| XML-RSS-Feed documentation | Contained in the XML-RSS-Feed distribution. |
XML::RSS::Feed - Persistant XML RSS Encapsulation
2.32
A quick and dirty non-POE example that uses a blocking sleep. The magic is in the late_breaking_news method that returns only headlines it hasn't seen.
use XML::RSS::Feed;
use LWP::Simple qw(get);
my $feed = XML::RSS::Feed->new(
url => "http://www.jbisbee.com/rdf/",
name => "jbisbee",
delay => 10,
debug => 1,
tmpdir => "/tmp", # optional caching
);
while (1) {
$feed->parse(get($feed->url));
print $_->headline . "\n" for $feed->late_breaking_news;
sleep($feed->delay);
}
ATTENTION! - If you want a non-blocking way to watch multiple RSS sources with one process use POE::Component::RSSAggregator.
If you want to fetch a feed, mark all the headlines as seen, then get events for any new headlines, pass 'init_headlines_seen => 1' to the constructor.
Identifier and hash lookup key for the RSS feed.
The URL of the RSS feed
Number of seconds between updates (defaults to 600)
Directory to keep a cached feed (using Storable) to keep persistance between instances.
Mark all headlines as seen from the intial fetch, and only report new headlines that appear from that point forward.
Turn debuging on.
Boolean value to use the headline as the id when URL isn't unique within a feed.
A class name sublcassed from XML::RSS::Headline
The max number of headlines to keep. (default is unlimited)
Pass in a xml string to parse with XML::RSS and then call process to process the results.
Calls pre_process, process_items, post_process, title, and link methods to process the parsed results of an RSS XML feed.
An array of hash refs which will eventually become XML::RSS::Headline objects. Look at XML::RSS::Headline->new() for acceptable arguments.
The title of the RSS feed.
The RSS channel link (normally a URL back to the homepage) of the RSS feed.
Mark all headlines from previous run as seen.
Turn an array refs of hash refs into XML::RSS::Headline objects and added to the internal list of headlines.
Post process cleanup, cache headlines (if tmpdir), and debug messages.
Create a new XML::RSS::Headline object and add it to the interal list. Check XML::RSS::Headline->new() for acceptable values for %args.
After fetching a feed for the first time, mark all headlines as seen so we don't generate a flood of events. Basically don't issue an event for any existing headlines, but for any headline from that point on.
Returns the number of headlines for the feed.
Just a boolean test to see if we've seen a headline or not.
Returns an array or array reference (based on context) of XML::RSS::Headline objects
Returns an array or the number of elements (based on context) of the latest XML::RSS::Headline objects.
If tmpdir is defined the rss info is cached.
Set the time of when the feed was last processed. If you pass in a value it will be used otherwise calls Time::HiRes::time.
The time (in epoch seconds) of when the feed was last processed.
The time (in epoch seconds and milliseconds) of when the feed was last processed.
The title of the RSS feed.
Turn on debugging messages
init is used so that we just load the current headlines and don't return all headlines. in other words we initialize them. Takes a boolean argument.
The identifier of an RSS feed.
Number of seconds between updates.
The url in the RSS feed with a link back to the site where the RSS feed came from.
The url in the RSS feed with a link back to the site where the RSS feed came from.
Within some RSS feeds the URL may not always be unique, in these cases you can use the headline as the unique id. The id is used to check whether or not a feed is new or has already been seen.
Ablity to use a subclass of XML::RSS::Headline. (See Perl Jobs example in XML::RSS::Headline::PerlJobs). This should just be the name of the subclass.
Temporay directory to store cached RSS XML between instances for persistance.
Boolean value to mark all headlines as seen from the intial fetch, and only report new headlines that appear from that point forward.
The maximum number of headlines you'd like to keep track of. (0 means infinate)
This should was deprecated because, the object shouldn't really know anything about fetching, it just processes the results. This method currently will always return false
This method was deprecated because, $feed->parse now returns a bool value. This method will always return false
Jeff Bisbee, <jbisbee at cpan.org>
Please report any bugs or feature requests to
bug-xml-rss-feed at rt.cpan.org, or through the web interface at
http://rt.cpan.org/NoAuth/ReportBug.html?Queue=XML-RSS-Feed.
I will be notified, and then you'll automatically be notified of progress on
your bug as I make changes.
