parser.c.svn-base

这是一个用于解析xml文件的类库。使用这个类库

		if ((ctxt->standalone == 1) ||
		    ((ctxt->hasExternalSubset == 0) &&
		     (ctxt->hasPErefs == 0))) {
		    xmlFatalErrMsgStr(ctxt, XML_ERR_UNDECLARED_ENTITY,
			 "PEReference: %%%s; not found\n", name);
	        } else {
		    /*
		     * [ VC: Entity Declared ]
		     * In a document with an external subset or external
		     * parameter entities with "standalone='no'", ...
		     * ... The declaration of a parameter entity must precede
		     * any reference to it...
		     */
		    if ((ctxt->validate) && (ctxt->vctxt.error != NULL)) {
		        xmlValidityError(ctxt, XML_WAR_UNDECLARED_ENTITY,
			                 "PEReference: %%%s; not found\n",
				         name);
		    } else 
		        xmlWarningMsg(ctxt, XML_WAR_UNDECLARED_ENTITY,
			              "PEReference: %%%s; not found\n",
				      name, NULL);
		    ctxt->valid = 0;
		}
	    } else if (ctxt->input->free != deallocblankswrapper) {
		    input = xmlNewBlanksWrapperInputStream(ctxt, entity);
		    xmlPushInput(ctxt, input);
	    } else {
	        if ((entity->etype == XML_INTERNAL_PARAMETER_ENTITY) ||
		    (entity->etype == XML_EXTERNAL_PARAMETER_ENTITY)) {
		    xmlChar start[4];
		    xmlCharEncoding enc;

		    /*
		     * handle the extra spaces added before and after
		     * c.f. http://www.w3.org/TR/REC-xml#as-PE
		     * this is done independently.
		     */
		    input = xmlNewEntityInputStream(ctxt, entity);
		    xmlPushInput(ctxt, input);

		    /* 
		     * Get the 4 first bytes and decode the charset
		     * if enc != XML_CHAR_ENCODING_NONE
		     * plug some encoding conversion routines.
		     * Note that, since we may have some non-UTF8
		     * encoding (like UTF16, bug 135229), the 'length'
		     * is not known, but we can calculate based upon
		     * the amount of data in the buffer.
		     */
		    GROW
		    if ((ctxt->input->end - ctxt->input->cur)>=4) {
			start[0] = RAW;
			start[1] = NXT(1);
			start[2] = NXT(2);
			start[3] = NXT(3);
			enc = xmlDetectCharEncoding(start, 4);
			if (enc != XML_CHAR_ENCODING_NONE) {
			    xmlSwitchEncoding(ctxt, enc);
			}
		    }

		    if ((entity->etype == XML_EXTERNAL_PARAMETER_ENTITY) &&
			(CMP5(CUR_PTR, '<', '?', 'x', 'm', 'l' )) &&
			(IS_BLANK_CH(NXT(5)))) {
			xmlParseTextDecl(ctxt);
		    }
		} else {
		    xmlFatalErrMsgStr(ctxt, XML_ERR_ENTITY_IS_PARAMETER,
			     "PEReference: %s is not a parameter entity\n",
				      name);
		}
	    }
	} else {
	    xmlFatalErr(ctxt, XML_ERR_PEREF_SEMICOL_MISSING, NULL);
	}
    }
}

/*
 * Macro used to grow the current buffer.
 */
#define growBuffer(buffer) {						\
    xmlChar *tmp;							\
    buffer##_size *= 2;							\
    tmp = (xmlChar *)							\
    		xmlRealloc(buffer, buffer##_size * sizeof(xmlChar));	\
    if (tmp == NULL) goto mem_error;					\
    buffer = tmp;							\
}

/**
 * xmlStringLenDecodeEntities:
 * @ctxt:  the parser context
 * @str:  the input string
 * @len: the string length
 * @what:  combination of XML_SUBSTITUTE_REF and XML_SUBSTITUTE_PEREF
 * @end:  an end marker xmlChar, 0 if none
 * @end2:  an end marker xmlChar, 0 if none
 * @end3:  an end marker xmlChar, 0 if none
 * 
 * Takes a entity string content and process to do the adequate substitutions.
 *
 * [67] Reference ::= EntityRef | CharRef
 *
 * [69] PEReference ::= '%' Name ';'
 *
 * Returns A newly allocated string with the substitution done. The caller
 *      must deallocate it !
 */
xmlChar *
xmlStringLenDecodeEntities(xmlParserCtxtPtr ctxt, const xmlChar *str, int len,
		      int what, xmlChar end, xmlChar  end2, xmlChar end3) {
    xmlChar *buffer = NULL;
    int buffer_size = 0;

    xmlChar *current = NULL;
    const xmlChar *last;
    xmlEntityPtr ent;
    int c,l;
    int nbchars = 0;

    if ((str == NULL) || (len < 0))
	return(NULL);
    last = str + len;

    if (ctxt->depth > 40) {
	xmlFatalErr(ctxt, XML_ERR_ENTITY_LOOP, NULL);
	return(NULL);
    }

