parserinternals.c

libxml,在UNIX/LINUX下非常重要的一个库,为XML相关应用提供方便.目前上载的是最新版本,若要取得最新版本,请参考里面的readme.

    {
        char buffer[150];

	snprintf(buffer, 149, "Bytes: 0x%02X 0x%02X 0x%02X 0x%02X\n",
			ctxt->input->cur[0], ctxt->input->cur[1],
			ctxt->input->cur[2], ctxt->input->cur[3]);
	__xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
		     "Input is not proper UTF-8, indicate encoding !\n%s",
		     BAD_CAST buffer, NULL);
    }
    *len = 1;
    return ((int) *cur);
}

/**
 * xmlCopyCharMultiByte:
 * @out:  pointer to an array of xmlChar
 * @val:  the char value
 *
 * append the char value in the array 
 *
 * Returns the number of xmlChar written
 */
int
xmlCopyCharMultiByte(xmlChar *out, int val) {
    if (out == NULL) return(0);
    /*
     * We are supposed to handle UTF8, check it's valid
     * From rfc2044: encoding of the Unicode values on UTF-8:
     *
     * UCS-4 range (hex.)           UTF-8 octet sequence (binary)
     * 0000 0000-0000 007F   0xxxxxxx
     * 0000 0080-0000 07FF   110xxxxx 10xxxxxx
     * 0000 0800-0000 FFFF   1110xxxx 10xxxxxx 10xxxxxx 
     */
    if  (val >= 0x80) {
	xmlChar *savedout = out;
	int bits;
	if (val <   0x800) { *out++= (val >>  6) | 0xC0;  bits=  0; }
	else if (val < 0x10000) { *out++= (val >> 12) | 0xE0;  bits=  6;}
	else if (val < 0x110000)  { *out++= (val >> 18) | 0xF0;  bits=  12; }
	else {
	    xmlErrEncodingInt(NULL, XML_ERR_INVALID_CHAR,
		    "Internal error, xmlCopyCharMultiByte 0x%X out of bound\n",
			      val);
	    return(0);
	}
	for ( ; bits >= 0; bits-= 6)
	    *out++= ((val >> bits) & 0x3F) | 0x80 ;
	return (out - savedout);
    }
    *out = (xmlChar) val;
    return 1;
}

/**
 * xmlCopyChar:
 * @len:  Ignored, compatibility
 * @out:  pointer to an array of xmlChar
 * @val:  the char value
 *
 * append the char value in the array 
 *
 * Returns the number of xmlChar written
 */

int
xmlCopyChar(int len ATTRIBUTE_UNUSED, xmlChar *out, int val) {
    if (out == NULL) return(0);
    /* the len parameter is ignored */
    if  (val >= 0x80) {
	return(xmlCopyCharMultiByte (out, val));
    }
    *out = (xmlChar) val;
    return 1;
}

/************************************************************************
 *									*
 *		Commodity functions to switch encodings			*
 *									*
 ************************************************************************/

/**
 * xmlSwitchEncoding:
 * @ctxt:  the parser context
 * @enc:  the encoding value (number)
 *
 * change the input functions when discovering the character encoding
 * of a given entity.
 *
 * Returns 0 in case of success, -1 otherwise
 */
int
xmlSwitchEncoding(xmlParserCtxtPtr ctxt, xmlCharEncoding enc)
{
    xmlCharEncodingHandlerPtr handler;

    if (ctxt == NULL) return(-1);
    switch (enc) {
	case XML_CHAR_ENCODING_ERROR:
	    __xmlErrEncoding(ctxt, XML_ERR_UNKNOWN_ENCODING,
	                   "encoding unknown\n", NULL, NULL);
	    break;
	case XML_CHAR_ENCODING_NONE:
	    /* let's assume it's UTF-8 without the XML decl */
	    ctxt->charset = XML_CHAR_ENCODING_UTF8;
	    return(0);
	case XML_CHAR_ENCODING_UTF8:
	    /* default encoding, no conversion should be needed */
	    ctxt->charset = XML_CHAR_ENCODING_UTF8;

