/* * parser.c : an XML 1.0 parser, namespaces and validity support are mostly * implemented on top of the SAX interfaces * * References: * The XML specification: * http://www.w3.org/TR/REC-xml * Original 1.0 version: * http://www.w3.org/TR/1998/REC-xml-19980210 * XML second edition working draft * http://www.w3.org/TR/2000/WD-xml-2e-20000814 * * Okay this is a big file, the parser core is around 7000 lines, then it * is followed by the progressive parser top routines, then the various * high level APIs to call the parser and a few miscellaneous functions. * A number of helper functions and deprecated ones have been moved to * parserInternals.c to reduce this file size. * As much as possible the functions are associated with their relative * production in the XML specification. A few productions defining the * different ranges of character are actually implanted either in * parserInternals.h or parserInternals.c * The DOM tree build is realized from the default SAX callbacks in * the module SAX.c. * The routines doing the validation checks are in valid.c and called either * from the SAX callbacks or as standalone functions using a preparsed * document. * * See Copyright for the status of this software. * * daniel@veillard.com */ #define IN_LIBXML #include "libxml.h" #if defined(WIN32) && !defined (__CYGWIN__) #define XML_DIR_SEP '\\' #else #define XML_DIR_SEP '/' #endif #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #ifdef LIBXML_CATALOG_ENABLED #include #endif #ifdef LIBXML_SCHEMAS_ENABLED #include #include #endif #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 /** * xmlParserMaxDepth: * * arbitrary depth limit for the XML documents that we allow to * process. This is not a limitation of the parser but a safety * boundary feature. */ unsigned int xmlParserMaxDepth = 1024; #define SAX2 1 #define XML_PARSER_BIG_BUFFER_SIZE 300 #define XML_PARSER_BUFFER_SIZE 100 #define SAX_COMPAT_MODE BAD_CAST "SAX compatibility mode document" /* * List of XML prefixed PI allowed by W3C specs */ static const char *xmlW3CPIs[] = { "xml-stylesheet", NULL }; /* DEPR void xmlParserHandleReference(xmlParserCtxtPtr ctxt); */ xmlEntityPtr xmlParseStringPEReference(xmlParserCtxtPtr ctxt, const xmlChar **str); static xmlParserErrors xmlParseExternalEntityPrivate(xmlDocPtr doc, xmlParserCtxtPtr oldctxt, xmlSAXHandlerPtr sax, void *user_data, int depth, const xmlChar *URL, const xmlChar *ID, xmlNodePtr *list); #ifdef LIBXML_LEGACY_ENABLED static void xmlAddEntityReference(xmlEntityPtr ent, xmlNodePtr firstNode, xmlNodePtr lastNode); #endif /* LIBXML_LEGACY_ENABLED */ static xmlParserErrors xmlParseBalancedChunkMemoryInternal(xmlParserCtxtPtr oldctxt, const xmlChar *string, void *user_data, xmlNodePtr *lst); static int xmlLoadEntityContent(xmlParserCtxtPtr ctxt, xmlEntityPtr entity); /************************************************************************ * * * Some factorized error routines * * * ************************************************************************/ /** * xmlErrAttributeDup: * @ctxt: an XML parser context * @prefix: the attribute prefix * @localname: the attribute localname * * Handle a redefinition of attribute error */ static void xmlErrAttributeDup(xmlParserCtxtPtr ctxt, const xmlChar * prefix, const xmlChar * localname) { if ((ctxt != NULL) && (ctxt->disableSAX != 0) && (ctxt->instate == XML_PARSER_EOF)) return; if (ctxt != NULL) ctxt->errNo = XML_ERR_ATTRIBUTE_REDEFINED; if (prefix == NULL) __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER, ctxt->errNo, XML_ERR_FATAL, NULL, 0, (const char *) localname, NULL, NULL, 0, 0, "Attribute %s redefined\n", localname); else __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER, ctxt->errNo, XML_ERR_FATAL, NULL, 0, (const char *) prefix, (const char *) localname, NULL, 0, 0, "Attribute %s:%s redefined\n", prefix, localname); if (ctxt != NULL) { ctxt->wellFormed = 0; if (ctxt->recovery == 0) ctxt->disableSAX = 1; } } /** * xmlFatalErr: * @ctxt: an XML parser context * @error: the error number * @extra: extra information string * * Handle a fatal parser error, i.e. violating Well-Formedness constraints */ static void xmlFatalErr(xmlParserCtxtPtr ctxt, xmlParserErrors error, const char *info) { const char *errmsg; if ((ctxt != NULL) && (ctxt->disableSAX != 0) && (ctxt->instate == XML_PARSER_EOF)) return; switch (error) { case XML_ERR_INVALID_HEX_CHARREF: errmsg = "CharRef: invalid hexadecimal value\n"; break; case XML_ERR_INVALID_DEC_CHARREF: errmsg = "CharRef: invalid decimal value\n"; break; case XML_ERR_INVALID_CHARREF: errmsg = "CharRef: invalid value\n"; break; case XML_ERR_INTERNAL_ERROR: errmsg = "internal error"; break; case XML_ERR_PEREF_AT_EOF: errmsg = "PEReference at end of document\n"; break; case XML_ERR_PEREF_IN_PROLOG: errmsg = "PEReference in prolog\n"; break; case XML_ERR_PEREF_IN_EPILOG: errmsg = "PEReference in epilog\n"; break; case XML_ERR_PEREF_NO_NAME: errmsg = "PEReference: no name\n"; break; case XML_ERR_PEREF_SEMICOL_MISSING: errmsg = "PEReference: expecting ';'\n"; break; case XML_ERR_ENTITY_LOOP: errmsg = "Detected an entity reference loop\n"; break; case XML_ERR_ENTITY_NOT_STARTED: errmsg = "EntityValue: \" or ' expected\n"; break; case XML_ERR_ENTITY_PE_INTERNAL: errmsg = "PEReferences forbidden in internal subset\n"; break; case XML_ERR_ENTITY_NOT_FINISHED: errmsg = "EntityValue: \" or ' expected\n"; break; case XML_ERR_ATTRIBUTE_NOT_STARTED: errmsg = "AttValue: \" or ' expected\n"; break; case XML_ERR_LT_IN_ATTRIBUTE: errmsg = "Unescaped '<' not allowed in attributes values\n"; break; case XML_ERR_LITERAL_NOT_STARTED: errmsg = "SystemLiteral \" or ' expected\n"; break; case XML_ERR_LITERAL_NOT_FINISHED: errmsg = "Unfinished System or Public ID \" or ' expected\n"; break; case XML_ERR_MISPLACED_CDATA_END: errmsg = "Sequence ']]>' not allowed in content\n"; break; case XML_ERR_URI_REQUIRED: errmsg = "SYSTEM or PUBLIC, the URI is missing\n"; break; case XML_ERR_PUBID_REQUIRED: errmsg = "PUBLIC, the Public Identifier is missing\n"; break; case XML_ERR_HYPHEN_IN_COMMENT: errmsg = "Comment must not contain '--' (double-hyphen)\n"; break; case XML_ERR_PI_NOT_STARTED: errmsg = "xmlParsePI : no target name\n"; break; case XML_ERR_RESERVED_XML_NAME: errmsg = "Invalid PI name\n"; break; case XML_ERR_NOTATION_NOT_STARTED: errmsg = "NOTATION: Name expected here\n"; break; case XML_ERR_NOTATION_NOT_FINISHED: errmsg = "'>' required to close NOTATION declaration\n"; break; case XML_ERR_VALUE_REQUIRED: errmsg = "Entity value required\n"; break; case XML_ERR_URI_FRAGMENT: errmsg = "Fragment not allowed"; break; case XML_ERR_ATTLIST_NOT_STARTED: errmsg = "'(' required to start ATTLIST enumeration\n"; break; case XML_ERR_NMTOKEN_REQUIRED: errmsg = "NmToken expected in ATTLIST enumeration\n"; break; case XML_ERR_ATTLIST_NOT_FINISHED: errmsg = "')' required to finish ATTLIST enumeration\n"; break; case XML_ERR_MIXED_NOT_STARTED: errmsg = "MixedContentDecl : '|' or ')*' expected\n"; break; case XML_ERR_PCDATA_REQUIRED: errmsg = "MixedContentDecl : '#PCDATA' expected\n"; break; case XML_ERR_ELEMCONTENT_NOT_STARTED: errmsg = "ContentDecl : Name or '(' expected\n"; break; case XML_ERR_ELEMCONTENT_NOT_FINISHED: errmsg = "ContentDecl : ',' '|' or ')' expected\n"; break; case XML_ERR_PEREF_IN_INT_SUBSET: errmsg = "PEReference: forbidden within markup decl in internal subset\n"; break; case XML_ERR_GT_REQUIRED: errmsg = "expected '>'\n"; break; case XML_ERR_CONDSEC_INVALID: errmsg = "XML conditional section '[' expected\n"; break; case XML_ERR_EXT_SUBSET_NOT_FINISHED: errmsg = "Content error in the external subset\n"; break; case XML_ERR_CONDSEC_INVALID_KEYWORD: errmsg = "conditional section INCLUDE or IGNORE keyword expected\n"; break; case XML_ERR_CONDSEC_NOT_FINISHED: errmsg = "XML conditional section not closed\n"; break; case XML_ERR_XMLDECL_NOT_STARTED: errmsg = "Text declaration '' expected\n"; break; case XML_ERR_EXT_ENTITY_STANDALONE: errmsg = "external parsed entities cannot be standalone\n"; break; case XML_ERR_ENTITYREF_SEMICOL_MISSING: errmsg = "EntityRef: expecting ';'\n"; break; case XML_ERR_DOCTYPE_NOT_FINISHED: errmsg = "DOCTYPE improperly terminated\n"; break; case XML_ERR_LTSLASH_REQUIRED: errmsg = "EndTag: 'errNo = error; __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_FATAL, NULL, 0, info, NULL, NULL, 0, 0, errmsg, info); if (ctxt != NULL) { ctxt->wellFormed = 0; if (ctxt->recovery == 0) ctxt->disableSAX = 1; } } /** * xmlFatalErrMsg: * @ctxt: an XML parser context * @error: the error number * @msg: the error message * * Handle a fatal parser error, i.e. violating Well-Formedness constraints */ static void xmlFatalErrMsg(xmlParserCtxtPtr ctxt, xmlParserErrors error, const char *msg) { if ((ctxt != NULL) && (ctxt->disableSAX != 0) && (ctxt->instate == XML_PARSER_EOF)) return; if (ctxt != NULL) ctxt->errNo = error; __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_FATAL, NULL, 0, NULL, NULL, NULL, 0, 0, msg); if (ctxt != NULL) { ctxt->wellFormed = 0; if (ctxt->recovery == 0) ctxt->disableSAX = 1; } } /** * xmlWarningMsg: * @ctxt: an XML parser context * @error: the error number * @msg: the error message * @str1: extra data * @str2: extra data * * Handle a warning. */ static void xmlWarningMsg(xmlParserCtxtPtr ctxt, xmlParserErrors error, const char *msg, const xmlChar *str1, const xmlChar *str2) { xmlStructuredErrorFunc schannel = NULL; if ((ctxt != NULL) && (ctxt->disableSAX != 0) && (ctxt->instate == XML_PARSER_EOF)) return; if ((ctxt != NULL) && (ctxt->sax != NULL) && (ctxt->sax->initialized == XML_SAX2_MAGIC)) schannel = ctxt->sax->serror; __xmlRaiseError(schannel, (ctxt->sax) ? ctxt->sax->warning : NULL, ctxt->userData, ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_WARNING, NULL, 0, (const char *) str1, (const char *) str2, NULL, 0, 0, msg, (const char *) str1, (const char *) str2); } /** * xmlValidityError: * @ctxt: an XML parser context * @error: the error number * @msg: the error message * @str1: extra data * * Handle a validity error. */ static void xmlValidityError(xmlParserCtxtPtr ctxt, xmlParserErrors error, const char *msg, const xmlChar *str1) { xmlStructuredErrorFunc schannel = NULL; if ((ctxt != NULL) && (ctxt->disableSAX != 0) && (ctxt->instate == XML_PARSER_EOF)) return; if (ctxt != NULL) { ctxt->errNo = error; if ((ctxt->sax != NULL) && (ctxt->sax->initialized == XML_SAX2_MAGIC)) schannel = ctxt->sax->serror; } __xmlRaiseError(schannel, ctxt->vctxt.error, ctxt->vctxt.userData, ctxt, NULL, XML_FROM_DTD, error, XML_ERR_ERROR, NULL, 0, (const char *) str1, NULL, NULL, 0, 0, msg, (const char *) str1); if (ctxt != NULL) { ctxt->valid = 0; } } /** * xmlFatalErrMsgInt: * @ctxt: an XML parser context * @error: the error number * @msg: the error message * @val: an integer value * * Handle a fatal parser