1 # $Id: Feed.pm 1958 2006-08-14 05:31:27Z btrott $
6 use base qw( Class::ErrorHandler );
12 our $VERSION = '0.21';
16 my $format = shift || 'Atom';
17 my $format_class = 'XML::Feed::' . $format;
18 eval "use $format_class";
19 Carp::croak("Unsupported format $format: $@") if $@;
20 my $feed = bless {}, join('::', __PACKAGE__, $format);
21 $feed->init_empty(@_) or return $class->error($feed->errstr);
29 my($stream, $specified_format) = @_;
30 return $class->error("Stream parameter is required") unless $stream;
31 my $feed = bless {}, $class;
33 if (UNIVERSAL::isa($stream, 'URI')) {
34 my $ua = LWP::UserAgent->new;
35 $ua->env_proxy; # force allowing of proxies
36 my $res = URI::Fetch->fetch($stream, UserAgent => $ua)
37 or return $class->error(URI::Fetch->errstr);
38 return $class->error("This feed has been permanently removed")
39 if $res->status == URI::Fetch::URI_GONE();
41 } elsif (ref($stream) eq 'SCALAR') {
43 } elsif (ref($stream)) {
44 while (read($stream, my($chunk), 8192)) {
49 or return $class->error("Can't open $stream: $!");
50 while (read $fh, my($chunk), 8192) {
55 return $class->error("Can't get feed XML content from $stream")
58 if ($specified_format) {
59 $format = $specified_format;
61 $format = $feed->identify_format(\$xml) or return $class->error($feed->errstr);
64 my $format_class = join '::', __PACKAGE__, $format;
65 eval "use $format_class";
66 return $class->error("Unsupported format $format: $@") if $@;
67 bless $feed, $format_class;
68 $feed->init_string(\$xml) or return $class->error($feed->errstr);
75 ## Auto-detect feed type based on first element. This is prone
76 ## to breakage, but then again we don't want to parse the whole
79 while ($$xml =~ /<(\S+)/sg) {
80 (my $t = $1) =~ tr/a-zA-Z0-9:\-\?!//cd;
81 my $first = substr $t, 0, 1;
82 $tag = $t, last unless $first eq '?' || $first eq '!';
84 return $feed->error("Cannot find first element") unless $tag;
86 if ($tag eq 'rss' || $tag eq 'RDF') {
88 } elsif ($tag eq 'feed') {
91 return $feed->error("Cannot detect feed type");
98 my @feeds = Feed::Find->find($uri)
99 or return $class->error(Feed::Find->errstr);
106 my $new = __PACKAGE__->new($format);
107 for my $field (qw( title link description language author copyright modified generator )) {
108 my $val = $feed->$field();
109 next unless defined $val;
112 for my $entry ($feed->entries) {
113 $new->add_entry($entry->convert($format));
121 my %ids = map { $_->id => 1 } $feed->entries;
122 for my $entry ($other->entries) {
123 $feed->add_entry($entry) unless $ids{$entry->id}++;
142 sub tagline { shift->description(@_) }
143 sub items { $_[0]->entries }
150 XML::Feed - Syndication feed parser and auto-discovery
155 my $feed = XML::Feed->parse(URI->new('http://example.com/atom.xml'))
156 or die XML::Feed->errstr;
157 print $feed->title, "\n";
158 for my $entry ($feed->entries) {
161 ## Find all of the syndication feeds on a given page, using
163 my @feeds = XML::Feed->find_feeds('http://example.com/');
167 I<XML::Feed> is a syndication feed parser for both RSS and Atom feeds. It
168 also implements feed auto-discovery for finding feeds, given a URI.
