static void
atom10_parse_feed_generator (xmlNodePtr cur, feedParserCtxtPtr ctxt, itemPtr ip, struct atom10ParserState *state)
{
gchar *ret, *version, *tmp = NULL, *uri;
ret = unhtmlize ((gchar *)xmlNodeListGetString (cur->doc, cur->xmlChildrenNode, 1));
if (ret && ret[0] != '\0') {
version = xml_get_ns_attribute (cur, "version", NULL);
if (version) {
tmp = g_strdup_printf ("%s %s", ret, version);
g_free (ret);
g_free (version);
ret = tmp;
}
uri = xml_get_ns_attribute (cur, "uri", NULL);
if (uri) {
tmp = g_markup_printf_escaped ("<a href=\"%s\">%s</a>", uri, ret);
g_free (uri);
g_free (ret);
ret = tmp;
}
ctxt->subscription->metadata = metadata_list_append (ctxt->subscription->metadata, "feedgenerator", tmp);
}
g_free (ret);
}
/**
* Parse Atom 1.0 text tags of all sorts.
*
* @param htmlified If set to 1, then HTML is returned.
* When set to 0, All HTML tags are removed
*
* @returns an escaped version of a text construct.
*/
static gchar *
atom10_parse_text_construct (xmlNodePtr cur, gboolean htmlified)
{
gchar *type, *tmp, *ret = NULL;
/* determine encoding mode */
type = xml_get_ns_attribute (cur, "type", NULL);
/* not sure what MIME types are necessary... */
/* This that need to be de-encoded and should not contain sub-tags.*/
if (!type || !strcmp(type, "text")) {
ret = (gchar *)xmlNodeListGetString (cur->doc, cur->xmlChildrenNode, 1);
if (ret) {
g_strchug (g_strchomp (ret));
if (htmlified) {
tmp = atom10_mark_up_text_content (ret);
g_free (ret);
ret = tmp;
}
}
} else if (!strcmp(type, "html")) {
ret = xhtml_extract (cur, 0, NULL);
if (!htmlified)
ret = unhtmlize (unxmlize (ret));
} else if (!strcmp (type, "xhtml")) {
/* The spec says to show the contents of the div tag that MUST be present */
ret = xhtml_extract (cur, 2, NULL);
if (!htmlified)
ret = unhtmlize (ret);
} else {
/* Invalid Atom feed */
ret = g_strdup ("This attribute was invalidly specified in this Atom feed.");
}
g_free (type);
return ret;
}
static gchar *
search_links (const gchar* data, gint linkType)
{
gchar *ptr;
const gchar *tmp = data;
gchar *result = NULL;
gchar *res;
gchar *tstr;
gchar *endptr;
while (1) {
ptr = common_strcasestr (tmp, "<link");
if (!ptr)
return NULL;
endptr = strchr (ptr, '>');
if (!endptr)
return NULL;
*endptr = '\0';
tstr = g_strdup (ptr);
*endptr = '>';
res = checkLinkRef (tstr, linkType);
g_free (tstr);
if (res) {
result = res;
break;
/* deactivated as long as we support only subscribing
to the first found link (BTW this code crashes on
sites like Groklaw!)
