ixmlparser.c

电驴下载工具eMule0.47aVeryCD的源代码,可作分析测试也可用于P2P软件的开发研究.

    return -1;
}

/*==============================================================================*
*   Parser_copyToken	
*       copy string in src into xml parser token buffer
*		Internal to parser only.	
*
*===============================================================================*/
static int
Parser_copyToken( IN Parser * xmlParser,
                  IN char *src,
                  IN int len )
{
    int i,
      c,
      cl;
    char *psrc,
     *pend;
    utf8char uch;

    if( !src || len <= 0 ) {
        return IXML_FAILED;
    }

    psrc = src;
    pend = src + len;

    while( psrc < pend ) {
        if( ( c = Parser_getChar( psrc, &cl ) ) <= 0 ) {
            return IXML_FAILED;
        }

        if( cl == 1 ) {
            Parser_appendTokBufChar( xmlParser, ( char )c );
            psrc++;
        } else {

            i = Parser_intToUTF8( c, uch );
            if( i == 0 ) {
                return IXML_FAILED;
            }

            Parser_appendTokBufStr( xmlParser, uch );
            psrc += cl;
        }
    }

    if( psrc > pend ) {
        return IXML_FAILED;
    } else {
        return IXML_SUCCESS;    // success
    }

}

/*==============================================================================*
*
*   Parser_skipString
*       Skips all characters in the string until it finds the skip key.
*       Then it skips the skip key and returns.
*       Internal to parser only
*
*===============================================================================*/
static int
Parser_skipString( INOUT char **pstrSrc,
                   IN const char *strSkipKey )
{
    if( !( *pstrSrc ) || !strSkipKey ) {
        return IXML_FAILED;
    }

    while( ( **pstrSrc )
           && strncmp( *pstrSrc, strSkipKey,
                       strlen( strSkipKey ) ) != 0 ) {
        ( *pstrSrc )++;
    }

    if( **pstrSrc == '\0' ) {
        return IXML_SYNTAX_ERR;
    }
    *pstrSrc = *pstrSrc + strlen( strSkipKey );

    return IXML_SUCCESS;        //success
}

/*==============================================================================*
*
* Function:	
* Returns:	
*			
*
*===============================================================================*/
static int
Parser_skipPI( INOUT char **pSrc )
{
    char *pEnd = NULL;

    assert( *pSrc );
    if( *pSrc == NULL ) {
        return IXML_FAILED;
    }

    if( ( strncasecmp( *pSrc, ( char * )XMLDECL, strlen( XMLDECL ) ) == 0 ) || ( strncasecmp( *pSrc, ( char * )XMLDECL2, strlen( XMLDECL2 ) ) == 0 ) ) {    // not allowed
        return IXML_SYNTAX_ERR;
    }

    if( strncasecmp( *pSrc, ( char * )BEGIN_PI, strlen( BEGIN_PI ) ) == 0 ) {
        pEnd = strstr( *pSrc, END_PI );
        if( ( pEnd != NULL ) && ( pEnd != *pSrc ) ) {
            *pSrc = pEnd + strlen( BEGIN_PI );
        } else {
            return IXML_SYNTAX_ERR;
        }
    }

    return IXML_SUCCESS;
}

/*==============================================================================*
*   Parser_skipXMLDecl:	
*       skips XML declarations.
*       Internal only to parser.			
*
*===============================================================================*/
static int
Parser_skipXMLDecl( INOUT Parser * xmlParser )
{
    int rc = IXML_FAILED;

    assert( xmlParser );
    if( xmlParser == NULL ) {
        return rc;
    }

    rc = Parser_skipString( &( xmlParser->curPtr ), END_PI );
    Parser_skipWhiteSpaces( xmlParser );
    return rc;

}

/*==============================================================================*
*   Parser_skipProlog
*       skip prolog
*       Internal to parser only.			
*
*===============================================================================*/
static int
Parser_skipProlog( INOUT Parser * xmlParser )
{
    int rc = IXML_SUCCESS;

    assert( xmlParser != NULL );
    if( xmlParser == NULL ) {
        return IXML_FAILED;
    }

