/** @todo Rewrite this function (use wbxml_tree_node_* functions) */
WBXML_DECLARE(WBXMLTreeNode *) wbxml_tree_add_xml_elt(WBXMLTree *tree,
WBXMLTreeNode *parent,
WB_UTINY *name)
{
const WBXMLTagEntry *tag_entry = NULL;
WBXMLTreeNode *node = NULL;
WBXMLTag *tag = NULL;
WB_UTINY *sep = NULL;
const WB_UTINY *namespace_name = NULL;
const WB_UTINY *element_name = NULL;
/* Separate the namespace from the element name */
sep = (WB_UTINY *)strrchr((const WB_TINY *) name, WBXML_NAMESPACE_SEPARATOR);
if (sep != NULL) {
/* Temporarily split the string by changing the separater to a null-terminator */
*sep = '\0';
namespace_name = name;
element_name = sep+1;
}
else {
/* No namespace, so just set it to an empty string (specifically, the null-terminator at the end of the elemet name */
namespace_name = name + strlen((const WB_TINY *) name);
element_name = name;
}
WBXML_DEBUG((WBXML_CONV, "Parsed element name: Namespace='%s', Element='%s'", namespace_name, element_name));
/* Update the current code page to match the one specified by the namespace */
tree->cur_code_page = wbxml_tables_get_code_page(tree->lang->nsTable, (const WB_TINY *) namespace_name);
/* Search for XML Tag Name in Table */
if ((tag_entry = wbxml_tables_get_tag_from_xml(tree->lang, tree->cur_code_page, element_name)) != NULL) {
tree->cur_code_page = tag_entry->wbxmlCodePage;
/* Found : token tag */
tag = wbxml_tag_create_token(tag_entry);
}
else {
/* Not found : literal tag */
tag = wbxml_tag_create_literal(name);
}
if (sep != NULL) {
/* We are done with the element and namespace names, so put the separator character back */
*sep = WBXML_NAMESPACE_SEPARATOR;
}
if (tag == NULL)
return NULL;
/* Create a new Node */
if ((node = wbxml_tree_node_create(WBXML_TREE_ELEMENT_NODE)) == NULL) {
wbxml_tag_destroy(tag);
return NULL;
}
/* Set Node Tag */
node->name = tag;
/* Add this Node to Tree */
if (!wbxml_tree_add_node(tree, parent, node)) {
wbxml_tree_node_destroy(node);
return NULL;
}
return node;
}
void wbxml_tree_clb_xml_end_element(void *ctx,
const XML_Char *localName)
{
WBXMLTreeClbCtx *tree_ctx = (WBXMLTreeClbCtx *) ctx;
WBXMLBuffer *content = NULL;
WBXMLTreeNode *node = NULL;
WBXMLError ret = WBXML_OK;
WBXML_DEBUG((WBXML_PARSER, "Expat element end callback ('%s')", localName));
/* If the node is flagged as binary node
* then the data is base64 encoded in the XML document
* and the data must be decoded in one step.
* Examples: Microsoft ActiveSync tags ConversationId or MIME
*/
node = tree_ctx->current;
if (node && node->type == WBXML_TREE_ELEMENT_NODE &&
node->name->type == WBXML_VALUE_TOKEN &&
node->name->u.token->options & WBXML_TAG_OPTION_BINARY)
{
if (node->content == NULL)
{
WBXML_DEBUG((WBXML_PARSER, " Binary tag: No content => no conversion!"));
} else {
WBXML_DEBUG((WBXML_PARSER, " Binary tag: Convert base64 data"));
ret = wbxml_buffer_decode_base64(node->content);
if (ret != WBXML_OK)
{
WBXML_DEBUG((WBXML_PARSER, " Binary tag: Base64 decoder failed!"));
tree_ctx->error = ret;
} else {
/* Add the buffer as a regular string node (since libwbxml doesn't
* offer a way to specify an opaque data node). The WBXML
* encoder is responsible for generating correct opaque data for
* nodes like this.
