-# $Id: Feed.pm 1920 2006-02-28 00:00:56Z btrott $
-
package XML::Feed;
use strict;
use base qw( Class::ErrorHandler );
use Feed::Find;
use URI::Fetch;
+use LWP::UserAgent;
use Carp;
-
-our $VERSION = '0.08';
+use Module::Pluggable search_path => "XML::Feed::Format",
+ require => 1,
+ sub_name => 'formatters';
+
+our $VERSION = '0.49';
+our $MULTIPLE_ENCLOSURES = 0;
+our @formatters;
+BEGIN {
+ @formatters = __PACKAGE__->formatters;
+}
sub new {
my $class = shift;
- my($format) = @_;
- $format ||= 'Atom';
- my $format_class = 'XML::Feed::' . $format;
+ my $format = shift || 'Atom';
+ my $format_class = 'XML::Feed::Format::' . $format;
eval "use $format_class";
Carp::croak("Unsupported format $format: $@") if $@;
- my $feed = bless {}, join('::', __PACKAGE__, $format);
- $feed->init_empty or return $class->error($feed->errstr);
+ my $feed = bless {}, join('::', __PACKAGE__, "Format", $format);
+ $feed->init_empty(@_) or return $class->error($feed->errstr);
$feed;
}
sub parse {
my $class = shift;
- my($stream) = @_;
+ my($stream, $specified_format) = @_;
return $class->error("Stream parameter is required") unless $stream;
my $feed = bless {}, $class;
my $xml = '';
if (UNIVERSAL::isa($stream, 'URI')) {
- my $res = URI::Fetch->fetch($stream)
+ my $ua = LWP::UserAgent->new;
+ $ua->env_proxy; # force allowing of proxies
+ my $res = URI::Fetch->fetch($stream, UserAgent => $ua)
or return $class->error(URI::Fetch->errstr);
return $class->error("This feed has been permanently removed")
if $res->status == URI::Fetch::URI_GONE();
}
return $class->error("Can't get feed XML content from $stream")
unless $xml;
- my $format = $feed->identify_format(\$xml)
- or return $class->error($feed->errstr);
- my $format_class = join '::', __PACKAGE__, $format;
+ my $format;
+ if ($specified_format) {
+ $format = $specified_format;
+ } else {
+ $format = $feed->identify_format(\$xml) or return $class->error($feed->errstr);
+ }
+
+ my $format_class = join '::', __PACKAGE__, "Format", $format;
eval "use $format_class";
return $class->error("Unsupported format $format: $@") if $@;
bless $feed, $format_class;
}
sub identify_format {
- my $feed = shift;
- my($xml) = @_;
+ my $feed = shift;
+ my($xml) = @_;
+ foreach my $class (@formatters) {
+ my ($name) = ($class =~ m!([^:]+)$!);
+ # TODO ugly
+ my $tmp = $$xml;
+ return $name if eval { $class->identify(\$tmp) };
+ return $feed->error($@) if $@;
+ }
+ return $feed->error("Cannot detect feed type");
+}
+
+sub _get_first_tag {
+ my $class = shift;
+ my ($xml) = @_;
