1 # $Id: Feed.pm 1958 2006-08-14 05:31:27Z btrott $
6 use base qw( Class::ErrorHandler );
11 use Module::Pluggable search_path => "XML::Feed::Format",
13 sub_name => 'formatters';
18 @formatters = __PACKAGE__->formatters;
23 my $format = shift || 'Atom';
24 my $format_class = 'XML::Feed::Format::' . $format;
25 eval "use $format_class";
26 Carp::croak("Unsupported format $format: $@") if $@;
27 my $feed = bless {}, join('::', __PACKAGE__, "Format", $format);
28 $feed->init_empty(@_) or return $class->error($feed->errstr);
36 my($stream, $specified_format) = @_;
37 return $class->error("Stream parameter is required") unless $stream;
38 my $feed = bless {}, $class;
40 if (UNIVERSAL::isa($stream, 'URI')) {
41 my $ua = LWP::UserAgent->new;
42 $ua->env_proxy; # force allowing of proxies
43 my $res = URI::Fetch->fetch($stream, UserAgent => $ua)
44 or return $class->error(URI::Fetch->errstr);
45 return $class->error("This feed has been permanently removed")
46 if $res->status == URI::Fetch::URI_GONE();
48 } elsif (ref($stream) eq 'SCALAR') {
50 } elsif (ref($stream)) {
51 while (read($stream, my($chunk), 8192)) {
56 or return $class->error("Can't open $stream: $!");
57 while (read $fh, my($chunk), 8192) {
62 return $class->error("Can't get feed XML content from $stream")
65 if ($specified_format) {
66 $format = $specified_format;
68 $format = $feed->identify_format(\$xml) or return $class->error($feed->errstr);
71 my $format_class = join '::', __PACKAGE__, "Format", $format;
72 eval "use $format_class";
73 return $class->error("Unsupported format $format: $@") if $@;
74 bless $feed, $format_class;
75 $feed->init_string(\$xml) or return $class->error($feed->errstr);
82 foreach my $class (@formatters) {
83 my ($name) = ($class =~ m!([^:]+)$!);
86 return $name if eval { $class->identify(\$tmp) };
87 return $feed->error($@) if $@;
89 return $feed->error("Cannot detect feed type");
97 ## Auto-detect feed type based on first element. This is prone
98 ## to breakage, but then again we don't want to parse the whole
101 while ($$xml =~ /<(\S+)/sg) {
102 (my $t = $1) =~ tr/a-zA-Z0-9:\-\?!//cd;
103 my $first = substr $t, 0, 1;
104 $tag = $t, last unless $first eq '?' || $first eq '!';
106 die ("Cannot find first element") unless $tag;
114 my @feeds = Feed::Find->find($uri)
115 or return $class->error(Feed::Find->errstr);
122 my $new = XML::Feed->new($format);
123 for my $field (qw( title link description language author copyright modified generator )) {
124 my $val = $feed->$field();
125 next unless defined $val;
128 for my $entry ($feed->entries) {
129 $new->add_entry($entry->convert($format));
137 my %ids = map { $_->id => 1 } $feed->entries;
138 for my $entry ($other->entries) {
139 $feed->add_entry($entry) unless $ids{$entry->id}++;
146 my $feed_format = ref($feed); $feed_format =~ s!^XML::Feed::Format::!!;
147 my $entry_format = ref($entry); $entry_format =~ s!^XML::Feed::Entry::Format::!!;
148 return $entry if $entry_format eq $feed_format;
149 return $entry->convert($feed_format);
168 sub tagline { shift->description(@_) }
169 sub items { $_[0]->entries }
176 XML::Feed - Syndication feed parser and auto-discovery
181 my $feed = XML::Feed->parse(URI->new('http://example.com/atom.xml'))
182 or die XML::Feed->errstr;
183 print $feed->title, "\n";
184 for my $entry ($feed->entries) {
187 ## Find all of the syndication feeds on a given page, using
189 my @feeds = XML::Feed->find_feeds('http://example.com/');
193 I<XML::Feed> is a syndication feed parser for both RSS and Atom feeds. It
194 also implements feed auto-discovery for finding feeds, given a URI.
196 I<XML::Feed> supports the following syndication feed formats:
210 The goal of I<XML::Feed> is to provide a unified API for parsing and using
211 the various syndication formats. The different flavors of RSS and Atom
212 handle data in different ways: date handling; summaries and content;
213 escaping and quoting; etc. This module attempts to remove those differences
214 by providing a wrapper around the formats and the classes implementing
215 those formats (I<XML::RSS> and I<XML::Atom::Feed>). For example, dates are
216 handled differently in each of the above formats. To provide a unified API for
217 date handling, I<XML::Feed> converts all date formats transparently into
218 I<DateTime> objects, which it then returns to the caller.
222 =head2 XML::Feed->new($format)
224 Creates a new empty I<XML::Feed> object using the format I<$format>.
226 $feed = XML::Feed->new('Atom');
227 $feed = XML::Feed->new('RSS');
228 $feed = XML::Feed->new('RSS', version => '0.91');
230 =head2 XML::Feed->parse($stream)
232 =head2 XML::Feed->parse($stream, $format)
234 Parses a syndication feed identified by I<$stream>. I<$stream> can be any
235 one of the following:
239 =item * Scalar reference
241 A reference to string containing the XML body of the feed.
245 An open filehandle from which the feed XML will be read.
249 The name of a file containing the feed XML.
253 A URI from which the feed XML will be retrieved.
257 I<$format> allows you to override format guessing.
259 =head2 XML::Feed->find_feeds($uri)
261 Given a URI I<$uri>, use auto-discovery to find all of the feeds linked
262 from that page (using I<E<lt>linkE<gt>> tags).
