Commit | Line | Data |
c4d4c98e |
1 | # $Id: Feed.pm 1958 2006-08-14 05:31:27Z btrott $ |
0d5e38d1 |
2 | |
3 | package XML::Feed; |
4 | use strict; |
5 | |
62d92771 |
6 | use base qw( Class::ErrorHandler ); |
62d92771 |
7 | use Feed::Find; |
fe71566d |
8 | use URI::Fetch; |
973e1f9e |
9 | use Carp; |
0d5e38d1 |
10 | |
c4d4c98e |
11 | our $VERSION = '0.12'; |
973e1f9e |
12 | |
13 | sub new { |
14 | my $class = shift; |
4e9c4625 |
15 | my $format = shift || 'Atom'; |
973e1f9e |
16 | my $format_class = 'XML::Feed::' . $format; |
17 | eval "use $format_class"; |
18 | Carp::croak("Unsupported format $format: $@") if $@; |
19 | my $feed = bless {}, join('::', __PACKAGE__, $format); |
4e9c4625 |
20 | $feed->init_empty(@_) or return $class->error($feed->errstr); |
973e1f9e |
21 | $feed; |
22 | } |
23 | |
24 | sub init_empty { 1 } |
0d5e38d1 |
25 | |
26 | sub parse { |
27 | my $class = shift; |
41e8c132 |
28 | my($stream, $specified_format) = @_; |
0d5e38d1 |
29 | return $class->error("Stream parameter is required") unless $stream; |
30 | my $feed = bless {}, $class; |
31 | my $xml = ''; |
32 | if (UNIVERSAL::isa($stream, 'URI')) { |
fe71566d |
33 | my $res = URI::Fetch->fetch($stream) |
34 | or return $class->error(URI::Fetch->errstr); |
35 | return $class->error("This feed has been permanently removed") |
b3b6d2fc |
36 | if $res->status == URI::Fetch::URI_GONE(); |
fe71566d |
37 | $xml = $res->content; |
0d5e38d1 |
38 | } elsif (ref($stream) eq 'SCALAR') { |
39 | $xml = $$stream; |
40 | } elsif (ref($stream)) { |
41 | while (read($stream, my($chunk), 8192)) { |
42 | $xml .= $chunk; |
43 | } |
44 | } else { |
45 | open my $fh, $stream |
46 | or return $class->error("Can't open $stream: $!"); |
47 | while (read $fh, my($chunk), 8192) { |
48 | $xml .= $chunk; |
49 | } |
50 | close $fh; |
51 | } |
52 | return $class->error("Can't get feed XML content from $stream") |
53 | unless $xml; |
41e8c132 |
54 | my $format; |
55 | if ($specified_format) { |
56 | $format = $specified_format; |
57 | } else { |
58 | $feed->identify_format(\$xml) or return $class->error($feed->errstr); |
59 | } |
60 | |
fe71566d |
61 | my $format_class = join '::', __PACKAGE__, $format; |
62 | eval "use $format_class"; |
63 | return $class->error("Unsupported format $format: $@") if $@; |
64 | bless $feed, $format_class; |
65 | $feed->init_string(\$xml) or return $class->error($feed->errstr); |
66 | $feed; |
67 | } |
68 | |
69 | sub identify_format { |
70 | my $feed = shift; |
71 | my($xml) = @_; |
0d5e38d1 |
72 | ## Auto-detect feed type based on first element. This is prone |
73 | ## to breakage, but then again we don't want to parse the whole |
74 | ## feed ourselves. |
a749d9b9 |
75 | my $tag; |
fe71566d |
76 | while ($$xml =~ /<(\S+)/sg) { |
62d92771 |
77 | (my $t = $1) =~ tr/a-zA-Z0-9:\-\?!//cd; |
78 | my $first = substr $t, 0, 1; |
79 | $tag = $t, last unless $first eq '?' || $first eq '!'; |
a749d9b9 |
80 | } |
fe71566d |
81 | return $feed->error("Cannot find first element") unless $tag; |
0d5e38d1 |
82 | $tag =~ s/^.*://; |
83 | if ($tag eq 'rss' || $tag eq 'RDF') { |
fe71566d |
84 | return 'RSS'; |
0d5e38d1 |
85 | } elsif ($tag eq 'feed') { |
fe71566d |
86 | return 'Atom'; |
0d5e38d1 |
87 | } else { |
fe71566d |
88 | return $feed->error("Cannot detect feed type"); |
0d5e38d1 |
89 | } |
0d5e38d1 |
90 | } |
91 | |
92 | sub find_feeds { |
93 | my $class = shift; |
94 | my($uri) = @_; |
62d92771 |
95 | my @feeds = Feed::Find->find($uri) |
96 | or return $class->error(Feed::Find->errstr); |
0d5e38d1 |
97 | @feeds; |
98 | } |
99 | |
973e1f9e |
100 | sub convert { |
101 | my $feed = shift; |