You can find documentation for this module with the perldoc command.
perldoc XML::RSS::Feed
You can also look for information at:
Special thanks to Rocco Caputo, Martijn van Beers, Sean Burke, Prakash Kailasa and Randal Schwartz for their help, guidance, patience, and bug reports. Guys thanks for actually taking time to use the code and give good, honest feedback.
Thank for to Carl Fürstenberg for providing feedback for new constructor param of 'init_headlines_seen' so you won't get flooded with headlines on the first fetch of the feed.
Thanks to Slaven Rezić for pointing out that t/008_store_retrieve.t pointed to broken rss tests on jbisbee.com (that I don't own anymore)
Thanks to Aaron Krowne for patch for XML::RSS::Headline to use guid as the unique id instead of url if its available.
Copyright 2006 Jeff Bisbee, all rights reserved.
This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself.
| XML-RSS-Feed documentation | Contained in the XML-RSS-Feed distribution. |
package XML::RSS::Feed; use strict; use warnings; use XML::RSS; use XML::RSS::Headline; use Time::HiRes; use Storable qw(store retrieve); use Carp qw(carp); use constant DEFAULT_DELAY => 3600; our $VERSION = 2.32; sub new { my ( $class, %args ) = @_; my $self = bless { process_count => 0, rss_headlines => [], rss_headline_ids => {}, max_headlines => 0, init_headlines_seen => 0, }, $class; foreach my $method ( keys %args ) { if ( $self->can($method) ) { $self->$method( $args{$method} ); } else { carp "Invalid argument '$method'"; } } $self->_load_cached_headlines if $self->{tmpdir}; $self->delay(DEFAULT_DELAY) unless $self->delay; return $self; } sub _load_cached_headlines { my ($self) = @_; my $filename_sto = $self->{tmpdir} . '/' . $self->name . '.sto'; my $filename_xml = $self->{tmpdir} . '/' . $self->name; if ( -s $filename_sto ) { my $cached = retrieve($filename_sto); my $title = $self->title || $cached->{title} || ''; $self->set_last_updated( $cached->{last_updated} ); $self->{process_count}++; $self->process( $cached->{items}, $title, $cached->{link} ); warn "[$self->{name}] Loaded Cached RSS Storable\n" if $self->{debug}; } elsif ( -T $filename_xml ) { # legacy XML caching if ( open( my $fh, '<', $filename_xml ) ) { my $xml = do { local $/ = undef, <$fh> }; close $fh; warn "[$self->{name}] Loaded Cached RSS XML\n" if $self->{debug}; $self->{process_count}++; $self->parse($xml); } else { carp "[$self->{name}] Failed to load XML cache $filename_xml\n"; } } else { warn "[$self->{name}] No Cache File Found\n" if $self->{debug}; } return; } sub _strip_whitespace { my ($string) = @_; $string =~ s/^\s+//; $string =~ s/\s+$//; return $string; } sub _mark_all_headlines_seen { my ($self) = @_; return unless $self->{process_count} || $self->{init_headlines_seen}; $self->{rss_headline_ids}{ $_->id } = 1 for $self->late_breaking_news; return; } sub parse { my ( $self, $xml ) = @_; my $rss = XML::RSS->new(); eval { $rss->parse($xml) }; unless ($@) { warn "[$self->{name}] Parsed RSS XML\n" if $self->{debug}; my $items = [ map { { item => $_ } } @{ $rss->{items} } ]; $self->process( $items, ( $self->title || $rss->{channel}{title} ), $rss->{channel}{link} ); return 1; } carp "[$self->{name}] [!!] Failed to parse RSS XML: $@\n"; return; } sub process { my ( $self, $items, $title, $link ) = @_; return unless $items; $self->pre_process; $self->process_items($items); $self->title($title) if $title; $self->link($link) if $link; $self->post_process; return 1; } sub pre_process { my ($self) = @_; $self->_mark_all_headlines_seen; return; } sub process_items { my ( $self, $items ) = @_; return unless $items; # used 'reverse' so order seen is preserved for my $item ( reverse @$items ) { $self->create_headline(%$item); } return 1; } sub post_process { my ($self) = @_; if ( $self->init ) { warn "[$self->{name}] " . $self->late_breaking_news . " New Headlines Found\n" if $self->{debug}; } else { $self->_mark_all_headlines_seen; $self->init(1); warn "[$self->{name}] " . $self->num_headlines . " Headlines