c14n.c

xml开源解析代码.版本为libxml2-2.6.29,可支持GB3212.网络消息发送XML时很有用.

 * Canonical XML applied to a document subset requires the search of the 
 * ancestor nodes of each orphan element node for attributes in the xml 
 * namespace, such as xml:lang and xml:space. These are copied into the 
 * element node except if a declaration of the same attribute is already 
 * in the attribute axis of the element (whether or not it is included in 
 * the document subset). This search and copying are omitted from the 
 * Exclusive XML Canonicalization method.
 *
 * Returns 0 on success or -1 on fail.
 */
static int
xmlC14NProcessAttrsAxis(xmlC14NCtxPtr ctx, xmlNodePtr cur, int parent_visible)
{
    xmlAttrPtr attr;
    xmlListPtr list;

    if ((ctx == NULL) || (cur == NULL) || (cur->type != XML_ELEMENT_NODE)) {
        xmlC14NErrParam("processing attributes axis");
        return (-1);
    }

    /*
     * Create a sorted list to store element attributes
     */
    list = xmlListCreate(NULL, (xmlListDataCompare) xmlC14NAttrsCompare);
    if (list == NULL) {
        xmlC14NErrInternal("creating attributes list");
        return (-1);
    }

    /* 
     * Add all visible attributes from current node. 
     */
    attr = cur->properties;
    while (attr != NULL) {
        /* check that attribute is visible */
        if (xmlC14NIsVisible(ctx, attr, cur)) {
            xmlListInsert(list, attr);
        }
        attr = attr->next;
    }

    /* 
     * include attributes in "xml" namespace defined in ancestors
     * (only for non-exclusive XML Canonicalization)
     */
    if (parent_visible && (!ctx->exclusive) && (cur->parent != NULL)
        && (!xmlC14NIsVisible(ctx, cur->parent, cur->parent->parent))) {
        /*
         * If XPath node-set is not specified then the parent is always 
         * visible!
         */
        cur = cur->parent;
        while (cur != NULL) {
            attr = cur->properties;
            while (attr != NULL) {
                if ((attr->ns != NULL)
                    && (xmlStrEqual(attr->ns->prefix, BAD_CAST "xml"))) {
                    if (xmlListSearch(list, attr) == NULL) {
                        xmlListInsert(list, attr);
                    }
                }
                attr = attr->next;
            }
            cur = cur->parent;
        }
    }

    /* 
     * print out all elements from list 
     */
    xmlListWalk(list, (xmlListWalker) xmlC14NPrintAttrs, (const void *) ctx);

    /* 
     * Cleanup
     */
    xmlListDelete(list);
    return (0);
}

/** 
 * xmlC14NCheckForRelativeNamespaces:
 * @ctx:		the C14N context
 * @cur:		the current element node
 *
 * Checks that current element node has no relative namespaces defined
 *
 * Returns 0 if the node has no relative namespaces or -1 otherwise.
 */
static int
xmlC14NCheckForRelativeNamespaces(xmlC14NCtxPtr ctx, xmlNodePtr cur)
{
    xmlNsPtr ns;

    if ((ctx == NULL) || (cur == NULL) || (cur->type != XML_ELEMENT_NODE)) {
        xmlC14NErrParam("checking for relative namespaces");
        return (-1);
    }

    ns = cur->nsDef;
    while (ns != NULL) {
        if (xmlStrlen(ns->href) > 0) {
            xmlURIPtr uri;

            uri = xmlParseURI((const char *) ns->href);
            if (uri == NULL) {
                xmlC14NErrInternal("parsing namespace uri");
                return (-1);
            }
            if (xmlStrlen((const xmlChar *) uri->scheme) == 0) {
                xmlC14NErrRelativeNamespace(uri->scheme);
                xmlFreeURI(uri);
                return (-1);
            }
            if ((xmlStrcasecmp((const xmlChar *) uri->scheme, BAD_CAST "urn") != 0)
                && (xmlStrcasecmp((const xmlChar *) uri->scheme, BAD_CAST "dav") !=0)
                && (xmlStrlen((const xmlChar *) uri->server) == 0)) {
                xmlC14NErrRelativeNamespace(uri->scheme);
                xmlFreeURI(uri);
                return (-1);
            }
            xmlFreeURI(uri);
        }
        ns = ns->next;
    }
    return (0);
}

