/* * parser.c : an XML 1.0 non-verifying parser * * See Copyright for the status of this software. * * Daniel.Veillard@w3.org */ #ifdef WIN32 #define HAVE_FCNTL_H #include #else #include #endif #include #include #include /* for memset() only */ #include #include #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" const char *xmlParserVersion = LIBXML_VERSION; /************************************************************************ * * * Input handling functions for progressive parsing * * * ************************************************************************/ /* #define DEBUG_INPUT */ #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) void 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 /** * xmlParserInputRead: * @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 CHARs read, or -1 in case of error, 0 indicate the * end of this entity */ int xmlParserInputRead(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); } /** * xmlParserInputGrow: * @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 CHARs read, or -1 in case of error, 0 indicate the * end of this entity */ int xmlParserInputGrow(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); } ret = xmlParserInputBufferGrow(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); } /** * xmlParserInputShrink: * @in: an XML parser input * * This function removes used input for the parser. */ void xmlParserInputShrink(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 * * * ************************************************************************/ int xmlSubstituteEntitiesDefaultValue = 0; int xmlDoValidityCheckingDefaultValue = 0; /* * Generic function for accessing stacks in the Parser Context */ #define PUSH_AND_POP(type, name) \ extern int name##Push(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"); \ exit(1); \ } \ } \ ctxt->name##Tab[ctxt->name##Nr] = value; \ ctxt->name = value; \ return(ctxt->name##Nr++); \ } \ extern type name##Pop(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(xmlParserInputPtr, input) PUSH_AND_POP(xmlNodePtr, node) /* * 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 CHAR to be parsed. * CUR returns the current CHAR 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 CHAR. Same as CUR is should be used only * to compare on ASCII based substring. * SKIP(n) Skip n CHAR, 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 CUR (ctxt->token ? ctxt->token : (*ctxt->input->cur)) #define SKIP(val) ctxt->input->cur += (val) #define NXT(val) ctxt->input->cur[(val)] #define CUR_PTR ctxt->input->cur #define SHRINK xmlParserInputShrink(ctxt->input); \ if ((*ctxt->input->cur == 0) && \ (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) \ xmlPopInput(ctxt) #define GROW xmlParserInputGrow(ctxt->input, INPUT_CHUNK); \ if ((*ctxt->input->cur == 0) && \ (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) \ xmlPopInput(ctxt) #define SKIP_BLANKS \ do { \ while (IS_BLANK(CUR)) NEXT; \ if (*ctxt->input->cur == '%') xmlParserHandlePEReference(ctxt); \ if (*ctxt->input->cur == '&') xmlParserHandleReference(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) && \ (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) { \ xmlPopInput(ctxt); \ } else { \ if (*(ctxt->input->cur) == '\n') { \ ctxt->input->line++; ctxt->input->col = 1; \ } else ctxt->input->col++; \ ctxt->input->cur++; \ if (*ctxt->input->cur == 0) \ xmlParserInputGrow(ctxt->input, INPUT_CHUNK); \ } \ if (*ctxt->input->cur == '%') xmlParserHandlePEReference(ctxt); \ if (*ctxt->input->cur == '&') xmlParserHandleReference(ctxt); \ }} /************************************************************************ * * * Commodity functions to handle entities processing * * * ************************************************************************/ /** * xmlPopInput: * @ctxt: an XML parser context * * xmlPopInput: the current input pointed by ctxt->input came to an end * pop it and return the next char. * * Returns the current CHAR in the parser context */ CHAR xmlPopInput(xmlParserCtxtPtr ctxt) { if (ctxt->inputNr == 1) return(0); /* End of main Input */ xmlFreeInputStream(inputPop(ctxt)); if ((*ctxt->input->cur == 0) && (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) return(xmlPopInput(ctxt)); return(CUR); } /** * xmlPushInput: * @ctxt: an XML parser context * @input: an XML parser input fragment (entity, XML fragment ...). * * xmlPushInput: switch to a new input stream which is stacked on top * of the previous one(s). */ void xmlPushInput(xmlParserCtxtPtr ctxt, xmlParserInputPtr input) { if (input == NULL) return; inputPush(ctxt, input); } /** * xmlFreeInputStream: * @input: an xmlP arserInputPtr * * Free up an input stream. */ void xmlFreeInputStream(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((CHAR *) input->base); if (input->buf != NULL) xmlFreeParserInputBuffer(input->buf); memset(input, -1, sizeof(xmlParserInput)); xmlFree(input); } /** * xmlNewInputStream: * @ctxt: an XML parser context * * Create a new input stream structure * Returns the new input stream or NULL */ xmlParserInputPtr xmlNewInputStream(xmlParserCtxtPtr ctxt) { xmlParserInputPtr input; input = (xmlParserInputPtr) xmlMalloc(sizeof(xmlParserInput)); if (input == NULL) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "malloc: couldn't allocate a new input stream\n"); 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; return(input); } /** * xmlNewEntityInputStream: * @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 */ xmlParserInputPtr xmlNewEntityInputStream(xmlParserCtxtPtr ctxt, xmlEntityPtr entity) { xmlParserInputPtr input; if (entity == NULL) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "internal: xmlNewEntityInputStream entity = NULL\n"); return(NULL); } if (entity->content == NULL) { switch (entity->type) { case XML_EXTERNAL_GENERAL_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->input)); 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: 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: 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 = xmlNewInputStream(ctxt); if (input == NULL) { return(NULL); } input->filename = (char *) entity->SystemID; /* TODO !!! char <- CHAR */ input->base = entity->content; input->cur = entity->content; return(input); } /** * xmlNewStringInputStream: * @ctxt: an XML parser context * @buffer: an memory buffer * * Create a new input stream based on a memory buffer. * Returns the new input stream */ xmlParserInputPtr xmlNewStringInputStream(xmlParserCtxtPtr ctxt, const CHAR *buffer) { xmlParserInputPtr input; if (buffer == NULL) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "internal: xmlNewStringInputStream string = NULL\n"); return(NULL); } input = xmlNewInputStream(ctxt); if (input == NULL) { return(NULL); } input->base = buffer; input->cur = buffer; return(input); } /** * xmlNewInputFromFile: * @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 */ xmlParserInputPtr xmlNewInputFromFile(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[1024]; 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 = xmlMemStrdup(ctxt->input->directory); } 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 = xmlMemStrdup(ctxt->directory); } if (buf == NULL) return(NULL); } if (directory == NULL) directory = xmlParserGetDirectory(filename); inputStream = xmlNewInputStream(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 = directory; return(inputStream); } /************************************************************************ * * * Commodity functions to handle parser contexts * * * ************************************************************************/ /** * xmlInitParserCtxt: * @ctxt: an XML parser context * * Initialize a parser context */ void xmlInitParserCtxt(xmlParserCtxtPtr ctxt) { xmlSAXHandler *sax; sax = (xmlSAXHandler *) xmlMalloc(sizeof(xmlSAXHandler)); if (sax == NULL) { fprintf(stderr, "xmlInitParserCtxt: out of memory\n"); } /* Allocate the Input stack */ ctxt->inputTab = (xmlParserInputPtr *) xmlMalloc(5 * sizeof(xmlParserInputPtr)); ctxt->inputNr = 0; ctxt->inputMax = 5; ctxt->input = NULL; ctxt->version = NULL; ctxt->encoding = NULL; ctxt->standalone = -1; ctxt->hasExternalSubset = 0; ctxt->hasPErefs = 0; ctxt->html = 0; ctxt->external = 0; ctxt->instate = XML_PARSER_PROLOG; ctxt->token = 0; ctxt->directory = NULL; /* Allocate the Node stack */ ctxt->nodeTab = (xmlNodePtr *) xmlMalloc(10 * sizeof(xmlNodePtr)); ctxt->nodeNr = 0; ctxt->nodeMax = 10; ctxt->node = NULL; if (sax == NULL) ctxt->sax = &xmlDefaultSAXHandler; else { ctxt->sax = sax; memcpy(sax, &xmlDefaultSAXHandler, sizeof(xmlSAXHandler)); } ctxt->userData = ctxt; ctxt->myDoc = NULL; ctxt->wellFormed = 1; ctxt->valid = 1; ctxt->validate = xmlDoValidityCheckingDefaultValue; ctxt->vctxt.userData = ctxt; ctxt->vctxt.error = xmlParserValidityError; ctxt->vctxt.warning = xmlParserValidityWarning; ctxt->replaceEntities = xmlSubstituteEntitiesDefaultValue; ctxt->record_info = 0; xmlInitNodeInfoSeq(&ctxt->node_seq); } /** * xmlFreeParserCtxt: * @ctxt: an XML parser context * * Free all the memory used by a parser context. However the parsed * document in ctxt->myDoc is not freed. */ void xmlFreeParserCtxt(xmlParserCtxtPtr ctxt) { xmlParserInputPtr input; if (ctxt == NULL) return; while ((input = inputPop(ctxt)) != NULL) { xmlFreeInputStream(input); } if (ctxt->nodeTab != NULL) xmlFree(ctxt->nodeTab); if (ctxt->inputTab != NULL) xmlFree(ctxt->inputTab); if (ctxt->version != NULL) xmlFree((char *) ctxt->version); if (ctxt->encoding != NULL) xmlFree((char *) ctxt->encoding); if ((ctxt->sax != NULL) && (ctxt->sax != &xmlDefaultSAXHandler)) xmlFree(ctxt->sax); if (ctxt->directory != NULL) xmlFree((char *) ctxt->directory); xmlFree(ctxt); } /** * xmlNewParserCtxt: * * Allocate and initialize a new parser context. * * Returns the xmlParserCtxtPtr or