gchar* t;
if(result == NULL)
result = res;
else {
t = g_strdup_printf("%s\n%s", result, res);
g_free(res);
g_free(result);
result = t;
}*/
}
tmp = endptr;
}
result = unhtmlize (result); /* URIs can contain escaped things.... All ampersands must be escaped, for example */
return result;
}
static xmlEntityPtr
xml_process_entities (void *ctxt, const xmlChar *name)
{
gchar *path;
xmlEntityPtr entity;
xmlEntityPtr found;
xmlChar *tmp;
static xmlDocPtr entities = NULL;
entity = xmlGetPredefinedEntity (name);
if (!entity) {
if (!entities) {
/* loading HTML entities from external DTD file */
entities = xmlNewDoc (BAD_CAST "1.0");
path = g_build_filename (g_get_user_data_dir (), PACKAGE, "/dtd/html.ent", NULL);
xmlCreateIntSubset (entities, BAD_CAST "HTML entities", NULL, BAD_CAST path);
g_free (path);
entities->extSubset = xmlParseDTD (entities->intSubset->ExternalID, entities->intSubset->SystemID);
}
if (NULL != (found = xmlGetDocEntity (entities, name))) {
/* returning as faked predefined entity... */
tmp = xmlStrdup (found->content);
tmp = BAD_CAST unhtmlize ((gchar*) tmp); /* arghh ... slow... */
entity = (xmlEntityPtr) g_new0 (xmlEntity, 1);
entity->type = XML_ENTITY_DECL;
entity->name = name;
entity->orig = NULL;
entity->content = tmp;
entity->length = g_utf8_strlen ((gchar*) tmp, -1);
entity->etype = XML_INTERNAL_PREDEFINED_ENTITY;
}
}
return entity;
}
/* reads a PIE feed URL and returns a new channel structure (even if
the feed could not be read) */
static void pie_parse(feedParserCtxtPtr ctxt, xmlNodePtr cur) {
gchar *tmp2, *tmp = NULL, *tmp3;
NsHandler *nsh;
parseChannelTagFunc pf;
while(TRUE) {
if(xmlStrcmp(cur->name, BAD_CAST"feed")) {
g_string_append(ctxt->feed->parseErrors, "<p>Could not find Atom/Echo/PIE header!</p>");
break;
}
/* parse feed contents */
cur = cur->xmlChildrenNode;
while(cur) {
if(!cur->name || cur->type != XML_ELEMENT_NODE) {
cur = cur->next;
continue;
}
/* check namespace of this tag */
if(cur->ns) {
if((cur->ns->href && (nsh = (NsHandler *)g_hash_table_lookup(ns_pie_ns_uri_table, (gpointer)cur->ns->href))) ||
(cur->ns->prefix && (nsh = (NsHandler *)g_hash_table_lookup(pie_nstable, (gpointer)cur->ns->prefix)))) {
pf = nsh->parseChannelTag;
if(pf)
(*pf)(ctxt, cur);
cur = cur->next;
continue;
} else {
/*g_print("unsupported namespace \"%s\"\n", cur->ns->prefix);*/
}
} /* explicitly no following else !!! */
if(!xmlStrcmp(cur->name, BAD_CAST"title")) {
tmp = unhtmlize(pie_parse_content_construct(cur));
if(tmp) {
if(ctxt->title)
g_free(ctxt->title);
ctxt->title = tmp;
}
} else if(!xmlStrcmp(cur->name, BAD_CAST"link")) {
tmp = xml_get_attribute (cur, "href");
if(tmp) {
/* 0.3 link : rel, type and href attribute */
tmp2 = xml_get_attribute (cur, "rel");
if(tmp2 && g_str_equal(tmp2, "alternate"))
subscription_set_homepage (ctxt->subscription, tmp);
