htmlparser.c

SIP 1.5.0源代码

	    if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
	        ctxt->sax->error(ctxt->userData, "htmlParseEntityRef: no name\n");
	    ctxt->wellFormed = 0;
	} else {
	    GROW;
	    if (CUR == ';') {
		*str = name;

		/*
		 * Lookup the entity in the table.
		 */
		ent = htmlEntityLookup(name);
		if (ent != NULL) /* OK that's ugly !!! */
		    NEXT;
	    } else {
		if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
		    ctxt->sax->error(ctxt->userData,
		                     "htmlParseEntityRef: expecting ';'\n");
		*str = name;
	    }
	}
    }
    return(ent);
}

/**
 * htmlParseAttValue:
 * @ctxt:  an HTML parser context
 *
 * parse a value for an attribute
 * Note: the parser won't do substitution of entities here, this
 * will be handled later in xmlStringGetNodeList, unless it was
 * asked for ctxt->replaceEntities != 0 
 *
 * Returns the AttValue parsed or NULL.
 */

xmlChar *
htmlParseAttValue(htmlParserCtxtPtr ctxt) {
    xmlChar *ret = NULL;

    if (CUR == '"') {
        NEXT;
	ret = htmlParseHTMLAttribute(ctxt, '"');
        if (CUR != '"') {
	    if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
		ctxt->sax->error(ctxt->userData, "AttValue: ' expected\n");
	    ctxt->wellFormed = 0;
	} else
	    NEXT;
    } else if (CUR == '\'') {
        NEXT;
	ret = htmlParseHTMLAttribute(ctxt, '\'');
        if (CUR != '\'') {
	    if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
		ctxt->sax->error(ctxt->userData, "AttValue: ' expected\n");
	    ctxt->wellFormed = 0;
	} else
	    NEXT;
    } else {
        /*
	 * That's an HTMLism, the attribute value may not be quoted
	 */
	ret = htmlParseHTMLAttribute(ctxt, 0);
	if (ret == NULL) {
	    if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
		ctxt->sax->error(ctxt->userData, "AttValue: no value found\n");
	    ctxt->wellFormed = 0;
	}
    }
    return(ret);
}

/**
 * htmlParseSystemLiteral:
 * @ctxt:  an HTML parser context
 * 
 * parse an HTML Literal
 *
 * [11] SystemLiteral ::= ('"' [^"]* '"') | ("'" [^']* "'")
 *
 * Returns the SystemLiteral parsed or NULL
 */

xmlChar *
htmlParseSystemLiteral(htmlParserCtxtPtr ctxt) {
    const xmlChar *q;
    xmlChar *ret = NULL;

    if (CUR == '"') {
        NEXT;
	q = CUR_PTR;
	while ((IS_CHAR(CUR)) && (CUR != '"'))
	    NEXT;
	if (!IS_CHAR(CUR)) {
	    if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
	        ctxt->sax->error(ctxt->userData, "Unfinished SystemLiteral\n");
	    ctxt->wellFormed = 0;
	} else {
	    ret = xmlStrndup(q, CUR_PTR - q);
	    NEXT;
        }
    } else if (CUR == '\'') {
        NEXT;
	q = CUR_PTR;
	while ((IS_CHAR(CUR)) && (CUR != '\''))
	    NEXT;
	if (!IS_CHAR(CUR)) {
	    if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
	        ctxt->sax->error(ctxt->userData, "Unfinished SystemLiteral\n");
	    ctxt->wellFormed = 0;
	} else {
	    ret = xmlStrndup(q, CUR_PTR - q);
	    NEXT;
        }
    } else {
	if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
	    ctxt->sax->error(ctxt->userData,
	                     "SystemLiteral \" or ' expected\n");
	ctxt->wellFormed = 0;
    }
    
    return(ret);
}

/**
 * htmlParsePubidLiteral:
 * @ctxt:  an HTML parser context
 *
 * parse an HTML public literal
 *
 * [12] PubidLiteral ::= '"' PubidChar* '"' | "'" (PubidChar - "'")* "'"
 *
 * Returns the PubidLiteral parsed or NULL.
 */

