINBPC(enc, p);
+}
+
+#undef VTABLE
+#define VTABLE VTABLE1, big2_toUtf8, big2_toUtf16
+
+#else /* not XML_MIN_SIZE */
+
+#undef PREFIX
+#define PREFIX(ident) big2_ ## ident
+#define MINBPC(enc) 2
+/* CHAR_MATCHES is guaranteed to have MINBPC bytes available. */
+#define BYTE_TYPE(enc, p) BIG2_BYTE_TYPE(enc, p)
+#define BYTE_TO_ASCII(enc, p) BIG2_BYTE_TO_ASCII(enc, p)
+#define CHAR_MATCHES(enc, p, c) BIG2_CHAR_MATCHES(enc, p, c)
+#define IS_NAME_CHAR(enc, p, n) 0
+#define IS_NAME_CHAR_MINBPC(enc, p) BIG2_IS_NAME_CHAR_MINBPC(enc, p)
+#define IS_NMSTRT_CHAR(enc, p, n) (0)
+#define IS_NMSTRT_CHAR_MINBPC(enc, p) BIG2_IS_NMSTRT_CHAR_MINBPC(enc, p)
+
+#include "xmltok_impl.c"
+
+#undef MINBPC
+#undef BYTE_TYPE
+#undef BYTE_TO_ASCII
+#undef CHAR_MATCHES
+#undef IS_NAME_CHAR
+#undef IS_NAME_CHAR_MINBPC
+#undef IS_NMSTRT_CHAR
+#undef IS_NMSTRT_CHAR_MINBPC
+#undef IS_INVALID_CHAR
+
+#endif /* not XML_MIN_SIZE */
+
+#ifdef XML_NS
+
+static const struct normal_encoding big2_encoding_ns = {
+ { VTABLE, 2, 0,
+#if BYTEORDER == 4321
+ 1
+#else
+ 0
+#endif
+ },
+ {
+#include "asciitab.h"
+#include "latin1tab.h"
+ },
+ STANDARD_VTABLE(big2_)
+};
+
+#endif
+
+static const struct normal_encoding big2_encoding = {
+ { VTABLE, 2, 0,
+#if BYTEORDER == 4321
+ 1
+#else
+ 0
+#endif
+ },
+ {
+#define BT_COLON BT_NMSTRT
+#include "asciitab.h"
+#undef BT_COLON
+#include "latin1tab.h"
+ },
+ STANDARD_VTABLE(big2_)
+};
+
+#if BYTEORDER != 1234
+
+#ifdef XML_NS
+
+static const struct normal_encoding internal_big2_encoding_ns = {
+ { VTABLE, 2, 0, 1 },
+ {
+#include "iasciitab.h"
+#include "latin1tab.h"
+ },
+ STANDARD_VTABLE(big2_)
+};
+
+#endif
+
+static const struct normal_encoding internal_big2_encoding = {
+ { VTABLE, 2, 0, 1 },
+ {
+#define BT_COLON BT_NMSTRT
+#include "iasciitab.h"
+#undef BT_COLON
+#include "latin1tab.h"
+ },
+ STANDARD_VTABLE(big2_)
+};
+
+#endif
+
+#undef PREFIX
+
+static int FASTCALL
+streqci(const char *s1, const char *s2)
+{
+ for (;;) {
+ char c1 = *s1++;
+ char c2 = *s2++;
+ if (ASCII_a <= c1 && c1 <= ASCII_z)
+ c1 += ASCII_A - ASCII_a;
+ if (ASCII_a <= c2 && c2 <= ASCII_z)
+ c2 += ASCII_A - ASCII_a;
+ if (c1 != c2)
+ return 0;
+ if (!c1)
+ break;
+ }
+ return 1;
+}
+
+static void PTRCALL
+initUpdatePosition(const ENCODING *enc, const char *ptr,
+ const char *end, POSITION *pos)
+{
+ normal_updatePosition(&utf8_encoding.enc, ptr, end, pos);
+}
+
+static int
+toAscii(const ENCODING *enc, const char *ptr, const char *end)
+{
+ char buf[1];
+ char *p = buf;
+ XmlUtf8Convert(enc, &ptr, end, &p, p + 1);
+ if (p == buf)
+ return -1;
+ else
+ return buf[0];
+}
+
+static int FASTCALL
+isSpace(int c)
+{
+ switch (c) {
+ case 0x20:
+ case 0xD:
+ case 0xA:
+ case 0x9:
+ return 1;
+ }
+ return 0;
+}
+
+/* Return 1 if there's just optional white space or there's an S
+ followed by name=val.
+*/
+static int
+parsePseudoAttribute(const ENCODING *enc,
+ const char *ptr,
+ const char *end,
+ const char **namePtr,
+ const char **nameEndPtr,
+ const char **valPtr,
+ const char **nextTokPtr)
+{
+ int c;
+ char open;
+ if (ptr == end) {
+ *namePtr = NULL;
+ return 1;
+ }
+ if (!isSpace(toAscii(enc, ptr, end))) {
+ *nextTokPtr = ptr;
+ return 0;
+ }
+ do {
+ ptr += enc->minBytesPerChar;
+ } while (isSpace(toAscii(enc, ptr, end)));
+ if (ptr == end) {
+ *namePtr = NULL;
+ return 1;
+ }
+ *namePtr = ptr;
+ for (;;) {
+ c = toAscii(enc, ptr, end);
+ if (c == -1) {
+ *nextTokPtr = ptr;
+ return 0;
+ }
+ if (c == ASCII_EQUALS) {
+ *nameEndPtr = ptr;
+ break;
+ }
+ if (isSpace(c)) {
+ *nameEndPtr = ptr;
+ do {
+ ptr += enc->minBytesPerChar;
+ } while (isSpace(c = toAscii(enc, ptr, end)));
+ if (c != ASCII_EQUALS) {
+ *nextTokPtr = ptr;
+ return 0;
+ }
+ break;
+ }
+ ptr += enc->minBytesPerChar;
+ }
+ if (ptr == *namePtr) {
+ *nextTokPtr = ptr;
+ return 0;
+ }
+ ptr += enc->minBytesPerChar;
+ c = toAscii(enc, ptr, end);
+ while (isSpace(c)) {
+ ptr += enc->minBytesPerChar;
+ c = toAscii(enc, ptr, end);
+ }
+ if (c != ASCII_QUOT && c != ASCII_APOS) {
+ *nextTokPtr = ptr;
+ return 0;
+ }
+ open = (char)c;
+ ptr += enc->minBytesPerChar;
+ *valPtr = ptr;
+ for (;; ptr += enc->minBytesPerChar) {
+ c = toAscii(enc, ptr, end);
+ if (c == open)
+ break;
+ if (!(ASCII_a <= c && c <= ASCII_z)
+ && !(ASCII_A <= c && c <= ASCII_Z)
+ && !(ASCII_0 <= c && c <= ASCII_9)
+ && c != ASCII_PERIOD
+ && c != ASCII_MINUS
+ && c != ASCII_UNDERSCORE) {
+ *nextTokPtr = ptr;
+ return 0;
+ }
+ }
+ *nextTokPtr = ptr + enc->minBytesPerChar;
+ return 1;
+}
+
+static const char KW_version[] = {
+ ASCII_v, ASCII_e, ASCII_r, ASCII_s, ASCII_i, ASCII_o, ASCII_n, '\0'
+};
+
+static const char KW_encoding[] = {
+ ASCII_e, ASCII_n, ASCII_c, ASCII_o, ASCII_d, ASCII_i, ASCII_n, ASCII_g, '\0'
+};
+
+static const char KW_standalone[] = {
+ ASCII_s, ASCII_t, ASCII_a, ASCII_n, ASCII_d, ASCII_a, ASCII_l, ASCII_o,
+ ASCII_n, ASCII_e, '\0'
+};
+
+static const char KW_yes[] = {
+ ASCII_y, ASCII_e, ASCII_s, '\0'
+};
+
+static const char KW_no[] = {
+ ASCII_n, ASCII_o, '\0'
+};
+
+static int
+doParseXmlDecl(const ENCODING *(*encodingFinder)(const ENCODING *,
+ const char *,
+ const char *),
+ int isGeneralTextEntity,
+ const ENCODING *enc,
+ const char *ptr,
+ const char *end,
+ const char **badPtr,
+ const char **versionPtr,
+ const char **versionEndPtr,
+ const char **encodingName,
+ const ENCODING **encoding,
+ int *standalone)
+{
+ const char *val = NULL;
+ const char *name = NULL;
+ const char *nameEnd = NULL;
+ ptr += 5 * enc->minBytesPerChar;
+ end -= 2 * enc->minBytesPerChar;
+ if (!parsePseudoAttribute(enc, ptr, end, &name, &nameEnd, &val, &ptr)
+ || !name) {
+ *badPtr = ptr;
+ return 0;
+ }
+ if (!XmlNameMatchesAscii(enc, name, nameEnd, KW_version)) {
+ if (!isGeneralTextEntity) {
+ *badPtr = name;
+ return 0;
+ }
+ }
+ else {
+ if (versionPtr)
+ *versionPtr = val;
+ if (versionEndPtr)
+ *versionEndPtr = ptr;
+ if (!parsePseudoAttribute(enc, ptr, end, &name, &nameEnd, &val, &ptr)) {
+ *badPtr = ptr;
+ return 0;
+ }
+ if (!name) {
+ if (isGeneralTextEntity) {
+ /* a TextDecl must have an EncodingDecl */
+ *badPtr = ptr;
+ return 0;
+ }
+ return 1;
+ }
+ }
+ if (XmlNameMatchesAscii(enc, name, nameEnd, KW_encoding)) {
+ int c = toAscii(enc, val, end);
+ if (!(ASCII_a <= c && c <= ASCII_z) && !(ASCII_A <= c && c <= ASCII_Z)) {
+ *badPtr = val;
+ return 0;
+ }
+ if (encodingName)
+ *encodingName = val;
+ if (encoding)
+ *encoding = encodingFinder(enc, val, ptr - enc->minBytesPerChar);
+ if (!parsePseudoAttribute(enc, ptr, end, &name, &nameEnd, &val, &ptr)) {
+ *badPtr = ptr;
+ return 0;
+ }
+ if (!name)
+ return 1;
+ }
+ if (!XmlNameMatchesAscii(enc, name, nameEnd, KW_standalone)
+ || isGeneralTextEntity) {
+ *badPtr = name;
+ return 0;
+ }
+ if (XmlNameMatchesAscii(enc, val, ptr - enc->minBytesPerChar, KW_yes)) {
+ if (standalone)
+ *standalone = 1;
+ }
+ else if (XmlNameMatchesAscii(enc, val, ptr - enc->minBytesPerChar, KW_no)) {
+ if (standalone)
+ *standalone = 0;
+ }
+ else {
+ *badPtr = val;
+ return 0;
+ }
+ while (isSpace(toAscii(enc, ptr, end)))
+ ptr += enc->minBytesPerChar;
+ if (ptr != end) {
+ *badPtr = ptr;
+ return 0;
+ }
+ return 1;
+}
+
+static int FASTCALL
+checkCharRefNumber(int result)
+{
+ switch (result >> 8) {
+ case 0xD8: case 0xD9: case 0xDA: case 0xDB:
+ case 0xDC: case 0xDD: case 0xDE: case 0xDF:
+ return -1;
+ case 0:
+ if (latin1_encoding.type[result] == BT_NONXML)
+ return -1;
+ break;
+ case 0xFF:
+ if (result == 0xFFFE || result == 0xFFFF)
+ return -1;
+ break;
+ }
+ return result;
+}
+
+int FASTCALL
+XmlUtf8Encode(int c, char *buf)
+{
+ enum {
+ /* minN is minimum legal resulting value for N byte sequence */
+ min2 = 0x80,
+ min3 = 0x800,
+ min4 = 0x10000
+ };
+
+ if (c < 0)
+ return 0;
+ if (c < min2) {
+ buf[0] = (char)(c | UTF8_cval1);
+ return 1;
+ }
+ if (c < min3) {
+ buf[0] = (char)((c >> 6) | UTF8_cval2);
+ buf[1] = (char)((c & 0x3f) | 0x80);
+ return 2;
+ }
+ if (c < min4) {
+ buf[0] = (char)((c >> 12) | UTF8_cval3);
+ buf[1] = (char)(((c >> 6) & 0x3f) | 0x80);
+ buf[2] = (char)((c & 0x3f) | 0x80);
+ return 3;
+ }
+ if (c < 0x110000) {
+ buf[0] = (char)((c >> 18) | UTF8_cval4);
+ buf[1] = (char)(((c >> 12) & 0x3f) | 0x80);
+ buf[2] = (char)(((c >> 6) & 0x3f) | 0x80);
+ buf[3] = (char)((c & 0x3f) | 0x80);
+ return 4;
+ }
+ return 0;
+}
+
+int FASTCALL
+XmlUtf16Encode(int charNum, unsigned short *buf)
+{
+ if (charNum < 0)
+ return 0;
+ if (charNum < 0x10000) {
+ buf[0] = (unsigned short)charNum;
+ return 1;
+ }
+ if (charNum < 0x110000) {
+ charNum -= 0x10000;
+ buf[0] = (unsigned short)((charNum >> 10) + 0xD800);
+ buf[1] = (unsigned short)((charNum & 0x3FF) + 0xDC00);
+ return 2;
+ }
+ return 0;
+}
+
+struct unknown_encoding {
+ struct normal_encoding normal;
+ CONVERTER convert;
+ void *userData;
+ unsigned short utf16[256];
+ char utf8[256][4];
+};
+
+#define AS_UNKNOWN_ENCODING(enc) ((const struct unknown_encoding *) (enc))
+
+int
+XmlSizeOfUnknownEncoding(void)
+{
+ return sizeof(struct unknown_encoding);
+}
+
+static int PTRFASTCALL
+unknown_isName(const ENCODING *enc, const char *p)
+{
+ const struct unknown_encoding *uenc = AS_UNKNOWN_ENCODING(enc);
+ int c = uenc->convert(uenc->userData, p);
+ if (c & ~0xFFFF)
+ return 0;
+ return UCS2_GET_NAMING(namePages, c >> 8, c & 0xFF);
+}
+
+static int PTRFASTCALL
+unknown_isNmstrt(const ENCODING *enc, const char *p)
+{
+ const struct unknown_encoding *uenc = AS_UNKNOWN_ENCODING(enc);
+ int c = uenc->convert(uenc->userData, p);
+ if (c & ~0xFFFF)
+ return 0;
+ return UCS2_GET_NAMING(nmstrtPages, c >> 8, c & 0xFF);
+}
+
+static int PTRFASTCALL
+unknown_isInvalid(const ENCODING *enc, const char *p)
+{
+ const struct unknown_encoding *uenc = AS_UNKNOWN_ENCODING(enc);
+ int c = uenc->convert(uenc->userData, p);
+ return (c & ~0xFFFF) || checkCharRefNumber(c) < 0;
+}
+
+static void PTRCALL
+unknown_toUtf8(const ENCODING *enc,
+ const char **fromP, const char *fromLim,
+ char **toP, const char *toLim)
+{
+ const struct unknown_encoding *uenc = AS_UNKNOWN_ENCODING(enc);
+ char buf[XML_UTF8_ENCODE_MAX];
+ for (;;) {
+ const char *utf8;
+ int n;
+ if (*fromP == fromLim)
+ break;
+ utf8 = uenc->utf8[(unsigned char)**fromP];
+ n = *utf8++;
+ if (n == 0) {
+ int c = uenc->convert(uenc->userData, *fromP);
+ n = XmlUtf8Encode(c, buf);
+ if (n > toLim - *toP)
+ break;
+ utf8 = buf;
+ *fromP += (AS_NORMAL_ENCODING(enc)->type[(unsigned char)**fromP]
+ - (BT_LEAD2 - 2));
+ }
+ else {
+ if (n > toLim - *toP)
+ break;
+ (*fromP)++;
+ }
+ do {
+ *(*toP)++ = *utf8++;
+ } while (--n != 0);
+ }
+}
+
+static void PTRCALL
+unknown_toUtf16(const ENCODING *enc,
+ const char **fromP, const char *fromLim,
+ unsigned short **toP, const unsigned short *toLim)
+{
+ const struct unknown_encoding *uenc = AS_UNKNOWN_ENCODING(enc);
+ while (*fromP != fromLim && *toP != toLim) {
+ unsigned short c = uenc->utf16[(unsigned char)**fromP];
+ if (c == 0) {
+ c = (unsigned short)
+ uenc->convert(uenc->userData, *fromP);
+ *fromP += (AS_NORMAL_ENCODING(enc)->type[(unsigned char)**fromP]
+ - (BT_LEAD2 - 2));
+ }
+ else
+ (*fromP)++;
+ *(*toP)++ = c;
+ }
+}
+
+ENCODING *
+XmlInitUnknownEncoding(void *mem,
+ int *table,
+ CONVERTER convert,
+ void *userData)
+{
+ int i;
+ struct unknown_encoding *e = (struct unknown_encoding *)mem;
+ for (i = 0; i < (int)sizeof(struct normal_encoding); i++)
+ ((char *)mem)[i] = ((char *)&latin1_encoding)[i];
+ for (i = 0; i < 128; i++)
+ if (latin1_encoding.type[i] != BT_OTHER
+ && latin1_encoding.type[i] != BT_NONXML
+ && table[i] != i)
+ return 0;
+ for (i = 0; i < 256; i++) {
+ int c = table[i];
+ if (c == -1) {
+ e->normal.type[i] = BT_MALFORM;
+ /* This shouldn't really get used. */
+ e->utf16[i] = 0xFFFF;
+ e->utf8[i][0] = 1;
+ e->utf8[i][1] = 0;
+ }
+ else if (c < 0) {
+ if (c < -4)
+ return 0;
+ e->normal.type[i] = (unsigned char)(BT_LEAD2 - (c + 2));
+ e->utf8[i][0] = 0;
+ e->utf16[i] = 0;
+ }
+ else if (c < 0x80) {
+ if (latin1_encoding.type[c] != BT_OTHER
+ && latin1_encoding.type[c] != BT_NONXML
+ && c != i)
+ return 0;
+ e->normal.type[i] = latin1_encoding.type[c];
+ e->utf8[i][0] = 1;
+ e->utf8[i][1] = (char)c;
+ e->utf16[i] = (unsigned short)(c == 0 ? 0xFFFF : c);
+ }
+ else if (checkCharRefNumber(c) < 0) {
+ e->normal.type[i] = BT_NONXML;
+ /* This shouldn't really get used. */
+ e->utf16[i] = 0xFFFF;
+ e->utf8[i][0] = 1;
+ e->utf8[i][1] = 0;
+ }
+ else {
+ if (c > 0xFFFF)
+ return 0;
+ if (UCS2_GET_NAMING(nmstrtPages, c >> 8, c & 0xff))
+ e->normal.type[i] = BT_NMSTRT;
+ else if (UCS2_GET_NAMING(namePages, c >> 8, c & 0xff))
+ e->normal.type[i] = BT_NAME;
+ else
+ e->normal.type[i] = BT_OTHER;
+ e->utf8[i][0] = (char)XmlUtf8Encode(c, e->utf8[i] + 1);
+ e->utf16[i] = (unsigned short)c;
+ }
+ }
+ e->userData = userData;
+ e->convert = convert;
+ if (convert) {
+ e->normal.isName2 = unknown_isName;
+ e->normal.isName3 = unknown_isName;
+ e->normal.isName4 = unknown_isName;
+ e->normal.isNmstrt2 = unknown_isNmstrt;
+ e->normal.isNmstrt3 = unknown_isNmstrt;
+ e->normal.isNmstrt4 = unknown_isNmstrt;
+ e->normal.isInvalid2 = unknown_isInvalid;
+ e->normal.isInvalid3 = unknown_isInvalid;
+ e->normal.isInvalid4 = unknown_isInvalid;
+ }
+ e->normal.enc.utf8Convert = unknown_toUtf8;
+ e->normal.enc.utf16Convert = unknown_toUtf16;
+ return &(e->normal.enc);
+}
+
+/* If this enumeration is changed, getEncodingIndex and encodings
+must also be changed. */
+enum {
+ UNKNOWN_ENC = -1,
+ ISO_8859_1_ENC = 0,
+ US_ASCII_ENC,
+ UTF_8_ENC,
+ UTF_16_ENC,
+ UTF_16BE_ENC,
+ UTF_16LE_ENC,
+ /* must match encodingNames up to here */
+ NO_ENC
+};
+
+static const char KW_ISO_8859_1[] = {
+ ASCII_I, ASCII_S, ASCII_O, ASCII_MINUS, ASCII_8, ASCII_8, ASCII_5, ASCII_9,
+ ASCII_MINUS, ASCII_1, '\0'
+};
+static const char KW_US_ASCII[] = {
+ ASCII_U, ASCII_S, ASCII_MINUS, ASCII_A, ASCII_S, ASCII_C, ASCII_I, ASCII_I,
+ '\0'
+};
+static const char KW_UTF_8[] = {
+ ASCII_U, ASCII_T, ASCII_F, ASCII_MINUS, ASCII_8, '\0'
+};
+static const char KW_UTF_16[] = {
+ ASCII_U, ASCII_T, ASCII_F, ASCII_MINUS, ASCII_1, ASCII_6, '\0'
+};
+static const char KW_UTF_16BE[] = {
+ ASCII_U, ASCII_T, ASCII_F, ASCII_MINUS, ASCII_1, ASCII_6, ASCII_B, ASCII_E,
+ '\0'
+};
+static const char KW_UTF_16LE[] = {
+ ASCII_U, ASCII_T, ASCII_F, ASCII_MINUS, ASCII_1, ASCII_6, ASCII_L, ASCII_E,
+ '\0'
+};
+
+static int FASTCALL
+getEncodingIndex(const char *name)
+{
+ static const char * const encodingNames[] = {
+ KW_ISO_8859_1,
+ KW_US_ASCII,
+ KW_UTF_8,
+ KW_UTF_16,
+ KW_UTF_16BE,
+ KW_UTF_16LE,
+ };
+ int i;
+ if (name == NULL)
+ return NO_ENC;
+ for (i = 0; i < (int)(sizeof(encodingNames)/sizeof(encodingNames[0])); i++)
+ if (streqci(name, encodingNames[i]))
+ return i;
+ return UNKNOWN_ENC;
+}
+
+/* For binary compatibility, we store the index of the encoding
+ specified at initialization in the isUtf16 member.
+*/
+
+#define INIT_ENC_INDEX(enc) ((int)(enc)->initEnc.isUtf16)
+#define SET_INIT_ENC_INDEX(enc, i) ((enc)->initEnc.isUtf16 = (char)i)
+
+/* This is what detects the encoding. encodingTable maps from
+ encoding indices to encodings; INIT_ENC_INDEX(enc) is the index of
+ the external (protocol) specified encoding; state is
+ XML_CONTENT_STATE if we're parsing an external text entity, and
+ XML_PROLOG_STATE otherwise.
+*/
+
+
+static int
+initScan(const ENCODING * const *encodingTable,
+ const INIT_ENCODING *enc,
+ int state,
+ const char *ptr,
+ const char *end,
+ const char **nextTokPtr)
+{
+ const ENCODING **encPtr;
+
+ if (ptr == end)
+ return XML_TOK_NONE;
+ encPtr = enc->encPtr;
+ if (ptr + 1 == end) {
+ /* only a single byte available for auto-detection */
+#ifndef XML_DTD /* FIXME */
+ /* a well-formed document entity must have more than one byte */
+ if (state != XML_CONTENT_STATE)
+ return XML_TOK_PARTIAL;
+#endif
+ /* so we're parsing an external text entity... */
+ /* if UTF-16 was externally specified, then we need at least 2 bytes */
+ switch (INIT_ENC_INDEX(enc)) {
+ case UTF_16_ENC:
+ case UTF_16LE_ENC:
+ case UTF_16BE_ENC:
+ return XML_TOK_PARTIAL;
+ }
+ switch ((unsigned char)*ptr) {
+ case 0xFE:
+ case 0xFF:
+ case 0xEF: /* possibly first byte of UTF-8 BOM */
+ if (INIT_ENC_INDEX(enc) == ISO_8859_1_ENC
+ && state == XML_CONTENT_STATE)
+ break;
+ /* fall through */
+ case 0x00:
+ case 0x3C:
+ return XML_TOK_PARTIAL;
+ }
+ }
+ else {
+ switch (((unsigned char)ptr[0] << 8) | (unsigned char)ptr[1]) {
+ case 0xFEFF:
+ if (INIT_ENC_INDEX(enc) == ISO_8859_1_ENC
+ && state == XML_CONTENT_STATE)
+ break;
+ *nextTokPtr = ptr + 2;
+ *encPtr = encodingTable[UTF_16BE_ENC];
+ return XML_TOK_BOM;
+ /* 00 3C is handled in the default case */
+ case 0x3C00:
+ if ((INIT_ENC_INDEX(enc) == UTF_16BE_ENC
+ || INIT_ENC_INDEX(enc) == UTF_16_ENC)
+ && state == XML_CONTENT_STATE)
+ break;
+ *encPtr = encodingTable[UTF_16LE_ENC];
+ return XmlTok(*encPtr, state, ptr, end, nextTokPtr);
+ case 0xFFFE:
+ if (INIT_ENC_INDEX(enc) == ISO_8859_1_ENC
+ && state == XML_CONTENT_STATE)
+ break;
+ *nextTokPtr = ptr + 2;
+ *encPtr = encodingTable[UTF_16LE_ENC];
+ return XML_TOK_BOM;
+ case 0xEFBB:
+ /* Maybe a UTF-8 BOM (EF BB BF) */
+ /* If there's an explicitly specified (external) encoding
+ of ISO-8859-1 or some flavour of UTF-16
+ and this is an external text entity,
+ don't look for the BOM,
+ because it might be a legal data.
+ */
+ if (state == XML_CONTENT_STATE) {
+ int e = INIT_ENC_INDEX(enc);
+ if (e == ISO_8859_1_ENC || e == UTF_16BE_ENC
+ || e == UTF_16LE_ENC || e == UTF_16_ENC)
+ break;
+ }
+ if (ptr + 2 == end)
+ return XML_TOK_PARTIAL;
+ if ((unsigned char)ptr[2] == 0xBF) {
+ *nextTokPtr = ptr + 3;
+ *encPtr = encodingTable[UTF_8_ENC];
+ return XML_TOK_BOM;
+ }
+ break;
+ default:
+ if (ptr[0] == '\0') {
+ /* 0 isn't a legal data character. Furthermore a document
+ entity can only start with ASCII characters. So the only
+ way this can fail to be big-endian UTF-16 if it it's an
+ external parsed general entity that's labelled as
+ UTF-16LE.
+ */
+ if (state == XML_CONTENT_STATE && INIT_ENC_INDEX(enc) == UTF_16LE_ENC)
+ break;
+ *encPtr = encodingTable[UTF_16BE_ENC];
+ return XmlTok(*encPtr, state, ptr, end, nextTokPtr);
+ }
+ else if (ptr[1] == '\0') {
+ /* We could recover here in the case:
+ - parsing an external entity
+ - second byte is 0
+ - no externally specified encoding
+ - no encoding declaration
+ by assuming UTF-16LE. But we don't, because this would mean when
+ presented just with a single byte, we couldn't reliably determine
+ whether we needed further bytes.
+ */
+ if (state == XML_CONTENT_STATE)
+ break;
+ *encPtr = encodingTable[UTF_16LE_ENC];
+ return XmlTok(*encPtr, state, ptr, end, nextTokPtr);
+ }
+ break;
+ }
+ }
+ *encPtr = encodingTable[INIT_ENC_INDEX(enc)];
+ return XmlTok(*encPtr, state, ptr, end, nextTokPtr);
+}
+
+
+#define NS(x) x
+#define ns(x) x
+#include "xmltok_ns.c"
+#undef NS
+#undef ns
+
+#ifdef XML_NS
+
+#define NS(x) x ## NS
+#define ns(x) x ## _ns
+
+#include "xmltok_ns.c"
+
+#undef NS
+#undef ns
+
+ENCODING *
+XmlInitUnknownEncodingNS(void *mem,
+ int *table,
+ CONVERTER convert,
+ void *userData)
+{
+ ENCODING *enc = XmlInitUnknownEncoding(mem, table, convert, userData);
+ if (enc)
+ ((struct normal_encoding *)enc)->type[ASCII_COLON] = BT_COLON;
+ return enc;
+}
+
+#endif /* XML_NS */
diff --git a/src/lib/expat/xmltok.h b/src/lib/expat/xmltok.h
new file mode 100644
index 00000000000..ca867aa6b42
--- /dev/null
+++ b/src/lib/expat/xmltok.h
@@ -0,0 +1,316 @@
+/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+ See the file COPYING for copying permission.
+*/
+
+#ifndef XmlTok_INCLUDED
+#define XmlTok_INCLUDED 1
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/* The following token may be returned by XmlContentTok */
+#define XML_TOK_TRAILING_RSQB -5 /* ] or ]] at the end of the scan; might be
+ start of illegal ]]> sequence */
+/* The following tokens may be returned by both XmlPrologTok and
+ XmlContentTok.
+*/
+#define XML_TOK_NONE -4 /* The string to be scanned is empty */
+#define XML_TOK_TRAILING_CR -3 /* A CR at the end of the scan;
+ might be part of CRLF sequence */
+#define XML_TOK_PARTIAL_CHAR -2 /* only part of a multibyte sequence */
+#define XML_TOK_PARTIAL -1 /* only part of a token */
+#define XML_TOK_INVALID 0
+
+/* The following tokens are returned by XmlContentTok; some are also
+ returned by XmlAttributeValueTok, XmlEntityTok, XmlCdataSectionTok.
+*/
+#define XML_TOK_START_TAG_WITH_ATTS 1
+#define XML_TOK_START_TAG_NO_ATTS 2
+#define XML_TOK_EMPTY_ELEMENT_WITH_ATTS 3 /* empty element tag <e/> */
+#define XML_TOK_EMPTY_ELEMENT_NO_ATTS 4
+#define XML_TOK_END_TAG 5
+#define XML_TOK_DATA_CHARS 6
+#define XML_TOK_DATA_NEWLINE 7
+#define XML_TOK_CDATA_SECT_OPEN 8
+#define XML_TOK_ENTITY_REF 9
+#define XML_TOK_CHAR_REF 10 /* numeric character reference */
+
+/* The following tokens may be returned by both XmlPrologTok and
+ XmlContentTok.
+*/
+#define XML_TOK_PI 11 /* processing instruction */
+#define XML_TOK_XML_DECL 12 /* XML decl or text decl */
+#define XML_TOK_COMMENT 13
+#define XML_TOK_BOM 14 /* Byte order mark */
+
+/* The following tokens are returned only by XmlPrologTok */
+#define XML_TOK_PROLOG_S 15
+#define XML_TOK_DECL_OPEN 16 /* <!foo */
+#define XML_TOK_DECL_CLOSE 17 /* > */
+#define XML_TOK_NAME 18
+#define XML_TOK_NMTOKEN 19
+#define XML_TOK_POUND_NAME 20 /* #name */
+#define XML_TOK_OR 21 /* | */
+#define XML_TOK_PERCENT 22
+#define XML_TOK_OPEN_PAREN 23
+#define XML_TOK_CLOSE_PAREN 24
+#define XML_TOK_OPEN_BRACKET 25
+#define XML_TOK_CLOSE_BRACKET 26
+#define XML_TOK_LITERAL 27
+#define XML_TOK_PARAM_ENTITY_REF 28
+#define XML_TOK_INSTANCE_START 29
+
+/* The following occur only in element type declarations */
+#define XML_TOK_NAME_QUESTION 30 /* name? */
+#define XML_TOK_NAME_ASTERISK 31 /* name* */
+#define XML_TOK_NAME_PLUS 32 /* name+ */
+#define XML_TOK_COND_SECT_OPEN 33 /* <![ */
+#define XML_TOK_COND_SECT_CLOSE 34 /* ]]> */
+#define XML_TOK_CLOSE_PAREN_QUESTION 35 /* )? */
+#define XML_TOK_CLOSE_PAREN_ASTERISK 36 /* )* */
+#define XML_TOK_CLOSE_PAREN_PLUS 37 /* )+ */
+#define XML_TOK_COMMA 38
+
+/* The following token is returned only by XmlAttributeValueTok */
+#define XML_TOK_ATTRIBUTE_VALUE_S 39
+
+/* The following token is returned only by XmlCdataSectionTok */
+#define XML_TOK_CDATA_SECT_CLOSE 40
+
+/* With namespace processing this is returned by XmlPrologTok for a
+ name with a colon.
+*/
+#define XML_TOK_PREFIXED_NAME 41
+
+#ifdef XML_DTD
+#define XML_TOK_IGNORE_SECT 42
+#endif /* XML_DTD */
+
+#ifdef XML_DTD
+#define XML_N_STATES 4
+#else /* not XML_DTD */
+#define XML_N_STATES 3
+#endif /* not XML_DTD */
+
+#define XML_PROLOG_STATE 0
+#define XML_CONTENT_STATE 1
+#define XML_CDATA_SECTION_STATE 2
+#ifdef XML_DTD
+#define XML_IGNORE_SECTION_STATE 3
+#endif /* XML_DTD */
+
+#define XML_N_LITERAL_TYPES 2
+#define XML_ATTRIBUTE_VALUE_LITERAL 0
+#define XML_ENTITY_VALUE_LITERAL 1
+
+/* The size of the buffer passed to XmlUtf8Encode must be at least this. */
+#define XML_UTF8_ENCODE_MAX 4
+/* The size of the buffer passed to XmlUtf16Encode must be at least this. */
+#define XML_UTF16_ENCODE_MAX 2
+
+typedef struct position {
+ /* first line and first column are 0 not 1 */
+ XML_Size lineNumber;
+ XML_Size columnNumber;
+} POSITION;
+
+typedef struct {
+ const char *name;
+ const char *valuePtr;
+ const char *valueEnd;
+ char normalized;
+} ATTRIBUTE;
+
+struct encoding;
+typedef struct encoding ENCODING;
+
+typedef int (PTRCALL *SCANNER)(const ENCODING *,
+ const char *,
+ const char *,
+ const char **);
+
+struct encoding {
+ SCANNER scanners[XML_N_STATES];
+ SCANNER literalScanners[XML_N_LITERAL_TYPES];
+ int (PTRCALL *sameName)(const ENCODING *,
+ const char *,
+ const char *);
+ int (PTRCALL *nameMatchesAscii)(const ENCODING *,
+ const char *,
+ const char *,
+ const char *);
+ int (PTRFASTCALL *nameLength)(const ENCODING *, const char *);
+ const char *(PTRFASTCALL *skipS)(const ENCODING *, const char *);
+ int (PTRCALL *getAtts)(const ENCODING *enc,
+ const char *ptr,
+ int attsMax,
+ ATTRIBUTE *atts);
+ int (PTRFASTCALL *charRefNumber)(const ENCODING *enc, const char *ptr);
+ int (PTRCALL *predefinedEntityName)(const ENCODING *,
+ const char *,
+ const char *);
+ void (PTRCALL *updatePosition)(const ENCODING *,
+ const char *ptr,
+ const char *end,
+ POSITION *);
+ int (PTRCALL *isPublicId)(const ENCODING *enc,
+ const char *ptr,
+ const char *end,
+ const char **badPtr);
+ void (PTRCALL *utf8Convert)(const ENCODING *enc,
+ const char **fromP,
+ const char *fromLim,
+ char **toP,
+ const char *toLim);
+ void (PTRCALL *utf16Convert)(const ENCODING *enc,
+ const char **fromP,
+ const char *fromLim,
+ unsigned short **toP,
+ const unsigned short *toLim);
+ int minBytesPerChar;
+ char isUtf8;
+ char isUtf16;
+};
+
+/* Scan the string starting at ptr until the end of the next complete
+ token, but do not scan past eptr. Return an integer giving the
+ type of token.
+
+ Return XML_TOK_NONE when ptr == eptr; nextTokPtr will not be set.
+
+ Return XML_TOK_PARTIAL when the string does not contain a complete
+ token; nextTokPtr will not be set.