/* else
FIXME: Maybe do something with other links? */
g_free(tmp2);
g_free(tmp);
} else {
/* 0.2 link : element content is the link, or non-alternate link in 0.3 */
tmp = (gchar *)xmlNodeListGetString(ctxt->doc, cur->xmlChildrenNode, 1);
if(tmp) {
subscription_set_homepage (ctxt->subscription, tmp);
g_free(tmp);
}
}
/* parse feed author */
} else if(!xmlStrcmp(cur->name, BAD_CAST"author")) {
/* parse feed author */
tmp = parseAuthor(cur);
if(tmp) {
ctxt->subscription->metadata = metadata_list_append(ctxt->subscription->metadata, "author", tmp);
g_free(tmp);
}
} else if (!xmlStrcmp (cur->name, BAD_CAST"tagline")) {
tmp = pie_parse_content_construct (cur);
if (tmp) {
metadata_list_set (&ctxt->subscription->metadata, "description", tmp);
g_free (tmp);
}
} else if(!xmlStrcmp(cur->name, BAD_CAST"generator")) {
tmp = unhtmlize((gchar *)xmlNodeListGetString(ctxt->doc, cur->xmlChildrenNode, 1));
if(tmp && tmp[0] != '\0') {
tmp2 = xml_get_attribute (cur, "version");
if(tmp2) {
tmp3 = g_strdup_printf("%s %s", tmp, tmp2);
g_free(tmp);
g_free(tmp2);
tmp = tmp3;
}
tmp2 = xml_get_attribute (cur, "url");
if(tmp2) {
tmp3 = g_strdup_printf("<a href=\"%s\">%s</a>", tmp2, tmp);
g_free(tmp2);
g_free(tmp);
tmp = tmp3;
}
ctxt->subscription->metadata = metadata_list_append(ctxt->subscription->metadata, "feedgenerator", tmp);
}
g_free(tmp);
} else if(!xmlStrcmp(cur->name, BAD_CAST"copyright")) {
tmp = pie_parse_content_construct(cur);
if(tmp) {
ctxt->subscription->metadata = metadata_list_append(ctxt->subscription->metadata, "copyright", tmp);
g_free(tmp);
}
} else if(!xmlStrcmp(cur->name, BAD_CAST"modified")) { /* Modified was last used in IETF draft 02) */
tmp = (gchar *)xmlNodeListGetString(ctxt->doc, cur->xmlChildrenNode, 1);
if(tmp) {
ctxt->subscription->metadata = metadata_list_append(ctxt->subscription->metadata, "pubDate", tmp);
ctxt->feed->time = date_parse_ISO8601 (tmp);
g_free(tmp);
}
} else if(!xmlStrcmp(cur->name, BAD_CAST"updated")) { /* Updated was added in IETF draft 03 */
tmp = (gchar *)xmlNodeListGetString(ctxt->doc, cur->xmlChildrenNode, 1);
if(tmp) {
ctxt->subscription->metadata = metadata_list_append(ctxt->subscription->metadata, "pubDate", tmp);
ctxt->feed->time = date_parse_ISO8601 (tmp);
g_free(tmp);
}
} else if(!xmlStrcmp(cur->name, BAD_CAST"contributor")) {
tmp = parseAuthor(cur);
if(tmp) {
ctxt->subscription->metadata = metadata_list_append(ctxt->subscription->metadata, "contributor", tmp);
g_free(tmp);
}
} else if((!xmlStrcmp(cur->name, BAD_CAST"entry"))) {
ctxt->item = parseEntry(ctxt, cur);
if(ctxt->item) {
if(0 == ctxt->item->time)
ctxt->item->time = ctxt->feed->time;
ctxt->items = g_list_append(ctxt->items, ctxt->item);
}
}
cur = cur->next;
}
break;
}
}
/* method to parse standard tags for each item element */