	    /*
	     * Errata on XML-1.0 June 20 2001
	     * Specific handling of the Byte Order Mark for
	     * UTF-8
	     */
	    if ((ctxt->input != NULL) &&
		(ctxt->input->cur[0] == 0xEF) &&
		(ctxt->input->cur[1] == 0xBB) &&
		(ctxt->input->cur[2] == 0xBF)) {
		ctxt->input->cur += 3;
	    }
	    return(0);
    case XML_CHAR_ENCODING_UTF16LE:
    case XML_CHAR_ENCODING_UTF16BE:
        /*The raw input characters are encoded
         *in UTF-16. As we expect this function
         *to be called after xmlCharEncInFunc, we expect
         *ctxt->input->cur to contain UTF-8 encoded characters.
         *So the raw UTF16 Byte Order Mark
         *has also been converted into
         *an UTF-8 BOM. Let's skip that BOM.
         */
        if ((ctxt->input != NULL) &&
            (ctxt->input->cur[0] == 0xEF) &&
            (ctxt->input->cur[1] == 0xBB) &&
            (ctxt->input->cur[2] == 0xBF)) {
            ctxt->input->cur += 3;
        }
	break ;
	default:
	    break;
    }
    handler = xmlGetCharEncodingHandler(enc);
    if (handler == NULL) {
	/*
	 * Default handlers.
	 */
	switch (enc) {
	    case XML_CHAR_ENCODING_ERROR:
		__xmlErrEncoding(ctxt, XML_ERR_UNKNOWN_ENCODING,
			       "encoding unknown\n", NULL, NULL);
		break;
	    case XML_CHAR_ENCODING_NONE:
		/* let's assume it's UTF-8 without the XML decl */
		ctxt->charset = XML_CHAR_ENCODING_UTF8;
		return(0);
	    case XML_CHAR_ENCODING_UTF8:
	    case XML_CHAR_ENCODING_ASCII:
		/* default encoding, no conversion should be needed */
		ctxt->charset = XML_CHAR_ENCODING_UTF8;
		return(0);
	    case XML_CHAR_ENCODING_UTF16LE:
		break;
	    case XML_CHAR_ENCODING_UTF16BE:
		break;
	    case XML_CHAR_ENCODING_UCS4LE:
		__xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
			       "encoding not supported %s\n",
			       BAD_CAST "USC4 little endian", NULL);
		break;
	    case XML_CHAR_ENCODING_UCS4BE:
		__xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
			       "encoding not supported %s\n",
			       BAD_CAST "USC4 big endian", NULL);
		break;
	    case XML_CHAR_ENCODING_EBCDIC:
		__xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
			       "encoding not supported %s\n",
			       BAD_CAST "EBCDIC", NULL);
		break;
	    case XML_CHAR_ENCODING_UCS4_2143:
		__xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
			       "encoding not supported %s\n",
			       BAD_CAST "UCS4 2143", NULL);
		break;
	    case XML_CHAR_ENCODING_UCS4_3412:
		__xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
			       "encoding not supported %s\n",
			       BAD_CAST "UCS4 3412", NULL);
		break;
	    case XML_CHAR_ENCODING_UCS2:
		__xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
			       "encoding not supported %s\n",
			       BAD_CAST "UCS2", NULL);
		break;
	    case XML_CHAR_ENCODING_8859_1:
	    case XML_CHAR_ENCODING_8859_2:
	    case XML_CHAR_ENCODING_8859_3:
	    case XML_CHAR_ENCODING_8859_4:
	    case XML_CHAR_ENCODING_8859_5:
	    case XML_CHAR_ENCODING_8859_6:
	    case XML_CHAR_ENCODING_8859_7:
	    case XML_CHAR_ENCODING_8859_8:
	    case XML_CHAR_ENCODING_8859_9:
		/*
		 * We used to keep the internal content in the
		 * document encoding however this turns being unmaintainable
		 * So xmlGetCharEncodingHandler() will return non-null
		 * values for this now.
		 */
		if ((ctxt->inputNr == 1) &&
		    (ctxt->encoding == NULL) &&
		    (ctxt->input->encoding != NULL)) {
		    ctxt->encoding = xmlStrdup(ctxt->input->encoding);
		}
		ctxt->charset = enc;
		return(0);
	    case XML_CHAR_ENCODING_2022_JP:
		__xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
			       "encoding not supported %s\n",
			       BAD_CAST "ISO-2022-JP", NULL);
		break;
	    case XML_CHAR_ENCODING_SHIFT_JIS:
		__xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
			       "encoding not supported %s\n",
			       BAD_CAST "Shift_JIS", NULL);
		break;
	    case XML_CHAR_ENCODING_EUC_JP:
		__xmlErrEncoding(ctxt, XML_ERR_UNSUPPORTED_ENCODING,
			       "encoding not supported %s\n",
			       BAD_CAST "EUC-JP", NULL);
		break;
	}
    }
    if (handler == NULL)
	return(-1);
    ctxt->charset = XML_CHAR_ENCODING_UTF8;
    return(xmlSwitchToEncoding(ctxt, handler));
}

/**
 * xmlSwitchInputEncoding:
 * @ctxt:  the parser context
 * @input:  the input stream
 * @handler:  the encoding handler
 *
 * change the input functions when discovering the character encoding
 * of a given entity.
 *
 * Returns 0 in case of success, -1 otherwise
 */
int
xmlSwitchInputEncoding(xmlParserCtxtPtr ctxt, xmlParserInputPtr input,
                       xmlCharEncodingHandlerPtr handler)
{
    int nbchars;

    if (handler == NULL)
        return (-1);
    if (input == NULL)
        return (-1);
    if (input->buf != NULL) {
        if (input->buf->encoder != NULL) {
            /*
             * Check in case the auto encoding detetection triggered
             * in already.
             */
            if (input->buf->encoder == handler)
                return (0);