+
+ Return XML_TOK_INVALID when the string does not start a valid
+ token; nextTokPtr will be set to point to the character which made
+ the token invalid.
+
+ Otherwise the string starts with a valid token; nextTokPtr will be
+ set to point to the character following the end of that token.
+
+ Each data character counts as a single token, but adjacent data
+ characters may be returned together. Similarly for characters in
+ the prolog outside literals, comments and processing instructions.
+*/
+
+
+#define XmlTok(enc, state, ptr, end, nextTokPtr) \
+ (((enc)->scanners[state])(enc, ptr, end, nextTokPtr))
+
+#define XmlPrologTok(enc, ptr, end, nextTokPtr) \
+ XmlTok(enc, XML_PROLOG_STATE, ptr, end, nextTokPtr)
+
+#define XmlContentTok(enc, ptr, end, nextTokPtr) \
+ XmlTok(enc, XML_CONTENT_STATE, ptr, end, nextTokPtr)
+
+#define XmlCdataSectionTok(enc, ptr, end, nextTokPtr) \
+ XmlTok(enc, XML_CDATA_SECTION_STATE, ptr, end, nextTokPtr)
+
+#ifdef XML_DTD
+
+#define XmlIgnoreSectionTok(enc, ptr, end, nextTokPtr) \
+ XmlTok(enc, XML_IGNORE_SECTION_STATE, ptr, end, nextTokPtr)
+
+#endif /* XML_DTD */
+
+/* This is used for performing a 2nd-level tokenization on the content
+ of a literal that has already been returned by XmlTok.
+*/
+#define XmlLiteralTok(enc, literalType, ptr, end, nextTokPtr) \
+ (((enc)->literalScanners[literalType])(enc, ptr, end, nextTokPtr))
+
+#define XmlAttributeValueTok(enc, ptr, end, nextTokPtr) \
+ XmlLiteralTok(enc, XML_ATTRIBUTE_VALUE_LITERAL, ptr, end, nextTokPtr)
+
+#define XmlEntityValueTok(enc, ptr, end, nextTokPtr) \
+ XmlLiteralTok(enc, XML_ENTITY_VALUE_LITERAL, ptr, end, nextTokPtr)
+
+#define XmlSameName(enc, ptr1, ptr2) (((enc)->sameName)(enc, ptr1, ptr2))
+
+#define XmlNameMatchesAscii(enc, ptr1, end1, ptr2) \
+ (((enc)->nameMatchesAscii)(enc, ptr1, end1, ptr2))
+
+#define XmlNameLength(enc, ptr) \
+ (((enc)->nameLength)(enc, ptr))
+
+#define XmlSkipS(enc, ptr) \
+ (((enc)->skipS)(enc, ptr))
+
+#define XmlGetAttributes(enc, ptr, attsMax, atts) \
+ (((enc)->getAtts)(enc, ptr, attsMax, atts))
+
+#define XmlCharRefNumber(enc, ptr) \
+ (((enc)->charRefNumber)(enc, ptr))
+
+#define XmlPredefinedEntityName(enc, ptr, end) \
+ (((enc)->predefinedEntityName)(enc, ptr, end))
+
+#define XmlUpdatePosition(enc, ptr, end, pos) \
+ (((enc)->updatePosition)(enc, ptr, end, pos))
+
+#define XmlIsPublicId(enc, ptr, end, badPtr) \
+ (((enc)->isPublicId)(enc, ptr, end, badPtr))
+
+#define XmlUtf8Convert(enc, fromP, fromLim, toP, toLim) \
+ (((enc)->utf8Convert)(enc, fromP, fromLim, toP, toLim))
+
+#define XmlUtf16Convert(enc, fromP, fromLim, toP, toLim) \
+ (((enc)->utf16Convert)(enc, fromP, fromLim, toP, toLim))
+
+typedef struct {
+ ENCODING initEnc;
+ const ENCODING **encPtr;
+} INIT_ENCODING;
+
+int XmlParseXmlDecl(int isGeneralTextEntity,
+ const ENCODING *enc,
+ const char *ptr,
+ const char *end,
+ const char **badPtr,
+ const char **versionPtr,
+ const char **versionEndPtr,
+ const char **encodingNamePtr,
+ const ENCODING **namedEncodingPtr,
+ int *standalonePtr);
+
+int XmlInitEncoding(INIT_ENCODING *, const ENCODING **, const char *name);
+const ENCODING *XmlGetUtf8InternalEncoding(void);
+const ENCODING *XmlGetUtf16InternalEncoding(void);
+int FASTCALL XmlUtf8Encode(int charNumber, char *buf);
+int FASTCALL XmlUtf16Encode(int charNumber, unsigned short *buf);
+int XmlSizeOfUnknownEncoding(void);
+
+
+typedef int (XMLCALL *CONVERTER) (void *userData, const char *p);
+
+ENCODING *
+XmlInitUnknownEncoding(void *mem,
+ int *table,
+ CONVERTER convert,
+ void *userData);
+
+int XmlParseXmlDeclNS(int isGeneralTextEntity,
+ const ENCODING *enc,
+ const char *ptr,
+ const char *end,
+ const char **badPtr,
+ const char **versionPtr,
+ const char **versionEndPtr,
+ const char **encodingNamePtr,
+ const ENCODING **namedEncodingPtr,
+ int *standalonePtr);
+
+int XmlInitEncodingNS(INIT_ENCODING *, const ENCODING **, const char *name);
+const ENCODING *XmlGetUtf8InternalEncodingNS(void);
+const ENCODING *XmlGetUtf16InternalEncodingNS(void);
+ENCODING *
+XmlInitUnknownEncodingNS(void *mem,
+ int *table,
+ CONVERTER convert,
+ void *userData);
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* not XmlTok_INCLUDED */
diff --git a/src/lib/expat/xmltok_impl.c b/src/lib/expat/xmltok_impl.c
new file mode 100644
index 00000000000..0ee57abb1f4
--- /dev/null
+++ b/src/lib/expat/xmltok_impl.c
@@ -0,0 +1,1779 @@
+/* Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+ See the file COPYING for copying permission.
+*/
+
+#ifndef IS_INVALID_CHAR
+#define IS_INVALID_CHAR(enc, ptr, n) (0)
+#endif
+
+#define INVALID_LEAD_CASE(n, ptr, nextTokPtr) \
+ case BT_LEAD ## n: \
+ if (end - ptr < n) \
+ return XML_TOK_PARTIAL_CHAR; \
+ if (IS_INVALID_CHAR(enc, ptr, n)) { \
+ *(nextTokPtr) = (ptr); \
+ return XML_TOK_INVALID; \
+ } \
+ ptr += n; \
+ break;
+
+#define INVALID_CASES(ptr, nextTokPtr) \
+ INVALID_LEAD_CASE(2, ptr, nextTokPtr) \
+ INVALID_LEAD_CASE(3, ptr, nextTokPtr) \
+ INVALID_LEAD_CASE(4, ptr, nextTokPtr) \
+ case BT_NONXML: \
+ case BT_MALFORM: \
+ case BT_TRAIL: \
+ *(nextTokPtr) = (ptr); \
+ return XML_TOK_INVALID;
+
+#define CHECK_NAME_CASE(n, enc, ptr, end, nextTokPtr) \
+ case BT_LEAD ## n: \
+ if (end - ptr < n) \
+ return XML_TOK_PARTIAL_CHAR; \
+ if (!IS_NAME_CHAR(enc, ptr, n)) { \
+ *nextTokPtr = ptr; \
+ return XML_TOK_INVALID; \
+ } \
+ ptr += n; \
+ break;
+
+#define CHECK_NAME_CASES(enc, ptr, end, nextTokPtr) \
+ case BT_NONASCII: \
+ if (!IS_NAME_CHAR_MINBPC(enc, ptr)) { \
+ *nextTokPtr = ptr; \
+ return XML_TOK_INVALID; \
+ } \
+ case BT_NMSTRT: \
+ case BT_HEX: \
+ case BT_DIGIT: \
+ case BT_NAME: \
+ case BT_MINUS: \
+ ptr += MINBPC(enc); \
+ break; \
+ CHECK_NAME_CASE(2, enc, ptr, end, nextTokPtr) \
+ CHECK_NAME_CASE(3, enc, ptr, end, nextTokPtr) \
+ CHECK_NAME_CASE(4, enc, ptr, end, nextTokPtr)
+
+#define CHECK_NMSTRT_CASE(n, enc, ptr, end, nextTokPtr) \
+ case BT_LEAD ## n: \
+ if (end - ptr < n) \
+ return XML_TOK_PARTIAL_CHAR; \
+ if (!IS_NMSTRT_CHAR(enc, ptr, n)) { \
+ *nextTokPtr = ptr; \
+ return XML_TOK_INVALID; \
+ } \
+ ptr += n; \
+ break;
+
+#define CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr) \
+ case BT_NONASCII: \
+ if (!IS_NMSTRT_CHAR_MINBPC(enc, ptr)) { \
+ *nextTokPtr = ptr; \
+ return XML_TOK_INVALID; \
+ } \
+ case BT_NMSTRT: \
+ case BT_HEX: \
+ ptr += MINBPC(enc); \
+ break; \
+ CHECK_NMSTRT_CASE(2, enc, ptr, end, nextTokPtr) \
+ CHECK_NMSTRT_CASE(3, enc, ptr, end, nextTokPtr) \
+ CHECK_NMSTRT_CASE(4, enc, ptr, end, nextTokPtr)
+
+#ifndef PREFIX
+#define PREFIX(ident) ident
+#endif
+
+/* ptr points to character following "<!-" */
+
+static int PTRCALL
+PREFIX(scanComment)(const ENCODING *enc, const char *ptr,
+ const char *end, const char **nextTokPtr)
+{
+ if (ptr != end) {
+ if (!CHAR_MATCHES(enc, ptr, ASCII_MINUS)) {
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ ptr += MINBPC(enc);
+ while (ptr != end) {
+ switch (BYTE_TYPE(enc, ptr)) {
+ INVALID_CASES(ptr, nextTokPtr)
+ case BT_MINUS:
+ if ((ptr += MINBPC(enc)) == end)
+ return XML_TOK_PARTIAL;
+ if (CHAR_MATCHES(enc, ptr, ASCII_MINUS)) {
+ if ((ptr += MINBPC(enc)) == end)
+ return XML_TOK_PARTIAL;
+ if (!CHAR_MATCHES(enc, ptr, ASCII_GT)) {
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_COMMENT;
+ }
+ break;
+ default:
+ ptr += MINBPC(enc);
+ break;
+ }
+ }
+ }
+ return XML_TOK_PARTIAL;
+}
+
+/* ptr points to character following "<!" */
+
+static int PTRCALL
+PREFIX(scanDecl)(const ENCODING *enc, const char *ptr,
+ const char *end, const char **nextTokPtr)
+{
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ switch (BYTE_TYPE(enc, ptr)) {
+ case BT_MINUS:
+ return PREFIX(scanComment)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+ case BT_LSQB:
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_COND_SECT_OPEN;
+ case BT_NMSTRT:
+ case BT_HEX:
+ ptr += MINBPC(enc);
+ break;
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ while (ptr != end) {
+ switch (BYTE_TYPE(enc, ptr)) {
+ case BT_PERCNT:
+ if (ptr + MINBPC(enc) == end)
+ return XML_TOK_PARTIAL;
+ /* don't allow <!ENTITY% foo "whatever"> */
+ switch (BYTE_TYPE(enc, ptr + MINBPC(enc))) {
+ case BT_S: case BT_CR: case BT_LF: case BT_PERCNT:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ /* fall through */
+ case BT_S: case BT_CR: case BT_LF:
+ *nextTokPtr = ptr;
+ return XML_TOK_DECL_OPEN;
+ case BT_NMSTRT:
+ case BT_HEX:
+ ptr += MINBPC(enc);
+ break;
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ }
+ return XML_TOK_PARTIAL;
+}
+
+static int PTRCALL
+PREFIX(checkPiTarget)(const ENCODING *enc, const char *ptr,
+ const char *end, int *tokPtr)
+{
+ int upper = 0;
+ *tokPtr = XML_TOK_PI;
+ if (end - ptr != MINBPC(enc)*3)
+ return 1;
+ switch (BYTE_TO_ASCII(enc, ptr)) {
+ case ASCII_x:
+ break;
+ case ASCII_X:
+ upper = 1;
+ break;
+ default:
+ return 1;
+ }
+ ptr += MINBPC(enc);
+ switch (BYTE_TO_ASCII(enc, ptr)) {
+ case ASCII_m:
+ break;
+ case ASCII_M:
+ upper = 1;
+ break;
+ default:
+ return 1;
+ }
+ ptr += MINBPC(enc);
+ switch (BYTE_TO_ASCII(enc, ptr)) {
+ case ASCII_l:
+ break;
+ case ASCII_L:
+ upper = 1;
+ break;
+ default:
+ return 1;
+ }
+ if (upper)
+ return 0;
+ *tokPtr = XML_TOK_XML_DECL;
+ return 1;
+}
+
+/* ptr points to character following "<?" */
+
+static int PTRCALL
+PREFIX(scanPi)(const ENCODING *enc, const char *ptr,
+ const char *end, const char **nextTokPtr)
+{
+ int tok;
+ const char *target = ptr;
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ switch (BYTE_TYPE(enc, ptr)) {
+ CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ while (ptr != end) {
+ switch (BYTE_TYPE(enc, ptr)) {
+ CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+ case BT_S: case BT_CR: case BT_LF:
+ if (!PREFIX(checkPiTarget)(enc, target, ptr, &tok)) {
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ ptr += MINBPC(enc);
+ while (ptr != end) {
+ switch (BYTE_TYPE(enc, ptr)) {
+ INVALID_CASES(ptr, nextTokPtr)
+ case BT_QUEST:
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ if (CHAR_MATCHES(enc, ptr, ASCII_GT)) {
+ *nextTokPtr = ptr + MINBPC(enc);
+ return tok;
+ }
+ break;
+ default:
+ ptr += MINBPC(enc);
+ break;
+ }
+ }
+ return XML_TOK_PARTIAL;
+ case BT_QUEST:
+ if (!PREFIX(checkPiTarget)(enc, target, ptr, &tok)) {
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ if (CHAR_MATCHES(enc, ptr, ASCII_GT)) {
+ *nextTokPtr = ptr + MINBPC(enc);
+ return tok;
+ }
+ /* fall through */
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ }
+ return XML_TOK_PARTIAL;
+}
+
+static int PTRCALL
+PREFIX(scanCdataSection)(const ENCODING *enc, const char *ptr,
+ const char *end, const char **nextTokPtr)
+{
+ static const char CDATA_LSQB[] = { ASCII_C, ASCII_D, ASCII_A,
+ ASCII_T, ASCII_A, ASCII_LSQB };
+ int i;
+ /* CDATA[ */
+ if (end - ptr < 6 * MINBPC(enc))
+ return XML_TOK_PARTIAL;
+ for (i = 0; i < 6; i++, ptr += MINBPC(enc)) {
+ if (!CHAR_MATCHES(enc, ptr, CDATA_LSQB[i])) {
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ }
+ *nextTokPtr = ptr;
+ return XML_TOK_CDATA_SECT_OPEN;
+}
+
+static int PTRCALL
+PREFIX(cdataSectionTok)(const ENCODING *enc, const char *ptr,
+ const char *end, const char **nextTokPtr)
+{
+ if (ptr == end)
+ return XML_TOK_NONE;
+ if (MINBPC(enc) > 1) {
+ size_t n = end - ptr;
+ if (n & (MINBPC(enc) - 1)) {
+ n &= ~(MINBPC(enc) - 1);
+ if (n == 0)
+ return XML_TOK_PARTIAL;
+ end = ptr + n;
+ }
+ }
+ switch (BYTE_TYPE(enc, ptr)) {
+ case BT_RSQB:
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ if (!CHAR_MATCHES(enc, ptr, ASCII_RSQB))
+ break;
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ if (!CHAR_MATCHES(enc, ptr, ASCII_GT)) {
+ ptr -= MINBPC(enc);
+ break;
+ }
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_CDATA_SECT_CLOSE;
+ case BT_CR:
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ if (BYTE_TYPE(enc, ptr) == BT_LF)
+ ptr += MINBPC(enc);
+ *nextTokPtr = ptr;
+ return XML_TOK_DATA_NEWLINE;
+ case BT_LF:
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_DATA_NEWLINE;
+ INVALID_CASES(ptr, nextTokPtr)
+ default:
+ ptr += MINBPC(enc);
+ break;
+ }
+ while (ptr != end) {
+ switch (BYTE_TYPE(enc, ptr)) {
+#define LEAD_CASE(n) \
+ case BT_LEAD ## n: \
+ if (end - ptr < n || IS_INVALID_CHAR(enc, ptr, n)) { \
+ *nextTokPtr = ptr; \
+ return XML_TOK_DATA_CHARS; \
+ } \
+ ptr += n; \
+ break;
+ LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
+#undef LEAD_CASE
+ case BT_NONXML:
+ case BT_MALFORM:
+ case BT_TRAIL:
+ case BT_CR:
+ case BT_LF:
+ case BT_RSQB:
+ *nextTokPtr = ptr;
+ return XML_TOK_DATA_CHARS;
+ default:
+ ptr += MINBPC(enc);
+ break;
+ }
+ }
+ *nextTokPtr = ptr;
+ return XML_TOK_DATA_CHARS;
+}
+
+/* ptr points to character following "</" */
+
+static int PTRCALL
+PREFIX(scanEndTag)(const ENCODING *enc, const char *ptr,
+ const char *end, const char **nextTokPtr)
+{
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ switch (BYTE_TYPE(enc, ptr)) {
+ CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ while (ptr != end) {
+ switch (BYTE_TYPE(enc, ptr)) {
+ CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+ case BT_S: case BT_CR: case BT_LF:
+ for (ptr += MINBPC(enc); ptr != end; ptr += MINBPC(enc)) {
+ switch (BYTE_TYPE(enc, ptr)) {
+ case BT_S: case BT_CR: case BT_LF:
+ break;
+ case BT_GT:
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_END_TAG;
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ }
+ return XML_TOK_PARTIAL;
+#ifdef XML_NS
+ case BT_COLON:
+ /* no need to check qname syntax here,
+ since end-tag must match exactly */
+ ptr += MINBPC(enc);
+ break;
+#endif
+ case BT_GT:
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_END_TAG;
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ }
+ return XML_TOK_PARTIAL;
+}
+
+/* ptr points to character following "&#X" */
+
+static int PTRCALL
+PREFIX(scanHexCharRef)(const ENCODING *enc, const char *ptr,
+ const char *end, const char **nextTokPtr)
+{
+ if (ptr != end) {
+ switch (BYTE_TYPE(enc, ptr)) {
+ case BT_DIGIT:
+ case BT_HEX:
+ break;
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ for (ptr += MINBPC(enc); ptr != end; ptr += MINBPC(enc)) {
+ switch (BYTE_TYPE(enc, ptr)) {
+ case BT_DIGIT:
+ case BT_HEX:
+ break;
+ case BT_SEMI:
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_CHAR_REF;
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ }
+ }
+ return XML_TOK_PARTIAL;
+}
+
+/* ptr points to character following "&#" */
+
+static int PTRCALL
+PREFIX(scanCharRef)(const ENCODING *enc, const char *ptr,
+ const char *end, const char **nextTokPtr)
+{
+ if (ptr != end) {
+ if (CHAR_MATCHES(enc, ptr, ASCII_x))
+ return PREFIX(scanHexCharRef)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+ switch (BYTE_TYPE(enc, ptr)) {
+ case BT_DIGIT:
+ break;
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ for (ptr += MINBPC(enc); ptr != end; ptr += MINBPC(enc)) {
+ switch (BYTE_TYPE(enc, ptr)) {
+ case BT_DIGIT:
+ break;
+ case BT_SEMI:
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_CHAR_REF;
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ }
+ }
+ return XML_TOK_PARTIAL;
+}
+
+/* ptr points to character following "&" */
+
+static int PTRCALL
+PREFIX(scanRef)(const ENCODING *enc, const char *ptr, const char *end,
+ const char **nextTokPtr)
+{
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ switch (BYTE_TYPE(enc, ptr)) {
+ CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
+ case BT_NUM:
+ return PREFIX(scanCharRef)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ while (ptr != end) {
+ switch (BYTE_TYPE(enc, ptr)) {
+ CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+ case BT_SEMI:
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_ENTITY_REF;
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ }
+ return XML_TOK_PARTIAL;
+}
+
+/* ptr points to character following first character of attribute name */
+
+static int PTRCALL
+PREFIX(scanAtts)(const ENCODING *enc, const char *ptr, const char *end,
+ const char **nextTokPtr)
+{
+#ifdef XML_NS
+ int hadColon = 0;
+#endif
+ while (ptr != end) {
+ switch (BYTE_TYPE(enc, ptr)) {
+ CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+#ifdef XML_NS
+ case BT_COLON:
+ if (hadColon) {
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ hadColon = 1;
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ switch (BYTE_TYPE(enc, ptr)) {
+ CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ break;
+#endif
+ case BT_S: case BT_CR: case BT_LF:
+ for (;;) {
+ int t;
+
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ t = BYTE_TYPE(enc, ptr);
+ if (t == BT_EQUALS)
+ break;
+ switch (t) {
+ case BT_S:
+ case BT_LF:
+ case BT_CR:
+ break;
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ }
+ /* fall through */
+ case BT_EQUALS:
+ {
+ int open;
+#ifdef XML_NS
+ hadColon = 0;
+#endif
+ for (;;) {
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ open = BYTE_TYPE(enc, ptr);
+ if (open == BT_QUOT || open == BT_APOS)
+ break;
+ switch (open) {
+ case BT_S:
+ case BT_LF:
+ case BT_CR:
+ break;
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ }
+ ptr += MINBPC(enc);
+ /* in attribute value */
+ for (;;) {
+ int t;
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ t = BYTE_TYPE(enc, ptr);
+ if (t == open)
+ break;
+ switch (t) {
+ INVALID_CASES(ptr, nextTokPtr)
+ case BT_AMP:
+ {
+ int tok = PREFIX(scanRef)(enc, ptr + MINBPC(enc), end, &ptr);
+ if (tok <= 0) {
+ if (tok == XML_TOK_INVALID)
+ *nextTokPtr = ptr;
+ return tok;
+ }
+ break;
+ }
+ case BT_LT:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ default:
+ ptr += MINBPC(enc);
+ break;
+ }
+ }
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ switch (BYTE_TYPE(enc, ptr)) {
+ case BT_S:
+ case BT_CR:
+ case BT_LF:
+ break;
+ case BT_SOL:
+ goto sol;
+ case BT_GT:
+ goto gt;
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ /* ptr points to closing quote */
+ for (;;) {
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ switch (BYTE_TYPE(enc, ptr)) {
+ CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
+ case BT_S: case BT_CR: case BT_LF:
+ continue;
+ case BT_GT:
+ gt:
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_START_TAG_WITH_ATTS;
+ case BT_SOL:
+ sol:
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ if (!CHAR_MATCHES(enc, ptr, ASCII_GT)) {
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_EMPTY_ELEMENT_WITH_ATTS;
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ break;
+ }
+ break;
+ }
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ }
+ return XML_TOK_PARTIAL;
+}
+
+/* ptr points to character following "<" */
+
+static int PTRCALL
+PREFIX(scanLt)(const ENCODING *enc, const char *ptr, const char *end,
+ const char **nextTokPtr)
+{
+#ifdef XML_NS
+ int hadColon;
+#endif
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ switch (BYTE_TYPE(enc, ptr)) {
+ CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
+ case BT_EXCL:
+ if ((ptr += MINBPC(enc)) == end)
+ return XML_TOK_PARTIAL;
+ switch (BYTE_TYPE(enc, ptr)) {
+ case BT_MINUS:
+ return PREFIX(scanComment)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+ case BT_LSQB:
+ return PREFIX(scanCdataSection)(enc, ptr + MINBPC(enc),
+ end, nextTokPtr);
+ }
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ case BT_QUEST:
+ return PREFIX(scanPi)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+ case BT_SOL:
+ return PREFIX(scanEndTag)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+#ifdef XML_NS
+ hadColon = 0;
+#endif
+ /* we have a start-tag */
+ while (ptr != end) {
+ switch (BYTE_TYPE(enc, ptr)) {
+ CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+#ifdef XML_NS
+ case BT_COLON:
+ if (hadColon) {
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ hadColon = 1;
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ switch (BYTE_TYPE(enc, ptr)) {
+ CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ break;
+#endif
+ case BT_S: case BT_CR: case BT_LF:
+ {
+ ptr += MINBPC(enc);
+ while (ptr != end) {
+ switch (BYTE_TYPE(enc, ptr)) {
+ CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
+ case BT_GT:
+ goto gt;
+ case BT_SOL:
+ goto sol;
+ case BT_S: case BT_CR: case BT_LF:
+ ptr += MINBPC(enc);
+ continue;
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ return PREFIX(scanAtts)(enc, ptr, end, nextTokPtr);
+ }
+ return XML_TOK_PARTIAL;
+ }
+ case BT_GT:
+ gt:
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_START_TAG_NO_ATTS;
+ case BT_SOL:
+ sol:
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ if (!CHAR_MATCHES(enc, ptr, ASCII_GT)) {
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_EMPTY_ELEMENT_NO_ATTS;
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ }
+ return XML_TOK_PARTIAL;
+}
+
+static int PTRCALL
+PREFIX(contentTok)(const ENCODING *enc, const char *ptr, const char *end,
+ const char **nextTokPtr)
+{
+ if (ptr == end)
+ return XML_TOK_NONE;
+ if (MINBPC(enc) > 1) {
+ size_t n = end - ptr;
+ if (n & (MINBPC(enc) - 1)) {
+ n &= ~(MINBPC(enc) - 1);
+ if (n == 0)
+ return XML_TOK_PARTIAL;
+ end = ptr + n;
+ }
+ }
+ switch (BYTE_TYPE(enc, ptr)) {
+ case BT_LT:
+ return PREFIX(scanLt)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+ case BT_AMP:
+ return PREFIX(scanRef)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+ case BT_CR:
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ return XML_TOK_TRAILING_CR;
+ if (BYTE_TYPE(enc, ptr) == BT_LF)
+ ptr += MINBPC(enc);
+ *nextTokPtr = ptr;
+ return XML_TOK_DATA_NEWLINE;
+ case BT_LF:
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_DATA_NEWLINE;
+ case BT_RSQB:
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ return XML_TOK_TRAILING_RSQB;
+ if (!CHAR_MATCHES(enc, ptr, ASCII_RSQB))
+ break;
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ return XML_TOK_TRAILING_RSQB;
+ if (!CHAR_MATCHES(enc, ptr, ASCII_GT)) {
+ ptr -= MINBPC(enc);
+ break;
+ }
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ INVALID_CASES(ptr, nextTokPtr)
+ default:
+ ptr += MINBPC(enc);
+ break;
+ }
+ while (ptr != end) {
+ switch (BYTE_TYPE(enc, ptr)) {
+#define LEAD_CASE(n) \
+ case BT_LEAD ## n: \
+ if (end - ptr < n || IS_INVALID_CHAR(enc, ptr, n)) { \
+ *nextTokPtr = ptr; \
+ return XML_TOK_DATA_CHARS; \
+ } \
+ ptr += n; \
+ break;
+ LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
+#undef LEAD_CASE
+ case BT_RSQB:
+ if (ptr + MINBPC(enc) != end) {
+ if (!CHAR_MATCHES(enc, ptr + MINBPC(enc), ASCII_RSQB)) {
+ ptr += MINBPC(enc);
+ break;
+ }
+ if (ptr + 2*MINBPC(enc) != end) {
+ if (!CHAR_MATCHES(enc, ptr + 2*MINBPC(enc), ASCII_GT)) {
+ ptr += MINBPC(enc);
+ break;
+ }
+ *nextTokPtr = ptr + 2*MINBPC(enc);
+ return XML_TOK_INVALID;
+ }
+ }
+ /* fall through */
+ case BT_AMP:
+ case BT_LT:
+ case BT_NONXML:
+ case BT_MALFORM:
+ case BT_TRAIL:
+ case BT_CR:
+ case BT_LF:
+ *nextTokPtr = ptr;
+ return XML_TOK_DATA_CHARS;
+ default:
+ ptr += MINBPC(enc);
+ break;
+ }
+ }
+ *nextTokPtr = ptr;
+ return XML_TOK_DATA_CHARS;
+}
+
+/* ptr points to character following "%" */
+
+static int PTRCALL
+PREFIX(scanPercent)(const ENCODING *enc, const char *ptr, const char *end,
+ const char **nextTokPtr)
+{
+ if (ptr == end)
+ return -XML_TOK_PERCENT;
+ switch (BYTE_TYPE(enc, ptr)) {
+ CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
+ case BT_S: case BT_LF: case BT_CR: case BT_PERCNT:
+ *nextTokPtr = ptr;
+ return XML_TOK_PERCENT;
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ while (ptr != end) {
+ switch (BYTE_TYPE(enc, ptr)) {
+ CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+ case BT_SEMI:
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_PARAM_ENTITY_REF;
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ }
+ return XML_TOK_PARTIAL;
+}
+
+static int PTRCALL
+PREFIX(scanPoundName)(const ENCODING *enc, const char *ptr, const char *end,
+ const char **nextTokPtr)
+{
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ switch (BYTE_TYPE(enc, ptr)) {
+ CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ while (ptr != end) {
+ switch (BYTE_TYPE(enc, ptr)) {
+ CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+ case BT_CR: case BT_LF: case BT_S:
+ case BT_RPAR: case BT_GT: case BT_PERCNT: case BT_VERBAR:
+ *nextTokPtr = ptr;
+ return XML_TOK_POUND_NAME;
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ }
+ return -XML_TOK_POUND_NAME;
+}
+
+static int PTRCALL
+PREFIX(scanLit)(int open, const ENCODING *enc,
+ const char *ptr, const char *end,
+ const char **nextTokPtr)
+{
+ while (ptr != end) {
+ int t = BYTE_TYPE(enc, ptr);
+ switch (t) {
+ INVALID_CASES(ptr, nextTokPtr)
+ case BT_QUOT:
+ case BT_APOS:
+ ptr += MINBPC(enc);
+ if (t != open)
+ break;
+ if (ptr == end)
+ return -XML_TOK_LITERAL;
+ *nextTokPtr = ptr;
+ switch (BYTE_TYPE(enc, ptr)) {
+ case BT_S: case BT_CR: case BT_LF:
+ case BT_GT: case BT_PERCNT: case BT_LSQB:
+ return XML_TOK_LITERAL;
+ default:
+ return XML_TOK_INVALID;
+ }
+ default:
+ ptr += MINBPC(enc);
+ break;
+ }
+ }
+ return XML_TOK_PARTIAL;
+}
+
+static int PTRCALL
+PREFIX(prologTok)(const ENCODING *enc, const char *ptr, const char *end,
+ const char **nextTokPtr)
+{
+ int tok;
+ if (ptr == end)
+ return XML_TOK_NONE;
+ if (MINBPC(enc) > 1) {
+ size_t n = end - ptr;
+ if (n & (MINBPC(enc) - 1)) {
+ n &= ~(MINBPC(enc) - 1);
+ if (n == 0)
+ return XML_TOK_PARTIAL;
+ end = ptr + n;
+ }
+ }
+ switch (BYTE_TYPE(enc, ptr)) {
+ case BT_QUOT:
+ return PREFIX(scanLit)(BT_QUOT, enc, ptr + MINBPC(enc), end, nextTokPtr);
+ case BT_APOS:
+ return PREFIX(scanLit)(BT_APOS, enc, ptr + MINBPC(enc), end, nextTokPtr);
+ case BT_LT:
+ {
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ switch (BYTE_TYPE(enc, ptr)) {
+ case BT_EXCL:
+ return PREFIX(scanDecl)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+ case BT_QUEST:
+ return PREFIX(scanPi)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+ case BT_NMSTRT:
+ case BT_HEX:
+ case BT_NONASCII:
+ case BT_LEAD2:
+ case BT_LEAD3:
+ case BT_LEAD4:
+ *nextTokPtr = ptr - MINBPC(enc);
+ return XML_TOK_INSTANCE_START;
+ }
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ case BT_CR:
+ if (ptr + MINBPC(enc) == end) {
+ *nextTokPtr = end;
+ /* indicate that this might be part of a CR/LF pair */
+ return -XML_TOK_PROLOG_S;
+ }
+ /* fall through */
+ case BT_S: case BT_LF:
+ for (;;) {
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ break;
+ switch (BYTE_TYPE(enc, ptr)) {
+ case BT_S: case BT_LF:
+ break;
+ case BT_CR:
+ /* don't split CR/LF pair */
+ if (ptr + MINBPC(enc) != end)
+ break;
+ /* fall through */
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_PROLOG_S;
+ }
+ }
+ *nextTokPtr = ptr;
+ return XML_TOK_PROLOG_S;
+ case BT_PERCNT:
+ return PREFIX(scanPercent)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+ case BT_COMMA:
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_COMMA;
+ case BT_LSQB:
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_OPEN_BRACKET;
+ case BT_RSQB:
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ return -XML_TOK_CLOSE_BRACKET;
+ if (CHAR_MATCHES(enc, ptr, ASCII_RSQB)) {
+ if (ptr + MINBPC(enc) == end)
+ return XML_TOK_PARTIAL;
+ if (CHAR_MATCHES(enc, ptr + MINBPC(enc), ASCII_GT)) {
+ *nextTokPtr = ptr + 2*MINBPC(enc);
+ return XML_TOK_COND_SECT_CLOSE;
+ }
+ }
+ *nextTokPtr = ptr;
+ return XML_TOK_CLOSE_BRACKET;
+ case BT_LPAR:
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_OPEN_PAREN;
+ case BT_RPAR:
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ return -XML_TOK_CLOSE_PAREN;
+ switch (BYTE_TYPE(enc, ptr)) {
+ case BT_AST:
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_CLOSE_PAREN_ASTERISK;
+ case BT_QUEST:
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_CLOSE_PAREN_QUESTION;
+ case BT_PLUS:
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_CLOSE_PAREN_PLUS;
+ case BT_CR: case BT_LF: case BT_S:
+ case BT_GT: case BT_COMMA: case BT_VERBAR:
+ case BT_RPAR:
+ *nextTokPtr = ptr;
+ return XML_TOK_CLOSE_PAREN;
+ }
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ case BT_VERBAR:
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_OR;
+ case BT_GT:
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_DECL_CLOSE;
+ case BT_NUM:
+ return PREFIX(scanPoundName)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+#define LEAD_CASE(n) \
+ case BT_LEAD ## n: \
+ if (end - ptr < n) \
+ return XML_TOK_PARTIAL_CHAR; \
+ if (IS_NMSTRT_CHAR(enc, ptr, n)) { \
+ ptr += n; \
+ tok = XML_TOK_NAME; \
+ break; \
+ } \
+ if (IS_NAME_CHAR(enc, ptr, n)) { \
+ ptr += n; \
+ tok = XML_TOK_NMTOKEN; \
+ break; \
+ } \
+ *nextTokPtr = ptr; \
+ return XML_TOK_INVALID;
+ LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
+#undef LEAD_CASE
+ case BT_NMSTRT:
+ case BT_HEX:
+ tok = XML_TOK_NAME;
+ ptr += MINBPC(enc);
+ break;
+ case BT_DIGIT:
+ case BT_NAME:
+ case BT_MINUS:
+#ifdef XML_NS
+ case BT_COLON:
+#endif
+ tok = XML_TOK_NMTOKEN;
+ ptr += MINBPC(enc);
+ break;
+ case BT_NONASCII:
+ if (IS_NMSTRT_CHAR_MINBPC(enc, ptr)) {
+ ptr += MINBPC(enc);
+ tok = XML_TOK_NAME;
+ break;
+ }
+ if (IS_NAME_CHAR_MINBPC(enc, ptr)) {
+ ptr += MINBPC(enc);
+ tok = XML_TOK_NMTOKEN;
+ break;
+ }
+ /* fall through */
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ while (ptr != end) {
+ switch (BYTE_TYPE(enc, ptr)) {
+ CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+ case BT_GT: case BT_RPAR: case BT_COMMA:
+ case BT_VERBAR: case BT_LSQB: case BT_PERCNT:
+ case BT_S: case BT_CR: case BT_LF:
+ *nextTokPtr = ptr;
+ return tok;
+#ifdef XML_NS
+ case BT_COLON:
+ ptr += MINBPC(enc);
+ switch (tok) {
+ case XML_TOK_NAME:
+ if (ptr == end)
+ return XML_TOK_PARTIAL;
+ tok = XML_TOK_PREFIXED_NAME;
+ switch (BYTE_TYPE(enc, ptr)) {
+ CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+ default:
+ tok = XML_TOK_NMTOKEN;
+ break;
+ }
+ break;
+ case XML_TOK_PREFIXED_NAME:
+ tok = XML_TOK_NMTOKEN;
+ break;
+ }
+ break;
+#endif
+ case BT_PLUS:
+ if (tok == XML_TOK_NMTOKEN) {
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_NAME_PLUS;
+ case BT_AST:
+ if (tok == XML_TOK_NMTOKEN) {
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_NAME_ASTERISK;
+ case BT_QUEST:
+ if (tok == XML_TOK_NMTOKEN) {
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_NAME_QUESTION;
+ default:
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ }
+ }
+ return -tok;
+}
+
+static int PTRCALL
+PREFIX(attributeValueTok)(const ENCODING *enc, const char *ptr,
+ const char *end, const char **nextTokPtr)
+{
+ const char *start;
+ if (ptr == end)
+ return XML_TOK_NONE;
+ start = ptr;
+ while (ptr != end) {
+ switch (BYTE_TYPE(enc, ptr)) {
+#define LEAD_CASE(n) \
+ case BT_LEAD ## n: ptr += n; break;
+ LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
+#undef LEAD_CASE
+ case BT_AMP:
+ if (ptr == start)
+ return PREFIX(scanRef)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+ *nextTokPtr = ptr;
+ return XML_TOK_DATA_CHARS;
+ case BT_LT:
+ /* this is for inside entity references */
+ *nextTokPtr = ptr;
+ return XML_TOK_INVALID;
+ case BT_LF:
+ if (ptr == start) {
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_DATA_NEWLINE;
+ }
+ *nextTokPtr = ptr;
+ return XML_TOK_DATA_CHARS;
+ case BT_CR:
+ if (ptr == start) {
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ return XML_TOK_TRAILING_CR;
+ if (BYTE_TYPE(enc, ptr) == BT_LF)
+ ptr += MINBPC(enc);
+ *nextTokPtr = ptr;
+ return XML_TOK_DATA_NEWLINE;
+ }
+ *nextTokPtr = ptr;
+ return XML_TOK_DATA_CHARS;
+ case BT_S:
+ if (ptr == start) {
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_ATTRIBUTE_VALUE_S;
+ }
+ *nextTokPtr = ptr;
+ return XML_TOK_DATA_CHARS;
+ default:
+ ptr += MINBPC(enc);
+ break;
+ }
+ }
+ *nextTokPtr = ptr;
+ return XML_TOK_DATA_CHARS;
+}
+
+static int PTRCALL
+PREFIX(entityValueTok)(const ENCODING *enc, const char *ptr,
+ const char *end, const char **nextTokPtr)
+{
+ const char *start;
+ if (ptr == end)
+ return XML_TOK_NONE;
+ start = ptr;
+ while (ptr != end) {
+ switch (BYTE_TYPE(enc, ptr)) {
+#define LEAD_CASE(n) \
+ case BT_LEAD ## n: ptr += n; break;
+ LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
+#undef LEAD_CASE
+ case BT_AMP:
+ if (ptr == start)
+ return PREFIX(scanRef)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+ *nextTokPtr = ptr;
+ return XML_TOK_DATA_CHARS;
+ case BT_PERCNT:
+ if (ptr == start) {
+ int tok = PREFIX(scanPercent)(enc, ptr + MINBPC(enc),
+ end, nextTokPtr);
+ return (tok == XML_TOK_PERCENT) ? XML_TOK_INVALID : tok;
+ }
+ *nextTokPtr = ptr;
+ return XML_TOK_DATA_CHARS;
+ case BT_LF:
+ if (ptr == start) {
+ *nextTokPtr = ptr + MINBPC(enc);
+ return XML_TOK_DATA_NEWLINE;
+ }
+ *nextTokPtr = ptr;
+ return XML_TOK_DATA_CHARS;
+ case BT_CR:
+ if (ptr == start) {
+ ptr += MINBPC(enc);
+ if (ptr == end)
+ return XML_TOK_TRAILING_CR;
+ if (BYTE_TYPE(enc, ptr) == BT_LF)
+ ptr += MINBPC(enc);
+ *nextTokPtr = ptr;
+ return XML_TOK_DATA_NEWLINE;
+ }
+ *nextTokPtr = ptr;
+ return XML_TOK_DATA_CHARS;
+ default:
+ ptr += MINBPC(enc);
+ break;
+ }
+ }
+ *nextTokPtr = ptr;
+ return XML_TOK_DATA_CHARS;
+}
+
+#ifdef XML_DTD
+
+static int PTRCALL
+PREFIX(ignoreSectionTok)(const ENCODING *enc, const char *ptr,
+ const char *end, const char **nextTokPtr)
+{
+ int level = 0;
+ if (MINBPC(enc) > 1) {
+ size_t n = end - ptr;
+ if (n & (MINBPC(enc) - 1)) {
+ n &= ~(MINBPC(enc) - 1);
+ end = ptr + n;
+ }
+ }
+ while (ptr != end) {
+ switch (BYTE_TYPE(enc, ptr)) {
+ INVALID_CASES(ptr, nextTokPtr)
+ case BT_LT:
+ if ((ptr += MINBPC(enc)) == end)
+ return XML_TOK_PARTIAL;
+ if (CHAR_MATCHES(enc, ptr, ASCII_EXCL)) {
+ if ((ptr += MINBPC(enc)) == end)
+ return XML_TOK_PARTIAL;
+ if (CHAR_MATCHES(enc, ptr, ASCII_LSQB)) {
+ ++level;
+ ptr += MINBPC(enc);
+ }
+ }
+ break;
+ case BT_RSQB:
+ if ((ptr += MINBPC(enc)) == end)
+ return XML_TOK_PARTIAL;
+ if (CHAR_MATCHES(enc, ptr, ASCII_RSQB)) {
+ if ((ptr += MINBPC(enc)) == end)
+ return XML_TOK_PARTIAL;
+ if (CHAR_MATCHES(enc, ptr, ASCII_GT)) {
+ ptr += MINBPC(enc);
+ if (level == 0) {
+ *nextTokPtr = ptr;
+ return XML_TOK_IGNORE_SECT;
+ }
+ --level;
+ }
+ }
+ break;
+ default:
+ ptr += MINBPC(enc);
+ break;
+ }
+ }
+ return XML_TOK_PARTIAL;
+}
+
+#endif /* XML_DTD */
+
+static int PTRCALL
+PREFIX(isPublicId)(const ENCODING *enc, const char *ptr, const char *end,
+ const char **badPtr)
+{
+ ptr += MINBPC(enc);
+ end -= MINBPC(enc);
+ for (; ptr != end; ptr += MINBPC(enc)) {
+ switch (BYTE_TYPE(enc, ptr)) {
+ case BT_DIGIT:
+ case BT_HEX:
+ case BT_MINUS:
+ case BT_APOS:
+ case BT_LPAR:
+ case BT_RPAR:
+ case BT_PLUS:
+ case BT_COMMA:
+ case BT_SOL:
+ case BT_EQUALS:
+ case BT_QUEST:
+ case BT_CR:
+ case BT_LF:
+ case BT_SEMI:
+ case BT_EXCL:
+ case BT_AST:
+ case BT_PERCNT:
+ case BT_NUM:
+#ifdef XML_NS
+ case BT_COLON:
+#endif
+ break;
+ case BT_S:
+ if (CHAR_MATCHES(enc, ptr, ASCII_TAB)) {
+ *badPtr = ptr;
+ return 0;
+ }
+ break;
+ case BT_NAME:
+ case BT_NMSTRT:
+ if (!(BYTE_TO_ASCII(enc, ptr) & ~0x7f))
+ break;
+ default:
+ switch (BYTE_TO_ASCII(enc, ptr)) {
+ case 0x24: /* $ */
+ case 0x40: /* @ */
+ break;
+ default:
+ *badPtr = ptr;
+ return 0;
+ }
+ break;
+ }
+ }
+ return 1;
+}
+
+/* This must only be called for a well-formed start-tag or empty
+ element tag. Returns the number of attributes. Pointers to the
+ first attsMax attributes are stored in atts.