itemPtr parseEntry(feedParserCtxtPtr ctxt, xmlNodePtr cur) {
xmlChar *xtmp;
gchar *tmp2, *tmp;
NsHandler *nsh;
parseItemTagFunc pf;
g_assert(NULL != cur);
ctxt->item = item_new();
cur = cur->xmlChildrenNode;
while(cur) {
if(!cur->name) {
g_warning("invalid XML: parser returns NULL value -> tag ignored!");
cur = cur->next;
continue;
}
/* check namespace of this tag */
if(cur->ns) {
if((cur->ns->href && (nsh = (NsHandler *)g_hash_table_lookup(ns_pie_ns_uri_table, (gpointer)cur->ns->href))) ||
(cur->ns->prefix && (nsh = (NsHandler *)g_hash_table_lookup(pie_nstable, (gpointer)cur->ns->prefix)))) {
if(NULL != (pf = nsh->parseItemTag))
(*pf)(ctxt, cur);
cur = cur->next;
continue;
} else {
/*g_print("unsupported namespace \"%s\"\n", cur->ns->prefix);*/
}
} /* explicitly no following else !!! */
if(!xmlStrcmp(cur->name, BAD_CAST"title")) {
if(NULL != (tmp = unhtmlize(pie_parse_content_construct(cur)))) {
item_set_title(ctxt->item, tmp);
g_free(tmp);
}
} else if(!xmlStrcmp(cur->name, BAD_CAST"link")) {
if(NULL != (tmp2 = xml_get_attribute(cur, "href"))) {
/* 0.3 link : rel, type and href attribute */
xtmp = xmlGetProp(cur, BAD_CAST"rel");
if(xtmp != NULL && !xmlStrcmp(xtmp, BAD_CAST"alternate"))
item_set_source(ctxt->item, tmp2);
/* else
FIXME: Maybe do something with other links? */
xmlFree(xtmp);
g_free(tmp2);
} else {
/* 0.2 link : element content is the link, or non-alternate link in 0.3 */
if(NULL != (tmp = (gchar *)xmlNodeListGetString(ctxt->doc, cur->xmlChildrenNode, 1))) {
item_set_source(ctxt->item, tmp);
g_free(tmp);
}
}
} else if(!xmlStrcmp(cur->name, BAD_CAST"author")) {
/* parse feed author */
tmp = parseAuthor(cur);
ctxt->item->metadata = metadata_list_append(ctxt->item->metadata, "author", tmp);
g_free(tmp);
} else if(!xmlStrcmp(cur->name, BAD_CAST"contributor")) {
/* parse feed contributors */
tmp = parseAuthor(cur);
ctxt->item->metadata = metadata_list_append(ctxt->item->metadata, "contributor", tmp);
g_free(tmp);
} else if(!xmlStrcmp(cur->name, BAD_CAST"id")) {
if(NULL != (tmp = (gchar *)xmlNodeListGetString(ctxt->doc, cur->xmlChildrenNode, 1))) {
item_set_id(ctxt->item, tmp);
g_free(tmp);
}
} else if(!xmlStrcmp(cur->name, BAD_CAST"issued")) {
/* FIXME: is <modified> or <issued> or <created> the time tag we want to display? */
if(NULL != (tmp = (gchar *)xmlNodeListGetString(ctxt->doc, cur->xmlChildrenNode, 1))) {
ctxt->item->time = date_parse_ISO8601 (tmp);
g_free(tmp);
}
} else if(!xmlStrcmp(cur->name, BAD_CAST"content")) {
/* <content> support */
if(NULL != (tmp = pie_parse_content_construct(cur))) {
item_set_description(ctxt->item, tmp);
g_free(tmp);
}
} else if(!xmlStrcmp(cur->name, BAD_CAST"summary")) {
/* <summary> can be used for short text descriptions, if there is no
<content> description we show the <summary> content */
