xmlparse.c

GNU/Linux程序开发指南书的所有例子源代码

void XML_SetProcessingInstructionHandler(XML_Parser parser,
					 XML_ProcessingInstructionHandler handler)
{
  processingInstructionHandler = handler;
}

void XML_SetDefaultHandler(XML_Parser parser,
			   XML_DefaultHandler handler)
{
  defaultHandler = handler;
}

void XML_SetUnparsedEntityDeclHandler(XML_Parser parser,
				      XML_UnparsedEntityDeclHandler handler)
{
  unparsedEntityDeclHandler = handler;
}

void XML_SetNotationDeclHandler(XML_Parser parser,
				XML_NotationDeclHandler handler)
{
  notationDeclHandler = handler;
}

void XML_SetExternalEntityRefHandler(XML_Parser parser,
				     XML_ExternalEntityRefHandler handler)
{
  externalEntityRefHandler = handler;
}

void XML_SetUnknownEncodingHandler(XML_Parser parser,
				   XML_UnknownEncodingHandler handler,
				   void *data)
{
  unknownEncodingHandler = handler;
  unknownEncodingHandlerData = data;
}

int XML_Parse(XML_Parser parser, const char *s, int len, int isFinal)
{
  if (len == 0) {
    if (!isFinal)
      return 1;
    errorCode = processor(parser, bufferPtr, parseEndPtr = bufferEnd, 0);
    if (errorCode == XML_ERROR_NONE)
      return 1;
    eventEndPtr = eventPtr;
    return 0;
  }
  else if (bufferPtr == bufferEnd) {
    const char *end;
    int nLeftOver;
    parseEndByteIndex += len;
    positionPtr = s;
    if (isFinal) {
      errorCode = processor(parser, s, parseEndPtr = s + len, 0);
      if (errorCode == XML_ERROR_NONE)
	return 1;
      eventEndPtr = eventPtr;
      return 0;
    }
    errorCode = processor(parser, s, parseEndPtr = s + len, &end);
    if (errorCode != XML_ERROR_NONE) {
      eventEndPtr = eventPtr;
      return 0;
    }
    XmlUpdatePosition(encoding, positionPtr, end, &position);
    nLeftOver = s + len - end;
    if (nLeftOver) {
      if (buffer == 0 || nLeftOver > bufferLim - buffer) {
	/* FIXME avoid integer overflow */
	buffer = buffer == 0 ? malloc(len * 2) : realloc(buffer, len * 2);
	if (!buffer) {
	  errorCode = XML_ERROR_NO_MEMORY;
	  eventPtr = eventEndPtr = 0;
	  return 0;
	}
	bufferLim = buffer + len * 2;
      }
      memcpy(buffer, end, nLeftOver);
      bufferPtr = buffer;
      bufferEnd = buffer + nLeftOver;
    }
    return 1;
  }
  else {
    memcpy(XML_GetBuffer(parser, len), s, len);
    return XML_ParseBuffer(parser, len, isFinal);
  }
}

int XML_ParseBuffer(XML_Parser parser, int len, int isFinal)
{
  const char *start = bufferPtr;
  positionPtr = start;
  bufferEnd += len;
  parseEndByteIndex += len;
  errorCode = processor(parser, start, parseEndPtr = bufferEnd,
			isFinal ? (const char **)0 : &bufferPtr);
  if (errorCode == XML_ERROR_NONE) {
    if (!isFinal)
      XmlUpdatePosition(encoding, positionPtr, bufferPtr, &position);
    return 1;
  }
  else {
    eventEndPtr = eventPtr;
    return 0;
  }
}