+*/
+
+static int PTRCALL
+PREFIX(getAtts)(const ENCODING *enc, const char *ptr,
+ int attsMax, ATTRIBUTE *atts)
+{
+ enum { other, inName, inValue } state = inName;
+ int nAtts = 0;
+ int open = 0; /* defined when state == inValue;
+ initialization just to shut up compilers */
+
+ for (ptr += MINBPC(enc);; ptr += MINBPC(enc)) {
+ switch (BYTE_TYPE(enc, ptr)) {
+#define START_NAME \
+ if (state == other) { \
+ if (nAtts < attsMax) { \
+ atts[nAtts].name = ptr; \
+ atts[nAtts].normalized = 1; \
+ } \
+ state = inName; \
+ }
+#define LEAD_CASE(n) \
+ case BT_LEAD ## n: START_NAME ptr += (n - MINBPC(enc)); break;
+ LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
+#undef LEAD_CASE
+ case BT_NONASCII:
+ case BT_NMSTRT:
+ case BT_HEX:
+ START_NAME
+ break;
+#undef START_NAME
+ case BT_QUOT:
+ if (state != inValue) {
+ if (nAtts < attsMax)
+ atts[nAtts].valuePtr = ptr + MINBPC(enc);
+ state = inValue;
+ open = BT_QUOT;
+ }
+ else if (open == BT_QUOT) {
+ state = other;
+ if (nAtts < attsMax)
+ atts[nAtts].valueEnd = ptr;
+ nAtts++;
+ }
+ break;
+ case BT_APOS:
+ if (state != inValue) {
+ if (nAtts < attsMax)
+ atts[nAtts].valuePtr = ptr + MINBPC(enc);
+ state = inValue;
+ open = BT_APOS;
+ }
+ else if (open == BT_APOS) {
+ state = other;
+ if (nAtts < attsMax)
+ atts[nAtts].valueEnd = ptr;
+ nAtts++;
+ }
+ break;
+ case BT_AMP:
+ if (nAtts < attsMax)
+ atts[nAtts].normalized = 0;
+ break;
+ case BT_S:
+ if (state == inName)
+ state = other;
+ else if (state == inValue
+ && nAtts < attsMax
+ && atts[nAtts].normalized
+ && (ptr == atts[nAtts].valuePtr
+ || BYTE_TO_ASCII(enc, ptr) != ASCII_SPACE
+ || BYTE_TO_ASCII(enc, ptr + MINBPC(enc)) == ASCII_SPACE
+ || BYTE_TYPE(enc, ptr + MINBPC(enc)) == open))
+ atts[nAtts].normalized = 0;
+ break;
+ case BT_CR: case BT_LF:
+ /* This case ensures that the first attribute name is counted
+ Apart from that we could just change state on the quote. */
+ if (state == inName)
+ state = other;
+ else if (state == inValue && nAtts < attsMax)
+ atts[nAtts].normalized = 0;
+ break;
+ case BT_GT:
+ case BT_SOL:
+ if (state != inValue)
+ return nAtts;
+ break;
+ default:
+ break;
+ }
+ }
+ /* not reached */
+}
+
+static int PTRFASTCALL
+PREFIX(charRefNumber)(const ENCODING *enc, const char *ptr)
+{
+ int result = 0;
+ /* skip &# */
+ ptr += 2*MINBPC(enc);
+ if (CHAR_MATCHES(enc, ptr, ASCII_x)) {
+ for (ptr += MINBPC(enc);
+ !CHAR_MATCHES(enc, ptr, ASCII_SEMI);
+ ptr += MINBPC(enc)) {
+ int c = BYTE_TO_ASCII(enc, ptr);
+ switch (c) {
+ case ASCII_0: case ASCII_1: case ASCII_2: case ASCII_3: case ASCII_4:
+ case ASCII_5: case ASCII_6: case ASCII_7: case ASCII_8: case ASCII_9:
+ result <<= 4;
+ result |= (c - ASCII_0);
+ break;
+ case ASCII_A: case ASCII_B: case ASCII_C:
+ case ASCII_D: case ASCII_E: case ASCII_F:
+ result <<= 4;
+ result += 10 + (c - ASCII_A);
+ break;
+ case ASCII_a: case ASCII_b: case ASCII_c:
+ case ASCII_d: case ASCII_e: case ASCII_f:
+ result <<= 4;
+ result += 10 + (c - ASCII_a);
+ break;
+ }
+ if (result >= 0x110000)
+ return -1;
+ }
+ }
+ else {
+ for (; !CHAR_MATCHES(enc, ptr, ASCII_SEMI); ptr += MINBPC(enc)) {
+ int c = BYTE_TO_ASCII(enc, ptr);
+ result *= 10;
+ result += (c - ASCII_0);
+ if (result >= 0x110000)
+ return -1;
+ }
+ }
+ return checkCharRefNumber(result);
+}
+
+static int PTRCALL
+PREFIX(predefinedEntityName)(const ENCODING *enc, const char *ptr,
+ const char *end)
+{
+ switch ((end - ptr)/MINBPC(enc)) {
+ case 2:
+ if (CHAR_MATCHES(enc, ptr + MINBPC(enc), ASCII_t)) {
+ switch (BYTE_TO_ASCII(enc, ptr)) {
+ case ASCII_l:
+ return ASCII_LT;
+ case ASCII_g:
+ return ASCII_GT;
+ }
+ }
+ break;
+ case 3:
+ if (CHAR_MATCHES(enc, ptr, ASCII_a)) {
+ ptr += MINBPC(enc);
+ if (CHAR_MATCHES(enc, ptr, ASCII_m)) {
+ ptr += MINBPC(enc);
+ if (CHAR_MATCHES(enc, ptr, ASCII_p))
+ return ASCII_AMP;
+ }
+ }
+ break;
+ case 4:
+ switch (BYTE_TO_ASCII(enc, ptr)) {
+ case ASCII_q:
+ ptr += MINBPC(enc);
+ if (CHAR_MATCHES(enc, ptr, ASCII_u)) {
+ ptr += MINBPC(enc);
+ if (CHAR_MATCHES(enc, ptr, ASCII_o)) {
+ ptr += MINBPC(enc);
+ if (CHAR_MATCHES(enc, ptr, ASCII_t))
+ return ASCII_QUOT;
+ }
+ }
+ break;
+ case ASCII_a:
+ ptr += MINBPC(enc);
+ if (CHAR_MATCHES(enc, ptr, ASCII_p)) {
+ ptr += MINBPC(enc);
+ if (CHAR_MATCHES(enc, ptr, ASCII_o)) {
+ ptr += MINBPC(enc);
+ if (CHAR_MATCHES(enc, ptr, ASCII_s))
+ return ASCII_APOS;
+ }
+ }
+ break;
+ }
+ }
+ return 0;
+}
+
+static int PTRCALL
+PREFIX(sameName)(const ENCODING *enc, const char *ptr1, const char *ptr2)
+{
+ for (;;) {
+ switch (BYTE_TYPE(enc, ptr1)) {
+#define LEAD_CASE(n) \
+ case BT_LEAD ## n: \
+ if (*ptr1++ != *ptr2++) \
+ return 0;
+ LEAD_CASE(4) LEAD_CASE(3) LEAD_CASE(2)
+#undef LEAD_CASE
+ /* fall through */
+ if (*ptr1++ != *ptr2++)
+ return 0;
+ break;
+ case BT_NONASCII:
+ case BT_NMSTRT:
+#ifdef XML_NS
+ case BT_COLON:
+#endif
+ case BT_HEX:
+ case BT_DIGIT:
+ case BT_NAME:
+ case BT_MINUS:
+ if (*ptr2++ != *ptr1++)
+ return 0;
+ if (MINBPC(enc) > 1) {
+ if (*ptr2++ != *ptr1++)
+ return 0;
+ if (MINBPC(enc) > 2) {
+ if (*ptr2++ != *ptr1++)
+ return 0;
+ if (MINBPC(enc) > 3) {
+ if (*ptr2++ != *ptr1++)
+ return 0;
+ }
+ }
+ }
+ break;
+ default:
+ if (MINBPC(enc) == 1 && *ptr1 == *ptr2)
+ return 1;
+ switch (BYTE_TYPE(enc, ptr2)) {
+ case BT_LEAD2:
+ case BT_LEAD3:
+ case BT_LEAD4:
+ case BT_NONASCII:
+ case BT_NMSTRT:
+#ifdef XML_NS
+ case BT_COLON:
+#endif
+ case BT_HEX:
+ case BT_DIGIT:
+ case BT_NAME:
+ case BT_MINUS:
+ return 0;
+ default:
+ return 1;
+ }
+ }
+ }
+ /* not reached */
+}
+
+static int PTRCALL
+PREFIX(nameMatchesAscii)(const ENCODING *enc, const char *ptr1,
+ const char *end1, const char *ptr2)
+{
+ for (; *ptr2; ptr1 += MINBPC(enc), ptr2++) {
+ if (ptr1 == end1)
+ return 0;
+ if (!CHAR_MATCHES(enc, ptr1, *ptr2))
+ return 0;
+ }
+ return ptr1 == end1;
+}
+
+static int PTRFASTCALL
+PREFIX(nameLength)(const ENCODING *enc, const char *ptr)
+{
+ const char *start = ptr;
+ for (;;) {
+ switch (BYTE_TYPE(enc, ptr)) {
+#define LEAD_CASE(n) \
+ case BT_LEAD ## n: ptr += n; break;
+ LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
+#undef LEAD_CASE
+ case BT_NONASCII:
+ case BT_NMSTRT:
+#ifdef XML_NS
+ case BT_COLON:
+#endif
+ case BT_HEX:
+ case BT_DIGIT:
+ case BT_NAME:
+ case BT_MINUS:
+ ptr += MINBPC(enc);
+ break;
+ default:
+ return (int)(ptr - start);
+ }
+ }
+}
+
+static const char * PTRFASTCALL
+PREFIX(skipS)(const ENCODING *enc, const char *ptr)
+{
+ for (;;) {
+ switch (BYTE_TYPE(enc, ptr)) {
+ case BT_LF:
+ case BT_CR:
+ case BT_S:
+ ptr += MINBPC(enc);
+ break;
+ default:
+ return ptr;
+ }
+ }
+}
+
+static void PTRCALL
+PREFIX(updatePosition)(const ENCODING *enc,
+ const char *ptr,
+ const char *end,
+ POSITION *pos)
+{
+ while (ptr != end) {
+ switch (BYTE_TYPE(enc, ptr)) {
+#define LEAD_CASE(n) \
+ case BT_LEAD ## n: \
+ ptr += n; \
+ break;
+ LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
+#undef LEAD_CASE
+ case BT_LF:
+ pos->columnNumber = (XML_Size)-1;
+ pos->lineNumber++;
+ ptr += MINBPC(enc);
+ break;
+ case BT_CR:
+ pos->lineNumber++;
+ ptr += MINBPC(enc);
+ if (ptr != end && BYTE_TYPE(enc, ptr) == BT_LF)
+ ptr += MINBPC(enc);
+ pos->columnNumber = (XML_Size)-1;
+ break;
+ default:
+ ptr += MINBPC(enc);
+ break;
+ }
+ pos->columnNumber++;
+ }
+}
+
+#undef DO_LEAD_CASE
+#undef MULTIBYTE_CASES
+#undef INVALID_CASES
+#undef CHECK_NAME_CASE
+#undef CHECK_NAME_CASES
+#undef CHECK_NMSTRT_CASE
+#undef CHECK_NMSTRT_CASES
+
diff --git a/src/lib/expat/xmltok_impl.h b/src/lib/expat/xmltok_impl.h
new file mode 100644
index 00000000000..da0ea60a657
--- /dev/null
+++ b/src/lib/expat/xmltok_impl.h
@@ -0,0 +1,46 @@
+/*
+Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+See the file COPYING for copying permission.
+*/
+
+enum {
+ BT_NONXML,
+ BT_MALFORM,
+ BT_LT,
+ BT_AMP,
+ BT_RSQB,
+ BT_LEAD2,
+ BT_LEAD3,
+ BT_LEAD4,
+ BT_TRAIL,
+ BT_CR,
+ BT_LF,
+ BT_GT,
+ BT_QUOT,
+ BT_APOS,
+ BT_EQUALS,
+ BT_QUEST,
+ BT_EXCL,
+ BT_SOL,
+ BT_SEMI,
+ BT_NUM,
+ BT_LSQB,
+ BT_S,
+ BT_NMSTRT,
+ BT_COLON,
+ BT_HEX,
+ BT_DIGIT,
+ BT_NAME,
+ BT_MINUS,
+ BT_OTHER, /* known not to be a name or name start character */
+ BT_NONASCII, /* might be a name or name start character */
+ BT_PERCNT,
+ BT_LPAR,
+ BT_RPAR,
+ BT_AST,
+ BT_PLUS,
+ BT_COMMA,
+ BT_VERBAR
+};
+
+#include <stddef.h>
diff --git a/src/lib/expat/xmltok_ns.c b/src/lib/expat/xmltok_ns.c
new file mode 100644
index 00000000000..d2f893836fb
--- /dev/null
+++ b/src/lib/expat/xmltok_ns.c
@@ -0,0 +1,106 @@
+const ENCODING *
+NS(XmlGetUtf8InternalEncoding)(void)
+{
+ return &ns(internal_utf8_encoding).enc;
+}
+
+const ENCODING *
+NS(XmlGetUtf16InternalEncoding)(void)
+{
+#if BYTEORDER == 1234
+ return &ns(internal_little2_encoding).enc;
+#elif BYTEORDER == 4321
+ return &ns(internal_big2_encoding).enc;
+#else
+ const short n = 1;
+ return (*(const char *)&n
+ ? &ns(internal_little2_encoding).enc
+ : &ns(internal_big2_encoding).enc);
+#endif
+}
+
+static const ENCODING * const NS(encodings)[] = {
+ &ns(latin1_encoding).enc,
+ &ns(ascii_encoding).enc,
+ &ns(utf8_encoding).enc,
+ &ns(big2_encoding).enc,
+ &ns(big2_encoding).enc,
+ &ns(little2_encoding).enc,
+ &ns(utf8_encoding).enc /* NO_ENC */
+};
+
+static int PTRCALL
+NS(initScanProlog)(const ENCODING *enc, const char *ptr, const char *end,
+ const char **nextTokPtr)
+{
+ return initScan(NS(encodings), (const INIT_ENCODING *)enc,
+ XML_PROLOG_STATE, ptr, end, nextTokPtr);
+}
+
+static int PTRCALL
+NS(initScanContent)(const ENCODING *enc, const char *ptr, const char *end,
+ const char **nextTokPtr)
+{
+ return initScan(NS(encodings), (const INIT_ENCODING *)enc,
+ XML_CONTENT_STATE, ptr, end, nextTokPtr);
+}
+
+int
+NS(XmlInitEncoding)(INIT_ENCODING *p, const ENCODING **encPtr,
+ const char *name)
+{
+ int i = getEncodingIndex(name);
+ if (i == UNKNOWN_ENC)
+ return 0;
+ SET_INIT_ENC_INDEX(p, i);
+ p->initEnc.scanners[XML_PROLOG_STATE] = NS(initScanProlog);
+ p->initEnc.scanners[XML_CONTENT_STATE] = NS(initScanContent);
+ p->initEnc.updatePosition = initUpdatePosition;
+ p->encPtr = encPtr;
+ *encPtr = &(p->initEnc);
+ return 1;
+}
+
+static const ENCODING *
+NS(findEncoding)(const ENCODING *enc, const char *ptr, const char *end)
+{
+#define ENCODING_MAX 128
+ char buf[ENCODING_MAX];
+ char *p = buf;
+ int i;
+ XmlUtf8Convert(enc, &ptr, end, &p, p + ENCODING_MAX - 1);
+ if (ptr != end)
+ return 0;
+ *p = 0;
+ if (streqci(buf, KW_UTF_16) && enc->minBytesPerChar == 2)
+ return enc;
+ i = getEncodingIndex(buf);
+ if (i == UNKNOWN_ENC)
+ return 0;
+ return NS(encodings)[i];
+}
+
+int
+NS(XmlParseXmlDecl)(int isGeneralTextEntity,
+ const ENCODING *enc,
+ const char *ptr,
+ const char *end,
+ const char **badPtr,
+ const char **versionPtr,
+ const char **versionEndPtr,
+ const char **encodingName,
+ const ENCODING **encoding,
+ int *standalone)
+{
+ return doParseXmlDecl(NS(findEncoding),
+ isGeneralTextEntity,
+ enc,
+ ptr,
+ end,
+ badPtr,
+ versionPtr,
+ versionEndPtr,
+ encodingName,
+ encoding,
+ standalone);
+}
diff --git a/src/lib/lib.mak b/src/lib/lib.mak
new file mode 100644
index 00000000000..224caa63cf3
--- /dev/null
+++ b/src/lib/lib.mak
@@ -0,0 +1,84 @@
+###########################################################################
+#
+# lib.mak
+#
+# MAME dependent library makefile
+#
+# Copyright (c) 1996-2007, Nicola Salmoria and the MAME Team.
+# Visit http://mamedev.org for licensing and usage restrictions.
+#
+###########################################################################
+
+
+LIBOBJ = $(OBJ)/lib
+
+OBJDIRS += \
+ $(LIBOBJ)/util \
+ $(LIBOBJ)/expat \
+ $(LIBOBJ)/zlib \
+
+
+
+#-------------------------------------------------
+# utility library objects
+#-------------------------------------------------
+
+UTILOBJS = \
+ $(LIBOBJ)/util/astring.o \
+ $(LIBOBJ)/util/avcomp.o \
+ $(LIBOBJ)/util/aviio.o \
+ $(LIBOBJ)/util/bitmap.o \
+ $(LIBOBJ)/util/cdrom.o \
+ $(LIBOBJ)/util/chd.o \
+ $(LIBOBJ)/util/corefile.o \
+ $(LIBOBJ)/util/corestr.o \
+ $(LIBOBJ)/util/coreutil.o \
+ $(LIBOBJ)/util/harddisk.o \
+ $(LIBOBJ)/util/huffman.o \
+ $(LIBOBJ)/util/jedparse.o \
+ $(LIBOBJ)/util/md5.o \
+ $(LIBOBJ)/util/options.o \
+ $(LIBOBJ)/util/palette.o \
+ $(LIBOBJ)/util/png.o \
+ $(LIBOBJ)/util/pool.o \
+ $(LIBOBJ)/util/sha1.o \
+ $(LIBOBJ)/util/unicode.o \
+ $(LIBOBJ)/util/unzip.o \
+ $(LIBOBJ)/util/xmlfile.o \
+
+$(OBJ)/libutil.a: $(UTILOBJS)
+
+
+
+#-------------------------------------------------
+# expat library objects
+#-------------------------------------------------
+
+EXPATOBJS = \
+ $(LIBOBJ)/expat/xmlparse.o \
+ $(LIBOBJ)/expat/xmlrole.o \
+ $(LIBOBJ)/expat/xmltok.o
+
+$(OBJ)/libexpat.a: $(EXPATOBJS)
+
+
+
+#-------------------------------------------------
+# zlib library objects
+#-------------------------------------------------
+
+ZLIBOBJS = \
+ $(LIBOBJ)/zlib/adler32.o \
+ $(LIBOBJ)/zlib/compress.o \
+ $(LIBOBJ)/zlib/crc32.o \
+ $(LIBOBJ)/zlib/deflate.o \
+ $(LIBOBJ)/zlib/gzio.o \
+ $(LIBOBJ)/zlib/inffast.o \
+ $(LIBOBJ)/zlib/inflate.o \
+ $(LIBOBJ)/zlib/infback.o \
+ $(LIBOBJ)/zlib/inftrees.o \
+ $(LIBOBJ)/zlib/trees.o \
+ $(LIBOBJ)/zlib/uncompr.o \
+ $(LIBOBJ)/zlib/zutil.o
+
+$(OBJ)/libz.a: $(ZLIBOBJS)
diff --git a/src/lib/util/astring.c b/src/lib/util/astring.c
new file mode 100644
index 00000000000..b2ae0c6c61d
--- /dev/null
+++ b/src/lib/util/astring.c
@@ -0,0 +1,596 @@
+/***************************************************************************
+
+ astring.c
+
+ Allocated string manipulation functions.
+
+ Copyright (c) 1996-2007, Nicola Salmoria and the MAME Team.
+ Visit http://mamedev.org for licensing and usage restrictions.
+
+****************************************************************************/
+
+#include "astring.h"
+#include "osdcore.h"
+
+#include <string.h>
+#include <stdarg.h>
+#include <stdlib.h>
+#include <ctype.h>
+
+
+
+/***************************************************************************
+ TYPE DEFINITIONS
+***************************************************************************/
+
+struct _astring
+{
+ char * text;
+ int alloclen;
+ char smallbuf[256 - sizeof(int) - sizeof(char *)];
+};
+
+
+
+/***************************************************************************
+ GLOBAL VARIABLES
+***************************************************************************/
+
+static char dummy_text[2] = { 0 };
+static astring dummy_astring = { dummy_text, 1, { 0 } };
+
+
+
+/***************************************************************************
+ INLINE FUNCTIONS
+***************************************************************************/
+
+/*-------------------------------------------------
+ ensure_room - ensure we have room for a
+ given string, or else reallocate our buffer
+-------------------------------------------------*/
+
+INLINE int ensure_room(astring *str, int length)
+{
+ char *newbuf, *oldbuf;
+ int alloclen;
+
+ /* always fail to expand the dummy */
+ if (str == &dummy_astring)
+ return FALSE;
+
+ /* if we have the room, do nothing */
+ if (str->alloclen >= length + 1)
+ return TRUE;
+
+ /* allocate a new buffer with some slop */
+ alloclen = length + 256;
+ newbuf = malloc(alloclen);
+ if (newbuf == NULL)
+ return FALSE;
+
+ /* swap in the new buffer and free the old one */
+ oldbuf = (str->text == str->smallbuf) ? NULL : str->text;
+ str->text = strcpy(newbuf, str->text);
+ str->alloclen = alloclen;
+ if (oldbuf != NULL)
+ free(oldbuf);
+
+ return TRUE;
+}
+
+
+/*-------------------------------------------------
+ safe_string_base - return a "safe" string
+ base for a given start index
+-------------------------------------------------*/
+
+INLINE char *safe_string_base(char *base, int start)
+{
+ int max = strlen(base);
+ return (start >= 0 && start < max) ? base + start : base + max;
+}
+
+
+/*-------------------------------------------------
+ normalize_substr - normalize substr parameters
+-------------------------------------------------*/
+
+INLINE void normalize_substr(int *start, int *count, int length)
+{
+ /* limit start */
+ if (*start < 0)
+ *start = 0;
+ else if (*start > length)
+ *start = length;
+
+ /* update count */
+ if (*count == -1 || *start + *count > length)
+ *count = length - *start;
+}
+
+
+
+/***************************************************************************
+ ASTRING ALLOCATION
+***************************************************************************/
+
+/*-------------------------------------------------
+ astring_alloc - allocate a new astring
+-------------------------------------------------*/
+
+astring *astring_alloc(void)
+{
+ astring *str;
+
+ /* allocate memory; if we fail, return the dummy */
+ str = malloc(sizeof(*str));
+ if (str == NULL)
+ return &dummy_astring;
+ memset(str, 0, sizeof(*str));
+
+ /* initialize the small buffer */
+ str->text = str->smallbuf;
+ str->alloclen = ARRAY_LENGTH(str->smallbuf);
+ return str;
+}
+
+
+/*-------------------------------------------------
+ astring_free - free an astring
+-------------------------------------------------*/
+
+void astring_free(astring *str)
+{
+ /* ignore attempts to free the dummy */
+ if (str == &dummy_astring)
+ return;
+
+ /* if we allocated additional memory, free that */
+ if (str->text != str->smallbuf)
+ free(str->text);
+ free(str);
+}
+
+
+
+/***************************************************************************
+ INLINE ASTRING CHANGES
+***************************************************************************/
+
+/*-------------------------------------------------
+ astring_cpy - copy one astring into another
+-------------------------------------------------*/
+
+astring *astring_cpy(astring *dst, const astring *src)
+{
+ return astring_cpyc(dst, src->text);
+}
+
+
+/*-------------------------------------------------
+ astring_cpyc - copy a C string into an astring
+-------------------------------------------------*/
+
+astring *astring_cpyc(astring *dst, const char *src)
+{
+ return astring_cpych(dst, src, strlen(src));
+}
+
+
+/*-------------------------------------------------
+ astring_cpych - copy a character array into
+ an astring
+-------------------------------------------------*/
+
+astring *astring_cpych(astring *dst, const char *src, int count)
+{
+ /* make room; if we fail or if dst is the dummy, do nothing */
+ if (!ensure_room(dst, count))
+ return dst;
+
+ /* copy the raw data and NULL-terminate */
+ if (count > 0)
+ memcpy(dst->text, src, count);
+ dst->text[count] = 0;
+ return dst;
+}
+
+
+/*-------------------------------------------------
+ astring_cpysubstr - copy a substring of one
+ string to another
+-------------------------------------------------*/
+
+astring *astring_cpysubstr(astring *dst, const astring *src, int start, int count)
+{
+ normalize_substr(&start, &count, strlen(src->text));
+ return astring_cpych(dst, src->text + start, count);
+}
+
+
+/*-------------------------------------------------
+ astring_ins - insert one astring into another
+-------------------------------------------------*/
+
+astring *astring_ins(astring *dst, int insbefore, const astring *src)
+{
+ return astring_insc(dst, insbefore, src->text);
+}
+
+
+/*-------------------------------------------------
+ astring_insc - insert a C string into an
+ astring
+-------------------------------------------------*/
+
+astring *astring_insc(astring *dst, int insbefore, const char *src)
+{
+ return astring_insch(dst, insbefore, src, strlen(src));
+}
+
+
+/*-------------------------------------------------
+ astring_insch - insert a character array
+ into an astring
+-------------------------------------------------*/
+
+astring *astring_insch(astring *dst, int insbefore, const char *src, int count)
+{
+ int dstlength = strlen(dst->text);
+
+ /* make room; if we fail or if dst is the dummy, do nothing */
+ if (!ensure_room(dst, dstlength + count))
+ return dst;
+
+ /* adjust insbefore to be logical */
+ if (insbefore < 0 || insbefore > dstlength)
+ insbefore = dstlength;
+
+ /* copy the data an NULL-terminate */
+ if (insbefore < dstlength)
+ memmove(dst->text + insbefore + count, dst->text + insbefore, dstlength - insbefore);
+ memcpy(dst->text + insbefore, src, count);
+ dst->text[dstlength + count] = 0;
+ return dst;
+}
+
+
+/*-------------------------------------------------
+ astring_inssubstr - insert a substring of
+ one string to another
+-------------------------------------------------*/
+
+astring *astring_inssubstr(astring *dst, int insbefore, const astring *src, int start, int count)
+{
+ normalize_substr(&start, &count, strlen(src->text));
+ return astring_insch(dst, insbefore, src->text + start, count);
+}
+
+
+/*-------------------------------------------------
+ astring_substr - extract a substring of
+ ourself, removing everything else
+-------------------------------------------------*/
+
+astring *astring_substr(astring *str, int start, int count)
+{
+ /* ignore attempts to do this on the dummy */
+ if (str == &dummy_astring)
+ return str;
+
+ /* normalize parameters */
+ normalize_substr(&start, &count, strlen(str->text));
+
+ /* move the data and NULL-terminate */
+ if (count > 0 && start > 0)
+ memmove(str->text, str->text + start, count);
+ str->text[count] = 0;
+ return str;
+}
+
+
+/*-------------------------------------------------
+ astring_printf - printf text into an astring
+-------------------------------------------------*/
+
+int astring_printf(astring *dst, const char *format, ...)