/**
 * xmlC14NProcessElementNode:
 * @ctx: 		the pointer to C14N context object
 * @cur:		the node to process
 *  		
 * Canonical XML v 1.0 (http://www.w3.org/TR/xml-c14n)
 *
 * Element Nodes
 * If the element is not in the node-set, then the result is obtained 
 * by processing the namespace axis, then the attribute axis, then 
 * processing the child nodes of the element that are in the node-set 
 * (in document order). If the element is in the node-set, then the result 
 * is an open angle bracket (<), the element QName, the result of 
 * processing the namespace axis, the result of processing the attribute 
 * axis, a close angle bracket (>), the result of processing the child 
 * nodes of the element that are in the node-set (in document order), an 
 * open angle bracket, a forward slash (/), the element QName, and a close 
 * angle bracket.
 *
 * Returns non-negative value on success or negative value on fail
 */
static int
xmlC14NProcessElementNode(xmlC14NCtxPtr ctx, xmlNodePtr cur, int visible)
{
    int ret;
    xmlC14NVisibleNsStack state;
    int parent_is_doc = 0;

    if ((ctx == NULL) || (cur == NULL) || (cur->type != XML_ELEMENT_NODE)) {
        xmlC14NErrParam("processing element node");
        return (-1);
    }

    /* 
     * Check relative relative namespaces:
     * implementations of XML canonicalization MUST report an operation
     * failure on documents containing relative namespace URIs.
     */
    if (xmlC14NCheckForRelativeNamespaces(ctx, cur) < 0) {
        xmlC14NErrInternal("checking for relative namespaces");
        return (-1);
    }


    /* 
     * Save ns_rendered stack position
     */
    memset(&state, 0, sizeof(state));
    xmlC14NVisibleNsStackSave(ctx->ns_rendered, &state);

    if (visible) {	
        if (ctx->parent_is_doc) {
	    /* save this flag into the stack */
	    parent_is_doc = ctx->parent_is_doc;
	    ctx->parent_is_doc = 0;
            ctx->pos = XMLC14N_INSIDE_DOCUMENT_ELEMENT;
        }
        xmlOutputBufferWriteString(ctx->buf, "<");

        if ((cur->ns != NULL) && (xmlStrlen(cur->ns->prefix) > 0)) {
            xmlOutputBufferWriteString(ctx->buf,
                                       (const char *) cur->ns->prefix);
            xmlOutputBufferWriteString(ctx->buf, ":");
        }
        xmlOutputBufferWriteString(ctx->buf, (const char *) cur->name);
    }

    if (!ctx->exclusive) {
        ret = xmlC14NProcessNamespacesAxis(ctx, cur, visible);
    } else {
        ret = xmlExcC14NProcessNamespacesAxis(ctx, cur, visible);
    }
    if (ret < 0) {
        xmlC14NErrInternal("processing namespaces axis");
        return (-1);
    }
    /* todo: shouldn't this go to "visible only"? */
    if(visible) {
	xmlC14NVisibleNsStackShift(ctx->ns_rendered);
    }
    
    ret = xmlC14NProcessAttrsAxis(ctx, cur, visible);
    if (ret < 0) {
	xmlC14NErrInternal("processing attributes axis");
    	return (-1);
    }

    if (visible) { 
        xmlOutputBufferWriteString(ctx->buf, ">");
    }
    if (cur->children != NULL) {
        ret = xmlC14NProcessNodeList(ctx, cur->children);
        if (ret < 0) {
            xmlC14NErrInternal("processing childrens list");
            return (-1);
        }
    }
    if (visible) {
        xmlOutputBufferWriteString(ctx->buf, "</");
        if ((cur->ns != NULL) && (xmlStrlen(cur->ns->prefix) > 0)) {
            xmlOutputBufferWriteString(ctx->buf,
                                       (const char *) cur->ns->prefix);
            xmlOutputBufferWriteString(ctx->buf, ":");
        }
        xmlOutputBufferWriteString(ctx->buf, (const char *) cur->name);
        xmlOutputBufferWriteString(ctx->buf, ">");
        if (parent_is_doc) {
	    /* restore this flag from the stack for next node */
            ctx->parent_is_doc = parent_is_doc;
	    ctx->pos = XMLC14N_AFTER_DOCUMENT_ELEMENT;
        }
    }

