/* 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;
}
/* 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;
}
}
static void
parseTitleInfo (xmlDocPtr doc, xmlNodePtr cur, FB2Content *fb)
{
xmlChar *content;
cur = cur->children;
while (cur != NULL) {
if (!xmlStrcmp(cur->name, (const xmlChar *)"p")) {
parseP(doc, cur, 1, fb);
} else if (!xmlStrcmp(cur->name, (const xmlChar *)"empty-line")) {
bufferAppend("\n", 1, fb);
} else if (!xmlStrcmp(cur->name, (const xmlChar *)"genre")) {
content = xmlNodeGetContent(cur->children);
if(content) {
/* add genre */
int i;
for (i=0; i<BUF_SIZE-1; i++) {
if (fb->genres[i] == NULL) {
fb->genres[i] = content;
fb->genres[i+1] = NULL;
fb->num_genres++;
break;
}
}
}
} else if (!xmlStrcmp(cur->name, (const xmlChar *)"book-title")) {
content = xmlNodeGetContent(cur->children);
if (content)
strncpy(fb->name, content, BUF_SIZE);
xmlFree(content);
} else if (!xmlStrcmp(cur->name, (const xmlChar *)"author")) {
parseAuthor(doc, cur, fb);
} else if (!xmlStrcmp(cur->name, (const xmlChar *)"annotation")) {
parseAnnotation(doc, cur, fb);
} else if (!xmlStrcmp(cur->name, (const xmlChar *)"coverpage")) {
xmlNodePtr c;
c = cur->children;
while (c) {
if (!xmlStrcmp(c->name, (const xmlChar *)"image")) {
xmlChar *cover_href;
cover_href = xmlGetProp(c, (const xmlChar *)"href");
strncpy(fb->cover_href, cover_href, BUF_SIZE);
xmlFree (cover_href);
break;
}
c = c->next;
}
/*
content = xmlNodeGetContent(cur->children);
if (content)
strncpy(fb->cover_href, content, BUF_SIZE);
*/
//if (cover_href)
/*xmlFree(content);*/
}
cur = cur->next;
}
return;
}