if(!item_get_description(ctxt->item)) {
if(NULL != (tmp = pie_parse_content_construct(cur))) {
item_set_description(ctxt->item, tmp);
g_free(tmp);
}
}
} else if(!xmlStrcmp(cur->name, BAD_CAST"copyright")) {
if(NULL != (tmp = (gchar *)xmlNodeListGetString(ctxt->doc, cur->xmlChildrenNode, 1))) {
ctxt->item->metadata = metadata_list_append(ctxt->item->metadata, "copyright", tmp);
g_free(tmp);
}
}
cur = cur->next;
}
/* after parsing we fill the infos into the itemPtr structure */
ctxt->item->readStatus = FALSE;
return ctxt->item;
}
/* method to parse standard tags for each item element */
itemPtr parseCDFItem(feedParserCtxtPtr ctxt, xmlNodePtr cur, CDFChannelPtr cp) {
gchar *tmp = NULL, *tmp2, *tmp3;
if(CDFToMetadataMapping == NULL) {
CDFToMetadataMapping = g_hash_table_new(g_str_hash, g_str_equal);
g_hash_table_insert(CDFToMetadataMapping, "author", "author");
g_hash_table_insert(CDFToMetadataMapping, "category", "category");
}
ctxt->item = item_new();
/* save the item link */
if(!(tmp = (gchar *)xmlGetProp(cur, BAD_CAST"href")))
tmp = (gchar *)xmlGetProp(cur, BAD_CAST"HREF");
if(tmp) {
item_set_source(ctxt->item, tmp);
g_free(tmp);
}
cur = cur->xmlChildrenNode;
while(cur) {
if(!cur->name || cur->type != XML_ELEMENT_NODE) {
cur = cur->next;
continue;
}
/* save first link to a channel image */
if(NULL != (tmp = g_ascii_strdown((gchar *)cur->name, -1))) {
if(NULL != (tmp2 = g_hash_table_lookup(CDFToMetadataMapping, tmp))) {
if(NULL != (tmp3 = (gchar *)xmlNodeListGetString(cur->doc, cur->xmlChildrenNode, TRUE))) {
ctxt->item->metadata = metadata_list_append(ctxt->item->metadata, tmp2, tmp3);
g_free(tmp3);
}
}
g_free(tmp);
}
if((!xmlStrcasecmp(cur->name, BAD_CAST"logo"))) {
if(!(tmp = (gchar *)xmlGetProp(cur, BAD_CAST"href")))
tmp = (gchar *)xmlGetProp(cur, BAD_CAST"HREF");
if(tmp) {
ctxt->item->metadata = metadata_list_append(ctxt->item->metadata, "imageUrl", tmp);
g_free(tmp);
}
} else if((!xmlStrcasecmp(cur->name, BAD_CAST"title"))) {
if(NULL != (tmp = unhtmlize((gchar *)xmlNodeListGetString(cur->doc, cur->xmlChildrenNode, 1)))) {
item_set_title(ctxt->item, tmp);
g_free(tmp);
}
} else if((!xmlStrcasecmp(cur->name, BAD_CAST"abstract"))) {
if(NULL != (tmp = (gchar *)xmlNodeListGetString(cur->doc, cur->xmlChildrenNode, 1))) {
item_set_description(ctxt->item, tmp);
g_free(tmp);
}
} else if((!xmlStrcasecmp(cur->name, BAD_CAST"a"))) {
if(!(tmp = (gchar *)xmlGetProp(cur, BAD_CAST"href")))
tmp = (gchar *)xmlGetProp(cur, BAD_CAST"HREF");
if(tmp) {
item_set_source(ctxt->item, tmp);
g_free(tmp);
}
}
cur = cur->next;
}
ctxt->item->readStatus = FALSE;
return ctxt->item;
}
static GrssFeedItem*
parse_rss_item (FeedRssHandler *parser, GrssFeedChannel *feed, xmlDocPtr doc, xmlNodePtr cur)
{
gchar *tmp;
gchar *tmp2;
gchar *tmp3;
time_t t;
GrssFeedItem *item;
g_assert (cur != NULL);
