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 }
176 sub prev_archive_link;
177 sub next_archive_link;
184 XML::Feed - Syndication feed parser and auto-discovery
189 my $feed = XML::Feed->parse(URI->new('http://example.com/atom.xml'))
190 or die XML::Feed->errstr;
191 print $feed->title, "\n";
192 for my $entry ($feed->entries) {
195 ## Find all of the syndication feeds on a given page, using
197 my @feeds = XML::Feed->find_feeds('http://example.com/');
201 I<XML::Feed> is a syndication feed parser for both RSS and Atom feeds. It
202 also implements feed auto-discovery for finding feeds, given a URI.
204 I<XML::Feed> supports the following syndication feed formats:
218 The goal of I<XML::Feed> is to provide a unified API for parsing and using
219 the various syndication formats. The different flavors of RSS and Atom
220 handle data in different ways: date handling; summaries and content;
221 escaping and quoting; etc. This module attempts to remove those differences
222 by providing a wrapper around the formats and the classes implementing
223 those formats (L<XML::RSS> and L<XML::Atom::Feed>). For example, dates are
224 handled differently in each of the above formats. To provide a unified API for
225 date handling, I<XML::Feed> converts all date formats transparently into
226 L<DateTime> objects, which it then returns to the caller.
230 =head2 XML::Feed->new($format)
232 Creates a new empty I<XML::Feed> object using the format I<$format>.
234 $feed = XML::Feed->new('Atom');
235 $feed = XML::Feed->new('RSS');
236 $feed = XML::Feed->new('RSS', version => '0.91');
238 =head2 XML::Feed->parse($stream)
240 =head2 XML::Feed->parse($stream, $format)
242 Parses a syndication feed identified by I<$stream> and returns an
243 I<XML::Feed> obhect. I<$stream> can be any
244 one of the following:
248 =item * Scalar reference
250 A reference to string containing the XML body of the feed.
254 An open filehandle from which the feed XML will be read.
258 The name of a file containing the feed XML.
262 A URI from which the feed XML will be retrieved.
266 I<$format> allows you to override format guessing.
268 =head2 XML::Feed->find_feeds($uri)
270 Given a URI I<$uri>, use auto-discovery to find all of the feeds linked
271 from that page (using I<E<lt>linkE<gt>> tags).
273 Returns a list of feed URIs.
275 =head2 XML::Feed->identify_format($xml)
277 Given the xml of a feed return what format it is in (C<Atom>, or some version of C<RSS>).
279 =head2 $feed->convert($format)
281 Converts the I<XML::Feed> object into the I<$format> format, and returns
284 =head2 $feed->splice($other_feed)
286 Splices in all of the entries from the feed I<$other_feed> into I<$feed>,
287 skipping posts that are already in I<$feed>.
291 Returns the format of the feed (C<Atom>, or some version of C<RSS>).
293 =head2 $feed->title([ $title ])
295 The title of the feed/channel.
297 =head2 $feed->base([ $base ])
299 The url base of the feed/channel.
301 =head2 $feed->link([ $uri ])
303 The permalink of the feed/channel.
305 =head2 $feed->tagline([ $tagline ])
307 The description or tagline of the feed/channel.
309 =head2 $feed->description([ $description ])
311 Alias for I<$feed-E<gt>tagline>.
313 =head2 $feed->author([ $author ])
315 The author of the feed/channel.
317 =head2 $feed->language([ $language ])
319 The language of the feed.
321 =head2 $feed->copyright([ $copyright ])
323 The copyright notice of the feed.
325 =head2 $feed->modified([ $modified ])
327 A I<DateTime> object representing the last-modified date of the feed.
329 If present, I<$modified> should be a I<DateTime> object.
331 =head2 $feed->generator([ $generator ])
333 The generator of the feed.
335 =head2 $feed->self_link ([ $uri ])
337 The Atom Self-link of the feed:
339 L<http://validator.w3.org/feed/docs/warning/MissingAtomSelfLink.html>
343 =head2 $feed->entries
345 A list of the entries/items in the feed. Returns an array containing
346 L<XML::Feed::Entry> objects.
350 A synonym (alias) for <$feed-E<gt>entries>.
352 =head2 $feed->add_entry($entry)
354 Adds an entry to the feed. I<$entry> should be an L<XML::Feed::Entry>
355 object in the correct format for the feed.