void *XML_GetBuffer(XML_Parser parser, int len)
{
  if (len > bufferLim - bufferEnd) {
    /* FIXME avoid integer overflow */
    int neededSize = len + (bufferEnd - bufferPtr);
    if (neededSize  <= bufferLim - buffer) {
      memmove(buffer, bufferPtr, bufferEnd - bufferPtr);
      bufferEnd = buffer + (bufferEnd - bufferPtr);
      bufferPtr = buffer;
    }
    else {
      char *newBuf;
      int bufferSize = bufferLim - bufferPtr;
      if (bufferSize == 0)
	bufferSize = INIT_BUFFER_SIZE;
      do {
	bufferSize *= 2;
      } while (bufferSize < neededSize);
      newBuf = malloc(bufferSize);
      if (newBuf == 0) {
	errorCode = XML_ERROR_NO_MEMORY;
	return 0;
      }
      bufferLim = newBuf + bufferSize;
      if (bufferPtr) {
	memcpy(newBuf, bufferPtr, bufferEnd - bufferPtr);
	free(buffer);
      }
      bufferEnd = newBuf + (bufferEnd - bufferPtr);
      bufferPtr = buffer = newBuf;
    }
  }
  return bufferEnd;
}

enum XML_Error XML_GetErrorCode(XML_Parser parser)
{
  return errorCode;
}

long XML_GetCurrentByteIndex(XML_Parser parser)
{
  if (eventPtr)
    return parseEndByteIndex - (parseEndPtr - eventPtr);
  return -1;
}

int XML_GetCurrentLineNumber(XML_Parser parser)
{
  if (eventPtr) {
    XmlUpdatePosition(encoding, positionPtr, eventPtr, &position);
    positionPtr = eventPtr;
  }
  return position.lineNumber + 1;
}

int XML_GetCurrentColumnNumber(XML_Parser parser)
{
  if (eventPtr) {
    XmlUpdatePosition(encoding, positionPtr, eventPtr, &position);
    positionPtr = eventPtr;
  }
  return position.columnNumber;
}

void XML_DefaultCurrent(XML_Parser parser)
{
  if (defaultHandler)
    reportDefault(parser, encoding, eventPtr, eventEndPtr);
}

const XML_LChar *XML_ErrorString(int code)
{
  static const XML_LChar *message[] = {
    0,
    XML_T("out of memory"),
    XML_T("syntax error"),
    XML_T("no element found"),
    XML_T("not well-formed"),
    XML_T("unclosed token"),
    XML_T("unclosed token"),
    XML_T("mismatched tag"),
    XML_T("duplicate attribute"),
    XML_T("junk after document element"),
    XML_T("illegal parameter entity reference"),
    XML_T("undefined entity"),
    XML_T("recursive entity reference"),
    XML_T("asynchronous entity"),
    XML_T("reference to invalid character number"),
    XML_T("reference to binary entity"),
    XML_T("reference to external entity in attribute"),
    XML_T("xml processing instruction not at start of external entity"),
    XML_T("unknown encoding"),
    XML_T("encoding specified in XML declaration is incorrect"),
    XML_T("unclosed CDATA section"),
    XML_T("error in processing external entity reference")
  };
  if (code > 0 && code < sizeof(message)/sizeof(message[0]))
    return message[code];
  return 0;
}

static
enum XML_Error contentProcessor(XML_Parser parser,
				const char *start,
				const char *end,
				const char **endPtr)
{
  return doContent(parser, 0, encoding, start, end, endPtr);
}

static
enum XML_Error externalEntityInitProcessor(XML_Parser parser,
					   const char *start,
					   const char *end,
					   const char **endPtr)
{
  enum XML_Error result = initializeEncoding(parser);
  if (result != XML_ERROR_NONE)
    return result;
  processor = externalEntityInitProcessor2;
  return externalEntityInitProcessor2(parser, start, end, endPtr);
}