item = grss_feed_item_new (feed);
/* try to get an item about id */
tmp = (gchar*) xmlGetProp (cur, BAD_CAST"about");
if (tmp) {
grss_feed_item_set_id (item, tmp);
grss_feed_item_set_source (item, tmp);
g_free (tmp);
}
cur = cur->xmlChildrenNode;
while (cur) {
if (cur->type != XML_ELEMENT_NODE || !cur->name) {
cur = cur->next;
continue;
}
/* check namespace of this tag */
if (cur->ns) {
if (ns_handler_item (parser->priv->handler, item, cur)) {
cur = cur->next;
continue;
}
}
if (!xmlStrcmp (cur->name, BAD_CAST"category")) {
tmp = (gchar*) xmlNodeListGetString (doc, cur->xmlChildrenNode, 1);
if (tmp) {
grss_feed_item_add_category (item, tmp);
g_free (tmp);
}
}
else if (!xmlStrcmp (cur->name, BAD_CAST"author")) {
tmp = (gchar*) xmlNodeListGetString (doc, cur->xmlChildrenNode, 1);
if (tmp) {
grss_feed_item_set_author (item, tmp);
g_free (tmp);
}
}
else if (!xmlStrcmp (cur->name, BAD_CAST"comments")) {
tmp = (gchar*) xmlNodeListGetString (doc, cur->xmlChildrenNode, 1);
if (tmp) {
grss_feed_item_set_comments_url (item, tmp);
g_free (tmp);
}
}
else if (!xmlStrcmp (cur->name, BAD_CAST"pubDate")) {
tmp = (gchar*) xmlNodeListGetString (doc, cur->xmlChildrenNode, 1);
if (tmp) {
t = date_parse_RFC822 (tmp);
grss_feed_item_set_publish_time (item, t);
g_free (tmp);
}
}
else if (!xmlStrcmp (cur->name, BAD_CAST"enclosure")) {
/* RSS 0.93 allows multiple enclosures */
tmp = (gchar*) xmlGetProp (cur, BAD_CAST"url");
if (tmp) {
gchar *type = (gchar*) xmlGetProp (cur, BAD_CAST"type");
gssize length = 0;
GrssFeedEnclosure *enclosure;
tmp2 = (gchar*) xmlGetProp (cur, BAD_CAST"length");
if (tmp2) {
length = atol (tmp2);
xmlFree (tmp2);
}
tmp3 = (gchar*) grss_feed_channel_get_homepage (feed);
if ((strstr (tmp, "://") == NULL) &&
(tmp3 != NULL) &&
(strstr (tmp3, "://") != NULL)) {
/* add base URL if necessary and possible */
tmp2 = g_strdup_printf ("%s/%s", tmp3, tmp);
xmlFree (tmp);
tmp = tmp2;
}
enclosure = grss_feed_enclosure_new (tmp);
grss_feed_enclosure_set_format (enclosure, type);
grss_feed_enclosure_set_length (enclosure, length);
grss_feed_item_add_enclosure (item, enclosure);
xmlFree (tmp);
xmlFree (type);
}
}
else if (!xmlStrcmp (cur->name, BAD_CAST"guid")) {
if (!grss_feed_item_get_id (item)) {
tmp = (gchar*) xmlNodeListGetString (doc, cur->xmlChildrenNode, 1);
if (tmp) {
if (strlen (tmp) > 0) {
grss_feed_item_set_id (item, tmp);
tmp2 = (gchar*) xmlGetProp (cur, BAD_CAST"isPermaLink");
if (!grss_feed_item_get_source (item) && (tmp2 == NULL || g_str_equal (tmp2, "true")))
grss_feed_item_set_source (item, tmp); /* Per the RSS 2.0 spec. */
if (tmp2)
xmlFree (tmp2);
}
xmlFree (tmp);
}
}
}
else if (!xmlStrcmp (cur->name, BAD_CAST"title")) {
tmp = unhtmlize ((gchar*) xmlNodeListGetString (doc, cur->xmlChildrenNode, TRUE));
if (tmp) {
grss_feed_item_set_title (item, tmp);