359 Returns an XML representation of the feed, in the format determined by
360 the current format of the I<$feed> object.
362 =head2 $feed->first_link ([ $uri ])
364 The Atom First-link for feed paging and archiving (RFC 5005).
366 L<http://tools.ietf.org/html/rfc5005>
368 =head2 $feed->last_link ([ $uri ])
370 The Atom Last-link for feed paging and archiving.
372 =head2 $feed->next_link ([ $uri ])
374 The Atom Next-link for feed paging and archiving.
376 =head2 $feed->previous_link ([ $uri ])
378 The Atom Previous-link for feed paging and archiving.
380 =head2 $feed->current_link ([ $uri ])
382 The Atom Current-link for feed paging and archiving.
384 =head2 $feed->next_archive_link ([ $uri ])
386 The Atom Next-link for feed paging and archiving.
388 =head2 $feed->prev_archive_link ([ $uri ])
390 The Atom Prev-Archive-link for feed paging and archiving.
392 =head1 PACKAGE VARIABLES
396 =item C<$XML::Feed::Format::RSS::PREFERRED_PARSER>
398 If you want to use another RSS parser class than XML::RSS (default), you can
399 change the class by setting C<$PREFERRED_PARSER> variable in the
400 XML::Feed::Format::RSS package.
402 $XML::Feed::Format::RSS::PREFERRED_PARSER = "XML::RSS::LibXML";
404 B<Note:> this will only work for parsing feeds, not creating feeds.
406 B<Note:> Only C<XML::RSS::LibXML> version 0.3004 is known to work at the moment.
408 =item C<$XML::Feed::MULTIPLE_ENCLOSURES>
410 Although the RSS specification states that there can be at most one enclosure per item
411 some feeds break this rule.
413 If this variable is set then C<XML::Feed> captures all of them and makes them available as a list.
415 Otherwise it returns the last enclosure parsed.
417 B<Note:> C<XML::RSS> version 1.44 is needed for this to work.
425 For reference, this cgi script will create valid, albeit nonsensical feeds
426 (according to C<http://feedvalidator.org> anyway) for Atom 1.0 and RSS 0.90,
433 use CGI::Carp qw(fatalsToBrowser);
438 my @args = ( $cgi->param('format') || "Atom" );
439 push @args, ( version => $cgi->param('version') ) if $cgi->param('version');
441 my $feed = XML::Feed->new(@args);
442 $feed->id("http://".time.rand()."/");
443 $feed->title('Test Feed');
444 $feed->link($cgi->url);
445 $feed->self_link($cgi->url( -query => 1, -full => 1, -rewrite => 1) );
446 $feed->modified(DateTime->now);
448 my $entry = XML::Feed::Entry->new();
449 $entry->id("http://".time.rand()."/");
450 $entry->link("http://example.com");
451 $entry->title("Test entry");
452 $entry->summary("Test summary");
453 $entry->content("Foo");
454 $entry->modified(DateTime->now);
455 $entry->author('test@example.com (Testy McTesterson)');
456 $feed->add_entry($entry);
458 my $mime = ("Atom" eq $feed->format) ? "application/atom+xml" : "application/rss+xml";
459 print $cgi->header($mime);
465 I<XML::Feed> is free software; you may redistribute it and/or modify it
466 under the same terms as Perl itself.
468 =head1 AUTHOR & COPYRIGHT
470 Except where otherwise noted, I<XML::Feed> is Copyright 2004-2008
471 Six Apart. All rights reserved.
475 For support contact the XML::Feed mailing list - xml-feed@perlhacks.com.
479 The latest version of I<XML::Feed> can be found at
481 http://github.com/davorg/XML-Feed