1 /*
2 * Copyright (C) 2006 Andrej Kacian <andrej@kacian.sk>
3 *
4 * This program is free software; you can redistribute it and/or
5 * modify it under the terms of the GNU General Public License as
6 * published by the Free Software Foundation; either version 2 of the
7 * License, or (at your option) any later version.
8 *
9 * This program is distributed in the hope that it will be useful,
10 * but WITHOUT ANY WARRANTY; without even the implied warranty of
11 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
12 * General Public License for more details.
13 *
14 * You should have received a copy of the GNU General Public
15 * License along with this program; if not, write to the
16 * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
17 * Boston, MA 02111-1307, USA.
18 */
19 #define __USE_GNU
20
21 #include <glib.h>
22 #include <expat.h>
23 #include <string.h>
24
25 #include <procheader.h>
26
27 #include "feed.h"
28 #include "feeditem.h"
29 #include "feeditemenclosure.h"
30 #include "date.h"
31 #include "parser.h"
32 #include "common/utils.h"
33
feed_parser_rss20_start(void * data,const gchar * el,const gchar ** attr)34 void feed_parser_rss20_start(void *data, const gchar *el, const gchar **attr)
35 {
36 FeedParserCtx *ctx = (FeedParserCtx *)data;
37 FeedItemEnclosure *enclosure = NULL;
38 gchar *url, *type, *size_s;
39 gulong size = -1;
40
41 /* ------------------- */
42 if( ctx->depth == 2 ) {
43 if( !strcmp(el, "item") ) { /* Start of new item */
44
45 if( ctx->curitem != NULL )
46 feed_item_free(ctx->curitem);
47
48 ctx->curitem = feed_item_new(ctx->feed);
49
50 } else ctx->location = 0;
51 /* ------------------- */
52 } else if( ctx->depth == 3 ) {
53 if( !strcmp(el, "enclosure") ) { /* Media enclosure */
54
55 url = feed_parser_get_attribute_value(attr, "url");
56 type = feed_parser_get_attribute_value(attr, "type");
57 size_s = feed_parser_get_attribute_value(attr, "length");
58 if( size_s != NULL )
59 size = (gulong)atol(size_s);
60
61 if( url != NULL && type != NULL && size > 0 ) {
62 if( (enclosure = feed_item_enclosure_new(url, type, size)) )
63 feed_item_set_enclosure(ctx->curitem, enclosure);
64 }
65
66 } else if( !strcmp(el, "guid") ) { /* Unique ID */
67 type = feed_parser_get_attribute_value(attr, "isPermaLink");
68 if( type != NULL && !strcmp(type, "false") )
69 feed_item_set_id_permalink(ctx->curitem, TRUE);
70 }
71 } else ctx->location = 0;
72
73 ctx->depth++;
74
75 }
76
feed_parser_rss20_end(void * data,const gchar * el)77 void feed_parser_rss20_end(void *data, const gchar *el)
78 {
79 FeedParserCtx *ctx = (FeedParserCtx *)data;
80 Feed *feed = ctx->feed;
81 gchar *text = NULL;
82
83 if( ctx->str != NULL )
84 text = g_strstrip(g_strdup(ctx->str->str));
85 else
86 text = "";
87
88 ctx->depth--;
89
90 switch( ctx->depth ) {
91
92 /* ------------------- */
93 case 0:
94
95 if( !strcmp(el, "rss") ) {
96 /* we finished parsing the feed */
97 ctx->feed->items = g_slist_reverse(ctx->feed->items);
98 }
99
100 break;
101
102 /* ------------------- */
103 case 1:
104
105 break; /* nothing to do at this depth */
106
107 /* ------------------- */
108 case 2:
109
110 /* decide if we just received </item>, so we can
111 * add a complete item to feed */
112 if( !strcmp(el, "item") ) {
113
114 /* append the complete feed item, if it is valid
115 * "All elements of an item are optional, however at least one
116 * of title or description must be present." */
117 if( ctx->curitem->title != NULL || ctx->curitem->summary != NULL ) {
118 ctx->feed->items =
119 g_slist_prepend(ctx->feed->items, (gpointer)ctx->curitem);
120 }
121
122 /* since it's in the linked list, lose this pointer */
123 ctx->curitem = NULL;
124
125 } else if( !strcmp(el, "title") ) { /* so it wasn't end of item */
126 FILL(feed->title)
127 } else if( !strcmp(el, "description" ) ) {
128 FILL(feed->description)
129 } else if( !strcmp(el, "dc:language") ) {
130 FILL(feed->language)
131 } else if( !strcmp(el, "author") ) {
132 FILL(feed->author)
133 } else if( !strcmp(el, "admin:generatorAgent") ) {
134 FILL(feed->generator)
135 } else if( !strcmp(el, "dc:date") ) {
136 feed->date = procheader_date_parse(NULL, text, 0);
137 } else if( !strcmp(el, "pubDate") ) {
138 feed->date = procheader_date_parse(NULL, text, 0);
139 }
140
141 break;
142
143 /* ------------------- */
144 case 3:
145
146 if( ctx->curitem == NULL ) {
147 break;
148 }
149
150 /* decide which field did we just get */
151 if( !strcmp(el, "title") ) {
152 FILL(ctx->curitem->title)
153 } else if( !strcmp(el, "author") ) {
154 FILL(ctx->curitem->author)
155 } else if( !strcmp(el, "description") ) {
156 FILL(ctx->curitem->summary)
157 } else if( !strcmp(el, "content:encoded") ) {
158 FILL(ctx->curitem->text)
159 } else if( !strcmp(el, "link") ) {
160 FILL(ctx->curitem->url)
161 } else if( !strcmp(el, "guid") ) {
162 FILL(ctx->curitem->id)
163 } else if( !strcmp(el, "wfw:commentRSS") || !strcmp(el, "wfw:commentRss") ) {
164 FILL(ctx->curitem->comments_url)
165 } else if( !strcmp(el, "dc:date") ) {
166 ctx->curitem->date_modified = procheader_date_parse(NULL, text, 0);
167 } else if( !strcmp(el, "pubDate") ) {
168 ctx->curitem->date_published = procheader_date_parse(NULL, text, 0);
169 } else if( !strcmp(el, "dc:creator")) {
170 FILL(ctx->curitem->author)
171 }
172
173 break;
174
175 }
176
177 if( ctx->str != NULL ) {
178 g_free(text);
179 g_string_free(ctx->str, TRUE);
180 ctx->str = NULL;
181 }
182 }
183