-# $Id: Feed.pm 1958 2006-08-14 05:31:27Z btrott $
+# $Id$
package XML::Feed;
use strict;
use URI::Fetch;
use LWP::UserAgent;
use Carp;
-
-our $VERSION = '0.23';
+use Module::Pluggable search_path => "XML::Feed::Format",
+ require => 1,
+ sub_name => 'formatters';
+
+our $VERSION = '0.46';
+our $MULTIPLE_ENCLOSURES = 0;
+our @formatters;
+BEGIN {
+ @formatters = __PACKAGE__->formatters;
+}
sub new {
my $class = shift;
}
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 $tag;
}
sub find_feeds {
handle data in different ways: date handling; summaries and content;
escaping and quoting; etc. This module attempts to remove those differences
by providing a wrapper around the formats and the classes implementing
-those formats (I<XML::RSS> and I<XML::Atom::Feed>). For example, dates are
+those formats (L<XML::RSS> and L<XML::Atom::Feed>). For example, dates are
handled differently in each of the above formats. To provide a unified API for
date handling, I<XML::Feed> converts all date formats transparently into
-I<DateTime> objects, which it then returns to the caller.
+L<DateTime> objects, which it then returns to the caller.
=head1 USAGE
=head2 XML::Feed->parse($stream, $format)
-Parses a syndication feed identified by I<$stream>. I<$stream> can be any
+Parses a syndication feed identified by I<$stream> and returns an
+I<XML::Feed> obhect. I<$stream> can be any
one of the following:
=over 4
=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 $feed->items
-A synonym for I<$feed->entries>.
+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 $feed->as_xml
=over 4
-=item C<$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