*/
if (wbxml_tree_add_text(tree_ctx->tree,
tree_ctx->current,
(const WB_UTINY*)wbxml_buffer_get_cstr(node->content),
wbxml_buffer_len(node->content)) == NULL)
{
WBXML_DEBUG((WBXML_PARSER, " Binary tag: Cannot add base64 decoded node!"));
tree_ctx->error = WBXML_ERROR_INTERNAL;
}
}
/* safe cleanup */
content = node->content;
node->content = NULL;
wbxml_buffer_destroy(content);
}
}
if (tree_ctx->expat_utf16) {
/** @todo Convert from UTF-16 to UTF-8 */
}
/* Check for Error */
if (tree_ctx->error != WBXML_OK)
return;
/* Are we skipping a whole node ? */
if (tree_ctx->skip_lvl > 0) {
if (tree_ctx->skip_lvl == 1)
{
/* End of skipped node */
#if defined( WBXML_SUPPORT_SYNCML )
if (WBXML_STRCMP(localName, "syncml:devinf:DevInf") == 0 ||
WBXML_STRCMP(localName, "syncml:dmddf1.2:MgmtTree") == 0) {
/* definitions first ... or some compilers don't like it */
WBXMLBuffer *embed_doc = NULL;
WBXMLTree *tree = NULL;
const WBXMLLangEntry *lang;
/* Get embedded DevInf or DM DDF Document */
embed_doc = wbxml_buffer_create(tree_ctx->input_buff + tree_ctx->skip_start,
XML_GetCurrentByteIndex(tree_ctx->xml_parser) - tree_ctx->skip_start,
XML_GetCurrentByteIndex(tree_ctx->xml_parser) - tree_ctx->skip_start + 10);
if (embed_doc == NULL) {
tree_ctx->error = WBXML_ERROR_NOT_ENOUGH_MEMORY;
wbxml_buffer_destroy(embed_doc);
return;
}
if (tree_ctx->expat_utf16) {
/** @todo Convert from UTF-16 to UTF-8 */
}
/* Check Buffer Creation and add the closing tag */
if ((WBXML_STRCMP(localName, "syncml:devinf:DevInf") == 0 &&
(!wbxml_buffer_append_cstr(embed_doc, "</DevInf>")))
||
(WBXML_STRCMP(localName, "syncml:dmddf1.2:MgmtTree") == 0 &&
(!wbxml_buffer_append_cstr(embed_doc, "</MgmtTree>"))))
{
tree_ctx->error = WBXML_ERROR_NOT_ENOUGH_MEMORY;
wbxml_buffer_destroy(embed_doc);
return;
}
/* Add doctype to give the XML parser a chance */
if (WBXML_STRCMP(localName, "syncml:dmddf1.2:MgmtTree") == 0 &&
tree_ctx->tree->lang->langID != WBXML_LANG_SYNCML_SYNCML12)
{
tree_ctx->error = WBXML_ERROR_UNKNOWN_XML_LANGUAGE;
wbxml_buffer_destroy(embed_doc);
return;
}
switch(tree_ctx->tree->lang->langID)
{
case WBXML_LANG_SYNCML_SYNCML10:
lang = wbxml_tables_get_table(WBXML_LANG_SYNCML_DEVINF10);
break;
case WBXML_LANG_SYNCML_SYNCML11:
lang = wbxml_tables_get_table(WBXML_LANG_SYNCML_DEVINF11);
break;
case WBXML_LANG_SYNCML_SYNCML12:
if (WBXML_STRCMP(localName, "syncml:dmddf1.2:MgmtTree") == 0) {
lang = wbxml_tables_get_table(WBXML_LANG_SYNCML_DMDDF12);
} else {
lang = wbxml_tables_get_table(WBXML_LANG_SYNCML_DEVINF12);
}
break;
default:
tree_ctx->error = WBXML_ERROR_UNKNOWN_XML_LANGUAGE;
wbxml_buffer_destroy(embed_doc);
return;
}
assert (lang!= NULL);
if (lang == NULL) {
tree_ctx->error = WBXML_ERROR_UNKNOWN_XML_LANGUAGE;
wbxml_buffer_destroy(embed_doc);
return;
}
/* DOCTYPE in reverse order */
if (!wbxml_buffer_insert_cstr(embed_doc,(WB_UTINY *) "\">\n", 0) || /* > */
!wbxml_buffer_insert_cstr(embed_doc, (WB_UTINY *) lang->publicID->xmlDTD, 0) || /* DTD */
!wbxml_buffer_insert_cstr(embed_doc, (WB_UTINY *) "\" \"", 0) || /* DTD */
!wbxml_buffer_insert_cstr(embed_doc, (WB_UTINY *) lang->publicID->xmlPublicID, 0) || /* Public ID */
!wbxml_buffer_insert_cstr(embed_doc, (WB_UTINY *) " PUBLIC \"", 0) || /* PUBLIC " */
!wbxml_buffer_insert_cstr(embed_doc, (WB_UTINY *) lang->publicID->xmlRootElt, 0) || /* Root Element */
!wbxml_buffer_insert_cstr(embed_doc, (WB_UTINY *) "<!DOCTYPE ", 0)) /* <!DOCTYPE */
{
tree_ctx->error = WBXML_ERROR_ENCODER_APPEND_DATA;
wbxml_buffer_destroy(embed_doc);
return;
}
WBXML_DEBUG((WBXML_PARSER, "\t Embedded Doc : '%s'", wbxml_buffer_get_cstr(embed_doc)));
/* Parse 'DevInf' Document */
if ((ret = wbxml_tree_from_xml(wbxml_buffer_get_cstr(embed_doc),
wbxml_buffer_len(embed_doc),
&tree)) != WBXML_OK)
{
tree_ctx->error = ret;
wbxml_buffer_destroy(embed_doc);
return;
}
/* Add Tree Node */
tree_ctx->current = wbxml_tree_add_tree(tree_ctx->tree,
tree_ctx->current,
tree);
if (tree_ctx->current == NULL)
{
tree_ctx->error = WBXML_ERROR_INTERNAL;
wbxml_tree_destroy(tree);
wbxml_buffer_destroy(embed_doc);
return;
}
/* Clean-up */
wbxml_buffer_destroy(embed_doc);
tree_ctx->skip_lvl = 0;
}
#endif /* WBXML_SUPPORT_SYNCML */
}
else {
tree_ctx->skip_lvl--;
return;
}
}
if (tree_ctx->current == NULL) {
tree_ctx->error = WBXML_ERROR_INTERNAL;
return;
}
if (tree_ctx->current->parent == NULL) {
/* This must be the Root Element */
if (tree_ctx->current != tree_ctx->tree->root) {
tree_ctx->error = WBXML_ERROR_INTERNAL;
}
}
else {
#if defined ( WBXML_SUPPORT_SYNCML )
/* Have we added a missing CDATA section ?
* If so, we assume that now that we have reached an end of Element,
* the CDATA section ended, and so we go back to parent.