xmlChar *
htmlParsePubidLiteral(htmlParserCtxtPtr ctxt) {
    const xmlChar *q;
    xmlChar *ret = NULL;
    /*
     * Name ::= (Letter | '_') (NameChar)*
     */
    if (CUR == '"') {
        NEXT;
	q = CUR_PTR;
	while (IS_PUBIDCHAR(CUR)) NEXT;
	if (CUR != '"') {
	    if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
	        ctxt->sax->error(ctxt->userData, "Unfinished PubidLiteral\n");
	    ctxt->wellFormed = 0;
	} else {
	    ret = xmlStrndup(q, CUR_PTR - q);
	    NEXT;
	}
    } else if (CUR == '\'') {
        NEXT;
	q = CUR_PTR;
	while ((IS_LETTER(CUR)) && (CUR != '\''))
	    NEXT;
	if (!IS_LETTER(CUR)) {
	    if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
	        ctxt->sax->error(ctxt->userData, "Unfinished PubidLiteral\n");
	    ctxt->wellFormed = 0;
	} else {
	    ret = xmlStrndup(q, CUR_PTR - q);
	    NEXT;
	}
    } else {
	if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
	    ctxt->sax->error(ctxt->userData, "SystemLiteral \" or ' expected\n");
	ctxt->wellFormed = 0;
    }
    
    return(ret);
}

/**
 * htmlParseCharData:
 * @ctxt:  an HTML parser context
 * @cdata:  int indicating whether we are within a CDATA section
 *
 * parse a CharData section.
 * if we are within a CDATA section ']]>' marks an end of section.
 *
 * [14] CharData ::= [^<&]* - ([^<&]* ']]>' [^<&]*)
 */

void
htmlParseCharData(htmlParserCtxtPtr ctxt, int cdata) {
    xmlChar *buf = NULL;
    int len = 0;
    int size = HTML_PARSER_BUFFER_SIZE;
    xmlChar q;

    buf = (xmlChar *) xmlMalloc(size * sizeof(xmlChar));
    if (buf == NULL) {
	fprintf(stderr, "malloc of %d byte failed\n", size);
	return;
    }

    q = CUR;
    while ((IS_CHAR(q)) && (q != '<') &&
           (q != '&')) {
	if ((q == ']') && (NXT(1) == ']') &&
	    (NXT(2) == '>')) {
	    if (cdata) break;
	    else {
		if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
		    ctxt->sax->error(ctxt->userData,
		       "Sequence ']]>' not allowed in content\n");
		ctxt->wellFormed = 0;
	    }
	}
	if (len + 1 >= size) {
	    size *= 2;
	    buf = xmlRealloc(buf, size * sizeof(xmlChar));
	    if (buf == NULL) {
		fprintf(stderr, "realloc of %d byte failed\n", size);
		return;
	    }
	}
	buf[len++] = q;
        NEXT;
	q = CUR;
    }
    if (len == 0) {
	xmlFree(buf);
	return;
    }

    /*
     * Ok the buffer is to be consumed as chars.
     */
    if (ctxt->sax != NULL) {
	if (areBlanks(ctxt, buf, len)) {
	    if (ctxt->sax->ignorableWhitespace != NULL)
		ctxt->sax->ignorableWhitespace(ctxt->userData, buf, len);
	} else {
	    if (ctxt->sax->characters != NULL)
		ctxt->sax->characters(ctxt->userData, buf, len);
        }
    }
    xmlFree(buf);
}

/**
 * htmlParseExternalID:
 * @ctxt:  an HTML parser context
 * @publicID:  a xmlChar** receiving PubidLiteral
 * @strict: indicate whether we should restrict parsing to only
 *          production [75], see NOTE below
 *
 * Parse an External ID or a Public ID
 *
 * NOTE: Productions [75] and [83] interract badly since [75] can generate
 *       'PUBLIC' S PubidLiteral S SystemLiteral
 *
 * [75] ExternalID ::= 'SYSTEM' S SystemLiteral
 *                   | 'PUBLIC' S PubidLiteral S SystemLiteral
 *
 * [83] PublicID ::= 'PUBLIC' S PubidLiteral
 *
 * Returns the function returns SystemLiteral and in the second
 *                case publicID receives PubidLiteral, is strict is off
 *                it is possible to return NULL and have publicID set.
 */

xmlChar *
htmlParseExternalID(htmlParserCtxtPtr ctxt, xmlChar **publicID, int strict) {
    xmlChar *URI = NULL;

