sax2.c.svn-base

这是一个用于解析xml文件的类库。使用这个类库

	ret->children = tmp;
	ret->last = tmp;
	if (tmp != NULL) {
	    tmp->doc = ret->doc;
	    tmp->parent = (xmlNodePtr) ret;
	}
    }

#ifdef LIBXML_VALID_ENABLED
    if ((!ctxt->html) && ctxt->validate && ctxt->wellFormed &&
        ctxt->myDoc && ctxt->myDoc->intSubset) {
	/*
	 * If we don't substitute entities, the validation should be
	 * done on a value with replaced entities anyway.
	 */
        if (!ctxt->replaceEntities) {
	    dup = xmlSAX2DecodeAttrEntities(ctxt, value, valueend);
	    if (dup == NULL) {
	        if (*valueend == 0) {
		    ctxt->valid &= xmlValidateOneAttribute(&ctxt->vctxt,
				    ctxt->myDoc, ctxt->node, ret, value);
		} else {
		    /*
		     * That should already be normalized.
		     * cheaper to finally allocate here than duplicate
		     * entry points in the full validation code
		     */
		    dup = xmlStrndup(value, valueend - value);

		    ctxt->valid &= xmlValidateOneAttribute(&ctxt->vctxt,
				    ctxt->myDoc, ctxt->node, ret, dup);
		}
	    } else {
	        /*
		 * dup now contains a string of the flattened attribute
		 * content with entities substitued. Check if we need to
		 * apply an extra layer of normalization.
		 * It need to be done twice ... it's an extra burden related
		 * to the ability to keep references in attributes
		 */
		if (ctxt->attsSpecial != NULL) {
		    xmlChar *nvalnorm;
		    xmlChar fn[50];
		    xmlChar *fullname;
		    
		    fullname = xmlBuildQName(localname, prefix, fn, 50);
		    if (fullname != NULL) {
			ctxt->vctxt.valid = 1;
		        nvalnorm = xmlValidCtxtNormalizeAttributeValue(
			                 &ctxt->vctxt, ctxt->myDoc,
					 ctxt->node, fullname, dup);
			if (ctxt->vctxt.valid != 1)
			    ctxt->valid = 0;

			if ((fullname != fn) && (fullname != localname))
			    xmlFree(fullname);
			if (nvalnorm != NULL) {
			    xmlFree(dup);
			    dup = nvalnorm;
			}
		    }
		}

		ctxt->valid &= xmlValidateOneAttribute(&ctxt->vctxt,
			        ctxt->myDoc, ctxt->node, ret, dup);
	    }
	} else {
	    /*
	     * if entities already have been substitued, then
	     * the attribute as passed is already normalized
	     */
	    dup = xmlStrndup(value, valueend - value);

	    ctxt->valid &= xmlValidateOneAttribute(&ctxt->vctxt,
	                             ctxt->myDoc, ctxt->node, ret, dup);
	}
    } else
#endif /* LIBXML_VALID_ENABLED */
           if (((ctxt->loadsubset & XML_SKIP_IDS) == 0) &&
	       (((ctxt->replaceEntities == 0) && (ctxt->external != 2)) ||
	        ((ctxt->replaceEntities != 0) && (ctxt->inSubset == 0)))) {
        /*
	 * when validating, the ID registration is done at the attribute
	 * validation level. Otherwise we have to do specific handling here.
	 */
	if (xmlIsID(ctxt->myDoc, ctxt->node, ret)) {
	    /* might be worth duplicate entry points and not copy */
	    if (dup == NULL)
	        dup = xmlStrndup(value, valueend - value);
	    xmlAddID(&ctxt->vctxt, ctxt->myDoc, dup, ret);
	} else if (xmlIsRef(ctxt->myDoc, ctxt->node, ret)) {
	    if (dup == NULL)
	        dup = xmlStrndup(value, valueend - value);
	    xmlAddRef(&ctxt->vctxt, ctxt->myDoc, dup, ret);
        } else if ((prefix == ctxt->str_xml) &&
	           (localname[0] == 'i') && (localname[1] == 'd') &&
		   (localname[2] == 0)) {
	    /*
	     * Add the xml:id value
	     *
	     * Open issue: normalization of the value.
	     */
	    if (dup == NULL)
	        dup = xmlStrndup(value, valueend - value);
	    if (xmlValidateNCName(dup, 1) != 0) {
	        xmlErrValid(ctxt, XML_DTD_XMLID_VALUE,
		      "xml:id : attribute value %s is not an NCName\n",
			    (const char *) dup, NULL);
	    }
	    xmlAddID(&ctxt->vctxt, ctxt->myDoc, dup, ret);
	}
    }
    if (dup != NULL)
	xmlFree(dup);
}

