parser.c

SIP（Session Initiation Protocol）是由IETF定义

			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)) {
			    if ((ctxt->sax != NULL) &&
				(ctxt->sax->error != NULL))
				ctxt->sax->error(ctxt->userData, 
				 "Char out of allowed range\n");
			    ctxt->errNo = XML_ERR_INVALID_ENCODING;
			    ctxt->wellFormed = 0;
			    ctxt->disableSAX = 1;
			}    
		    } else
		      /* 2-byte code */
		        ctxt->input->cur += 2;
		} else
		    /* 1-byte code */
		    ctxt->input->cur++;
	    } else {
		/*
		 * Assume it's a fixed lenght encoding (1) with
		 * a compatibke encoding for the ASCII set, since
		 * XML constructs only use < 128 chars
		 */
	        ctxt->input->cur++;
	    }
	    ctxt->nbChars++;
	    if (*ctxt->input->cur == 0)
		xmlParserInputGrow(ctxt->input, INPUT_CHUNK);
	}
    }
    if (*ctxt->input->cur == '%') xmlParserHandlePEReference(ctxt);
    if (*ctxt->input->cur == '&') xmlParserHandleReference(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 advertized 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->sax != NULL) && (ctxt->sax->error != NULL))
	ctxt->sax->error(ctxt->userData, 
			 "Input is not proper UTF-8, indicate encoding !\n");
    ctxt->errNo = XML_ERR_INVALID_ENCODING;

    ctxt->encoding = xmlStrdup(BAD_CAST "ISO-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 actaully 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 lenght
 */

int
xmlCurrentChar(xmlParserCtxtPtr ctxt, int *len) {
    if (ctxt->token != 0) {
	*len = 0;
	return(ctxt->token);
    }	
    if (ctxt->encoding == NULL) {
	/*
	 * 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 (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)) {
		if ((ctxt->sax != NULL) &&
		    (ctxt->sax->error != NULL))
		    ctxt->sax->error(ctxt->userData, 
				     "Char out of allowed range\n");
		ctxt->errNo = XML_ERR_INVALID_ENCODING;
		ctxt->wellFormed = 0;
		ctxt->disableSAX = 1;
	    }    
	    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 lenght encoding (1) with
     * a compatibke 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:
    /*
     * If we detect an UTF8 error that probably mean that the
     * input encoding didn't get properly advertized 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->sax != NULL) && (ctxt->sax->error != NULL))
	ctxt->sax->error(ctxt->userData, 
			 "Input is not proper UTF-8, indicate encoding !\n");
    ctxt->errNo = XML_ERR_INVALID_ENCODING;

    ctxt->encoding = xmlStrdup(BAD_CAST "ISO-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 actaully span multiple
 * bytes in the input buffer.
 *
 * Returns the current char value and its lenght
 */

int
xmlStringCurrentChar(xmlParserCtxtPtr ctxt, const xmlChar *cur, int *len) {
    if (ctxt->encoding == NULL) {
	/*
	 * 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[2] & 0x3f;
	    }
	    if (!IS_CHAR(val)) {
		if ((ctxt->sax != NULL) &&
		    (ctxt->sax->error != NULL))
		    ctxt->sax->error(ctxt->userData, 
				     "Char out of allowed range\n");
		ctxt->errNo = XML_ERR_INVALID_ENCODING;
		ctxt->wellFormed = 0;
		ctxt->disableSAX = 1;
	    }    
	    return(val);
	} else {
	    /* 1-byte code */
	    *len = 1;
	    return((int) *cur);
	}
    }
    /*
     * Assume it's a fixed lenght encoding (1) with
     * a compatibke encoding for the ASCII set, since
     * XML constructs only use < 128 chars
     */
    *len = 1;
    return((int) *cur);
encoding_error:
    /*
     * If we detect an UTF8 error that probably mean that the
     * input encoding didn't get properly advertized 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->sax != NULL) && (ctxt->sax->error != NULL))
	ctxt->sax->error(ctxt->userData, 
			 "Input is not proper UTF-8, indicate encoding !\n");
    ctxt->errNo = XML_ERR_INVALID_ENCODING;

    *len = 1;
    return((int) *cur);
}

/**
 * xmlCopyChar:
 * @len:  pointer to the length of the char read (or zero)
 * @array:  pointer to an arry of xmlChar
 * @val:  the char value
 *
 * append the char value in the array 
 *
 * Returns the number of xmlChar written
 */

int
xmlCopyChar(int len, xmlChar *out, int val) {
    /*
     * 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 (len == 0) {
	if (val < 0) len = 0;
	else if (val < 0x80) len = 1;
	else if (val < 0x800) len = 2;
	else if (val < 0x10000) len = 3;
	else if (val < 0x110000) len = 4;
	if (len == 0) {
	    fprintf(stderr, "Internal error, xmlCopyChar 0x%X out of bound\n",
		    val);
	    return(0);
	}
    }
    if (len > 1) {
	int bits; 

        if      (val <    0x80) {  *out++=  val;                bits= -6; }
        else if (val <   0x800) {  *out++= (val >>  6) | 0xC0;  bits=  0; }
        else if (val < 0x10000) {  *out++= (val >> 12) | 0xE0;  bits=  6; }
        else                  {    *out++= (val >> 18) | 0xF0;  bits= 12; }
 
        for ( ; bits >= 0; bits-= 6)
            *out++= ((val >> bits) & 0x3F) | 0x80 ;

        return(len);
    }
    *out = (xmlChar) val;
    return(1);
}

/**
 * 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 cur, res = 0;

    do {
	cur = CUR;
	while (IS_BLANK(cur)) {
	    NEXT;
	    cur = CUR;
	    res++;
	}
	while ((cur == 0) && (ctxt->inputNr > 1) &&
	       (ctxt->instate != XML_PARSER_COMMENT)) {
	    xmlPopInput(ctxt);
	    cur = CUR;
	}
	if (*ctxt->input->cur == '%') xmlParserHandlePEReference(ctxt);
	if (*ctxt->input->cur == '&') xmlParserHandleReference(ctxt);
    } while (IS_BLANK(cur));
    return(res);
}

/************************************************************************
 *									*
 *	Commodity functions to handle entities processing		*
 *									*
 ************************************************************************/

/**
 * 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->inputNr == 1) return(0); /* End of main Input */
    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;
    inputPush(ctxt, input);
    GROW;
}

/**
 * xmlFreeInputStream:
 * @input:  an xmlParserInputPtr
 *
 * Free up an input stream.
 */
void
xmlFreeInputStream(xmlParserInputPtr input) {
    if (input == NULL) return;

    if (input->filename != NULL) xmlFree((char *) input->filename);
    if (input->directory != NULL) xmlFree((char *) input->directory);
    if (input->encoding != NULL) xmlFree((char *) input->encoding);
    if (input->version != NULL) xmlFree((char *) input->version);
    if ((input->free != NULL) && (input->base != NULL))
        input->free((xmlChar *) input->base);
    if (input->buf != NULL) 
        xmlFreeParserInputBuffer(input->buf);
    memset(input, -1, sizeof(xmlParserInput));
    xmlFree(input);
}

/**
 * xmlNewInputStream:
 * @ctxt:  an XML parser context
 *
 * Create a new input stream structure
 * Returns the new input stream or NULL