264 Returns a list of feed URIs.
266 =head2 XML::Feed->identify_format($xml)
268 Given the xml of a feed return what format it is in (C<Atom>, or some version of C<RSS>).
270 =head2 $feed->convert($format)
272 Converts the I<XML::Feed> object into the I<$format> format, and returns
275 =head2 $feed->splice($other_feed)
277 Splices in all of the entries from the feed I<$other_feed> into I<$feed>,
278 skipping posts that are already in I<$feed>.
282 Returns the format of the feed (C<Atom>, or some version of C<RSS>).
284 =head2 $feed->title([ $title ])
286 The title of the feed/channel.
288 =head2 $feed->base([ $base ])
290 The url base of the feed/channel.
292 =head2 $feed->link([ $uri ])
294 The permalink of the feed/channel.
296 =head2 $feed->tagline([ $tagline ])
298 The description or tagline of the feed/channel.
300 =head2 $feed->description([ $description ])
302 Alias for I<$feed-E<gt>tagline>.
304 =head2 $feed->author([ $author ])
306 The author of the feed/channel.
308 =head2 $feed->language([ $language ])
310 The language of the feed.
312 =head2 $feed->copyright([ $copyright ])
314 The copyright notice of the feed.
316 =head2 $feed->modified([ $modified ])
318 A I<DateTime> object representing the last-modified date of the feed.
320 If present, I<$modified> should be a I<DateTime> object.
322 =head2 $feed->generator([ $generator ])
324 The generator of the feed.
326 =head2 $feed->self_link ([ $uri ])
328 The Atom Self-link of the feed:
330 L<http://validator.w3.org/feed/docs/warning/MissingAtomSelfLink.html>
334 =head2 $feed->entries
336 A list of the entries/items in the feed. Returns an array containing
337 I<XML::Feed::Entry> objects.
341 A synonym for I<$feed->entries>.
343 =head2 $feed->add_entry($entry)
345 Adds an entry to the feed. I<$entry> should be an I<XML::Feed::Entry>
346 object in the correct format for the feed.
350 Returns an XML representation of the feed, in the format determined by
351 the current format of the I<$feed> object.
353 =head1 PACKAGE VARIABLES
357 =item C<$XML::Feed::RSS::PREFERRED_PARSER>
359 If you want to use another RSS parser class than XML::RSS (default), you can
360 change the class by setting C<$PREFERRED_PARSER> variable in XML::Feed::RSS
363 $XML::Feed::RSS::PREFERRED_PARSER = "XML::RSS::LibXML";
365 B<Note:> this will only work for parsing feeds, not creating feeds.
371 For reference, this cgi script will create valid, albeit nonsensical feeds
372 (according to C<http://feedvalidator.org> anyway) for Atom 1.0 and RSS 0.90,
379 use CGI::Carp qw(fatalsToBrowser);
384 my @args = ( $cgi->param('format') || "Atom" );
385 push @args, ( version => $cgi->param('version') ) if $cgi->param('version');
387 my $feed = XML::Feed->new(@args);
388 $feed->id("http://".time.rand()."/");
389 $feed->title('Test Feed');
390 $feed->link($cgi->url);
391 $feed->self_link($cgi->url( -query => 1, -full => 1, -rewrite => 1) );
392 $feed->modified(DateTime->now);
394 my $entry = XML::Feed::Entry->new();
395 $entry->id("http://".time.rand()."/");
396 $entry->link("http://example.com");
397 $entry->title("Test entry");
398 $entry->summary("Test summary");
399 $entry->content("Foo");
400 $entry->modified(DateTime->now);
401 $entry->author('test@example.com (Testy McTesterson)');
402 $feed->add_entry($entry);
404 my $mime = ("Atom" eq $feed->format) ? "application/atom+xml" : "application/rss+xml";
405 print $cgi->header($mime);
411 I<XML::Feed> is free software; you may redistribute it and/or modify it
412 under the same terms as Perl itself.
414 =head1 AUTHOR & COPYRIGHT
416 Except where otherwise noted, I<XML::Feed> is Copyright 2004-2008
417 Six Apart, cpan@sixapart.com. All rights reserved.
421 The latest version of I<XML::Feed> can be found at
423 http://code.sixapart.com/svn/XML-Feed/trunk/