    if ((UPPER == 'S') && (UPP(1) == 'Y') &&
         (UPP(2) == 'S') && (UPP(3) == 'T') &&
	 (UPP(4) == 'E') && (UPP(5) == 'M')) {
        SKIP(6);
	if (!IS_BLANK(CUR)) {
	    if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
		ctxt->sax->error(ctxt->userData,
		    "Space required after 'SYSTEM'\n");
	    ctxt->wellFormed = 0;
	}
        SKIP_BLANKS;
	URI = htmlParseSystemLiteral(ctxt);
	if (URI == NULL) {
	    if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
	        ctxt->sax->error(ctxt->userData,
	          "htmlParseExternalID: SYSTEM, no URI\n");
	    ctxt->wellFormed = 0;
        }
    } else if ((UPPER == 'P') && (UPP(1) == 'U') &&
	       (UPP(2) == 'B') && (UPP(3) == 'L') &&
	       (UPP(4) == 'I') && (UPP(5) == 'C')) {
        SKIP(6);
	if (!IS_BLANK(CUR)) {
	    if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
		ctxt->sax->error(ctxt->userData,
		    "Space required after 'PUBLIC'\n");
	    ctxt->wellFormed = 0;
	}
        SKIP_BLANKS;
	*publicID = htmlParsePubidLiteral(ctxt);
	if (*publicID == NULL) {
	    if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
	        ctxt->sax->error(ctxt->userData, 
	          "htmlParseExternalID: PUBLIC, no Public Identifier\n");
	    ctxt->wellFormed = 0;
	}
        SKIP_BLANKS;
        if ((CUR == '"') || (CUR == '\'')) {
	    URI = htmlParseSystemLiteral(ctxt);
	}
    }
    return(URI);
}

/**
 * htmlParseComment:
 * @ctxt:  an HTML parser context
 *
 * Parse an XML (SGML) comment <!-- .... -->
 *
 * [15] Comment ::= '<!--' ((Char - '-') | ('-' (Char - '-')))* '-->'
 */
void
htmlParseComment(htmlParserCtxtPtr ctxt) {
    xmlChar *buf = NULL;
    int len = 0;
    int size = HTML_PARSER_BUFFER_SIZE;
    register xmlChar s, r, q;

    /*
     * Check that there is a comment right here.
     */
    if ((CUR != '<') || (NXT(1) != '!') ||
        (NXT(2) != '-') || (NXT(3) != '-')) return;

    buf = (xmlChar *) xmlMalloc(size * sizeof(xmlChar));
    if (buf == NULL) {
	fprintf(stderr, "malloc of %d byte failed\n", size);
	return;
    }
    q = r = '-'; /* 0 or '-' to cover our ass against <!--> and <!---> ? !!! */
    SKIP(4);
    s = CUR;
    
    while (IS_CHAR(s) &&
           ((s != '>') || (r != '-') || (q != '-'))) {
	if (len + 1 >= size) {
	    size *= 2;
	    buf = xmlRealloc(buf, size * sizeof(xmlChar));
	    if (buf == NULL) {
		fprintf(stderr, "realloc of %d byte failed\n", size);
		return;
	    }
	}
	buf[len++] = s;
        NEXT;
	q = r;
	r = s;
	s = CUR;
    }
    buf[len - 2] = 0;
    if (!IS_CHAR(s)) {
	if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
	    ctxt->sax->error(ctxt->userData, "Comment not terminated \n<!--%.50s\n", buf);
	ctxt->wellFormed = 0;
    } else {
        NEXT;
	if ((ctxt->sax != NULL) && (ctxt->sax->comment != NULL)) {
	    ctxt->sax->comment(ctxt->userData, buf);
	}
    }
    xmlFree(buf);
}

/**
 * htmlParseCharRef:
 * @ctxt:  an HTML parser context
 *
 * parse Reference declarations
 *
 * [66] CharRef ::= '&#' [0-9]+ ';' |
 *                  '&#x' [0-9a-fA-F]+ ';'
 *
 * Returns the value parsed (as an int)
 */
int
htmlParseCharRef(htmlParserCtxtPtr ctxt) {
    int val = 0;

    if ((CUR == '&') && (NXT(1) == '#') &&
        (NXT(2) == 'x')) {
	SKIP(3);
	while (CUR != ';') {
	    if ((CUR >= '0') && (CUR <= '9')) 
	        val = val