NULL */ xmlParserCtxtPtr xmlNewParserCtxt() { xmlParserCtxtPtr ctxt; ctxt = (xmlParserCtxtPtr) xmlMalloc(sizeof(xmlParserCtxt)); if (ctxt == NULL) { fprintf(stderr, "xmlNewParserCtxt : cannot allocate context\n"); perror("malloc"); return(NULL); } xmlInitParserCtxt(ctxt); return(ctxt); } /** * xmlClearParserCtxt: * @ctxt: an XML parser context * * Clear (release owned resources) and reinitialize a parser context */ void xmlClearParserCtxt(xmlParserCtxtPtr ctxt) { xmlClearNodeInfoSeq(&ctxt->node_seq); xmlInitParserCtxt(ctxt); } /************************************************************************ * * * Commodity functions to handle entities * * * ************************************************************************/ void xmlParserHandleReference(xmlParserCtxtPtr ctxt); void xmlParserHandlePEReference(xmlParserCtxtPtr ctxt); /** * xmlParseCharRef: * @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) */ int xmlParseCharRef(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 { 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 == ';') NEXT; } else if ((CUR == '&') && (NXT(1) == '#')) { SKIP(2); while (CUR != ';') { if ((CUR >= '0') && (CUR <= '9')) val = val * 10 + (CUR - '0'); else { 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 == ';') NEXT; } else { 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 { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "CharRef: invalid CHAR value %d\n", val); ctxt->wellFormed = 0; } return(0); } /** * xmlParserHandleReference: * @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 */ void xmlParserHandleReference(xmlParserCtxtPtr ctxt) { xmlParserInputPtr input; CHAR *name; xmlEntityPtr ent = NULL; if (ctxt->token != 0) return; if (CUR != '&') return; GROW; if ((CUR == '&') && (NXT(1) == '#')) { switch(ctxt->instate) { case XML_PARSER_CDATA_SECTION: return; case XML_PARSER_COMMENT: return; case XML_PARSER_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: 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: 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: 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_DECL: /* we just ignore it there */ return; case XML_PARSER_ENTITY_VALUE: /* * NOTE: in the case of entity values, we don't do the * substitution here since we need the litteral * entity value to be able to save the internal * subset of the document. * This will be handled by xmlDecodeEntities */ return; case XML_PARSER_CONTENT: case XML_PARSER_ATTRIBUTE_VALUE: /* !!! this may not be Ok for UTF-8, multibyte sequence */ ctxt->token = xmlParseCharRef(ctxt); return; } return; } switch(ctxt->instate) { case XML_PARSER_CDATA_SECTION: return; case XML_PARSER_COMMENT: return; case XML_PARSER_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: 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: 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 litteral * entity value to be able to save the internal * subset of the document. * This will be handled by xmlDecodeEntities */ 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 xmlDecodeEntities */ 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: 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 = xmlScanName(ctxt); if (name == NULL) { 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)) != ';') { 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) { 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) { 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 = xmlNewEntityInputStream(ctxt, ent); xmlPushInput(ctxt, input); xmlFree(name); return; } /** * xmlParserHandlePEReference: * @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 */ void xmlParserHandlePEReference(xmlParserCtxtPtr ctxt) { CHAR *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_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: 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: /* we just ignore it there */ return; case XML_PARSER_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 litteral * entity value to be able to save the internal * subset of the document. * This will be handled by xmlDecodeEntities */ 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 xmlParseMarkupDecl */ if ((ctxt->external == 0) && (ctxt->inputNr == 1)) return; } NEXT; name = xmlParseName(ctxt); if (name == NULL) { 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 = xmlNewEntityInputStream(ctxt, entity); xmlPushInput(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 { 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 = (CHAR *) xmlRealloc(buffer, buffer##_size * sizeof(CHAR)); \ if (buffer == NULL) { \ perror("realloc failed"); \ exit(1); \ } \ } /** * xmlDecodeEntities: * @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 CHAR, 0 if none * @end2: an end marker CHAR, 0 if none * @end3: an end marker CHAR, 0 if none * * [67] Reference ::= EntityRef | CharRef * * [69] PEReference ::= '%' Name ';' * * Returns