+{
+ char tempbuf[4096];
+ va_list args;
+ int result;
+
+ /* sprintf into the temporary buffer */
+ va_start(args, format);
+ result = vsprintf(tempbuf, format, args);
+ va_end(args);
+
+ /* set the result */
+ astring_cpyc(dst, tempbuf);
+ return result;
+}
+
+
+
+/***************************************************************************
+ ASTRING QUERIES
+***************************************************************************/
+
+/*-------------------------------------------------
+ astring_c - return a pointer to a C string
+ from an astring
+-------------------------------------------------*/
+
+const char *astring_c(const astring *str)
+{
+ return str->text;
+}
+
+
+/*-------------------------------------------------
+ astring_len - return the length of an astring
+-------------------------------------------------*/
+
+int astring_len(const astring *str)
+{
+ return strlen(str->text);
+}
+
+
+/*-------------------------------------------------
+ astring_cmp - compare one astring to another
+-------------------------------------------------*/
+
+int astring_cmp(const astring *str1, const astring *str2)
+{
+ return astring_cmpc(str1, str2->text);
+}
+
+
+/*-------------------------------------------------
+ astring_cmpc - compare a C string to an astring
+-------------------------------------------------*/
+
+int astring_cmpc(const astring *str1, const char *str2)
+{
+ const char *s1 = str1->text;
+
+ /* loop while equal until we hit the end of strings */
+ while (*s1 != 0 && *str2 != 0 && *s1 == *str2)
+ s1++, str2++;
+ return *s1 - *str2;
+}
+
+
+/*-------------------------------------------------
+ astring_cmpch - compare a character array to
+ an astring
+-------------------------------------------------*/
+
+int astring_cmpch(const astring *str1, const char *str2, int count)
+{
+ const char *s1 = str1->text;
+ int result;
+
+ /* loop while equal until we hit the end of strings */
+ while (count-- > 0 && *s1 != 0 && *str2 != 0 && *s1 == *str2)
+ s1++, str2++;
+ result = (count == -1) ? 0 : *s1 - *str2;
+ if (result == 0 && *s1 != 0)
+ result = 1;
+ return result;
+}
+
+
+/*-------------------------------------------------
+ astring_cmpsubstr - compare a substring to
+ an astring
+-------------------------------------------------*/
+
+int astring_cmpsubstr(const astring *str1, const astring *str2, int start, int count)
+{
+ normalize_substr(&start, &count, strlen(str2->text));
+ return astring_cmpch(str1, str2->text + start, count);
+}
+
+
+/*-------------------------------------------------
+ astring_icmp - case-insenstive compare one
+ astring to another
+-------------------------------------------------*/
+
+int astring_icmp(const astring *str1, const astring *str2)
+{
+ return astring_icmpc(str1, str2->text);
+}
+
+
+/*-------------------------------------------------
+ astring_icmpc - case-insenstive compare a C
+ string to an astring
+-------------------------------------------------*/
+
+int astring_icmpc(const astring *str1, const char *str2)
+{
+ const char *s1 = str1->text;
+
+ /* loop while equal until we hit the end of strings */
+ while (*s1 != 0 && *str2 != 0 && tolower(*s1) == tolower(*str2))
+ s1++, str2++;
+ return tolower(*s1) - tolower(*str2);
+}
+
+
+/*-------------------------------------------------
+ astring_icmpch - case-insenstive compare a
+ character array to an astring
+-------------------------------------------------*/
+
+int astring_icmpch(const astring *str1, const char *str2, int count)
+{
+ const char *s1 = str1->text;
+ int result;
+
+ /* loop while equal until we hit the end of strings */
+ while (count-- > 0 && *s1 != 0 && *str2 != 0 && tolower(*s1) == tolower(*str2))
+ s1++, str2++;
+ result = (count == -1) ? 0 : tolower(*s1) - tolower(*str2);
+ if (result == 0 && *s1 != 0)
+ result = 1;
+ return result;
+}
+
+
+/*-------------------------------------------------
+ astring_icmpsubstr - case-insenstive compare a
+ substring to an astring
+-------------------------------------------------*/
+
+int astring_icmpsubstr(const astring *str1, const astring *str2, int start, int count)
+{
+ normalize_substr(&start, &count, strlen(str2->text));
+ return astring_icmpch(str1, str2->text + start, count);
+}
+
+
+/*-------------------------------------------------
+ astring_chr - return the index of a character
+ in an astring
+-------------------------------------------------*/
+
+int astring_chr(const astring *str, int start, int ch)
+{
+ char *result = strchr(safe_string_base(str->text, start), ch);
+ return (result != NULL) ? (result - str->text) : -1;
+}
+
+
+/*-------------------------------------------------
+ astring_rchr - return the index of a character
+ in an astring, searching from the end
+-------------------------------------------------*/
+
+int astring_rchr(const astring *str, int start, int ch)
+{
+ char *result = strrchr(safe_string_base(str->text, start), ch);
+ return (result != NULL) ? (result - str->text) : -1;
+}
+
+
+/*-------------------------------------------------
+ astring_find - find one astring in another
+-------------------------------------------------*/
+
+int astring_find(const astring *str, int start, const astring *search)
+{
+ return astring_findc(str, start, search->text);
+}
+
+
+/*-------------------------------------------------
+ astring_findc - find a C string in an astring
+-------------------------------------------------*/
+
+int astring_findc(const astring *str, int start, const char *search)
+{
+ char *result = strstr(safe_string_base(str->text, start), search);
+ return (result != NULL) ? (result - str->text) : -1;
+}
+
+
+
+/***************************************************************************
+ ASTRING UTILITIES
+***************************************************************************/
+
+/*-------------------------------------------------
+ astring_delchr - delete all instances of
+ 'ch'
+-------------------------------------------------*/
+
+astring *astring_delchr(astring *str, int ch)
+{
+ char *src, *dst;
+
+ /* simple deletion */
+ for (src = dst = str->text; *src != 0; src++)
+ if (*src != ch)
+ *dst++ = *src;
+ *dst = 0;
+
+ return str;
+}
+
+
+/*-------------------------------------------------
+ astring_replacechr - replace all instances of
+ 'ch' with 'newch'
+-------------------------------------------------*/
+
+astring *astring_replacechr(astring *str, int ch, int newch)
+{
+ char *text;
+
+ /* simple replacement */
+ for (text = str->text; *text != 0; text++)
+ if (*text == ch)
+ *text = newch;
+
+ return str;
+}
+
+
+/*-------------------------------------------------
+ astring_toupper - convert string to all
+ upper-case
+-------------------------------------------------*/
+
+astring *astring_toupper(astring *str)
+{
+ char *text;
+
+ /* just toupper() on all characters */
+ for (text = str->text; *text != 0; text++)
+ *text = toupper(*text);
+
+ return str;
+}
+
+
+/*-------------------------------------------------
+ astring_tolower - convert string to all
+ lower-case
+-------------------------------------------------*/
+
+astring *astring_tolower(astring *str)
+{
+ char *text;
+
+ /* just toupper() on all characters */
+ for (text = str->text; *text != 0; text++)
+ *text = tolower(*text);
+
+ return str;
+}
+
+
+/*-------------------------------------------------
+ astring_trimspace - remove all space
+ characters from beginning/end
+-------------------------------------------------*/
+
+astring *astring_trimspace(astring *str)
+{
+ char *ptr;
+
+ /* first remove stuff from the end */
+ for (ptr = str->text + strlen(str->text) - 1; ptr >= str->text && isspace(*ptr); ptr--)
+ *ptr = 0;
+
+ /* then count how much to remove from the beginning */
+ for (ptr = str->text; *ptr != 0 && isspace(*ptr); ptr++) ;
+ if (ptr > str->text)
+ astring_substr(str, ptr - str->text, -1);
+
+ return str;
+}
diff --git a/src/lib/util/astring.h b/src/lib/util/astring.h
new file mode 100644
index 00000000000..3d80216383c
--- /dev/null
+++ b/src/lib/util/astring.h
@@ -0,0 +1,228 @@
+/***************************************************************************
+
+ astring.h
+
+ Allocated string manipulation functions.
+
+ Copyright (c) 1996-2007, Nicola Salmoria and the MAME Team.
+ Visit http://mamedev.org for licensing and usage restrictions.
+
+***************************************************************************/
+
+#pragma once
+
+#ifndef __ASTRING_H__
+#define __ASTRING_H__
+
+#include "pool.h"
+
+
+/***************************************************************************
+ TYPE DEFINITIONS
+***************************************************************************/
+
+typedef struct _astring astring;
+
+
+
+/***************************************************************************
+ FUNCTION PROTOTYPES
+***************************************************************************/
+
+
+/* ----- astring allocation ----- */
+
+/* allocate a new astring */
+astring *astring_alloc(void);
+
+/* free an astring */
+void astring_free(astring *str);
+
+
+
+/* ----- inline astring changes ----- */
+
+/* copy one astring into another */
+astring *astring_cpy(astring *dst, const astring *src);
+
+/* copy a C string into an astring */
+astring *astring_cpyc(astring *dst, const char *src);
+
+/* copy a character array into an astring */
+astring *astring_cpych(astring *dst, const char *src, int count);
+
+/* copy a substring of one string into another */
+astring *astring_cpysubstr(astring *dst, const astring *src, int start, int count);
+
+/* insert one astring into another */
+astring *astring_ins(astring *dst, int insbefore, const astring *src);
+
+/* insert a C string into an astring */
+astring *astring_insc(astring *dst, int insbefore, const char *src);
+
+/* insert a character array into an astring */
+astring *astring_insch(astring *dst, int insbefore, const char *src, int count);
+
+/* insert a substring of one string into another */
+astring *astring_inssubstr(astring *dst, int insbefore, const astring *src, int start, int count);
+
+/* extract a substring of ourself, removing everything else */
+astring *astring_substr(astring *str, int start, int count);
+
+/* formatted printf to an astring */
+int astring_printf(astring *dst, const char *format, ...);
+
+
+
+/* ----- astring queries ----- */
+
+/* return a pointer to a C string from an astring */
+const char *astring_c(const astring *str);
+
+/* return the length of an astring */
+int astring_len(const astring *str);
+
+/* compare two astrings */
+int astring_cmp(const astring *str1, const astring *str2);
+
+/* compare an astring to a C string */
+int astring_cmpc(const astring *str1, const char *str2);
+
+/* compare an astring to a character buffer */
+int astring_cmpch(const astring *str1, const char *str2, int count);
+
+/* compare an astring to a character buffer */
+int astring_cmpsubstr(const astring *str1, const astring *str2, int start, int count);
+
+/* case-insenstive compare two astrings */
+int astring_icmp(const astring *str1, const astring *str2);
+
+/* case-insenstive compare an astring to a C string */
+int astring_icmpc(const astring *str1, const char *str2);
+
+/* case-insenstive compare an astring to a character buffer */
+int astring_icmpch(const astring *str1, const char *str2, int count);
+
+/* case-insenstive compare an astring to a character buffer */
+int astring_icmpsubstr(const astring *str1, const astring *str2, int start, int count);
+
+/* search an astring for a character, returning offset or -1 if not found */
+int astring_chr(const astring *str, int start, int ch);
+
+/* reverse search an astring for a character, returning offset or -1 if not found */
+int astring_rchr(const astring *str, int start, int ch);
+
+/* search in an astring for another astring, returning offset or -1 if not found */
+int astring_find(const astring *str, int start, const astring *search);
+
+/* search in an astring for a C string, returning offset or -1 if not found */
+int astring_findc(const astring *str, int start, const char *search);
+
+
+
+/* ----- astring utilties ----- */
+
+/* delete all instances of 'ch' */
+astring *astring_delchr(astring *str, int ch);
+
+/* replace all instances of 'ch' with 'newch' */
+astring *astring_replacechr(astring *str, int ch, int newch);
+
+/* convert string to all upper-case */
+astring *astring_toupper(astring *str);
+
+/* convert string to all lower-case */
+astring *astring_tolower(astring *str);
+
+/* remove all space characters from beginning/end */
+astring *astring_trimspace(astring *str);
+
+
+
+/***************************************************************************
+ INLINE FUNCTIONS
+***************************************************************************/
+
+/* allocate a new duplicate of an astring */
+INLINE astring *astring_dup(const astring *str)
+{
+ return astring_cpy(astring_alloc(), str);
+}
+
+/* allocate a new duplicate of an astring */
+INLINE astring *astring_dupc(const char *str)
+{
+ return astring_cpyc(astring_alloc(), str);
+}
+
+/* allocate a new duplicate of an astring */
+INLINE astring *astring_dupch(const char *str, int count)
+{
+ return astring_cpych(astring_alloc(), str, count);
+}
+
+/* allocate a duplicate of a substring */
+INLINE astring *astring_dupsubstr(const astring *str, int start, int count)
+{
+ return astring_cpysubstr(astring_alloc(), str, start, count);
+}
+
+
+/* reset an astring to an empty string */
+INLINE astring *astring_reset(astring *dst)
+{
+ return astring_cpyc(dst, "");
+}
+
+
+/* concatenate one astring to the end of another */
+INLINE astring *astring_cat(astring *dst, const astring *src)
+{
+ return astring_ins(dst, -1, src);
+}
+
+/* concatenate a C string to the end of an astring */
+INLINE astring *astring_catc(astring *dst, const char *src)
+{
+ return astring_insc(dst, -1, src);
+}
+
+/* concatenate a character array to the end of an astring */
+INLINE astring *astring_catch(astring *dst, const char *src, int count)
+{
+ return astring_insch(dst, -1, src, count);
+}
+
+/* concatenate a substring of one string into another */
+INLINE astring *astring_catsubstr(astring *dst, const astring *src, int start, int count)
+{
+ return astring_inssubstr(dst, -1, src, start, count);
+}
+
+
+/* assemble an astring from 2 C strings */
+INLINE astring *astring_assemble_2(astring *dst, const char *src1, const char *src2)
+{
+ return astring_catc(astring_cpyc(dst, src1), src2);
+}
+
+/* assemble an astring from 3 C strings */
+INLINE astring *astring_assemble_3(astring *dst, const char *src1, const char *src2, const char *src3)
+{
+ return astring_catc(astring_assemble_2(dst, src1, src2), src3);
+}
+
+/* assemble an astring from 4 C strings */
+INLINE astring *astring_assemble_4(astring *dst, const char *src1, const char *src2, const char *src3, const char *src4)
+{
+ return astring_catc(astring_assemble_3(dst, src1, src2, src3), src4);
+}
+
+/* assemble an astring from 5 C strings */
+INLINE astring *astring_assemble_5(astring *dst, const char *src1, const char *src2, const char *src3, const char *src4, const char *src5)
+{
+ return astring_catc(astring_assemble_4(dst, src1, src2, src3, src4), src5);
+}
+
+
+#endif /* __ASTRING_H__ */
diff --git a/src/lib/util/avcomp.c b/src/lib/util/avcomp.c
new file mode 100644
index 00000000000..d3144e01175
--- /dev/null
+++ b/src/lib/util/avcomp.c
@@ -0,0 +1,840 @@
+/*
+To do:
+* verify Cb,Y,Cr,Y ordering
+* swap the ordering(?)
+* add backchannel support for samples as well?
+*/
+/***************************************************************************
+
+ avcomp.c
+
+ Audio/video compression and decompression helpers.
+
+ Copyright (c) 1996-2007, Nicola Salmoria and the MAME Team.
+ Visit http://mamedev.org for licensing and usage restrictions.
+
+****************************************************************************
+
+ Each frame is compressed as a unit. The raw data is of the form:
+ (all multibyte values are stored in big-endian format)
+
+ 'chav' (4 bytes) - fixed header data to identify the format
+ metasize (1 byte) - size of metadata in bytes (max=255 bytes)
+ channels (1 byte) - number of audio channels
+ samples (2 bytes) - number of samples per audio stream
+ width (2 bytes) - width of video data
+ height (2 bytes) - height of video data (high bit set means interlaced)
+ <metadata> - as raw bytes
+ <audio stream 0> - as signed 16-bit samples
+ <audio stream 1> - as signed 16-bit samples
+ ...
+ <video data> - as a raw array of 8-bit YUY data in (Cb,Y,Cr,Y) order
+
+ When compressed, the data is stored as follows:
+ (all multibyte values are stored in big-endian format)
+
+ metasize (1 byte) - size of metadata in bytes
+ channels (1 byte) - number of audio channels
+ samples (2 bytes) - number of samples per audio stream
+ width (2 bytes) - width of video data
+ height (2 bytes) - height of video data (high bit set means interlaced)
+ audhuffsize (1 byte) - size of the audio Huffman table
+ str0size (2 bytes) - compressed size of stream 0
+ str1size (2 bytes) - compressed size of stream 1
+ ...
+ <metadata> - as raw data
+ <audio huffman table> - Huffman table for audio decoding
+ <audio stream 0 data> - Huffman-compressed deltas
+ <audio stream 1 data> - Huffman-compressed deltas
+ <...>
+ <video huffman table> - Huffman table for video decoding
+ <video data> - compressed data
+
+****************************************************************************
+
+ Attempted techniques that have not been worthwhile:
+
+ * Attempted to use integer DCTs from the IJG code; even the "slow"
+ variants produce a lot of error and thus kill our compression ratio,
+ since our compression is based on error not bitrate.
+
+ * Tried various other predictors for the lossless video encoding, but
+ none tended to give any significant gain over predicting the
+ previous pixel.
+
+***************************************************************************/
+
+#include "avcomp.h"
+#include "huffman.h"
+#include "chd.h"
+
+#include <math.h>
+
+
+
+/***************************************************************************
+ TYPE DEFINITIONS
+***************************************************************************/
+
+struct _avcomp_state
+{
+ /* decompression parameters */
+ UINT32 decodemask;
+ UINT8 * videobuffer;
+ UINT32 videostride;
+ UINT32 videoxor;
+ UINT32 audioxor;
+
+ /* video parameters */
+ UINT32 maxwidth, maxheight;
+
+ /* audio parameters */
+ UINT32 maxchannels;
+
+ /* intermediate data */
+ UINT8 * deltadata;
+ UINT8 * audiodata;
+
+ /* huffman contexts */
+ huffman_context * ycontext;
+ huffman_context * ccontext;
+ huffman_context * audiocontext;
+};
+
+
+
+/***************************************************************************
+ PROTOTYPES
+***************************************************************************/
+
+/* encoding helpers */
+static avcomp_error encode_audio(avcomp_state *state, int channels, int samples, const UINT8 *source, UINT8 *dest, UINT8 *sizes);
+static avcomp_error encode_video(avcomp_state *state, int width, int height, int interlaced, const UINT8 *source, UINT8 *dest, UINT32 *complength);
+static avcomp_error encode_video_lossless(avcomp_state *state, int width, int height, int interlaced, const UINT8 *source, UINT8 *dest, UINT32 *complength);
+
+/* decoding helpers */
+static avcomp_error decode_audio(avcomp_state *state, int channels, int samples, const UINT8 *source, UINT8 *dest, const UINT8 *sizes, UINT32 destxor);
+static avcomp_error decode_video(avcomp_state *state, int width, int height, int interlaced, const UINT8 *source, UINT32 complength, UINT8 *dest);
+static avcomp_error decode_video_lossless(avcomp_state *state, int width, int height, int interlaced, const UINT8 *source, UINT32 complength, UINT8 *dest, UINT32 deststride, UINT32 destxor);
+
+
+
+/***************************************************************************
+ IMPLEMENTATION
+***************************************************************************/
+
+/*-------------------------------------------------
+ avcomp_init - allocate and initialize a
+ new state block for compression or
+ decompression
+-------------------------------------------------*/
+
+avcomp_state *avcomp_init(UINT32 maxwidth, UINT32 maxheight, UINT32 maxchannels)
+{
+ huffman_error hufferr;
+ avcomp_state *state;
+
+ /* allocate memory for state block */
+ state = malloc(sizeof(*state));
+ if (state == NULL)
+ return NULL;
+
+ /* clear the buffers */
+ memset(state, 0, sizeof(*state));
+
+ /* fill in sensible decompression defaults */
+ state->decodemask = AVCOMP_DECODE_DEFAULT;
+
+ /* compute the core info */
+ state->maxwidth = maxwidth;
+ state->maxheight = maxheight;
+ state->maxchannels = maxchannels;
+
+ /* now allocate data buffers */
+ state->deltadata = malloc(state->maxwidth * state->maxheight * 2);
+ state->audiodata = malloc(65536 * state->maxchannels * 2);
+ if (state->deltadata == NULL || state->audiodata == NULL)
+ goto cleanup;
+
+ /* create huffman contexts */
+ hufferr = huffman_create_context(&state->ycontext, 12);
+ if (hufferr != HUFFERR_NONE)
+ goto cleanup;
+ hufferr = huffman_create_context(&state->ccontext, 12);
+ if (hufferr != HUFFERR_NONE)
+ goto cleanup;
+ hufferr = huffman_create_context(&state->audiocontext, 12);
+ if (hufferr != HUFFERR_NONE)
+ goto cleanup;
+
+ return state;
+
+cleanup:
+ avcomp_free(state);
+ return NULL;
+}
+
+
+/*-------------------------------------------------
+ avcomp_free - free a state block
+-------------------------------------------------*/
+
+void avcomp_free(avcomp_state *state)
+{
+ /* free the data buffers */
+ if (state->deltadata != NULL)
+ free(state->deltadata);
+ if (state->audiodata != NULL)
+ free(state->audiodata);
+
+ /* free the contexts */
+ if (state->ycontext != NULL)
+ huffman_free_context(state->ycontext);
+ if (state->ccontext != NULL)
+ huffman_free_context(state->ccontext);
+ if (state->audiocontext != NULL)
+ huffman_free_context(state->audiocontext);
+
+ free(state);
+}
+
+
+/*-------------------------------------------------
+ avcomp_decompress_config - configure compression
+ parameters
+-------------------------------------------------*/
+
+void avcomp_decompress_config(avcomp_state *state, UINT32 decodemask, UINT8 *videobuffer, UINT32 videostride, UINT32 videoxor, UINT32 audioxor)
+{
+ state->decodemask = decodemask;
+ state->videobuffer = videobuffer;
+ state->videostride = videostride;
+ state->videoxor = videoxor;
+ state->audioxor = audioxor;
+}
+
+
+
+/***************************************************************************
+ ENCODING/DECODING FRONTENDS
+***************************************************************************/
+
+/*-------------------------------------------------
+ avcomp_encode_data - encode a block of data
+ into a compressed data stream
+-------------------------------------------------*/
+
+avcomp_error avcomp_encode_data(avcomp_state *state, const UINT8 *source, UINT8 *dest, UINT32 *complength)
+{
+ UINT32 metasize, channels, samples, width, height;
+ UINT32 srcoffs, dstoffs;
+ avcomp_error err;
+ int interlaced;
+
+ /* validate the header */
+ if (source[0] != 'c' || source[1] != 'h' || source[2] != 'a' || source[3] != 'v')
+ return AVCERR_INVALID_DATA;
+
+ /* extract info from the header */
+ metasize = source[4];
+ channels = source[5];
+ samples = (source[6] << 8) + source[7];
+ width = (source[8] << 8) + source[9];
+ height = (source[10] << 8) + source[11];
+ interlaced = (height >> 15) & 1;
+ height &= 0x7fff;
+
+ /* validate the info from the header */
+ if (width > state->maxwidth || height > state->maxheight)
+ return AVCERR_VIDEO_TOO_LARGE;
+ if (channels > state->maxchannels)
+ return AVCERR_AUDIO_TOO_LARGE;
+
+ /* write the basics to the new header */
+ dest[0] = metasize;
+ dest[1] = channels;
+ dest[2] = samples >> 8;
+ dest[3] = samples;
+ dest[4] = width >> 8;
+ dest[5] = width;
+ dest[6] = (interlaced << 7) | (height >> 8);
+ dest[7] = height;
+
+ /* starting offsets */
+ srcoffs = 12;
+ dstoffs = 9 + 2 * channels;
+
+ /* copy the metadata first */
+ if (metasize > 0)
+ {
+ memcpy(dest + dstoffs, source + srcoffs, metasize);
+ srcoffs += metasize;
+ dstoffs += metasize;
+ }
+
+ /* encode the audio channels */
+ if (channels > 0)
+ {
+ int chnum;
+
+ /* encode the audio */
+ err = encode_audio(state, channels, samples, source + srcoffs, dest + dstoffs, &dest[8]);
+ if (err != AVCERR_NONE)
+ return err;
+
+ /* advance the pointers past the data */
+ srcoffs += channels * samples * 2;
+ dstoffs += dest[8];
+ for (chnum = 0; chnum < channels; chnum++)
+ dstoffs += (dest[9 + 2 * chnum] << 8) + dest[10 + 2 * chnum];
+ }
+
+ /* encode the video data */
+ if (width > 0 && height > 0)
+ {
+ UINT32 vidlength;
+
+ /* encode the video */
+ err = encode_video(state, width, height, interlaced, source + srcoffs, dest + dstoffs, &vidlength);
+ if (err != AVCERR_NONE)
+ return err;
+
+ /* advance the pointers past the data */
+ srcoffs += width * height * 2;
+ dstoffs += vidlength;
+ }
+
+ /* set the total compression */
+ *complength = dstoffs;
+ return AVCERR_NONE;
+}
+
+
+/*-------------------------------------------------
+ avcomp_decode_data - decode both
+ audio and video from a raw data stream
+-------------------------------------------------*/
+
+avcomp_error avcomp_decode_data(avcomp_state *state, const UINT8 *source, UINT32 complength, UINT8 *dest)
+{
+ UINT32 metasize, channels, samples, width, height;
+ UINT32 srcoffs, dstoffs, totalsize;
+ avcomp_error err;
+ int interlaced;
+ int chnum;
+
+ /* extract info from the header */
+ if (complength < 8)
+ return AVCERR_INVALID_DATA;
+ metasize = source[0];
+ channels = source[1];
+ samples = (source[2] << 8) + source[3];
+ width = (source[4] << 8) + source[5];
+ height = (source[6] << 8) + source[7];
+ interlaced = height >> 15;
+ height &= 0x7fff;
+
+ /* validate the info from the header */
+ if (width > state->maxwidth || height > state->maxheight)
+ return AVCERR_VIDEO_TOO_LARGE;
+ if (channels > state->maxchannels)
+ return AVCERR_AUDIO_TOO_LARGE;
+
+ /* validate that the sizes make sense */
+ if (complength < 9 + 2 * channels)
+ return AVCERR_INVALID_DATA;
+ totalsize = 9 + 2 * channels + source[8];
+ for (chnum = 0; chnum < channels; chnum++)
+ totalsize += (source[9 + 2 * chnum] << 8) | source[10 + 2 * chnum];
+ if (totalsize >= complength)
+ return AVCERR_INVALID_DATA;
+
+ /* starting offsets */
+ srcoffs = 9 + 2 * channels;
+ dstoffs = 12;
+
+ /* write the basics to the new header */
+ dest[0] = 'c';
+ dest[1] = 'h';
+ dest[2] = 'a';
+ dest[3] = 'v';
+ dest[4] = metasize;
+ dest[5] = channels;
+ dest[6] = samples >> 8;
+ dest[7] = samples;
+ dest[8] = width >> 8;
+ dest[9] = width;
+ dest[10] = (interlaced << 7) | (height >> 8);
+ dest[11] = height;
+
+ /* copy the metadata first */
+ if (metasize > 0)
+ {
+ if ((state->decodemask & AVCOMP_DECODE_META) != 0)
+ memcpy(dest + dstoffs, source + srcoffs, metasize);
+ srcoffs += metasize;
+ dstoffs += metasize;
+ }
+
+ /* decode the audio channels */
+ if (channels > 0)
+ {
+ /* decode the audio */
+ err = decode_audio(state, channels, samples, source + srcoffs, dest + dstoffs, &source[8], state->audioxor);
+ if (err != AVCERR_NONE)
+ return err;
+
+ /* advance the pointers past the data */
+ srcoffs += source[8];
+ for (chnum = 0; chnum < channels; chnum++)
+ srcoffs += (source[9 + 2 * chnum] << 8) + source[10 + 2 * chnum];
+ dstoffs += channels * samples * 2;
+ }
+
+ /* decode the video data */
+ if (width > 0 && height > 0 && (state->decodemask & AVCOMP_DECODE_VIDEO) != 0)
+ {
+ /* decode the video */
+ err = decode_video(state, width, height, interlaced, source + srcoffs, complength - srcoffs, dest + dstoffs);
+ if (err != AVCERR_NONE)
+ return err;
+ }
+ return AVCERR_NONE;
+}
+
+
+
+/***************************************************************************
+ ENCODING HELPERS
+***************************************************************************/
+
+/*-------------------------------------------------
+ encode_audio - encode raw audio data
+ to the destination
+-------------------------------------------------*/
+
+static avcomp_error encode_audio(avcomp_state *state, int channels, int samples, const UINT8 *source, UINT8 *dest, UINT8 *sizes)
+{
+ UINT32 size, bytesleft;
+ huffman_error hufferr;
+ int chnum, sampnum;
+ const UINT8 *input;
+ UINT8 *output;
+
+ /* get input/output pointers */
+ input = source;
+ output = state->audiodata;
+
+ /* extract audio deltas */
+ for (chnum = 0; chnum < channels; chnum++)
+ {
+ INT16 prev = 0;
+
+ for (sampnum = 0; sampnum < samples; sampnum++)
+ {
+ INT16 cursample, delta;
+
+ /* read current sample (big endian) */
+ cursample = *input++ << 8;
+ cursample |= *input++;
+
+ /* compute delta against previous sample in this channel */
+ delta = cursample - prev;
+ prev = cursample;
+
+ /* store the delta (big endian) */
+ *output++ = delta >> 8;
+ *output++ = delta & 0xff;
+ }
+ }
+
+ /* reset our output pointer and huffman encode */
+ input = state->audiodata;
+ output = dest;
+ bytesleft = channels * samples * 2;
+
+ /* compute a single huffman tree for all the audio data */
+ hufferr = huffman_compute_tree(state->audiocontext, input, bytesleft, 1);
+ if (hufferr != HUFFERR_NONE)
+ return AVCERR_COMPRESSION_ERROR;
+ hufferr = huffman_export_tree(state->audiocontext, output, bytesleft, &size);
+ if (hufferr != HUFFERR_NONE)
+ return AVCERR_COMPRESSION_ERROR;
+ output += size;
+ *sizes++ = size;
+ bytesleft -= size;
+
+ /* encode each stream separately */
+ for (chnum = 0; chnum < channels; chnum++)
+ {
+ /* encode the data */
+ hufferr = huffman_encode_data(state->audiocontext, input, samples * 2, output, bytesleft, &size);
+ if (hufferr != HUFFERR_NONE)
+ return AVCERR_COMPRESSION_ERROR;
+ input += samples * 2;
+ output += size;
+ bytesleft = (bytesleft > size) ? (bytesleft - size) : 0;
+
+ /* store the size of this stream */
+ *sizes++ = size >> 8;
+ *sizes++ = size;
+ }
+
+ /* if we didn't compress enough, set the huffman table size to 0 and memcpy the data */
+ if (bytesleft == 0)
+ {
+ /* back up to the beginning */
+ input = source;
+ output = dest;
+ sizes -= 1 + 2 * channels;
+
+ /* huffman table size of 0 */
+ *sizes++ = 0;
+
+ /* write the full size of each stream and then copy it */
+ size = samples * 2;
+ for (chnum = 0; chnum < channels; chnum++)
+ {
+ /* copy the data */
+ memcpy(output, input, size);
+ input += size;
+ output += size;
+
+ /* store the size of this stream */
+ *sizes++ = size >> 8;
+ *sizes++ = size;
+ }
+ }
+
+ return AVCERR_NONE;
+}
+
+
+/*-------------------------------------------------
+ encode_video - encode raw video data
+ to the destination
+-------------------------------------------------*/
+
+static avcomp_error encode_video(avcomp_state *state, int width, int height, int interlaced, const UINT8 *source, UINT8 *dest, UINT32 *complength)
+{
+ /* only lossless supported at this time */
+ return encode_video_lossless(state, width, height, interlaced, source, dest, complength);
+}
+
+
+/*-------------------------------------------------
+ encode_video_lossless - do a lossless video
+ encoding using deltas and huffman encoding
+-------------------------------------------------*/
+
+static avcomp_error encode_video_lossless(avcomp_state *state, int width, int height, int interlaced, const UINT8 *source, UINT8 *dest, UINT32 *complength)
+{
+ UINT32 srcbytes = width * height * 2;
+ int rowbytes = width * 2;
+ huffman_error hufferr;
+ UINT32 outbytes;
+ UINT8 *output;
+ int x, y;
+
+ /* loop over rows */
+ output = state->deltadata;
+ for (y = 0; y < height; y++)
+ {
+ const UINT8 *src = &source[rowbytes * y];
+ UINT8 lasty = (y == 0) ? 0x80 : src[-rowbytes + 1];
+ UINT8 lastcb = (y == 0) ? 0x80 : src[-rowbytes + 0];
+ UINT8 lastcr = (y == 0) ? 0x80 : src[-rowbytes + 2];
+
+ /* loop over columns */
+ for (x = 0; x < rowbytes; x += 4)
+ {
+ *output++ = *src - lastcb;
+ lastcb = *src++;
+ *output++ = *src - lasty;
+ lasty = *src++;
+ *output++ = *src - lastcr;
+ lastcr = *src++;
+ *output++ = *src - lasty;
+ lasty = *src++;
+ }
+ }
+
+ /* set up the output; first byte is 0x80 to indicate lossless encoding */
+ output = dest;
+ *output++ = 0x80;
+
+ /* now encode to the destination using two trees, one for the Y and one for the Cr/Cb */
+
+ /* compute the histograms for the data */
+ hufferr = huffman_compute_tree(state->ycontext, state->deltadata + 0, srcbytes, 2);
+ if (hufferr != HUFFERR_NONE)
+ return AVCERR_COMPRESSION_ERROR;
+ hufferr = huffman_compute_tree(state->ccontext, state->deltadata + 1, srcbytes, 2);
+ if (hufferr != HUFFERR_NONE)
+ return AVCERR_COMPRESSION_ERROR;
+
+ /* export the two trees to the data stream */
+ hufferr = huffman_export_tree(state->ycontext, output, 256, &outbytes);
+ if (hufferr != HUFFERR_NONE)
+ return AVCERR_COMPRESSION_ERROR;
+ output += outbytes;
+ hufferr = huffman_export_tree(state->ccontext, output, 256, &outbytes);
+ if (hufferr != HUFFERR_NONE)
+ return AVCERR_COMPRESSION_ERROR;
+ output += outbytes;
+
+ /* encode the data using the two trees */
+ hufferr = huffman_encode_data_interleaved_2(state->ycontext, state->ccontext, state->deltadata, srcbytes, output, srcbytes, &outbytes);
+ if (hufferr != HUFFERR_NONE)
+ return AVCERR_COMPRESSION_ERROR;
+ output += outbytes;
+
+ /* set the final length */
+ *complength = output - dest;
+ return AVCERR_NONE;
+}
+
+
+
+/***************************************************************************
+ DECODING HELPERS
+***************************************************************************/
+
+/*-------------------------------------------------
+ decode_audio - decode audio from a
+ compressed data stream
+-------------------------------------------------*/
+
+static avcomp_error decode_audio(avcomp_state *state, int channels, int samples, const UINT8 *source, UINT8 *dest, const UINT8 *sizes, UINT32 destxor)
+{
+ huffman_error hufferr;
+ int chnum, sampnum;
+ const UINT8 *input;
+ UINT32 actsize;
+ UINT8 *output;
+ UINT16 size;
+
+ /* set up input/output pointers */
+ input = source;
+ output = dest;
+
+ /* if no huffman length, just copy the data */
+ size = *sizes++;
+ if (size == 0)
+ {
+ /* loop over channels */
+ for (chnum = 0; chnum < channels; chnum++)
+ {
+ /* extract the size of this channel */
+ size = *sizes++ << 8;
+ size |= *sizes++;
+
+ /* copy the data */
+ if ((state->decodemask & AVCOMP_DECODE_AUDIO(chnum)) != 0)
+ {
+ for (sampnum = 0; sampnum < samples; sampnum++)
+ {
+ output[0 ^ destxor] = input[0];
+ output[1 ^ destxor] = input[1];
+ output += 2;
+ input += 2;
+ }
+ }
+ else
+ {
+ input += size;
+ output += size;
+ }
+ }
+ return AVCERR_NONE;
+ }
+
+ /* extract the huffman tree */
+ input = source;
+ hufferr = huffman_import_tree(state->audiocontext, input, size, &actsize);
+ if (hufferr != HUFFERR_NONE || actsize != size)
+ return AVCERR_INVALID_DATA;
+ input += actsize;
+
+ /* now loop over channels and decode their data */
+ for (chnum = 0; chnum < channels; chnum++)
+ {
+ /* extract the size of this channel */
+ size = *sizes++ << 8;
+ size |= *sizes++;
+
+ /* decode the data */
+ if ((state->decodemask & AVCOMP_DECODE_AUDIO(chnum)) != 0)
+ {
+ hufferr = huffman_decode_data(state->audiocontext, input, size, output, samples * 2, &actsize);
+ if (hufferr != HUFFERR_NONE || actsize != size)
+ return AVCERR_INVALID_DATA;
+ }
+
+ /* advance */
+ input += size;
+ output += samples * 2;
+ }
+
+ /* reset input/output pointers */
+ output = dest;
+
+ /* reassemble audio from the deltas */
+ for (chnum = 0; chnum < channels; chnum++)
+ {
+ INT16 prev = 0;
+
+ if ((state->decodemask & AVCOMP_DECODE_AUDIO(chnum)) != 0)
+ {
+ for (sampnum = 0; sampnum < samples; sampnum++)
+ {
+ INT16 cursample;
+
+ /* read current sample (big endian) */
+ cursample = output[0] << 8;
+ cursample |= output[1];
+
+ /* compute delta against previous sample in this channel */
+ cursample += prev;
+ prev = cursample;
+
+ /* store the delta (big endian) */
+ output[0 ^ destxor] = cursample >> 8;
+ output[1 ^ destxor] = cursample;
+ output += 2;
+ }
+ }
+ else
+ output += samples * 2;
+ }
+ return AVCERR_NONE;
+}
+
+
+/*-------------------------------------------------
+ decode_video - decode video from a
+ compressed data stream
+-------------------------------------------------*/
+
+static avcomp_error decode_video(avcomp_state *state, int width, int height, int interlaced, const UINT8 *source, UINT32 complength, UINT8 *dest)
+{
+ UINT32 targetstride, targetxor;
+ UINT8 *target;
+
+ /* if we have a target buffer, use that instead */
+ if (state->videobuffer != NULL)
+ {
+ target = state->videobuffer;
+ targetstride = state->videostride;
+ targetxor = state->videoxor;
+ }
+ else
+ {
+ target = dest;
+ targetstride = width * 2;
+ targetxor = 0;
+ }
+
+ /* if the high bit of the first byte is set, we decode losslessly */
+ if (source[0] & 0x80)
+ return decode_video_lossless(state, width, height, interlaced, source, complength, target, targetstride, targetxor);
+ else
+ return AVCERR_INVALID_DATA;
+}
+
+
+/*-------------------------------------------------
+ decode_video_lossless - do a lossless video
+ decoding using deltas and huffman encoding
+-------------------------------------------------*/
+
+static avcomp_error decode_video_lossless(avcomp_state *state, int width, int height, int interlaced, const UINT8 *source, UINT32 complength, UINT8 *dest, UINT32 deststride, UINT32 destxor)
+{
+ const UINT8 *sourceend = source + complength;
+ const huffman_lookup_value *table1, *table2;
+ huffman_error hufferr;
+ UINT32 bitbuf = 0;
+ UINT32 actsize;
+ int sbits = 0;
+ int x, y;
+
+ /* skip the first byte */
+ source++;
+
+ /* decode the table and data */
+ hufferr = huffman_import_tree(state->ycontext, source, sourceend - source, &actsize);
+ if (hufferr != HUFFERR_NONE)
+ return AVCERR_INVALID_DATA;
+ source += actsize;
+ hufferr = huffman_import_tree(state->ccontext, source, sourceend - source, &actsize);
+ if (hufferr != HUFFERR_NONE)
+ return AVCERR_INVALID_DATA;
+ source += actsize;
+
+ /* get the lookup tables */
+ hufferr = huffman_get_lookup_table(state->ycontext, &table1);
+ if (hufferr != HUFFERR_NONE)
+ return AVCERR_OUT_OF_MEMORY;
+ hufferr = huffman_get_lookup_table(state->ccontext, &table2);
+ if (hufferr != HUFFERR_NONE)
+ return AVCERR_OUT_OF_MEMORY;
+
+ /* loop over rows */
+ for (y = 0; y < height; y++)
+ {
+ UINT8 *dst = &dest[y * deststride];
+ UINT8 lasty = (y == 0) ? 0x80 : (dst - deststride)[1 ^ destxor];
+ UINT8 lastcb = (y == 0) ? 0x80 : (dst - deststride)[0 ^ destxor];
+ UINT8 lastcr = (y == 0) ? 0x80 : (dst - deststride)[2 ^ destxor];
+
+ /* loop over columns */
+ for (x = 0; x < width * 2; x += 4)
+ {
+ huffman_lookup_value lookup;
+
+ /* keep the buffer full */
+ while (sbits <= 24)
+ {
+ bitbuf |= *source++ << (24 - sbits);
+ sbits += 8;
+ }
+
+ /* do the Cb component */
+ lookup = table1[bitbuf >> 20];
+ dst[(x + 0) ^ destxor] = lastcb += lookup >> 8;
+ lookup &= 0x1f;
+ bitbuf <<= lookup;
+ sbits -= lookup;
+
+ /* do the Y component */
+ lookup = table2[bitbuf >> 20];
+ dst[(x + 1) ^ destxor] = lasty += lookup >> 8;
+ lookup &= 0x1f;
+ bitbuf <<= lookup;
+ sbits -= lookup;
+
+ /* keep the buffer full */
+ while (sbits <= 24)
+ {
+ bitbuf |= *source++ << (24 - sbits);
+ sbits += 8;
+ }
+
+ /* do the Cr component */
+ lookup = table1[bitbuf >> 20];
+ dst[(x + 2) ^ destxor] = lastcr += lookup >> 8;
+ lookup &= 0x1f;
+ bitbuf <<= lookup;
+ sbits -= lookup;
+
+ /* do the Y component */
+ lookup = table2[bitbuf >> 20];
+ dst[(x + 3) ^ destxor] = lasty += lookup >> 8;
+ lookup &= 0x1f;
+ bitbuf <<= lookup;
+ sbits -= lookup;
+ }
+ }
+
+ return AVCERR_NONE;
+}
diff --git a/src/lib/util/avcomp.h b/src/lib/util/avcomp.h
new file mode 100644
index 00000000000..9a729e3c1c1
--- /dev/null
+++ b/src/lib/util/avcomp.h
@@ -0,0 +1,60 @@
+/***************************************************************************
+
+ avcomp.h
+
+ Audio/video compression and decompression helpers.