/**
 * xmlSAX2StartElementNs:
 * @ctx:  the user data (XML parser context)
 * @localname:  the local name of the element
 * @prefix:  the element namespace prefix if available
 * @URI:  the element namespace name if available
 * @nb_namespaces:  number of namespace definitions on that node
 * @namespaces:  pointer to the array of prefix/URI pairs namespace definitions
 * @nb_attributes:  the number of attributes on that node
 * @nb_defaulted:  the number of defaulted attributes.
 * @attributes:  pointer to the array of (localname/prefix/URI/value/end)
 *               attribute values.
 *
 * SAX2 callback when an element start has been detected by the parser.
 * It provides the namespace informations for the element, as well as
 * the new namespace declarations on the element.
 */
void
xmlSAX2StartElementNs(void *ctx,
                      const xmlChar *localname,
		      const xmlChar *prefix,
		      const xmlChar *URI,
		      int nb_namespaces,
		      const xmlChar **namespaces,
		      int nb_attributes,
		      int nb_defaulted,
		      const xmlChar **attributes)
{
    xmlParserCtxtPtr ctxt = (xmlParserCtxtPtr) ctx;
    xmlNodePtr ret;
    xmlNodePtr parent = ctxt->node;
    xmlNsPtr last = NULL, ns;
    const xmlChar *uri, *pref;
    int i, j;

    /*
     * First check on validity:
     */
    if (ctxt->validate && (ctxt->myDoc->extSubset == NULL) && 
        ((ctxt->myDoc->intSubset == NULL) ||
	 ((ctxt->myDoc->intSubset->notations == NULL) && 
	  (ctxt->myDoc->intSubset->elements == NULL) &&
	  (ctxt->myDoc->intSubset->attributes == NULL) && 
	  (ctxt->myDoc->intSubset->entities == NULL)))) {
	xmlErrValid(ctxt, XML_ERR_NO_DTD,
	  "Validation failed: no DTD found !", NULL, NULL);
	ctxt->validate = 0;
    }

    /*
     * allocate the node
     */
    if (ctxt->freeElems != NULL) {
        ret = ctxt->freeElems;
	ctxt->freeElems = ret->next;
	ctxt->freeElemsNr--;
	memset(ret, 0, sizeof(xmlNode));
	ret->type = XML_ELEMENT_NODE;

	if (ctxt->dictNames)
	    ret->name = localname;
	else
	    ret->name = xmlStrdup(localname);

	if ((__xmlRegisterCallbacks) && (xmlRegisterNodeDefaultValue))
	    xmlRegisterNodeDefaultValue(ret);
    } else {
	if (ctxt->dictNames)
	    ret = xmlNewDocNodeEatName(ctxt->myDoc, NULL, 
	                               (xmlChar *) localname, NULL);
	else
	    ret = xmlNewDocNode(ctxt->myDoc, NULL, localname, NULL);
	if (ret == NULL) {
	    ctxt->errNo = XML_ERR_NO_MEMORY;
	    ctxt->instate = XML_PARSER_EOF;
	    ctxt->disableSAX = 1;
	    return;
	}
    }
    if (ctxt->linenumbers) {
	if (ctxt->input != NULL) {
	    if (ctxt->input->line < 65535)
		ret->line = (short) ctxt->input->line;
	    else
	        ret->line = 65535;
	}
    }