    Parser_skipWhiteSpaces( xmlParser );

    if( strncmp( xmlParser->curPtr, ( char * )XMLDECL, strlen( XMLDECL ) ) == 0 ) { // <?xml
        rc = Parser_skipXMLDecl( xmlParser );
        if( rc != IXML_SUCCESS ) {
            return rc;
        }
    }

    rc = Parser_skipMisc( xmlParser );
    if( ( rc == IXML_SUCCESS ) && strncmp( xmlParser->curPtr, ( char * )BEGIN_DOCTYPE, strlen( BEGIN_DOCTYPE ) ) == 0 ) {   // <! DOCTYPE
        xmlParser->curPtr++;
        rc = Parser_skipDocType( &( xmlParser->curPtr ) );
    }

    if( rc == IXML_SUCCESS ) {
        rc = Parser_skipMisc( xmlParser );
    }

    return rc;
}

/*==============================================================================*
*
* Function:
* Returns:
*       Skips all characters in the string until it finds the skip key.
*       Then it skips the skip key and returns.
*
*===============================================================================*/
static int
Parser_skipComment( INOUT char **pstrSrc )
{
    char *pStrFound = NULL;

    assert( ( *pstrSrc ) != NULL );
    if( *pstrSrc == NULL ) {
        return IXML_FAILED;
    }

    pStrFound = strstr( *pstrSrc, END_COMMENT );
    if( ( pStrFound != NULL ) && ( pStrFound != *pstrSrc ) &&
        ( *( pStrFound - 1 ) != '-' ) ) {
        *pstrSrc = pStrFound + strlen( END_COMMENT );
    } else {
        return IXML_SYNTAX_ERR;
    }

    return IXML_SUCCESS;
}

/*==============================================================================*
*   Parser_skipDocType
*       skips document type declaration
*
*===============================================================================*/
static int
Parser_skipDocType( INOUT char **pstr )
{
    char *pCur = *pstr;
    char *pNext = NULL;         // default there is no nested <
    int num = 1;

    assert( ( *pstr ) != NULL );
    if( *pstr == NULL ) {
        return IXML_FAILED;
    }

    while( ( pCur != NULL ) && ( num != 0 ) && ( *pCur != 0 ) ) {
        if( *pCur == '<' ) {
            num++;
        } else if( *pCur == '>' ) {
            num--;
        } else if( *pCur == '"' ) {
            pNext = strchr( pCur + 1, '"' );
            if( pNext == NULL ) {
                return IXML_SYNTAX_ERR;
            }

            pCur = pNext;
        }

        pCur++;
    }

    if( num == 0 ) {
        *pstr = pCur;
        return IXML_SUCCESS;
    } else {
        return IXML_SYNTAX_ERR;
    }
}

/*==============================================================================*
*
*   Parser_skipMisc:	
*       skip comment, PI and white space 
*			
*
*===============================================================================*/
static int
Parser_skipMisc( IN Parser * xmlParser )
{
    int rc = IXML_SUCCESS;
    int done = FALSE;

    while( ( done == FALSE ) && ( rc == IXML_SUCCESS ) ) {
        if( strncasecmp( xmlParser->curPtr, ( char * )BEGIN_COMMENT, strlen( BEGIN_COMMENT ) ) == 0 ) { // <!--
            rc = Parser_skipComment( &( xmlParser->curPtr ) );

        } else if( ( strncasecmp( xmlParser->curPtr, ( char * )XMLDECL, strlen( XMLDECL ) ) == 0 ) || ( strncasecmp( xmlParser->curPtr, ( char * )XMLDECL2, strlen( XMLDECL2 ) ) == 0 ) ) { // <?xml or <?xml?
            rc = IXML_SYNTAX_ERR;
        } else if( strncasecmp( xmlParser->curPtr, ( char * )BEGIN_PI, strlen( BEGIN_PI ) ) == 0 ) {    // <?
            rc = Parser_skipString( &( xmlParser->curPtr ), END_PI );
        } else {
            done = TRUE;
        }