+
+ Copyright (c) 1996-2007, Nicola Salmoria and the MAME Team.
+ Visit http://mamedev.org for licensing and usage restrictions.
+
+***************************************************************************/
+
+#ifndef __AVCOMP_H__
+
+#include "osdcore.h"
+
+
+/***************************************************************************
+ CONSTANTS
+***************************************************************************/
+
+/* errors */
+enum _avcomp_error
+{
+ AVCERR_NONE = 0,
+ AVCERR_INVALID_DATA,
+ AVCERR_VIDEO_TOO_LARGE,
+ AVCERR_AUDIO_TOO_LARGE,
+ AVCERR_OUT_OF_MEMORY,
+ AVCERR_COMPRESSION_ERROR
+};
+typedef enum _avcomp_error avcomp_error;
+
+/* default decompression parameters */
+#define AVCOMP_DECODE_META (1 << 0)
+#define AVCOMP_DECODE_VIDEO (1 << 1)
+#define AVCOMP_DECODE_AUDIO(x) (1 << (2 + (x)))
+#define AVCOMP_DECODE_DEFAULT (~0)
+
+
+
+/***************************************************************************
+ TYPE DEFINITIONS
+***************************************************************************/
+
+typedef struct _avcomp_state avcomp_state;
+
+
+
+/***************************************************************************
+ PROTOTYPES
+***************************************************************************/
+
+avcomp_state *avcomp_init(UINT32 maxwidth, UINT32 maxheight, UINT32 maxchannels);
+void avcomp_free(avcomp_state *state);
+void avcomp_decompress_config(avcomp_state *state, UINT32 decodemask, UINT8 *videobuffer, UINT32 videostride, UINT32 videoxor, UINT32 audioxor);
+
+avcomp_error avcomp_encode_data(avcomp_state *state, const UINT8 *source, UINT8 *dest, UINT32 *complength);
+avcomp_error avcomp_decode_data(avcomp_state *state, const UINT8 *source, UINT32 complength, UINT8 *dest);
+
+#endif
diff --git a/src/lib/util/aviio.c b/src/lib/util/aviio.c
new file mode 100644
index 00000000000..5fa464dcdb4
--- /dev/null
+++ b/src/lib/util/aviio.c
@@ -0,0 +1,2744 @@
+/***************************************************************************
+
+ aviio.c
+
+ AVI movie format parsing helpers.
+
+ Copyright (c) 1996-2007, Nicola Salmoria and the MAME Team.
+ Visit http://mamedev.org for licensing and usage restrictions.
+
+***************************************************************************/
+
+#include "aviio.h"
+
+
+/***************************************************************************
+ CONSTANTS
+***************************************************************************/
+
+#define FILETYPE_READ 1
+#define FILETYPE_CREATE 2
+
+#define MAX_RIFF_SIZE (2UL * 1024 * 1024 * 1024 - 1024) /* just under 2GB */
+#define MAX_AVI_SIZE_IN_GB (256)
+#define FOUR_GB ((UINT64)1 << 32)
+
+#define MAX_SOUND_CHANNELS 2
+#define SOUND_BUFFER_MSEC 2000 /* microseconds of sond buffering */
+
+#define CHUNKTYPE_RIFF AVI_FOURCC('R','I','F','F')
+#define CHUNKTYPE_LIST AVI_FOURCC('L','I','S','T')
+#define CHUNKTYPE_JUNK AVI_FOURCC('J','U','N','K')
+#define CHUNKTYPE_AVIH AVI_FOURCC('a','v','i','h')
+#define CHUNKTYPE_STRH AVI_FOURCC('s','t','r','h')
+#define CHUNKTYPE_STRF AVI_FOURCC('s','t','r','f')
+#define CHUNKTYPE_IDX1 AVI_FOURCC('i','d','x','1')
+#define CHUNKTYPE_INDX AVI_FOURCC('i','n','d','x')
+#define CHUNKTYPE_XXDB AVI_FOURCC(0x00,0x00,'d','b')
+#define CHUNKTYPE_XXDC AVI_FOURCC(0x00,0x00,'d','c')
+#define CHUNKTYPE_XXWB AVI_FOURCC(0x00,0x00,'w','b')
+#define CHUNKTYPE_IXXX AVI_FOURCC('i','x',0x00,0x00)
+#define CHUNKTYPE_XX_MASK AVI_FOURCC(0x00,0x00,0xff,0xff)
+
+#define LISTTYPE_AVI AVI_FOURCC('A','V','I',' ')
+#define LISTTYPE_AVIX AVI_FOURCC('A','V','I','X')
+#define LISTTYPE_HDRL AVI_FOURCC('h','d','r','l')
+#define LISTTYPE_STRL AVI_FOURCC('s','t','r','l')
+#define LISTTYPE_MOVI AVI_FOURCC('m','o','v','i')
+
+#define STREAMTYPE_VIDS AVI_FOURCC('v','i','d','s')
+#define STREAMTYPE_AUDS AVI_FOURCC('a','u','d','s')
+
+#define HANDLER_DIB AVI_FOURCC('D','I','B',' ')
+#define HANDLER_HFYU AVI_FOURCC('h','f','y','u')
+
+#define FORMAT_UYVY AVI_FOURCC('U','Y','V','Y')
+#define FORMAT_VYUY AVI_FOURCC('V','Y','U','Y')
+#define FORMAT_YUY2 AVI_FOURCC('Y','U','Y','2')
+#define FORMAT_HFYU AVI_FOURCC('H','F','Y','U')
+
+/* main AVI header files */
+#define AVIF_HASINDEX 0x00000010
+#define AVIF_MUSTUSEINDEX 0x00000020
+#define AVIF_ISINTERLEAVED 0x00000100
+#define AVIF_COPYRIGHTED 0x00010000
+#define AVIF_WASCAPTUREFILE 0x00020000
+
+/* index definitions */
+#define AVI_INDEX_OF_INDEXES 0x00
+#define AVI_INDEX_OF_CHUNKS 0x01
+#define AVI_INDEX_IS_DATA 0x80
+#define AVI_INDEX_2FIELD 0x01
+
+/* HuffYUV definitions */
+#define HUFFYUV_PREDICT_LEFT 0
+#define HUFFYUV_PREDICT_GRADIENT 1
+#define HUFFYUV_PREDICT_MEDIAN 2
+#define HUFFYUV_PREDICT_DECORR 0x40
+
+
+
+/***************************************************************************
+ TYPE DEFINITIONS
+***************************************************************************/
+
+typedef struct _avi_chunk avi_chunk;
+struct _avi_chunk
+{
+ UINT64 offset; /* file offset of chunk header */
+ UINT64 size; /* size of this chunk */
+ UINT32 type; /* type of this chunk */
+ UINT32 listtype; /* type of this list (if we are a list) */
+};
+
+
+typedef struct _avi_chunk_list avi_chunk_list;
+struct _avi_chunk_list
+{
+ UINT64 offset; /* offset in the file of header */
+ UINT32 length; /* length of the chunk including header */
+};
+
+
+typedef struct _huffyuv_table huffyuv_table;
+struct _huffyuv_table
+{
+ UINT8 shift[256]; /* bit shift amounts */
+ UINT32 bits[256]; /* bit match values */
+ UINT32 mask[256]; /* bit mask values */
+ UINT16 baselookup[65536]; /* base lookup table */
+ UINT16 * extralookup; /* extra lookup tables */
+};
+
+
+typedef struct _huffyuv_data huffyuv_data;
+struct _huffyuv_data
+{
+ UINT8 predictor; /* predictor */
+ huffyuv_table table[3]; /* array of tables */
+};
+
+
+typedef struct _avi_stream avi_stream;
+struct _avi_stream
+{
+ UINT32 type; /* subtype of stream */
+ UINT32 format; /* format of stream data */
+
+ UINT32 rate; /* timescale for stream */
+ UINT32 scale; /* duration of one sample in the stream */
+ UINT32 samples; /* number of samples */
+
+ avi_chunk_list * chunk; /* list of chunks */
+ UINT32 chunks; /* chunks currently known */
+ UINT32 chunksalloc; /* number of chunks allocated */
+
+ UINT32 width; /* width of video */
+ UINT32 height; /* height of video */
+ UINT32 depth; /* depth of video */
+ UINT8 interlace; /* interlace parameters */
+ huffyuv_data * huffyuv; /* huffyuv decompression data */
+
+ UINT16 channels; /* audio channels */
+ UINT16 samplebits; /* audio bits per sample */
+ UINT32 samplerate; /* audio sample rate */
+
+ /* only used when creating */
+ UINT64 saved_strh_offset; /* writeoffset of strh chunk */
+ UINT64 saved_indx_offset; /* writeoffset of indx chunk */
+};
+
+
+struct _avi_file
+{
+ /* shared data */
+ osd_file * file; /* pointer to open file */
+ int type; /* type of access (read/create) */
+ avi_movie_info info; /* movie info structure */
+ UINT8 * tempbuffer; /* temporary buffer */
+ UINT32 tempbuffersize; /* size of the temporary buffer */
+
+ /* only used when reading */
+ int streams; /* number of streams */
+ avi_stream * stream; /* allocated array of stream information */
+ avi_chunk rootchunk; /* dummy root chunk that wraps the whole file */
+
+ /* only used when creating */
+ UINT64 writeoffs; /* current file write offset */
+ UINT64 riffbase; /* base of the current RIFF */
+
+ avi_chunk chunkstack[8]; /* stack of chunks we are writing */
+ int chunksp; /* stack pointer for the current chunk */
+
+ UINT64 saved_movi_offset; /* writeoffset of movi list */
+ UINT64 saved_avih_offset; /* writeoffset of avih chunk */
+
+ INT16 * soundbuf; /* buffer for sound data */
+ UINT32 soundbuf_samples; /* length of sound buffer in samples */
+ UINT32 soundbuf_chansamples[MAX_SOUND_CHANNELS]; /* samples in buffer for each channel */
+ UINT32 soundbuf_chunks; /* number of chunks completed so far */
+ UINT32 soundbuf_frames; /* number of frames ahead of the video */
+};
+
+
+
+/***************************************************************************
+ PROTOTYPES
+***************************************************************************/
+
+/* stream helpers */
+
+/* core chunk read routines */
+static avi_error get_first_chunk(avi_file *file, const avi_chunk *parent, avi_chunk *newchunk);
+static avi_error get_next_chunk(avi_file *file, const avi_chunk *parent, avi_chunk *newchunk);
+static avi_error find_first_chunk(avi_file *file, UINT32 findme, const avi_chunk *container, avi_chunk *result);
+static avi_error find_next_chunk(avi_file *file, UINT32 findme, const avi_chunk *container, avi_chunk *result);
+static avi_error get_next_chunk_internal(avi_file *file, const avi_chunk *parent, avi_chunk *newchunk, UINT64 offset);
+static avi_error read_chunk_data(avi_file *file, const avi_chunk *chunk, UINT8 **buffer);
+
+/* chunk read helpers */
+static avi_error read_movie_data(avi_file *file);
+static avi_error extract_movie_info(avi_file *file);
+static avi_error parse_avih_chunk(avi_file *file, avi_chunk *avih);
+static avi_error parse_strh_chunk(avi_file *file, avi_stream *stream, avi_chunk *strh);
+static avi_error parse_strf_chunk(avi_file *file, avi_stream *stream, avi_chunk *strf);
+static avi_error parse_indx_chunk(avi_file *file, avi_stream *stream, avi_chunk *strf);
+static avi_error parse_idx1_chunk(avi_file *file, UINT64 baseoffset, avi_chunk *idx1);
+
+/* core chunk write routines */
+static avi_error chunk_open(avi_file *file, UINT32 type, UINT32 listtype, UINT32 estlength);
+static avi_error chunk_close(avi_file *file);
+static avi_error chunk_write(avi_file *file, UINT32 type, const void *data, UINT32 length);
+static avi_error chunk_overwrite(avi_file *file, UINT32 type, const void *data, UINT32 length, UINT64 *offset, int initial_write);
+
+/* chunk write helpers */
+static avi_error write_initial_headers(avi_file *file);
+static avi_error write_avih_chunk(avi_file *file, int initial_write);
+static avi_error write_strh_chunk(avi_file *file, avi_stream *stream, int initial_write);
+static avi_error write_strf_chunk(avi_file *file, avi_stream *stream);
+static avi_error write_indx_chunk(avi_file *file, avi_stream *stream, int initial_write);
+static avi_error write_idx1_chunk(avi_file *file);
+
+/* sound buffering helpers */
+static avi_error soundbuf_initialize(avi_file *file);
+static avi_error soundbuf_write_chunk(avi_file *file, UINT32 framenum);
+static avi_error soundbuf_flush(avi_file *file, int only_flush_full);
+
+/* YUY helpers */
+static avi_error yuv_decompress_to_yuy16(avi_stream *stream, const UINT8 *data, UINT32 numbytes, bitmap_t *bitmap);
+static avi_error yuy16_compress_to_yuy(avi_stream *stream, const bitmap_t *bitmap, UINT8 *data, UINT32 numbytes);
+
+/* HuffYUV helpers */
+static avi_error huffyuv_extract_tables(avi_stream *stream, const UINT8 *chunkdata, UINT32 size);
+static avi_error huffyuv_decompress_to_yuy16(avi_stream *stream, const UINT8 *data, UINT32 numbytes, bitmap_t *bitmap);
+
+/* debugging */
+static void printf_chunk_recursive(avi_file *file, avi_chunk *chunk, int indent);
+
+
+
+/***************************************************************************
+ INLINE FUNCTIONS
+***************************************************************************/
+
+/*-------------------------------------------------
+ fetch_16bits - read 16 bits in LSB order
+ from the given buffer
+-------------------------------------------------*/
+
+INLINE UINT16 fetch_16bits(const UINT8 *data)
+{
+ return data[0] | (data[1] << 8);
+}
+
+
+/*-------------------------------------------------
+ fetch_32bits - read 32 bits in LSB order
+ from the given buffer
+-------------------------------------------------*/
+
+INLINE UINT32 fetch_32bits(const UINT8 *data)
+{
+ return data[0] | (data[1] << 8) | (data[2] << 16) | (data[3] << 24);
+}
+
+
+/*-------------------------------------------------
+ fetch_64bits - read 64 bits in LSB order
+ from the given buffer
+-------------------------------------------------*/
+
+INLINE UINT64 fetch_64bits(const UINT8 *data)
+{
+ return (UINT64)data[0] | ((UINT64)data[1] << 8) |
+ ((UINT64)data[2] << 16) | ((UINT64)data[3] << 24) |
+ ((UINT64)data[4] << 32) | ((UINT64)data[5] << 40) |
+ ((UINT64)data[6] << 48) | ((UINT64)data[7] << 56);
+}
+
+
+/*-------------------------------------------------
+ put_16bits - write 16 bits in LSB order
+ to the given buffer
+-------------------------------------------------*/
+
+INLINE void put_16bits(UINT8 *data, UINT16 value)
+{
+ data[0] = value >> 0;
+ data[1] = value >> 8;
+}
+
+
+/*-------------------------------------------------
+ put_32bits - write 32 bits in LSB order
+ to the given buffer
+-------------------------------------------------*/
+
+INLINE void put_32bits(UINT8 *data, UINT32 value)
+{
+ data[0] = value >> 0;
+ data[1] = value >> 8;
+ data[2] = value >> 16;
+ data[3] = value >> 24;
+}
+
+
+/*-------------------------------------------------
+ put_64bits - write 64 bits in LSB order
+ to the given buffer
+-------------------------------------------------*/
+
+INLINE void put_64bits(UINT8 *data, UINT64 value)
+{
+ data[0] = value >> 0;
+ data[1] = value >> 8;
+ data[2] = value >> 16;
+ data[3] = value >> 24;
+ data[4] = value >> 32;
+ data[5] = value >> 40;
+ data[6] = value >> 48;
+ data[7] = value >> 56;
+}
+
+
+/*-------------------------------------------------
+ get_video_stream - return a pointer to the
+ video stream
+-------------------------------------------------*/
+
+INLINE avi_stream *get_video_stream(avi_file *file)
+{
+ int streamnum;
+
+ /* find the video stream */
+ for (streamnum = 0; streamnum < file->streams; streamnum++)
+ if (file->stream[streamnum].type == STREAMTYPE_VIDS)
+ return &file->stream[streamnum];
+
+ return NULL;
+}
+
+
+/*-------------------------------------------------
+ get_audio_stream - return a pointer to the
+ audio stream for the 'n'th channel
+-------------------------------------------------*/
+
+INLINE avi_stream *get_audio_stream(avi_file *file, int channel, int *offset)
+{
+ int streamnum;
+
+ /* find the audios stream */
+ for (streamnum = 0; streamnum < file->streams; streamnum++)
+ if (file->stream[streamnum].type == STREAMTYPE_AUDS)
+ {
+ if (channel < file->stream[streamnum].channels)
+ {
+ if (offset != NULL)
+ *offset = channel;
+ return &file->stream[streamnum];
+ }
+ channel -= file->stream[streamnum].channels;
+ }
+
+ return NULL;
+}
+
+
+/*-------------------------------------------------
+ set_stream_chunk_info - set the chunk info
+ for a given chunk within a stream
+-------------------------------------------------*/
+
+INLINE avi_error set_stream_chunk_info(avi_stream *stream, UINT32 index, UINT64 offset, UINT32 length)
+{
+ /* if we need to allocate more, allocate more */
+ if (index >= stream->chunksalloc)
+ {
+ UINT32 newcount = MAX(index, stream->chunksalloc + 1000);
+ stream->chunk = realloc(stream->chunk, newcount * sizeof(stream->chunk[0]));
+ if (stream->chunk == NULL)
+ return AVIERR_NO_MEMORY;
+ stream->chunksalloc = newcount;
+ }
+
+ /* set the data */
+ stream->chunk[index].offset = offset;
+ stream->chunk[index].length = length;
+
+ /* update the number of chunks */
+ stream->chunks = MAX(stream->chunks, index + 1);
+ return AVIERR_NONE;
+}
+
+
+/*-------------------------------------------------
+ compute_idx1_size - compute the size of the
+ idx1 chunk
+-------------------------------------------------*/
+
+INLINE UINT32 compute_idx1_size(avi_file *file)
+{
+ int chunks = 0;
+ int strnum;
+
+ /* count chunks in streams */
+ for (strnum = 0; strnum < file->streams; strnum++)
+ chunks += file->stream[strnum].chunks;
+
+ return chunks * 16 + 8;
+}
+
+
+/*-------------------------------------------------
+ get_chunkid_for_stream - make a chunk id for
+ a given stream
+-------------------------------------------------*/
+
+INLINE UINT32 get_chunkid_for_stream(avi_file *file, avi_stream *stream)
+{
+ UINT32 chunkid;
+
+ chunkid = AVI_FOURCC('0' + (stream - file->stream) / 10, '0' + (stream - file->stream) % 10, 0, 0);
+ if (stream->type == STREAMTYPE_VIDS)
+ chunkid |= (stream->format == 0) ? CHUNKTYPE_XXDB : CHUNKTYPE_XXDC;
+ else if (stream->type == STREAMTYPE_AUDS)
+ chunkid |= CHUNKTYPE_XXWB;
+
+ return chunkid;
+}
+
+
+/*-------------------------------------------------
+ framenum_to_samplenum - given a video frame
+ number, get the first sample number
+-------------------------------------------------*/
+
+INLINE UINT32 framenum_to_samplenum(avi_file *file, UINT32 framenum)
+{
+ return ((UINT64)file->info.audio_samplerate * (UINT64)framenum * (UINT64)file->info.video_sampletime + file->info.video_timescale - 1) / (UINT64)file->info.video_timescale;
+}
+
+
+/*-------------------------------------------------
+ expand_tempbuffer - expand the file's
+ tempbuffer if necessary to contain the
+ requested amount of data
+-------------------------------------------------*/
+
+INLINE avi_error expand_tempbuffer(avi_file *file, UINT32 length)
+{
+ /* expand the tempbuffer to hold the data if necessary */
+ if (length > file->tempbuffersize)
+ {
+ file->tempbuffersize = 2 * length;
+ file->tempbuffer = realloc(file->tempbuffer, file->tempbuffersize);
+ if (file->tempbuffer == NULL)
+ return AVIERR_NO_MEMORY;
+ }
+ return AVIERR_NONE;
+}
+
+
+
+/***************************************************************************
+ IMPLEMENTATION
+***************************************************************************/
+
+/*-------------------------------------------------
+ avi_open - open an AVI movie file for read
+-------------------------------------------------*/
+
+avi_error avi_open(const char *filename, avi_file **file)
+{
+ avi_file *newfile = NULL;
+ file_error filerr;
+ avi_error avierr;
+ UINT64 length;
+
+ /* allocate the file */
+ newfile = malloc(sizeof(*newfile));
+ if (newfile == NULL)
+ return AVIERR_NO_MEMORY;
+ memset(newfile, 0, sizeof(*newfile));
+ newfile->type = FILETYPE_READ;
+
+ /* open the file */
+ filerr = osd_open(filename, OPEN_FLAG_READ, &newfile->file, &length);
+ if (filerr != FILERR_NONE)
+ {
+ avierr = AVIERR_CANT_OPEN_FILE;
+ goto error;
+ }
+
+ /* make a root atom */
+ newfile->rootchunk.offset = 0;
+ newfile->rootchunk.size = length;
+ newfile->rootchunk.type = 0;
+ newfile->rootchunk.listtype = 0;
+
+ /* parse the data */
+ avierr = read_movie_data(newfile);
+ if (avierr != AVIERR_NONE)
+ goto error;
+
+ *file = newfile;
+ return AVIERR_NONE;
+
+error:
+ /* clean up after an error */
+ if (newfile != NULL)
+ {
+ if (newfile->file != NULL)
+ osd_close(newfile->file);
+ free(newfile);
+ }
+ return avierr;
+}
+
+
+/*-------------------------------------------------
+ avi_create - create a new QuickTime movie file
+-------------------------------------------------*/
+
+avi_error avi_create(const char *filename, const avi_movie_info *info, avi_file **file)
+{
+ avi_file *newfile = NULL;
+ file_error filerr;
+ avi_stream *stream;
+ avi_error avierr;
+ UINT64 length;
+
+ /* validate video info */
+ if ((info->video_format != 0 && info->video_format != FORMAT_UYVY && info->video_format != FORMAT_VYUY && info->video_format != FORMAT_YUY2) ||
+ info->video_width == 0 ||
+ info->video_height == 0 ||
+ info->video_depth == 0 || info->video_depth % 8 != 0)
+ return AVIERR_UNSUPPORTED_VIDEO_FORMAT;
+
+ /* validate audio info */
+ if (info->audio_format != 0 ||
+ info->audio_channels > MAX_SOUND_CHANNELS ||
+ info->audio_samplebits != 16)
+ return AVIERR_UNSUPPORTED_AUDIO_FORMAT;
+
+ /* allocate the file */
+ newfile = malloc(sizeof(*newfile));
+ if (newfile == NULL)
+ return AVIERR_NO_MEMORY;
+ memset(newfile, 0, sizeof(*newfile));
+ newfile->type = FILETYPE_CREATE;
+
+ /* open the file */
+ filerr = osd_open(filename, OPEN_FLAG_WRITE | OPEN_FLAG_CREATE | OPEN_FLAG_CREATE_PATHS, &newfile->file, &length);
+ if (filerr != FILERR_NONE)
+ {
+ avierr = AVIERR_CANT_OPEN_FILE;
+ goto error;
+ }
+
+ /* copy the movie info */
+ newfile->info = *info;
+ newfile->info.video_numsamples = 0;
+ newfile->info.audio_numsamples = 0;
+
+ /* allocate two streams */
+ newfile->stream = malloc(2 * sizeof(newfile->stream[0]));
+ if (newfile->stream == NULL)
+ {
+ avierr = AVIERR_NO_MEMORY;
+ goto error;
+ }
+ memset(newfile->stream, 0, 2 * sizeof(newfile->stream[0]));
+
+ /* initialize the video track */
+ stream = &newfile->stream[newfile->streams++];
+ stream->type = STREAMTYPE_VIDS;
+ stream->format = newfile->info.video_format;
+ stream->rate = newfile->info.video_timescale;
+ stream->scale = newfile->info.video_sampletime;
+ stream->width = newfile->info.video_width;
+ stream->height = newfile->info.video_height;
+ stream->depth = newfile->info.video_depth;
+
+ /* initialize the audio track */
+ if (newfile->info.audio_channels > 0)
+ {
+ stream = &newfile->stream[newfile->streams++];
+ stream->type = STREAMTYPE_AUDS;
+ stream->format = newfile->info.audio_format;
+ stream->rate = newfile->info.audio_timescale;
+ stream->scale = newfile->info.audio_sampletime;
+ stream->channels = newfile->info.audio_channels;
+ stream->samplebits = newfile->info.audio_samplebits;
+ stream->samplerate = newfile->info.audio_samplerate;
+
+ /* initialize the sound buffering */
+ avierr = soundbuf_initialize(newfile);
+ if (avierr != AVIERR_NONE)
+ goto error;
+ }
+
+ /* write the initial headers */
+ avierr = write_initial_headers(newfile);
+
+ *file = newfile;
+ return AVIERR_NONE;
+
+error:
+ /* clean up after an error */
+ if (newfile != NULL)
+ {
+ if (newfile->stream != NULL)
+ free(newfile->stream);
+ if (newfile->file != NULL)
+ {
+ osd_close(newfile->file);
+ osd_rmfile(filename);
+ }
+ free(newfile);
+ }
+ return avierr;
+}
+
+
+/*-------------------------------------------------
+ avi_close - close a QuickTime movie file
+-------------------------------------------------*/
+
+avi_error avi_close(avi_file *file)
+{
+ avi_error avierr = AVIERR_NONE;
+ int strnum;
+
+ /* if we're creating a new file, finalize it by writing out the non-media chunks */
+ if (file->type == FILETYPE_CREATE)
+ {
+ /* flush any pending sound data */
+ avierr = soundbuf_flush(file, FALSE);
+
+ /* close the movi chunk */
+ if (avierr == AVIERR_NONE)
+ avierr = chunk_close(file);
+
+ /* if this is the first RIFF chunk, write an idx1 */
+ if (avierr == AVIERR_NONE && file->riffbase == 0)
+ avierr = write_idx1_chunk(file);
+
+ /* close the RIFF chunk */
+ if (avierr == AVIERR_NONE)
+ avierr = chunk_close(file);
+
+ /* update the avih chunk */
+ if (avierr == AVIERR_NONE)
+ avierr = write_avih_chunk(file, FALSE);
+
+ /* update the strh and indx chunks for each stream */
+ for (strnum = 0; strnum < file->streams; strnum++)
+ {
+ if (avierr == AVIERR_NONE)
+ avierr = write_strh_chunk(file, &file->stream[strnum], FALSE);
+ if (avierr == AVIERR_NONE)
+ avierr = write_indx_chunk(file, &file->stream[strnum], FALSE);
+ }
+ }
+
+ /* close the file */
+ osd_close(file->file);
+
+ /* free the stream-specific data */
+ for (strnum = 0; strnum < file->streams; strnum++)
+ {
+ avi_stream *stream = &file->stream[strnum];
+ if (stream->huffyuv != NULL)
+ {
+ huffyuv_data *huffyuv = stream->huffyuv;
+ int table;
+
+ for (table = 0; table < ARRAY_LENGTH(huffyuv->table); table++)
+ if (huffyuv->table[table].extralookup != NULL)
+ free(huffyuv->table[table].extralookup);
+ free(huffyuv);
+ }
+ if (stream->chunk != NULL)
+ free(stream->chunk);
+ }
+
+ /* free the file itself */
+ if (file->soundbuf != NULL)
+ free(file->soundbuf);
+ if (file->stream != NULL)
+ free(file->stream);
+ if (file->tempbuffer != NULL)
+ free(file->tempbuffer);
+ free(file);
+ return avierr;
+}
+
+
+/*-------------------------------------------------
+ avi_printf_chunks - print the chunks in a file
+-------------------------------------------------*/
+
+void avi_printf_chunks(avi_file *file)
+{
+ printf_chunk_recursive(file, &file->rootchunk, 0);
+}
+
+
+/*-------------------------------------------------
+ avi_error_string - get the error string for
+ an avi_error
+-------------------------------------------------*/
+
+const char *avi_error_string(avi_error err)
+{
+ switch (err)
+ {
+ case AVIERR_NONE: return "success";
+ case AVIERR_END: return "hit end of file";
+ case AVIERR_INVALID_DATA: return "invalid data";
+ case AVIERR_NO_MEMORY: return "out of memory";
+ case AVIERR_READ_ERROR: return "read error";
+ case AVIERR_WRITE_ERROR: return "write error";
+ case AVIERR_STACK_TOO_DEEP: return "stack overflow";
+ case AVIERR_UNSUPPORTED_FEATURE: return "unsupported feature";
+ case AVIERR_CANT_OPEN_FILE: return "unable to open file";
+ case AVIERR_INCOMPATIBLE_AUDIO_STREAMS: return "found incompatible audio streams";
+ case AVIERR_INVALID_SAMPLERATE: return "found invalid sample rate";
+ case AVIERR_INVALID_STREAM: return "invalid stream";
+ case AVIERR_INVALID_FRAME: return "invalid frame index";
+ case AVIERR_INVALID_BITMAP: return "invalid bitmap";
+ case AVIERR_UNSUPPORTED_VIDEO_FORMAT: return "unsupported video format";
+ case AVIERR_UNSUPPORTED_AUDIO_FORMAT: return "unsupported audio format";
+ case AVIERR_EXCEEDED_SOUND_BUFFER: return "sound buffer overflow";
+ default: return "undocumented error";
+ }
+}
+
+
+/*-------------------------------------------------
+ avi_get_movie_info - return a pointer to the
+ movie info
+-------------------------------------------------*/
+
+const avi_movie_info *avi_get_movie_info(avi_file *file)
+{
+ return &file->info;
+}
+
+
+/*-------------------------------------------------
+ avi_frame_to_sample - convert a frame index
+ to a sample index
+-------------------------------------------------*/
+
+UINT32 avi_first_sample_in_frame(avi_file *file, UINT32 framenum)
+{
+ return framenum_to_samplenum(file, framenum);
+}
+
+
+/*-------------------------------------------------
+ avi_read_video_frame_yuy16 - read video data
+ for a particular frame from the AVI file,
+ converting to YUY16 format
+-------------------------------------------------*/
+
+avi_error avi_read_video_frame_yuy16(avi_file *file, UINT32 framenum, bitmap_t *bitmap)
+{
+ avi_error avierr = AVIERR_NONE;
+ UINT32 bytes_read, chunkid;
+ file_error filerr;
+ avi_stream *stream;
+
+ /* get the video stream */
+ stream = get_video_stream(file);
+ if (stream == NULL)
+ return AVIERR_INVALID_STREAM;
+
+ /* validate our ability to handle the data */
+ if (stream->format != FORMAT_UYVY && stream->format != FORMAT_VYUY && stream->format != FORMAT_YUY2 && stream->format != FORMAT_HFYU)
+ return AVIERR_UNSUPPORTED_VIDEO_FORMAT;
+
+ /* assume one chunk == one frame */
+ if (framenum >= stream->chunks)
+ return AVIERR_INVALID_FRAME;
+
+ /* we only support YUY-style bitmaps (16bpp) */
+ if (bitmap->format != BITMAP_FORMAT_YUY16 || bitmap->width < stream->width || bitmap->height < stream->height)
+ return AVIERR_INVALID_BITMAP;
+
+ /* expand the tempbuffer to hold the data if necessary */
+ avierr = expand_tempbuffer(file, stream->chunk[framenum].length);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+
+ /* read in the data */
+ filerr = osd_read(file->file, file->tempbuffer, stream->chunk[framenum].offset, stream->chunk[framenum].length, &bytes_read);
+ if (filerr != FILERR_NONE || bytes_read != stream->chunk[framenum].length)
+ return AVIERR_READ_ERROR;
+
+ /* validate this is good data */
+ chunkid = fetch_32bits(&file->tempbuffer[0]);
+ if (chunkid == get_chunkid_for_stream(file, stream))
+ {
+ /* HuffYUV-compressed */
+ if (stream->format == FORMAT_HFYU)
+ avierr = huffyuv_decompress_to_yuy16(stream, file->tempbuffer + 8, stream->chunk[framenum].length - 8, bitmap);
+
+ /* other YUV-compressed */
+ else
+ avierr = yuv_decompress_to_yuy16(stream, file->tempbuffer + 8, stream->chunk[framenum].length - 8, bitmap);
+ }
+ else
+ avierr = AVIERR_INVALID_DATA;
+
+ return avierr;
+}
+
+
+/*-------------------------------------------------
+ avi_read_sound_samples - read sound sample
+ data from an AVI file
+-------------------------------------------------*/
+
+avi_error avi_read_sound_samples(avi_file *file, int channel, UINT32 firstsample, UINT32 numsamples, INT16 *output)
+{
+ avi_error avierr = AVIERR_NONE;
+ UINT32 bytes_per_sample;
+ file_error filerr;
+ avi_stream *stream;
+ int offset;
+
+ /* get the audio stream */