g_free (tmp);
}
}
else if (!xmlStrcmp (cur->name, BAD_CAST"link")) {
tmp = unhtmlize ((gchar*) xmlNodeListGetString (doc, cur->xmlChildrenNode, TRUE));
if (tmp) {
grss_feed_item_set_source (item, tmp);
g_free (tmp);
}
}
else if (!xmlStrcmp (cur->name, BAD_CAST"description")) {
tmp = xhtml_extract (cur, 0, NULL);
if (tmp) {
/* don't overwrite content:encoded descriptions... */
if (!grss_feed_item_get_description (item))
grss_feed_item_set_description (item, tmp);
g_free (tmp);
}
}
else if (!xmlStrcmp (cur->name, BAD_CAST"source")) {
tmp = (gchar*) xmlGetProp (cur, BAD_CAST"url");
tmp2 = unhtmlize ((gchar*) xmlNodeListGetString (doc, cur->xmlChildrenNode, 1));
if (tmp) {
grss_feed_item_set_real_source (item, g_strchomp (tmp), tmp2 ? g_strchomp (tmp2) : NULL);
g_free (tmp);
}
if (tmp2)
g_free (tmp2);
}
cur = cur->next;
}
return item;
}
static void
parse_channel (FeedRssHandler *parser, GrssFeedChannel *feed, xmlDocPtr doc, xmlNodePtr cur) {
gchar *tmp;
time_t t;
g_assert (NULL != cur);
cur = cur->xmlChildrenNode;
while (cur) {
if (cur->type != XML_ELEMENT_NODE || cur->name == NULL) {
cur = cur->next;
continue;
}
/* check namespace of this tag */
if (cur->ns) {
if (ns_handler_channel (parser->priv->handler, feed, cur)) {
cur = cur->next;
continue;
}
}
if (!xmlStrcmp (cur->name, BAD_CAST"copyright")) {
tmp = (gchar*) xmlNodeListGetString (doc, cur->xmlChildrenNode, 1);
if (tmp) {
grss_feed_channel_set_copyright (feed, tmp);
g_free (tmp);
}
}
else if (!xmlStrcmp (cur->name, BAD_CAST"webMaster") || !xmlStrcmp (cur->name, BAD_CAST"publisher")) {
tmp = (gchar*) xmlNodeListGetString (doc, cur->xmlChildrenNode, 1);
if (tmp) {
grss_feed_channel_set_webmaster (feed, tmp);
g_free (tmp);
}
}
else if (!xmlStrcmp (cur->name, BAD_CAST"language")) {
tmp = (gchar*) xmlNodeListGetString (doc, cur->xmlChildrenNode, 1);
if (tmp) {
grss_feed_channel_set_language (feed, tmp);
g_free (tmp);
}
}
else if (!xmlStrcmp (cur->name, BAD_CAST"managingEditor")) {
tmp = (gchar*) xmlNodeListGetString (doc, cur->xmlChildrenNode, 1);
if (tmp) {
grss_feed_channel_set_editor (feed, tmp);
g_free (tmp);
}
}
else if (!xmlStrcmp (cur->name, BAD_CAST"lastBuildDate")) {
tmp = (gchar*) xmlNodeListGetString (doc, cur->xmlChildrenNode, 1);
if (tmp) {
t = date_parse_RFC822 (tmp);
grss_feed_channel_set_update_time (feed, t);
g_free (tmp);
}
}
else if (!xmlStrcmp (cur->name, BAD_CAST"generator")) {
tmp = (gchar*) xmlNodeListGetString (doc, cur->xmlChildrenNode, 1);
if (tmp) {
grss_feed_channel_set_generator (feed, tmp);
g_free (tmp);
}
}
else if (!xmlStrcmp (cur->name, BAD_CAST"pubDate")) {
if (NULL != (tmp = (gchar*) xmlNodeListGetString (doc, cur->xmlChildrenNode, 1))) {
t = date_parse_RFC822 (tmp);
grss_feed_channel_set_publish_time (feed, t);
g_free (tmp);
}
}
else if (!xmlStrcmp (cur->name, BAD_CAST"ttl")) {