static
enum XML_Error externalEntityInitProcessor2(XML_Parser parser,
					    const char *start,
					    const char *end,
					    const char **endPtr)
{
  const char *next;
  int tok = XmlContentTok(encoding, start, end, &next);
  switch (tok) {
  case XML_TOK_BOM:
    start = next;
    break;
  case XML_TOK_PARTIAL:
    if (endPtr) {
      *endPtr = start;
      return XML_ERROR_NONE;
    }
    eventPtr = start;
    return XML_ERROR_UNCLOSED_TOKEN;
  case XML_TOK_PARTIAL_CHAR:
    if (endPtr) {
      *endPtr = start;
      return XML_ERROR_NONE;
    }
    eventPtr = start;
    return XML_ERROR_PARTIAL_CHAR;
  }
  processor = externalEntityInitProcessor3;
  return externalEntityInitProcessor3(parser, start, end, endPtr);
}

static
enum XML_Error externalEntityInitProcessor3(XML_Parser parser,
					    const char *start,
					    const char *end,
					    const char **endPtr)
{
  const char *next;
  int tok = XmlContentTok(encoding, start, end, &next);
  switch (tok) {
  case XML_TOK_XML_DECL:
    {
      enum XML_Error result = processXmlDecl(parser, 1, start, next);
      if (result != XML_ERROR_NONE)
	return result;
      start = next;
    }
    break;
  case XML_TOK_PARTIAL:
    if (endPtr) {
      *endPtr = start;
      return XML_ERROR_NONE;
    }
    eventPtr = start;
    return XML_ERROR_UNCLOSED_TOKEN;
  case XML_TOK_PARTIAL_CHAR:
    if (endPtr) {
      *endPtr = start;
      return XML_ERROR_NONE;
    }
    eventPtr = start;
    return XML_ERROR_PARTIAL_CHAR;
  }
  processor = externalEntityContentProcessor;
  tagLevel = 1;
  return doContent(parser, 1, encoding, start, end, endPtr);
}

static
enum XML_Error externalEntityContentProcessor(XML_Parser parser,
					      const char *start,
					      const char *end,
					      const char **endPtr)
{
  return doContent(parser, 1, encoding, start, end, endPtr);
}