    /* 
     * Restore ns_rendered stack position
     */
    xmlC14NVisibleNsStackRestore(ctx->ns_rendered, &state);
    return (0);
}

/**
 * xmlC14NProcessNode:
 * @ctx: 		the pointer to C14N context object
 * @cur:		the node to process
 *  		
 * Processes the given node
 *
 * Returns non-negative value on success or negative value on fail
 */
static int
xmlC14NProcessNode(xmlC14NCtxPtr ctx, xmlNodePtr cur)
{
    int ret = 0;
    int visible;

    if ((ctx == NULL) || (cur == NULL)) {
        xmlC14NErrParam("processing node");
        return (-1);
    }

    visible = xmlC14NIsVisible(ctx, cur, cur->parent);
    switch (cur->type) {
        case XML_ELEMENT_NODE:
            ret = xmlC14NProcessElementNode(ctx, cur, visible);
            break;
        case XML_CDATA_SECTION_NODE:
        case XML_TEXT_NODE:
            /*
             * Text Nodes
             * the string value, except all ampersands are replaced 
             * by &amp;, all open angle brackets (<) are replaced by &lt;, all closing 
             * angle brackets (>) are replaced by &gt;, and all #xD characters are 
             * replaced by &#xD;.
             */
            /* cdata sections are processed as text nodes */
            /* todo: verify that cdata sections are included in XPath nodes set */
            if ((visible) && (cur->content != NULL)) {
                xmlChar *buffer;

                buffer = xmlC11NNormalizeText(cur->content);
                if (buffer != NULL) {
                    xmlOutputBufferWriteString(ctx->buf,
                                               (const char *) buffer);
                    xmlFree(buffer);
                } else {
                    xmlC14NErrInternal("normalizing text node");
                    return (-1);
                }
            }
            break;
        case XML_PI_NODE:
            /* 
             * Processing Instruction (PI) Nodes- 
             * The opening PI symbol (<?), the PI target name of the node, 
             * a leading space and the string value if it is not empty, and 
             * the closing PI symbol (?>). If the string value is empty, 
             * then the leading space is not added. Also, a trailing #xA is 
             * rendered after the closing PI symbol for PI children of the 
             * root node with a lesser document order than the document 
             * element, and a leading #xA is rendered before the opening PI 
             * symbol of PI children of the root node with a greater document 
             * order than the document element.
             */
            if (visible) {
                if (ctx->pos == XMLC14N_AFTER_DOCUMENT_ELEMENT) {
                    xmlOutputBufferWriteString(ctx->buf, "\x0A<?");
                } else {
                    xmlOutputBufferWriteString(ctx->buf, "<?");
                }

                xmlOutputBufferWriteString(ctx->buf,
                                           (const char *) cur->name);
                if ((cur->content != NULL) && (*(cur->content) != '\0')) {
                    xmlChar *buffer;

                    xmlOutputBufferWriteString(ctx->buf, " ");

                    /* todo: do we need to normalize pi? */
                    buffer = xmlC11NNormalizePI(cur->content);
                    if (buffer != NULL) {
                        xmlOutputBufferWriteString(ctx->buf,
                                                   (const char *) buffer);
                        xmlFree(buffer);
                    } else {
                        xmlC14NErrInternal("normalizing pi node");
                        return (-1);
                    }
                }

                if (ctx->pos == XMLC14N_BEFORE_DOCUMENT_ELEMENT) {
                    xmlOutputBufferWriteString(ctx->buf, "?>\x0A");
                } else {
                    xmlOutputBufferWriteString(ctx->buf, "?>");
                }
            }
            break;
        case XML_COMMENT_NODE:
            /*
             * Comment Nodes
             * Nothing if generating canonical XML without  comments. For 
             * canonical XML with comments, generate the opening comment 
             * symbol (<!--), the string value of the node, and the 
             * closing comment symbol (-->). Also, a trailing #xA is rendered 
             * after the closing comment symbol for comment children of the 
             * root node with a lesser document order than the document 
             * element, and a leading #xA is rendered before the opening 
             * comment symbol of comment children of the root node with a 
             * greater document order than the document element. (Comment 
             * children of the root node represent comments outside of the 
             * top-level document element and outside of the document type 
             * declaration).
             */
            if (visible && ctx->with_comments) {
                if (ctx->pos == XMLC14N_AFTER_DOCUMENT_ELEMENT) {
                    xmlOutputBufferWriteString(ctx->buf, "\x0A<!--");
                } else {
                    xmlOutputBufferWriteString(ctx->buf, "<!--");
                }

                if (cur->content != NULL) {
                    xmlChar *buffer;