+ stream = get_audio_stream(file, channel, &offset);
+ if (stream == NULL)
+ return AVIERR_INVALID_STREAM;
+
+ /* validate our ability to handle the data */
+ if (stream->format != 0 || (stream->samplebits != 8 && stream->samplebits != 16))
+ return AVIERR_UNSUPPORTED_AUDIO_FORMAT;
+
+ /* verify we are in range */
+ if (firstsample >= stream->samples)
+ return AVIERR_INVALID_FRAME;
+ if (firstsample + numsamples > stream->samples)
+ numsamples = stream->samples - firstsample;
+
+ /* determine bytes per sample */
+ bytes_per_sample = (stream->samplebits / 8) * stream->channels;
+
+ /* loop until all samples have been extracted */
+ while (numsamples > 0)
+ {
+ UINT32 chunkbase = 0, chunkend = 0, chunkid;
+ UINT32 bytes_read, samples_this_chunk;
+ int chunknum, sampnum;
+
+ /* locate the chunk with the first sample */
+ for (chunknum = 0; chunknum < stream->chunks; chunknum++)
+ {
+ chunkend = chunkbase + (stream->chunk[chunknum].length - 8) / bytes_per_sample;
+ if (firstsample < chunkend)
+ break;
+ chunkbase = chunkend;
+ }
+
+ /* expand the tempbuffer to hold the data if necessary */
+ avierr = expand_tempbuffer(file, stream->chunk[chunknum].length);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+
+ /* read in the data */
+ filerr = osd_read(file->file, file->tempbuffer, stream->chunk[chunknum].offset, stream->chunk[chunknum].length, &bytes_read);
+ if (filerr != FILERR_NONE || bytes_read != stream->chunk[chunknum].length)
+ return AVIERR_READ_ERROR;
+
+ /* validate this is good data */
+ chunkid = fetch_32bits(&file->tempbuffer[0]);
+ if (chunkid != get_chunkid_for_stream(file, stream))
+ return AVIERR_INVALID_DATA;
+
+ /* determine how many samples to copy */
+ samples_this_chunk = chunkend - firstsample;
+ samples_this_chunk = MIN(samples_this_chunk, numsamples);
+
+ /* extract 16-bit samples from the chunk */
+ if (stream->samplebits == 16)
+ {
+ const INT16 *base = (const INT16 *)(file->tempbuffer + 8);
+ base += stream->channels * (firstsample - chunkbase) + offset;
+ for (sampnum = 0; sampnum < samples_this_chunk; sampnum++)
+ {
+ *output++ = LITTLE_ENDIANIZE_INT16(*base);
+ base += stream->channels;
+ }
+ }
+
+ /* extract 8-bit samples from the chunk */
+ else if (stream->samplebits == 8)
+ {
+ const UINT8 *base = (const UINT8 *)(file->tempbuffer + 8);
+ base += stream->channels * (firstsample - chunkbase) + offset;
+ for (sampnum = 0; sampnum < samples_this_chunk; sampnum++)
+ {
+ *output++ = (*base << 8) - 0x8000;
+ base += stream->channels;
+ }
+ }
+
+ /* update our counters */
+ firstsample += samples_this_chunk;
+ numsamples -= samples_this_chunk;
+ }
+
+ return avierr;
+}
+
+
+/*-------------------------------------------------
+ avi_append_video_frame_yuy16 - append a frame
+ of video in YUY16 format
+-------------------------------------------------*/
+
+avi_error avi_append_video_frame_yuy16(avi_file *file, const bitmap_t *bitmap)
+{
+ avi_stream *stream = get_video_stream(file);
+ avi_error avierr;
+ UINT32 maxlength;
+
+ /* double check bitmap format */
+ if (bitmap->format != BITMAP_FORMAT_YUY16)
+ return AVIERR_INVALID_BITMAP;
+
+ /* write out any sound data first */
+ avierr = soundbuf_write_chunk(file, stream->chunks);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+
+ /* make sure we have enough room */
+ maxlength = 2 * stream->width * stream->height;
+ avierr = expand_tempbuffer(file, maxlength);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+
+ /* now compress the data */
+ avierr = yuy16_compress_to_yuy(stream, bitmap, file->tempbuffer, maxlength);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+
+ /* set the info for this new chunk */
+ avierr = set_stream_chunk_info(stream, stream->chunks, file->writeoffs, maxlength + 8);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+ stream->samples = file->info.video_numsamples = stream->chunks;
+
+ /* write the data */
+ return chunk_write(file, get_chunkid_for_stream(file, stream), file->tempbuffer, maxlength);
+}
+
+
+/*-------------------------------------------------
+ avi_append_sound_samples - append sound
+ samples
+-------------------------------------------------*/
+
+avi_error avi_append_sound_samples(avi_file *file, int channel, const INT16 *samples, UINT32 numsamples)
+{
+ UINT32 sampoffset = file->soundbuf_chansamples[channel];
+ UINT32 sampnum;
+
+ /* see if we have enough room in the buffer */
+ if (sampoffset + numsamples > file->soundbuf_samples)
+ return AVIERR_EXCEEDED_SOUND_BUFFER;
+
+ /* append samples to the buffer in little-endian format */
+ for (sampnum = 0; sampnum < numsamples; sampnum++)
+ {
+ INT16 data = *samples++;
+ data = LITTLE_ENDIANIZE_INT16(data);
+ file->soundbuf[sampoffset++ * file->info.audio_channels + channel] = data;
+ }
+ file->soundbuf_chansamples[channel] = sampoffset;
+
+ /* flush any full sound chunks to disk */
+ return soundbuf_flush(file, TRUE);
+}
+
+
+/*-------------------------------------------------
+ read_chunk_data - read a chunk's data into
+ memory
+-------------------------------------------------*/
+
+static avi_error read_chunk_data(avi_file *file, const avi_chunk *chunk, UINT8 **buffer)
+{
+ file_error filerr;
+ UINT32 bytes_read;
+
+ /* allocate memory for the data */
+ *buffer = malloc(chunk->size);
+ if (*buffer == NULL)
+ return AVIERR_NO_MEMORY;
+
+ /* read from the file */
+ filerr = osd_read(file->file, *buffer, chunk->offset + 8, chunk->size, &bytes_read);
+ if (filerr != FILERR_NONE || bytes_read != chunk->size)
+ {
+ free(*buffer);
+ *buffer = NULL;
+ return AVIERR_READ_ERROR;
+ }
+
+ return AVIERR_NONE;
+}
+
+
+/*-------------------------------------------------
+ get_first_chunk - get information about the
+ first chunk in a container
+-------------------------------------------------*/
+
+static avi_error get_first_chunk(avi_file *file, const avi_chunk *parent, avi_chunk *newchunk)
+{
+ UINT64 startoffset = (parent != NULL && parent->type != 0) ? parent->offset + 12 : 0;
+ if (parent != NULL && parent->type != CHUNKTYPE_LIST && parent->type != CHUNKTYPE_RIFF && parent->type != 0)
+ return AVIERR_INVALID_DATA;
+ return get_next_chunk_internal(file, parent, newchunk, startoffset);
+}
+
+
+/*-------------------------------------------------
+ get_next_chunk - get information about the
+ next chunk in a container
+-------------------------------------------------*/
+
+static avi_error get_next_chunk(avi_file *file, const avi_chunk *parent, avi_chunk *newchunk)
+{
+ UINT64 nextoffset = newchunk->offset + 8 + newchunk->size + (newchunk->size & 1);
+ return get_next_chunk_internal(file, parent, newchunk, nextoffset);
+}
+
+
+/*-------------------------------------------------
+ find_first_chunk - get information about the
+ first chunk of a particular type in a container
+-------------------------------------------------*/
+
+static avi_error find_first_chunk(avi_file *file, UINT32 findme, const avi_chunk *container, avi_chunk *result)
+{
+ avi_error avierr;
+
+ for (avierr = get_first_chunk(file, container, result); avierr == AVIERR_NONE; avierr = get_next_chunk(file, container, result))
+ if (result->type == findme)
+ return AVIERR_NONE;
+
+ return avierr;
+}
+
+
+/*-------------------------------------------------
+ find_next_chunk - get information about the
+ next chunk of a particular type in a container
+-------------------------------------------------*/
+
+static avi_error find_next_chunk(avi_file *file, UINT32 findme, const avi_chunk *container, avi_chunk *result)
+{
+ avi_error avierr;
+
+ for (avierr = get_next_chunk(file, container, result); avierr == AVIERR_NONE; avierr = get_next_chunk(file, container, result))
+ if (result->type == findme)
+ return AVIERR_NONE;
+
+ return avierr;
+}
+
+
+/*-------------------------------------------------
+ find_first_list - get information about the
+ first list of a particular type in a container
+-------------------------------------------------*/
+
+static avi_error find_first_list(avi_file *file, UINT32 findme, const avi_chunk *container, avi_chunk *result)
+{
+ avi_error avierr;
+
+ for (avierr = find_first_chunk(file, CHUNKTYPE_LIST, container, result); avierr == AVIERR_NONE; avierr = find_next_chunk(file, CHUNKTYPE_LIST, container, result))
+ if (result->listtype == findme)
+ return AVIERR_NONE;
+
+ return avierr;
+}
+
+
+/*-------------------------------------------------
+ find_next_list - get information about the
+ next list of a particular type in a container
+-------------------------------------------------*/
+
+static avi_error find_next_list(avi_file *file, UINT32 findme, const avi_chunk *container, avi_chunk *result)
+{
+ avi_error avierr;
+
+ for (avierr = find_next_chunk(file, CHUNKTYPE_LIST, container, result); avierr == AVIERR_NONE; avierr = find_next_chunk(file, CHUNKTYPE_LIST, container, result))
+ if (result->listtype == findme)
+ return AVIERR_NONE;
+
+ return avierr;
+}
+
+
+/*-------------------------------------------------
+ get_next_chunk_internal - fetch the next
+ chunk relative to the current one
+-------------------------------------------------*/
+
+static avi_error get_next_chunk_internal(avi_file *file, const avi_chunk *parent, avi_chunk *newchunk, UINT64 offset)
+{
+ file_error filerr;
+ UINT8 buffer[12];
+ UINT32 bytesread;
+
+ /* NULL parent implies the root */
+ if (parent == NULL)
+ parent = &file->rootchunk;
+
+ /* start at the current offset */
+ newchunk->offset = offset;
+
+ /* if we're past the bounds of the parent, bail */
+ if (newchunk->offset + 8 >= parent->offset + 8 + parent->size)
+ return AVIERR_END;
+
+ /* read the header */
+ filerr = osd_read(file->file, buffer, newchunk->offset, 8, &bytesread);
+ if (filerr != FILERR_NONE || bytesread != 8)
+ return AVIERR_INVALID_DATA;
+
+ /* fill in the new chunk */
+ newchunk->type = fetch_32bits(&buffer[0]);
+ newchunk->size = fetch_32bits(&buffer[4]);
+
+ /* if we are a list, fetch the list type */
+ if (newchunk->type == CHUNKTYPE_LIST || newchunk->type == CHUNKTYPE_RIFF)
+ {
+ filerr = osd_read(file->file, &buffer[8], newchunk->offset + 8, 4, &bytesread);
+ if (filerr != FILERR_NONE || bytesread != 4)
+ return AVIERR_INVALID_DATA;
+ newchunk->listtype = fetch_32bits(&buffer[8]);
+ }
+
+ return AVIERR_NONE;
+}
+
+
+/*-------------------------------------------------
+ read_movie_data - get data about a movie
+-------------------------------------------------*/
+
+static avi_error read_movie_data(avi_file *file)
+{
+ avi_chunk riff, hdrl, avih, strl, strh, strf, indx, movi, idx1;
+ avi_error avierr;
+ int strindex;
+
+ /* find the RIFF chunk */
+ avierr = find_first_chunk(file, CHUNKTYPE_RIFF, NULL, &riff);
+ if (avierr != AVIERR_NONE)
+ goto error;
+
+ /* verify that the RIFF type is AVI */
+ if (riff.listtype != LISTTYPE_AVI)
+ {
+ avierr = AVIERR_INVALID_DATA;
+ goto error;
+ }
+
+ /* find the hdrl LIST chunk within the RIFF */
+ avierr = find_first_list(file, LISTTYPE_HDRL, &riff, &hdrl);
+ if (avierr != AVIERR_NONE)
+ goto error;
+
+ /* find the avih chunk */
+ avierr = find_first_chunk(file, CHUNKTYPE_AVIH, &hdrl, &avih);
+ if (avierr != AVIERR_NONE)
+ goto error;
+
+ /* parse the avih chunk */
+ avierr = parse_avih_chunk(file, &avih);
+ if (avierr != AVIERR_NONE)
+ goto error;
+
+ /* loop over strl LIST chunks */
+ strindex = 0;
+ for (avierr = find_first_list(file, LISTTYPE_STRL, &hdrl, &strl); avierr == AVIERR_NONE; avierr = find_next_list(file, LISTTYPE_STRL, &hdrl, &strl))
+ {
+ /* if we have too many, it's a bad file */
+ if (strindex >= file->streams)
+ goto error;
+
+ /* find the strh chunk */
+ avierr = find_first_chunk(file, CHUNKTYPE_STRH, &strl, &strh);
+ if (avierr != AVIERR_NONE)
+ goto error;
+
+ /* parse the data */
+ avierr = parse_strh_chunk(file, &file->stream[strindex], &strh);
+ if (avierr != AVIERR_NONE)
+ goto error;
+
+ /* find the strf chunk */
+ avierr = find_first_chunk(file, CHUNKTYPE_STRF, &strl, &strf);
+ if (avierr != AVIERR_NONE)
+ goto error;
+
+ /* parse the data */
+ avierr = parse_strf_chunk(file, &file->stream[strindex], &strf);
+ if (avierr != AVIERR_NONE)
+ goto error;
+
+ /* find the indx chunk, if present */
+ avierr = find_first_chunk(file, CHUNKTYPE_INDX, &strl, &indx);
+ if (avierr == AVIERR_NONE)
+ avierr = parse_indx_chunk(file, &file->stream[strindex], &indx);
+
+ /* next stream */
+ strindex++;
+ }
+
+ /* normalize the error after parsing the stream headers */
+ if (avierr == AVIERR_END)
+ avierr = AVIERR_NONE;
+ if (avierr != AVIERR_NONE)
+ goto error;
+
+ /* find the base of the movi data */
+ avierr = find_first_list(file, LISTTYPE_MOVI, &riff, &movi);
+ if (avierr != AVIERR_NONE)
+ goto error;
+
+ /* find and parse the idx1 chunk within the RIFF (if present) */
+ avierr = find_first_chunk(file, CHUNKTYPE_IDX1, &riff, &idx1);
+ if (avierr == AVIERR_NONE)
+ avierr = parse_idx1_chunk(file, movi.offset + 8, &idx1);
+ avierr = AVIERR_NONE;
+
+ /* now extract the movie info */
+ avierr = extract_movie_info(file);
+
+error:
+ return avierr;
+}
+
+
+/*-------------------------------------------------
+ extract_movie_info - extract the movie info
+ from the streams we've read
+-------------------------------------------------*/
+
+static avi_error extract_movie_info(avi_file *file)
+{
+ avi_stream *audiostream;
+ avi_stream *stream;
+
+ /* get the video stream */
+ stream = get_video_stream(file);
+ if (stream != NULL)
+ {
+ /* fill in the info */
+ file->info.video_format = stream->format;
+ file->info.video_timescale = stream->rate;
+ file->info.video_sampletime = stream->scale;
+ file->info.video_numsamples = stream->samples;
+ file->info.video_width = stream->width;
+ file->info.video_height = stream->height;
+ }
+
+ /* get the first audio stream */
+ stream = get_audio_stream(file, 0, NULL);
+ if (stream != NULL)
+ {
+ /* fill in the info */
+ file->info.audio_format = stream->format;
+ file->info.audio_timescale = stream->rate;
+ file->info.audio_sampletime = stream->scale;
+ file->info.audio_numsamples = stream->samples;
+ file->info.audio_channels = 1;
+ file->info.audio_samplebits = stream->samplebits;
+ file->info.audio_samplerate = stream->samplerate;
+ }
+
+ /* now make sure all other audio streams are valid */
+ audiostream = stream;
+ while (1)
+ {
+ /* get the stream info */
+ stream = get_audio_stream(file, file->info.audio_channels, NULL);
+ if (stream == NULL)
+ break;
+ file->info.audio_channels++;
+
+ /* verify compatibility */
+ if (file->info.audio_format != stream->format ||
+ file->info.audio_timescale != stream->rate ||
+ file->info.audio_sampletime != stream->scale ||
+ file->info.audio_numsamples != stream->samples ||
+ file->info.audio_samplebits != stream->samplebits ||
+ file->info.audio_samplerate != stream->samplerate)
+ return AVIERR_INCOMPATIBLE_AUDIO_STREAMS;
+ }
+
+ return AVIERR_NONE;
+}
+
+
+/*-------------------------------------------------
+ parse_avih_chunk - parse an avih header
+ chunk
+-------------------------------------------------*/
+
+static avi_error parse_avih_chunk(avi_file *file, avi_chunk *avih)
+{
+ UINT8 *chunkdata = NULL;
+ avi_error avierr;
+
+ /* read the data */
+ avierr = read_chunk_data(file, avih, &chunkdata);
+ if (avierr != AVIERR_NONE)
+ goto error;
+
+ /* extract the data */
+ file->streams = fetch_32bits(&chunkdata[24]);
+
+ /* allocate memory for the streams */
+ file->stream = malloc(sizeof(*file->stream) * file->streams);
+ if (file->stream == NULL)
+ goto error;
+ memset(file->stream, 0, sizeof(*file->stream) * file->streams);
+
+error:
+ if (chunkdata != NULL)
+ free(chunkdata);
+ return avierr;
+}
+
+
+/*-------------------------------------------------
+ parse_strh_chunk - parse a strh header
+ chunk
+-------------------------------------------------*/
+
+static avi_error parse_strh_chunk(avi_file *file, avi_stream *stream, avi_chunk *strh)
+{
+ UINT8 *chunkdata = NULL;
+ avi_error avierr;
+
+ /* read the data */
+ avierr = read_chunk_data(file, strh, &chunkdata);
+ if (avierr != AVIERR_NONE)
+ goto error;
+
+ /* extract the data */
+ stream->type = fetch_32bits(&chunkdata[0]);
+ stream->scale = fetch_32bits(&chunkdata[20]);
+ stream->rate = fetch_32bits(&chunkdata[24]);
+ stream->samples = fetch_32bits(&chunkdata[32]);
+
+error:
+ if (chunkdata != NULL)
+ free(chunkdata);
+ return avierr;
+}
+
+
+/*-------------------------------------------------
+ parse_strf_chunk - parse a strf header
+ chunk
+-------------------------------------------------*/
+
+static avi_error parse_strf_chunk(avi_file *file, avi_stream *stream, avi_chunk *strf)
+{
+ UINT8 *chunkdata = NULL;
+ avi_error avierr;
+
+ /* read the data */
+ avierr = read_chunk_data(file, strf, &chunkdata);
+ if (avierr != AVIERR_NONE)
+ goto error;
+
+ /* audio and video streams have differing headers */
+ if (stream->type == STREAMTYPE_VIDS)
+ {
+ stream->width = fetch_32bits(&chunkdata[4]);
+ stream->height = fetch_32bits(&chunkdata[8]);
+ stream->depth = fetch_16bits(&chunkdata[14]);
+ stream->format = fetch_32bits(&chunkdata[16]);
+
+ /* extra extraction for HuffYUV data */
+ if (stream->format == FORMAT_HFYU && strf->size >= 56)
+ {
+ avierr = huffyuv_extract_tables(stream, chunkdata, strf->size);
+ if (avierr != AVIERR_NONE)
+ goto error;
+ }
+ }
+ else if (stream->type == STREAMTYPE_AUDS)
+ {
+ stream->channels = fetch_16bits(&chunkdata[2]);
+ stream->samplebits = fetch_16bits(&chunkdata[14]);
+ stream->samplerate = fetch_32bits(&chunkdata[4]);
+ }
+
+error:
+ if (chunkdata != NULL)
+ free(chunkdata);
+ return avierr;
+}
+
+
+/*-------------------------------------------------
+ parse_indx_chunk - parse an indx chunk
+-------------------------------------------------*/
+
+static avi_error parse_indx_chunk(avi_file *file, avi_stream *stream, avi_chunk *strf)
+{
+ UINT32 entries, entry, id;
+ UINT8 *chunkdata = NULL;
+ UINT16 longs_per_entry;
+ UINT8 subtype, type;
+ UINT64 baseoffset;
+ avi_error avierr;
+
+ /* read the data */
+ avierr = read_chunk_data(file, strf, &chunkdata);
+ if (avierr != AVIERR_NONE)
+ goto error;
+
+ /* extract the data */
+ longs_per_entry = fetch_16bits(&chunkdata[0]);
+ subtype = chunkdata[2];
+ type = chunkdata[3];
+ entries = fetch_32bits(&chunkdata[4]);
+ id = fetch_32bits(&chunkdata[8]);
+ baseoffset = fetch_64bits(&chunkdata[12]);
+
+ /* if this is a superindex, loop over entries and call ourselves recursively */
+ if (type == AVI_INDEX_OF_INDEXES)
+ {
+ /* validate the size of each entry */
+ if (longs_per_entry != 4)
+ return AVIERR_INVALID_DATA;
+
+ /* loop over entries and create subchunks for each */
+ for (entry = 0; entry < entries; entry++)
+ {
+ const UINT8 *base = &chunkdata[24 + entry * 16];
+ file_error filerr;
+ avi_chunk subchunk;
+ UINT32 bytes_read;
+ UINT8 buffer[8];
+
+ /* go read the subchunk */
+ subchunk.offset = fetch_64bits(&base[0]);
+ filerr = osd_read(file->file, buffer, subchunk.offset, sizeof(buffer), &bytes_read);
+ if (filerr != FILERR_NONE || bytes_read != sizeof(buffer))
+ {
+ avierr = AVIERR_READ_ERROR;
+ break;
+ }
+
+ /* parse the data */
+ subchunk.type = fetch_32bits(&buffer[0]);
+ subchunk.size = fetch_32bits(&buffer[4]);
+
+ /* recursively parse each referenced chunk; stop if we hit an error */
+ avierr = parse_indx_chunk(file, stream, &subchunk);
+ if (avierr != AVIERR_NONE)
+ break;
+ }
+ }
+
+ /* otherwise, this is a standard index */
+ else if (type == AVI_INDEX_OF_CHUNKS)
+ {
+ /* validate the size of each entry */
+ if (longs_per_entry != 2 && longs_per_entry != 3)
+ return AVIERR_INVALID_DATA;
+
+ /* loop over entries and parse out the data */
+ for (entry = 0; entry < entries; entry++)
+ {
+ const UINT8 *base = &chunkdata[24 + entry * 4 * longs_per_entry];
+ UINT32 offset = fetch_32bits(&base[0]);
+ UINT32 size = fetch_32bits(&base[4]);
+
+ /* set the info for this chunk and advance */
+ avierr = set_stream_chunk_info(stream, stream->chunks++, baseoffset + offset - 8, size + 8);
+ if (avierr != AVIERR_NONE)
+ break;
+ }
+ }
+
+error:
+ if (chunkdata != NULL)
+ free(chunkdata);
+ return avierr;
+}
+
+
+/*-------------------------------------------------
+ parse_idx1_chunk - parse an idx1 chunk
+-------------------------------------------------*/
+
+static avi_error parse_idx1_chunk(avi_file *file, UINT64 baseoffset, avi_chunk *idx1)
+{
+ UINT8 *chunkdata = NULL;
+ avi_error avierr;
+ UINT32 entries;
+ UINT32 entry;
+
+ /* read the data */
+ avierr = read_chunk_data(file, idx1, &chunkdata);
+ if (avierr != AVIERR_NONE)
+ goto error;
+
+ /* loop over entries */
+ entries = idx1->size / 16;
+ for (entry = 0; entry < entries; entry++)
+ {
+ const UINT8 *base = &chunkdata[entry * 16];
+ UINT32 chunkid = fetch_32bits(&base[0]);
+ UINT32 offset = fetch_32bits(&base[8]);
+ UINT32 size = fetch_32bits(&base[12]);
+ avi_stream *stream;
+ int streamnum;
+
+ /* determine the stream index */
+ streamnum = ((chunkid >> 8) & 0xff) - '0';
+ streamnum += 10 * ((chunkid & 0xff) - '0');
+ if (streamnum >= file->streams)
+ {
+ avierr = AVIERR_INVALID_DATA;
+ goto error;
+ }
+ stream = &file->stream[streamnum];
+
+ /* set the appropriate entry */
+ avierr = set_stream_chunk_info(stream, stream->chunks++, baseoffset + offset, size + 8);
+ if (avierr != AVIERR_NONE)
+ goto error;
+ }
+
+error:
+ if (chunkdata != NULL)
+ free(chunkdata);
+ return avierr;
+}
+
+
+/*-------------------------------------------------
+ chunk_open - open a new chunk for writing
+-------------------------------------------------*/
+
+static avi_error chunk_open(avi_file *file, UINT32 type, UINT32 listtype, UINT32 estlength)
+{
+ file_error filerr;
+ avi_chunk *chunk;
+ UINT32 written;
+
+ /* if we're out of stack entries, bail */
+ if (file->chunksp >= ARRAY_LENGTH(file->chunkstack))
+ return AVIERR_STACK_TOO_DEEP;
+ chunk = &file->chunkstack[file->chunksp++];
+
+ /* set up the chunk information */
+ chunk->offset = file->writeoffs;
+ chunk->size = estlength;
+ chunk->type = type;
+ chunk->listtype = listtype;
+
+ /* non-list types */
+ if (type != CHUNKTYPE_RIFF && type != CHUNKTYPE_LIST)
+ {
+ UINT8 buffer[8];
+
+ /* populate the header */
+ put_32bits(&buffer[0], chunk->type);
+ put_32bits(&buffer[4], chunk->size);
+
+ /* write the header */
+ filerr = osd_write(file->file, buffer, file->writeoffs, sizeof(buffer), &written);
+ if (filerr != FILERR_NONE || written != sizeof(buffer))
+ return AVIERR_WRITE_ERROR;
+ file->writeoffs += written;
+ }
+
+ /* list types */
+ else
+ {
+ UINT8 buffer[12];
+
+ /* populate the header */
+ put_32bits(&buffer[0], chunk->type);
+ put_32bits(&buffer[4], chunk->size);
+ put_32bits(&buffer[8], chunk->listtype);
+
+ /* write the header */
+ filerr = osd_write(file->file, buffer, file->writeoffs, sizeof(buffer), &written);
+ if (filerr != FILERR_NONE || written != sizeof(buffer))
+ return AVIERR_WRITE_ERROR;
+ file->writeoffs += written;
+ }
+
+ return AVIERR_NONE;
+}
+
+
+/*-------------------------------------------------
+ chunk_close - finish writing an chunk
+-------------------------------------------------*/
+
+static avi_error chunk_close(avi_file *file)
+{
+ avi_chunk *chunk = &file->chunkstack[--file->chunksp];
+ UINT64 chunksize = file->writeoffs - (chunk->offset + 8);
+ UINT32 written;
+
+ /* error if we don't fit into 32 bits */
+ if (chunksize != (UINT32)chunksize)
+ return AVIERR_INVALID_DATA;
+
+ /* write the final size if it is different from the guess */
+ if (chunk->size != chunksize)
+ {
+ file_error filerr;
+ UINT8 buffer[4];
+
+ put_32bits(&buffer[0], (UINT32)chunksize);
+ filerr = osd_write(file->file, buffer, chunk->offset + 4, sizeof(buffer), &written);
+ if (filerr != FILERR_NONE || written != sizeof(buffer))
+ return AVIERR_WRITE_ERROR;
+ }
+
+ /* round up to the next word */
+ file->writeoffs += chunksize & 1;
+
+ return AVIERR_NONE;
+}
+
+
+/*-------------------------------------------------
+ chunk_write - write an chunk and its data
+-------------------------------------------------*/
+
+static avi_error chunk_write(avi_file *file, UINT32 type, const void *data, UINT32 length)
+{
+ file_error filerr;
+ avi_error avierr;
+ UINT32 written;
+
+ /* if we are getting too big, split the RIFF */
+ if ((file->riffbase == 0 && file->writeoffs + length + compute_idx1_size(file) >= MAX_RIFF_SIZE) ||
+ file->writeoffs + length - file->riffbase >= MAX_RIFF_SIZE)
+ {
+ /* close the movi list */
+ avierr = chunk_close(file);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+
+ /* write the idx1 chunk if this is the first */
+ if (file->riffbase == 0)
+ {
+ avierr = write_idx1_chunk(file);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+ }
+
+ /* close the RIFF */
+ avierr = chunk_close(file);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+
+ /* open a new RIFF */
+ file->riffbase = file->writeoffs;
+ avierr = chunk_open(file, CHUNKTYPE_RIFF, LISTTYPE_AVIX, 0);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+
+ /* open a nested movi list */
+ file->saved_movi_offset = file->writeoffs;
+ avierr = chunk_open(file, CHUNKTYPE_LIST, LISTTYPE_MOVI, 0);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+ }
+
+ /* open the chunk */
+ avierr = chunk_open(file, type, 0, length);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+
+ /* write the data */
+ filerr = osd_write(file->file, data, file->writeoffs, length, &written);
+ if (filerr != FILERR_NONE || written != length)
+ return AVIERR_WRITE_ERROR;
+ file->writeoffs += written;
+
+ /* close the chunk */
+ return chunk_close(file);
+}
+
+
+/*-------------------------------------------------
+ chunk_overwrite - write a chunk in two passes;
+ first pass writes to the end of file and
+ saves the offset; second pass overwrites the
+ original
+-------------------------------------------------*/
+
+static avi_error chunk_overwrite(avi_file *file, UINT32 type, const void *data, UINT32 length, UINT64 *offset, int initial_write)
+{
+ UINT64 savedoffset = 0;
+ avi_error avierr;
+
+ /* if this is our initial write, save the offset */
+ if (initial_write)
+ *offset = file->writeoffs;
+
+ /* otherwise, remember the current write offset and replace it with the original */
+ else
+ {
+ savedoffset = file->writeoffs;
+ file->writeoffs = *offset;
+ }
+
+ /* write the chunk */
+ avierr = chunk_write(file, type, data, length);
+
+ /* if this isn't the initial write, restore the previous offset */
+ if (!initial_write)
+ file->writeoffs = savedoffset;
+
+ return avierr;
+}
+
+
+/*-------------------------------------------------
+ write_initial_headers - write out the inital
+ set of AVI and stream headers
+-------------------------------------------------*/
+
+static avi_error write_initial_headers(avi_file *file)
+{
+ avi_error avierr;
+ int strnum;
+
+ /* reset the write pointer */
+ file->writeoffs = 0;
+
+ /* open a RIFF chunk */
+ avierr = chunk_open(file, CHUNKTYPE_RIFF, LISTTYPE_AVI, 0);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+
+ /* open a hdlr LIST */
+ avierr = chunk_open(file, CHUNKTYPE_LIST, LISTTYPE_HDRL, 0);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+
+ /* write an avih chunk */
+ avierr = write_avih_chunk(file, TRUE);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+
+ /* for each stream, write a strl LIST */
+ for (strnum = 0; strnum < file->streams; strnum++)
+ {
+ /* open a strl LIST */
+ avierr = chunk_open(file, CHUNKTYPE_LIST, LISTTYPE_STRL, 0);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+
+ /* write the strh chunk */
+ avierr = write_strh_chunk(file, &file->stream[strnum], TRUE);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+
+ /* write the strf chunk */
+ avierr = write_strf_chunk(file, &file->stream[strnum]);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+
+ /* write the indx chunk */
+ avierr = write_indx_chunk(file, &file->stream[strnum], TRUE);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+
+ /* close the strl LIST */
+ avierr = chunk_close(file);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+ }
+
+ /* close the hdlr LIST */
+ avierr = chunk_close(file);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+
+ /* open a movi LIST */
+ file->saved_movi_offset = file->writeoffs;
+ avierr = chunk_open(file, CHUNKTYPE_LIST, LISTTYPE_MOVI, 0);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+
+ return avierr;
+}
+
+
+/*-------------------------------------------------
