/* * parserold.c : the 1.8.11 XML parser core added for compatibility * * See Copyright for the status of this software. * * Daniel.Veillard@w3.org */ #ifdef WIN32 #include "win32config.h" #else #include "config.h" #endif #include #include /* for memset() only */ #ifdef HAVE_CTYPE_H #include #endif #ifdef HAVE_STDLIB_H #include #endif #ifdef HAVE_SYS_STAT_H #include #endif #ifdef HAVE_FCNTL_H #include #endif #ifdef HAVE_UNISTD_H #include #endif #ifdef HAVE_ZLIB_H #include #endif #include "xmlmemory.h" #include "tree.h" #include "parser.h" #include "entities.h" #include "encoding.h" #include "valid.h" #include "parserInternals.h" #include "xmlIO.h" #include "xml-error.h" #define XML_PARSER_BIG_BUFFER_SIZE 1000 #define XML_PARSER_BUFFER_SIZE 100 /* * List of XML prefixed PI allowed by W3C specs */ static const char *xmlW3CPIs[] = { "xml-stylesheet", NULL }; /************************************************************************ * * * Tweaks for plugging in the old parser * * * ************************************************************************/ static void xmlOldFreeInputStream(xmlParserInputPtr input); static xmlChar * xmlOldParseName(xmlParserCtxtPtr ctxt); static xmlEntityPtr xmlOldParseEntityRef(xmlParserCtxtPtr ctxt); static void xmlOldParsePEReference(xmlParserCtxtPtr ctxt); static void xmlOldParseReference(xmlParserCtxtPtr ctxt); static xmlChar * xmlOldParseVersionInfo(xmlParserCtxtPtr ctxt); static xmlChar * xmlOldParseEncodingDecl(xmlParserCtxtPtr ctxt); static void xmlOldParseElement(xmlParserCtxtPtr ctxt); static xmlChar * xmlOldScanName(xmlParserCtxtPtr ctxt); static xmlEntityPtr xmlOldParseStringEntityRef(xmlParserCtxtPtr ctxt, const xmlChar ** str); static xmlEntityPtr xmlOldParseStringPEReference(xmlParserCtxtPtr ctxt, const xmlChar **str); extern xmlChar * xmlCharStrdup(const char *cur); /************************************************************************ * * * Input handling functions for progressive parsing * * * ************************************************************************/ /* #define DEBUG_INPUT */ /* #define DEBUG_STACK */ /* #define DEBUG_PUSH */ #define INPUT_CHUNK 250 /* we need to keep enough input to show errors in context */ #define LINE_LEN 80 #ifdef DEBUG_INPUT #define CHECK_BUFFER(in) check_buffer(in) static void old_check_buffer(xmlParserInputPtr in) { if (in->base != in->buf->buffer->content) { fprintf(stderr, "xmlParserInput: base mismatch problem\n"); } if (in->cur < in->base) { fprintf(stderr, "xmlParserInput: cur < base problem\n"); } if (in->cur > in->base + in->buf->buffer->use) { fprintf(stderr, "xmlParserInput: cur > base + use problem\n"); } fprintf(stderr,"buffer %x : content %x, cur %d, use %d, size %d\n", (int) in, (int) in->buf->buffer->content, in->cur - in->base, in->buf->buffer->use, in->buf->buffer->size); } #else #define CHECK_BUFFER(in) #endif /** * xmlOldParserInputRead: * @in: an XML parser input * @len: an indicative size for the lookahead * * This function refresh the input for the parser. It doesn't try to * preserve pointers to the input buffer, and discard already read data * * Returns the number of xmlChars read, or -1 in case of error, 0 indicate the * end of this entity */ static int xmlOldParserInputRead(xmlParserInputPtr in, int len) { int ret; int used; int index; #ifdef DEBUG_INPUT fprintf(stderr, "Read\n"); #endif if (in->buf == NULL) return(-1); if (in->base == NULL) return(-1); if (in->cur == NULL) return(-1); if (in->buf->buffer == NULL) return(-1); CHECK_BUFFER(in); used = in->cur - in->buf->buffer->content; ret = xmlBufferShrink(in->buf->buffer, used); if (ret > 0) { in->cur -= ret; in->consumed += ret; } ret = xmlParserInputBufferRead(in->buf, len); if (in->base != in->buf->buffer->content) { /* * the buffer has been realloced */ index = in->cur - in->base; in->base = in->buf->buffer->content; in->cur = &in->buf->buffer->content[index]; } CHECK_BUFFER(in); return(ret); } /** * xmlOldParserInputGrow: * @in: an XML parser input * @len: an indicative size for the lookahead * * This function increase the input for the parser. It tries to * preserve pointers to the input buffer, and keep already read data * * Returns the number of xmlChars read, or -1 in case of error, 0 indicate the * end of this entity */ static int xmlOldParserInputGrow(xmlParserInputPtr in, int len) { int ret; int index; #ifdef DEBUG_INPUT fprintf(stderr, "Grow\n"); #endif if (in->buf == NULL) return(-1); if (in->base == NULL) return(-1); if (in->cur == NULL) return(-1); if (in->buf->buffer == NULL) return(-1); CHECK_BUFFER(in); index = in->cur - in->base; if (in->buf->buffer->use > index + INPUT_CHUNK) { CHECK_BUFFER(in); return(0); } if ((in->buf->httpIO != NULL) || (in->buf->ftpIO != NULL) || (in->buf->file != NULL) || #ifdef HAVE_ZLIB_H (in->buf->gzfile != NULL) || #endif (in->buf->fd >= 0)) ret = xmlParserInputBufferGrow(in->buf, len); else return(0); /* * NOTE : in->base may be a "dandling" i.e. freed pointer in this * block, but we use it really as an integer to do some * pointer arithmetic. Insure will raise it as a bug but in * that specific case, that's not ! */ if (in->base != in->buf->buffer->content) { /* * the buffer has been realloced */ index = in->cur - in->base; in->base = in->buf->buffer->content; in->cur = &in->buf->buffer->content[index]; } CHECK_BUFFER(in); return(ret); } /** * xmlOldParserInputShrink: * @in: an XML parser input * * This function removes used input for the parser. */ static void xmlOldParserInputShrink(xmlParserInputPtr in) { int used; int ret; int index; #ifdef DEBUG_INPUT fprintf(stderr, "Shrink\n"); #endif if (in->buf == NULL) return; if (in->base == NULL) return; if (in->cur == NULL) return; if (in->buf->buffer == NULL) return; CHECK_BUFFER(in); used = in->cur - in->buf->buffer->content; if (used > INPUT_CHUNK) { ret = xmlBufferShrink(in->buf->buffer, used - LINE_LEN); if (ret > 0) { in->cur -= ret; in->consumed += ret; } } CHECK_BUFFER(in); if (in->buf->buffer->use > INPUT_CHUNK) { return; } xmlParserInputBufferRead(in->buf, 2 * INPUT_CHUNK); if (in->base != in->buf->buffer->content) { /* * the buffer has been realloced */ index = in->cur - in->base; in->base = in->buf->buffer->content; in->cur = &in->buf->buffer->content[index]; } CHECK_BUFFER(in); } /************************************************************************ * * * Parser stacks related functions and macros * * * ************************************************************************/ extern int xmlSubstituteEntitiesDefaultValue; extern int xmlDoValidityCheckingDefaultValue; extern int xmlKeepBlanksDefaultValue; xmlEntityPtr xmlOldParseStringEntityRef(xmlParserCtxtPtr ctxt, const xmlChar ** str); /* * Generic function for accessing stacks in the Parser Context */ #define PUSH_AND_POP(scope, type, name) \ scope int name##OldPush(xmlParserCtxtPtr ctxt, type value) { \ if (ctxt->name##Nr >= ctxt->name##Max) { \ ctxt->name##Max *= 2; \ ctxt->name##Tab = (void *) xmlRealloc(ctxt->name##Tab, \ ctxt->name##Max * sizeof(ctxt->name##Tab[0])); \ if (ctxt->name##Tab == NULL) { \ fprintf(stderr, "realloc failed !