if (NULL != (tmp = (gchar*) xmlNodeListGetString (doc, cur->xmlChildrenNode, TRUE))) {
grss_feed_channel_set_update_interval (feed, atoi (tmp));
g_free (tmp);
}
}
else if (!xmlStrcmp (cur->name, BAD_CAST"title")) {
if (NULL != (tmp = unhtmlize ((gchar*) xmlNodeListGetString (doc, cur->xmlChildrenNode, TRUE)))) {
grss_feed_channel_set_title (feed, tmp);
g_free (tmp);
}
}
/*
<alink> has been found at least in Xinhua News Agency RSS feeds
*/
else if (!xmlStrcmp (cur->name, BAD_CAST"link") || !xmlStrcmp (cur->name, BAD_CAST"alink")) {
if (NULL != (tmp = unhtmlize ((gchar*) xmlNodeListGetString (doc, cur->xmlChildrenNode, TRUE)))) {
grss_feed_channel_set_homepage (feed, tmp);
g_free (tmp);
}
}
else if (!xmlStrcmp (cur->name, BAD_CAST"description")) {
tmp = (gchar*) xmlNodeListGetString (doc, cur->xmlChildrenNode, TRUE);
if (tmp) {
grss_feed_channel_set_description (feed, tmp);
g_free (tmp);
}
}
else if (!xmlStrcmp (cur->name, BAD_CAST"cloud")) {
parse_rss_cloud (feed, cur);
}
cur = cur->next;
}
}
/* method to parse standard tags for the channel element */
static void parseCDFChannel(feedParserCtxtPtr ctxt, xmlNodePtr cur, CDFChannelPtr cp) {
gchar *tmp, *tmp2, *tmp3;
cur = cur->xmlChildrenNode;
while(cur) {
if(!cur->name || cur->type != XML_ELEMENT_NODE) {
cur = cur->next;
continue;
}
if((!xmlStrcasecmp(cur->name, BAD_CAST"logo"))) {
tmp = (gchar *)xmlGetProp(cur, BAD_CAST"HREF");
if(tmp) {
tmp = (gchar *)xmlGetProp(cur, BAD_CAST"href");
metadata_list_set (&ctxt->subscription->metadata, "imageUrl", tmp);
g_free(tmp);
}
} else if((!xmlStrcasecmp(cur->name, BAD_CAST"a"))) {
xmlChar *value = xmlGetProp(cur, BAD_CAST"HREF");
if(value) {
subscription_set_homepage (ctxt->subscription, (gchar *)value);
xmlFree(value);
}
} else if((!xmlStrcasecmp(cur->name, BAD_CAST"item"))) {
ctxt->item = parseCDFItem(ctxt, cur, cp);
if(ctxt->item) {
if(0 == ctxt->item->time)
ctxt->item->time = cp->time;
ctxt->items = g_list_append(ctxt->items, ctxt->item);
}
} else if(!xmlStrcasecmp(cur->name, BAD_CAST "title")) {
tmp = (gchar *)xmlNodeListGetString(cur->doc, cur->xmlChildrenNode, TRUE);
if(tmp) {
tmp = unhtmlize(tmp);
if(ctxt->title)
g_free(ctxt->title);
ctxt->title = tmp;
}
} else if (!xmlStrcasecmp(cur->name, BAD_CAST "abstract")) {
tmp = (gchar *)xmlNodeListGetString (cur->doc, cur->xmlChildrenNode, TRUE);
if (tmp) {
metadata_list_set (&ctxt->subscription->metadata, "description", tmp);
xmlFree (tmp);
}
} else {
tmp = g_ascii_strdown((gchar *)cur->name, -1);
tmp2 = g_hash_table_lookup(channelHash, tmp);
if(tmp2) {
tmp3 = (gchar *)xmlNodeListGetString(cur->doc, cur->xmlChildrenNode, TRUE);
if(tmp3) {
ctxt->subscription->metadata = metadata_list_append(ctxt->subscription->metadata, tmp2, tmp3);
g_free(tmp3);
}
}
g_free(tmp);
}
cur = cur->next;
}
}