            /*
             * "UTF-16" can be used for both LE and BE
             if ((!xmlStrncmp(BAD_CAST input->buf->encoder->name,
             BAD_CAST "UTF-16", 6)) &&
             (!xmlStrncmp(BAD_CAST handler->name,
             BAD_CAST "UTF-16", 6))) {
             return(0);
             }
             */

            /*
             * Note: this is a bit dangerous, but that's what it
             * takes to use nearly compatible signature for different
             * encodings.
             */
            xmlCharEncCloseFunc(input->buf->encoder);
            input->buf->encoder = handler;
            return (0);
        }
        input->buf->encoder = handler;

        /*
         * Is there already some content down the pipe to convert ?
         */
        if ((input->buf->buffer != NULL) && (input->buf->buffer->use > 0)) {
            int processed;
	    unsigned int use;

            /*
             * Specific handling of the Byte Order Mark for 
             * UTF-16
             */
            if ((handler->name != NULL) &&
                (!strcmp(handler->name, "UTF-16LE") ||
                 !strcmp(handler->name, "UTF-16")) &&
                (input->cur[0] == 0xFF) && (input->cur[1] == 0xFE)) {
                input->cur += 2;
            }
            if ((handler->name != NULL) &&
                (!strcmp(handler->name, "UTF-16BE")) &&
                (input->cur[0] == 0xFE) && (input->cur[1] == 0xFF)) {
                input->cur += 2;
            }
            /*
             * Errata on XML-1.0 June 20 2001
             * Specific handling of the Byte Order Mark for
             * UTF-8
             */
            if ((handler->name != NULL) &&
                (!strcmp(handler->name, "UTF-8")) &&
                (input->cur[0] == 0xEF) &&
                (input->cur[1] == 0xBB) && (input->cur[2] == 0xBF)) {
                input->cur += 3;
            }

            /*
             * Shrink the current input buffer.
             * Move it as the raw buffer and create a new input buffer
             */
            processed = input->cur - input->base;
            xmlBufferShrink(input->buf->buffer, processed);
            input->buf->raw = input->buf->buffer;
            input->buf->buffer = xmlBufferCreate();
	    input->buf->rawconsumed = processed;
	    use = input->buf->raw->use;

            if (ctxt->html) {
                /*
                 * convert as much as possible of the buffer
                 */
                nbchars = xmlCharEncInFunc(input->buf->encoder,
                                           input->buf->buffer,
                                           input->buf->raw);
            } else {
                /*
                 * convert just enough to get
                 * '<?xml version="1.0" encoding="xxx"?>'
                 * parsed with the autodetected encoding
                 * into the parser reading buffer.
                 */
                nbchars = xmlCharEncFirstLine(input->buf->encoder,
                                              input->buf->buffer,
                                              input->buf->raw);
            }
            if (nbchars < 0) {
                xmlErrInternal(ctxt,
                               "switching encoding: encoder error\n",
                               NULL);
                return (-1);
            }
	    input->buf->rawconsumed += use - input->buf->raw->use;
            input->base = input->cur = input->buf->buffer->content;
            input->end = &input->base[input->buf->buffer->use];

        }
        return (0);
    } else {
        if ((input->length == 0) || (input->buf == NULL)) {
            /*
             * When parsing a static memory array one must know the
             * size to be able to convert the buffer.
             */
            xmlErrInternal(ctxt, "switching encoding : no input\n", NULL);
            return (-1);
        } else {
            int processed;

            /*
             * Shrink the current input buffer.
             * Move it as the raw buffer and create a new input buffer
             */
            processed = input->cur - input->base;

            input->buf->raw = xmlBufferCreate();
            xmlBufferAdd(input->buf->raw, input->cur,
                         input->length - processed);
            input->buf->buffer = xmlBufferCreate();

            /*
             * convert as much as possible of the raw input
             * to the parser reading buffer.
             */
            nbchars = xmlCharEncInFunc(input->buf->encoder,
                                       input->buf->buffer,
                                       input->buf->raw);
            if (nbchars < 0) {
                xmlErrInternal(ctxt,
                               "switching encoding: encoder error\n",
                               NULL);
                return (-1);
            }

            /*
             * Conversion succeeded, get rid of the old buffer
             */
            if ((input->free != NULL) && (input->base != NULL))
                input->free((xmlChar *) input->base);
            input->base = input->cur = input->buf->buffer->content;
            input->end = &input->base[input->buf->buffer->use];
        }
    }
    return (0);
}

/**
 * xmlSwitchToEncoding:
 * @ctxt:  the parser context
 * @handler:  the encoding handler