A newly allocated string with the substitution done. The caller * must deallocate it ! */ CHAR * xmlDecodeEntities(xmlParserCtxtPtr ctxt, int len, int what, CHAR end, CHAR end2, CHAR end3) { CHAR *buffer = NULL; int buffer_size = 0; CHAR *out = NULL; CHAR *current = NULL; xmlEntityPtr ent; int nbchars = 0; unsigned int max = (unsigned int) len; CHAR cur; /* * allocate a translation buffer. */ buffer_size = 1000; buffer = (CHAR *) xmlMalloc(buffer_size * sizeof(CHAR)); 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 = xmlParseCharRef(ctxt); *out++ = val; nbchars += 3; } else if ((cur == '&') && (what & XML_SUBSTITUTE_REF)) { ent = xmlParseEntityRef(ctxt); if ((ent != NULL) && (ctxt->replaceEntities != 0)) { current = ent->content; while (*current != 0) { *out++ = *current++; if (out - buffer > buffer_size - 100) { int index = out - buffer; growBuffer(buffer); out = &buffer[index]; } } nbchars += 3 + xmlStrlen(ent->name); } else if (ent != NULL) { int i = xmlStrlen(ent->name); const CHAR *cur = ent->name; nbchars += i + 2; *out++ = '&'; if (out - buffer > buffer_size - i - 100) { 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; xmlParsePEReference(ctxt); /* * Pop-up of finished entities. */ while ((CUR == 0) && (ctxt->inputNr > 1)) xmlPopInput(ctxt); break; } else { /* invalid for UTF-8 , use COPY(out); !!!!!! */ *out++ = cur; nbchars++; if (out - buffer > buffer_size - 100) { int index = out - buffer; growBuffer(buffer); out = &buffer[index]; } NEXT; } cur = CUR; } *out++ = 0; return(buffer); } /************************************************************************ * * * Commodity functions to handle encodings * * * ************************************************************************/ /** * xmlSwitchEncoding: * @ctxt: the parser context * @len: the len of @cur * * change the input functions when discovering the character encoding * of a given entity. * */ void xmlSwitchEncoding(xmlParserCtxtPtr ctxt, xmlCharEncoding enc) { switch (enc) { case XML_CHAR_ENCODING_ERROR: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "encoding unknown\n"); ctxt->wellFormed = 0; break; case XML_CHAR_ENCODING_NONE: /* let's assume it's UTF-8 without the XML decl */ return; case XML_CHAR_ENCODING_UTF8: /* default encoding, no conversion should be needed */ return; case XML_CHAR_ENCODING_UTF16LE: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "char encoding UTF16 little endian not supported\n"); break; case XML_CHAR_ENCODING_UTF16BE: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "char encoding UTF16 big endian not supported\n"); break; case XML_CHAR_ENCODING_UCS4LE: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "char encoding USC4 little endian not supported\n"); break; case XML_CHAR_ENCODING_UCS4BE: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "char encoding USC4 big endian not supported\n"); break; case XML_CHAR_ENCODING_EBCDIC: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "char encoding EBCDIC not supported\n"); break; case XML_CHAR_ENCODING_UCS4_2143: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "char encoding UCS4 2143 not supported\n"); break; case XML_CHAR_ENCODING_UCS4_3412: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "char encoding UCS4 3412 not supported\n"); break; case XML_CHAR_ENCODING_UCS2: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "char encoding UCS2 not supported\n"); break; case XML_CHAR_ENCODING_8859_1: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "char encoding ISO_8859_1 ISO Latin 1 not supported\n"); break; case XML_CHAR_ENCODING_8859_2: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "char encoding ISO_8859_2 ISO Latin 2 not supported\n"); break; case XML_CHAR_ENCODING_8859_3: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "char encoding ISO_8859_3 not supported\n"); break; case XML_CHAR_ENCODING_8859_4: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "char encoding ISO_8859_4 not supported\n"); break; case XML_CHAR_ENCODING_8859_5: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "char encoding ISO_8859_5 not supported\n"); break; case XML_CHAR_ENCODING_8859_6: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "char encoding ISO_8859_6 not supported\n"); break; case XML_CHAR_ENCODING_8859_7: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "char encoding ISO_8859_7 not supported\n"); break; case XML_CHAR_ENCODING_8859_8: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "char encoding ISO_8859_8 not supported\n"); break; case XML_CHAR_ENCODING_8859_9: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "char encoding ISO_8859_9 not supported\n"); break; case XML_CHAR_ENCODING_2022_JP: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "char encoding ISO-2022-JPnot supported\n"); break; case XML_CHAR_ENCODING_SHIFT_JIS: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "char encoding Shift_JISnot supported\n"); break; case