    /*
     * allocate a translation buffer.
     */
    buffer_size = XML_PARSER_BIG_BUFFER_SIZE;
    buffer = (xmlChar *) xmlMallocAtomic(buffer_size * sizeof(xmlChar));
    if (buffer == NULL) goto mem_error;

    /*
     * OK loop until we reach one of the ending char or a size limit.
     * we are operating on already parsed values.
     */
    if (str < last)
	c = CUR_SCHAR(str, l);
    else
        c = 0;
    while ((c != 0) && (c != end) && /* non input consuming loop */
	   (c != end2) && (c != end3)) {

	if (c == 0) break;
        if ((c == '&') && (str[1] == '#')) {
	    int val = xmlParseStringCharRef(ctxt, &str);
	    if (val != 0) {
		COPY_BUF(0,buffer,nbchars,val);
	    }
	} else if ((c == '&') && (what & XML_SUBSTITUTE_REF)) {
	    if (xmlParserDebugEntities)
		xmlGenericError(xmlGenericErrorContext,
			"String decoding Entity Reference: %.30s\n",
			str);
	    ent = xmlParseStringEntityRef(ctxt, &str);
	    if ((ent != NULL) &&
		(ent->etype == XML_INTERNAL_PREDEFINED_ENTITY)) {
		if (ent->content != NULL) {
		    COPY_BUF(0,buffer,nbchars,ent->content[0]);
		} else {
		    xmlFatalErrMsg(ctxt, XML_ERR_INTERNAL_ERROR,
			    "predefined entity has no content\n");
		}
	    } else if ((ent != NULL) && (ent->content != NULL)) {
		xmlChar *rep;

		ctxt->depth++;
		rep = xmlStringDecodeEntities(ctxt, ent->content, what,
			                      0, 0, 0);
		ctxt->depth--;
		if (rep != NULL) {
		    current = rep;
		    while (*current != 0) { /* non input consuming loop */
			buffer[nbchars++] = *current++;
			if (nbchars >
		            buffer_size - XML_PARSER_BUFFER_SIZE) {
			    growBuffer(buffer);
			}
		    }
		    xmlFree(rep);
		}
	    } else if (ent != NULL) {
		int i = xmlStrlen(ent->name);
		const xmlChar *cur = ent->name;

		buffer[nbchars++] = '&';
		if (nbchars > buffer_size - i - XML_PARSER_BUFFER_SIZE) {
		    growBuffer(buffer);
		}
		for (;i > 0;i--)
		    buffer[nbchars++] = *cur++;
		buffer[nbchars++] = ';';
	    }
	} else if (c == '%' && (what & XML_SUBSTITUTE_PEREF)) {
	    if (xmlParserDebugEntities)
		xmlGenericError(xmlGenericErrorContext,
			"String decoding PE Reference: %.30s\n", str);
	    ent = xmlParseStringPEReference(ctxt, &str);
	    if (ent != NULL) {
		xmlChar *rep;

		ctxt->depth++;
		rep = xmlStringDecodeEntities(ctxt, ent->content, what,
			                      0, 0, 0);
		ctxt->depth--;
		if (rep != NULL) {
		    current = rep;
		    while (*current != 0) { /* non input consuming loop */
			buffer[nbchars++] = *current++;
			if (nbchars >
		            buffer_size - XML_PARSER_BUFFER_SIZE) {
			    growBuffer(buffer);
			}
		    }
		    xmlFree(rep);
		}
	    }
	} else {
	    COPY_BUF(l,buffer,nbchars,c);
	    str += l;
	    if (nbchars > buffer_size - XML_PARSER_BUFFER_SIZE) {
	      growBuffer(buffer);
	    }
	}
	if (str < last)
	    c = CUR_SCHAR(str, l);
	else
	    c = 0;
    }
    buffer[nbchars++] = 0;
    return(buffer);

mem_error:
    xmlErrMemory(ctxt, NULL);
    return(NULL);
}

/**
 * xmlStringDecodeEntities:
 * @ctxt:  the parser context
 * @str:  the input string
 * @what:  combination of XML_SUBSTITUTE_REF and XML_SUBSTITUTE_PEREF
 * @end:  an end marker xmlChar, 0 if none
 * @end2:  an end marker xmlChar, 0 if none
 * @end3:  an end marker xmlChar, 0 if none
 * 
 * Takes a entity string content and process to do the adequate substitutions.
 *
 * [67] Reference ::= EntityRef | CharRef
 *
 * [69] PEReference ::= '%' Name ';'
 *
 * Returns A newly allocated string with the substitution done. The caller
 *      must deallocate it !
 */
xmlChar *
xmlStringDecodeEntities(xmlParserCtxtPtr ctxt, const xmlChar *str, int what,
		        xmlChar end, xmlChar  end2, xmlChar end3) {
    return(xmlStringLenDecodeEntities(ctxt, str, xmlStrlen(str), what,
           end, end2, end3));
}

/************************************************************************
 *									*
 *		Commodity functions, cleanup needed ?			*
 *									*
 ************************************************************************/

