4 use base qw( Class::ErrorHandler );
9 use Module::Pluggable search_path => "XML::Feed::Format",
11 sub_name => 'formatters';
13 our $VERSION = '0.49';
14 our $MULTIPLE_ENCLOSURES = 0;
17 @formatters = __PACKAGE__->formatters;
22 my $format = shift || 'Atom';
23 my $format_class = 'XML::Feed::Format::' . $format;
24 eval "use $format_class";
25 Carp::croak("Unsupported format $format: $@") if $@;
26 my $feed = bless {}, join('::', __PACKAGE__, "Format", $format);
27 $feed->init_empty(@_) or return $class->error($feed->errstr);
35 my($stream, $specified_format) = @_;
36 return $class->error("Stream parameter is required") unless $stream;
37 my $feed = bless {}, $class;
39 if (UNIVERSAL::isa($stream, 'URI')) {
40 my $ua = LWP::UserAgent->new;
41 $ua->env_proxy; # force allowing of proxies
42 my $res = URI::Fetch->fetch($stream, UserAgent => $ua)
43 or return $class->error(URI::Fetch->errstr);
44 return $class->error("This feed has been permanently removed")
45 if $res->status == URI::Fetch::URI_GONE();
47 } elsif (ref($stream) eq 'SCALAR') {
49 } elsif (ref($stream)) {
50 while (read($stream, my($chunk), 8192)) {
55 or return $class->error("Can't open $stream: $!");
56 while (read $fh, my($chunk), 8192) {
61 return $class->error("Can't get feed XML content from $stream")
64 if ($specified_format) {
65 $format = $specified_format;
67 $format = $feed->identify_format(\$xml) or return $class->error($feed->errstr);
70 my $format_class = join '::', __PACKAGE__, "Format", $format;
71 eval "use $format_class";
72 return $class->error("Unsupported format $format: $@") if $@;
73 bless $feed, $format_class;
74 $feed->init_string(\$xml) or return $class->error($feed->errstr);
81 foreach my $class (@formatters) {
82 my ($name) = ($class =~ m!([^:]+)$!);
85 return $name if eval { $class->identify(\$tmp) };
86 return $feed->error($@) if $@;
88 return $feed->error("Cannot detect feed type");
96 ## Auto-detect feed type based on first element. This is prone
97 ## to breakage, but then again we don't want to parse the whole
100 while ($$xml =~ /<(\S+)/sg) {
101 (my $t = $1) =~ tr/a-zA-Z0-9:\-\?!//cd;
102 my $first = substr $t, 0, 1;
103 $tag = $t, last unless $first eq '?' || $first eq '!';
105 die ("Cannot find first element") unless $tag;
113 my @feeds = Feed::Find->find($uri)
114 or return $class->error(Feed::Find->errstr);
121 my $new = XML::Feed->new($format);
122 for my $field (qw( title link description language author copyright modified generator )) {
123 my $val = $feed->$field();
124 next unless defined $val;
127 for my $entry ($feed->entries) {
128 $new->add_entry($entry->convert($format));
136 my %ids = map { $_->id => 1 } $feed->entries;
137 for my $entry ($other->entries) {
138 $feed->add_entry($entry) unless $ids{$entry->id}++;
145 my $feed_format = ref($feed); $feed_format =~ s!^XML::Feed::Format::!!;
146 my $entry_format = ref($entry); $entry_format =~ s!^XML::Feed::Entry::Format::!!;
147 return $entry if $entry_format eq $feed_format;
148 return $entry->convert($feed_format);
167 sub tagline { shift->description(@_) }
168 sub items { $_[0]->entries }
175 XML::Feed - Syndication feed parser and auto-discovery
180 my $feed = XML::Feed->parse(URI->new('http://example.com/atom.xml'))
181 or die XML::Feed->errstr;
182 print $feed->title, "\n";
183 for my $entry ($feed->entries) {
186 ## Find all of the syndication feeds on a given page, using
188 my @feeds = XML::Feed->find_feeds('http://example.com/');
192 I<XML::Feed> is a syndication feed parser for both RSS and Atom feeds. It
193 also implements feed auto-discovery for finding feeds, given a URI.