    if (ctxt->myDoc->children == NULL) {
        xmlAddChild((xmlNodePtr) ctxt->myDoc, (xmlNodePtr) ret);
    } else if (parent == NULL) {
        parent = ctxt->myDoc->children;
    }
    /*
     * Build the namespace list
     */
    for (i = 0,j = 0;j < nb_namespaces;j++) {
        pref = namespaces[i++];
	uri = namespaces[i++];
	ns = xmlNewNs(NULL, uri, pref);
	if (ns != NULL) {
	    if (last == NULL) {
	        ret->nsDef = last = ns;
	    } else {
	        last->next = ns;
		last = ns;
	    }
	    if ((URI != NULL) && (prefix == pref))
		ret->ns = ns;
	} else {
	    ctxt->errNo = XML_ERR_NO_MEMORY;
	    ctxt->instate = XML_PARSER_EOF;
	    ctxt->disableSAX = 1;
	    return;
	}
#ifdef LIBXML_VALID_ENABLED
	if ((!ctxt->html) && ctxt->validate && ctxt->wellFormed &&
	    ctxt->myDoc && ctxt->myDoc->intSubset) {
	    ctxt->valid &= xmlValidateOneNamespace(&ctxt->vctxt, ctxt->myDoc,
	                                           ret, prefix, ns, uri);
	}
#endif /* LIBXML_VALID_ENABLED */
    }
    ctxt->nodemem = -1;

    /*
     * We are parsing a new node.
     */
    nodePush(ctxt, ret);

    /*
     * Link the child element
     */
    if (parent != NULL) {
        if (parent->type == XML_ELEMENT_NODE) {
	    xmlAddChild(parent, ret);
	} else {
	    xmlAddSibling(parent, ret);
	}
    }

    /*
     * Insert the defaulted attributes from the DTD only if requested:
     */
    if ((nb_defaulted != 0) &&
        ((ctxt->loadsubset & XML_COMPLETE_ATTRS) == 0))
	nb_attributes -= nb_defaulted;

    /*
     * Search the namespace if it wasn't already found
     */
    if ((URI != NULL) && (ret->ns == NULL)) {
	ret->ns = xmlSearchNs(ctxt->myDoc, parent, prefix);
	if (ret->ns == NULL) {
	    ns = xmlNewNs(ret, NULL, prefix);
	    if ((ctxt->sax != NULL) && (ctxt->sax->warning != NULL))
		ctxt->sax->warning(ctxt->userData, 
		     "Namespace prefix %s was not found\n", prefix);
	}
    }

    /*
     * process all the other attributes
     */
    if (nb_attributes > 0) {
        for (j = 0,i = 0;i < nb_attributes;i++,j+=5) {
	    xmlSAX2AttributeNs(ctxt, attributes[j], attributes[j+1],
	                       attributes[j+3], attributes[j+4]);
	}
    }

#ifdef LIBXML_VALID_ENABLED
    /*
     * If it's the Document root, finish the DTD validation and
     * check the document root element for validity
     */
    if ((ctxt->validate) && (ctxt->vctxt.finishDtd == 0)) {
	int chk;

	chk = xmlValidateDtdFinal(&ctxt->vctxt, ctxt->myDoc);
	if (chk <= 0)
	    ctxt->valid = 0;
	if (chk < 0)
	    ctxt->wellFormed = 0;
	ctxt->valid &= xmlValidateRoot(&ctxt->vctxt, ctxt->myDoc);
	ctxt->vctxt.finishDtd = 1;
    }
#endif /* LIBXML_VALID_ENABLED */
}

/**
 * xmlSAX2EndElementNs:
 * @ctx:  the user data (XML parser context)
 * @localname:  the local name of the element
 * @prefix:  the element namespace prefix if available
 * @URI:  the element namespace name if available
 *
 * SAX2 callback when an element end has been detected by the parser.
 * It provides the namespace informations for the element.
 */
void
xmlSAX2EndElementNs(void *ctx,
                    const xmlChar * localname ATTRIBUTE_UNUSED,
                    const xmlChar * prefix ATTRIBUTE_UNUSED,
		    const xmlChar * URI ATTRIBUTE_UNUSED)
{
    xmlParserCtxtPtr ctxt = (xmlParserCtxtPtr) ctx;
    xmlParserNodeInfo node_info;
    xmlNodePtr cur = ctxt->node;

    /* Capture end position and add node */
    if ((ctxt->record_info) && (cur != NULL)) {
        node_info.end_pos = ctxt->input->cur - ctxt->input->base;
        node_info.end_line = ctxt->input->line;
        node_info.node = cur;
        xmlParserAddNodeInfo(ctxt, &node_info);
    }
    ctxt->nodemem = -1;

#ifdef LIBXML_VALID_ENABLED
    if (ctxt->validate && ctxt->wellFormed &&
        ctxt->myDoc && ctxt->myDoc->intSubset)
        ctxt->valid &= xmlValidateOneElement(&ctxt->vctxt, ctxt->myDoc, cur);
#endif /* LIBXML_VALID_ENABLED */