102 | my($format) = @_; |
103 | my $new = __PACKAGE__->new($format); |
ecac864a |
104 | for my $field (qw( title link description language author copyright modified generator )) { |
23103173 |
105 | my $val = $feed->$field(); |
106 | next unless defined $val; |
107 | $new->$field($val); |
973e1f9e |
108 | } |
109 | for my $entry ($feed->entries) { |
110 | $new->add_entry($entry->convert($format)); |
111 | } |
112 | $new; |
113 | } |
114 | |
23103173 |
115 | sub splice { |
116 | my $feed = shift; |
117 | my($other) = @_; |
118 | my %ids = map { $_->id => 1 } $feed->entries; |
119 | for my $entry ($other->entries) { |
120 | $feed->add_entry($entry) unless $ids{$entry->id}++; |
121 | } |
122 | } |
123 | |
0d5e38d1 |
124 | sub format; |
125 | sub title; |
126 | sub link; |
127 | sub description; |
128 | sub language; |
973e1f9e |
129 | sub author; |
0d5e38d1 |
130 | sub copyright; |
131 | sub modified; |
132 | sub generator; |
973e1f9e |
133 | sub add_entry; |
0d5e38d1 |
134 | sub entries; |
973e1f9e |
135 | sub as_xml; |
0d5e38d1 |
136 | |
973e1f9e |
137 | sub tagline { shift->description(@_) } |
0d5e38d1 |
138 | sub items { $_[0]->entries } |
139 | |
140 | 1; |
141 | __END__ |
142 | |
143 | =head1 NAME |
144 | |
145 | XML::Feed - Syndication feed parser and auto-discovery |
146 | |
147 | =head1 SYNOPSIS |
148 | |
149 | use XML::Feed; |
150 | my $feed = XML::Feed->parse(URI->new('http://example.com/atom.xml')) |
151 | or die XML::Feed->errstr; |
152 | print $feed->title, "\n"; |
153 | for my $entry ($feed->entries) { |
154 | } |
155 | |
156 | ## Find all of the syndication feeds on a given page, using |
157 | ## auto-discovery. |
158 | my @feeds = XML::Feed->find_feeds('http://example.com/'); |
159 | |
160 | =head1 DESCRIPTION |
161 | |
162 | I<XML::Feed> is a syndication feed parser for both RSS and Atom feeds. It |
163 | also implements feed auto-discovery for finding feeds, given a URI. |
164 | |
165 | I<XML::Feed> supports the following syndication feed formats: |
166 | |
167 | =over 4 |
168 | |
169 | =item * RSS 0.91 |
170 | |
171 | =item * RSS 1.0 |
172 | |
173 | =item * RSS 2.0 |
174 | |
175 | =item * Atom |
176 | |
177 | =back |
178 | |
179 | The goal of I<XML::Feed> is to provide a unified API for parsing and using |
180 | the various syndication formats. The different flavors of RSS and Atom |
181 | handle data in different ways: date handling; summaries and content; |
182 | escaping and quoting; etc. This module attempts to remove those differences |
183 | by providing a wrapper around the formats and the classes implementing |
184 | those formats (I<XML::RSS> and I<XML::Atom::Feed>). For example, dates are |
185 | handled differently in each of the above formats. To provide a unified API for |
186 | date handling, I<XML::Feed> converts all date formats transparently into |
187 | I<DateTime> objects, which it then returns to the caller. |
188 | |
189 | =head1 USAGE |
190 | |
973e1f9e |
191 | =head2 XML::Feed->new($format) |
192 | |
193 | Creates a new empty I<XML::Feed> object using the format I<$format>. |
194 | |
0d5e38d1 |
195 | =head2 XML::Feed->parse($stream) |
196 | |
41e8c132 |
197 | =head2 XML::Feed->parse($stream, $format) |
198 | |
0d5e38d1 |
199 | Parses a syndication feed identified by I<$stream>. I<$stream> can be any |
200 | one of the following: |
201 | |
202 | =over 4 |
203 | |
204 | =item * Scalar reference |
205 | |
206 | A reference to string containing the XML body of the feed. |
207 | |
208 | =item * Filehandle |
209 | |
210 | An open filehandle from which the feed XML will be read. |
211 | |
212 | =item * File name |
213 | |
