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.

    if (in->base != in->buf->buffer->content) {
        /*
	 * the buffer has been reallocated
	 */
	indx = in->cur - in->base;
	in->base = in->buf->buffer->content;
	in->cur = &in->buf->buffer->content[indx];
    }
    in->end = &in->buf->buffer->content[in->buf->buffer->use];

    CHECK_BUFFER(in);
}

/************************************************************************
 *									*
 * 		UTF8 character input and related functions		*
 *									*
 ************************************************************************/

/**
 * xmlNextChar:
 * @ctxt:  the XML parser context
 *
 * Skip to the next char input char.
 */

void
xmlNextChar(xmlParserCtxtPtr ctxt)
{
    if ((ctxt == NULL) || (ctxt->instate == XML_PARSER_EOF) ||
        (ctxt->input == NULL))
        return;

    if (ctxt->charset == XML_CHAR_ENCODING_UTF8) {
        if ((*ctxt->input->cur == 0) &&
            (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0) &&
            (ctxt->instate != XML_PARSER_COMMENT)) {
            /*
             * If we are at the end of the current entity and
             * the context allows it, we pop consumed entities
             * automatically.
             * the auto closing should be blocked in other cases
             */
            xmlPopInput(ctxt);
        } else {
            const unsigned char *cur;
            unsigned char c;

            /*
             *   2.11 End-of-Line Handling
             *   the literal two-character sequence "#xD#xA" or a standalone
             *   literal #xD, an XML processor must pass to the application
             *   the single character #xA.
             */
            if (*(ctxt->input->cur) == '\n') {
                ctxt->input->line++; ctxt->input->col = 1;
            } else
                ctxt->input->col++;

            /*
             * 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 
             *
             * Check for the 0x110000 limit too
             */
            cur = ctxt->input->cur;

            c = *cur;
            if (c & 0x80) {
	        if (c == 0xC0)
		    goto encoding_error;
                if (cur[1] == 0)
                    xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
                if ((cur[1] & 0xc0) != 0x80)
                    goto encoding_error;
                if ((c & 0xe0) == 0xe0) {
                    unsigned int val;

                    if (cur[2] == 0)
                        xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
                    if ((cur[2] & 0xc0) != 0x80)
                        goto encoding_error;
                    if ((c & 0xf0) == 0xf0) {
                        if (cur[3] == 0)
                            xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
                        if (((c & 0xf8) != 0xf0) ||
                            ((cur[3] & 0xc0) != 0x80))
                            goto encoding_error;
                        /* 4-byte code */
                        ctxt->input->cur += 4;
                        val = (cur[0] & 0x7) << 18;
                        val |= (cur[1] & 0x3f) << 12;
                        val |= (cur[2] & 0x3f) << 6;
                        val |= cur[3] & 0x3f;
                    } else {
                        /* 3-byte code */
                        ctxt->input->cur += 3;
                        val = (cur[0] & 0xf) << 12;
                        val |= (cur[1] & 0x3f) << 6;
                        val |= cur[2] & 0x3f;
                    }
                    if (((val > 0xd7ff) && (val < 0xe000)) ||
                        ((val > 0xfffd) && (val < 0x10000)) ||
                        (val >= 0x110000)) {
			xmlErrEncodingInt(ctxt, XML_ERR_INVALID_CHAR,
					  "Char 0x%X out of allowed range\n",
					  val);
                    }
                } else
                    /* 2-byte code */
                    ctxt->input->cur += 2;
            } else
                /* 1-byte code */
                ctxt->input->cur++;

            ctxt->nbChars++;
            if (*ctxt->input->cur == 0)
                xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
        }
    } else {
        /*
         * Assume it's a fixed length encoding (1) with
         * a compatible encoding for the ASCII set, since
         * XML constructs only use < 128 chars
         */