static enum XML_Error
doContent(XML_Parser parser,
	  int startTagLevel,
	  const ENCODING *enc,
	  const char *s,
	  const char *end,
	  const char **nextPtr)
{
  const ENCODING *internalEnc = XmlGetInternalEncoding();
  const char *dummy;
  const char **eventPP;
  const char **eventEndPP;
  if (enc == encoding) {
    eventPP = &eventPtr;
    *eventPP = s;
    eventEndPP = &eventEndPtr;
  }
  else
    eventPP = eventEndPP = &dummy;
  for (;;) {
    const char *next;
    int tok = XmlContentTok(enc, s, end, &next);
    *eventEndPP = next;
    switch (tok) {
    case XML_TOK_TRAILING_CR:
      if (nextPtr) {
	*nextPtr = s;
	return XML_ERROR_NONE;
      }
      *eventEndPP = end;
      if (characterDataHandler) {
	XML_Char c = XML_T('\n');
	characterDataHandler(handlerArg, &c, 1);
      }
      else if (defaultHandler)
	reportDefault(parser, enc, s, end);
      if (startTagLevel == 0)
	return XML_ERROR_NO_ELEMENTS;
      if (tagLevel != startTagLevel)
	return XML_ERROR_ASYNC_ENTITY;
      return XML_ERROR_NONE;
    case XML_TOK_NONE:
      if (nextPtr) {
	*nextPtr = s;
	return XML_ERROR_NONE;
      }
      if (startTagLevel > 0) {
	if (tagLevel != startTagLevel)
	  return XML_ERROR_ASYNC_ENTITY;
	return XML_ERROR_NONE;
      }
      return XML_ERROR_NO_ELEMENTS;
    case XML_TOK_INVALID:
      *eventPP = next;
      return XML_ERROR_INVALID_TOKEN;
    case XML_TOK_PARTIAL:
      if (nextPtr) {
	*nextPtr = s;
	return XML_ERROR_NONE;
      }
      return XML_ERROR_UNCLOSED_TOKEN;
    case XML_TOK_PARTIAL_CHAR:
      if (nextPtr) {
	*nextPtr = s;
	return XML_ERROR_NONE;
      }
      return XML_ERROR_PARTIAL_CHAR;
    case XML_TOK_ENTITY_REF:
      {
	const XML_Char *name;
	ENTITY *entity;
	XML_Char ch = XmlPredefinedEntityName(enc,
					      s + enc->minBytesPerChar,
					      next - enc->minBytesPerChar);
	if (ch) {
	  if (characterDataHandler)
	    characterDataHandler(handlerArg, &ch, 1);
	  else if (defaultHandler)
	    reportDefault(parser, enc, s, next);
	  break;
	}
	name = poolStoreString(&dtd.pool, enc,
				s + enc->minBytesPerChar,
				next - enc->minBytesPerChar);
	if (!name)
	  return XML_ERROR_NO_MEMORY;
	entity = (ENTITY *)lookup(&dtd.generalEntities, name, 0);
	poolDiscard(&dtd.pool);
	if (!entity) {
	  if (dtd.complete || dtd.standalone)
	    return XML_ERROR_UNDEFINED_ENTITY;
	  if (defaultHandler)
	    reportDefault(parser, enc, s, next);
	  break;
	}
	if (entity->open)
	  return XML_ERROR_RECURSIVE_ENTITY_REF;
	if (entity->notation)
	  return XML_ERROR_BINARY_ENTITY_REF;
	if (entity) {
	  if (entity->textPtr) {
	    enum XML_Error result;
	    if (defaultHandler) {
	      reportDefault(parser, enc, s, next);
	      break;
	    }
	    /* Protect against the possibility that somebody sets
	       the defaultHandler from inside another handler. */
	    *eventEndPP = *eventPP;
	    entity->open = 1;
	    result = doContent(parser,
			       tagLevel,
			       internalEnc,
			       (char *)entity->textPtr,
			       (char *)(entity->textPtr + entity->textLen),
			       0);
	    entity->open = 0;
	    if (result)
	      return result;
	  }
	  else if (externalEntityRefHandler) {
	    const XML_Char *openEntityNames;
	    entity->open = 1;
	    openEntityNames = getOpenEntityNames(parser);
	    entity->open = 0;
	    if (!openEntityNames)
	      return XML_ERROR_NO_MEMORY;
	    if (!externalEntityRefHandler(parser, openEntityNames, dtd.base, entity->systemId, entity->publicId))
	      return XML_ERROR_EXTERNAL_ENTITY_HANDLING;
	    poolDiscard(&tempPool);
	  }
	  else if (defaultHandler)
	    reportDefault(parser, enc, s, next);
	}
	break;
      }
    case XML_TOK_START_TAG_WITH_ATTS:
      if (!startElementHandler) {
	enum XML_Error result = storeAtts(parser, enc, 0, s);
	if (result)
	  return result;
      }
      /* fall through */
    case XML_TOK_START_TAG_NO_ATTS:
      {
	TAG *tag;
	if (freeTagList) {
	  tag = freeTagList;
	  freeTagList = freeTagList->parent;
	}
	else {
	  tag = malloc(sizeof(TAG));
	  if (!tag)
	    return XML_ERROR_NO_MEMORY;
	  tag->buf = malloc(INIT_TAG_BUF_SIZE);
	  if (!tag->buf)
	    return XML_ERROR_NO_MEMORY;
	  tag->bufEnd = tag->buf + INIT_TAG_BUF_SIZE;
	}
	tag->parent = tagStack;
	tagStack = tag;
	tag->rawName = s + enc->minBytesPerChar;
	tag->rawNameLength = XmlNameLength(enc, tag->rawName);
	if (nextPtr) {
	  if (tag->rawNameLength > tag->bufEnd - tag->buf) {
	    int bufSize = tag->rawNameLength * 4;
	    bufSize = ROUND_UP(bufSize, sizeof(XML_Char));
	    tag->buf = realloc(tag->buf, bufSize);