        Parser_skipWhiteSpaces( xmlParser );
    }

    return rc;
}

/*==============================================================================*
*
*   Parser_getNextToken
*       return the length of next token in tokenBuff
*			
*
*===============================================================================*/
static int
Parser_getNextToken( IN Parser * xmlParser )
{
    int tokenLength = 0;
    int temp,
      tlen;
    int rc;

    Parser_clearTokenBuf( xmlParser );

    if( *( xmlParser->curPtr ) == '\0' ) {
        return 0;
    }
    // skip XML instructions
    rc = Parser_skipMisc( xmlParser );
    if( rc != IXML_SUCCESS ) {
        return 0;
    }
    // Attribute value logic must come first, since all text untokenized until end-quote
    if( *( xmlParser->curPtr ) == QUOTE ) {
        tokenLength = 1;
    } else if( *( xmlParser->curPtr ) == SINGLEQUOTE ) {
        tokenLength = 1;
    } else if( *( xmlParser->curPtr ) == LESSTHAN ) {   // Check for start tags
        temp = Parser_UTF8ToInt( xmlParser->curPtr + 1, &tlen );
        if( temp == '/' ) {
            tokenLength = 2;    // token is '</' end tag
        } else if( Parser_isNameChar( temp, FALSE ) == TRUE ) {
            tokenLength = 1;    // '<' found, so return '<' token
        } else {
            return 0;           //error
        }
    } else if( *( xmlParser->curPtr ) == EQUALS ) { // Check for '=' token, return it as a token
        tokenLength = 1;
    } else if( *( xmlParser->curPtr ) == SLASH ) {
        if( *( xmlParser->curPtr + 1 ) == GREATERTHAN ) {   // token '/>' found
            tokenLength = 2;
            xmlParser->savePtr = xmlParser->curPtr; // fix
        }
    } else if( *( xmlParser->curPtr ) == GREATERTHAN ) {    // > found, so return it as a token
        tokenLength = 1;
    } else if( Parser_isNameChar( Parser_UTF8ToInt( xmlParser->curPtr, &tlen ), FALSE ) ) { // Check for name tokens, name found, so find out how long it is
        int iIndex = tlen;

        while( Parser_isNameChar
               ( Parser_UTF8ToInt( xmlParser->curPtr + iIndex, &tlen ),
                 TRUE ) ) {
            iIndex += tlen;
        }
        tokenLength = iIndex;
    } else {
        return 0;
    }

    // Copy the token to the return string
    if( Parser_copyToken( xmlParser, xmlParser->curPtr, tokenLength ) !=
        IXML_SUCCESS ) {
        return 0;
    }

    xmlParser->curPtr += tokenLength;
    return tokenLength;
}

/*==============================================================================*
*
*   Parser_getNameSpace	
*       return the namespce as defined as prefix.
*       Internal to parser only			
*
*===============================================================================*/
static char *
Parser_getNameSpace( IN Parser * xmlParser,
                     IN char *prefix )
{
    IXML_ElementStack *pCur;
    IXML_NamespaceURI *pNsUri;

    pCur = xmlParser->pCurElement;
    if( strcmp( pCur->prefix, prefix ) != 0 ) {
        pNsUri = pCur->pNsURI;
        while( pNsUri != NULL ) {
            if( strcmp( pNsUri->prefix, prefix ) == 0 ) {
                return pNsUri->nsURI;
            }
            pNsUri = pNsUri->nextNsURI;
        }
    } else {
        return pCur->namespaceUri;
    }

    return NULL;

}

/*==============================================================================*
*
*   Parser_addNamespace
*       Add a namespace definition
*       Internal to parser only			
*
*===============================================================================*/