170 I<XML::Feed> supports the following syndication feed formats:
184 The goal of I<XML::Feed> is to provide a unified API for parsing and using
185 the various syndication formats. The different flavors of RSS and Atom
186 handle data in different ways: date handling; summaries and content;
187 escaping and quoting; etc. This module attempts to remove those differences
188 by providing a wrapper around the formats and the classes implementing
189 those formats (I<XML::RSS> and I<XML::Atom::Feed>). For example, dates are
190 handled differently in each of the above formats. To provide a unified API for
191 date handling, I<XML::Feed> converts all date formats transparently into
192 I<DateTime> objects, which it then returns to the caller.
196 =head2 XML::Feed->new($format)
198 Creates a new empty I<XML::Feed> object using the format I<$format>.
200 $feed = XML::Feed->new('Atom');
201 $feed = XML::Feed->new('RSS');
202 $feed = XML::Feed->new('RSS', version => '0.91');
204 =head2 XML::Feed->parse($stream)
206 =head2 XML::Feed->parse($stream, $format)
208 Parses a syndication feed identified by I<$stream>. I<$stream> can be any
209 one of the following:
213 =item * Scalar reference
215 A reference to string containing the XML body of the feed.
219 An open filehandle from which the feed XML will be read.
223 The name of a file containing the feed XML.
227 A URI from which the feed XML will be retrieved.
231 C<$format> allows you to override format guessing.
233 =head2 XML::Feed->find_feeds($uri)
235 Given a URI I<$uri>, use auto-discovery to find all of the feeds linked
236 from that page (using I<E<lt>linkE<gt>> tags).
238 Returns a list of feed URIs.
240 =head2 $feed->convert($format)
242 Converts the I<XML::Feed> object into the I<$format> format, and returns
245 =head2 $feed->splice($other_feed)
247 Splices in all of the entries from the feed I<$other_feed> into I<$feed>,
248 skipping posts that are already in I<$feed>.
252 Returns the format of the feed (C<Atom>, or some version of C<RSS>).
254 =head2 $feed->title([ $title ])
256 The title of the feed/channel.
258 =head2 $feed->link([ $uri ])
260 The permalink of the feed/channel.
262 =head2 $feed->tagline([ $tagline ])
264 The description or tagline of the feed/channel.
266 =head2 $feed->description([ $description ])
268 Alias for I<$feed-E<gt>tagline>.
270 =head2 $feed->author([ $author ])
272 The author of the feed/channel.
274 =head2 $feed->language([ $language ])
276 The language of the feed.
278 =head2 $feed->copyright([ $copyright ])
280 The copyright notice of the feed.
282 =head2 $feed->modified([ $modified ])
284 A I<DateTime> object representing the last-modified date of the feed.
286 If present, I<$modified> should be a I<DateTime> object.
288 =head2 $feed->generator([ $generator ])
290 The generator of the feed.
292 =head2 $feed->self_link ([ $uri ])
294 The Atom Self-link of the feed:
296 L<http://validator.w3.org/feed/docs/warning/MissingAtomSelfLink.html>
300 =head2 $feed->entries
302 A list of the entries/items in the feed. Returns an array containing
303 I<XML::Feed::Entry> objects.
305 =head2 $feed->add_entry($entry)
307 Adds an entry to the feed. I<$entry> should be an I<XML::Feed::Entry>
308 object in the correct format for the feed.
312 Returns an XML representation of the feed, in the format determined by
313 the current format of the I<$feed> object.
315 =head1 PACKAGE VARIABLES
319 =item C<$XML::Feed::RSS::PREFERRED_PARSER>
321 If you want to use another RSS parser class than XML::RSS (default), you can
322 change the class by setting C<$PREFERRED_PARSER> variable in XML::Feed::RSS
325 $XML::Feed::RSS::PREFERRED_PARSER = "XML::RSS::LibXML";
327 B<Note:> this will only work for parsing feeds, not creating feeds.
333 I<XML::Feed> is free software; you may redistribute it and/or modify it
334 under the same terms as Perl itself.
336 =head1 AUTHOR & COPYRIGHT
338 Except where otherwise noted, I<XML::Feed> is Copyright 2004-2008
339 Six Apart, cpan@sixapart.com. All rights reserved.
343 The latest version of I<XML::Feed> can be found at
345 http://code.sixapart.com/svn/XML-Feed/trunk/