\n"); \ return(0); \ } \ } \ ctxt->name##Tab[ctxt->name##Nr] = value; \ ctxt->name = value; \ return(ctxt->name##Nr++); \ } \ scope type name##OldPop(xmlParserCtxtPtr ctxt) { \ type ret; \ if (ctxt->name##Nr <= 0) return(0); \ ctxt->name##Nr--; \ if (ctxt->name##Nr > 0) \ ctxt->name = ctxt->name##Tab[ctxt->name##Nr - 1]; \ else \ ctxt->name = NULL; \ ret = ctxt->name##Tab[ctxt->name##Nr]; \ ctxt->name##Tab[ctxt->name##Nr] = 0; \ return(ret); \ } \ PUSH_AND_POP(static, xmlParserInputPtr, input) PUSH_AND_POP(static, xmlNodePtr, node) PUSH_AND_POP(static, xmlChar*, name) /* * Macros for accessing the content. Those should be used only by the parser, * and not exported. * * Dirty macros, i.e. one need to make assumption on the context to use them * * CUR_PTR return the current pointer to the xmlChar to be parsed. * CUR returns the current xmlChar value, i.e. a 8 bit value if compiled * in ISO-Latin or UTF-8, and the current 16 bit value if compiled * in UNICODE mode. This should be used internally by the parser * only to compare to ASCII values otherwise it would break when * running with UTF-8 encoding. * NXT(n) returns the n'th next xmlChar. Same as CUR is should be used only * to compare on ASCII based substring. * SKIP(n) Skip n xmlChar, and must also be used only to skip ASCII defined * strings within the parser. * * Clean macros, not dependent of an ASCII context, expect UTF-8 encoding * * CURRENT Returns the current char value, with the full decoding of * UTF-8 if we are using this mode. It returns an int. * NEXT Skip to the next character, this does the proper decoding * in UTF-8 mode. It also pop-up unfinished entities on the fly. * COPY(to) copy one char to *to, increment CUR_PTR and to accordingly */ #define RAW (ctxt->token ? -1 : (*ctxt->input->cur)) #define CUR (ctxt->token ? ctxt->token : (*ctxt->input->cur)) #define SKIP(val) ctxt->nbChars += (val),ctxt->input->cur += (val) #define NXT(val) ctxt->input->cur[(val)] #define CUR_PTR ctxt->input->cur #define SHRINK xmlOldParserInputShrink(ctxt->input); \ if ((*ctxt->input->cur == 0) && \ (xmlOldParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) \ xmlOldPopInput(ctxt) #define GROW xmlOldParserInputGrow(ctxt->input, INPUT_CHUNK); \ if ((*ctxt->input->cur == 0) && \ (xmlOldParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) \ xmlOldPopInput(ctxt) #define SKIP_BLANKS \ do { \ while (IS_BLANK(CUR)) NEXT; \ while ((CUR == 0) && (ctxt->inputNr > 1)) \ xmlOldPopInput(ctxt); \ if (*ctxt->input->cur == '%') xmlOldParserHandlePEReference(ctxt); \ if (*ctxt->input->cur == '&') xmlOldParserHandleReference(ctxt); \ } while (IS_BLANK(CUR)); #define CURRENT (*ctxt->input->cur) #define NEXT { \ if (ctxt->token != 0) ctxt->token = 0; \ else { \ if ((*ctxt->input->cur == 0) && \ (xmlOldParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) { \ xmlOldPopInput(ctxt); \ } else { \ if (*(ctxt->input->cur) == '\n') { \ ctxt->input->line++; ctxt->input->col = 1; \ } else ctxt->input->col++; \ ctxt->input->cur++; \ ctxt->nbChars++; \ if (*ctxt->input->cur == 0) \ xmlOldParserInputGrow(ctxt->input, INPUT_CHUNK); \ } \ if (*ctxt->input->cur == '%') xmlOldParserHandlePEReference(ctxt); \ if (*ctxt->input->cur == '&') xmlOldParserHandleReference(ctxt); \ }} /************************************************************************ * * * Commodity functions to handle entities processing * * * ************************************************************************/ /** * xmlOldPopInput: * @ctxt: an XML parser context * * xmlOldPopInput: the current input pointed by ctxt->input came to an end * pop it and return the next char. * * Returns the current xmlChar in the parser context */ static xmlChar xmlOldPopInput(xmlParserCtxtPtr ctxt) { if (ctxt->inputNr == 1) return(0); /* End of main Input */ xmlOldFreeInputStream(inputOldPop(ctxt)); if ((*ctxt->input->cur == 0) && (xmlOldParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) return(xmlOldPopInput(ctxt)); return(CUR); } /** * xmlOldPushInput: * @ctxt: an XML parser context * @input: an XML parser input fragment (entity, XML fragment ...). * * xmlOldPushInput: switch to a new input stream which is stacked on top * of the previous one(s). */ void xmlOldPushInput(xmlParserCtxtPtr ctxt, xmlParserInputPtr input) { if (input == NULL) return; inputOldPush(ctxt, input); } /** * xmlOldFreeInputStream: * @input: an xmlParserInputPtr * * Free up an input stream. */ static void xmlOldFreeInputStream(xmlParserInputPtr input) { if (input == NULL) return; if (input->filename != NULL) xmlFree((char *) input->filename); if (input->directory != NULL) xmlFree((char *) input->directory); if ((input->free != NULL) && (input->base != NULL)) input->free((xmlChar *) input->base); if (input->buf != NULL) xmlFreeParserInputBuffer(input->buf); memset(input, -1, sizeof(xmlParserInput)); xmlFree(input); } /** * xmlOldNewInputStream: * @ctxt: an XML parser context * * Create a new input stream structure * Returns the new input stream or NULL */ static xmlParserInputPtr xmlOldNewInputStream(xmlParserCtxtPtr ctxt) { xmlParserInputPtr input; input = (xmlParserInputPtr) xmlMalloc(sizeof(xmlParserInput)); if (input == NULL) { ctxt->errNo = XML_ERR_NO_MEMORY; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "malloc: couldn't allocate a new input stream\n"); ctxt->errNo = XML_ERR_NO_MEMORY; return(NULL); } input->filename = NULL; input->directory = NULL; input->base = NULL; input->cur = NULL; input->buf = NULL; input->line = 1; input->col = 1; input->buf = NULL; input->free = NULL; input->consumed = 0; input->length = 0; return(input); } /** * xmlOldNewEntityInputStream: * @ctxt: an XML parser context * @entity: an Entity pointer * * Create a new input stream based on an xmlEntityPtr * * Returns the new input stream or NULL */ static xmlParserInputPtr xmlOldNewEntityInputStream(xmlParserCtxtPtr ctxt, xmlEntityPtr entity) { xmlParserInputPtr input; if (entity == NULL) { ctxt->errNo = XML_ERR_INTERNAL_ERROR; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "internal: xmlOldNewEntityInputStream entity = NULL\n"); ctxt->errNo = XML_ERR_INTERNAL_ERROR; return(NULL); } if (entity->content == NULL) { switch (entity->type) { case XML_EXTERNAL_GENERAL_UNPARSED_ENTITY: ctxt->errNo = XML_ERR_UNPARSED_ENTITY; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "xmlNewEntityInputStream unparsed entity !\n"); break; case XML_EXTERNAL_GENERAL_PARSED_ENTITY: case XML_EXTERNAL_PARAMETER_ENTITY: return(xmlLoadExternalEntity((char *) entity->SystemID, (char *) entity->ExternalID, ctxt)); case XML_INTERNAL_GENERAL_ENTITY: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Internal entity %s without content !\n", entity->name); break; case XML_INTERNAL_PARAMETER_ENTITY: ctxt->errNo = XML_ERR_INTERNAL_ERROR; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Internal parameter entity %s without content !\n", entity->name); break; case XML_INTERNAL_PREDEFINED_ENTITY: ctxt->errNo = XML_ERR_INTERNAL_ERROR; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Predefined entity %s without content !\n", entity->name); break; } return(NULL); } input = xmlOldNewInputStream(ctxt); if (input == NULL) { return(NULL); } input->filename = (char *) entity->SystemID; /* TODO !!! char <- xmlChar */ input->base = entity->content; input->cur = entity->content; input->length = entity->length; return(input); } /** * xmlOldNewStringInputStream: * @ctxt: an XML parser context * @buffer: an memory buffer * * Create a new input stream based on a memory buffer. * Returns the new input stream */ static xmlParserInputPtr xmlOldNewStringInputStream(xmlParserCtxtPtr ctxt, const xmlChar *buffer) { xmlParserInputPtr input; if (buffer == NULL) { ctxt->errNo = XML_ERR_INTERNAL_ERROR; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "internal: xmlOldNewStringInputStream string = NULL\n"); return(NULL); } input = xmlOldNewInputStream(ctxt); if (input == NULL) { return(NULL); } input->base = buffer; input->cur = buffer; input->length = xmlStrlen(buffer); return(input); } /** * xmlOldNewInputFromFile: * @ctxt: an XML parser context * @filename: the filename to use as entity * * Create a new input stream based on a file. * * Returns the new input stream or NULL in case of error */ static xmlParserInputPtr xmlOldNewInputFromFile(xmlParserCtxtPtr ctxt, const char *filename) { xmlParserInputBufferPtr buf; xmlParserInputPtr inputStream; char *directory = NULL; if (ctxt == NULL) return(NULL); buf = xmlParserInputBufferCreateFilename(filename, XML_CHAR_ENCODING_NONE); if (buf == NULL) { char name[XML_PARSER_BIG_BUFFER_SIZE]; if ((ctxt->input != NULL) && (ctxt->input->directory != NULL)) { #ifdef WIN32 sprintf(name, "%s\\%s", ctxt->input->directory, filename); #else sprintf(name, "%s/%s", ctxt->input->directory, filename); #endif buf = xmlParserInputBufferCreateFilename(name, XML_CHAR_ENCODING_NONE); if (buf != NULL) directory = xmlParserGetDirectory(name); } if ((buf == NULL) && (ctxt->directory != NULL)) { #ifdef WIN32 sprintf(name, "%s\\%s", ctxt->directory, filename); #else sprintf(name, "%s/%s", ctxt->directory, filename); #endif buf = xmlParserInputBufferCreateFilename(name, XML_CHAR_ENCODING_NONE); if (buf != NULL) directory = xmlParserGetDirectory(name); } if (buf == NULL) return(NULL); } if (directory == NULL) directory = xmlParserGetDirectory(filename); inputStream = xmlOldNewInputStream(ctxt); if (inputStream == NULL) { if (directory != NULL) xmlFree((char *) directory); return(NULL); } inputStream->filename = xmlMemStrdup(filename); inputStream->directory = directory; inputStream->buf = buf; inputStream->base = inputStream->buf->buffer->content; inputStream->cur = inputStream->buf->buffer->content; if ((ctxt->directory == NULL) && (directory != NULL)) ctxt->directory = (char *) xmlStrdup((const xmlChar *) directory); return(inputStream); } /************************************************************************ * * * Commodity functions to handle entities * * * ************************************************************************/ static void xmlOldParserHandleReference(xmlParserCtxtPtr ctxt); static void xmlOldParserHandlePEReference(xmlParserCtxtPtr ctxt); xmlEntityPtr xmlOldParseStringPEReference(xmlParserCtxtPtr ctxt, const xmlChar **str); /** * xmlOldParseCharRef: * @ctxt: an XML parser context * * parse Reference declarations * * [66] CharRef ::= '&#' [0-9]+ ';' | * '&#x' [0-9a-fA-F]+ ';' * * [ WFC: Legal Character ] * Characters referred to using character references must match the * production for Char. * * Returns the value parsed (as an int), 0 in case of error */ static int xmlOldParseCharRef(xmlParserCtxtPtr ctxt) { int val = 0; if (ctxt->token != 0) { val = ctxt->token; ctxt->token = 0; return(val); } if ((CUR == '&') && (NXT(1) == '#') && (NXT(2) == 'x')) { SKIP(3); while (CUR != ';') { if ((CUR >= '0') && (CUR <= '9')) val = val * 16 + (CUR - '0'); else if ((CUR >= 'a') && (CUR <= 'f')) val = val * 16 + (CUR - 'a') + 10; else if ((CUR >= 'A') && (CUR <= 'F')) val = val * 16 + (CUR - 'A') + 10; else { ctxt->errNo = XML_ERR_INVALID_HEX_CHARREF; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "xmlParseCharRef: invalid hexadecimal value\n"); ctxt->wellFormed = 0; val = 0; break; } NEXT; } if (CUR == ';') SKIP(1); /* on purpose to avoid reentrancy problems with NEXT */ } else if ((CUR == '&') && (NXT(1) == '#')) { SKIP(2); while (CUR != ';') { if ((CUR >= '0') && (CUR <= '9')) val = val * 10 + (CUR - '0'); else { ctxt->errNo = XML_ERR_INVALID_DEC_CHARREF; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "xmlParseCharRef: invalid decimal value\n"); ctxt->wellFormed = 0; val = 0; break; } NEXT; } if (CUR == ';') SKIP(1); /* on purpose to avoid reentrancy problems with NEXT */ } else { ctxt->errNo = XML_ERR_INVALID_CHARREF; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "xmlParseCharRef: invalid value\n"); ctxt->wellFormed = 0; } /* * [ WFC: Legal Character ] * Characters referred to using character references must match the * production for Char. */ if (IS_CHAR(val)) { return(val); } else { ctxt->errNo = XML_ERR_INVALID_CHAR; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "CharRef: invalid xmlChar value %d\n", val); ctxt->wellFormed = 0; } return(0); } /** * xmlOldParseStringCharRef: * @ctxt: an XML parser context * @str: a pointer to an index in the string * * parse Reference declarations, variant parsing from a string rather * than an an input flow. * * [66] CharRef ::= '&#' [0-9]+ ';' | * '&#x' [0-9a-fA-F]+ ';' * * [ WFC: Legal Character ] * Characters referred to using character references must match the * production for Char. * * Returns the value parsed (as an int), 0 in case of error, str will be * updated to the current value of the index */ static int xmlOldParseStringCharRef(xmlParserCtxtPtr ctxt, const xmlChar **str) { const xmlChar *ptr; xmlChar cur; int val = 0; if ((str == NULL) || (*str == NULL)) return(0); ptr = *str; cur = *ptr; if ((cur == '&') && (ptr[1] == '#') && (ptr[2] == 'x')) { ptr += 3; cur = *ptr; while (cur != ';') { if ((cur >= '0') && (cur <= '9')) val = val * 16 + (cur - '0'); else if ((cur >= 'a') && (cur <= 'f')) val = val * 16 + (cur - 'a') + 10; else if ((cur >= 'A') && (cur <= 'F')) val = val * 16 + (cur - 'A') + 10; else { ctxt->errNo = XML_ERR_INVALID_HEX_CHARREF; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "xmlParseCharRef: invalid hexadecimal value\n"); ctxt->wellFormed = 0; val = 0; break; } ptr++; cur = *ptr; } if (cur == ';') ptr++; } else if ((cur == '&') && (ptr[1] == '#')){ ptr += 2; cur = *ptr; while (cur != ';') { if ((cur >= '0') && (cur <= '9')) val = val * 10 + (cur - '0'); else { ctxt->errNo = XML_ERR_INVALID_DEC_CHARREF; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "xmlParseCharRef: invalid decimal value\n"); ctxt->wellFormed = 0; val = 0; break; } ptr++; cur = *ptr; } if (cur == ';') ptr++; } else { ctxt->errNo = XML_ERR_INVALID_CHARREF; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "xmlParseCharRef: invalid value\n"); ctxt->wellFormed = 0; return(0); } *str = ptr; /* * [ WFC: Legal Character ] * Characters referred to using character references must match the * production for Char. */ if (IS_CHAR(val)) { return(val); } else { ctxt->errNo = XML_ERR_INVALID_CHAR; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "CharRef: invalid xmlChar value %d\n", val); ctxt->wellFormed = 0; } return(0); } /** * xmlOldParserHandleReference: * @ctxt: the parser context * * [67] Reference ::= EntityRef | CharRef * * [68] EntityRef ::= '&' Name ';' * * [ WFC: Entity Declared ] * the Name given in the entity reference must match that in an entity * declaration, except that well-formed documents need not declare any * of the following entities: amp, lt, gt, apos, quot. * * [ WFC: Parsed Entity ] * An entity reference must not contain the name of an unparsed entity * * [66] CharRef ::= '&#' [0-9]+ ';' | * '&#x' [0-9a-fA-F]+ ';' * * A PEReference may have been detectect in the current input stream * the handling is done accordingly to * http://www.w3.org/TR/REC-xml#entproc */ static void xmlOldParserHandleReference(xmlParserCtxtPtr ctxt) { xmlParserInputPtr input; xmlChar *name; xmlEntityPtr ent = NULL; if (ctxt->token != 0) { return; } if (CUR != '&') return; GROW; if ((CUR == '&') && (NXT(1) == '#')) { switch(ctxt->instate) { case XML_PARSER_ENTITY_DECL: case XML_PARSER_PI: case XML_PARSER_CDATA_SECTION: case XML_PARSER_COMMENT: /* we just ignore it there */ return; case XML_PARSER_START_TAG: return; case XML_PARSER_END_TAG: return; case XML_PARSER_EOF: ctxt->errNo = XML_ERR_CHARREF_AT_EOF; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "CharRef at EOF\n"); ctxt->wellFormed = 0; return; case XML_PARSER_PROLOG: case XML_PARSER_START: case XML_PARSER_MISC: ctxt->errNo = XML_ERR_CHARREF_IN_PROLOG; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "CharRef in prolog!\n"); ctxt->wellFormed = 0; return; case XML_PARSER_EPILOG: ctxt->errNo = XML_ERR_CHARREF_IN_EPILOG; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "CharRef in epilog!\n"); ctxt->wellFormed = 0; return; case XML_PARSER_DTD: ctxt->errNo = XML_ERR_CHARREF_IN_DTD; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "CharRef are forbiden in DTDs!\n"); ctxt->wellFormed = 0; return; case XML_PARSER_ENTITY_VALUE: /* * NOTE: in the case of entity values, we don't do the * substitution here since we need the literal * entity value to be able to save the internal * subset of the document. * This will be handled by xmlOldDecodeEntities */ return; case XML_PARSER_CONTENT: case XML_PARSER_ATTRIBUTE_VALUE: /* !!! this may not be Ok for UTF-8, multibyte sequence */ ctxt->token = xmlOldParseCharRef(ctxt); return; } return; } switch(ctxt->instate) { case XML_PARSER_CDATA_SECTION: return; case XML_PARSER_PI: case XML_PARSER_COMMENT: return; case XML_PARSER_START_TAG: return; case XML_PARSER_END_TAG: return; case XML_PARSER_EOF: ctxt->errNo = XML_ERR_ENTITYREF_AT_EOF; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Reference at EOF\n"); ctxt->wellFormed = 0; return; case XML_PARSER_PROLOG: case XML_PARSER_START: case XML_PARSER_MISC: ctxt->errNo = XML_ERR_ENTITYREF_IN_PROLOG; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Reference in prolog!