                    /* todo: do we need to normalize comment? */
                    buffer = xmlC11NNormalizeComment(cur->content);
                    if (buffer != NULL) {
                        xmlOutputBufferWriteString(ctx->buf,
                                                   (const char *) buffer);
                        xmlFree(buffer);
                    } else {
                        xmlC14NErrInternal("normalizing comment node");
                        return (-1);
                    }
                }

                if (ctx->pos == XMLC14N_BEFORE_DOCUMENT_ELEMENT) {
                    xmlOutputBufferWriteString(ctx->buf, "-->\x0A");
                } else {
                    xmlOutputBufferWriteString(ctx->buf, "-->");
                }
            }
            break;
        case XML_DOCUMENT_NODE:
        case XML_DOCUMENT_FRAG_NODE:   /* should be processed as document? */
#ifdef LIBXML_DOCB_ENABLED
        case XML_DOCB_DOCUMENT_NODE:   /* should be processed as document? */
#endif
#ifdef LIBXML_HTML_ENABLED
        case XML_HTML_DOCUMENT_NODE:   /* should be processed as document? */
#endif
            if (cur->children != NULL) {
                ctx->pos = XMLC14N_BEFORE_DOCUMENT_ELEMENT;
                ctx->parent_is_doc = 1;
                ret = xmlC14NProcessNodeList(ctx, cur->children);
            }
            break;

        case XML_ATTRIBUTE_NODE:
            xmlC14NErrInvalidNode("XML_ATTRIBUTE_NODE", "processing node");
            return (-1);
        case XML_NAMESPACE_DECL:
            xmlC14NErrInvalidNode("XML_NAMESPACE_DECL", "processing node");
            return (-1);
        case XML_ENTITY_REF_NODE:
            xmlC14NErrInvalidNode("XML_ENTITY_REF_NODE", "processing node");
            return (-1);
        case XML_ENTITY_NODE:
            xmlC14NErrInvalidNode("XML_ENTITY_NODE", "processing node");
            return (-1);

        case XML_DOCUMENT_TYPE_NODE:
        case XML_NOTATION_NODE:
        case XML_DTD_NODE:
        case XML_ELEMENT_DECL:
        case XML_ATTRIBUTE_DECL:
        case XML_ENTITY_DECL:
#ifdef LIBXML_XINCLUDE_ENABLED
        case XML_XINCLUDE_START:
        case XML_XINCLUDE_END:
#endif
            /* 
             * should be ignored according to "W3C Canonical XML" 
             */
            break;
        default:
            xmlC14NErrUnknownNode(cur->type, "processing node");
            return (-1);
    }

    return (ret);
}

/**
 * xmlC14NProcessNodeList:
 * @ctx: 		the pointer to C14N context object
 * @cur:		the node to start from
 *  		
 * Processes all nodes in the row starting from cur.
 *
 * Returns non-negative value on success or negative value on fail
 */
static int
xmlC14NProcessNodeList(xmlC14NCtxPtr ctx, xmlNodePtr cur)
{
    int ret;

    if (ctx == NULL) {
        xmlC14NErrParam("processing node list");
        return (-1);
    }

    for (ret = 0; cur != NULL && ret >= 0; cur = cur->next) {
        ret = xmlC14NProcessNode(ctx, cur);
    }
    return (ret);
}


/**
 * xmlC14NFreeCtx:
 * @ctx: the pointer to C14N context object
 * 		
 * Cleanups the C14N context object.
 */

static void
xmlC14NFreeCtx(xmlC14NCtxPtr ctx)
{
    if (ctx == NULL) {
        xmlC14NErrParam("freeing context");
        return;
    }

    if (ctx->ns_rendered != NULL) {
        xmlC14NVisibleNsStackDestroy(ctx->ns_rendered);
    }