/**
 * areBlanks:
 * @ctxt:  an XML parser context
 * @str:  a xmlChar *
 * @len:  the size of @str
 * @blank_chars: we know the chars are blanks
 *
 * Is this a sequence of blank chars that one can ignore ?
 *
 * Returns 1 if ignorable 0 otherwise.
 */

static int areBlanks(xmlParserCtxtPtr ctxt, const xmlChar *str, int len,
                     int blank_chars) {
    int i, ret;
    xmlNodePtr lastChild;

    /*
     * Don't spend time trying to differentiate them, the same callback is
     * used !
     */
    if (ctxt->sax->ignorableWhitespace == ctxt->sax->characters)
	return(0);

    /*
     * Check for xml:space value.
     */
    if (*(ctxt->space) == 1)
	return(0);

    /*
     * Check that the string is made of blanks
     */
    if (blank_chars == 0) {
	for (i = 0;i < len;i++)
	    if (!(IS_BLANK_CH(str[i]))) return(0);
    }

    /*
     * Look if the element is mixed content in the DTD if available
     */
    if (ctxt->node == NULL) return(0);
    if (ctxt->myDoc != NULL) {
	ret = xmlIsMixedElement(ctxt->myDoc, ctxt->node->name);
        if (ret == 0) return(1);
        if (ret == 1) return(0);
    }

    /*
     * Otherwise, heuristic :-\
     */
    if (RAW != '<') return(0);
    if ((ctxt->node->children == NULL) &&
	(RAW == '<') && (NXT(1) == '/')) return(0);

    lastChild = xmlGetLastChild(ctxt->node);
    if (lastChild == NULL) {
        if ((ctxt->node->type != XML_ELEMENT_NODE) &&
            (ctxt->node->content != NULL)) return(0);
    } else if (xmlNodeIsText(lastChild))
        return(0);
    else if ((ctxt->node->children != NULL) &&
             (xmlNodeIsText(ctxt->node->children)))
        return(0);
    return(1);
}

/************************************************************************
 *									*
 *		Extra stuff for namespace support			*
 *	Relates to http://www.w3.org/TR/WD-xml-names			*
 *									*
 ************************************************************************/

/**
 * xmlSplitQName:
 * @ctxt:  an XML parser context
 * @name:  an XML parser context
 * @prefix:  a xmlChar ** 
 *
 * parse an UTF8 encoded XML qualified name string
 *
 * [NS 5] QName ::= (Prefix ':')? LocalPart
 *
 * [NS 6] Prefix ::= NCName
 *
 * [NS 7] LocalPart ::= NCName
 *
 * Returns the local part, and prefix is updated
 *   to get the Prefix if any.
 */

xmlChar *
xmlSplitQName(xmlParserCtxtPtr ctxt, const xmlChar *name, xmlChar **prefix) {
    xmlChar buf[XML_MAX_NAMELEN + 5];
    xmlChar *buffer = NULL;
    int len = 0;
    int max = XML_MAX_NAMELEN;
    xmlChar *ret = NULL;
    const xmlChar *cur = name;
    int c;

    *prefix = NULL;

    if (cur == NULL) return(NULL);

#ifndef XML_XML_NAMESPACE
    /* xml: prefix is not really a namespace */
    if ((cur[0] == 'x') && (cur[1] == 'm') &&
        (cur[2] == 'l') && (cur[3] == ':'))
	return(xmlStrdup(name));
#endif

    /* nasty but well=formed */
    if (cur[0] == ':')
	return(xmlStrdup(name));

    c = *cur++;
    while ((c != 0) && (c != ':') && (len < max)) { /* tested bigname.xml */
	buf[len++] = c;
	c = *cur++;
    }
    if (len >= max) {
	/*
	 * Okay someone managed to make a huge name, so he's ready to pay
	 * for the processing speed.
	 */
	max = len * 2;
	
	buffer = (xmlChar *) xmlMallocAtomic(max * sizeof(xmlChar));
	if (buffer == NULL) {
	    xmlErrMemory(ctxt, NULL);
	    return(NULL);
	}
	memcpy(buffer, buf, len);
	while ((c != 0) && (c != ':')) { /* tested bigname.xml */
	    if (len + 10 > max) {
		max *= 2;
		buffer = (xmlChar *) xmlRealloc(buffer,
						max * sizeof(xmlChar));
		if (buffer == NULL) {
		    xmlErrMemory(ctxt, NULL);
		    return(NULL);
		}
	    }
	    buffer[len++] = c;
	    c = *cur++;
	}
	buffer[len] = 0;
    }
    
    /* nasty but well=formed
    if ((c == ':') && (*cur == 0)) {
	return(xmlStrdup(name));
    } */

    if (buffer == NULL)
	ret = xmlStrndup(buf, len);
    else {
	ret = buffer;
	buffer = NULL;
	max = XML_MAX_NAMELEN;
    }


    if (c == ':') {
	c = *cur;
        *prefix = ret;
	if (c == 0) {