\n"); ctxt->wellFormed = 0; return; case XML_PARSER_EPILOG: ctxt->errNo = XML_ERR_ENTITYREF_IN_EPILOG; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Reference in epilog!\n"); ctxt->wellFormed = 0; return; case XML_PARSER_ENTITY_VALUE: /* * NOTE: in the case of entity values, we don't do the * substitution here since we need the literal * entity value to be able to save the internal * subset of the document. * This will be handled by xmlOldDecodeEntities */ return; case XML_PARSER_ATTRIBUTE_VALUE: /* * NOTE: in the case of attributes values, we don't do the * substitution here unless we are in a mode where * the parser is explicitely asked to substitute * entities. The SAX callback is called with values * without entity substitution. * This will then be handled by xmlOldDecodeEntities */ return; case XML_PARSER_ENTITY_DECL: /* * we just ignore it there * the substitution will be done once the entity is referenced */ return; case XML_PARSER_DTD: ctxt->errNo = XML_ERR_ENTITYREF_IN_DTD; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Entity references are forbiden in DTDs!\n"); ctxt->wellFormed = 0; return; case XML_PARSER_CONTENT: return; } NEXT; name = xmlOldScanName(ctxt); if (name == NULL) { ctxt->errNo = XML_ERR_ENTITYREF_NO_NAME; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Entity reference: no name\n"); ctxt->wellFormed = 0; ctxt->token = '&'; return; } if (NXT(xmlStrlen(name)) != ';') { ctxt->errNo = XML_ERR_ENTITYREF_SEMICOL_MISSING; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Entity reference: ';' expected\n"); ctxt->wellFormed = 0; ctxt->token = '&'; xmlFree(name); return; } SKIP(xmlStrlen(name) + 1); if (ctxt->sax != NULL) { if (ctxt->sax->getEntity != NULL) ent = ctxt->sax->getEntity(ctxt->userData, name); } /* * [ WFC: Entity Declared ] * the Name given in the entity reference must match that in an entity * declaration, except that well-formed documents need not declare any * of the following entities: amp, lt, gt, apos, quot. */ if (ent == NULL) ent = xmlGetPredefinedEntity(name); if (ent == NULL) { ctxt->errNo = XML_ERR_UNDECLARED_ENTITY; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Entity reference: entity %s not declared\n", name); ctxt->wellFormed = 0; xmlFree(name); return; } /* * [ WFC: Parsed Entity ] * An entity reference must not contain the name of an unparsed entity */ if (ent->type == XML_EXTERNAL_GENERAL_UNPARSED_ENTITY) { ctxt->errNo = XML_ERR_UNPARSED_ENTITY; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Entity reference to unparsed entity %s\n", name); ctxt->wellFormed = 0; } if (ent->type == XML_INTERNAL_PREDEFINED_ENTITY) { ctxt->token = ent->content[0]; xmlFree(name); return; } input = xmlOldNewEntityInputStream(ctxt, ent); xmlOldPushInput(ctxt, input); xmlFree(name); return; } /** * xmlOldParserHandlePEReference: * @ctxt: the parser context * * [69] PEReference ::= '%' Name ';' * * [ WFC: No Recursion ] * TODO A parsed entity must not contain a recursive * reference to itself, either directly or indirectly. * * [ WFC: Entity Declared ] * In a document without any DTD, a document with only an internal DTD * subset which contains no parameter entity references, or a document * with "standalone='yes'", ... ... The declaration of a parameter * entity must precede any reference to it... * * [ VC: Entity Declared ] * In a document with an external subset or external parameter entities * with "standalone='no'", ... ... The declaration of a parameter entity * must precede any reference to it... * * [ WFC: In DTD ] * Parameter-entity references may only appear in the DTD. * NOTE: misleading but this is handled. * * A PEReference may have been detected in the current input stream * the handling is done accordingly to * http://www.w3.org/TR/REC-xml#entproc * i.e. * - Included in literal in entity values * - Included as Paraemeter Entity reference within DTDs */ static void xmlOldParserHandlePEReference(xmlParserCtxtPtr ctxt) { xmlChar *name; xmlEntityPtr entity = NULL; xmlParserInputPtr input; if (ctxt->token != 0) { return; } if (CUR != '%') return; switch(ctxt->instate) { case XML_PARSER_CDATA_SECTION: return; case XML_PARSER_COMMENT: return; case XML_PARSER_START_TAG: return; case XML_PARSER_END_TAG: return; case XML_PARSER_EOF: ctxt->errNo = XML_ERR_PEREF_AT_EOF; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "PEReference at EOF\n"); ctxt->wellFormed = 0; return; case XML_PARSER_PROLOG: case XML_PARSER_START: case XML_PARSER_MISC: ctxt->errNo = XML_ERR_PEREF_IN_PROLOG; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "PEReference in prolog!\n"); ctxt->wellFormed = 0; return; case XML_PARSER_ENTITY_DECL: case XML_PARSER_CONTENT: case XML_PARSER_ATTRIBUTE_VALUE: case XML_PARSER_PI: /* we just ignore it there */ return; case XML_PARSER_EPILOG: ctxt->errNo = XML_ERR_PEREF_IN_EPILOG; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "PEReference in epilog!\n"); ctxt->wellFormed = 0; return; case XML_PARSER_ENTITY_VALUE: /* * NOTE: in the case of entity values, we don't do the * substitution here since we need the literal * entity value to be able to save the internal * subset of the document. * This will be handled by xmlOldDecodeEntities */ return; case XML_PARSER_DTD: /* * [WFC: Well-Formedness Constraint: PEs in Internal Subset] * In the internal DTD subset, parameter-entity references * can occur only where markup declarations can occur, not * within markup declarations. * In that case this is handled in xmlOldParseMarkupDecl */ if ((ctxt->external == 0) && (ctxt->inputNr == 1)) return; } NEXT; name = xmlOldParseName(ctxt); if (name == NULL) { ctxt->errNo = XML_ERR_PEREF_NO_NAME; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "xmlHandlePEReference: no name\n"); ctxt->wellFormed = 0; } else { if (CUR == ';') { NEXT; if ((ctxt->sax != NULL) && (ctxt->sax->getParameterEntity != NULL)) entity = ctxt->sax->getParameterEntity(ctxt->userData, name); if (entity == NULL) { /* * [ WFC: Entity Declared ] * In a document without any DTD, a document with only an * internal DTD subset which contains no parameter entity * references, or a document with "standalone='yes'", ... * ... The declaration of a parameter entity must precede * any reference to it... */ if ((ctxt->standalone == 1) || ((ctxt->hasExternalSubset == 0) && (ctxt->hasPErefs == 0))) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "PEReference: %%%s; not found\n", name); ctxt->wellFormed = 0; } else { /* * [ VC: Entity Declared ] * In a document with an external subset or external * parameter entities with "standalone='no'", ... * ... The declaration of a parameter entity must precede * any reference to it... */ if ((ctxt->sax != NULL) && (ctxt->sax->warning != NULL)) ctxt->sax->warning(ctxt->userData, "PEReference: %%%s; not found\n", name); ctxt->valid = 0; } } else { if ((entity->type == XML_INTERNAL_PARAMETER_ENTITY) || (entity->type == XML_EXTERNAL_PARAMETER_ENTITY)) { /* * TODO !!!! handle the extra spaces added before and after * c.f. http://www.w3.org/TR/REC-xml#as-PE * TODO !!!! Avoid quote processing in parameters value * c.f. http://www.w3.org/TR/REC-xml#inliteral */ input = xmlOldNewEntityInputStream(ctxt, entity); xmlOldPushInput(ctxt, input); } else { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "xmlHandlePEReference: %s is not a parameter entity\n", name); ctxt->wellFormed = 0; } } } else { ctxt->errNo = XML_ERR_PEREF_SEMICOL_MISSING; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "xmlHandlePEReference: expecting ';'\n"); ctxt->wellFormed = 0; } xmlFree(name); } } /* * Macro used to grow the current buffer. */ #define growBuffer(buffer) { \ buffer##_size *= 2; \ buffer = (xmlChar *) \ xmlRealloc(buffer, buffer##_size * sizeof(xmlChar)); \ if (buffer == NULL) { \ perror("realloc failed"); \ return(NULL); \ } \ } /** * xmlOldDecodeEntities: * @ctxt: the parser context * @what: combination of XML_SUBSTITUTE_REF and XML_SUBSTITUTE_PEREF * @len: the len to decode (in bytes !), -1 for no size limit * @end: an end marker xmlChar, 0 if none * @end2: an end marker xmlChar, 0 if none * @end3: an end marker xmlChar, 0 if none * * [67] Reference ::= EntityRef | CharRef * * [69] PEReference ::= '%' Name ';' * * Returns A newly allocated string with the substitution done. The caller * must deallocate it ! */ static xmlChar * xmlOldDecodeEntities(xmlParserCtxtPtr ctxt, int len, int what, xmlChar end, xmlChar end2, xmlChar end3) { xmlChar *buffer = NULL; int buffer_size = 0; xmlChar *out = NULL; xmlChar *current = NULL; xmlEntityPtr ent; int nbchars = 0; unsigned int max = (unsigned int) len; xmlChar cur; /* * allocate a translation buffer. */ buffer_size = XML_PARSER_BIG_BUFFER_SIZE; buffer = (xmlChar *) xmlMalloc(buffer_size * sizeof(xmlChar)); if (buffer == NULL) { perror("xmlDecodeEntities: malloc failed"); return(NULL); } out = buffer; /* * Ok loop until we reach one of the ending char or a size limit. */ cur = CUR; while ((nbchars < max) && (cur != end) && (cur != end2) && (cur != end3)) { if (cur == 0) break; if ((cur == '&') && (NXT(1) == '#')) { int val = xmlOldParseCharRef(ctxt); *out++ = val; nbchars += 3; } else if ((cur == '&') && (what & XML_SUBSTITUTE_REF)) { ent = xmlOldParseEntityRef(ctxt); if ((ent != NULL) && (ctxt->replaceEntities != 0)) { current = ent->content; while (*current != 0) { *out++ = *current++; if (out - buffer > buffer_size - XML_PARSER_BUFFER_SIZE) { int index = out - buffer; growBuffer(buffer); out = &buffer[index]; } } nbchars += 3 + xmlStrlen(ent->name); } else if (ent != NULL) { int i = xmlStrlen(ent->name); const xmlChar *cur = ent->name; nbchars += i + 2; *out++ = '&'; if (out - buffer > buffer_size - i - XML_PARSER_BUFFER_SIZE) { int index = out - buffer; growBuffer(buffer); out = &buffer[index]; } for (;i > 0;i--) *out++ = *cur++; *out++ = ';'; } } else if (cur == '%' && (what & XML_SUBSTITUTE_PEREF)) { /* * a PEReference induce to switch the entity flow, * we break here to flush the current set of chars * parsed if any. We will be called back later. */ if (nbchars != 0) break; xmlOldParsePEReference(ctxt); /* * Pop-up of finished entities. */ while ((CUR == 0) && (ctxt->inputNr > 1)) xmlOldPopInput(ctxt); break; } else { /* invalid for UTF-8 , use COPY(out); !!!!!! */ *out++ = cur; nbchars++; if (out - buffer > buffer_size - XML_PARSER_BUFFER_SIZE) { int index = out - buffer; growBuffer(buffer); out = &buffer[index]; } NEXT; } cur = CUR; } *out++ = 0; return(buffer); } /** * xmlOldStringDecodeEntities: * @ctxt: the parser context * @str: the input string * @what: combination of XML_SUBSTITUTE_REF and XML_SUBSTITUTE_PEREF * @end: an end marker xmlChar, 0 if none * @end2: an end marker xmlChar, 0 if none * @end3: an end marker xmlChar, 0 if none * * [67] Reference ::= EntityRef | CharRef * * [69] PEReference ::= '%' Name ';' * * Returns A newly allocated string with the substitution done. The caller * must deallocate it ! */ static xmlChar * xmlOldStringDecodeEntities(xmlParserCtxtPtr ctxt, const xmlChar *str, int what, xmlChar end, xmlChar end2, xmlChar end3) { xmlChar *buffer = NULL; int buffer_size = 0; xmlChar *out = NULL; xmlChar *current = NULL; xmlEntityPtr ent; xmlChar cur; /* * allocate a translation buffer. */ buffer_size = XML_PARSER_BIG_BUFFER_SIZE; buffer = (xmlChar *) xmlMalloc(buffer_size * sizeof(xmlChar)); if (buffer == NULL) { perror("xmlStringDecodeEntities: malloc failed"); return(NULL); } out = buffer; /* * Ok loop until we reach one of the ending char or a size limit. */ cur = *str; while ((cur != 0) && (cur != end) && (cur != end2) && (cur != end3)) { if (cur == 0) break; if ((cur == '&') && (str[1] == '#')) { int val = xmlOldParseStringCharRef(ctxt, &str); if (val != 0) *out++ = val; } else if ((cur == '&') && (what & XML_SUBSTITUTE_REF)) { ent = xmlOldParseStringEntityRef(ctxt, &str); if ((ent != NULL) && (ctxt->replaceEntities != 0)) { current = ent->content; while (*current != 0) { *out++ = *current++; if (out - buffer > buffer_size - XML_PARSER_BUFFER_SIZE) { int index = out - buffer; growBuffer(buffer); out = &buffer[index]; } } } else if (ent != NULL) { int i = xmlStrlen(ent->name); const xmlChar *cur = ent->name; *out++ = '&'; if (out - buffer > buffer_size - i - XML_PARSER_BUFFER_SIZE) { int index = out - buffer; growBuffer(buffer); out = &buffer[index]; } for (;i > 0;i--) *out++ = *cur++; *out++ = ';'; } } else if (cur == '%' && (what & XML_SUBSTITUTE_PEREF)) { ent = xmlOldParseStringPEReference(ctxt, &str); if (ent != NULL) { current = ent->content; while (*current != 0) { *out++ = *current++; if (out - buffer > buffer_size - XML_PARSER_BUFFER_SIZE) { int index = out - buffer; growBuffer(buffer); out = &buffer[index]; } } } } else { /* invalid for UTF-8 , use COPY(out); !!!!!! */ *out++ = cur; if (out - buffer > buffer_size - XML_PARSER_BUFFER_SIZE) { int index = out - buffer; growBuffer(buffer); out = &buffer[index]; } str++; } cur = *str; } *out = 0; return(buffer); } /************************************************************************ * * * Commodity functions, cleanup needed ? * * * ************************************************************************/ /** * areBlanksOld: * @ctxt: an XML parser context * @str: a xmlChar * * @len: the size of @str * * Is this a sequence of blank chars that one can ignore ? * * Returns 1 if ignorable 0 otherwise. */ static int areBlanksOld(xmlParserCtxtPtr ctxt, const xmlChar *str, int len) { int i, ret; xmlNodePtr lastChild; /* * Check that the string is made of blanks */ for (i = 0;i < len;i++) if (!(IS_BLANK(str[i]))) return(0); /* * Look if the element is mixed content in the Dtd if available */ if (ctxt->myDoc != NULL) { ret = xmlIsMixedElement(ctxt->myDoc, ctxt->node->name); if (ret == 0) return(1); if (ret == 1) return(0); } /* * Do we allow an heuristic on white space */ if (ctxt->keepBlanks) return(0); if (CUR != '<') return(0); if (ctxt->node == NULL) return(0); if ((ctxt->node->childs == NULL) && (CUR == '<') && (NXT(1) == '/')) return(0); lastChild = xmlGetLastChild(ctxt->node); if (lastChild == NULL) { if (ctxt->node->content != NULL) return(0); } else if (xmlNodeIsText(lastChild)) return(0); else if ((ctxt->node->childs != NULL) && (xmlNodeIsText(ctxt->node->childs))) return(0); return(1); } /** * xmlOldHandleEntity: * @ctxt: an XML parser context * @entity: an XML entity pointer. * * Default handling of defined entities, when should we define a new input * stream ? When do we just handle that as a set of chars ? * * OBSOLETE: to be removed at some point. */ static void xmlOldHandleEntity(xmlParserCtxtPtr ctxt, xmlEntityPtr entity) { int len; xmlParserInputPtr input; if (entity->content == NULL) { ctxt->errNo = XML_ERR_INTERNAL_ERROR; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "xmlHandleEntity %s: content == NULL\n", entity->name); ctxt->wellFormed = 0; return; } len = xmlStrlen(entity->content); if (len <= 2) goto handle_as_char; /* * Redefine its content as an input stream. */ input = xmlOldNewEntityInputStream(ctxt, entity); xmlOldPushInput(ctxt, input); return; handle_as_char: /* * Just handle the content as a set of chars. */ if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL)) ctxt->sax->characters(ctxt->userData, entity->content, len); } /************************************************************************ * * * Extra stuff for namespace support * * Relates to http://www.w3.org/TR/WD-xml-names * * * ************************************************************************/ /** * xmlOldNamespaceParseNCName: * @ctxt: an XML parser context * * parse an XML namespace name. * * [NS 3] NCName ::= (Letter | '_') (NCNameChar)* * * [NS 4] NCNameChar ::= Letter | Digit | '.' | '-' | '_' | * CombiningChar | Extender * * Returns the namespace name or NULL */ static xmlChar * xmlOldNamespaceParseNCName(xmlParserCtxtPtr ctxt) { xmlChar buf[XML_MAX_NAMELEN]; int len = 0; if (!IS_LETTER(CUR) && (CUR != '_')) return(NULL); while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) || (CUR == '.') || (CUR == '-') || (CUR == '_') || (IS_COMBINING(CUR)) || (IS_EXTENDER(CUR))) { buf[len++] = CUR; NEXT; if (len >= XML_MAX_NAMELEN) { fprintf(stderr, "xmlNamespaceParseNCName: reached XML_MAX_NAMELEN limit\n"); while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) || (CUR == '.') || (CUR == '-') || (CUR == '_') || (IS_COMBINING(CUR)) || (IS_EXTENDER(CUR))) NEXT; break; } } return(xmlStrndup(buf, len)); } /** * xmlOldNamespaceParseQName: * @ctxt: an XML parser context * @prefix: a xmlChar ** * * parse an XML qualified name * * [NS 5] QName ::= (Prefix ':')? LocalPart * * [NS 6] Prefix ::= NCName * * [NS 7] LocalPart ::= NCName * * Returns the local part, and prefix is updated * to get the Prefix if any. */ static xmlChar * xmlOldNamespaceParseQName(xmlParserCtxtPtr ctxt, xmlChar **prefix) { xmlChar *ret = NULL; *prefix = NULL; ret = xmlOldNamespaceParseNCName(ctxt); if (CUR == ':') { *prefix = ret; NEXT; ret = xmlOldNamespaceParseNCName(ctxt); } return(ret); } /** * xmlOldParseQuotedString: * @ctxt: an XML parser context * * [OLD] Parse and return a string between quotes or doublequotes * To be removed at next drop of binary compatibility * * Returns the string parser or NULL. */ static xmlChar * xmlOldParseQuotedString(xmlParserCtxtPtr ctxt) { xmlChar *buf = NULL; int len = 0; int size = XML_PARSER_BUFFER_SIZE; xmlChar c; buf = (xmlChar *) xmlMalloc(size * sizeof(xmlChar)); if (buf == NULL) { fprintf(stderr, "malloc of %d byte failed\n", size); return(NULL); } if (CUR == '"') { NEXT; c = CUR; while (IS_CHAR(c) && (c != '"')) { if (len + 1 >= size) { size *= 2; buf = xmlRealloc(buf, size * sizeof(xmlChar)); if (buf == NULL) { fprintf(stderr, "realloc of %d byte failed\n", size); return(NULL); } } buf[len++] = c; NEXT; c = CUR; } if (c != '"') { ctxt->errNo = XML_ERR_STRING_NOT_CLOSED; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "String not closed \"%.50s\"\n", buf); ctxt->wellFormed = 0; } else { NEXT; } } else if (CUR == '\''){ NEXT; c = CUR; while (IS_CHAR(c) && (c != '\'')) { if (len + 1 >= size) { size *= 2; buf = xmlRealloc(buf, size * sizeof(xmlChar)); if (buf == NULL) { fprintf(stderr, "realloc of %d byte failed\n", size); return(NULL); } } buf[len++] = c; NEXT; c = CUR; } if (CUR != '\'') { ctxt->errNo = XML_ERR_STRING_NOT_CLOSED; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "String not closed \"%.50s\"\n", buf); ctxt->wellFormed = 0; } else { NEXT; } } return(buf); } /** * xmlOldParseNamespace: * @ctxt: an XML parser context * * [OLD] xmlOldParseNamespace: parse specific PI '')) { /* * We can have "ns" or "prefix" attributes * Old encoding as 'href' or 'AS' attributes is still supported */ if ((CUR == 'n') && (NXT(1) == 's')) { garbage = 0; SKIP(2); SKIP_BLANKS; if (CUR != '=') continue; NEXT; SKIP_BLANKS; href = xmlOldParseQuotedString(ctxt); SKIP_BLANKS; } else if ((CUR == 'h') && (NXT(1) == 'r') && (NXT(2) == 'e') && (NXT(3) == 'f')) { garbage = 0; SKIP(4); SKIP_BLANKS; if (CUR != '=') continue; NEXT; SKIP_BLANKS; href = xmlOldParseQuotedString(ctxt); SKIP_BLANKS; } else if ((CUR == 'p') && (NXT(1) == 'r') && (NXT(2) == 'e') && (NXT(3) == 'f') && (NXT(4) == 'i') && (NXT(5) == 'x')) { garbage = 0; SKIP(6); SKIP_BLANKS; if (CUR != '=') continue; NEXT; SKIP_BLANKS; prefix = xmlOldParseQuotedString(ctxt); SKIP_BLANKS; } else if ((CUR == 'A') && (NXT(1) == 'S')) { garbage = 0; SKIP(2); SKIP_BLANKS; if (CUR != '=') continue; NEXT; SKIP_BLANKS; prefix = xmlOldParseQuotedString(ctxt); SKIP_BLANKS; } else if ((CUR == '?') && (NXT(1) == '>')) { garbage = 0; NEXT; } else { /* * Found garbage when parsing the namespace */ if (!garbage) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "xmlParseNamespace found garbage\n"); } ctxt->errNo = XML_ERR_NS_DECL_ERROR; ctxt->wellFormed = 0; NEXT; } } MOVETO_ENDTAG(CUR_PTR); NEXT; /* * Register the DTD. if (href != NULL) if ((ctxt->sax != NULL) && (ctxt->sax->globalNamespace != NULL)) ctxt->sax->globalNamespace(ctxt->userData, href, prefix); */ if (prefix != NULL) xmlFree(prefix); if (href != NULL) xmlFree(href); } /************************************************************************ * * * The parser itself * * Relates to http://www.w3.org/TR/REC-xml * * * ************************************************************************/ /** * xmlOldScanName: * @ctxt: an XML parser context * * Trickery: parse an XML name but without consuming the input flow * Needed for rollback cases. * * [4] NameChar ::= Letter | Digit | '.' | '-' | '_' | ':' | * CombiningChar | Extender * * [5] Name ::= (Letter | '_' | ':') (NameChar)* * * [6] Names ::= Name (S Name)* * * Returns the Name parsed or NULL */ static xmlChar * xmlOldScanName(xmlParserCtxtPtr ctxt) { xmlChar buf[XML_MAX_NAMELEN]; int len = 0; GROW; if (!IS_LETTER(CUR) && (CUR != '_') && (CUR != ':')) { return(NULL); } while ((IS_LETTER(NXT(len))) || (IS_DIGIT(NXT(len))) || (NXT(len) == '.') || (NXT(len) == '-') || (NXT(len) == '_') || (NXT(len) == ':') || (IS_COMBINING(NXT(len))) || (IS_EXTENDER(NXT(len)))) { buf[len] = NXT(len); len++; if (len >= XML_MAX_NAMELEN) { fprintf(stderr, "xmlScanName: reached XML_MAX_NAMELEN limit\n"); while ((IS_LETTER(NXT(len))) || (IS_DIGIT(NXT(len))) || (NXT(len) == '.') || (NXT(len) == '-') || (NXT(len) == '_') || (NXT(len) == ':') || (IS_COMBINING(NXT(len))) || (IS_EXTENDER(NXT(len)))) len++; break; } } return(xmlStrndup(buf, len)); } /** * xmlOldParseName: * @ctxt: an XML parser context * * parse an XML name. * * [4] NameChar ::= Letter | Digit | '.' | '-' | '_' | ':' | * CombiningChar | Extender * * [5] Name ::= (Letter | '_' | ':') (NameChar)* * * [6] Names ::= Name (S Name)* * * Returns the Name parsed or NULL */ static xmlChar * xmlOldParseName(xmlParserCtxtPtr ctxt) { xmlChar buf[XML_MAX_NAMELEN]; int len = 0; xmlChar cur; GROW; cur = CUR; if (!IS_LETTER(cur) && (cur != '_') && (cur != ':')) { return(NULL); } while ((IS_LETTER(cur)) || (IS_DIGIT(cur)) || (cur == '.') || (cur == '-') || (cur == '_') || (cur == ':') || (IS_COMBINING(cur)) || (IS_EXTENDER(cur))) { buf[len++] = cur; NEXT; cur = CUR; if (len >= XML_MAX_NAMELEN) { fprintf(stderr, "xmlParseName: reached XML_MAX_NAMELEN limit\n"); while ((IS_LETTER(cur)) || (IS_DIGIT(cur)) || (cur == '.') || (cur == '-') || (cur == '_') || (cur == ':') || (IS_COMBINING(cur)) || (IS_EXTENDER(cur))) { NEXT; cur = CUR; } break; } } return(xmlStrndup(buf, len)); } /** * xmlOldParseStringName: * @ctxt: an XML parser context * @str: a pointer to an index in the string * * parse an XML name. * * [4] NameChar ::= Letter | Digit | '.' | '-' | '_' | ':' | * CombiningChar | Extender * * [5] Name ::= (Letter | '_' | ':') (NameChar)* * * [6] Names ::= Name (S Name)* * * Returns the Name parsed or NULL. The str pointer * is updated to the current location in the string. */ static xmlChar * xmlOldParseStringName(xmlParserCtxtPtr ctxt, const xmlChar** str) { const xmlChar *ptr; const xmlChar *start; xmlChar cur; if ((str == NULL) || (*str == NULL)) return(NULL); start = ptr = *str; cur = *ptr; if (!IS_LETTER(cur) && (cur != '_') && (cur != ':')) { return(NULL); } while ((IS_LETTER(cur)) || (IS_DIGIT(cur)) || (cur == '.') || (cur == '-') || (cur == '_') || (cur == ':') || (IS_COMBINING(cur)) || (IS_EXTENDER(cur))) { ptr++; cur = *ptr; } *str = ptr; return(xmlStrndup(start, ptr - start )); } /** * xmlOldParseNmtoken: * @ctxt: an XML parser context * * parse an XML Nmtoken. * * [7] Nmtoken ::= (NameChar)+ * * [8] Nmtokens ::= Nmtoken (S Nmtoken)* * * Returns the Nmtoken parsed or NULL */ static xmlChar * xmlOldParseNmtoken(xmlParserCtxtPtr ctxt) { xmlChar buf[XML_MAX_NAMELEN]; int len = 0; GROW; while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) || (CUR == '.') || (CUR == '-') || (CUR == '_') || (CUR == ':') || (IS_COMBINING(CUR)) || (IS_EXTENDER(CUR))) { buf[len++] = CUR; NEXT; if (len >= XML_MAX_NAMELEN) { fprintf(stderr, "xmlParseNmtoken: reached XML_MAX_NAMELEN limit\n"); while ((IS_LETTER(CUR)) || (IS_DIGIT(CUR)) || (CUR == '.') || (CUR == '-') || (CUR == '_') || (CUR == ':') || (IS_COMBINING(CUR)) || (IS_EXTENDER(CUR))) NEXT; break; } } return(xmlStrndup(buf, len)); } /** * xmlOldParseEntityValue: * @ctxt: an XML parser context * @orig: if non-NULL store a copy of the original entity value * * parse a value for ENTITY decl. * * [9] EntityValue ::= '"' ([^%&"] | PEReference | Reference)* '"' | * "'" ([^%&'] | PEReference | Reference)* "'" * * Returns the EntityValue parsed with reference substitued or NULL */ static xmlChar * xmlOldParseEntityValue(xmlParserCtxtPtr ctxt, xmlChar **orig) { xmlChar *buf = NULL; int len = 0; int size = XML_PARSER_BUFFER_SIZE; xmlChar c; xmlChar stop; xmlChar *ret = NULL; xmlParserInputPtr input; if (CUR == '"') stop = '"'; else if (CUR == '\'') stop = '\''; else { ctxt->errNo = XML_ERR_ENTITY_NOT_STARTED; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "EntityValue: \" or ' expected\n"); ctxt->wellFormed = 0; return(NULL); } buf = (xmlChar *) xmlMalloc(size * sizeof(xmlChar)); if (buf == NULL) { fprintf(stderr, "malloc of %d byte failed\n", size); return(NULL); } /* * The content of the entity definition is copied in a buffer. */ ctxt->instate = XML_PARSER_ENTITY_VALUE; input = ctxt->input; GROW; NEXT; c = CUR; /* * NOTE: 4.4.5 Included in Literal * When a parameter entity reference appears in a literal entity * value, ... a single or double quote character in the replacement * text is always treated as a normal data character and will not * terminate the literal. * In practice it means we stop the loop only when back at parsing * the initial entity and the quote is found */ while (IS_CHAR(c) && ((c != stop) || (ctxt->input != input))) { if (len + 1 >= size) { size *= 2; buf = xmlRealloc(buf, size * sizeof(xmlChar)); if (buf == NULL) { fprintf(stderr, "realloc of %d byte failed\n", size); return(NULL); } } buf[len++] = c; NEXT; /* * Pop-up of finished entities. */ while ((CUR == 0) && (ctxt->inputNr > 1)) xmlOldPopInput(ctxt); c = CUR; if (c == 0) { GROW; c = CUR; } } buf[len] = 0; /* * Then PEReference entities are substituted. */ if (c != stop) { ctxt->errNo = XML_ERR_ENTITY_NOT_FINISHED; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "EntityValue: \" expected\n"); ctxt->wellFormed = 0; } else { NEXT; /* * NOTE: 4.4.7 Bypassed * When a general entity reference appears in the EntityValue in * an entity declaration, it is bypassed and left as is. * so XML_SUBSTITUTE_REF is not set. */ ret = xmlOldStringDecodeEntities(ctxt, buf, XML_SUBSTITUTE_PEREF, 0, 0, 0); if (orig != NULL) *orig = buf; else xmlFree(buf); } return(ret); } /** * xmlOldParseAttValue: * @ctxt: an XML parser context * * parse a value for an attribute * Note: the parser won't do substitution of entities here, this * will be handled later in xmlStringGetNodeList * * [10] AttValue ::= '"' ([^<&"] | Reference)* '"' | * "'" ([^<&'] | Reference)* "'" * * 3.3.3 Attribute-Value Normalization: * Before the value of an attribute is passed to the application or * checked for validity, the XML processor must normalize it as follows: * - a character reference is processed by appending the referenced * character to the attribute value * - an entity reference is processed by recursively processing the * replacement text of the entity * - a whitespace character (#x20, #xD, #xA, #x9) is processed by * appending #x20 to the normalized value, except that only a single * #x20 is appended for a "#xD#xA" sequence that is part of an external * parsed entity or the literal entity value of an internal parsed entity * - other characters are processed by appending them to the normalized value * If the declared value is not CDATA, then the XML processor must further * process the normalized attribute value by discarding any leading and * trailing space (#x20) characters, and by replacing sequences of space * (#x20) characters by a single space (#x20) character. * All attributes for which no declaration has been read should be treated * by a non-validating parser as if declared CDATA. * * Returns the AttValue parsed or NULL. The value has to be freed by the caller. */ static xmlChar * xmlOldParseAttValue(xmlParserCtxtPtr ctxt) { xmlChar limit = 0; xmlChar *buffer = NULL; int buffer_size = 0; xmlChar *out = NULL; xmlChar *current = NULL; xmlEntityPtr ent; xmlChar cur; SHRINK; if (CUR == '"') { ctxt->instate = XML_PARSER_ATTRIBUTE_VALUE; limit = '"'; NEXT; } else if (CUR == '\'') { limit = '\''; ctxt->instate = XML_PARSER_ATTRIBUTE_VALUE; NEXT; } else { ctxt->errNo = XML_ERR_ATTRIBUTE_NOT_STARTED; if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "AttValue: \" or ' expected\n"); ctxt->wellFormed = 0; return(NULL); } /* * allocate a translation buffer. */ buffer_size = XML_PARSER_BUFFER_SIZE; buffer = (xmlChar *) xmlMalloc(buffer_size * sizeof(xmlChar)); if (buffer == NULL) { perror("xmlParseAttValue: malloc failed"); return(NULL); } out = buffer; /* * Ok loop until we reach one of the ending char or a size limit. */ cur = CUR; while ((ctxt->token != 0) || ((cur != limit) && (cur != '<'))) { if (cur == 0) break; if (cur == '&') { GROW; } if (ctxt->token == '&') { /* * The reparsing will be done in xmlStringGetNodeList() * called by the attribute() function in SAX.c */ static xmlChar quote[6] = "&"; if (out - buffer > buffer_size - 10) { int index = out - buffer; growBuffer(buffer); out = &buffer[index]; } current = "e[0]; while (*current != 0) { /* non input consuming */ *out++ = *current++; } NEXT; } else if ((cur == '&') && (NXT(1) == '#')) { int val = xmlOldParseCharRef(ctxt); *out++ = val; if (out - buffer > buffer_size - 10) { int index = out - buffer; growBuffer(buffer); out = &buffer[index]; } } else if (cur == '&') { ent = xmlOldParseEntityRef(ctxt); if ((ent != NULL) && (ctxt->replaceEntities != 0)) { current = ent->content; while (*current != 0) { *out++ = *current++; if (out - buffer > buffer_size - 10) { int index = out - buffer; growBuffer(buffer); out = &buffer[index]; } } } else if (ent != NULL) { int i = xmlStrlen(ent->name); const xmlChar *cur = ent->name; *out++ = '&'; while (out - buffer > buffer_size - i - 10) { int index = out - buffer; growBuffer(buffer); out = &buffer[index]; } for (;i > 0;i--) *out++ = *cur++; *out++ = ';'; } } else { /* invalid for UTF-8 , use COPY(out); !!!!!! */ if ((ctxt->token == 0) && ((cur == 0x20) || (cur == 0xD) || (cur == 0xA) || (cur == 0x9))) { *out++ = 0x20; if (out - buffer > buffer_size - 10) { int index = out - buffer; growBuffer(buffer); out = &buffer[index]; } } else { *out++ = cur; if (out - buffer > buffer_size - 10) { int index = out - buffer; growBuffer(buffer); out = &buffer[index]; } } NEXT; } cur = CUR; } *out++ = 0; if (CUR == '<') { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Unescaped '<' not allowed in attributes values\n"); ctxt->errNo = XML_ERR_LT_IN_ATTRIBUTE; ctxt->wellFormed = 0; } else if (CUR != limit) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "AttValue: ' expected\n"); ctxt->errNo = XML_ERR_ATTRIBUTE_NOT_FINISHED; ctxt->wellFormed = 0; } else NEXT; return(buffer); } /** * xmlOldParseSystemLiteral: * @ctxt: an XML parser context * * parse an XML Literal * * [11] SystemLiteral ::= ('"' [^"]* '"') | ("'" [^']* "'") * * Returns the SystemLiteral parsed or NULL */ static xmlChar * xmlOldParseSystemLiteral(xmlParserCtxtPtr ctxt) { xmlChar *buf = NULL; int len = 0; int size = XML_PARSER_BUFFER_SIZE; xmlChar cur; xmlChar