*/
if ((tree_ctx->current != NULL) && (tree_ctx->current->type == WBXML_TREE_CDATA_NODE))
tree_ctx->current = tree_ctx->current->parent;
#endif /* WBXML_SUPPORT_SYNCML */
/* Go back one step upper in the tree */
tree_ctx->current = tree_ctx->current->parent;
}
}
void wbxml_tree_clb_xml_characters(void *ctx,
const XML_Char *ch,
int len)
{
WBXMLTreeNode *node;
WBXMLTreeClbCtx *tree_ctx = (WBXMLTreeClbCtx *) ctx;
WBXML_DEBUG((WBXML_PARSER, "Expat text callback"));
if (tree_ctx->expat_utf16) {
/** @todo Convert from UTF-16 to UTF-8 */
}
/* Check for Error */
if (tree_ctx->error != WBXML_OK)
return;
/* Are we skipping a whole node ? */
if (tree_ctx->skip_lvl > 0)
return;
#if defined ( WBXML_SUPPORT_SYNCML )
/* Specific treatment for SyncML */
switch (wbxml_tree_node_get_syncml_data_type(tree_ctx->current)) {
case WBXML_SYNCML_DATA_TYPE_DIRECTORY_VCARD:
case WBXML_SYNCML_DATA_TYPE_VCALENDAR:
case WBXML_SYNCML_DATA_TYPE_VCARD:
case WBXML_SYNCML_DATA_TYPE_VOBJECT:
/* SyncML has some real design bugs
* because the authors of the specification did not understand XML.
*
* There must be a hack to preserve the CRLFs of vFormat objects.
* The only chance to do this is the detection of the vFormat itself
* and the conversion of every LF to a CRLF.
*
* The line breaks are always in a single text node.
* So a CR is appended to get a CRLF at the end.
*/
if (len == 1 && ch[0] == '\n') /* line break - LF */
{
ch = "\r\n";
len = 2;
}
/* Do not break here.
* The CDATA handling is required for vFormat objects too.
*/
case WBXML_SYNCML_DATA_TYPE_CLEAR:
/*
* Add a missing CDATA section node
*
* Example:
* <Add>
* <CmdID>6</CmdID>
* <Meta><Type xmlns='syncml:metinf'>text/x-vcard</Type></Meta>
* <Item>
* <Source>
* <LocURI>pas-id-3F4B790300000000</LocURI>
* </Source>
* <Data>BEGIN:VCARD
* VERSION:2.1
* X-EVOLUTION-FILE-AS:Ximian, Inc.
* N:
* LABEL;WORK;ENCODING=QUOTED-PRINTABLE:401 Park Drive 3 West=0ABoston, MA
* 02215=0AUSA
* TEL;WORK;VOICE:(617) 236-0442
* TEL;WORK;FAX:(617) 236-8630
* EMAIL;INTERNET:[EMAIL PROTECTED]
* URL:www.ximian.com/
* ORG:Ximian, Inc.
* NOTE:Welcome to the Ximian Addressbook.
* UID:pas-id-3F4B790300000000
* END:VCARD</Data>
* </Item>
* </Add>
*
* The end of CDATA section is assumed to be reached when parsing the end
* of </Data> element.
*
* This kind of document is erroneous, but we must handle it.
* Normally, this should be:
*
* ...
* <Data><!CDATA[[BEGIN:VCARD
* VERSION:2.1
* X-EVOLUTION-FILE-AS:Ximian, Inc.
* ...
* UID:pas-id-3F4B790300000000
* END:VCARD
* ]]></Data>
* ...
*/
/*
* We add a missing CDATA section if we are not already in a CDATA section.
*
* We don't add a CDATA section if we have already added a CDATA section. This
* permits to correctly handle good XML documents like this:
*
* ...
* <Data><!CDATA[[BEGIN:VCARD
* VERSION:2.1
* X-EVOLUTION-FILE-AS:Ximian, Inc.
* ...
* UID:pas-id-3F4B790300000000
* END:VCARD
* ]]>
* </Data>
* ...
*
* In this example, the spaces beetwen "]]>" and "</Data>" must not be added
* to a CDATA section.