        if (*(ctxt->input->cur) == '\n') {
            ctxt->input->line++; ctxt->input->col = 1;
        } else
            ctxt->input->col++;
        ctxt->input->cur++;
        ctxt->nbChars++;
        if (*ctxt->input->cur == 0)
            xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
    }
    if ((*ctxt->input->cur == '%') && (!ctxt->html))
        xmlParserHandlePEReference(ctxt);
    if ((*ctxt->input->cur == 0) &&
        (xmlParserInputGrow(ctxt->input, INPUT_CHUNK) <= 0))
        xmlPopInput(ctxt);
    return;
encoding_error:
    /*
     * If we detect an UTF8 error that probably mean that the
     * input encoding didn't get properly advertised in the
     * declaration header. Report the error and switch the encoding
     * to ISO-Latin-1 (if you don't like this policy, just declare the
     * encoding !)
     */
    if ((ctxt == NULL) || (ctxt->input == NULL) ||
        (ctxt->input->end - ctxt->input->cur < 4)) {
	__xmlErrEncoding(ctxt, XML_ERR_INVALID_CHAR,
		     "Input is not proper UTF-8, indicate encoding !\n",
		     NULL, NULL);
    } else {
        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);
    }
    ctxt->charset = XML_CHAR_ENCODING_8859_1;
    ctxt->input->cur++;
    return;
}

/**
 * xmlCurrentChar:
 * @ctxt:  the XML parser context
 * @len:  pointer to the length of the char read
 *
 * The current char value, if using UTF-8 this may actually span multiple
 * bytes in the input buffer. Implement the end of line normalization:
 * 2.11 End-of-Line Handling
 * Wherever an external parsed entity or the literal entity value
 * of an internal parsed entity contains either the literal two-character
 * sequence "#xD#xA" or a standalone literal #xD, an XML processor
 * must pass to the application the single character #xA.
 * This behavior can conveniently be produced by normalizing all
 * line breaks to #xA on input, before parsing.)
 *
 * Returns the current char value and its length
 */

int
xmlCurrentChar(xmlParserCtxtPtr ctxt, int *len) {
    if ((ctxt == NULL) || (len == NULL) || (ctxt->input == NULL)) return(0);
    if (ctxt->instate == XML_PARSER_EOF)
	return(0);

    if ((*ctxt->input->cur >= 0x20) && (*ctxt->input->cur <= 0x7F)) {
	    *len = 1;
	    return((int) *ctxt->input->cur);
    }
    if (ctxt->charset == XML_CHAR_ENCODING_UTF8) {
	/*
	 * 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 
	 *
	 * Check for the 0x110000 limit too
	 */
	const unsigned char *cur = ctxt->input->cur;
	unsigned char c;
	unsigned int val;

	c = *cur;
	if (c & 0x80) {
	    if (c == 0xC0)
		goto encoding_error;
	    if (cur[1] == 0)
		xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
	    if ((cur[1] & 0xc0) != 0x80)
		goto encoding_error;
	    if ((c & 0xe0) == 0xe0) {

		if (cur[2] == 0)
		    xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
		if ((cur[2] & 0xc0) != 0x80)
		    goto encoding_error;
		if ((c & 0xf0) == 0xf0) {
		    if (cur[3] == 0)
			xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
		    if (((c & 0xf8) != 0xf0) ||
			((cur[3] & 0xc0) != 0x80))
			goto encoding_error;
		    /* 4-byte code */
		    *len = 4;
		    val = (cur[0] & 0x7) << 18;
		    val |= (cur[1] & 0x3f) << 12;
		    val |= (cur[2] & 0x3f) << 6;
		    val |= cur[3] & 0x3f;
		} else {
		  /* 3-byte code */
		    *len = 3;
		    val = (cur[0] & 0xf) << 12;
		    val |= (cur[1] & 0x3f) << 6;
		    val |= cur[2] & 0x3f;
		}
	    } else {
	      /* 2-byte code */
		*len = 2;
		val = (cur[0] & 0x1f) << 6;
		val |= cur[1] & 0x3f;
	    }
	    if (!IS_CHAR(val)) {
	        xmlErrEncodingInt(ctxt, XML_ERR_INVALID_CHAR,
				  "Char 0x%X out of allowed range\n", val);
	    }    
	    return(val);
	} else {
	    /* 1-byte code */
	    *len = 1;
	    if (*ctxt->input->cur == 0xD) {
		if (ctxt->input->cur[1] == 0xA) {
		    ctxt->nbChars++;
		    ctxt->input->cur++;
		}
		return(0xA);
	    }
	    return((int) *ctxt->input->cur);
	}
    }
    /*
     * Assume it's a fixed length encoding (1) with
     * a compatible encoding for the ASCII set, since
     * XML constructs only use < 128 chars
     */
    *len = 1;
    if (*ctxt->input->cur == 0xD) {
	if (ctxt->input->cur[1] == 0xA) {
	    ctxt->nbChars++;
	    ctxt->input->cur++;
	}
	return(0xA);
    }
    return((int) *ctxt->input->cur);
encoding_error:
    /*
     * An encoding problem may arise from a truncated input buffer
     * splitting a character in the middle. In that case do not raise
     * an error but return 0 to endicate an end of stream problem
     */
    if (ctxt->input->end - ctxt->input->cur < 4) {
	*len = 0;
	return(0);
    }