+ write_avih_chunk - write the avih header
+ chunk
+-------------------------------------------------*/
+
+static avi_error write_avih_chunk(avi_file *file, int initial_write)
+{
+ avi_stream *video = get_video_stream(file);
+ UINT8 buffer[56];
+
+ /* reset the buffer */
+ memset(buffer, 0, sizeof(buffer));
+
+ put_32bits(&buffer[0], 1000000 * (INT64)video->scale / video->rate); /* dwMicroSecPerFrame */
+ put_32bits(&buffer[12], AVIF_HASINDEX | AVIF_ISINTERLEAVED); /* dwFlags */
+ put_32bits(&buffer[16], video->samples); /* dwTotalFrames */
+ put_32bits(&buffer[24], file->streams); /* dwStreams */
+ put_32bits(&buffer[32], video->width); /* dwWidth */
+ put_32bits(&buffer[36], video->height); /* dwHeight */
+
+ /* (over)write the chunk */
+ return chunk_overwrite(file, CHUNKTYPE_AVIH, buffer, sizeof(buffer), &file->saved_avih_offset, initial_write);
+}
+
+
+/*-------------------------------------------------
+ write_strh_chunk - write the strh header
+ chunk
+-------------------------------------------------*/
+
+static avi_error write_strh_chunk(avi_file *file, avi_stream *stream, int initial_write)
+{
+ UINT8 buffer[56];
+
+ /* reset the buffer */
+ memset(buffer, 0, sizeof(buffer));
+
+ put_32bits(&buffer[0], stream->type); /* fccType */
+ put_32bits(&buffer[20], stream->scale); /* dwScale */
+ put_32bits(&buffer[24], stream->rate); /* dwRate */
+ put_32bits(&buffer[32], stream->samples); /* dwLength */
+ put_32bits(&buffer[40], 10000); /* dwQuality */
+
+ /* video-stream specific data */
+ if (stream->type == STREAMTYPE_VIDS)
+ {
+ put_32bits(&buffer[4], /* fccHandler */
+ (stream->format == FORMAT_HFYU) ? HANDLER_HFYU : HANDLER_DIB);
+ put_32bits(&buffer[36], /* dwSuggestedBufferSize */
+ stream->width * stream->height * 4);
+ put_16bits(&buffer[52], stream->width); /* rcFrame.right */
+ put_16bits(&buffer[54], stream->height); /* rcFrame.bottom */
+ }
+
+ /* audio-stream specific data */
+ if (stream->type == STREAMTYPE_AUDS)
+ {
+ put_32bits(&buffer[36], /* dwSuggestedBufferSize */
+ stream->samplerate * stream->channels * (stream->samplebits / 8));
+ put_32bits(&buffer[44], /* dwSampleSize */
+ stream->channels * (stream->samplebits / 8));
+ }
+
+ /* write the chunk */
+ return chunk_overwrite(file, CHUNKTYPE_STRH, buffer, sizeof(buffer), &stream->saved_strh_offset, initial_write);
+}
+
+
+/*-------------------------------------------------
+ write_strf_chunk - write the strf header
+ chunk
+-------------------------------------------------*/
+
+static avi_error write_strf_chunk(avi_file *file, avi_stream *stream)
+{
+ /* video stream */
+ if (stream->type == STREAMTYPE_VIDS)
+ {
+ UINT8 buffer[40];
+
+ /* reset the buffer */
+ memset(buffer, 0, sizeof(buffer));
+
+ put_32bits(&buffer[0], sizeof(buffer)); /* biSize */
+ put_32bits(&buffer[4], stream->width); /* biWidth */
+ put_32bits(&buffer[8], stream->height); /* biHeight */
+ put_16bits(&buffer[12], 1); /* biPlanes */
+ put_16bits(&buffer[14], stream->depth); /* biBitCount */
+ put_32bits(&buffer[16], stream->format); /* biCompression */
+ put_32bits(&buffer[20], /* biSizeImage */
+ stream->width * stream->height * (stream->depth + 7) / 8);
+
+ /* write the chunk */
+ return chunk_write(file, CHUNKTYPE_STRF, buffer, sizeof(buffer));
+ }
+
+ /* audio stream */
+ if (stream->type == STREAMTYPE_AUDS)
+ {
+ UINT8 buffer[16];
+
+ /* reset the buffer */
+ memset(buffer, 0, sizeof(buffer));
+
+ put_16bits(&buffer[0], 1); /* wFormatTag */
+ put_16bits(&buffer[2], stream->channels); /* nChannels */
+ put_32bits(&buffer[4], stream->samplerate); /* nSamplesPerSec */
+ put_32bits(&buffer[8], /* nAvgBytesPerSec */
+ stream->samplerate * stream->channels * (stream->samplebits / 8));
+ put_16bits(&buffer[12], /* nBlockAlign */
+ stream->channels * (stream->samplebits / 8));
+ put_16bits(&buffer[14], stream->samplebits); /* wBitsPerSample */
+
+ /* write the chunk */
+ return chunk_write(file, CHUNKTYPE_STRF, buffer, sizeof(buffer));
+ }
+
+ return AVIERR_INVALID_DATA;
+}
+
+
+/*-------------------------------------------------
+ write_indx_chunk - write the indx header
+ chunk
+-------------------------------------------------*/
+
+static avi_error write_indx_chunk(avi_file *file, avi_stream *stream, int initial_write)
+{
+ UINT8 buffer[24 + 16 * MAX_AVI_SIZE_IN_GB / 4];
+ UINT32 chunkid, indexchunkid;
+ UINT32 master_entries = 0;
+
+ /* reset the buffer */
+ memset(buffer, 0, sizeof(buffer));
+
+ /* construct the chunk ID and index chunk ID */
+ chunkid = get_chunkid_for_stream(file, stream);
+ indexchunkid = AVI_FOURCC('i', 'x', '0' + (stream - file->stream) / 10, '0' + (stream - file->stream) % 10);
+
+ /* loop over chunks of 4GB and write out indexes for them first */
+ if (!initial_write && file->riffbase != 0)
+ {
+ UINT64 currentbase;
+ for (currentbase = 0; currentbase < file->writeoffs; currentbase += FOUR_GB)
+ {
+ UINT64 currentend = currentbase + FOUR_GB;
+ UINT32 chunks_this_index = 0;
+ UINT32 bytes_this_index = 0;
+ avi_error avierr;
+ UINT32 chunknum;
+ UINT8 *tempbuf;
+
+ /* count chunks that are in this region */
+ for (chunknum = 0; chunknum < stream->chunks; chunknum++)
+ if (stream->chunk[chunknum].offset >= currentbase && stream->chunk[chunknum].offset < currentend)
+ chunks_this_index++;
+
+ /* if no chunks, skip */
+ if (chunks_this_index == 0)
+ continue;
+
+ /* allocate memory */
+ tempbuf = malloc(24 + 8 * chunks_this_index);
+ if (tempbuf == NULL)
+ return AVIERR_NO_MEMORY;
+ memset(tempbuf, 0, 24 + 8 * chunks_this_index);
+
+ /* make a regular index */
+ put_16bits(&tempbuf[0], 2); /* wLongsPerEntry */
+ tempbuf[2] = 0; /* bIndexSubType */
+ tempbuf[3] = AVI_INDEX_OF_CHUNKS; /* bIndexType */
+ put_32bits(&tempbuf[4], chunks_this_index); /* nEntriesInUse */
+ put_32bits(&tempbuf[8], chunkid); /* dwChunkId */
+ put_64bits(&tempbuf[12], currentbase); /* qwBaseOffset */
+
+ /* now fill in the indexes */
+ chunks_this_index = 0;
+ for (chunknum = 0; chunknum < stream->chunks; chunknum++)
+ if (stream->chunk[chunknum].offset >= currentbase && stream->chunk[chunknum].offset < currentend)
+ {
+ put_32bits(&tempbuf[24 + 8 * chunks_this_index + 0], stream->chunk[chunknum].offset + 8 - currentbase);
+ put_32bits(&tempbuf[24 + 8 * chunks_this_index + 4], stream->chunk[chunknum].length - 8);
+ bytes_this_index += stream->chunk[chunknum].length;
+ chunks_this_index++;
+ }
+
+ /* write the offset of this index to the master table */
+ put_64bits(&buffer[24 + 16 * master_entries + 0], file->writeoffs);
+ put_32bits(&buffer[24 + 16 * master_entries + 8], 24 + 8 * chunks_this_index + 8);
+ if (stream->type == STREAMTYPE_VIDS)
+ put_32bits(&buffer[24 + 16 * master_entries + 12], chunks_this_index);
+ else if (stream->type == STREAMTYPE_AUDS)
+ put_32bits(&buffer[24 + 16 * master_entries + 12], bytes_this_index / ((stream->samplebits / 8) * stream->channels));
+ master_entries++;
+
+ /* write the index */
+ avierr = chunk_write(file, indexchunkid, tempbuf, 24 + 8 * chunks_this_index);
+ free(tempbuf);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+ }
+ }
+
+ /* build up the master index */
+ if (master_entries != 0)
+ {
+ put_16bits(&buffer[0], 4); /* wLongsPerEntry */
+ buffer[2] = 0; /* bIndexSubType */
+ buffer[3] = AVI_INDEX_OF_INDEXES; /* bIndexType */
+ put_32bits(&buffer[4], master_entries); /* nEntriesInUse */
+ put_32bits(&buffer[8], chunkid); /* dwChunkId */
+ }
+
+ /* (over)write the chunk */
+ return chunk_overwrite(file, (master_entries == 0) ? CHUNKTYPE_JUNK : CHUNKTYPE_INDX, buffer, sizeof(buffer), &stream->saved_indx_offset, initial_write);
+}
+
+
+/*-------------------------------------------------
+ write_idx1_chunk - write the idx1 chunk
+-------------------------------------------------*/
+
+static avi_error write_idx1_chunk(avi_file *file)
+{
+ UINT32 tempbuflength = compute_idx1_size(file) - 8;
+ UINT32 curchunk[2] = { 0 };
+ UINT32 curoffset;
+ avi_error avierr;
+ UINT8 *tempbuf;
+
+ /* allocate a temporary buffer */
+ tempbuf = malloc(tempbuflength);
+ if (tempbuf == NULL)
+ return AVIERR_NO_MEMORY;
+
+ /* fill it in */
+ for (curoffset = 0; curoffset < tempbuflength; curoffset += 16)
+ {
+ UINT64 minoffset = ~(UINT64)0;
+ int strnum, minstr = 0;
+
+ /* determine which stream has the next chunk */
+ for (strnum = 0; strnum < file->streams; strnum++)
+ if (curchunk[strnum] < file->stream[strnum].chunks && file->stream[strnum].chunk[curchunk[strnum]].offset < minoffset)
+ {
+ minoffset = file->stream[strnum].chunk[curchunk[strnum]].offset;
+ minstr = strnum;
+ }
+
+ /* make an entry for this index */
+ put_32bits(&tempbuf[curoffset + 0], get_chunkid_for_stream(file, &file->stream[minstr]));
+ put_32bits(&tempbuf[curoffset + 4], 0x0010 /* AVIIF_KEYFRAME */);
+ put_32bits(&tempbuf[curoffset + 8], minoffset - (file->saved_movi_offset + 8));
+ put_32bits(&tempbuf[curoffset + 12], file->stream[minstr].chunk[curchunk[minstr]].length - 8);
+
+ /* advance the chunk counter for this stream */
+ curchunk[minstr]++;
+ }
+
+ /* write the chunk */
+ avierr = chunk_write(file, CHUNKTYPE_IDX1, tempbuf, tempbuflength);
+ free(tempbuf);
+ return avierr;
+}
+
+
+/*-------------------------------------------------
+ soundbuf_initialize - initialize the sound
+ buffering system
+-------------------------------------------------*/
+
+static avi_error soundbuf_initialize(avi_file *file)
+{
+ avi_stream *audio = get_audio_stream(file, 0, NULL);
+ avi_stream *video = get_video_stream(file);
+
+ /* we require a video stream */
+ if (video == NULL)
+ return AVIERR_UNSUPPORTED_VIDEO_FORMAT;
+
+ /* skip if no audio stream */
+ if (audio == NULL)
+ return AVIERR_NONE;
+
+ /* determine the number of samples we want in our buffer; 2 seconds should be enough */
+ file->soundbuf_samples = file->info.audio_samplerate * SOUND_BUFFER_MSEC / 1000;
+
+ /* allocate a buffer */
+ file->soundbuf = malloc(file->soundbuf_samples * file->info.audio_channels * sizeof(file->soundbuf[0]));
+ if (file->soundbuf == NULL)
+ return AVIERR_NO_MEMORY;
+ memset(file->soundbuf, 0, file->soundbuf_samples * file->info.audio_channels * sizeof(file->soundbuf[0]));
+
+ /* determine the number of frames to be ahead (0.75secs) */
+ file->soundbuf_frames = ((UINT64)video->rate * 75) / ((UINT64)video->scale * 100) + 1;
+ return AVIERR_NONE;
+}
+
+
+/*-------------------------------------------------
+ soundbuf_write_chunk - write the initial
+ chunk data
+-------------------------------------------------*/
+
+static avi_error soundbuf_write_chunk(avi_file *file, UINT32 framenum)
+{
+ avi_stream *stream = get_audio_stream(file, 0, NULL);
+ avi_error avierr;
+ UINT32 chunknum;
+ UINT32 length;
+
+ /* skip if no audio stream */
+ if (stream == NULL)
+ return AVIERR_NONE;
+
+ /* determine the length of this chunk */
+ if (framenum == 0)
+ length = framenum_to_samplenum(file, file->soundbuf_frames);
+ else
+ length = framenum_to_samplenum(file, framenum + 1 + file->soundbuf_frames) - framenum_to_samplenum(file, framenum + file->soundbuf_frames);
+ length *= stream->channels * sizeof(INT16);
+
+ /* set the info for this new chunk */
+ chunknum = stream->chunks;
+ avierr = set_stream_chunk_info(stream, chunknum, file->writeoffs, length + 8);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+
+ /* then do the initial write */
+ return chunk_overwrite(file, get_chunkid_for_stream(file, stream), file->soundbuf, length, &stream->chunk[chunknum].offset, TRUE);
+}
+
+
+/*-------------------------------------------------
+ soundbuf_flush - flush data from the sound
+ buffers
+-------------------------------------------------*/
+
+static avi_error soundbuf_flush(avi_file *file, int only_flush_full)
+{
+ avi_stream *stream = get_audio_stream(file, 0, NULL);
+ INT32 channelsamples = file->soundbuf_samples;
+ INT32 processedsamples = 0;
+ UINT32 bytes_per_sample;
+ UINT32 finalchunks;
+ avi_error avierr;
+ UINT32 chunknum;
+ UINT32 chunkid;
+ int channel;
+
+ /* skip if no stream */
+ if (stream == NULL)
+ return AVIERR_NONE;
+
+ /* get the chunk ID for this stream */
+ chunkid = get_chunkid_for_stream(file, stream);
+ bytes_per_sample = stream->channels * sizeof(INT16);
+ finalchunks = stream->chunks;
+
+ /* find out how many samples we've accumulated */
+ for (channel = 0; channel < stream->channels; channel++)
+ channelsamples = MIN(channelsamples, file->soundbuf_chansamples[channel]);
+
+ /* loop over pending sound chunks */
+ for (chunknum = file->soundbuf_chunks; chunknum < stream->chunks; chunknum++)
+ {
+ avi_chunk_list *chunk = &stream->chunk[chunknum];
+ UINT32 chunksamples = (chunk->length - 8) / bytes_per_sample;
+
+ /* stop if we don't have enough to satisfy this chunk */
+ if (only_flush_full && channelsamples < chunksamples)
+ break;
+
+ /* if we don't have all the samples we need, pad with 0's */
+ if (channelsamples > 0 && channelsamples < chunksamples)
+ {
+ if (processedsamples + chunksamples > file->soundbuf_samples)
+ return AVIERR_EXCEEDED_SOUND_BUFFER;
+ memset(&file->soundbuf[(processedsamples + channelsamples) * stream->channels], 0, (chunksamples - channelsamples) * bytes_per_sample);
+ }
+
+ /* if we're completely out of samples, clear the buffer entirely and use the end */
+ else if (channelsamples <= 0)
+ {
+ processedsamples = file->soundbuf_samples - chunksamples;
+ memset(&file->soundbuf[processedsamples * stream->channels], 0, chunksamples * bytes_per_sample);
+ chunkid = CHUNKTYPE_JUNK;
+ finalchunks--;
+ }
+
+ /* copy the sample data in */
+ avierr = chunk_overwrite(file, chunkid, &file->soundbuf[processedsamples * stream->channels], chunk->length - 8, &chunk->offset, FALSE);
+ if (avierr != AVIERR_NONE)
+ return avierr;
+
+ /* add up the samples */
+ if (channelsamples > 0)
+ file->info.audio_numsamples = stream->samples += channelsamples;
+
+ /* advance past those */
+ processedsamples += chunksamples;
+ channelsamples -= chunksamples;
+ channelsamples = MAX(0, channelsamples);
+ }
+
+ /* if we have a non-zero offset, shift everything down */
+ if (processedsamples > 0)
+ {
+ /* first account for the samples we processed */
+ memmove(&file->soundbuf[0], &file->soundbuf[processedsamples * stream->channels], (file->soundbuf_samples - processedsamples) * bytes_per_sample);
+ for (channel = 0; channel < stream->channels; channel++)
+ file->soundbuf_chansamples[channel] -= processedsamples;
+ }
+
+ /* update the final chunk count */
+ if (!only_flush_full)
+ stream->chunks = finalchunks;
+
+ /* account for flushed chunks */
+ file->soundbuf_chunks = chunknum;
+ return AVIERR_NONE;
+}
+
+
+/*-------------------------------------------------
+ yuv_decompress_to_yuy16 - decompress a YUV
+ encoded frame to a YUY16 bitmap
+-------------------------------------------------*/
+
+static avi_error yuv_decompress_to_yuy16(avi_stream *stream, const UINT8 *data, UINT32 numbytes, bitmap_t *bitmap)
+{
+ const UINT16 *dataend = (const UINT16 *)(data + numbytes);
+ int x, y;
+
+ /* compressed video */
+ for (y = 0; y < stream->height; y++)
+ {
+ const UINT16 *source = (const UINT16 *)data + y * stream->width;
+ UINT16 *dest = (UINT16 *)bitmap->base + y * bitmap->rowpixels;
+
+ /* switch off the compression */
+ switch (stream->format)
+ {
+ case FORMAT_UYVY:
+ for (x = 0; x < stream->width && source < dataend; x++)
+ *dest++ = *source++;
+ break;
+
+ case FORMAT_VYUY:
+ case FORMAT_YUY2:
+ for (x = 0; x < stream->width && source < dataend; x++)
+ {
+ UINT16 pix = *source++;
+ *dest++ = (pix >> 8) | (pix << 8);
+ }
+ break;
+ }
+ }
+
+ return AVIERR_NONE;
+}
+
+
+/*-------------------------------------------------
+ yuy16_compress_to_yuy - "compress" a YUY16
+ bitmap to a YUV encoded frame
+-------------------------------------------------*/
+
+static avi_error yuy16_compress_to_yuy(avi_stream *stream, const bitmap_t *bitmap, UINT8 *data, UINT32 numbytes)
+{
+ const UINT16 *dataend = (const UINT16 *)(data + numbytes);
+ int x, y;
+
+ /* compressed video */
+ for (y = 0; y < stream->height; y++)
+ {
+ const UINT16 *source = (UINT16 *)bitmap->base + y * bitmap->rowpixels;
+ UINT16 *dest = (UINT16 *)data + y * stream->width;
+
+ /* switch off the compression */
+ switch (stream->format)
+ {
+ case FORMAT_UYVY:
+ for (x = 0; x < stream->width && dest < dataend; x++)
+ *dest++ = *source++;
+ break;
+
+ case FORMAT_VYUY:
+ case FORMAT_YUY2:
+ for (x = 0; x < stream->width && source < dataend; x++)
+ {
+ UINT16 pix = *source++;
+ *dest++ = (pix >> 8) | (pix << 8);
+ }
+ break;
+ }
+ }
+
+ return AVIERR_NONE;
+}
+
+
+/*-------------------------------------------------
+ huffyuv_extract_tables - extract HuffYUV
+ tables
+-------------------------------------------------*/
+
+static avi_error huffyuv_extract_tables(avi_stream *stream, const UINT8 *chunkdata, UINT32 size)
+{
+ const UINT8 *chunkend = chunkdata + size;
+ avi_error avierr = AVIERR_NONE;
+ int tabnum;
+
+ /* allocate memory for the data */
+ stream->huffyuv = malloc(sizeof(*stream->huffyuv));
+ if (stream->huffyuv == NULL)
+ {
+ avierr = AVIERR_NO_MEMORY;
+ goto error;
+ }
+
+ /* extract predictor information */
+ if (&chunkdata[40] >= chunkend)
+ return AVIERR_INVALID_DATA;
+ stream->huffyuv->predictor = chunkdata[40];
+
+ /* make sure it's the left predictor */
+ if ((stream->huffyuv->predictor & ~HUFFYUV_PREDICT_DECORR) != HUFFYUV_PREDICT_LEFT)
+ return AVIERR_UNSUPPORTED_VIDEO_FORMAT;
+
+ /* make sure it's 16bpp YUV data */
+ if (chunkdata[41] != 16)
+ return AVIERR_UNSUPPORTED_VIDEO_FORMAT;
+ chunkdata += 44;
+
+ /* loop over tables */
+ for (tabnum = 0; tabnum < 3; tabnum++)
+ {
+ huffyuv_table *table = &stream->huffyuv->table[tabnum];
+ UINT32 curbits, bitadd;
+ UINT16 bitsat16 = 0;
+ int offset = 0, bits;
+
+ /* loop until we populate the whole table */
+ while (offset < 256)
+ {
+ int data, shift, count, i;
+
+ /* extract the next run */
+ if (chunkdata >= chunkend)
+ {
+ avierr = AVIERR_INVALID_DATA;
+ goto error;
+ }
+ data = *chunkdata++;
+ shift = data & 0x1f;
+ count = data >> 5;
+
+ /* zero count means next whole byte is a count */
+ if (count == 0)
+ {
+ if (chunkdata >= chunkend)
+ {
+ avierr = AVIERR_INVALID_DATA;
+ goto error;
+ }
+ count = *chunkdata++;
+ }
+ for (i = 0; i < count; i++)
+ table->shift[offset++] = shift;
+ }
+
+ /* now determine bit patterns and masks */
+ curbits = 0;
+ for (bits = 31; bits >= 0; bits--)
+ {
+ bitadd = 1 << (32 - bits);
+
+ /* make sure we've cleared out all the bits below */
+ if ((curbits & (bitadd - 1)) != 0)
+ {
+ avierr = AVIERR_INVALID_DATA;
+ goto error;
+ }
+
+ /* find all entries with this shift count and assign them */
+ for (offset = 0; offset < 256; offset++)
+ if (table->shift[offset] == bits)
+ {
+ table->bits[offset] = curbits;
+ table->mask[offset] = ~(bitadd - 1);
+ curbits += bitadd;
+ }
+
+ /* remember the bit pattern when we complete all the 17-bit codes */
+ if (bits == 17)
+ bitsat16 = curbits >> 16;
+ }
+
+ /* allocate the number of extra lookup tables we need */
+ if (bitsat16 > 0)
+ {
+ table->extralookup = malloc(bitsat16 * 65536 * sizeof(table->extralookup[0]));
+ if (table->extralookup == NULL)
+ {
+ avierr = AVIERR_NO_MEMORY;
+ goto error;
+ }
+ for (offset = 0; offset < bitsat16; offset++)
+ table->baselookup[offset] = (offset << 8) | 0;
+ }
+
+ /* then create lookup tables */
+ for (offset = 0; offset < 256; offset++)
+ if (table->shift[offset] > 16)
+ {
+ UINT16 *tablebase = table->extralookup + (table->bits[offset] >> 16) * 65536;
+ UINT32 start = table->bits[offset] & 0xffff;
+ UINT32 end = start + ((1 << (32 - table->shift[offset])) - 1);
+ while (start <= end)
+ tablebase[start++] = (offset << 8) | (table->shift[offset] - 16);
+ }
+ else if (table->shift[offset] > 0)
+ {
+ UINT32 start = table->bits[offset] >> 16;
+ UINT32 end = start + ((1 << (16 - table->shift[offset])) - 1);
+ while (start <= end)
+ table->baselookup[start++] = (offset << 8) | table->shift[offset];
+ }
+ }
+
+error:
+ if (avierr != AVIERR_NONE && stream->huffyuv != NULL)
+ {
+ free(stream->huffyuv);
+ stream->huffyuv = NULL;
+ }
+ return avierr;
+}
+
+
+/*-------------------------------------------------
+ huffyuv_decompress_to_yuy16 - decompress a
+ HuffYUV-encoded frame to a YUY16 bitmap
+-------------------------------------------------*/
+
+static avi_error huffyuv_decompress_to_yuy16(avi_stream *stream, const UINT8 *data, UINT32 numbytes, bitmap_t *bitmap)
+{
+ huffyuv_data *huffyuv = stream->huffyuv;
+ int prevlines = (stream->height > 288) ? 2 : 1;
+ UINT8 lastprevy = 0, lastprevcb = 0, lastprevcr = 0;
+ UINT8 lasty = 0, lastcb = 0, lastcr = 0;
+ UINT8 bitsinbuffer = 0;
+ UINT32 bitbuffer = 0;
+ UINT32 dataoffs = 0;
+ int x, y;
+
+ /* compressed video */
+ for (y = 0; y < stream->height; y++)
+ {
+ UINT16 *dest = (UINT16 *)bitmap->base + y * bitmap->rowpixels;
+
+ /* handle the first four bytes independently */
+ x = 0;
+ if (y == 0)
+ {
+ /* first DWORD is stored as YUY2 */
+ lasty = data[dataoffs++];
+ lastcb = data[dataoffs++];
+ *dest++ = (lasty << 8) | lastcb;
+ lasty = data[dataoffs++];
+ lastcr = data[dataoffs++];
+ *dest++ = (lasty << 8) | lastcr;
+ x = 2;
+ }
+
+ /* loop over pixels */
+ for ( ; x < stream->width; x++)
+ {
+ huffyuv_table *ytable = &huffyuv->table[0];
+ huffyuv_table *ctable = &huffyuv->table[1 + (x & 1)];
+ UINT16 pixel, huffdata;
+ int shift;
+
+ /* fill up the buffer; they store little-endian DWORDs, so we XOR with 3 */
+ while (bitsinbuffer <= 24 && dataoffs < numbytes)
+ {
+ bitbuffer |= data[dataoffs++ ^ 3] << (24 - bitsinbuffer);
+ bitsinbuffer += 8;
+ }
+
+ /* look up the Y component */
+ huffdata = ytable->baselookup[bitbuffer >> 16];
+ shift = huffdata & 0xff;
+ if (shift == 0)
+ {
+ bitsinbuffer -= 16;
+ bitbuffer <<= 16;
+
+ /* fill up the buffer; they store little-endian DWORDs, so we XOR with 3 */
+ while (bitsinbuffer <= 24 && dataoffs < numbytes)
+ {
+ bitbuffer |= data[dataoffs++ ^ 3] << (24 - bitsinbuffer);
+ bitsinbuffer += 8;
+ }
+
+ huffdata = ytable->extralookup[(huffdata >> 8) * 65536 + (bitbuffer >> 16)];
+ shift = huffdata & 0xff;
+ }
+ bitsinbuffer -= shift;
+ bitbuffer <<= shift;
+ pixel = huffdata & 0xff00;
+
+ /* fill up the buffer; they store little-endian DWORDs, so we XOR with 3 */
+ while (bitsinbuffer <= 24 && dataoffs < numbytes)
+ {
+ bitbuffer |= data[dataoffs++ ^ 3] << (24 - bitsinbuffer);
+ bitsinbuffer += 8;
+ }
+
+ /* look up the Cb/Cr component */
+ huffdata = ctable->baselookup[bitbuffer >> 16];
+ shift = huffdata & 0xff;
+ if (shift == 0)
+ {
+ bitsinbuffer -= 16;
+ bitbuffer <<= 16;
+
+ /* fill up the buffer; they store little-endian DWORDs, so we XOR with 3 */
+ while (bitsinbuffer <= 24 && dataoffs < numbytes)
+ {
+ bitbuffer |= data[dataoffs++ ^ 3] << (24 - bitsinbuffer);
+ bitsinbuffer += 8;
+ }
+
+ huffdata = ctable->extralookup[(huffdata >> 8) * 65536 + (bitbuffer >> 16)];
+ shift = huffdata & 0xff;
+ }
+ bitsinbuffer -= shift;
+ bitbuffer <<= shift;
+ *dest++ = pixel | (huffdata >> 8);
+ }
+ }
+
+ /* apply deltas */
+ lasty = lastcb = lastcr = 0;
+ lastprevy = lastprevcb = lastprevcr = 0;
+ for (y = 0; y < stream->height; y++)
+ {
+ UINT16 *dest = (UINT16 *)bitmap->base + y * bitmap->rowpixels;
+ UINT16 *prevrow = dest - prevlines * bitmap->rowpixels;
+
+ /* left predict or gradient predict */
+ if ((huffyuv->predictor & ~HUFFYUV_PREDICT_DECORR) == HUFFYUV_PREDICT_LEFT ||
+ ((huffyuv->predictor & ~HUFFYUV_PREDICT_DECORR) == HUFFYUV_PREDICT_GRADIENT))
+ {
+ /* first do left deltas */
+ for (x = 0; x < stream->width; x += 2)
+ {
+ UINT16 pixel0 = dest[x + 0];
+ UINT16 pixel1 = dest[x + 1];
+
+ lasty += pixel0 >> 8;
+ lastcb += pixel0;
+ dest[x + 0] = (lasty << 8) | lastcb;
+
+ lasty += pixel1 >> 8;
+ lastcr += pixel1;
+ dest[x + 1] = (lasty << 8) | lastcr;
+ }
+
+ /* for gradient, we then add in the previous row */
+ if ((huffyuv->predictor & ~HUFFYUV_PREDICT_DECORR) == HUFFYUV_PREDICT_GRADIENT && y >= prevlines)
+ for (x = 0; x < stream->width; x++)
+ {
+ UINT16 curpix = dest[x];
+ UINT16 prevpix = prevrow[x];
+ UINT8 ysum = (curpix >> 8) + (prevpix >> 8);
+ UINT8 csum = curpix + prevpix;
+ dest[x] = (ysum << 8) | csum;
+ }
+ }
+
+ /* median predict on rows > 0 */
+ else if ((huffyuv->predictor & ~HUFFYUV_PREDICT_DECORR) == HUFFYUV_PREDICT_MEDIAN && y >= prevlines)
+ {
+ for (x = 0; x < stream->width; x += 2)
+ {
+ UINT16 prevpixel0 = prevrow[x + 0];
+ UINT16 prevpixel1 = prevrow[x + 1];
+ UINT16 pixel0 = dest[x + 0];
+ UINT16 pixel1 = dest[x + 1];
+ UINT8 a, b, c;
+
+ /* compute previous, above, and (prev + above - above-left) */
+ a = lasty;
+ b = prevpixel0 >> 8;
+ c = lastprevy;
+ lastprevy = b;
+ if (a > b) { UINT8 tmp = a; a = b; b = tmp; }
+ if (a > c) { UINT8 tmp = a; a = c; c = tmp; }
+ if (b > c) { UINT8 tmp = b; b = c; c = tmp; }
+ lasty = (pixel0 >> 8) + b;
+
+ /* compute previous, above, and (prev + above - above-left) */
+ a = lastcb;
+ b = prevpixel0 & 0xff;
+ c = lastprevcb;
+ lastprevcb = b;
+ if (a > b) { UINT8 tmp = a; a = b; b = tmp; }
+ if (a > c) { UINT8 tmp = a; a = c; c = tmp; }
+ if (b > c) { UINT8 tmp = b; b = c; c = tmp; }
+ lastcb = (pixel0 & 0xff) + b;
+ dest[x + 0] = (lasty << 8) | lastcb;
+
+ /* compute previous, above, and (prev + above - above-left) */
+ a = lasty;
+ b = prevpixel1 >> 8;
+ c = lastprevy;
+ lastprevy = b;
+ if (a > b) { UINT8 tmp = a; a = b; b = tmp; }
+ if (a > c) { UINT8 tmp = a; a = c; c = tmp; }
+ if (b > c) { UINT8 tmp = b; b = c; c = tmp; }
+ lasty = (pixel1 >> 8) + b;
+
+ /* compute previous, above, and (prev + above - above-left) */
+ a = lastcr;
+ b = prevpixel1 & 0xff;
+ c = lastprevcr;
+ lastprevcr = b;
+ if (a > b) { UINT8 tmp = a; a = b; b = tmp; }
+ if (a > c) { UINT8 tmp = a; a = c; c = tmp; }
+ if (b > c) { UINT8 tmp = b; b = c; c = tmp; }
+ lastcr = (pixel1 & 0xff) + b;
+ dest[x + 1] = (lasty << 8) | lastcr;
+ }
+ }
+ }
+
+ return AVIERR_NONE;
+}
+
+
+static void u64toa(UINT64 val, char *output)
+{
+ UINT32 lo = (UINT32)(val & 0xffffffff);
+ UINT32 hi = (UINT32)(val >> 32);
+ if (hi != 0)
+ sprintf(output, "%X%08X", hi, lo);
+ else
+ sprintf(output, "%X", lo);
+}
+
+
+/*-------------------------------------------------
+ printf_chunk_recursive - print information
+ about a chunk recursively
+-------------------------------------------------*/
+
+static void printf_chunk_recursive(avi_file *file, avi_chunk *container, int indent)
+{
+ char size[20], offset[20];
+ avi_chunk curchunk;
+ int avierr;
+
+ /* iterate over chunks in this container */
+ for (avierr = get_first_chunk(file, container, &curchunk); avierr == AVIERR_NONE; avierr = get_next_chunk(file, container, &curchunk))
+ {
+ UINT32 chunksize = curchunk.size;
+ int recurse = FALSE;
+
+ u64toa(curchunk.size, size);
+ u64toa(curchunk.offset, offset);
+ printf("%*schunk = %c%c%c%c, size=%s (%s)\n", indent, "",
+ (UINT8)(curchunk.type >> 0),
+ (UINT8)(curchunk.type >> 8),
+ (UINT8)(curchunk.type >> 16),
+ (UINT8)(curchunk.type >> 24),
+ size, offset);
+
+ /* certain chunks are just containers; recurse into them */
+ switch (curchunk.type)
+ {
+ /* basic containers */
+ case CHUNKTYPE_RIFF:
+ case CHUNKTYPE_LIST:
+ printf("%*stype = %c%c%c%c\n", indent, "",
+ (UINT8)(curchunk.listtype >> 0),
+ (UINT8)(curchunk.listtype >> 8),
+ (UINT8)(curchunk.listtype >> 16),
+ (UINT8)(curchunk.listtype >> 24));
+ recurse = TRUE;
+ chunksize = 0;
+ break;
+ }
+
+ /* print data within the chunk */
+ if (chunksize > 0 && curchunk.size < 1024 * 1024)
+ {
+ UINT8 *data = NULL;
+ int i;
+
+ /* read the data for a chunk */
+ avierr = read_chunk_data(file, &curchunk, &data);
+ if (avierr == AVIERR_NONE)
+ {
+ int bytes = MIN(512, chunksize);
+ for (i = 0; i < bytes; i++)
+ {
+ if (i % 16 == 0) printf("%*s ", indent, "");
+ printf("%02X ", data[i]);
+ if (i % 16 == 15) printf("\n");
+ }
+ if (chunksize % 16 != 0) printf("\n");
+ free(data);
+ }
+ }
+
+ /* if we're recursing, dive down */
+ if (recurse)
+ printf_chunk_recursive(file, &curchunk, indent + 4);
+ }
+
+ /* if we didn't get a legitimate error, indicate that */
+ if (avierr != AVIERR_END)
+ printf("[chunk error %d]\n", avierr);
+}
diff --git a/src/lib/util/aviio.h b/src/lib/util/aviio.h
new file mode 100644
index 00000000000..41f9eba826d
--- /dev/null
+++ b/src/lib/util/aviio.h
@@ -0,0 +1,123 @@
+/***************************************************************************
+
+ aviio.h
+
+ AVI movie format parsing helpers.