error, i.e. violating Well-Formedness constraints */ static void xmlFatalErrMsgInt(xmlParserCtxtPtr ctxt, xmlParserErrors error, const char *msg, int val) { if ((ctxt != NULL) && (ctxt->disableSAX != 0) && (ctxt->instate == XML_PARSER_EOF)) return; if (ctxt != NULL) ctxt->errNo = error; __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_FATAL, NULL, 0, NULL, NULL, NULL, val, 0, msg, val); if (ctxt != NULL) { ctxt->wellFormed = 0; if (ctxt->recovery == 0) ctxt->disableSAX = 1; } } /** * xmlFatalErrMsgStrIntStr: * @ctxt: an XML parser context * @error: the error number * @msg: the error message * @str1: an string info * @val: an integer value * @str2: an string info * * Handle a fatal parser error, i.e. violating Well-Formedness constraints */ static void xmlFatalErrMsgStrIntStr(xmlParserCtxtPtr ctxt, xmlParserErrors error, const char *msg, const xmlChar *str1, int val, const xmlChar *str2) { if ((ctxt != NULL) && (ctxt->disableSAX != 0) && (ctxt->instate == XML_PARSER_EOF)) return; if (ctxt != NULL) ctxt->errNo = error; __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_FATAL, NULL, 0, (const char *) str1, (const char *) str2, NULL, val, 0, msg, str1, val, str2); if (ctxt != NULL) { ctxt->wellFormed = 0; if (ctxt->recovery == 0) ctxt->disableSAX = 1; } } /** * xmlFatalErrMsgStr: * @ctxt: an XML parser context * @error: the error number * @msg: the error message * @val: a string value * * Handle a fatal parser error, i.e. violating Well-Formedness constraints */ static void xmlFatalErrMsgStr(xmlParserCtxtPtr ctxt, xmlParserErrors error, const char *msg, const xmlChar * val) { if ((ctxt != NULL) && (ctxt->disableSAX != 0) && (ctxt->instate == XML_PARSER_EOF)) return; if (ctxt != NULL) ctxt->errNo = error; __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_FATAL, NULL, 0, (const char *) val, NULL, NULL, 0, 0, msg, val); if (ctxt != NULL) { ctxt->wellFormed = 0; if (ctxt->recovery == 0) ctxt->disableSAX = 1; } } /** * xmlErrMsgStr: * @ctxt: an XML parser context * @error: the error number * @msg: the error message * @val: a string value * * Handle a non fatal parser error */ static void xmlErrMsgStr(xmlParserCtxtPtr ctxt, xmlParserErrors error, const char *msg, const xmlChar * val) { if ((ctxt != NULL) && (ctxt->disableSAX != 0) && (ctxt->instate == XML_PARSER_EOF)) return; if (ctxt != NULL) ctxt->errNo = error; __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_PARSER, error, XML_ERR_ERROR, NULL, 0, (const char *) val, NULL, NULL, 0, 0, msg, val); } /** * xmlNsErr: * @ctxt: an XML parser context * @error: the error number * @msg: the message * @info1: extra information string * @info2: extra information string * * Handle a fatal parser error, i.e. violating Well-Formedness constraints */ static void xmlNsErr(xmlParserCtxtPtr ctxt, xmlParserErrors error, const char *msg, const xmlChar * info1, const xmlChar * info2, const xmlChar * info3) { if ((ctxt != NULL) && (ctxt->disableSAX != 0) && (ctxt->instate == XML_PARSER_EOF)) return; if (ctxt != NULL) ctxt->errNo = error; __xmlRaiseError(NULL, NULL, NULL, ctxt, NULL, XML_FROM_NAMESPACE, error, XML_ERR_ERROR, NULL, 0, (const char *) info1, (const char *) info2, (const char *) info3, 0, 0, msg, info1, info2, info3); if (ctxt != NULL) ctxt->nsWellFormed = 0; } /************************************************************************ * * * Library wide options * * * ************************************************************************/ /** * xmlHasFeature: * @feature: the feature to be examined * * Examines if the library has been compiled with a given feature. * * Returns a non-zero value if the feature exist, otherwise zero. * Returns zero (0) if the feature does not exist or an unknown * unknown feature is requested, non-zero otherwise. */ int xmlHasFeature(xmlFeature feature) { switch (feature) { case XML_WITH_THREAD: #ifdef LIBXML_THREAD_ENABLED return(1); #else return(0); #endif case XML_WITH_TREE: #ifdef LIBXML_TREE_ENABLED return(1); #else return(0); #endif case XML_WITH_OUTPUT: #ifdef LIBXML_OUTPUT_ENABLED return(1); #else return(0); #endif case XML_WITH_PUSH: #ifdef LIBXML_PUSH_ENABLED return(1); #else return(0); #endif case XML_WITH_READER: #ifdef LIBXML_READER_ENABLED return(1); #else return(0); #endif case XML_WITH_PATTERN: #ifdef LIBXML_PATTERN_ENABLED return(1); #else return(0); #endif case XML_WITH_WRITER: #ifdef LIBXML_WRITER_ENABLED return(1); #else return(0); #endif case XML_WITH_SAX1: #ifdef LIBXML_SAX1_ENABLED return(1); #else return(0); #endif case XML_WITH_FTP: #ifdef LIBXML_FTP_ENABLED return(1); #else return(0); #endif case XML_WITH_HTTP: #ifdef LIBXML_HTTP_ENABLED return(1); #else return(0); #endif case XML_WITH_VALID: #ifdef LIBXML_VALID_ENABLED return(1); #else return(0); #endif case XML_WITH_HTML: #ifdef LIBXML_HTML_ENABLED return(1); #else return(0); #endif case XML_WITH_LEGACY: #ifdef LIBXML_LEGACY_ENABLED return(1); #else return(0); #endif case XML_WITH_C14N: #ifdef LIBXML_C14N_ENABLED return(1); #else return(0); #endif case XML_WITH_CATALOG: #ifdef LIBXML_CATALOG_ENABLED return(1); #else return(0); #endif case XML_WITH_XPATH: #ifdef LIBXML_XPATH_ENABLED return(1); #else return(0); #endif case XML_WITH_XPTR: #ifdef LIBXML_XPTR_ENABLED return(1); #else return(0); #endif case XML_WITH_XINCLUDE: #ifdef LIBXML_XINCLUDE_ENABLED return(1); #else return(0); #endif case XML_WITH_ICONV: #ifdef LIBXML_ICONV_ENABLED return(1); #else return(0); #endif case XML_WITH_ISO8859X: #ifdef LIBXML_ISO8859X_ENABLED return(1); #else return(0); #endif case XML_WITH_UNICODE: #ifdef LIBXML_UNICODE_ENABLED return(1); #else return(0); #endif case XML_WITH_REGEXP: #ifdef LIBXML_REGEXP_ENABLED return(1); #else return(0); #endif case XML_WITH_AUTOMATA: #ifdef LIBXML_AUTOMATA_ENABLED return(1); #else return(0); #endif case XML_WITH_EXPR: #ifdef LIBXML_EXPR_ENABLED return(1); #else return(0); #endif case XML_WITH_SCHEMAS: #ifdef LIBXML_SCHEMAS_ENABLED return(1); #else return(0); #endif case XML_WITH_SCHEMATRON: #ifdef LIBXML_SCHEMATRON_ENABLED return(1); #else return(0); #endif case XML_WITH_MODULES: #ifdef LIBXML_MODULES_ENABLED return(1); #else return(0); #endif case XML_WITH_DEBUG: #ifdef LIBXML_DEBUG_ENABLED return(1); #else return(0); #endif case XML_WITH_DEBUG_MEM: #ifdef DEBUG_MEMORY_LOCATION return(1); #else return(0); #endif case XML_WITH_DEBUG_RUN: #ifdef LIBXML_DEBUG_RUNTIME return(1); #else return(0); #endif case XML_WITH_ZLIB: #ifdef LIBXML_ZLIB_ENABLED return(1); #else return(0); #endif default: break; } return(0); } /************************************************************************ * * * SAX2 defaulted attributes handling * * * ************************************************************************/ /** * xmlDetectSAX2: * @ctxt: an XML parser context * * Do the SAX2 detection and specific intialization */ static void xmlDetectSAX2(xmlParserCtxtPtr ctxt) { if (ctxt == NULL) return; #ifdef LIBXML_SAX1_ENABLED if ((ctxt->sax) && (ctxt->sax->initialized == XML_SAX2_MAGIC) && ((ctxt->sax->startElementNs != NULL) || (ctxt->sax->endElementNs != NULL))) ctxt->sax2 = 1; #else ctxt->sax2 = 1; #endif /* LIBXML_SAX1_ENABLED */ ctxt->str_xml = xmlDictLookup(ctxt->dict, BAD_CAST "xml", 3); ctxt->str_xmlns = xmlDictLookup(ctxt->dict, BAD_CAST "xmlns", 5); ctxt->str_xml_ns = xmlDictLookup(ctxt->dict, XML_XML_NAMESPACE, 36); if ((ctxt->str_xml==NULL) || (ctxt->str_xmlns==NULL) || (ctxt->str_xml_ns == NULL)) { xmlErrMemory(ctxt, NULL); } } typedef struct _xmlDefAttrs xmlDefAttrs; typedef xmlDefAttrs *xmlDefAttrsPtr; struct _xmlDefAttrs { int nbAttrs; /* number of defaulted attributes on that element */ int maxAttrs; /* the size of the array */ const xmlChar *values[4]; /* array of localname/prefix/values */ }; /** * xmlAttrNormalizeSpace: * @src: the source string * @dst: the target string * * Normalize the space in non CDATA attribute values: * If the attribute type 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. * Note that the size of dst need to be at least src, and if one doesn't need * to preserve dst (and it doesn't come from a dictionary or read-only) then * passing src as dst is just fine. * * Returns a pointer to the normalized value (dst) or NULL if no conversion * is needed. */ static xmlChar * xmlAttrNormalizeSpace(const xmlChar *src, xmlChar *dst) { if ((src == NULL) || (dst == NULL)) return(NULL); while (*src == 0x20) src++; while (*src != 0) { if (*src == 0x20) { while (*src == 0x20) src++; if (*src != 0) *dst++ = 0x20; } else { *dst++ = *src++; } } *dst = 0; if (dst == src) return(NULL); return(dst); } /** * xmlAttrNormalizeSpace2: * @src: the source string * * Normalize the space in non CDATA attribute values, a slightly more complex * front end to avoid allocation problems when running on attribute values * coming from the input. * * Returns a pointer to the normalized value (dst) or NULL if no conversion * is needed. */ static const xmlChar * xmlAttrNormalizeSpace2(xmlParserCtxtPtr ctxt, const xmlChar *src, int *len) { int i; int remove_head = 0; int need_realloc = 0; const xmlChar *cur; if ((ctxt == NULL) || (src == NULL) || (len == NULL)) return(NULL); i = *len; if (i <= 0) return(NULL); cur = src; while (*cur == 0x20) { cur++; remove_head++; } while (*cur != 0) { if (*cur == 0x20) { cur++; if ((*cur == 0x20) || (*cur == 0)) { need_realloc = 1; break; } } else cur++; } if (need_realloc) { xmlChar *ret; ret = xmlStrndup(src + remove_head, i - remove_head + 1); if (ret == NULL) { xmlErrMemory(ctxt, NULL); return(NULL); } xmlAttrNormalizeSpace(ret, ret); *len = (int) strlen((const char *)ret); return(ret); } else if (remove_head) { *len -= remove_head; return(src + remove_head); } return(NULL); } /** * xmlAddDefAttrs: * @ctxt: an XML parser context * @fullname: the element fullname * @fullattr: the attribute fullname * @value: the attribute value * * Add a defaulted attribute for an element */ static void xmlAddDefAttrs(xmlParserCtxtPtr ctxt, const xmlChar *fullname, const xmlChar *fullattr, const xmlChar *value) { xmlDefAttrsPtr defaults; int len; const xmlChar *name; const xmlChar *prefix; /* * Allows to detect attribute redefinitions */ if (ctxt->attsSpecial != NULL) { if (xmlHashLookup2(ctxt->attsSpecial, fullname, fullattr) != NULL) return; } if (ctxt->attsDefault == NULL) { ctxt->attsDefault = xmlHashCreateDict(10, ctxt->dict); if (ctxt->attsDefault == NULL) goto mem_error; } /* * split the element name into prefix:localname , the string found * are within the DTD and then not associated to namespace names. */ name = xmlSplitQName3(fullname, &len); if (name == NULL) { name = xmlDictLookup(ctxt->dict, fullname, -1); prefix = NULL; } else { name = xmlDictLookup(ctxt->dict, name, -1); prefix = xmlDictLookup(ctxt->dict, fullname, len); } /* * make sure there is some storage */ defaults = xmlHashLookup2(ctxt->attsDefault, name, prefix); if (defaults == NULL) { defaults = (xmlDefAttrsPtr) xmlMalloc(sizeof(xmlDefAttrs) + (4 * 4) * sizeof(const xmlChar *)); if (defaults == NULL) goto mem_error; defaults->nbAttrs = 0; defaults->maxAttrs = 4; if (xmlHashUpdateEntry2(ctxt->attsDefault, name, prefix, defaults, NULL) < 0) { xmlFree(defaults); goto mem_error; } } else if (defaults->nbAttrs >= defaults->maxAttrs) { xmlDefAttrsPtr temp; temp = (xmlDefAttrsPtr) xmlRealloc(defaults, sizeof(xmlDefAttrs) + (2 * defaults->maxAttrs * 4) * sizeof(const xmlChar *)); if (temp == NULL) goto mem_error; defaults = temp; defaults->maxAttrs *= 2; if (xmlHashUpdateEntry2(ctxt->attsDefault, name, prefix, defaults, NULL) < 0) { xmlFree(defaults); goto mem_error; } } /* * Split the element name into prefix:localname , the string found * are within the DTD and hen not associated to namespace names. */ name = xmlSplitQName3(fullattr, &len); if (name == NULL) { name = xmlDictLookup(ctxt->dict, fullattr, -1); prefix = NULL; } else { name = xmlDictLookup(ctxt->dict, name, -1); prefix = xmlDictLookup(ctxt->dict, fullattr, len); } defaults->values[4 * defaults->nbAttrs] = name; defaults->values[4 * defaults->nbAttrs + 1] = prefix; /* intern the string and precompute the end */ len = xmlStrlen(value); value = xmlDictLookup(ctxt->dict, value, len); defaults->values[4 * defaults->nbAttrs + 2] = value; defaults->values[4 * defaults->nbAttrs + 3] = value + len; defaults->nbAttrs++; return; mem_error: xmlErrMemory(ctxt, NULL); return; } /** * xmlAddSpecialAttr: * @ctxt: an XML parser context * @fullname: the element fullname * @fullattr: the attribute fullname * @type: the attribute type * * Register this attribute type */ static void xmlAddSpecialAttr(xmlParserCtxtPtr ctxt, const xmlChar *fullname, const xmlChar *fullattr, int type) { if (ctxt->attsSpecial == NULL) { ctxt->attsSpecial = xmlHashCreateDict(10, ctxt->dict); if (ctxt->attsSpecial == NULL) goto mem_error; } if (xmlHashLookup2(ctxt->attsSpecial, fullname, fullattr) != NULL) return; xmlHashAddEntry2(ctxt->attsSpecial, fullname, fullattr, (void *) (long) type); return; mem_error: xmlErrMemory(ctxt, NULL); return; } /** * xmlCleanSpecialAttrCallback: * * Removes CDATA attributes from the special attribute table */ static void xmlCleanSpecialAttrCallback(void *payload, void *data, const xmlChar *fullname, const xmlChar *fullattr, const xmlChar *unused ATTRIBUTE_UNUSED) { xmlParserCtxtPtr ctxt = (xmlParserCtxtPtr) data; if (((long) payload) == XML_ATTRIBUTE_CDATA) { xmlHashRemoveEntry2(ctxt->attsSpecial, fullname, fullattr, NULL); } } /** * xmlCleanSpecialAttr: * @ctxt: an XML parser context * * Trim the list of attributes defined to remove all those of type * CDATA as they are not special. This call should be done when finishing * to parse the DTD and before starting to parse the document root. */ static void xmlCleanSpecialAttr(xmlParserCtxtPtr ctxt) { if (ctxt->attsSpecial == NULL) return; xmlHashScanFull(ctxt->attsSpecial, xmlCleanSpecialAttrCallback, ctxt); if (xmlHashSize(ctxt->attsSpecial) == 0) { xmlHashFree(ctxt->attsSpecial, NULL); ctxt->attsSpecial = NULL; } return; } /** * xmlCheckLanguageID: * @lang: pointer to the string value * * Checks that the value conforms to the LanguageID production: * * NOTE: this is somewhat deprecated, those productions were removed from * the XML Second edition. * * [33] LanguageID ::= Langcode ('-' Subcode)* * [34] Langcode ::= ISO639Code | IanaCode | UserCode * [35] ISO639Code ::= ([a-z] | [A-Z]) ([a-z] | [A-Z]) * [36] IanaCode ::= ('i' | 'I') '-' ([a-z] | [A-Z])+ * [37] UserCode ::= ('x' | 'X') '-' ([a-z] | [A-Z])+ * [38] Subcode ::= ([a-z] | [A-Z])+ * * Returns 1 if correct 0 otherwise **/ int xmlCheckLanguageID(const xmlChar * lang) { const xmlChar *cur = lang; if (cur == NULL) return (0); if (((cur[0] == 'i') && (cur[1] == '-')) || ((cur[0] == 'I') && (cur[1] == '-'))) { /* * IANA code */ cur += 2; while (((cur[0] >= 'A') && (cur[0] <= 'Z')) || /* non input consuming */ ((cur[0] >= 'a') && (cur[0] <= 'z'))) cur++; } else if (((cur[0] == 'x') && (cur[1] == '-')) || ((cur[0] == 'X') && (cur[1] == '-'))) { /* * User code */ cur += 2; while (((cur[0] >= 'A') && (cur[0] <= 'Z')) || /* non input consuming */ ((cur[0] >= 'a') && (cur[0] <= 'z'))) cur++; } else if (((cur[0] >= 'A') && (cur[0] <= 'Z')) || ((cur[0] >= 'a') && (cur[0] <= 'z'))) { /* * ISO639 */ cur++; if (((cur[0] >= 'A') && (cur[0] <= 'Z')) || ((cur[0] >= 'a') && (cur[0] <= 'z'))) cur++; else return (0); } else return (0); while (cur[0] != 0) { /* non input consuming */ if (cur[0] != '-') return (0); cur++; if (((cur[0] >= 'A') && (cur[0] <= 'Z')) || ((cur[0] >= 'a') && (cur[0] <= 'z'))) cur++; else return (0); while (((cur[0] >= 'A') && (cur[0] <= 'Z')) || /* non input consuming */ ((cur[0] >= 'a') && (cur[0] <= 'z'))) cur++; } return (1); } /************************************************************************ * * * Parser stacks related functions and macros * * * ************************************************************************/ xmlEntityPtr xmlParseStringEntityRef(xmlParserCtxtPtr ctxt, const xmlChar ** str); #ifdef SAX2 /** * nsPush: * @ctxt: an XML parser context * @prefix: the namespace prefix or NULL * @URL: the namespace name * * Pushes a new parser namespace on top of the ns stack * * Returns -1 in case of error, -2 if the namespace should be discarded * and the index in the stack otherwise. */ static int nsPush(xmlParserCtxtPtr ctxt, const xmlChar *prefix, const xmlChar *URL) { if (ctxt->options & XML_PARSE_NSCLEAN) { int i; for (i = 0;i < ctxt->nsNr;i += 2) { if (ctxt->nsTab[i] == prefix) { /* in scope */ if (ctxt->nsTab[i + 1] == URL) return(-2); /* out of scope keep it */ break; } } } if ((ctxt->nsMax == 0) || (ctxt->nsTab == NULL)) { ctxt->nsMax = 10; ctxt->nsNr = 0; ctxt->nsTab = (const xmlChar **) xmlMalloc(ctxt->nsMax * sizeof(xmlChar *)); if (ctxt->nsTab == NULL) { xmlErrMemory(ctxt, NULL); ctxt->nsMax = 0; return (-1); } } else if (ctxt->nsNr >= ctxt->nsMax) { const xmlChar ** tmp; ctxt->nsMax *= 2; tmp = (const xmlChar **) xmlRealloc((char *) ctxt->nsTab, ctxt->nsMax * sizeof(ctxt->nsTab[0])); if (tmp == NULL) { xmlErrMemory(ctxt, NULL); ctxt->nsMax /= 2; return (-1); } ctxt->nsTab = tmp; } ctxt->nsTab[ctxt->nsNr++] = prefix; ctxt->nsTab[ctxt->nsNr++] = URL; return (ctxt->nsNr); } /** * nsPop: * @ctxt: an XML parser context * @nr: the number to pop * * Pops the top @nr parser prefix/namespace from the ns stack * * Returns the number of namespaces removed */ static int nsPop(xmlParserCtxtPtr ctxt, int nr) { int i; if (ctxt->nsTab == NULL) return(0); if (ctxt->nsNr < nr) { xmlGenericError(xmlGenericErrorContext, "Pbm popping %d NS\n", nr); nr = ctxt->nsNr; } if (ctxt->nsNr <= 0) return (0); for (i = 0;i < nr;i++) { ctxt->nsNr--; ctxt->nsTab[ctxt->nsNr] = NULL; } return(nr); } #endif static int xmlCtxtGrowAttrs(xmlParserCtxtPtr ctxt, int nr) { const xmlChar **atts; int *attallocs; int maxatts; if (ctxt->atts == NULL) { maxatts = 55; /* allow for 10 attrs by default */ atts = (const xmlChar **) xmlMalloc(maxatts * sizeof(xmlChar *)); if (atts == NULL) goto mem_error; ctxt->atts = atts; attallocs = (int *) xmlMalloc((maxatts / 5) * sizeof(int)); if (attallocs == NULL) goto mem_error; ctxt->attallocs = attallocs; ctxt->maxatts = maxatts; } else if (nr + 5 > ctxt->maxatts) { maxatts = (nr + 5) * 2; atts = (const xmlChar **) xmlRealloc((void *) ctxt->atts, maxatts * sizeof(const xmlChar *)); if (atts == NULL) goto mem_error; ctxt->atts = atts; attallocs = (int *) xmlRealloc((void *) ctxt->attallocs, (maxatts / 5) * sizeof(int)); if (attallocs == NULL) goto mem_error; ctxt->attallocs = attallocs; ctxt->maxatts = maxatts; } return(ctxt->maxatts); mem_error: xmlErrMemory(ctxt, NULL); return(-1); } /** * inputPush: * @ctxt: an XML parser context * @value: the parser input * * Pushes a new parser input on top of the input stack * * Returns 0 in case of error, the index in the stack otherwise */ int inputPush(xmlParserCtxtPtr ctxt, xmlParserInputPtr value) { if ((ctxt == NULL) || (value == NULL)) return(0); if (ctxt->inputNr >= ctxt->inputMax) { ctxt->inputMax *= 2; ctxt->inputTab = (xmlParserInputPtr *) xmlRealloc(ctxt->inputTab, ctxt->inputMax * sizeof(ctxt->inputTab[0])); if (ctxt->inputTab == NULL) { xmlErrMemory(ctxt, NULL); return (0); } } ctxt->inputTab[ctxt->inputNr] = value; ctxt->input = value; return (ctxt->inputNr++); } /** * inputPop: * @ctxt: an XML parser context * * Pops the top parser input from the input stack * * Returns the input just removed */ xmlParserInputPtr inputPop(xmlParserCtxtPtr ctxt) { xmlParserInputPtr ret; if (ctxt == NULL) return(NULL); if (ctxt->inputNr <= 0) return (NULL); ctxt->inputNr--; if (ctxt->inputNr > 0) ctxt->input = ctxt->inputTab[ctxt->inputNr - 1]; else ctxt->input = NULL; ret = ctxt->inputTab[ctxt->inputNr]; ctxt->inputTab[ctxt->inputNr] = NULL; return (ret); } /** * nodePush: * @ctxt: an XML parser context * @value: the element node * * Pushes a new element node on top of the node stack * * Returns 0 in case of error, the index in the stack otherwise */ int nodePush(xmlParserCtxtPtr ctxt, xmlNodePtr value) { if (ctxt == NULL) return(0); if (ctxt->nodeNr >= ctxt->nodeMax) { xmlNodePtr *tmp; tmp = (xmlNodePtr *) xmlRealloc(ctxt->nodeTab, ctxt->nodeMax * 2 * sizeof(ctxt->nodeTab[0])); if (tmp == NULL) { xmlErrMemory(ctxt, NULL); return (0); } ctxt->nodeTab = tmp; ctxt->nodeMax *= 2; } if (((unsigned int) ctxt->nodeNr) > xmlParserMaxDepth) { xmlFatalErrMsgInt(ctxt, XML_ERR_INTERNAL_ERROR, "Excessive depth in document: change xmlParserMaxDepth = %d\n", xmlParserMaxDepth); ctxt->instate = XML_PARSER_EOF; return(0); } ctxt->nodeTab[ctxt->nodeNr] = value; ctxt->node = value; return (ctxt->nodeNr++); } /** * nodePop: * @ctxt: an XML parser context * * Pops the top element node from the node stack * * Returns the node just removed */ xmlNodePtr nodePop(xmlParserCtxtPtr ctxt) { xmlNodePtr ret; if (ctxt == NULL) return(NULL); if (ctxt->nodeNr <= 0) return (NULL); ctxt->nodeNr--; if (ctxt->nodeNr > 0) ctxt->node = ctxt->nodeTab[ctxt->nodeNr - 1]; else ctxt->node = NULL; ret = ctxt->nodeTab[ctxt->nodeNr]; ctxt->nodeTab[ctxt->nodeNr] = NULL; return (ret); } #ifdef LIBXML_PUSH_ENABLED /** * nameNsPush: * @ctxt: an XML parser context * @value: the element name * @prefix: the element prefix * @URI: the element namespace name * * Pushes a new element name/prefix/URL on top of the name stack * * Returns -1 in case of error, the index in the stack otherwise */ static int nameNsPush(xmlParserCtxtPtr ctxt, const xmlChar * value, const xmlChar *prefix, const xmlChar *URI, int nsNr) { if (ctxt->nameNr >= ctxt->nameMax) { const xmlChar * *tmp; void **tmp2; ctxt->nameMax *= 2; tmp = (const xmlChar * *) xmlRealloc((xmlChar * *)ctxt->nameTab, ctxt->nameMax * sizeof(ctxt->nameTab[0])); if (tmp == NULL) { ctxt->nameMax /= 2; goto mem_error; } ctxt->nameTab = tmp; tmp2 = (void **) xmlRealloc((void * *)ctxt->pushTab, ctxt->nameMax * 3 * sizeof(ctxt->pushTab[0])); if (tmp2 == NULL) { ctxt->nameMax /= 2; goto mem_error; } ctxt->pushTab = tmp2; } ctxt->nameTab[ctxt->nameNr] = value; ctxt->name = value; ctxt->pushTab[ctxt->nameNr * 3] = (void *) prefix; ctxt->pushTab[ctxt->nameNr * 3 + 1] = (void *) URI; ctxt->pushTab[ctxt->nameNr * 3 + 2] = (void *) (long) nsNr; return (ctxt->nameNr++); mem_error: xmlErrMemory(ctxt, NULL); return (-1); } /** * nameNsPop: * @ctxt: an XML parser context * * Pops the top element/prefix/URI name from the name stack * * Returns the name just removed */ static const xmlChar * nameNsPop(xmlParserCtxtPtr ctxt) { const xmlChar *ret; if (ctxt->nameNr <= 0) return (NULL); ctxt->nameNr--; if (ctxt->nameNr > 0) ctxt->name = ctxt->nameTab[ctxt->nameNr - 1]; else ctxt->name = NULL; ret = ctxt->nameTab[ctxt->nameNr]; ctxt->nameTab[ctxt->nameNr] = NULL; return (ret); } #endif /* LIBXML_PUSH_ENABLED */ /** * namePush: * @ctxt: an XML parser context * @value: the element name * * Pushes a new element name on top of the name stack * * Returns -1 in case of error, the index in the stack otherwise */ int namePush(xmlParserCtxtPtr ctxt, const xmlChar * value) { if (ctxt == NULL) return (-1); if (ctxt->nameNr >= ctxt->nameMax) { const xmlChar * *tmp; ctxt->nameMax *= 2; tmp = (const xmlChar * *) xmlRealloc((xmlChar * *)ctxt->nameTab, ctxt->nameMax * sizeof(ctxt->nameTab[0])); if (tmp == NULL) { ctxt->nameMax /= 2; goto mem_error; } ctxt->nameTab = tmp; } ctxt->nameTab[ctxt->nameNr] = value; ctxt->name = value; return (ctxt->nameNr++); mem_error: xmlErrMemory(ctxt, NULL); return (-1); } /** * namePop: * @ctxt: an XML parser context * * Pops the top element name from the name stack * * Returns the name just removed */ const xmlChar * namePop(xmlParserCtxtPtr ctxt) { const xmlChar *ret; if ((ctxt == NULL) || (ctxt->nameNr <= 0)) return (NULL); ctxt->nameNr--; if (ctxt->nameNr > 0) ctxt->name = ctxt->nameTab[ctxt->nameNr - 1]; else ctxt->name = NULL; ret = ctxt->nameTab[ctxt->nameNr]; ctxt->nameTab[ctxt->nameNr] = NULL; return (ret); } static int spacePush(xmlParserCtxtPtr ctxt, int val) { if (ctxt->spaceNr >= ctxt->spaceMax) { int *tmp; ctxt->spaceMax *= 2; tmp = (int *) xmlRealloc(ctxt->spaceTab, ctxt->spaceMax * sizeof(ctxt->spaceTab[0])); if (tmp == NULL) { xmlErrMemory(ctxt, NULL); return(0); } ctxt->spaceTab = tmp; } ctxt->spaceTab[ctxt->spaceNr] = val; ctxt->space = &ctxt->spaceTab[ctxt->spaceNr]; return(ctxt->spaceNr++); } static int spacePop(xmlParserCtxtPtr ctxt) { int ret; if (ctxt->spaceNr <= 0) return(0); ctxt->spaceNr--; if (ctxt->spaceNr > 0) ctxt->space = &ctxt->spaceTab[ctxt->spaceNr - 1]; else ctxt->space = &ctxt->spaceTab[0]; ret = ctxt->spaceTab[ctxt->spaceNr]; ctxt->spaceTab[ctxt->spaceNr] = -1; return(ret); } /* * Macros for accessing the content. Those should be used only by the parser, * and not exported. * * Dirty macros, i.e. one often need to make assumption on the context to * use them * * CUR_PTR return the current pointer to the xmlChar to be parsed. * To be used with extreme caution since operations consuming * characters may move the input buffer to a different location ! * CUR returns the current xmlChar value, i.e. a 8 bit value if compiled * This should be used internally by the parser * only to compare to ASCII values otherwise it would break when * running with UTF-8 encoding. * RAW same as CUR but in the input buffer, bypass any token * extraction that may have been done * 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 without newlines within the parser. * NEXT1(l) Skip 1 xmlChar, and must also be used only to skip 1 non-newline ASCII * defined char within the parser. * Clean macros, not dependent of an ASCII context, expect UTF-8 encoding * * NEXT Skip to the next character, this does the proper decoding * in UTF-8 mode. It also pop-up unfinished entities on the fly. * NEXTL(l) Skip the current unicode character of l xmlChars long. * CUR_CHAR(l) returns the current unicode character (int), set l * to the number of xmlChars used for the encoding [0-5]. * CUR_SCHAR same but operate on a string instead of the context * COPY_BUF copy the current unicode char to the target buffer, increment * the index * GROW, SHRINK handling of input buffers */ #define RAW (*ctxt->input->cur) #define CUR (*ctxt->input->cur) #define NXT(val) ctxt->input->cur[(val)] #define CUR_PTR ctxt->input->cur #define CMP4( s, c1, c2, c3, c4 ) \ ( ((unsigned char *) s)[ 0 ] == c1 && ((unsigned char *) s)[ 1 ] == c2 && \ ((unsigned char *) s)[ 2 ] == c3 && ((unsigned char *) s)[ 3 ] == c4 ) #define CMP5( s, c1, c2, c3, c4, c5 ) \ ( CMP4( s, c1, c2, c3, c4 ) && ((unsigned char *) s)[ 4 ] == c5 ) #define CMP6( s, c1, c2, c3, c4, c5, c6 ) \ ( CMP5( s, c1, c2, c3, c4, c5 ) && ((unsigned char *) s)[ 5 ] == c6 ) #define CMP7( s, c1, c2, c3, c4, c5, c6, c7 ) \ ( CMP6( s, c1, c2, c3, c4, c5, c6 ) && ((unsigned char *) s)[ 6 ] == c7 ) #define CMP8( s, c1, c2, c3, c4, c5, c6, c7, c8 ) \ ( CMP7( s, c1, c2, c3, c4, c5, c6, c7 ) && ((unsigned char *) s)[ 7 ] == c8 ) #define CMP9( s, c1, c2, c3, c4, c5, c6, c7, c8, c9 ) \ ( CMP8( s, c1, c2, c3, c4, c5, c6, c7, c8 ) && \ ((unsigned char *) s)[ 8 ] == c9 ) #define CMP10( s, c1, c2, c3, c4, c5, c6, c7, c8, c9, c10 ) \ ( CMP9( s, c1, c2, c3, c4, c5, c6, c7, c8, c9 ) && \ ((unsigned char *) s)[ 9 ] == c10 ) #define SKIP(val) do { \ ctxt->nbChars += (val),ctxt->input->cur += (val),ctxt->input->col+=(val); \ if (*ctxt->input->cur == '%') xmlParserHandlePEReference(ctxt); \ if ((*ctxt->input->cur == 0) && \ (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) \ xmlPopInput(ctxt); \ } while (0) #define SKIPL(val) do { \ int skipl; \ for(skipl=0; skiplinput->cur) == '\n') { \ ctxt->input->line++; ctxt->input->col = 1; \ } else ctxt->input->col++; \ ctxt->nbChars++; \ ctxt->input->cur++; \ } \ if (*ctxt->input->cur == '%') xmlParserHandlePEReference(ctxt); \ if ((*ctxt->input->cur == 0) && \ (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) \ xmlPopInput(ctxt); \ } while (0) #define SHRINK if ((ctxt->progressive == 0) && \ (ctxt->input->cur - ctxt->input->base > 2 * INPUT_CHUNK) && \ (ctxt->input->end - ctxt->input->cur < 2 * INPUT_CHUNK)) \ xmlSHRINK (ctxt); static void xmlSHRINK (xmlParserCtxtPtr ctxt) { xmlParserInputShrink(ctxt->input); if ((*ctxt->input->cur == 0) && (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) xmlPopInput(ctxt); } #define GROW if ((ctxt->progressive == 0) && \ (ctxt->input->end - ctxt->input->cur < INPUT_CHUNK)) \ xmlGROW (ctxt); static void xmlGROW (xmlParserCtxtPtr ctxt) { xmlParserInputGrow(ctxt->input, INPUT_CHUNK); if ((*ctxt->input->cur == 0) && (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0)) xmlPopInput(ctxt); } #define SKIP_BLANKS xmlSkipBlankChars(ctxt) #define NEXT xmlNextChar(ctxt) #define NEXT1 { \ ctxt->input->col++; \ ctxt->input->cur++; \ ctxt->nbChars++; \ if (*ctxt->input->cur == 0) \ xmlParserInputGrow(ctxt->input, INPUT_CHUNK); \ } #define NEXTL(l) do { \ if (*(ctxt->input->cur) == '\n') { \ ctxt->input->line++; ctxt->input->col = 1; \ } else ctxt->input->col++; \ ctxt->input->cur += l; \ if (*ctxt->input->cur == '%') xmlParserHandlePEReference(ctxt); \ } while (0) #define CUR_CHAR(l) xmlCurrentChar(ctxt, &l) #define CUR_SCHAR(s, l) xmlStringCurrentChar(ctxt, s, &l) #define COPY_BUF(l,b,i,v) \ if (l == 1) b[i++] = (xmlChar) v; \ else i += xmlCopyCharMultiByte(&b[i],v) /** * xmlSkipBlankChars: * @ctxt: the XML parser context * * skip all blanks character found at that point in the input streams. * It pops up finished entities in the process if allowable at that point. * * Returns the number of space chars skipped */ int xmlSkipBlankChars(xmlParserCtxtPtr ctxt) { int res = 0; /* * It's Okay to use CUR/NEXT here since all the blanks are on * the ASCII range. */ if ((ctxt->inputNr == 1) && (ctxt->instate != XML_PARSER_DTD)) { const xmlChar *cur; /* * if we are in the document content, go really fast */ cur = ctxt->input->cur; while (IS_BLANK_CH(*cur)) { if (*cur == '\n') { ctxt->input->line++; ctxt->input->col = 1; } cur++; res++; if (*cur == 0) { ctxt->input->cur = cur; xmlParserInputGrow(ctxt->input, INPUT_CHUNK); cur = ctxt->input->cur; } } ctxt->input->cur = cur; } else { int cur; do { cur = CUR; while (IS_BLANK_CH(cur)) { /* CHECKED tstblanks.xml */ NEXT; cur = CUR; res++; } while ((cur == 0) && (ctxt->inputNr > 1) && (ctxt->instate != XML_PARSER_COMMENT)) { xmlPopInput(ctxt); cur = CUR; } /* * Need to handle support of entities branching here */ if (*ctxt->input->cur == '%') xmlParserHandlePEReference(ctxt); } while (IS_BLANK(cur)); /* CHECKED tstblanks.xml */ } return(res); } /************************************************************************ * * * Commodity functions to handle entities * * * ************************************************************************/ /** * 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 xmlChar in the parser context */ xmlChar xmlPopInput(xmlParserCtxtPtr ctxt) { if ((ctxt == NULL) || (ctxt->inputNr <= 1)) return(0); if (xmlParserDebugEntities) xmlGenericError(xmlGenericErrorContext, "Popping input %d\n", ctxt->inputNr); 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; if (xmlParserDebugEntities) { if ((ctxt->input != NULL) && (ctxt->input->filename)) xmlGenericError(xmlGenericErrorContext, "%s(%d): ", ctxt->input->filename, ctxt->input->line); xmlGenericError(xmlGenericErrorContext, "Pushing input %d : %.30s\n", ctxt->inputNr+1, input->cur); } inputPush(ctxt, input); GROW; } /** * 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), 0 in case of error */ int xmlParseCharRef(xmlParserCtxtPtr ctxt) { unsigned int val = 0; int count = 0; unsigned int outofrange = 0; /* * Using RAW/CUR/NEXT is okay since we are working on ASCII range here */ if ((RAW == '&') && (NXT(1) == '#') && (NXT(2) == 'x')) { SKIP(3); GROW; while (RAW != ';') { /* loop blocked by count */ if (count++ > 20) { count = 0; GROW; } if ((RAW >= '0') && (RAW <= '9')) val = val * 16 + (CUR - '0'); else if ((RAW >= 'a') && (RAW <= 'f') && (count < 20)) val = val * 16 + (CUR - 'a') + 10; else if ((RAW >= 'A') && (RAW <= 'F') && (count < 20)) val = val * 16 + (CUR - 'A') + 10; else { xmlFatalErr(ctxt, XML_ERR_INVALID_HEX_CHARREF, NULL); val = 0; break; } if (val > 0x10FFFF) outofrange = val; NEXT; count++; } if (RAW == ';') { /* on purpose to avoid reentrancy problems with NEXT and SKIP */ ctxt->input->col++; ctxt->nbChars ++; ctxt->input->cur++; } } else if ((RAW == '&') && (NXT(1) == '#')) { SKIP(2); GROW; while (RAW != ';') { /* loop blocked by count */ if (count++ > 20) { count = 0; GROW; } if ((RAW >= '0') && (RAW <= '9')) val = val * 10 + (CUR - '0'); else { xmlFatalErr(ctxt, XML_ERR_INVALID_DEC_CHARREF, NULL); val = 0; break; } if (val > 0x10FFFF) outofrange = val; NEXT; count++; } if (RAW == ';') { /* on purpose to avoid reentrancy problems with NEXT and SKIP */ ctxt->input->col++; ctxt->nbChars ++; ctxt->input->cur++; } } else { xmlFatalErr(ctxt, XML_ERR_INVALID_CHARREF, NULL); } /* * [ WFC: Legal Character ] * Characters referred to using character references must match the * production for Char. */ if ((IS_CHAR(val) && (outofrange == 0))) { return(val); } else { xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, "xmlParseCharRef: invalid xmlChar value %d\n", val); } return(0); } /** * xmlParseStringCharRef: * @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 xmlParseStringCharRef(xmlParserCtxtPtr ctxt, const xmlChar **str) { const xmlChar *ptr; xmlChar cur; unsigned int val = 0; unsigned int outofrange = 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 != ';') { /* Non input consuming loop */ 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 { xmlFatalErr(ctxt, XML_ERR_INVALID_HEX_CHARREF, NULL); val = 0; break; } if (val > 0x10FFFF) outofrange = val; ptr++; cur = *ptr; } if (cur == ';') ptr++; } else if ((cur == '&') && (ptr[1] == '#')){ ptr += 2; cur = *ptr; while (cur != ';') { /* Non input consuming loops */ if ((cur >= '0') && (cur <= '9')) val = val * 10 + (cur - '0'); else { xmlFatalErr(ctxt, XML_ERR_INVALID_DEC_CHARREF, NULL); val = 0; break; } if (val > 0x10FFFF) outofrange = val; ptr++; cur = *ptr; } if (cur == ';') ptr++; } else { xmlFatalErr(ctxt, XML_ERR_INVALID_CHARREF, NULL); return(0); } *str = ptr; /* * [ WFC: Legal Character ] * Characters referred to using character references must match the * production for Char. */ if ((IS_CHAR(val) && (outofrange == 0))) { return(val); } else { xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, "xmlParseStringCharRef: invalid xmlChar value %d\n", val); } return(0); } /** * xmlNewBlanksWrapperInputStream: * @ctxt: an XML parser context * @entity: an Entity pointer * * Create a new input stream for wrapping * blanks around a PEReference * * Returns the new input stream or NULL */ static void deallocblankswrapper (xmlChar *str) {xmlFree(str);} static xmlParserInputPtr xmlNewBlanksWrapperInputStream(xmlParserCtxtPtr ctxt, xmlEntityPtr entity) { xmlParserInputPtr input; xmlChar *buffer; size_t length; if (entity == NULL) { xmlFatalErr(ctxt, XML_ERR_INTERNAL_ERROR, "xmlNewBlanksWrapperInputStream entity\n"); return(NULL); } if (xmlParserDebugEntities) xmlGenericError(xmlGenericErrorContext, "new blanks wrapper for entity: %s\n", entity->name); input = xmlNewInputStream(ctxt); if (input == NULL) { return(NULL); } length = xmlStrlen(entity->name) + 5; buffer = xmlMallocAtomic(length); if (buffer == NULL) { xmlErrMemory(ctxt, NULL); xmlFree(input); return(NULL); } buffer [0] = ' '; buffer [1] = '%'; buffer [length-3] = ';'; buffer [length-2] = ' '; buffer [length-1] = 0; memcpy(buffer + 2, entity->name, length - 5); input->free = deallocblankswrapper; input->base = buffer; input->cur = buffer; input->length = length; input->end = &buffer[length]; return(input); } /** * xmlParserHandlePEReference: * @ctxt: the parser context * * [69] PEReference ::= '%' Name ';' * * [ WFC: No Recursion ] * 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 Parameter Entity reference within DTDs */ void xmlParserHandlePEReference(xmlParserCtxtPtr ctxt) { const xmlChar *name; xmlEntityPtr entity = NULL; xmlParserInputPtr input; if (RAW != '%') 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: xmlFatalErr(ctxt, XML_ERR_PEREF_AT_EOF, NULL); return; case XML_PARSER_PROLOG: case XML_PARSER_START: case XML_PARSER_MISC: xmlFatalErr(ctxt, XML_ERR_PEREF_IN_PROLOG, NULL); return; case XML_PARSER_ENTITY_DECL: case XML_PARSER_CONTENT: case XML_PARSER_ATTRIBUTE_VALUE: case XML_PARSER_PI: case XML_PARSER_SYSTEM_LITERAL: case XML_PARSER_PUBLIC_LITERAL: /* we just ignore it there */ return; case XML_PARSER_EPILOG: xmlFatalErr(ctxt, XML_ERR_PEREF_IN_EPILOG, NULL); 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 xmlStringDecodeEntities */ 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; if (IS_BLANK_CH(NXT(1)) || NXT(1) == 0) return; break; case XML_PARSER_IGNORE: return; } NEXT; name = xmlParseName(ctxt); if (xmlParserDebugEntities) xmlGenericError(xmlGenericErrorContext, "PEReference: %s\n", name); if (name == NULL) { xmlFatalErr(ctxt, XML_ERR_PEREF_NO_NAME, NULL); } else { if (RAW == ';') { 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))) { xmlFatalErrMsgStr(ctxt, XML_ERR_UNDECLARED_ENTITY, "PEReference: %%%s; not found\n", name); } 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->validate) && (ctxt->vctxt.error != NULL)) { xmlValidityError(ctxt, XML_WAR_UNDECLARED_ENTITY, "PEReference: %%%s; not found\n", name); } else xmlWarningMsg(ctxt, XML_WAR_UNDECLARED_ENTITY, "PEReference: %%%s; not found\n", name, NULL); ctxt->valid = 0; } } else if (ctxt->input->free != deallocblankswrapper) { input = xmlNewBlanksWrapperInputStream(ctxt, entity); xmlPushInput(ctxt, input); } else { if ((entity->etype == XML_INTERNAL_PARAMETER_ENTITY) || (entity->etype == XML_EXTERNAL_PARAMETER_ENTITY)) { xmlChar start[4]; xmlCharEncoding enc; /* * handle the extra spaces added before and after * c.f. http://www.w3.org/TR/REC-xml#as-PE * this is done independently. */ input = xmlNewEntityInputStream(ctxt, entity); xmlPushInput(ctxt, input); /* * Get the 4 first bytes and decode the charset * if enc != XML_CHAR_ENCODING_NONE * plug some encoding conversion routines. * Note that, since we may have some non-UTF8 * encoding (like UTF16, bug 135229), the 'length' * is not known, but we can calculate based upon * the amount of data in the buffer. */ GROW if ((ctxt->input->end - ctxt->input->cur)>=4) { start[0] = RAW; start[1] = NXT(1); start[2] = NXT(2); start[3] = NXT(3); enc = xmlDetectCharEncoding(start, 4); if (enc != XML_CHAR_ENCODING_NONE) { xmlSwitchEncoding(ctxt, enc); } } if ((entity->etype == XML_EXTERNAL_PARAMETER_ENTITY) && (CMP5(CUR_PTR, '<', '?', 'x', 'm', 'l' )) && (IS_BLANK_CH(NXT(5)))) { xmlParseTextDecl(ctxt); } } else { xmlFatalErrMsgStr(ctxt, XML_ERR_ENTITY_IS_PARAMETER, "PEReference: %s is not a parameter entity\n", name); } } } else { xmlFatalErr(ctxt, XML_ERR_PEREF_SEMICOL_MISSING, NULL); } } } /* * Macro used to grow the current buffer. */ #define growBuffer(buffer) { \ xmlChar *tmp; \ buffer##_size *= 2; \ tmp = (xmlChar *) \ xmlRealloc(buffer, buffer##_size * sizeof(xmlChar)); \ if (tmp == NULL) goto mem_error; \ buffer = tmp; \ } /** * xmlStringLenDecodeEntities: * @ctxt: the parser context * @str: the input string * @len: the string length * @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 * * Takes a entity string content and process to do the adequate substitutions. * * [67] Reference ::= EntityRef | CharRef * * [69] PEReference ::= '%' Name ';' * * Returns A newly allocated string with the substitution done. The caller * must deallocate it ! */ xmlChar * xmlStringLenDecodeEntities(xmlParserCtxtPtr ctxt, const xmlChar *str, int len, int what, xmlChar end, xmlChar end2, xmlChar end3) { xmlChar *buffer = NULL; int buffer_size = 0; xmlChar *current = NULL; xmlChar *rep = NULL; const xmlChar *last; xmlEntityPtr ent; int c,l; int nbchars = 0; if ((ctxt == NULL) || (str == NULL) || (len < 0)) return(NULL); last = str + len; if (ctxt->depth > 40) { xmlFatalErr(ctxt, XML_ERR_ENTITY_LOOP, NULL); return(NULL); } /* * allocate a translation buffer. */ buffer_size = XML_PARSER_BIG_BUFFER_SIZE; buffer = (xmlChar *) xmlMallocAtomic(buffer_size * sizeof(xmlChar)); if (buffer == NULL) goto mem_error; /* * OK loop until we reach one of the ending char or a size limit. * we are operating on already parsed values. */ if (str < last) c = CUR_SCHAR(str, l); else c = 0; while ((c != 0) && (c != end) && /* non input consuming loop */ (c != end2) && (c != end3)) { if (c == 0) break; if ((c == '&') && (str[1] == '#')) { int val = xmlParseStringCharRef(ctxt, &str); if (val != 0) { COPY_BUF(0,buffer,nbchars,val); } if (nbchars > buffer_size - XML_PARSER_BUFFER_SIZE) { growBuffer(buffer); } } else if ((c == '&') && (what & XML_SUBSTITUTE_REF)) { if (xmlParserDebugEntities) xmlGenericError(xmlGenericErrorContext, "String decoding Entity Reference: %.30s\n", str); ent = xmlParseStringEntityRef(ctxt, &str); if ((ent != NULL) && (ent->etype == XML_INTERNAL_PREDEFINED_ENTITY)) { if (ent->content != NULL) { COPY_BUF(0,buffer,nbchars,ent->content[0]); if (nbchars > buffer_size - XML_PARSER_BUFFER_SIZE) { growBuffer(buffer); } } else { xmlFatalErrMsg(ctxt, XML_ERR_INTERNAL_ERROR, "predefined entity has no content\n"); } } else if ((ent != NULL) && (ent->content != NULL)) { ctxt->depth++; rep = xmlStringDecodeEntities(ctxt, ent->content, what, 0, 0, 0); ctxt->depth--; if (rep != NULL) { current = rep; while (*current != 0) { /* non input consuming loop */ buffer[nbchars++] = *current++; if (nbchars > buffer_size - XML_PARSER_BUFFER_SIZE) { growBuffer(buffer); } } xmlFree(rep); rep = NULL; } } else if (ent != NULL) { int i = xmlStrlen(ent->name); const xmlChar *cur = ent->name; buffer[nbchars++] = '&'; if (nbchars > buffer_size - i - XML_PARSER_BUFFER_SIZE) { growBuffer(buffer); } for (;i > 0;i--) buffer[nbchars++] = *cur++; buffer[nbchars++] = ';'; } } else if (c == '%' && (what & XML_SUBSTITUTE_PEREF)) { if (xmlParserDebugEntities) xmlGenericError(xmlGenericErrorContext, "String decoding PE Reference: %.30s\n", str); ent = xmlParseStringPEReference(ctxt, &str); if (ent != NULL) { if (ent->content == NULL) { if (xmlLoadEntityContent(ctxt, ent) < 0) { } } ctxt->depth++; rep = xmlStringDecodeEntities(ctxt, ent->content, what, 0, 0, 0); ctxt->depth--; if (rep != NULL) { current = rep; while (*current != 0) { /* non input consuming loop */ buffer[nbchars++] = *current++; if (nbchars > buffer_size - XML_PARSER_BUFFER_SIZE) { growBuffer(buffer); } } xmlFree(rep); rep = NULL; } } } else { COPY_BUF(l,buffer,nbchars,c); str += l; if (nbchars > buffer_size - XML_PARSER_BUFFER_SIZE) { growBuffer(buffer); } } if (str < last) c = CUR_SCHAR(str, l); else c = 0; } buffer[nbchars++] = 0; return(buffer); mem_error: xmlErrMemory(ctxt, NULL); if (rep != NULL) xmlFree(rep); if (buffer != NULL) xmlFree(buffer); return(NULL); } /** * xmlStringDecodeEntities: * @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 * * Takes a entity string content and process to do the adequate substitutions. * * [67] Reference ::= EntityRef | CharRef * * [69] PEReference ::= '%' Name ';' * * Returns A newly allocated string with the substitution done. The caller * must deallocate it ! */ xmlChar * xmlStringDecodeEntities(xmlParserCtxtPtr ctxt, const xmlChar *str, int what, xmlChar end, xmlChar end2, xmlChar end3) { if ((ctxt == NULL) || (str == NULL)) return(NULL); return(xmlStringLenDecodeEntities(ctxt, str, xmlStrlen(str), what, end, end2, end3)); } /************************************************************************ * * * Commodity functions, cleanup needed ? * * * ************************************************************************/ /** * areBlanks: * @ctxt: an XML parser context * @str: a xmlChar * * @len: the size of @str * @blank_chars: we know the chars are blanks * * Is this a sequence of blank chars that one can ignore ? * * Returns 1 if ignorable 0 otherwise. */ static int areBlanks(xmlParserCtxtPtr ctxt, const xmlChar *str, int len, int blank_chars) { int i, ret; xmlNodePtr lastChild; /* * Don't spend time trying to differentiate them, the same callback is * used ! */ if (ctxt->sax->ignorableWhitespace == ctxt->sax->characters) return(0); /* * Check for xml:space value. */ if ((ctxt->space == NULL) || (*(ctxt->space) == 1) || (*(ctxt->space) == -2)) return(0); /* * Check that the string is made of blanks */ if (blank_chars == 0) { for (i = 0;i < len;i++) if (!