    /*
     * If we detect an UTF8 error that probably mean that the
     * input encoding didn't get properly advertised in the
     * declaration header. Report the error and switch the encoding
     * to ISO-Latin-1 (if you don't like this policy, just declare the
     * encoding !)
     */
    {
        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);
    }
    ctxt->charset = XML_CHAR_ENCODING_8859_1; 
    *len = 1;
    return((int) *ctxt->input->cur);
}

/**
 * xmlStringCurrentChar:
 * @ctxt:  the XML parser context
 * @cur:  pointer to the beginning of the char
 * @len:  pointer to the length of the char read
 *
 * The current char value, if using UTF-8 this may actually span multiple
 * bytes in the input buffer.
 *
 * Returns the current char value and its length
 */

int
xmlStringCurrentChar(xmlParserCtxtPtr ctxt, const xmlChar * cur, int *len)
{
    if ((len == NULL) || (cur == NULL)) return(0);
    if ((ctxt == NULL) || (ctxt->charset == XML_CHAR_ENCODING_UTF8)) {
        /*
         * 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 
         *
         * Check for the 0x110000 limit too
         */
        unsigned char c;
        unsigned int val;

        c = *cur;
        if (c & 0x80) {
            if ((cur[1] & 0xc0) != 0x80)
                goto encoding_error;
            if ((c & 0xe0) == 0xe0) {

                if ((cur[2] & 0xc0) != 0x80)
                    goto encoding_error;
                if ((c & 0xf0) == 0xf0) {
                    if (((c & 0xf8) != 0xf0) || ((cur[3] & 0xc0) != 0x80))
                        goto encoding_error;
                    /* 4-byte code */
                    *len = 4;
                    val = (cur[0] & 0x7) << 18;
                    val |= (cur[1] & 0x3f) << 12;
                    val |= (cur[2] & 0x3f) << 6;
                    val |= cur[3] & 0x3f;
                } else {
                    /* 3-byte code */
                    *len = 3;
                    val = (cur[0] & 0xf) << 12;
                    val |= (cur[1] & 0x3f) << 6;
                    val |= cur[2] & 0x3f;
                }
            } else {
                /* 2-byte code */
                *len = 2;
                val = (cur[0] & 0x1f) << 6;
                val |= cur[1] & 0x3f;
            }
            if (!IS_CHAR(val)) {
	        xmlErrEncodingInt(ctxt, XML_ERR_INVALID_CHAR,
				  "Char 0x%X out of allowed range\n", val);
            }
            return (val);
        } else {
            /* 1-byte code */
            *len = 1;
            return ((int) *cur);
        }
    }
    /*
     * Assume it's a fixed length encoding (1) with
     * a compatible encoding for the ASCII set, since
     * XML constructs only use < 128 chars
     */
    *len = 1;
    return ((int) *cur);
encoding_error:

    /*
     * An encoding problem may arise from a truncated input buffer
     * splitting a character in the middle. In that case do not raise
     * an error but return 0 to endicate an end of stream problem
     */
    if ((ctxt == NULL) || (ctxt->input == NULL) ||
        (ctxt->input->end - ctxt->input->cur < 4)) {
	*len = 0;
	return(0);
    }
    /*
     * If we detect an UTF8 error that probably mean that the
     * input encoding didn't get properly advertised in the
     * declaration header. Report the error and switch the encoding
     * to ISO-Latin-1 (if you don't like this policy, just declare the
     * encoding !)
     */