XML_CHAR_ENCODING_EUC_JP: if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "char encoding EUC-JPnot supported\n"); break; } } /************************************************************************ * * * Commodity functions to handle CHARs * * * ************************************************************************/ /** * xmlStrndup: * @cur: the input CHAR * * @len: the len of @cur * * a strndup for array of CHAR's * * Returns a new CHAR * or NULL */ CHAR * xmlStrndup(const CHAR *cur, int len) { CHAR *ret = xmlMalloc((len + 1) * sizeof(CHAR)); if (ret == NULL) { fprintf(stderr, "malloc of %ld byte failed\n", (len + 1) * (long)sizeof(CHAR)); return(NULL); } memcpy(ret, cur, len * sizeof(CHAR)); ret[len] = 0; return(ret); } /** * xmlStrdup: * @cur: the input CHAR * * * a strdup for array of CHAR's * * Returns a new CHAR * or NULL */ CHAR * xmlStrdup(const CHAR *cur) { const CHAR *p = cur; while (IS_CHAR(*p)) p++; return(xmlStrndup(cur, p - cur)); } /** * xmlCharStrndup: * @cur: the input char * * @len: the len of @cur * * a strndup for char's to CHAR's * * Returns a new CHAR * or NULL */ CHAR * xmlCharStrndup(const char *cur, int len) { int i; CHAR *ret = xmlMalloc((len + 1) * sizeof(CHAR)); if (ret == NULL) { fprintf(stderr, "malloc of %ld byte failed\n", (len + 1) * (long)sizeof(CHAR)); return(NULL); } for (i = 0;i < len;i++) ret[i] = (CHAR) cur[i]; ret[len] = 0; return(ret); } /** * xmlCharStrdup: * @cur: the input char * * @len: the len of @cur * * a strdup for char's to CHAR's * * Returns a new CHAR * or NULL */ CHAR * xmlCharStrdup(const char *cur) { const char *p = cur; while (*p != '\0') p++; return(xmlCharStrndup(cur, p - cur)); } /** * xmlStrcmp: * @str1: the first CHAR * * @str2: the second CHAR * * * a strcmp for CHAR's * * Returns the integer result of the comparison */ int xmlStrcmp(const CHAR *str1, const CHAR *str2) { register int tmp; do { tmp = *str1++ - *str2++; if (tmp != 0) return(tmp); } while ((*str1 != 0) && (*str2 != 0)); return (*str1 - *str2); } /** * xmlStrncmp: * @str1: the first CHAR * * @str2: the second CHAR * * @len: the max comparison length * * a strncmp for CHAR's * * Returns the integer result of the comparison */ int xmlStrncmp(const CHAR *str1, const CHAR *str2, int len) { register int tmp; if (len <= 0) return(0); do { tmp = *str1++ - *str2++; if (tmp != 0) return(tmp); len--; if (len <= 0) return(0); } while ((*str1 != 0) && (*str2 != 0)); return (*str1 - *str2); } /** * xmlStrchr: * @str: the CHAR * array * @val: the CHAR to search * * a strchr for CHAR's * * Returns the CHAR * for the first occurence or NULL. */ const CHAR * xmlStrchr(const CHAR *str, CHAR val) { while (*str != 0) { if (*str == val) return((CHAR *) str); str++; } return(NULL); } /** * xmlStrstr: * @str: the CHAR * array (haystack) * @val: the CHAR to search (needle) * * a strstr for CHAR's * * Returns the CHAR * for the first occurence or NULL. */ const CHAR * xmlStrstr(const CHAR *str, CHAR *val) { int n; if (str == NULL) return(NULL); if (val == NULL) return(NULL); n = xmlStrlen(val); if (n == 0) return(str); while (*str != 0) { if (*str == *val) { if (!xmlStrncmp(str, val, n)) return((const CHAR *) str); } str++; } return(NULL); } /** * xmlStrsub: * @str: the CHAR * array (haystack) * @start: the index of the first char (zero based) * @len: the length of the substring * * Extract a substring of a given string * * Returns the CHAR * for the first occurence or NULL. */ CHAR * xmlStrsub(const CHAR *str, int start, int len) { int i; if (str == NULL) return(NULL); if (start < 0) return(NULL); if (len < 0) return(NULL); for (i = 0;i < start;i++) { if (*str == 0) return(NULL); str++; } if (*str == 0) return(NULL); return(xmlStrndup(str, len)); } /** * xmlStrlen: * @str: the CHAR * array * * lenght of a CHAR's string * * Returns the number of CHAR contained in the ARRAY. */ int xmlStrlen(const CHAR *str) { int len = 0; if (str == NULL) return(0); while (*str != 0) { str++; len++; } return(len); } /** * xmlStrncat: * @cur: the original CHAR * array * @add: the CHAR * array added * @len: the length of @add * * a strncat for array of CHAR's * * Returns a new CHAR * containing the concatenated string. */ CHAR * xmlStrncat(CHAR *cur, const CHAR *add, int len) { int size; CHAR *ret; if ((add == NULL) || (len == 0)) return(cur); if (cur == NULL) return(xmlStrndup(add, len)); size = xmlStrlen(cur); ret = xmlRealloc(cur, (size + len + 1) * sizeof(CHAR)); if (ret == NULL) { fprintf(stderr, "xmlStrncat: realloc of %ld byte failed\n", (size + len + 1) * (long)sizeof(CHAR)); return(cur); } memcpy(&ret[size], add, len * sizeof(CHAR)); ret[size + len] = 0; return(ret); } /** * xmlStrcat: * @cur: the original CHAR * array * @add: the CHAR * array added * * a strcat for array of CHAR's * * Returns a new CHAR * containing the concatenated string. */ CHAR * xmlStrcat(CHAR *cur, const CHAR *add) { const CHAR *p = add; if (add == NULL) return(cur); if (cur == NULL) return(xmlStrdup(add)); while (IS_CHAR(*p)) p++; return(xmlStrncat(cur, add, p - add)); } /************************************************************************ * * * Commodity functions, cleanup needed ? * * * ************************************************************************/ /** * areBlanks: * @ctxt: an XML parser context * @str: a CHAR * * @len: the size of @str * * Is this a sequence of blank chars that one can ignore ? * * TODO: Whether white space are significant has to be checked accordingly * to DTD informations if available * * Returns 1 if ignorable 0 otherwise. */ static int areBlanks(xmlParserCtxtPtr ctxt, const CHAR *str, int len) { int i, ret; xmlNodePtr lastChild; for (i = 0;i < len;i++) if (!(IS_BLANK(str[i]))) return(0); if (CUR != '<') return(0); if (ctxt->node == NULL) return(0); if (ctxt->myDoc != NULL) { ret = xmlIsMixedElement(ctxt->myDoc, ctxt->node->name); if (ret == 0) return(1); if (ret == 1) return(0); } /* * heuristic */ 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); } /** * xmlHandleEntity: * @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. */ void xmlHandleEntity(xmlParserCtxtPtr ctxt, xmlEntityPtr entity) { int len; xmlParserInputPtr input; if (entity->content == NULL) { 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 = xmlNewEntityInputStream(ctxt, entity); xmlPushInput(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); } /* * Forward definition for recusive behaviour. */ void xmlParsePEReference(xmlParserCtxtPtr ctxt); void xmlParseReference(xmlParserCtxtPtr ctxt); /************************************************************************ * * * Extra stuff for namespace support * * Relates to http://www.w3.org/TR/WD-xml-names * * * ************************************************************************/ /** * xmlNamespaceParseNCName: * @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 */ CHAR * xmlNamespaceParseNCName(xmlParserCtxtPtr ctxt) { CHAR 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)); } /** * xmlNamespaceParseQName: * @ctxt: an XML parser context * @prefix: a CHAR ** * * parse an XML qualified name * * [NS 5] QName ::= (Prefix ':')? LocalPart * * [NS 6] Prefix ::= NCName * * [NS 7] LocalPart ::= NCName * * Returns the function returns the local part, and prefix is updated * to get the Prefix if any. */ CHAR * xmlNamespaceParseQName(xmlParserCtxtPtr ctxt, CHAR **prefix) { CHAR *ret = NULL; *prefix = NULL; ret = xmlNamespaceParseNCName(ctxt); if (CUR == ':') { *prefix = ret; NEXT; ret = xmlNamespaceParseNCName(ctxt); } return(ret); } /** * xmlSplitQName: * @name: an XML parser context * @prefix: a CHAR ** * * parse an XML qualified name string * * [NS 5] QName ::= (Prefix ':')? LocalPart * * [NS 6] Prefix ::= NCName * * [NS 7] LocalPart ::= NCName * * Returns the function returns the local part, and prefix is updated * to get the Prefix if any. */ CHAR * xmlSplitQName(const CHAR *name, CHAR **prefix) { CHAR *ret = NULL; const CHAR *q; const CHAR *cur = name; *prefix = NULL; /* xml: prefix is not really a namespace */ if ((cur[0] == 'x') && (cur[1] == 'm') && (cur[2] == 'l') && (cur[3] == ':')) return(xmlStrdup(name)); if (!IS_LETTER(*cur) && (*cur != '_')) return(NULL); q = cur++; while ((IS_LETTER(*cur)) || (IS_DIGIT(*cur)) || (*cur == '.') || (*cur == '-') || (*cur == '_') || (IS_COMBINING(*cur)) || (IS_EXTENDER(*cur))) cur++; ret = xmlStrndup(q, cur - q); if (*cur == ':') { cur++; if (!IS_LETTER(*cur) && (*cur != '_')) return(ret); *prefix = ret; q = cur++; while ((IS_LETTER(*cur)) || (IS_DIGIT(*cur)) || (*cur == '.') || (*cur == '-') || (*cur == '_') || (IS_COMBINING(*cur)) || (IS_EXTENDER(*cur))) cur++; ret = xmlStrndup(q, cur - q); } return(ret); } /** * xmlNamespaceParseNSDef: * @ctxt: an XML parser context * * parse a namespace prefix declaration * * [NS 1] NSDef ::= PrefixDef Eq SystemLiteral * * [NS 2] PrefixDef ::= 'xmlns' (':' NCName)? * * Returns the namespace name */ CHAR * xmlNamespaceParseNSDef(xmlParserCtxtPtr ctxt) { CHAR *name = NULL; if ((CUR == 'x') && (NXT(1) == 'm') && (NXT(2) == 'l') && (NXT(3) == 'n') && (NXT(4) == 's')) { SKIP(5); if (CUR == ':') { NEXT; name = xmlNamespaceParseNCName(ctxt); } } return(name); } /** * xmlParseQuotedString: * @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. */ CHAR * xmlParseQuotedString(xmlParserCtxtPtr ctxt) { CHAR *ret = NULL; const CHAR *q; if (CUR == '"') { NEXT; q = CUR_PTR; while (IS_CHAR(CUR) && (CUR != '"')) NEXT; if (CUR != '"') { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "String not closed \"%.50s\"\n", q); ctxt->wellFormed = 0; } else { ret = xmlStrndup(q, CUR_PTR - q); NEXT; } } else if (CUR == '\''){ NEXT; q = CUR_PTR; while (IS_CHAR(CUR) && (CUR != '\'')) NEXT; if (CUR != '\'') { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "String not closed \"%.50s\"\n", q); ctxt->wellFormed = 0; } else { ret = xmlStrndup(q, CUR_PTR - q); NEXT; } } return(ret); } /** * xmlParseNamespace: * @ctxt: an XML parser context * * [OLD] xmlParseNamespace: 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 = xmlParseQuotedString(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 = xmlParseQuotedString(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 = xmlParseQuotedString(ctxt); SKIP_BLANKS; } else if ((CUR == 'A') && (NXT(1) == 'S')) { garbage = 0; SKIP(2); SKIP_BLANKS; if (CUR != '=') continue; NEXT; SKIP_BLANKS; prefix = xmlParseQuotedString(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->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 * * * ************************************************************************/ /** * xmlScanName: * @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 */ CHAR * xmlScanName(xmlParserCtxtPtr ctxt) { CHAR 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)); } /** * xmlParseName: * @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 */ CHAR * xmlParseName(xmlParserCtxtPtr ctxt) { CHAR buf[XML_MAX_NAMELEN]; int len = 0; CHAR 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)); } /** * xmlParseNmtoken: * @ctxt: an XML parser context * * parse an XML Nmtoken. * * [7] Nmtoken ::= (NameChar)+ * * [8] Nmtokens ::= Nmtoken (S Nmtoken)* * * Returns the Nmtoken parsed or NULL */ CHAR * xmlParseNmtoken(xmlParserCtxtPtr ctxt) { CHAR 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)); } /** * xmlParseEntityValue: * @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 */ CHAR * xmlParseEntityValue(xmlParserCtxtPtr ctxt, CHAR **orig) { CHAR *ret = NULL; const CHAR *org = NULL; const CHAR *tst = NULL; const CHAR *temp = NULL; xmlParserInputPtr input; SHRINK; if (CUR == '"') { ctxt->instate = XML_PARSER_ENTITY_VALUE; input = ctxt->input; NEXT; org = CUR_PTR; /* * 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 ((CUR != '"') || (ctxt->input != input)) { tst = CUR_PTR; /* * 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. */ if (ctxt->input != input) temp = xmlDecodeEntities(ctxt, -1, XML_SUBSTITUTE_PEREF, 0, 0, 0); else temp = xmlDecodeEntities(ctxt, -1, XML_SUBSTITUTE_PEREF, '"', 0, 0); /* * Pop-up of finished entities. */ while ((CUR == 0) && (ctxt->inputNr > 1)) xmlPopInput(ctxt); if ((temp == NULL) && (tst == CUR_PTR)) { ret = xmlStrndup((CHAR *) "", 0); break; } if ((temp[0] == 0) && (tst == CUR_PTR)) { xmlFree((char *)temp); ret = xmlStrndup((CHAR *) "", 0); break; } ret = xmlStrcat(ret, temp); if (temp != NULL) xmlFree((char *)temp); GROW; } if (CUR != '"') { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "EntityValue: \" expected\n"); ctxt->wellFormed = 0; } else { if (orig != NULL) /* !!!!!!!!! */ *orig = xmlStrndup(org, CUR_PTR - org); if (ret == NULL) ret = xmlStrndup((CHAR *) "", 0); NEXT; } } else if (CUR == '\'') { ctxt->instate = XML_PARSER_ENTITY_VALUE; input = ctxt->input; NEXT; org = CUR_PTR; /* * 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 ((CUR != '\'') || (ctxt->input != input)) { tst = CUR_PTR; /* * 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. */ if (ctxt->input != input) temp = xmlDecodeEntities(ctxt, -1, XML_SUBSTITUTE_PEREF, 0, 0, 0); else temp = xmlDecodeEntities(ctxt, -1, XML_SUBSTITUTE_PEREF, '\'', 0, 0); /* * Pop-up of finished entities. */ while ((CUR == 0) && (ctxt->inputNr > 1)) xmlPopInput(ctxt); if ((temp == NULL) && (tst == CUR_PTR)) { ret = xmlStrndup((CHAR *) "", 0); break; } if ((temp[0] == 0) && (tst == CUR_PTR)) { xmlFree((char *)temp); ret = xmlStrndup((CHAR *) "", 0); break; } ret = xmlStrcat(ret, temp); if (temp != NULL) xmlFree((char *)temp); GROW; } if (CUR != '\'') { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "EntityValue: ' expected\n"); ctxt->wellFormed = 0; } else { if (orig != NULL) /* !!!!!!!!! */ *orig = xmlStrndup(org, CUR_PTR - org); if (ret == NULL) ret = xmlStrndup((CHAR *) "", 0); NEXT; } } else { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "EntityValue: \" or ' expected\n"); ctxt->wellFormed = 0; } return(ret); } /** * xmlParseAttValue: * @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)* "'" * * Returns the AttValue parsed or NULL. */ CHAR * xmlParseAttValue(xmlParserCtxtPtr ctxt) { CHAR *ret = NULL; SHRINK; if (CUR == '"') { ctxt->instate = XML_PARSER_ATTRIBUTE_VALUE; NEXT; ret = xmlDecodeEntities(ctxt, -1, XML_SUBSTITUTE_REF, '"', '<', 0); if (CUR == '<') { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Unescaped '<' not allowed in attributes values\n"); ctxt->wellFormed = 0; } if (CUR != '"') { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "AttValue: ' expected\n"); ctxt->wellFormed = 0; } else NEXT; } else if (CUR == '\'') { ctxt->instate = XML_PARSER_ATTRIBUTE_VALUE; NEXT; ret = xmlDecodeEntities(ctxt, -1, XML_SUBSTITUTE_REF, '\'', '<', 0); if (CUR == '<') { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Unescaped '<' not allowed in attributes values\n"); ctxt->wellFormed = 0; } if (CUR != '\'') { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "AttValue: ' expected\n"); ctxt->wellFormed = 0; } else NEXT; } else { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "AttValue: \" or ' expected\n"); ctxt->wellFormed = 0; } return(ret); } /** * xmlParseSystemLiteral: * @ctxt: an XML parser context * * parse an XML Literal * * [11] SystemLiteral ::= ('"' [^"]* '"') | ("'" [^']* "'") * * Returns the SystemLiteral parsed or NULL */ CHAR * xmlParseSystemLiteral(xmlParserCtxtPtr ctxt) { const CHAR *q; CHAR *ret = NULL; SHRINK; if (CUR == '"') { NEXT; q = CUR_PTR; while ((IS_CHAR(CUR)) && (CUR != '"')) NEXT; if (!IS_CHAR(CUR)) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Unfinished SystemLiteral\n"); ctxt->wellFormed = 0; } else { ret = xmlStrndup(q, CUR_PTR - q); NEXT; } } else if (CUR == '\'') { NEXT; q = CUR_PTR; while ((IS_CHAR(CUR)) && (CUR != '\'')) NEXT; if (!IS_CHAR(CUR)) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Unfinished SystemLiteral\n"); ctxt->wellFormed = 0; } else { ret = xmlStrndup(q, CUR_PTR - q); NEXT; } } else { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "SystemLiteral \" or ' expected\n"); ctxt->wellFormed = 0; } return(ret); } /** * xmlParsePubidLiteral: * @ctxt: an XML parser context * * parse an XML public literal * * [12] PubidLiteral ::= '"' PubidChar* '"' | "'" (PubidChar - "'")* "'" * * Returns the PubidLiteral parsed or NULL. */ CHAR * xmlParsePubidLiteral(xmlParserCtxtPtr ctxt) { const CHAR *q; CHAR *ret = NULL; /* * Name ::= (Letter | '_') (NameChar)* */ SHRINK; if (CUR == '"') { NEXT; q = CUR_PTR; while (IS_PUBIDCHAR(CUR)) NEXT; if (CUR != '"') { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Unfinished PubidLiteral\n"); ctxt->wellFormed = 0; } else { ret = xmlStrndup(q, CUR_PTR - q); NEXT; } } else if (CUR == '\'') { NEXT; q = CUR_PTR; while ((IS_LETTER(CUR)) && (CUR != '\'')) NEXT; if (!IS_LETTER(CUR)) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Unfinished PubidLiteral\n"); ctxt->wellFormed = 0; } else { ret = xmlStrndup(q, CUR_PTR - q); NEXT; } } else { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "SystemLiteral \" or ' expected\n"); ctxt->wellFormed = 0; } return(ret); } /** * xmlParseCharData: * @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 ::= [^<&]* - ([^<&]* ']]>' [^<&]*) */ void xmlParseCharData(xmlParserCtxtPtr ctxt, int cdata) { CHAR buf[1000]; int nbchar = 0; CHAR cur; SHRINK; /* * !!!!!!!!!!!! * NOTE: NXT(0) is used here to avoid breaking on < or & * entities substitutions. */ cur = CUR; while ((IS_CHAR(cur)) && (cur != '<') && (cur != '&')) { if ((cur == ']') && (NXT(1) == ']') && (NXT(2) == '>')) { if (cdata) break; else { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Sequence ']]>' not allowed in content\n"); ctxt->wellFormed = 0; } } buf[nbchar++] = CUR; if (nbchar == 1000) { /* * Ok the segment is to be consumed as chars. */ if (ctxt->sax != NULL) { if (areBlanks(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 (areBlanks(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); } } } } /** * xmlParseExternalID: * @ctxt: an XML parser context * @publicID: a CHAR** 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. */ CHAR * xmlParseExternalID(xmlParserCtxtPtr ctxt, CHAR **publicID, int strict) { CHAR *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->wellFormed = 0; } SKIP_BLANKS; URI = xmlParseSystemLiteral(ctxt); if (URI == NULL) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "xmlParseExternalID: SYSTEM, no URI\n"); 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->wellFormed = 0; } SKIP_BLANKS; *publicID = xmlParsePubidLiteral(ctxt); if (*publicID == NULL) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "xmlParseExternalID: PUBLIC, no Public Identifier\n"); 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->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 CHAR *ptr = CUR_PTR; if (!IS_BLANK(*ptr)) return(NULL); while (IS_BLANK(*ptr)) ptr++; if ((*ptr != '\'') || (*ptr != '"')) return(NULL); } SKIP_BLANKS; URI = xmlParseSystemLiteral(ctxt); if (URI == NULL) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "xmlParseExternalID: PUBLIC, no URI\n"); ctxt->wellFormed = 0; } } return(URI); } /** * xmlParseComment: * @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 ::= '' */ void xmlParseComment(xmlParserCtxtPtr ctxt) { const CHAR *q, *start; const CHAR *r; CHAR *val; /* * Check that there is a comment right here. */ if ((CUR != '<') || (NXT(1) != '!') || (NXT(2) != '-') || (NXT(3) != '-')) return; ctxt->instate = XML_PARSER_COMMENT; SHRINK; SKIP(4); start = q = CUR_PTR; NEXT; r = CUR_PTR; NEXT; while (IS_CHAR(CUR) && ((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->wellFormed = 0; } NEXT;r++;q++; } if (!IS_CHAR(CUR)) { if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL)) ctxt->sax->error(ctxt->userData, "Comment not terminated \n