Initialized\n" if $self->{debug}; } $self->{process_count}++; $self->cache; $self->set_last_updated; return; } sub create_headline { my ( $self, %args ) = @_; my $hlobj = $self->{hlobj} || 'XML::RSS::Headline'; $args{headline_as_id} = $self->{headline_as_id}; my $headline = $hlobj->new(%args); return unless $headline; unshift( @{ $self->{rss_headlines} }, $headline ) unless $self->seen_headline( $headline->id ); # remove the oldest if the new headline put us over the max_headlines limit if ( $self->max_headlines ) { while ( $self->num_headlines > $self->max_headlines ) { my $garbage = pop @{ $self->{rss_headlines} }; # just in case max_headlines < number of headlines in the feed $self->{rss_headline_ids}{ $garbage->id } = 1; warn "[$self->{name}] Exceeded maximum headlines, removing " . "oldest headline\n" if $self->{debug}; } } return; } sub num_headlines { my ($self) = @_; return scalar @{ $self->{rss_headlines} }; } sub seen_headline { my ( $self, $id ) = @_; return 1 if exists $self->{rss_headline_ids}{$id}; return; } sub headlines { my ($self) = @_; return wantarray ? @{ $self->{rss_headlines} } : $self->{rss_headlines}; } sub late_breaking_news { my ($self) = @_; my @list = grep { !$self->seen_headline( $_->id ); } @{ $self->{rss_headlines} }; return wantarray ? @list : scalar @list; } sub cache { my ($self) = @_; return unless $self->tmpdir; if ( -d $self->tmpdir && $self->num_headlines ) { my $tmp_filename = $self->tmpdir . '/' . $self->{name} . '.sto'; eval { store( $self->_build_dump_structure, $tmp_filename ) }; if ($@) { carp "[$self->{name}] Could not cache RSS XML to $tmp_filename\n"; return; } else { warn "[$self->{name}] Cached RSS Storable to $tmp_filename\n" if $self->{debug}; return 1; } } return; } sub _build_dump_structure { my ($self) = @_; my $cached = {}; $cached->{title} = $self->title; $cached->{link} = $self->link; $cached->{last_updated} = $self->{timestamp_hires}; $cached->{items} = []; for my $headline ( $self->headlines ) { push @{ $cached->{items} }, { headline => $headline->headline, url => $headline->url, description => $headline->description, first_seen => $headline->first_seen_hires, }; } return $cached; } sub set_last_updated { my ( $self, $hires_time ) = @_; $self->{hires_timestamp} = $hires_time if $hires_time; $self->{hires_timestamp} = Time::HiRes::time() unless $self->{hires_timestamp}; return; } sub last_updated { my ($self) = @_; return int $self->{hires_timestamp}; } sub last_updated_hires { my ($self) = @_; return $self->{hires_timestamp}; } sub title { my ( $self, $title ) = @_; if ($title) { $title = _strip_whitespace($title); $self->{title} = $title if $title; } return $self->{title}; } sub debug { my $self = shift @_; $self->{debug} = shift if @_; return $self->{debug}; } sub init { my $self = shift @_; $self->{init} = shift if @_; return $self->{init}; } sub name { my $self = shift; $self->{name} = shift if @_; return $self->{name}; } sub delay { my $self = shift @_; $self->{delay} = shift if @_; return $self->{delay}; } sub link { my $self = shift @_; $self->{link} = shift if @_; return $self->{link}; } sub url { my $self = shift @_; $self->{url} = shift if @_; return $self->{url}; } sub headline_as_id { my ( $self, $bool ) = @_; if ( defined $bool ) { $self->{headline_as_id} = $bool; $_->headline_as_id($bool) for $self->headlines; } return $self->{headline_as_id}; } sub hlobj { my ( $self, $hlobj ) = @_; $self->{hlobj} = $hlobj if defined $hlobj; return $self->{hlobj}; } sub tmpdir { my $self = shift @_; $self->{tmpdir} = shift if @_; return $self->{tmpdir}; } sub init_headlines_seen { my $self = shift @_; $self->{init_headlines_seen} = shift if @_; return $self->{init_headlines_seen}; } sub max_headlines { my $self = shift @_; $self->{max_headlines} = shift if @_; return $self->{max_headlines}; } sub failed_to_fetch { carp __PACKAGE__ . '::failed_to_fetch has been deprecated'; return; } sub failed_to_parse { carp __PACKAGE__ . '::failed_to_parse has been deprecated'; return; } 1; __END__