*/
if ((tree_ctx->current != NULL) &&
(tree_ctx->current->type != WBXML_TREE_CDATA_NODE) &&
!((tree_ctx->current->children != NULL) &&
(tree_ctx->current->children->type == WBXML_TREE_CDATA_NODE)))
{
/* Add CDATA Node */
tree_ctx->current = wbxml_tree_add_cdata(tree_ctx->tree, tree_ctx->current);
if (tree_ctx->current == NULL) {
tree_ctx->error = WBXML_ERROR_INTERNAL;
return;
}
}
/* Now we can add the Text Node */
break;
default:
/* NOP */
break;
} /* switch */
#endif /* WBXML_SUPPORT_SYNCML */
/* We expect that "byte array" or BLOB types are
* encoded in Base 64 in the XML code, since they may contain binary data.
*/
node = tree_ctx->current;
if (node && node->type == WBXML_TREE_ELEMENT_NODE &&
node->name->type == WBXML_VALUE_TOKEN &&
node->name->u.token->options & WBXML_TAG_OPTION_BINARY)
{
WBXML_DEBUG((WBXML_PARSER, " Binary tag: Caching base64 encoded data for later conversion."));
if (node->content == NULL)
{
node->content = wbxml_buffer_create(ch, len, 1);
if (node->content == NULL)
tree_ctx->error = WBXML_ERROR_NOT_ENOUGH_MEMORY;
} else {
if (!wbxml_buffer_append_data(node->content, ch, len))
tree_ctx->error = WBXML_ERROR_NOT_ENOUGH_MEMORY;
}
return;
}
/* Add Text Node */
if (wbxml_tree_add_text(tree_ctx->tree,
tree_ctx->current,
(const WB_UTINY*) ch,
len) == NULL)
{
tree_ctx->error = WBXML_ERROR_INTERNAL;
}
}
void wbxml_tree_clb_xml_start_element(void *ctx,
const XML_Char *localName,
const XML_Char **attrs)
{
WBXMLTreeClbCtx *tree_ctx = (WBXMLTreeClbCtx *) ctx;
const WBXMLLangEntry *lang_table = NULL;
WBXML_DEBUG((WBXML_PARSER, "Expat element start callback ('%s')", localName));
if (tree_ctx->expat_utf16) {
/** @todo Convert from UTF-16 to UTF-8 */
}
/* Check for Error */
if (tree_ctx->error != WBXML_OK)
return;
/* Are we skipping a whole node ? */
if (tree_ctx->skip_lvl > 0) {
tree_ctx->skip_lvl++;
return;
}
if (tree_ctx->current == NULL) {
/* This is the Root Element */
if (tree_ctx->tree->lang == NULL) {
/* Language Table not already found: Search again */
lang_table = wbxml_tables_search_table(wbxml_tables_get_main(),
NULL,
NULL,
(const WB_UTINY *) localName);
if (lang_table == NULL) {
/* Damn, this is an unknown language for us... */
tree_ctx->error = WBXML_ERROR_UNKNOWN_XML_LANGUAGE;
return;
}
else {
/* Well, we hope this was the Language we are searching for.. let's try with it :| */
tree_ctx->tree->lang = lang_table;
}
}
}
#if defined( WBXML_SUPPORT_SYNCML )
/* If this is an embedded (not root) document, skip it
* Actually SyncML DevInf and DM DDF are known as such
* potentially embedded documents.
*/
if ((
(WBXML_STRCMP(localName, "syncml:devinf:DevInf") == 0) ||
(WBXML_STRCMP(localName, "syncml:dmddf1.2:MgmtTree") == 0)
)&&
(tree_ctx->current != NULL))
{
tree_ctx->skip_start = XML_GetCurrentByteIndex(tree_ctx->xml_parser);
/* Skip this node */
tree_ctx->skip_lvl++;
return;
}
#endif /* WBXML_SUPPORT_SYNCML */
/* Add Element Node */
tree_ctx->current = wbxml_tree_add_xml_elt_with_attrs(tree_ctx->tree,
tree_ctx->current,
(WB_UTINY *) localName,
(const WB_UTINY**) attrs);
if (tree_ctx->current == NULL) {
tree_ctx->error = WBXML_ERROR_NOT_ENOUGH_MEMORY;
}
}