    /*
     * end of parsing of this node.
     */
    nodePop(ctxt);
}

/**
 * xmlSAX2Reference:
 * @ctx: the user data (XML parser context)
 * @name:  The entity name
 *
 * called when an entity xmlSAX2Reference is detected. 
 */
void
xmlSAX2Reference(void *ctx, const xmlChar *name)
{
    xmlParserCtxtPtr ctxt = (xmlParserCtxtPtr) ctx;
    xmlNodePtr ret;

#ifdef DEBUG_SAX
    xmlGenericError(xmlGenericErrorContext,
	    "SAX.xmlSAX2Reference(%s)\n", name);
#endif
    if (name[0] == '#')
	ret = xmlNewCharRef(ctxt->myDoc, name);
    else
	ret = xmlNewReference(ctxt->myDoc, name);
#ifdef DEBUG_SAX_TREE
    xmlGenericError(xmlGenericErrorContext,
	    "add xmlSAX2Reference %s to %s \n", name, ctxt->node->name);
#endif
    xmlAddChild(ctxt->node, ret);
}

/**
 * xmlSAX2Characters:
 * @ctx: the user data (XML parser context)
 * @ch:  a xmlChar string
 * @len: the number of xmlChar
 *
 * receiving some chars from the parser.
 */
void
xmlSAX2Characters(void *ctx, const xmlChar *ch, int len)
{
    xmlParserCtxtPtr ctxt = (xmlParserCtxtPtr) ctx;
    xmlNodePtr lastChild;

#ifdef DEBUG_SAX
    xmlGenericError(xmlGenericErrorContext,
	    "SAX.xmlSAX2Characters(%.30s, %d)\n", ch, len);
#endif
    /*
     * Handle the data if any. If there is no child
     * add it as content, otherwise if the last child is text,
     * concatenate it, else create a new node of type text.
     */

    if (ctxt->node == NULL) {
#ifdef DEBUG_SAX_TREE
	xmlGenericError(xmlGenericErrorContext,
		"add chars: ctxt->node == NULL !\n");
#endif
        return;
    }
    lastChild = ctxt->node->last;
#ifdef DEBUG_SAX_TREE
    xmlGenericError(xmlGenericErrorContext,
	    "add chars to %s \n", ctxt->node->name);
#endif

    /*
     * Here we needed an accelerator mechanism in case of very large
     * elements. Use an attribute in the structure !!!
     */
    if (lastChild == NULL) {
        lastChild = xmlSAX2TextNode(ctxt, ch, len);
	if (lastChild != NULL) {
	    ctxt->node->children = lastChild;
	    ctxt->node->last = lastChild;
	    lastChild->parent = ctxt->node;
	    lastChild->doc = ctxt->node->doc;
	    ctxt->nodelen = len;
	    ctxt->nodemem = len + 1;
	}
    } else {
	int coalesceText = (lastChild != NULL) &&
	    (lastChild->type == XML_TEXT_NODE) &&
	    (lastChild->name == xmlStringText);
	if ((coalesceText) && (ctxt->nodemem != 0)) {
	    /*
	     * The whole point of maintaining nodelen and nodemem,
	     * xmlTextConcat is too costly, i.e. compute length,
	     * reallocate a new buffer, move data, append ch. Here
	     * We try to minimaze realloc() uses and avoid copying
	     * and recomputing length over and over.
	     */
	    if ((ctxt->nodemem == ctxt->nodelen + 1) &&
	        (xmlDictOwns(ctxt->dict, lastChild->content))) {
		lastChild->content = xmlStrdup(lastChild->content);
	    }
	    if (ctxt->nodelen + len >= ctxt->nodemem) {
		xmlChar *newbuf;
		int size;

		size = ctxt->nodemem + len;
		size *= 2;
                newbuf = (xmlChar *) xmlRealloc(lastChild->content,size);
		if (newbuf == NULL) {
		    if ((ctxt->sax != NULL) && (ctxt->sax->error != NULL))
			ctxt->sax->error(ctxt->userData, 
			     "SAX.xmlSAX2Characters(): out of memory\n");
		    ctxt->errNo = XML_ERR_NO_MEMORY;
		    ctxt->instate = XML_PARSER_EOF;
		    ctxt->disableSAX = 1;
		    return;
		}
		ctxt->nodemem = size;
		lastChild->content = newbuf;
	    }
	    m