stop; SHRINK; if (CUR == '"') { NEXT; stop = '"'; } else if (CUR == '\'') { NEXT; stop = '\''; } else { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "SystemLiteral \" or ' expected\n"); ctxt->errNo = XML_ERR_LITERAL_NOT_STARTED; ctxt->wellFormed = 0; return(NULL); } buf = (xmlChar *) xmlMalloc(size * sizeof(xmlChar)); if (buf == NULL) { fprintf(stderr, "malloc of %d byte failed\n", size); return(NULL); } cur = CUR; while ((IS_CHAR(cur)) && (cur != stop)) { if (len + 1 >= size) { size *= 2; buf = xmlRealloc(buf, size * sizeof(xmlChar)); if (buf == NULL) { fprintf(stderr, "realloc of %d byte failed\n", size); return(NULL); } } buf[len++] = cur; NEXT; cur = CUR; if (cur == 0) { GROW; SHRINK; cur = CUR; } } buf[len] = 0; if (!IS_CHAR(cur)) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Unfinished SystemLiteral\n"); ctxt->errNo = XML_ERR_LITERAL_NOT_FINISHED; ctxt->wellFormed = 0; } else { NEXT; } return(buf); } /** * xmlOldParsePubidLiteral: * @ctxt: an XML parser context * * parse an XML public literal * * [12] PubidLiteral ::= '"' PubidChar* '"' | "'" (PubidChar - "'")* "'" * * Returns the PubidLiteral parsed or NULL. */ static xmlChar * xmlOldParsePubidLiteral(xmlParserCtxtPtr ctxt) { xmlChar *buf = NULL; int len = 0; int size = XML_PARSER_BUFFER_SIZE; xmlChar cur; xmlChar stop; SHRINK; if (CUR == '"') { NEXT; stop = '"'; } else if (CUR == '\'') { NEXT; stop = '\''; } else { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "SystemLiteral \" or ' expected\n"); ctxt->errNo = XML_ERR_LITERAL_NOT_STARTED; ctxt->wellFormed = 0; return(NULL); } buf = (xmlChar *) xmlMalloc(size * sizeof(xmlChar)); if (buf == NULL) { fprintf(stderr, "malloc of %d byte failed\n", size); return(NULL); } cur = CUR; while ((IS_PUBIDCHAR(cur)) && (cur != stop)) { if (len + 1 >= size) { size *= 2; buf = xmlRealloc(buf, size * sizeof(xmlChar)); if (buf == NULL) { fprintf(stderr, "realloc of %d byte failed\n", size); return(NULL); } } buf[len++] = cur; NEXT; cur = CUR; if (cur == 0) { GROW; SHRINK; cur = CUR; } } buf[len] = 0; if (cur != stop) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Unfinished PubidLiteral\n"); ctxt->errNo = XML_ERR_LITERAL_NOT_FINISHED; ctxt->wellFormed = 0; } else { NEXT; } return(buf); } /** * xmlOldParseCharData: * @ctxt: an XML parser context * @cdata: int indicating whether we are within a CDATA section * * parse a CharData section. * if we are within a CDATA section ']]>' marks an end of section. * * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*) */ static void xmlOldParseCharData(xmlParserCtxtPtr ctxt, int cdata) { xmlChar buf[XML_PARSER_BIG_BUFFER_SIZE]; int nbchar = 0; xmlChar cur; SHRINK; cur = CUR; while (((cur != '<') || (ctxt->token == '<')) && ((cur != '&') || (ctxt->token == '&')) && (IS_CHAR(cur))) { if ((cur == ']') && (NXT(1) == ']') && (NXT(2) == '>')) { if (cdata) break; else { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->warning(ctxt->userData, "Sequence ']]>' not allowed in content\n"); ctxt->errNo = XML_ERR_MISPLACED_CDATA_END; } } buf[nbchar++] = CUR; if (nbchar == XML_PARSER_BIG_BUFFER_SIZE) { /* * Ok the segment is to be consumed as chars. */ if (ctxt->sax != NULL) { if (areBlanksOld(ctxt, buf, nbchar)) { if (ctxt->sax->ignorableWhitespace != NULL) ctxt->sax->ignorableWhitespace(ctxt->userData, buf, nbchar); } else { if (ctxt->sax->characters != NULL) ctxt->sax->characters(ctxt->userData, buf, nbchar); } } nbchar = 0; } NEXT; cur = CUR; } if (nbchar != 0) { /* * Ok the segment is to be consumed as chars. */ if (ctxt->sax != NULL) { if (areBlanksOld(ctxt, buf, nbchar)) { if (ctxt->sax->ignorableWhitespace != NULL) ctxt->sax->ignorableWhitespace(ctxt->userData, buf, nbchar); } else { if (ctxt->sax->characters != NULL) ctxt->sax->characters(ctxt->userData, buf, nbchar); } } } } /** * xmlOldParseExternalID: * @ctxt: an XML parser context * @publicID: a xmlChar** receiving PubidLiteral * @strict: indicate whether we should restrict parsing to only * production [75], see NOTE below * * Parse an External ID or a Public ID * * NOTE: Productions [75] and [83] interract badly since [75] can generate * 'PUBLIC' S PubidLiteral S SystemLiteral * * [75] ExternalID ::= 'SYSTEM' S SystemLiteral * | 'PUBLIC' S PubidLiteral S SystemLiteral * * [83] PublicID ::= 'PUBLIC' S PubidLiteral * * Returns the function returns SystemLiteral and in the second * case publicID receives PubidLiteral, is strict is off * it is possible to return NULL and have publicID set. */ static xmlChar * xmlOldParseExternalID(xmlParserCtxtPtr ctxt, xmlChar **publicID, int strict) { xmlChar *URI = NULL; SHRINK; if ((CUR == 'S') && (NXT(1) == 'Y') && (NXT(2) == 'S') && (NXT(3) == 'T') && (NXT(4) == 'E') && (NXT(5) == 'M')) { SKIP(6); if (!IS_BLANK(CUR)) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Space required after 'SYSTEM'\n"); ctxt->errNo = XML_ERR_SPACE_REQUIRED; ctxt->wellFormed = 0; } SKIP_BLANKS; URI = xmlOldParseSystemLiteral(ctxt); if (URI == NULL) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "xmlParseExternalID: SYSTEM, no URI\n"); ctxt->errNo = XML_ERR_URI_REQUIRED; ctxt->wellFormed = 0; } } else if ((CUR == 'P') && (NXT(1) == 'U') && (NXT(2) == 'B') && (NXT(3) == 'L') && (NXT(4) == 'I') && (NXT(5) == 'C')) { SKIP(6); if (!IS_BLANK(CUR)) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Space required after 'PUBLIC'\n"); ctxt->errNo = XML_ERR_SPACE_REQUIRED; ctxt->wellFormed = 0; } SKIP_BLANKS; *publicID = xmlOldParsePubidLiteral(ctxt); if (*publicID == NULL) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "xmlParseExternalID: PUBLIC, no Public Identifier\n"); ctxt->errNo = XML_ERR_PUBID_REQUIRED; ctxt->wellFormed = 0; } if (strict) { /* * We don't handle [83] so "S SystemLiteral" is required. */ if (!IS_BLANK(CUR)) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Space required after the Public Identifier\n"); ctxt->errNo = XML_ERR_SPACE_REQUIRED; ctxt->wellFormed = 0; } } else { /* * We handle [83] so we return immediately, if * "S SystemLiteral" is not detected. From a purely parsing * point of view that's a nice mess. */ const xmlChar *ptr; GROW; ptr = CUR_PTR; if (!IS_BLANK(*ptr)) return(NULL); while (IS_BLANK(*ptr)) ptr++; if ((*ptr != '\'') || (*ptr != '"')) return(NULL); } SKIP_BLANKS; URI = xmlOldParseSystemLiteral(ctxt); if (URI == NULL) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "xmlParseExternalID: PUBLIC, no URI\n"); ctxt->errNo = XML_ERR_URI_REQUIRED; ctxt->wellFormed = 0; } } return(URI); } /** * xmlOldParseComment: * @ctxt: an XML parser context * * Skip an XML (SGML) comment * The spec says that "For compatibility, the string "--" (double-hyphen) * must not occur within comments. " * * [15] Comment ::= '' */ static void xmlOldParseComment(xmlParserCtxtPtr ctxt) { xmlChar *buf = NULL; int len = 0; int size = XML_PARSER_BUFFER_SIZE; xmlChar q; xmlChar r; xmlChar cur; xmlParserInputState state; /* * Check that there is a comment right here. */ if ((CUR != '<') || (NXT(1) != '!') || (NXT(2) != '-') || (NXT(3) != '-')) return; state = ctxt->instate; ctxt->instate = XML_PARSER_COMMENT; SHRINK; SKIP(4); buf = (xmlChar *) xmlMalloc(size * sizeof(xmlChar)); if (buf == NULL) { fprintf(stderr, "malloc of %d byte failed\n", size); ctxt->instate = state; return; } q = CUR; NEXT; r = CUR; NEXT; cur = CUR; while (IS_CHAR(cur) && ((cur != '>') || (r != '-') || (q != '-'))) { if ((r == '-') && (q == '-')) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Comment must not contain '--' (double-hyphen)`\n"); ctxt->errNo = XML_ERR_HYPHEN_IN_COMMENT; ctxt->wellFormed = 0; } if (len + 1 >= size) { size *= 2; buf = xmlRealloc(buf, size * sizeof(xmlChar)); if (buf == NULL) { fprintf(stderr, "realloc of %d byte failed\n", size); ctxt->instate = state; return; } } buf[len++] = q; q = r; r = cur; NEXT; cur = CUR; if (cur == 0) { SHRINK; GROW; cur = CUR; } } buf[len] = 0; if (!IS_CHAR(cur)) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Comment not terminated \n