+
+
## Auto-detect feed type based on first element. This is prone
## to breakage, but then again we don't want to parse the whole
## feed ourselves.
my $first = substr $t, 0, 1;
$tag = $t, last unless $first eq '?' || $first eq '!';
}
- return $feed->error("Cannot find first element") unless $tag;
+ die ("Cannot find first element") unless $tag;
$tag =~ s/^.*://;
- if ($tag eq 'rss' || $tag eq 'RDF') {
- return 'RSS';
- } elsif ($tag eq 'feed') {
- return 'Atom';
- } else {
- return $feed->error("Cannot detect feed type");
- }
+ return $tag;
}
sub find_feeds {
sub convert {
my $feed = shift;
my($format) = @_;
- my $new = __PACKAGE__->new($format);
+ my $new = XML::Feed->new($format);
for my $field (qw( title link description language author copyright modified generator )) {
my $val = $feed->$field();
next unless defined $val;
}
}
+sub _convert_entry {
+ my $feed = shift;
+ my $entry = shift;
+ my $feed_format = ref($feed); $feed_format =~ s!^XML::Feed::Format::!!;
+ my $entry_format = ref($entry); $entry_format =~ s!^XML::Feed::Entry::Format::!!;
+ return $entry if $entry_format eq $feed_format;
+ return $entry->convert($feed_format);
+}
+
+sub base;
sub format;
sub title;
sub link;
+sub self_link;
sub description;
sub language;
sub author;
sub add_entry;
sub entries;
sub as_xml;
+sub id;
+sub image;
sub tagline { shift->description(@_) }
sub items { $_[0]->entries }
+# RFC 5005
+sub first_link;
+sub last_link;
+sub previous_link;
+sub next_link;
+sub current_link;
+sub prev_archive_link;
+sub next_archive_link;
+
1;
__END__
handle data in different ways: date handling; summaries and content;
escaping and quoting; etc. This module attempts to remove those differences
by providing a wrapper around the formats and the classes implementing
-those formats (I<XML::RSS> and I<XML::Atom::Feed>). For example, dates are
+those formats (L<XML::RSS> and L<XML::Atom::Feed>). For example, dates are
handled differently in each of the above formats. To provide a unified API for
date handling, I<XML::Feed> converts all date formats transparently into
-I<DateTime> objects, which it then returns to the caller.
+L<DateTime> objects, which it then returns to the caller.
=head1 USAGE
Creates a new empty I<XML::Feed> object using the format I<$format>.
+ $feed = XML::Feed->new('Atom');
+ $feed = XML::Feed->new('RSS');
+ $feed = XML::Feed->new('RSS', version => '0.91');
+
=head2 XML::Feed->parse($stream)
-Parses a syndication feed identified by I<$stream>. I<$stream> can be any
+=head2 XML::Feed->parse($stream, $format)
+
+Parses a syndication feed identified by I<$stream> and returns an
+I<XML::Feed> obhect. I<$stream> can be any
one of the following:
=over 4
=back
+I<$format> allows you to override format guessing.
+
=head2 XML::Feed->find_feeds($uri)
Given a URI I<$uri>, use auto-discovery to find all of the feeds linked
Returns a list of feed URIs.
+=head2 XML::Feed->identify_format($xml)
+
+Given the xml of a feed return what format it is in (C<Atom>, or some version of C<RSS>).
+
=head2 $feed->convert($format)
Converts the I<XML::Feed> object into the I<$format> format, and returns
The title of the feed/channel.
+=head2 $feed->base([ $base ])
+
+The url base of the feed/channel.
+
=head2 $feed->link([ $uri ])
The permalink of the feed/channel.
The generator of the feed.
+=head2 $feed->self_link ([ $uri ])
+
+The Atom Self-link of the feed:
+
+L<http://validator.w3.org/feed/docs/warning/MissingAtomSelfLink.html>
+
+A string.
+
=head2 $feed->entries
A list of the entries/items in the feed. Returns an array containing
-I<XML::Feed::Entry> objects.
+L<XML::Feed::Entry> objects.
+
+=head2 $feed->items
+
+A synonym (alias) for <$feed-E<gt>entries>.
=head2 $feed->add_entry($entry)
-Adds an entry to the feed. I<$entry> should be an I<XML::Feed::Entry>
+Adds an entry to the feed. I<$entry> should be an L<XML::Feed::Entry>
object in the correct format for the feed.
=head2 $feed->as_xml
Returns an XML representation of the feed, in the format determined by
the current format of the I<$feed> object.
+=head2 $feed->first_link ([ $uri ])
+
+The Atom First-link for feed paging and archiving (RFC 5005).
+
+L<http://tools.ietf.org/html/rfc5005>
+
+=head2 $feed->last_link ([ $uri ])
+
+The Atom Last-link for feed paging and archiving.
+
+=head2 $feed->next_link ([ $uri ])
+
+The Atom Next-link for feed paging and archiving.