214 | The name of a file containing the feed XML. |
215 | |
216 | =item * URI object |
217 | |
218 | A URI from which the feed XML will be retrieved. |
219 | |
220 | =back |
221 | |
41e8c132 |
222 | C<$format> allows you to override format guessing. |
223 | |
0d5e38d1 |
224 | =head2 XML::Feed->find_feeds($uri) |
225 | |
226 | Given a URI I<$uri>, use auto-discovery to find all of the feeds linked |
227 | from that page (using I<E<lt>linkE<gt>> tags). |
228 | |
229 | Returns a list of feed URIs. |
230 | |
973e1f9e |
231 | =head2 $feed->convert($format) |
232 | |
233 | Converts the I<XML::Feed> object into the I<$format> format, and returns |
234 | the new object. |
235 | |
23103173 |
236 | =head2 $feed->splice($other_feed) |
237 | |
238 | Splices in all of the entries from the feed I<$other_feed> into I<$feed>, |
239 | skipping posts that are already in I<$feed>. |
240 | |
0d5e38d1 |
241 | =head2 $feed->format |
242 | |
243 | Returns the format of the feed (C<Atom>, or some version of C<RSS>). |
244 | |
973e1f9e |
245 | =head2 $feed->title([ $title ]) |
0d5e38d1 |
246 | |
247 | The title of the feed/channel. |
248 | |
973e1f9e |
249 | =head2 $feed->link([ $uri ]) |
0d5e38d1 |
250 | |
251 | The permalink of the feed/channel. |
252 | |
973e1f9e |
253 | =head2 $feed->tagline([ $tagline ]) |
0d5e38d1 |
254 | |
255 | The description or tagline of the feed/channel. |
256 | |
973e1f9e |
257 | =head2 $feed->description([ $description ]) |
0d5e38d1 |
258 | |
259 | Alias for I<$feed-E<gt>tagline>. |
260 | |
973e1f9e |
261 | =head2 $feed->author([ $author ]) |
262 | |
263 | The author of the feed/channel. |
264 | |
265 | =head2 $feed->language([ $language ]) |
0d5e38d1 |
266 | |
267 | The language of the feed. |
268 | |
973e1f9e |
269 | =head2 $feed->copyright([ $copyright ]) |
0d5e38d1 |
270 | |
271 | The copyright notice of the feed. |
272 | |
973e1f9e |
273 | =head2 $feed->modified([ $modified ]) |
0d5e38d1 |
274 | |
275 | A I<DateTime> object representing the last-modified date of the feed. |
276 | |
973e1f9e |
277 | If present, I<$modified> should be a I<DateTime> object. |
278 | |
279 | =head2 $feed->generator([ $generator ]) |
0d5e38d1 |
280 | |
281 | The generator of the feed. |
282 | |
283 | =head2 $feed->entries |
284 | |
285 | A list of the entries/items in the feed. Returns an array containing |
286 | I<XML::Feed::Entry> objects. |
287 | |
973e1f9e |
288 | =head2 $feed->add_entry($entry) |
289 | |
290 | Adds an entry to the feed. I<$entry> should be an I<XML::Feed::Entry> |
291 | object in the correct format for the feed. |
292 | |
293 | =head2 $feed->as_xml |
294 | |
295 | Returns an XML representation of the feed, in the format determined by |
296 | the current format of the I<$feed> object. |
297 | |
298 | =head1 PACKAGE VARIABLES |
299 | |
300 | =over 4 |
301 | |
302 | =item C<$XML::Feed::RSS::PREFERRED_PARSER> |
303 | |
304 | If you want to use another RSS parser class than XML::RSS (default), you can |
305 | change the class by setting C<$PREFERRED_PARSER> variable in XML::Feed::RSS |
306 | package. |
307 | |
308 | $XML::Feed::RSS::PREFERRED_PARSER = "XML::RSS::LibXML"; |
309 | |
310 | B<Note:> this will only work for parsing feeds, not creating feeds. |
311 | |
312 | =back |
313 | |
0d5e38d1 |
314 | =head1 LICENSE |
315 | |
316 | I<XML::Feed> is free software; you may redistribute it and/or modify it |
317 | under the same terms as Perl itself. |
318 | |
319 | =head1 AUTHOR & COPYRIGHT |
320 | |
973e1f9e |
321 | Except where otherwise noted, I<XML::Feed> is Copyright 2004-2005 |
322 | Six Apart, cpan@sixapart.com. All rights reserved. |
0d5e38d1 |
323 | |
324 | =cut |