(IS_BLANK_CH(str[i]))) return(0); } /* * Look if the element is mixed content in the DTD if available */ 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); } /* * Otherwise, heuristic :-\ */ if ((RAW != '<') && (RAW != 0xD)) return(0); if ((ctxt->node->children == NULL) && (RAW == '<') && (NXT(1) == '/')) return(0); lastChild = xmlGetLastChild(ctxt->node); if (lastChild == NULL) { if ((ctxt->node->type != XML_ELEMENT_NODE) && (ctxt->node->content != NULL)) return(0); } else if (xmlNodeIsText(lastChild)) return(0); else if ((ctxt->node->children != NULL) && (xmlNodeIsText(ctxt->node->children))) return(0); return(1); } /************************************************************************ * * * Extra stuff for namespace support * * Relates to http://www.w3.org/TR/WD-xml-names * * * ************************************************************************/ /** * xmlSplitQName: * @ctxt: an XML parser context * @name: an XML parser context * @prefix: a xmlChar ** * * parse an UTF8 encoded XML qualified name string * * [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. */ xmlChar * xmlSplitQName(xmlParserCtxtPtr ctxt, const xmlChar *name, xmlChar **prefix) { xmlChar buf[XML_MAX_NAMELEN + 5]; xmlChar *buffer = NULL; int len = 0; int max = XML_MAX_NAMELEN; xmlChar *ret = NULL; const xmlChar *cur = name; int c; if (prefix == NULL) return(NULL); *prefix = NULL; if (cur == NULL) return(NULL); #ifndef XML_XML_NAMESPACE /* xml: prefix is not really a namespace */ if ((cur[0] == 'x') && (cur[1] == 'm') && (cur[2] == 'l') && (cur[3] == ':')) return(xmlStrdup(name)); #endif /* nasty but well=formed */ if (cur[0] == ':') return(xmlStrdup(name)); c = *cur++; while ((c != 0) && (c != ':') && (len < max)) { /* tested bigname.xml */ buf[len++] = c; c = *cur++; } if (len >= max) { /* * Okay someone managed to make a huge name, so he's ready to pay * for the processing speed. */ max = len * 2; buffer = (xmlChar *) xmlMallocAtomic(max * sizeof(xmlChar)); if (buffer == NULL) { xmlErrMemory(ctxt, NULL); return(NULL); } memcpy(buffer, buf, len); while ((c != 0) && (c != ':')) { /* tested bigname.xml */ if (len + 10 > max) { xmlChar *tmp; max *= 2; tmp = (xmlChar *) xmlRealloc(buffer, max * sizeof(xmlChar)); if (tmp == NULL) { xmlFree(buffer); xmlErrMemory(ctxt, NULL); return(NULL); } buffer = tmp; } buffer[len++] = c; c = *cur++; } buffer[len] = 0; } if ((c == ':') && (*cur == 0)) { if (buffer != NULL) xmlFree(buffer); *prefix = NULL; return(xmlStrdup(name)); } if (buffer == NULL) ret = xmlStrndup(buf, len); else { ret = buffer; buffer = NULL; max = XML_MAX_NAMELEN; } if (c == ':') { c = *cur; *prefix = ret; if (c == 0) { return(xmlStrndup(BAD_CAST "", 0)); } len = 0; /* * Check that the first character is proper to start * a new name */ if (!(((c >= 0x61) && (c <= 0x7A)) || ((c >= 0x41) && (c <= 0x5A)) || (c == '_') || (c == ':'))) { int l; int first = CUR_SCHAR(cur, l); if (!IS_LETTER(first) && (first != '_')) { xmlFatalErrMsgStr(ctxt, XML_NS_ERR_QNAME, "Name %s is not XML Namespace compliant\n", name); } } cur++; while ((c != 0) && (len < max)) { /* tested bigname2.xml */ buf[len++] = c; c = *cur++; } if (len >= max) { /* * Okay someone managed to make a huge name, so he's ready to pay * for the processing speed. */ max = len * 2; buffer = (xmlChar *) xmlMallocAtomic(max * sizeof(xmlChar)); if (buffer == NULL) { xmlErrMemory(ctxt, NULL); return(NULL); } memcpy(buffer, buf, len); while (c != 0) { /* tested bigname2.xml */ if (len + 10 > max) { xmlChar *tmp; max *= 2; tmp = (xmlChar *) xmlRealloc(buffer, max * sizeof(xmlChar)); if (tmp == NULL) { xmlErrMemory(ctxt, NULL); xmlFree(buffer); return(NULL); } buffer = tmp; } buffer[len++] = c; c = *cur++; } buffer[len] = 0; } if (buffer == NULL) ret = xmlStrndup(buf, len); else { ret = buffer; } } return(ret); } /************************************************************************ * * * The parser itself * * Relates to http://www.w3.org/TR/REC-xml * * * ************************************************************************/ static const xmlChar * xmlParseNameComplex(xmlParserCtxtPtr ctxt); static xmlChar * xmlParseAttValueInternal(xmlParserCtxtPtr ctxt, int *len, int *alloc, int normalize); /** * xmlParseName: * @ctxt: an XML parser context * * parse an XML name. * * [4] NameChar ::= Letter | Digit | '.' | '-' | '_' | ':' | * CombiningChar | Extender * * [5] Name ::= (Letter | '_' | ':') (NameChar)* * * [6] Names ::= Name (#x20 Name)* * * Returns the Name parsed or NULL */ const xmlChar * xmlParseName(xmlParserCtxtPtr ctxt) { const xmlChar *in; const xmlChar *ret; int count = 0; GROW; /* * Accelerator for simple ASCII names */ in = ctxt->input->cur; if (((*in >= 0x61) && (*in <= 0x7A)) || ((*in >= 0x41) && (*in <= 0x5A)) || (*in == '_') || (*in == ':')) { in++; while (((*in >= 0x61) && (*in <= 0x7A)) || ((*in >= 0x41) && (*in <= 0x5A)) || ((*in >= 0x30) && (*in <= 0x39)) || (*in == '_') || (*in == '-') || (*in == ':') || (*in == '.')) in++; if ((*in > 0) && (*in < 0x80)) { count = in - ctxt->input->cur; ret = xmlDictLookup(ctxt->dict, ctxt->input->cur, count); ctxt->input->cur = in; ctxt->nbChars += count; ctxt->input->col += count; if (ret == NULL) xmlErrMemory(ctxt, NULL); return(ret); } } return(xmlParseNameComplex(ctxt)); } /** * xmlParseNameAndCompare: * @ctxt: an XML parser context * * parse an XML name and compares for match * (specialized for endtag parsing) * * Returns NULL for an illegal name, (xmlChar*) 1 for success * and the name for mismatch */ static const xmlChar * xmlParseNameAndCompare(xmlParserCtxtPtr ctxt, xmlChar const *other) { register const xmlChar *cmp = other; register const xmlChar *in; const xmlChar *ret; GROW; in = ctxt->input->cur; while (*in != 0 && *in == *cmp) { ++in; ++cmp; ctxt->input->col++; } if (*cmp == 0 && (*in == '>' || IS_BLANK_CH (*in))) { /* success */ ctxt->input->cur = in; return (const xmlChar*) 1; } /* failure (or end of input buffer), check with full function */ ret = xmlParseName (ctxt); /* strings coming from the dictionnary direct compare possible */ if (ret == other) { return (const xmlChar*) 1; } return ret; } static const xmlChar * xmlParseNameComplex(xmlParserCtxtPtr ctxt) { int len = 0, l; int c; int count = 0; /* * Handler for more complex cases */ GROW; c = CUR_CHAR(l); if ((c == ' ') || (c == '>') || (c == '/') || /* accelerators */ (!IS_LETTER(c) && (c != '_') && (c != ':'))) { return(NULL); } while ((c != ' ') && (c != '>') && (c != '/') && /* test bigname.xml */ ((IS_LETTER(c)) || (IS_DIGIT(c)) || (c == '.') || (c == '-') || (c == '_') || (c == ':') || (IS_COMBINING(c)) || (IS_EXTENDER(c)))) { if (count++ > 100) { count = 0; GROW; } len += l; NEXTL(l); c = CUR_CHAR(l); } if ((*ctxt->input->cur == '\n') && (ctxt->input->cur[-1] == '\r')) return(xmlDictLookup(ctxt->dict, ctxt->input->cur - (len + 1), len)); return(xmlDictLookup(ctxt->dict, ctxt->input->cur - len, len)); } /** * xmlParseStringName: * @ctxt: an XML parser context * @str: a pointer to the string pointer (IN/OUT) * * parse an XML name. * * [4] NameChar ::= Letter | Digit | '.' | '-' | '_' | ':' | * CombiningChar | Extender * * [5] Name ::= (Letter | '_' | ':') (NameChar)* * * [6] Names ::= Name (#x20 Name)* * * Returns the Name parsed or NULL. The @str pointer * is updated to the current location in the string. */ static xmlChar * xmlParseStringName(xmlParserCtxtPtr ctxt, const xmlChar** str) { xmlChar buf[XML_MAX_NAMELEN + 5]; const xmlChar *cur = *str; int len = 0, l; int c; c = CUR_SCHAR(cur, l); if (!IS_LETTER(c) && (c != '_') && (c != ':')) { return(NULL); } while ((IS_LETTER(c)) || (IS_DIGIT(c)) || /* test bigentname.xml */ (c == '.') || (c == '-') || (c == '_') || (c == ':') || (IS_COMBINING(c)) || (IS_EXTENDER(c))) { COPY_BUF(l,buf,len,c); cur += l; c = CUR_SCHAR(cur, l); if (len >= XML_MAX_NAMELEN) { /* test bigentname.xml */ /* * Okay someone managed to make a huge name, so he's ready to pay * for the processing speed. */ xmlChar *buffer; int max = len * 2; buffer = (xmlChar *) xmlMallocAtomic(max * sizeof(xmlChar)); if (buffer == NULL) { xmlErrMemory(ctxt, NULL); return(NULL); } memcpy(buffer, buf, len); while ((IS_LETTER(c)) || (IS_DIGIT(c)) || /* test bigentname.xml */ (c == '.') || (c == '-') || (c == '_') || (c == ':') || (IS_COMBINING(c)) || (IS_EXTENDER(c))) { if (len + 10 > max) { xmlChar *tmp; max *= 2; tmp = (xmlChar *) xmlRealloc(buffer, max * sizeof(xmlChar)); if (tmp == NULL) { xmlErrMemory(ctxt, NULL); xmlFree(buffer); return(NULL); } buffer = tmp; } COPY_BUF(l,buffer,len,c); cur += l; c = CUR_SCHAR(cur, l); } buffer[len] = 0; *str = cur; return(buffer); } } *str = cur; return(xmlStrndup(buf, len)); } /** * xmlParseNmtoken: * @ctxt: an XML parser context * * parse an XML Nmtoken. * * [7] Nmtoken ::= (NameChar)+ * * [8] Nmtokens ::= Nmtoken (#x20 Nmtoken)* * * Returns the Nmtoken parsed or NULL */ xmlChar * xmlParseNmtoken(xmlParserCtxtPtr ctxt) { xmlChar buf[XML_MAX_NAMELEN + 5]; int len = 0, l; int c; int count = 0; GROW; c = CUR_CHAR(l); while ((IS_LETTER(c)) || (IS_DIGIT(c)) || /* test bigtoken.xml */ (c == '.') || (c == '-') || (c == '_') || (c == ':') || (IS_COMBINING(c)) || (IS_EXTENDER(c))) { if (count++ > 100) { count = 0; GROW; } COPY_BUF(l,buf,len,c); NEXTL(l); c = CUR_CHAR(l); if (len >= XML_MAX_NAMELEN) { /* * Okay someone managed to make a huge token, so he's ready to pay * for the processing speed. */ xmlChar *buffer; int max = len * 2; buffer = (xmlChar *) xmlMallocAtomic(max * sizeof(xmlChar)); if (buffer == NULL) { xmlErrMemory(ctxt, NULL); return(NULL); } memcpy(buffer, buf, len); while ((IS_LETTER(c)) || (IS_DIGIT(c)) || /* test bigtoken.xml */ (c == '.') || (c == '-') || (c == '_') || (c == ':') || (IS_COMBINING(c)) || (IS_EXTENDER(c))) { if (count++ > 100) { count = 0; GROW; } if (len + 10 > max) { xmlChar *tmp; max *= 2; tmp = (xmlChar *) xmlRealloc(buffer, max * sizeof(xmlChar)); if (tmp == NULL) { xmlErrMemory(ctxt, NULL); xmlFree(buffer); return(NULL); } buffer = tmp; } COPY_BUF(l,buffer,len,c); NEXTL(l); c = CUR_CHAR(l); } buffer[len] = 0; return(buffer); } } if (len == 0) return(NULL); 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 declarations * * [9] EntityValue ::= '"' ([^%&"] | PEReference | Reference)* '"' | * "'" ([^%&'] | PEReference | Reference)* "'" * * Returns the EntityValue parsed with reference substituted or NULL */ xmlChar * xmlParseEntityValue(xmlParserCtxtPtr ctxt, xmlChar **orig) { xmlChar *buf = NULL; int len = 0; int size = XML_PARSER_BUFFER_SIZE; int c, l; xmlChar stop; xmlChar *ret = NULL; const xmlChar *cur = NULL; xmlParserInputPtr input; if (RAW == '"') stop = '"'; else if (RAW == '\'') stop = '\''; else { xmlFatalErr(ctxt, XML_ERR_ENTITY_NOT_STARTED, NULL); return(NULL); } buf = (xmlChar *) xmlMallocAtomic(size * sizeof(xmlChar)); if (buf == NULL) { xmlErrMemory(ctxt, NULL); 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_CHAR(l); /* * 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) || /* checked */ (ctxt->input != input))) { if (len + 5 >= size) { xmlChar *tmp; size *= 2; tmp = (xmlChar *) xmlRealloc(buf, size * sizeof(xmlChar)); if (tmp == NULL) { xmlErrMemory(ctxt, NULL); xmlFree(buf); return(NULL); } buf = tmp; } COPY_BUF(l,buf,len,c); NEXTL(l); /* * Pop-up of finished entities. */ while ((RAW == 0) && (ctxt->inputNr > 1)) /* non input consuming */ xmlPopInput(ctxt); GROW; c = CUR_CHAR(l); if (c == 0) { GROW; c = CUR_CHAR(l); } } buf[len] = 0; /* * Raise problem w.r.t. '&' and '%' being used in non-entities * reference constructs. Note Charref will be handled in * xmlStringDecodeEntities() */ cur = buf; while (*cur != 0) { /* non input consuming */ if ((*cur == '%') || ((*cur == '&') && (cur[1] != '#'))) { xmlChar *name; xmlChar tmp = *cur; cur++; name = xmlParseStringName(ctxt, &cur); if ((name == NULL) || (*cur != ';')) { xmlFatalErrMsgInt(ctxt, XML_ERR_ENTITY_CHAR_ERROR, "EntityValue: '%c' forbidden except for entities references\n", tmp); } if ((tmp == '%') && (ctxt->inSubset == 1) && (ctxt->inputNr == 1)) { xmlFatalErr(ctxt, XML_ERR_ENTITY_PE_INTERNAL, NULL); } if (name != NULL) xmlFree(name); if (*cur == 0) break; } cur++; } /* * Then PEReference entities are substituted. */ if (c != stop) { xmlFatalErr(ctxt, XML_ERR_ENTITY_NOT_FINISHED, NULL); xmlFree(buf); } 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 here. */ ret = xmlStringDecodeEntities(ctxt, buf, XML_SUBSTITUTE_PEREF, 0, 0, 0); if (orig != NULL) *orig = buf; else xmlFree(buf); } return(ret); } /** * xmlParseAttValueComplex: * @ctxt: an XML parser context * @len: the resulting attribute len * @normalize: wether to apply the inner normalization * * parse a value for an attribute, this is the fallback function * of xmlParseAttValue() when the attribute parsing requires handling * of non-ASCII characters, or normalization compaction. * * Returns the AttValue parsed or NULL. The value has to be freed by the caller. */ static xmlChar * xmlParseAttValueComplex(xmlParserCtxtPtr ctxt, int *attlen, int normalize) { xmlChar limit = 0; xmlChar *buf = NULL; xmlChar *rep = NULL; int len = 0; int buf_size = 0; int c, l, in_space = 0; xmlChar *current = NULL; xmlEntityPtr ent; if (NXT(0) == '"') { ctxt->instate = XML_PARSER_ATTRIBUTE_VALUE; limit = '"'; NEXT; } else if (NXT(0) == '\'') { limit = '\''; ctxt->instate = XML_PARSER_ATTRIBUTE_VALUE; NEXT; } else { xmlFatalErr(ctxt, XML_ERR_ATTRIBUTE_NOT_STARTED, NULL); return(NULL); } /* * allocate a translation buffer. */ buf_size = XML_PARSER_BUFFER_SIZE; buf = (xmlChar *) xmlMallocAtomic(buf_size * sizeof(xmlChar)); if (buf == NULL) goto mem_error; /* * OK loop until we reach one of the ending char or a size limit. */ c = CUR_CHAR(l); while ((NXT(0) != limit) && /* checked */ (IS_CHAR(c)) && (c != '<')) { if (c == 0) break; if (c == '&') { in_space = 0; if (NXT(1) == '#') { int val = xmlParseCharRef(ctxt); if (val == '&') { if (ctxt->replaceEntities) { if (len > buf_size - 10) { growBuffer(buf); } buf[len++] = '&'; } else { /* * The reparsing will be done in xmlStringGetNodeList() * called by the attribute() function in SAX.c */ if (len > buf_size - 10) { growBuffer(buf); } buf[len++] = '&'; buf[len++] = '#'; buf[len++] = '3'; buf[len++] = '8'; buf[len++] = ';'; } } else if (val != 0) { if (len > buf_size - 10) { growBuffer(buf); } len += xmlCopyChar(0, &buf[len], val); } } else { ent = xmlParseEntityRef(ctxt); if ((ent != NULL) && (ent->etype == XML_INTERNAL_PREDEFINED_ENTITY)) { if (len > buf_size - 10) { growBuffer(buf); } if ((ctxt->replaceEntities == 0) && (ent->content[0] == '&')) { buf[len++] = '&'; buf[len++] = '#'; buf[len++] = '3'; buf[len++] = '8'; buf[len++] = ';'; } else { buf[len++] = ent->content[0]; } } else if ((ent != NULL) && (ctxt->replaceEntities != 0)) { if (ent->etype != XML_INTERNAL_PREDEFINED_ENTITY) { rep = xmlStringDecodeEntities(ctxt, ent->content, XML_SUBSTITUTE_REF, 0, 0, 0); if (rep != NULL) { current = rep; while (*current != 0) { /* non input consuming */ buf[len++] = *current++; if (len > buf_size - 10) { growBuffer(buf); } } xmlFree(rep); rep = NULL; } } else { if (len > buf_size - 10) { growBuffer(buf); } if (ent->content != NULL) buf[len++] = ent->content[0]; } } else if (ent != NULL) { int i = xmlStrlen(ent->name); const xmlChar *cur = ent->name; /* * This may look absurd but is needed to detect * entities problems */ if ((ent->etype != XML_INTERNAL_PREDEFINED_ENTITY) && (ent->content != NULL)) { rep = xmlStringDecodeEntities(ctxt, ent->content, XML_SUBSTITUTE_REF, 0, 0, 0); if (rep != NULL) { xmlFree(rep); rep = NULL; } } /* * Just output the reference */ buf[len++] = '&'; if (len > buf_size - i - 10) { growBuffer(buf); } for (;i > 0;i--) buf[len++] = *cur++; buf[len++] = ';'; } } } else { if ((c == 0x20) || (c == 0xD) || (c == 0xA) || (c == 0x9)) { if ((len != 0) || (!normalize)) { if ((!normalize) || (!in_space)) { COPY_BUF(l,buf,len,0x20); if (len > buf_size - 10) { growBuffer(buf); } } in_space = 1; } } else { in_space = 0; COPY_BUF(l,buf,len,c); if (len > buf_size - 10) { growBuffer(buf); } } NEXTL(l); } GROW; c = CUR_CHAR(l); } if ((in_space) && (normalize)) { while (buf[len - 1] == 0x20) len--; } buf[len] = 0; if (RAW == '<') { xmlFatalErr(ctxt, XML_ERR_LT_IN_ATTRIBUTE, NULL); } else if (RAW != limit) { if ((c != 0) && (!IS_CHAR(c))) { xmlFatalErrMsg(ctxt, XML_ERR_INVALID_CHAR, "invalid character in attribute value\n"); } else { xmlFatalErrMsg(ctxt, XML_ERR_ATTRIBUTE_NOT_FINISHED, "AttValue: ' expected\n"); } } else NEXT; if (attlen != NULL) *attlen = len; return(buf); mem_error: xmlErrMemory(ctxt, NULL); if (buf != NULL) xmlFree(buf); if (rep != NULL) xmlFree(rep); return(NULL); } /** * 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)* "'" * * 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. */ xmlChar * xmlParseAttValue(xmlParserCtxtPtr ctxt) { if ((ctxt == NULL) || (ctxt->input == NULL)) return(NULL); return(xmlParseAttValueInternal(ctxt, NULL, NULL, 0)); } /** * xmlParseSystemLiteral: * @ctxt: an XML parser context * * parse an XML Literal * * [11] SystemLiteral ::= ('"' [^"]* '"') | ("'" [^']* "'") * * Returns the SystemLiteral parsed or NULL */ xmlChar * xmlParseSystemLiteral(xmlParserCtxtPtr ctxt) { xmlChar *buf = NULL; int len = 0; int size = XML_PARSER_BUFFER_SIZE; int cur, l; xmlChar stop; int state = ctxt->instate; int count = 0; SHRINK; if (RAW == '"') { NEXT; stop = '"'; } else if (RAW == '\'') { NEXT; stop = '\''; } else { xmlFatalErr(ctxt, XML_ERR_LITERAL_NOT_STARTED, NULL); return(NULL); } buf = (xmlChar *) xmlMallocAtomic(size * sizeof(xmlChar)); if (buf == NULL) { xmlErrMemory(ctxt, NULL); return(NULL); } ctxt->instate = XML_PARSER_SYSTEM_LITERAL; cur = CUR_CHAR(l); while ((IS_CHAR(cur)) && (cur != stop)) { /* checked */ if (len + 5 >= size) { xmlChar *tmp; size *= 2; tmp = (xmlChar *) xmlRealloc(buf, size * sizeof(xmlChar)); if (tmp == NULL) { xmlFree(buf); xmlErrMemory(ctxt, NULL); ctxt->instate = (xmlParserInputState) state; return(NULL); } buf = tmp; } count++; if (count > 50) { GROW; count = 0; } COPY_BUF(l,buf,len,cur); NEXTL(l); cur = CUR_CHAR(l); if (cur == 0) { GROW; SHRINK; cur = CUR_CHAR(l); } } buf[len] = 0; ctxt->instate = (xmlParserInputState) state; if (!IS_CHAR(cur)) { xmlFatalErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED, NULL); } else { NEXT; } return(buf); } /** * xmlParsePubidLiteral: * @ctxt: an XML parser context * * parse an XML public literal * * [12] PubidLiteral ::= '"' PubidChar* '"' | "'" (PubidChar - "'")* "'" * * Returns the PubidLiteral parsed or NULL. */ xmlChar * xmlParsePubidLiteral(xmlParserCtxtPtr ctxt) { xmlChar *buf = NULL; int len = 0; int size = XML_PARSER_BUFFER_SIZE; xmlChar cur; xmlChar stop; int count = 0; xmlParserInputState oldstate = ctxt->instate; SHRINK; if (RAW == '"') { NEXT; stop = '"'; } else if (RAW == '\'') { NEXT; stop = '\''; } else { xmlFatalErr(ctxt, XML_ERR_LITERAL_NOT_STARTED, NULL); return(NULL); } buf = (xmlChar *) xmlMallocAtomic(size * sizeof(xmlChar)); if (buf == NULL) { xmlErrMemory(ctxt, NULL); return(NULL); } ctxt->instate = XML_PARSER_PUBLIC_LITERAL; cur = CUR; while ((IS_PUBIDCHAR_CH(cur)) && (cur != stop)) { /* checked */ if (len + 1 >= size) { xmlChar *tmp; size *= 2; tmp = (xmlChar *) xmlRealloc(buf, size * sizeof(xmlChar)); if (tmp == NULL) { xmlErrMemory(ctxt, NULL); xmlFree(buf); return(NULL); } buf = tmp; } buf[len++] = cur; count++; if (count > 50) { GROW; count = 0; } NEXT; cur = CUR; if (cur == 0) { GROW; SHRINK; cur = CUR; } } buf[len] = 0; if (cur != stop) { xmlFatalErr(ctxt, XML_ERR_LITERAL_NOT_FINISHED, NULL); } else { NEXT; } ctxt->instate = oldstate; return(buf); } void xmlParseCharDataComplex(xmlParserCtxtPtr ctxt, int cdata); /* * used for the test in the inner loop of the char data testing */ static const unsigned char test_char_data[256] = { 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x09, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 0x9, CR/LF separated */ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x00, 0x27, /* & */ 0x28, 0x29, 0x2A, 0x2B, 0x2C, 0x2D, 0x2E, 0x2F, 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, 0x38, 0x39, 0x3A, 0x3B, 0x00, 0x3D, 0x3E, 0x3F, /* < */ 0x40, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, 0x48, 0x49, 0x4A, 0x4B, 0x4C, 0x4D, 0x4E, 0x4F, 0x50, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57, 0x58, 0x59, 0x5A, 0x5B, 0x5C, 0x00, 0x5E, 0x5F, /* ] */ 0x60, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F, 0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, 0x7A, 0x7B, 0x7C, 0x7D, 0x7E, 0x7F, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* non-ascii */ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00 }; /** * 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. * * The right angle bracket (>) may be represented using the string ">", * and must, for compatibility, be escaped using ">" or a character * reference when it appears in the string "]]>" in content, when that * string is not marking the end of a CDATA section. * * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*) */ void xmlParseCharData(xmlParserCtxtPtr ctxt, int cdata) { const xmlChar *in; int nbchar = 0; int line = ctxt->input->line; int col = ctxt->input->col; int ccol; SHRINK; GROW; /* * Accelerated common case where input don't need to be * modified before passing it to the handler. */ if (!cdata) { in = ctxt->input->cur; do { get_more_space: while (*in == 0x20) { in++; ctxt->input->col++; } if (*in == 0xA) { do { ctxt->input->line++; ctxt->input->col = 1; in++; } while (*in == 0xA); goto get_more_space; } if (*in == '<') { nbchar = in - ctxt->input->cur; if (nbchar > 0) { const xmlChar *tmp = ctxt->input->cur; ctxt->input->cur = in; if ((ctxt->sax != NULL) && (ctxt->sax->ignorableWhitespace != ctxt->sax->characters)) { if (areBlanks(ctxt, tmp, nbchar, 1)) { if (ctxt->sax->ignorableWhitespace != NULL) ctxt->sax->ignorableWhitespace(ctxt->userData, tmp, nbchar); } else { if (ctxt->sax->characters != NULL) ctxt->sax->characters(ctxt->userData, tmp, nbchar); if (*ctxt->space == -1) *ctxt->space = -2; } } else if ((ctxt->sax != NULL) && (ctxt->sax->characters != NULL)) { ctxt->sax->characters(ctxt->userData, tmp, nbchar); } } return; } get_more: ccol = ctxt->input->col; while (test_char_data[*in]) { in++; ccol++; } ctxt->input->col = ccol; if (*in == 0xA) { do { ctxt->input->line++; ctxt->input->col = 1; in++; } while (*in == 0xA); goto get_more; } if (*in == ']') { if ((in[1] == ']') && (in[2] == '>')) { xmlFatalErr(ctxt, XML_ERR_MISPLACED_CDATA_END, NULL); ctxt->input->cur = in; return; } in++; ctxt->input->col++; goto get_more; } nbchar = in - ctxt->input->cur; if (nbchar > 0) { if ((ctxt->sax != NULL) && (ctxt->sax->ignorableWhitespace != ctxt->sax->characters) && (IS_BLANK_CH(*ctxt->input->cur))) { const xmlChar *tmp = ctxt->input->cur; ctxt->input->cur = in; if (areBlanks(ctxt, tmp, nbchar, 0)) { if (ctxt->sax->ignorableWhitespace != NULL) ctxt->sax->ignorableWhitespace(ctxt->userData, tmp, nbchar); } else { if (ctxt->sax->characters != NULL) ctxt->sax->characters(ctxt->userData, tmp, nbchar); if (*ctxt->space == -1) *ctxt->space = -2; } line = ctxt->input->line; col = ctxt->input->col; } else if (ctxt->sax != NULL) { if (ctxt->sax->characters != NULL) ctxt->sax->characters(ctxt->userData, ctxt->input->cur, nbchar); line = ctxt->input->line; col = ctxt->input->col; } } ctxt->input->cur = in; if (*in == 0xD) { in++; if (*in == 0xA) { ctxt->input->cur = in; in++; ctxt->input->line++; ctxt->input->col = 1; continue; /* while */ } in--; } if (*in == '<') { return; } if (*in == '&') { return; } SHRINK; GROW; in = ctxt->input->cur; } while (((*in >= 0x20) && (*in <= 0x7F)) || (*in == 0x09)); nbchar = 0; } ctxt->input->line = line; ctxt->input->col = col; xmlParseCharDataComplex(ctxt, cdata); } /** * xmlParseCharDataComplex: * @ctxt: an XML parser context * @cdata: int indicating whether we are within a CDATA section * * parse a CharData section.this is the fallback function * of xmlParseCharData() when the parsing requires handling * of non-ASCII characters. */ void xmlParseCharDataComplex(xmlParserCtxtPtr ctxt, int cdata) { xmlChar buf[XML_PARSER_BIG_BUFFER_SIZE + 5]; int nbchar = 0; int cur, l; int count = 0; SHRINK; GROW; cur = CUR_CHAR(l); while ((cur != '<') && /* checked */ (cur != '&') && (IS_CHAR(cur))) /* test also done in xmlCurrentChar() */ { if ((cur == ']') && (NXT(1) == ']') && (NXT(2) == '>')) { if (cdata) break; else { xmlFatalErr(ctxt, XML_ERR_MISPLACED_CDATA_END, NULL); } } COPY_BUF(l,buf,nbchar,cur); if (nbchar >= XML_PARSER_BIG_BUFFER_SIZE) { buf[nbchar] = 0; /* * OK the segment is to be consumed as chars. */ if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) { if (areBlanks(ctxt, buf, nbchar, 0)) { 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); if ((ctxt->sax->characters != ctxt->sax->ignorableWhitespace) && (*ctxt->space == -1)) *ctxt->space = -2; } } nbchar = 0; } count++; if (count > 50) { GROW; count = 0; } NEXTL(l); cur = CUR_CHAR(l); } if (nbchar != 0) { buf[nbchar] = 0; /* * OK the segment is to be consumed as chars. */ if ((ctxt->sax != NULL) && (!ctxt->disableSAX)) { if (areBlanks(ctxt, buf, nbchar, 0)) { 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); if ((ctxt->sax->characters != ctxt->sax->ignorableWhitespace) && (*ctxt->space == -1)) *ctxt->space = -2; } } } if ((cur != 0) && (!IS_CHAR(cur))) { /* Generate the error and skip the offending character */ xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, "PCDATA invalid Char value %d\n", cur); NEXTL(l); } } /** * xmlParseExternalID: * @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] interact 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. */ xmlChar * xmlParseExternalID(xmlParserCtxtPtr ctxt, xmlChar **publicID, int strict) { xmlChar *URI = NULL; SHRINK; *publicID = NULL; if (CMP6(CUR_PTR, 'S', 'Y', 'S', 'T', 'E', 'M')) { SKIP(6); if (!IS_BLANK_CH(CUR)) { xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, "Space required after 'SYSTEM'\n"); } SKIP_BLANKS; URI = xmlParseSystemLiteral(ctxt); if (URI == NULL) { xmlFatalErr(ctxt, XML_ERR_URI_REQUIRED, NULL); } } else if (CMP6(CUR_PTR, 'P', 'U', 'B', 'L', 'I', 'C')) { SKIP(6); if (!IS_BLANK_CH(CUR)) { xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, "Space required after 'PUBLIC'\n"); } SKIP_BLANKS; *publicID = xmlParsePubidLiteral(ctxt); if (*publicID == NULL) { xmlFatalErr(ctxt, XML_ERR_PUBID_REQUIRED, NULL); } if (strict) { /* * We don't handle [83] so "S SystemLiteral" is required. */ if (!IS_BLANK_CH(CUR)) { xmlFatalErrMsg(ctxt, XML_ERR_SPACE_REQUIRED, "Space required after the Public Identifier\n"); } } 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_CH(*ptr)) return(NULL); while (IS_BLANK_CH(*ptr)) ptr++; /* TODO: dangerous, fix ! */ if ((*ptr != '\'') && (*ptr != '"')) return(NULL); } SKIP_BLANKS; URI = xmlParseSystemLiteral(ctxt); if (URI == NULL) { xmlFatalErr(ctxt, XML_ERR_URI_REQUIRED, NULL); } } return(URI); } /** * xmlParseCommentComplex: * @ctxt: an XML parser context * @buf: the already parsed part of the buffer * @len: number of bytes filles in the buffer * @size: allocated size of the buffer * * Skip an XML (SGML) comment * The spec says that "For compatibility, the string "--" (double-hyphen) * must not occur within comments. " * This is the slow routine in case the accelerator for ascii didn't work * * [15] Comment ::= '' */ static void xmlParseCommentComplex(xmlParserCtxtPtr ctxt, xmlChar *buf, int len, int size) { int q, ql; int r, rl; int cur, l; xmlParserInputPtr input = ctxt->input; int count = 0; if (buf == NULL) { len = 0; size = XML_PARSER_BUFFER_SIZE; buf = (xmlChar *) xmlMallocAtomic(size * sizeof(xmlChar)); if (buf == NULL) { xmlErrMemory(ctxt, NULL); return; } } GROW; /* Assure there's enough input data */ q = CUR_CHAR(ql); if (q == 0) goto not_terminated; if (!IS_CHAR(q)) { xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, "xmlParseComment: invalid xmlChar value %d\n", q); xmlFree (buf); return; } NEXTL(ql); r = CUR_CHAR(rl); if (r == 0) goto not_terminated; if (!IS_CHAR(r)) { xmlFatalErrMsgInt(ctxt, XML_ERR_INVALID_CHAR, "xmlParseComment: invalid xmlChar value %d\n", q); xmlFree (buf); return; } NEXTL(rl); cur = CUR_CHAR(l); if (cur == 0) goto not_terminated; while (IS_CHAR(cur) && /* checked */ ((cur != '>') || (r != '-') || (q != '-'))) { if ((r == '-') && (q == '-')) { xmlFatalErr(ctxt, XML_ERR_HYPHEN_IN_COMMENT, NULL); } if (len + 5 >= size) { xmlChar *new_buf; size *= 2; new_buf = (xmlChar *) xmlRealloc(buf, size * sizeof(xmlChar)); if (new_buf == NULL) { xmlFree (buf); xmlErrMemory(ctxt, NULL); return; } buf = new_buf; } COPY_BUF(ql,buf,len,q); q = r; ql = rl; r = cur; rl = l; count++; if (count > 50) { GROW; count = 0; } NEXTL(l); cur = CUR_CHAR(l); if (cur == 0) { SHRINK; GROW; cur = CUR_CHAR(l); } } buf[len] = 0; if (cur == 0) { xmlFatalErrMsgStr(ctxt, XML_ERR_COMMENT_NOT_FINISHED, "Comment not terminated \n * The spec says that "For compatibility, the string "--" (double-hyphen) * must not occur within comments. " * * [15] Comment ::= '' */ void xmlParseComment(xmlParserCtxtPtr ctxt) { xmlChar *buf = NULL; int size = XML_PARSER_BUFFER_SIZE; int len = 0; xmlParserInputState state; const xmlChar *in; int nbchar = 0, ccol; /* * Check that there is a comment right here. */ if ((RAW != '<') || (NXT(1) != '!') || (NXT(2) != '-') || (NXT(3) != '-')) return; state = ctxt->instate; ctxt->instate = XML_PARSER_COMMENT; SKIP(4); SHRINK; GROW; /* * Accelerated common case where input don't need to be * modified before passing it to the handler. */ in = ctxt->input->cur; do { if (*in == 0xA) { do { ctxt->input->line++; ctxt->input->col = 1; in++; } while (*in == 0xA); } get_more: ccol = ctxt->input->col; while (((*in > '-') && (*in <= 0x7F)) || ((*in >= 0x20) && (*in < '-')) || (*in == 0x09)) { in++; ccol++; } ctxt->input->col = ccol; if (*in == 0xA) { do { ctxt->input->line++; ctxt->input->col = 1; in++; } while (*in == 0xA); goto get_more; } nbchar = in - ctxt->input->cur; /* * save current set of data */ if (nbchar > 0) { if ((ctxt->sax != NULL) && (ctxt->sax->comment != NULL)) { if (buf == NULL) { if ((*in == '-') && (in[1] == '-')) size = nbchar + 1; else size = XML_PARSER_BUFFER_SIZE + nbchar; buf = (xmlChar *) xmlMallocAtomic(size * sizeof(xmlChar)); if (buf == NULL) { xmlErrMemory(ctxt, NULL); ctxt->instate = state; return; } len = 0; } else if (len + nbchar + 1 >= size) { xmlChar *new_buf; size += len + nbchar + XML_PARSER_BUFFER_SIZE; new_buf = (xmlChar *) xmlRealloc(buf, size * sizeof(xmlChar)); if (new_buf == NULL) { xmlFree (buf); xmlErrMemory(ctxt, NULL); ctxt->instate = state; return; } buf = new_buf; } memcpy(&buf[len], ctxt->input->cur, nbchar); len += nbchar; buf[len] = 0; } } ctxt->input->cur = in; if (*in == 0xA) { in++; ctxt->input->line++; ctxt->input->col = 1; } if (*in == 0xD) { in++; if (*in == 0xA) { ctxt->input->cur = in; in++; ctxt->input->line++; ctxt->input->col = 1; continue; /* while */ } in--; } SHRINK; GROW; in = ctxt->input->cur; if (*in == '-') { if (in[1] == '-') { if (in[2] == '>') { SKIP(3); if ((ctxt->sax != NULL) && (ctxt->sax->comment != NULL) && (!ctxt->disableSAX)) { if (buf != NULL) ctxt->sax->comment(ctxt->userData, buf); else ctxt->sax->comment(ctxt->userData, BAD_CAST ""); } if (buf != NULL) xmlFree(buf); ctxt->instate = state; return; } if (buf != NULL) xmlFatalErrMsgStr(ctxt, XML_ERR_COMMENT_NOT_FINISHED, "Comment not terminated \n