+
+=head2 $feed->previous_link ([ $uri ])
+
+The Atom Previous-link for feed paging and archiving.
+
+=head2 $feed->current_link ([ $uri ])
+
+The Atom Current-link for feed paging and archiving.
+
+=head2 $feed->next_archive_link ([ $uri ])
+
+The Atom Next-link for feed paging and archiving.
+
+=head2 $feed->prev_archive_link ([ $uri ])
+
+The Atom Prev-Archive-link for feed paging and archiving.
+
=head1 PACKAGE VARIABLES
=over 4
-=item C<$XML::Feed::RSS::PREFERRED_PARSER>
+=item C<$XML::Feed::Format::RSS::PREFERRED_PARSER>
If you want to use another RSS parser class than XML::RSS (default), you can
-change the class by setting C<$PREFERRED_PARSER> variable in XML::Feed::RSS
-package.
+change the class by setting C<$PREFERRED_PARSER> variable in the
+XML::Feed::Format::RSS package.
- $XML::Feed::RSS::PREFERRED_PARSER = "XML::RSS::LibXML";
+ $XML::Feed::Format::RSS::PREFERRED_PARSER = "XML::RSS::LibXML";
B<Note:> this will only work for parsing feeds, not creating feeds.
+B<Note:> Only C<XML::RSS::LibXML> version 0.3004 is known to work at the moment.
+
+=item C<$XML::Feed::MULTIPLE_ENCLOSURES>
+
+Although the RSS specification states that there can be at most one enclosure per item
+some feeds break this rule.
+
+If this variable is set then C<XML::Feed> captures all of them and makes them available as a list.
+
+Otherwise it returns the last enclosure parsed.
+
+B<Note:> C<XML::RSS> version 1.44 is needed for this to work.
+
=back
+=cut
+
+=head1 VALID FEEDS
+
+For reference, this cgi script will create valid, albeit nonsensical feeds
+(according to C<http://feedvalidator.org> anyway) for Atom 1.0 and RSS 0.90,
+0.91, 1.0 and 2.0.
+
+ #!perl -w
+
+ use strict;
+ use CGI;
+ use CGI::Carp qw(fatalsToBrowser);
+ use DateTime;
+ use XML::Feed;
+
+ my $cgi = CGI->new;
+ my @args = ( $cgi->param('format') || "Atom" );
+ push @args, ( version => $cgi->param('version') ) if $cgi->param('version');
+
+ my $feed = XML::Feed->new(@args);
+ $feed->id("http://".time.rand()."/");
+ $feed->title('Test Feed');
+ $feed->link($cgi->url);
+ $feed->self_link($cgi->url( -query => 1, -full => 1, -rewrite => 1) );
+ $feed->modified(DateTime->now);
+
+ my $entry = XML::Feed::Entry->new();
+ $entry->id("http://".time.rand()."/");
+ $entry->link("http://example.com");
+ $entry->title("Test entry");
+ $entry->summary("Test summary");
+ $entry->content("Foo");
+ $entry->modified(DateTime->now);
+ $entry->author('test@example.com (Testy McTesterson)');
+ $feed->add_entry($entry);
+
+ my $mime = ("Atom" eq $feed->format) ? "application/atom+xml" : "application/rss+xml";
+ print $cgi->header($mime);
+ print $feed->as_xml;
+
+
=head1 LICENSE
I<XML::Feed> is free software; you may redistribute it and/or modify it
=head1 AUTHOR & COPYRIGHT
-Except where otherwise noted, I<XML::Feed> is Copyright 2004-2005
-Six Apart, cpan@sixapart.com. All rights reserved.
+Except where otherwise noted, I<XML::Feed> is Copyright 2004-2008
+Six Apart. All rights reserved.
+
+=head1 SUPPORT
+
+For support contact the XML::Feed mailing list - xml-feed@perlhacks.com.
+
+=head1 SOURCE CODE
+
+The latest version of I<XML::Feed> can be found at
+
+ http://github.com/davorg/XML-Feed
=cut