+
+ Copyright (c) 1996-2007, Nicola Salmoria and the MAME Team.
+ Visit http://mamedev.org for licensing and usage restrictions.
+
+***************************************************************************/
+
+#ifndef __AVIIO_H__
+
+#include "osdcore.h"
+#include "bitmap.h"
+
+
+/***************************************************************************
+ CONSTANTS
+***************************************************************************/
+
+enum _avi_error
+{
+ AVIERR_NONE = 0,
+ AVIERR_END,
+ AVIERR_INVALID_DATA,
+ AVIERR_NO_MEMORY,
+ AVIERR_READ_ERROR,
+ AVIERR_WRITE_ERROR,
+ AVIERR_STACK_TOO_DEEP,
+ AVIERR_UNSUPPORTED_FEATURE,
+ AVIERR_CANT_OPEN_FILE,
+ AVIERR_INCOMPATIBLE_AUDIO_STREAMS,
+ AVIERR_INVALID_SAMPLERATE,
+ AVIERR_INVALID_STREAM,
+ AVIERR_INVALID_FRAME,
+ AVIERR_INVALID_BITMAP,
+ AVIERR_UNSUPPORTED_VIDEO_FORMAT,
+ AVIERR_UNSUPPORTED_AUDIO_FORMAT,
+ AVIERR_EXCEEDED_SOUND_BUFFER
+};
+typedef enum _avi_error avi_error;
+
+
+enum _avi_datatype
+{
+ AVIDATA_VIDEO,
+ AVIDATA_AUDIO_CHAN0,
+ AVIDATA_AUDIO_CHAN1,
+ AVIDATA_AUDIO_CHAN2,
+ AVIDATA_AUDIO_CHAN3,
+ AVIDATA_AUDIO_CHAN4,
+ AVIDATA_AUDIO_CHAN5,
+ AVIDATA_AUDIO_CHAN6,
+ AVIDATA_AUDIO_CHAN7
+};
+typedef enum _avi_datatype avi_datatype;
+
+
+
+/***************************************************************************
+ MACROS
+***************************************************************************/
+
+#define AVI_FOURCC(a,b,c,d) ((a) | ((b) << 8) | ((c) << 16) | ((d) << 24))
+
+#define FORMAT_UYVY AVI_FOURCC('U','Y','V','Y')
+#define FORMAT_VYUY AVI_FOURCC('V','Y','U','Y')
+#define FORMAT_YUY2 AVI_FOURCC('Y','U','Y','2')
+#define FORMAT_HFYU AVI_FOURCC('H','F','Y','U')
+
+
+
+/***************************************************************************
+ TYPE DEFINITIONS
+***************************************************************************/
+
+typedef struct _avi_file avi_file;
+
+
+typedef struct _avi_movie_info avi_movie_info;
+struct _avi_movie_info
+{
+ UINT32 video_format; /* format of video data */
+ UINT32 video_timescale; /* timescale for video data */
+ UINT32 video_sampletime; /* duration of a single video sample (frame) */
+ UINT32 video_numsamples; /* total number of video samples */
+ UINT32 video_width; /* width of the video */
+ UINT32 video_height; /* height of the video */
+ UINT32 video_depth; /* depth of the video */
+
+ UINT32 audio_format; /* format of audio data */
+ UINT32 audio_timescale; /* timescale for audio data */
+ UINT32 audio_sampletime; /* duration of a single audio sample */
+ UINT32 audio_numsamples; /* total number of audio samples */
+ UINT32 audio_channels; /* number of audio channels */
+ UINT32 audio_samplebits; /* number of bits per channel */
+ UINT32 audio_samplerate; /* sample rate of audio */
+};
+
+
+
+/***************************************************************************
+ PROTOTYPES
+***************************************************************************/
+
+avi_error avi_open(const char *filename, avi_file **file);
+avi_error avi_create(const char *filename, const avi_movie_info *info, avi_file **file);
+avi_error avi_close(avi_file *file);
+
+void avi_printf_chunks(avi_file *file);
+const char *avi_error_string(avi_error err);
+
+const avi_movie_info *avi_get_movie_info(avi_file *file);
+UINT32 avi_first_sample_in_frame(avi_file *file, UINT32 framenum);
+
+avi_error avi_read_video_frame_yuy16(avi_file *file, UINT32 framenum, bitmap_t *bitmap);
+avi_error avi_read_sound_samples(avi_file *file, int channel, UINT32 firstsample, UINT32 numsamples, INT16 *output);
+
+avi_error avi_append_video_frame_yuy16(avi_file *file, const bitmap_t *bitmap);
+avi_error avi_append_sound_samples(avi_file *file, int channel, const INT16 *samples, UINT32 numsamples);
+
+#endif
diff --git a/src/lib/util/bitmap.c b/src/lib/util/bitmap.c
new file mode 100644
index 00000000000..a7e5d5e46f8
--- /dev/null
+++ b/src/lib/util/bitmap.c
@@ -0,0 +1,263 @@
+/***************************************************************************
+
+ bitmap.c
+
+ Core bitmap routines.
+
+ Copyright (c) 1996-2007, Nicola Salmoria and the MAME Team.
+ Visit http://mamedev.org for licensing and usage restrictions.
+
+***************************************************************************/
+
+#include "bitmap.h"
+
+
+
+/***************************************************************************
+ BITMAP ALLOCATION/CONFIGURATION
+***************************************************************************/
+
+/*-------------------------------------------------
+ bitmap_alloc -- allocate memory for a new
+ bitmap of the given format
+-------------------------------------------------*/
+
+bitmap_t *bitmap_alloc(int width, int height, bitmap_format format)
+{
+ return bitmap_alloc_slop(width, height, 0, 0, format);
+}
+
+
+/*-------------------------------------------------
+ bitmap_alloc_slop -- allocate a new bitmap with
+ additional slop on the borders
+-------------------------------------------------*/
+
+bitmap_t *bitmap_alloc_slop(int width, int height, int xslop, int yslop, bitmap_format format)
+{
+ int bpp = bitmap_format_to_bpp(format);
+ size_t allocbytes;
+ bitmap_t *bitmap;
+ int rowpixels;
+
+ /* fail if invalid format */
+ if (bpp == 0)
+ return NULL;
+
+ /* allocate the bitmap itself */
+ bitmap = malloc(sizeof(*bitmap));
+ if (bitmap == NULL)
+ return NULL;
+ memset(bitmap, 0, sizeof(*bitmap));
+
+ /* round the width to a multiple of 8 and add some padding */
+ rowpixels = (width + 2 * xslop + 7) & ~7;
+
+ /* allocate memory for the bitmap itself */
+ allocbytes = rowpixels * (height + 2 * yslop) * bpp / 8;
+ bitmap->alloc = malloc(allocbytes);
+ if (bitmap->alloc == NULL)
+ {
+ free(bitmap);
+ return NULL;
+ }
+
+ /* clear to 0 by default */
+ memset(bitmap->alloc, 0, allocbytes);
+
+ /* fill in the data */
+ bitmap->format = format;
+ bitmap->width = width;
+ bitmap->height = height;
+ bitmap->bpp = bpp;
+ bitmap->rowpixels = rowpixels;
+ bitmap->base = (UINT8 *)bitmap->alloc + (rowpixels * yslop + xslop) * (bpp / 8);
+
+ return bitmap;
+}
+
+
+/*-------------------------------------------------
+ bitmap_wrap -- wrap an existing memory buffer
+ as a bitmap
+-------------------------------------------------*/
+
+bitmap_t *bitmap_wrap(void *base, int width, int height, int rowpixels, bitmap_format format)
+{
+ int bpp = bitmap_format_to_bpp(format);
+ bitmap_t *bitmap;
+
+ /* fail if invalid format */
+ if (bpp == 0)
+ return NULL;
+
+ /* allocate memory */
+ bitmap = malloc(sizeof(*bitmap));
+ if (bitmap == NULL)
+ return NULL;
+ memset(bitmap, 0, sizeof(*bitmap));
+
+ /* fill in the data */
+ bitmap->format = format;
+ bitmap->width = width;
+ bitmap->height = height;
+ bitmap->bpp = bpp;
+ bitmap->rowpixels = rowpixels;
+ bitmap->base = base;
+
+ return bitmap;
+}
+
+
+/*-------------------------------------------------
+ bitmap_set_palette -- associate a palette with
+ a bitmap
+-------------------------------------------------*/
+
+void bitmap_set_palette(bitmap_t *bitmap, palette_t *palette)
+{
+ /* first dereference any existing palette */
+ if (bitmap->palette != NULL)
+ {
+ palette_deref(bitmap->palette);
+ bitmap->palette = NULL;
+ }
+
+ /* then reference any new palette */
+ if (palette != NULL)
+ {
+ palette_ref(palette);
+ bitmap->palette = palette;
+ }
+}
+
+
+/*-------------------------------------------------
+ bitmap_free -- release memory allocated for
+ a bitmap
+-------------------------------------------------*/
+
+void bitmap_free(bitmap_t *bitmap)
+{
+ /* dereference the palette */
+ if (bitmap->palette != NULL)
+ palette_deref(bitmap->palette);
+
+ /* free any allocated memory */
+ if (bitmap->alloc != NULL)
+ free(bitmap->alloc);
+
+ /* free the bitmap */
+ free(bitmap);
+}
+
+
+
+/***************************************************************************
+ BITMAP DRAWING
+***************************************************************************/
+
+/*-------------------------------------------------
+ bitmap_fill -- fill a bitmap with a solid
+ color
+-------------------------------------------------*/
+
+void bitmap_fill(bitmap_t *dest, const rectangle *cliprect, UINT32 color)
+{
+ rectangle fill;
+ int x, y;
+
+ /* default to the whole bitmap */
+ fill.min_x = fill.min_y = 0;
+ fill.max_x = dest->width - 1;
+ fill.max_y = dest->height - 1;
+
+ /* if we have a cliprect, intersect with that */
+ if (cliprect != NULL)
+ sect_rect(&fill, cliprect);
+
+ /* early out if nothing to do */
+ if (fill.min_x > fill.max_x || fill.min_y > fill.max_y)
+ return;
+
+ /* based on the bpp go from there */
+ switch (dest->bpp)
+ {
+ case 8:
+ /* 8bpp always uses memset */
+ for (y = fill.min_y; y <= fill.max_y; y++)
+ memset(BITMAP_ADDR8(dest, y, fill.min_x), (UINT8)color, fill.max_x + 1 - fill.min_x);
+ break;
+
+ case 16:
+ /* 16bpp can use memset if the bytes are equal */
+ if ((UINT8)(color >> 8) == (UINT8)color)
+ {
+ for (y = fill.min_y; y <= fill.max_y; y++)
+ memset(BITMAP_ADDR16(dest, y, fill.min_x), (UINT8)color, (fill.max_x + 1 - fill.min_x) * 2);
+ }
+ else
+ {
+ for (y = fill.min_y; y <= fill.max_y; y++)
+ {
+ UINT16 *destrow = BITMAP_ADDR16(dest, y, 0);
+ for (x = fill.min_x; x <= fill.max_x; x++)
+ destrow[x] = (UINT16)color;
+ }
+ }
+ break;
+
+ case 32:
+ /* 32bpp can use memset if the bytes are equal */
+ if ((UINT8)(color >> 8) == (UINT8)color && (UINT16)(color >> 16) == (UINT16)color)
+ {
+ for (y = fill.min_y; y <= fill.max_y; y++)
+ memset(BITMAP_ADDR32(dest, y, fill.min_x), (UINT8)color, (fill.max_x + 1 - fill.min_x) * 4);
+ }
+ else
+ {
+ for (y = fill.min_y; y <= fill.max_y; y++)
+ {
+ UINT32 *destrow = BITMAP_ADDR32(dest, y, 0);
+ for (x = fill.min_x; x <= fill.max_x; x++)
+ destrow[x] = (UINT32)color;
+ }
+ }
+ break;
+ }
+}
+
+
+
+/***************************************************************************
+ BITMAP UTILITIES
+***************************************************************************/
+
+/*-------------------------------------------------
+ bitmap_format_to_bpp - given a format, return
+ the bpp
+-------------------------------------------------*/
+
+int bitmap_format_to_bpp(bitmap_format format)
+{
+ /* choose a depth for the format */
+ switch (format)
+ {
+ case BITMAP_FORMAT_INDEXED8:
+ return 8;
+
+ case BITMAP_FORMAT_INDEXED16:
+ case BITMAP_FORMAT_RGB15:
+ case BITMAP_FORMAT_YUY16:
+ return 16;
+
+ case BITMAP_FORMAT_INDEXED32:
+ case BITMAP_FORMAT_RGB32:
+ case BITMAP_FORMAT_ARGB32:
+ return 32;
+
+ default:
+ break;
+ }
+ return 0;
+}
diff --git a/src/lib/util/bitmap.h b/src/lib/util/bitmap.h
new file mode 100644
index 00000000000..5b4a5badbd0
--- /dev/null
+++ b/src/lib/util/bitmap.h
@@ -0,0 +1,169 @@
+/***************************************************************************
+
+ bitmap.h
+
+ Core bitmap routines.
+
+ Copyright (c) 1996-2007, Nicola Salmoria and the MAME Team.
+ Visit http://mamedev.org for licensing and usage restrictions.
+
+***************************************************************************/
+
+#pragma once
+
+#ifndef __BITMAP_H__
+#define __BITMAP_H__
+
+#include "osdcore.h"
+#include "palette.h"
+
+
+/***************************************************************************
+ TYPE DEFINITIONS
+***************************************************************************/
+
+/* bitmap_format describes the various bitmap formats we use */
+enum _bitmap_format
+{
+ BITMAP_FORMAT_INVALID = 0, /* invalid format */
+ BITMAP_FORMAT_INDEXED8, /* 8bpp indexed */
+ BITMAP_FORMAT_INDEXED16, /* 16bpp indexed */
+ BITMAP_FORMAT_INDEXED32, /* 32bpp indexed */
+ BITMAP_FORMAT_RGB15, /* 15bpp 5-5-5 RGB */
+ BITMAP_FORMAT_RGB32, /* 32bpp 8-8-8 RGB */
+ BITMAP_FORMAT_ARGB32, /* 32bpp 8-8-8-8 ARGB */
+ BITMAP_FORMAT_YUY16, /* 16bpp 8-8 Y/Cb, Y/Cr in sequence */
+ BITMAP_FORMAT_LAST
+};
+typedef enum _bitmap_format bitmap_format;
+
+
+/* rectangles describe a bitmap portion */
+typedef struct _rectangle rectangle;
+struct _rectangle
+{
+ int min_x; /* minimum X, or left coordinate */
+ int max_x; /* maximum X, or right coordinate (inclusive) */
+ int min_y; /* minimum Y, or top coordinate */
+ int max_y; /* maximum Y, or bottom coordinate (inclusive) */
+};
+
+
+/* bitmaps describe a rectangular array of pixels */
+typedef struct _bitmap_t bitmap_t;
+struct _bitmap_t
+{
+ void * alloc; /* pointer to allocated pixel memory */
+ void * base; /* pointer to pixel (0,0) (adjusted for padding) */
+ int rowpixels; /* pixels per row (including padding) */
+ int width; /* width of the bitmap */
+ int height; /* height of the bitmap */
+ bitmap_format format; /* format of the bitmap */
+ int bpp; /* bits per pixel */
+ palette_t * palette; /* optional palette */
+};
+
+
+
+/***************************************************************************
+ MACROS
+***************************************************************************/
+
+/* Macros for accessing bitmap pixels */
+#define BITMAP_ADDR(bitmap, type, y, x) \
+ ((type *)(bitmap)->base + (y) * (bitmap)->rowpixels + (x))
+
+#define BITMAP_ADDR8(bitmap, y, x) BITMAP_ADDR(bitmap, UINT8, y, x)
+#define BITMAP_ADDR16(bitmap, y, x) BITMAP_ADDR(bitmap, UINT16, y, x)
+#define BITMAP_ADDR32(bitmap, y, x) BITMAP_ADDR(bitmap, UINT32, y, x)
+
+
+
+/***************************************************************************
+ FUNCTION PROTOTYPES
+***************************************************************************/
+
+
+/* ----- bitmap allocation ----- */
+
+/* allocate a new bitmap of the given dimensions and format */
+bitmap_t *bitmap_alloc(int width, int height, bitmap_format format);
+
+/* allocate a new bitmap with additional slop on the borders */
+bitmap_t *bitmap_alloc_slop(int width, int height, int xslop, int yslop, bitmap_format format);
+
+/* wrap a bitmap object around an existing array of data */
+bitmap_t *bitmap_wrap(void *base, int width, int height, int rowpixels, bitmap_format format);
+
+/* associate a palette with a bitmap */
+void bitmap_set_palette(bitmap_t *bitmap, palette_t *palette);
+
+/* free allocated data for a bitmap */
+void bitmap_free(bitmap_t *bitmap);
+
+
+
+/* ----- bitmap drawing ----- */
+
+/* fill a bitmap with a single color, clipping to the given rectangle */
+void bitmap_fill(bitmap_t *dest, const rectangle *clip, rgb_t color);
+
+
+
+/* ----- bitmap utilities ----- */
+
+/* return the number of bits per pixel for a given bitmap format */
+int bitmap_format_to_bpp(bitmap_format format);
+
+
+
+
+/***************************************************************************
+ INLINE FUNCTIONS
+***************************************************************************/
+
+/*-------------------------------------------------
+ sect_rect - compute the intersection of two
+ rectangles
+-------------------------------------------------*/
+
+INLINE void sect_rect(rectangle *dst, const rectangle *src)
+{
+ if (src->min_x > dst->min_x) dst->min_x = src->min_x;
+ if (src->max_x < dst->max_x) dst->max_x = src->max_x;
+ if (src->min_y > dst->min_y) dst->min_y = src->min_y;
+ if (src->max_y < dst->max_y) dst->max_y = src->max_y;
+}
+
+
+/*-------------------------------------------------
+ union_rect - compute the union of two
+ rectangles
+-------------------------------------------------*/
+
+INLINE void union_rect(rectangle *dst, const rectangle *src)
+{
+ if (src->min_x < dst->min_x) dst->min_x = src->min_x;
+ if (src->max_x > dst->max_x) dst->max_x = src->max_x;
+ if (src->min_y < dst->min_y) dst->min_y = src->min_y;
+ if (src->max_y > dst->max_y) dst->max_y = src->max_y;
+}
+
+
+/*-------------------------------------------------
+ plot_box - draw a filled rectangle into a
+ bitmap of arbitrary depth
+-------------------------------------------------*/
+
+INLINE void plot_box(bitmap_t *bitmap, int x, int y, int width, int height, UINT32 color)
+{
+ rectangle clip;
+ clip.min_x = x;
+ clip.min_y = y;
+ clip.max_x = x + width - 1;
+ clip.max_y = y + height - 1;
+ bitmap_fill(bitmap, &clip, color);
+}
+
+
+#endif /* __BITMAP_H__ */
diff --git a/src/lib/util/cdrom.c b/src/lib/util/cdrom.c
new file mode 100644
index 00000000000..82287ad30eb
--- /dev/null
+++ b/src/lib/util/cdrom.c
@@ -0,0 +1,617 @@
+/***************************************************************************
+
+ Generic MAME CD-ROM utilties - build IDE and SCSI CD-ROMs on top of this
+
+ Copyright (c) 1996-2007, Nicola Salmoria and the MAME Team.
+ Visit http://mamedev.org for licensing and usage restrictions.
+
+****************************************************************************
+
+ IMPORTANT:
+ "physical" block addresses are the actual addresses on the emulated CD.
+ "chd" block addresses are the block addresses in the CHD file.
+ Because we pad each track to a hunk boundry, these addressing
+ schemes will differ after track 1!
+
+***************************************************************************/
+
+#include "cdrom.h"
+
+
+
+/***************************************************************************
+ DEBUGGING
+***************************************************************************/
+
+#define VERBOSE (0)
+
+
+
+/***************************************************************************
+ TYPE DEFINITIONS
+***************************************************************************/
+
+struct _cdrom_file
+{
+ chd_file * chd; /* CHD file */
+ cdrom_toc cdtoc; /* TOC for the CD */
+ UINT32 hunksectors; /* sectors per hunk */
+ UINT32 cachehunk; /* which hunk is cached */
+ UINT8 * cache; /* cache of the current hunk */
+};
+
+
+
+/***************************************************************************
+ FUNCTION PROTOTYPES
+***************************************************************************/
+
+static chd_error read_sector_into_cache(cdrom_file *file, UINT32 lbasector, UINT32 *sectoroffs, UINT32 *tracknum);
+static chd_error parse_metadata(chd_file *chd, cdrom_toc *toc);
+
+
+
+/***************************************************************************
+ INLINE FUNCTIONS
+***************************************************************************/
+
+/*-------------------------------------------------
+ physical_to_chd_lba - find the CHD LBA
+ and the track number
+-------------------------------------------------*/
+
+INLINE UINT32 physical_to_chd_lba(cdrom_file *file, UINT32 physlba, UINT32 *tracknum)
+{
+ UINT32 chdlba;
+ int track;
+
+ /* loop until our current LBA is less than the start LBA of the next track */
+ for (track = 0; track < file->cdtoc.numtrks; track++)
+ if (physlba < file->cdtoc.tracks[track + 1].physframeofs)
+ {
+ chdlba = physlba - file->cdtoc.tracks[track].physframeofs + file->cdtoc.tracks[track].chdframeofs;
+ if (tracknum != NULL)
+ *tracknum = track;
+ return chdlba;
+ }
+
+ return physlba;
+}
+
+
+
+/***************************************************************************
+ BASE FUNCTIONALITY
+***************************************************************************/
+
+/*-------------------------------------------------
+ cdrom_open - "open" a CD-ROM file from an
+ already-opened CHD file
+-------------------------------------------------*/
+
+cdrom_file *cdrom_open(chd_file *chd)
+{
+ const chd_header *header = chd_get_header(chd);
+ int i;
+ cdrom_file *file;
+ UINT32 physofs, chdofs;
+ chd_error err;
+
+ /* punt if no CHD */
+ if (!chd)
+ return NULL;
+
+ /* validate the CHD information */
+ if (header->hunkbytes % CD_FRAME_SIZE != 0)
+ return NULL;
+
+ /* allocate memory for the CD-ROM file */
+ file = malloc(sizeof(cdrom_file));
+ if (file == NULL)
+ return NULL;
+
+ /* fill in the data */
+ file->chd = chd;
+ file->hunksectors = header->hunkbytes / CD_FRAME_SIZE;
+ file->cachehunk = -1;
+
+ /* read the CD-ROM metadata */
+ err = parse_metadata(chd, &file->cdtoc);
+ if (err != CHDERR_NONE)
+ {
+ free(file);
+ return NULL;
+ }
+
+ #if VERBOSE
+ logerror("CD has %d tracks\n", file->cdtoc.numtrks);
+ #endif
+
+ /* calculate the starting frame for each track, keeping in mind that CHDMAN
+ pads tracks out with extra frames to fit hunk size boundries
+ */
+ physofs = chdofs = 0;
+ for (i = 0; i < file->cdtoc.numtrks; i++)
+ {
+ file->cdtoc.tracks[i].physframeofs = physofs;
+ file->cdtoc.tracks[i].chdframeofs = chdofs;
+
+ physofs += file->cdtoc.tracks[i].frames;
+ chdofs += file->cdtoc.tracks[i].frames;
+ chdofs += file->cdtoc.tracks[i].extraframes;
+
+ #if VERBOSE
+ logerror("Track %02d is format %d subtype %d datasize %d subsize %d frames %d extraframes %d physofs %d chdofs %d\n", i+1,
+ file->cdtoc.tracks[i].trktype,
+ file->cdtoc.tracks[i].subtype,
+ file->cdtoc.tracks[i].datasize,
+ file->cdtoc.tracks[i].subsize,
+ file->cdtoc.tracks[i].frames,
+ file->cdtoc.tracks[i].extraframes,
+ file->cdtoc.tracks[i].physframeofs,
+ file->cdtoc.tracks[i].chdframeofs);
+ #endif
+ }
+
+ /* fill out dummy entries for the last track to help our search */
+ file->cdtoc.tracks[i].physframeofs = physofs;
+ file->cdtoc.tracks[i].chdframeofs = chdofs;
+
+ /* allocate a cache */
+ file->cache = malloc(chd_get_header(chd)->hunkbytes);
+ if (file->cache == NULL)
+ {
+ free(file);
+ return NULL;
+ }
+
+ return file;
+}
+
+
+/*-------------------------------------------------
+ cdrom_close - "close" a CD-ROM file
+-------------------------------------------------*/
+
+void cdrom_close(cdrom_file *file)
+{
+ if (file == NULL)
+ return;
+
+ /* free the cache */
+ if (file->cache)
+ free(file->cache);
+ free(file);
+}
+
+
+
+/***************************************************************************
+ CORE READ ACCESS
+***************************************************************************/
+
+/*-------------------------------------------------
+ cdrom_read_data - read one or more sectors
+ from a CD-ROM
+-------------------------------------------------*/
+
+UINT32 cdrom_read_data(cdrom_file *file, UINT32 lbasector, void *buffer, UINT32 datatype)
+{
+ UINT32 tracktype, tracknum, sectoroffs;
+ chd_error err;
+
+ if (file == NULL)
+ return 0;
+
+ /* cache in the sector */
+ err = read_sector_into_cache(file, lbasector, &sectoroffs, &tracknum);
+ if (err != CHDERR_NONE)
+ return 0;
+
+ /* copy out the requested sector */
+ tracktype = file->cdtoc.tracks[tracknum].trktype;
+ if ((datatype == tracktype) || (datatype == CD_TRACK_RAW_DONTCARE))
+ {
+ memcpy(buffer, &file->cache[sectoroffs * CD_FRAME_SIZE], file->cdtoc.tracks[tracknum].datasize);
+ }
+ else
+ {
+ /* return 2048 bytes of mode1 data from a 2336 byte mode 1 or 2 raw sector */
+ if ((datatype == CD_TRACK_MODE1) && ((tracktype == CD_TRACK_MODE1_RAW) || (tracktype == CD_TRACK_MODE2_RAW)))
+ {
+ memcpy(buffer, &file->cache[(sectoroffs * CD_FRAME_SIZE) + 16], 2048);
+ return 1;
+ }
+
+ /* return 2048 bytes of mode1 data from a mode2 form1 sector */
+ if ((datatype == CD_TRACK_MODE1) && (tracktype == CD_TRACK_MODE2_FORM1))
+ {
+ memcpy(buffer, &file->cache[(sectoroffs * CD_FRAME_SIZE) + 24], 2048);
+ return 1;
+ }
+
+ /* return mode 2 2336 byte data from a RAW sector */
+ if ((datatype == CD_TRACK_MODE2) && ((tracktype == CD_TRACK_MODE1_RAW) || (tracktype == CD_TRACK_MODE2_RAW)))
+ {
+ memcpy(buffer, &file->cache[(sectoroffs * CD_FRAME_SIZE) + 16], 2048);
+ return 1;
+ }
+
+ #if VERBOSE
+ logerror("CDROM: Conversion from type %d to type %d not supported!\n", tracktype, datatype);
+ #endif
+ return 0;
+ }
+ return 1;
+}
+
+
+/*-------------------------------------------------
+ cdrom_read_subcode - read subcode data for
+ a sector
+-------------------------------------------------*/
+
+UINT32 cdrom_read_subcode(cdrom_file *file, UINT32 lbasector, void *buffer)
+{
+ UINT32 sectoroffs, tracknum;
+ chd_error err;
+
+ if (file == NULL)
+ return ~0;
+
+ /* cache in the sector */
+ err = read_sector_into_cache(file, lbasector, &sectoroffs, &tracknum);
+ if (err != CHDERR_NONE)
+ return 0;
+
+ /* copy out the requested data */
+ memcpy(buffer, &file->cache[(sectoroffs * CD_FRAME_SIZE) + file->cdtoc.tracks[tracknum].datasize], file->cdtoc.tracks[tracknum].subsize);
+ return 1;
+}
+
+
+
+/***************************************************************************
+ HANDY UTILITIES
+***************************************************************************/
+
+/*-------------------------------------------------
+ cdrom_get_track - get the track number
+ for a physical frame number
+-------------------------------------------------*/
+
+UINT32 cdrom_get_track(cdrom_file *file, UINT32 frame)
+{
+ UINT32 track = 0;
+
+ if (file == NULL)
+ return ~0;
+
+ /* convert to a CHD sector offset and get track information */
+ physical_to_chd_lba(file, frame, &track);
+ return track;
+}
+
+
+/*-------------------------------------------------
+ cdrom_get_track_start - get the frame number
+ that a track starts at
+-------------------------------------------------*/
+
+UINT32 cdrom_get_track_start(cdrom_file *file, UINT32 track)
+{
+ if (file == NULL)
+ return ~0;
+
+ /* handle lead-out specially */
+ if (track == 0xaa)
+ track = file->cdtoc.numtrks;
+
+ return file->cdtoc.tracks[track].physframeofs;
+}
+
+
+
+/***************************************************************************
+ TOC UTILITIES
+***************************************************************************/
+
+/*-------------------------------------------------
+ cdrom_get_last_track - returns the last track
+ number
+-------------------------------------------------*/
+
+int cdrom_get_last_track(cdrom_file *file)
+{
+ if (file == NULL)
+ return -1;
+
+ return file->cdtoc.numtrks;
+}
+
+
+/*-------------------------------------------------
+ cdrom_get_adr_control - get the ADR | CONTROL
+ for a track
+-------------------------------------------------*/
+
+int cdrom_get_adr_control(cdrom_file *file, int track)
+{
+ if (file == NULL)
+ return -1;
+
+ if (track == 0xaa || file->cdtoc.tracks[track].trktype == CD_TRACK_AUDIO)
+ {
+ return 0x10; // audio track, subchannel is position
+ }
+
+ return 0x14; // data track, subchannel is position
+}
+
+
+/*-------------------------------------------------
+ cdrom_get_track_type - return the track type
+-------------------------------------------------*/
+
+int cdrom_get_track_type(cdrom_file *file, int track)
+{
+ if (file == NULL)
+ return -1;
+
+ return file->cdtoc.tracks[track].trktype;
+}
+
+
+/*-------------------------------------------------
+ cdrom_get_toc - return the TOC data for a
+ CD-ROM
+-------------------------------------------------*/
+
+const cdrom_toc *cdrom_get_toc(cdrom_file *file)
+{
+ if (file == NULL)
+ return NULL;
+
+ return &file->cdtoc;
+}
+
+
+
+/***************************************************************************
+ EXTRA UTILITIES
+***************************************************************************/
+
+/*-------------------------------------------------
+ cdrom_convert_type_string_to_track_info -
+ take a string and convert it into track type
+ and track data size
+-------------------------------------------------*/
+
+void cdrom_convert_type_string_to_track_info(const char *typestring, cdrom_track_info *info)
+{
+ if (!strcmp(typestring, "MODE1"))
+ {
+ info->trktype = CD_TRACK_MODE1;
+ info->datasize = 2048;
+ }
+ else if (!strcmp(typestring, "MODE1_RAW"))
+ {
+ info->trktype = CD_TRACK_MODE1_RAW;
+ info->datasize = 2352;
+ }
+ else if (!strcmp(typestring, "MODE2"))
+ {
+ info->trktype = CD_TRACK_MODE2;
+ info->datasize = 2336;
+ }
+ else if (!strcmp(typestring, "MODE2_FORM1"))
+ {
+ info->trktype = CD_TRACK_MODE2_FORM1;
+ info->datasize = 2048;
+ }
+ else if (!strcmp(typestring, "MODE2_FORM2"))
+ {
+ info->trktype = CD_TRACK_MODE2_FORM2;
+ info->datasize = 2324;
+ }
+ else if (!strcmp(typestring, "MODE2_FORM_MIX"))
+ {
+ info->trktype = CD_TRACK_MODE2_FORM_MIX;
+ info->datasize = 2336;
+ }
+ else if (!strcmp(typestring, "MODE2_RAW"))
+ {
+ info->trktype = CD_TRACK_MODE2_RAW;
+ info->datasize = 2352;
+ }
+ else if (!strcmp(typestring, "AUDIO"))
+ {
+ info->trktype = CD_TRACK_AUDIO;
+ info->datasize = 2352;
+ }
+}
+
+
+/*-------------------------------------------------
+ cdrom_convert_subtype_string_to_track_info -
+ take a string and convert it into track subtype
+ and track subcode data size
+-------------------------------------------------*/
+
+void cdrom_convert_subtype_string_to_track_info(const char *typestring, cdrom_track_info *info)
+{
+ if (!strcmp(typestring, "RW"))
+ {
+ info->subtype = CD_SUB_NORMAL;
+ info->subsize = 96;
+ }
+ else if (!strcmp(typestring, "RW_RAW"))
+ {
+ info->subtype = CD_SUB_RAW;
+ info->subsize = 96;
+ }
+}
+
+
+/*-------------------------------------------------
+ cdrom_get_type_string - get the string
+ associated with the given type
+-------------------------------------------------*/
+
+const char *cdrom_get_type_string(const cdrom_track_info *info)
+{
+ switch (info->trktype)
+ {
+ case CD_TRACK_MODE1: return "MODE1";
+ case CD_TRACK_MODE1_RAW: return "MODE1_RAW";
+ case CD_TRACK_MODE2: return "MODE2";
+ case CD_TRACK_MODE2_FORM1: return "MODE2_FORM1";
+ case CD_TRACK_MODE2_FORM2: return "MODE2_FORM2";
+ case CD_TRACK_MODE2_FORM_MIX: return "MODE2_FORM_MIX";
+ case CD_TRACK_MODE2_RAW: return "MODE2_RAW";
+ case CD_TRACK_AUDIO: return "AUDIO";
+ default: return "UNKNOWN";
+ }
+}
+
+
+/*-------------------------------------------------
+ cdrom_get_subtype_string - get the string
+ associated with the given subcode type
+-------------------------------------------------*/
+
+const char *cdrom_get_subtype_string(const cdrom_track_info *info)
+{
+ switch (info->subtype)
+ {
+ case CD_SUB_NORMAL: return "RW";
+ case CD_SUB_RAW: return "RW_RAW";
+ default: return "NONE";
+ }
+}
+
+
+
+/***************************************************************************
+ INTERNAL UTILITIES
+***************************************************************************/
+
+/*-------------------------------------------------
+ read_sector_into_cache - cache a sector at
+ the given physical LBA
+-------------------------------------------------*/
+
+static chd_error read_sector_into_cache(cdrom_file *file, UINT32 lbasector, UINT32 *sectoroffs, UINT32 *tracknum)
+{
+ UINT32 chdsector, hunknum;
+ chd_error err;
+
+ /* convert to a CHD sector offset and get track information */
+ *tracknum = 0;
+ chdsector = physical_to_chd_lba(file, lbasector, tracknum);
+ hunknum = chdsector / file->hunksectors;
+ *sectoroffs = chdsector % file->hunksectors;
+
+ /* if we haven't cached this hunk, read it now */
+ if (file->cachehunk != hunknum)
+ {
+ err = chd_read(file->chd, hunknum, file->cache);
+ if (err !=