195 I<XML::Feed> supports the following syndication feed formats:
209 The goal of I<XML::Feed> is to provide a unified API for parsing and using
210 the various syndication formats. The different flavors of RSS and Atom
211 handle data in different ways: date handling; summaries and content;
212 escaping and quoting; etc. This module attempts to remove those differences
213 by providing a wrapper around the formats and the classes implementing
214 those formats (L<XML::RSS> and L<XML::Atom::Feed>). For example, dates are
215 handled differently in each of the above formats. To provide a unified API for
216 date handling, I<XML::Feed> converts all date formats transparently into
217 L<DateTime> objects, which it then returns to the caller.
221 =head2 XML::Feed->new($format)
223 Creates a new empty I<XML::Feed> object using the format I<$format>.
225 $feed = XML::Feed->new('Atom');
226 $feed = XML::Feed->new('RSS');
227 $feed = XML::Feed->new('RSS', version => '0.91');
229 =head2 XML::Feed->parse($stream)
231 =head2 XML::Feed->parse($stream, $format)
233 Parses a syndication feed identified by I<$stream> and returns an
234 I<XML::Feed> obhect. 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 $feed->entries
336 A list of the entries/items in the feed. Returns an array containing
337 L<XML::Feed::Entry> objects.
341 A synonym (alias) for <$feed-E<gt>entries>.
343 =head2 $feed->add_entry($entry)
345 Adds an entry to the feed. I<$entry> should be an L<XML::Feed::Entry>
346 object in the correct format for the feed.
350 Returns an XML representation of the feed, in the format determined by
351 the current format of the I<$feed> object.
353 =head1 PACKAGE VARIABLES
357 =item C<$XML::Feed::Format::RSS::PREFERRED_PARSER>
359 If you want to use another RSS parser class than XML::RSS (default), you can
360 change the class by setting C<$PREFERRED_PARSER> variable in the
361 XML::Feed::Format::RSS package.
363 $XML::Feed::Format::RSS::PREFERRED_PARSER = "XML::RSS::LibXML";
365 B<Note:> this will only work for parsing feeds, not creating feeds.
367 B<Note:> Only C<XML::RSS::LibXML> version 0.3004 is known to work at the moment.
369 =item C<$XML::Feed::MULTIPLE_ENCLOSURES>
371 Although the RSS specification states that there can be at most one enclosure per item
372 some feeds break this rule.
374 If this variable is set then C<XML::Feed> captures all of them and makes them available as a list.
376 Otherwise it returns the last enclosure parsed.
378 B<Note:> C<XML::RSS> version 1.44 is needed for this to work.
386 For reference, this cgi script will create valid, albeit nonsensical feeds
387 (according to C<http://feedvalidator.org> anyway) for Atom 1.0 and RSS 0.90,
394 use CGI::Carp qw(fatalsToBrowser);
399 my @args = ( $cgi->param('format') || "Atom" );
400 push @args, ( version => $cgi->param('version') ) if $cgi->param('version');
402 my $feed = XML::Feed->new(@args);
403 $feed->id("http://".time.rand()."/");
404 $feed->title('Test Feed');
405 $feed->link($cgi->url);
406 $feed->self_link($cgi->url( -query => 1, -full => 1, -rewrite => 1) );
407 $feed->modified(DateTime->now);
409 my $entry = XML::Feed::Entry->new();
410 $entry->id("http://".time.rand()."/");
411 $entry->link("http://example.com");
412 $entry->title("Test entry");
413 $entry->summary("Test summary");
414 $entry->content("Foo");
415 $entry->modified(DateTime->now);
416 $entry->author('test@example.com (Testy McTesterson)');
417 $feed->add_entry($entry);
419 my $mime = ("Atom" eq $feed->format) ? "application/atom+xml" : "application/rss+xml";
420 print $cgi->header($mime);
426 I<XML::Feed> is free software; you may redistribute it and/or modify it
427 under the same terms as Perl itself.
429 =head1 AUTHOR & COPYRIGHT
431 Except where otherwise noted, I<XML::Feed> is Copyright 2004-2008
432 Six Apart. All rights reserved.
436 For support contact the XML::Feed mailing list - xml-feed@perlhacks.com.
440 The latest version of I<XML::Feed> can be found at
442 http://github.com/davorg/XML-Feed