Updated Syndicated Content to use new improved RSS feed reader.
This commit is contained in:
JT Smith
parent
40927d91aa
commit
c53feaf14f
2 changed files with 327 additions and 326 deletions
|
|
@ -86,8 +86,8 @@ sub www_view {
|
||||||
unless (defined $rssFile) {
|
unless (defined $rssFile) {
|
||||||
$rssFile = $cache->setByHTTP($_[0]->get("rssUrl"),3600);
|
$rssFile = $cache->setByHTTP($_[0]->get("rssUrl"),3600);
|
||||||
}
|
}
|
||||||
$rssFile =~ s#(<title>)(.*?)(</title>)#$1.encode_entities($2).$3#ges;
|
$rssFile =~ s#(<title>)(.*?)(</title>)#$1.encode_entities(decode_entities($2)).$3#ges;
|
||||||
eval{parseXML(\%rss, \$rssFile)};
|
eval{parseRSS(\%rss, \$rssFile)};
|
||||||
if ($@) {
|
if ($@) {
|
||||||
WebGUI::ErrorHandler::warn($_[0]->get("rssUrl")." ".$@);
|
WebGUI::ErrorHandler::warn($_[0]->get("rssUrl")." ".$@);
|
||||||
}
|
}
|
||||||
|
|
@ -96,7 +96,7 @@ sub www_view {
|
||||||
$var{"channel.link"} = $rss{link};
|
$var{"channel.link"} = $rss{link};
|
||||||
$var{"channel.description"} = $rss{description};
|
$var{"channel.description"} = $rss{description};
|
||||||
my @items;
|
my @items;
|
||||||
foreach my $item (@{$rss{items}}) {
|
foreach my $item (@{$rss{item}}) {
|
||||||
push (@items,{
|
push (@items,{
|
||||||
link=>$item->{link},
|
link=>$item->{link},
|
||||||
title=>$item->{title},
|
title=>$item->{title},
|
||||||
|
|
|
||||||
|
|
@ -1,334 +1,299 @@
|
||||||
#!/usr/bin/perl
|
#!/usr/bin/perl
|
||||||
package XML::RSSLite;
|
package XML::RSSLite;
|
||||||
use strict;
|
use strict;
|
||||||
use vars qw(@ISA @EXPORT @EXPORT_OK %EXPORT_TAGS $VERSION);
|
use vars qw($VERSION);
|
||||||
use Exporter;
|
|
||||||
@ISA = ('Exporter');
|
|
||||||
@EXPORT = qw/parseXML usableXML/;
|
|
||||||
@EXPORT_OK = qw/parseXML usableXML
|
|
||||||
isRSS isRDF isSN isWL
|
|
||||||
xml_content_string xml_content_array/;
|
|
||||||
$VERSION = '0.08';
|
|
||||||
|
|
||||||
|
$VERSION = 0.11;
|
||||||
|
|
||||||
sub parseXML {
|
sub import{
|
||||||
my ($rr, $cr) = @_;
|
no strict 'refs';
|
||||||
|
shift;
|
||||||
die "Parms to 'parse' must be refs to a hash and XML content!"
|
my $pkg = scalar caller();
|
||||||
unless (ref($rr) and ref($cr));
|
*{"${pkg}::parseRSS"} = \&parseRSS;
|
||||||
|
*{"${pkg}::parseXML"} = \&parseXML if grep($_ eq 'parseXML', @_);
|
||||||
return unless $$cr; ## Gotta have some content to parse
|
|
||||||
|
|
||||||
my $type = usableXML($cr)
|
|
||||||
or die "Content must be RSS/RDF/ScriptingNews/Weblog XML " .
|
|
||||||
"(or something pretty close)";
|
|
||||||
|
|
||||||
preprocess($cr);
|
|
||||||
|
|
||||||
if ($type == 1 or $type == 2) {
|
|
||||||
parseRSS($rr, $cr);
|
|
||||||
} elsif ($type == 3) {
|
|
||||||
parseSN($rr, $cr);
|
|
||||||
} elsif ($type == 4) {
|
|
||||||
parseWL($rr, $cr);
|
|
||||||
} else {
|
|
||||||
die "Screwed up XML type-checking somehow!";
|
|
||||||
}
|
}
|
||||||
|
|
||||||
|
|
||||||
|
sub parseRSS {
|
||||||
|
my ($rr, $cref) = @_;
|
||||||
|
|
||||||
|
die "$rr is not a hash reference" unless ref($rr) eq 'HASH';
|
||||||
|
die "$cref is not a scalar reference" unless ref($cref) eq 'SCALAR';
|
||||||
|
|
||||||
|
# Gotta have some content to parse
|
||||||
|
return unless $$cref;
|
||||||
|
|
||||||
|
preprocess($cref);
|
||||||
|
{
|
||||||
|
_parseRSS($rr, $cref), last if index(${$cref}, '<rss')+1;
|
||||||
|
_parseRDF($rr, $cref), last if index(${$cref}, '<rdf:RDF')+1;
|
||||||
|
_parseSN( $rr, $cref), last if index(${$cref}, '<scriptingnews')+1;
|
||||||
|
_parseWL( $rr, $cref), last if index(${$cref}, '<weblog')+1;
|
||||||
|
die "Content must be RSS|RDF|ScriptingNews|Weblog|reasonably close";
|
||||||
|
}
|
||||||
postprocess($rr);
|
postprocess($rr);
|
||||||
}
|
}
|
||||||
|
|
||||||
sub preprocess {
|
sub preprocess {
|
||||||
my $cr = shift;
|
my $cref = shift;
|
||||||
|
$$cref =~ y/\r\n/\n/s;
|
||||||
##
|
$$cref =~ y{\n\t ~0-9\-+!@#$%^&*()_=a-zA-Z[]\\;':",./<>?}{ }cs;
|
||||||
## Help create "well-formed" XML so parser doesn't puke by
|
#XXX $$cref =~ s/&(?!0[a-zA-Z0-9]+|#\d+);/amp/gs;
|
||||||
## 1. Making unix-style line endings
|
#XXX Do we wish to (re)allow escaped HTML?!
|
||||||
## 2. Using & for & (this screws up urls, but we fix it later)
|
$$cref =~ s{(?:<|<)/?(?:b|i|h\d|p|center|quote|strong)(?:>|>)}{}gsi;
|
||||||
## 3. Removing objectionable characters
|
|
||||||
##
|
|
||||||
$$cr =~ s|<(/*)rss\d+:(.*?)>|<$1$2>|g;
|
|
||||||
$$cr =~ s|<([^<> ]+)\s+(.+?)\s+/>|<$1 $2></$1>|g;
|
|
||||||
$$cr =~ s/[\012\015]{1,2}/\n/g;
|
|
||||||
$$cr =~ s/&(?!([a-zA-Z0-9]+|#\d+);)/&/g;
|
|
||||||
$$cr =~ y/\~\[\n 0-9a-zA-Z_\!\@\#\$\%\^\&\*\(\)\-\+\=\:\;\"\'\<\>\,\.\/?\]/ /c;
|
|
||||||
|
|
||||||
## Tidy up for debugging by starting open tags on new line
|
|
||||||
# $content =~ s|(?!\n)<(?!/)|\n<|gs;
|
|
||||||
}
|
}
|
||||||
|
|
||||||
|
sub _parseRSS {
|
||||||
|
parseXML($_[0], $_[1], 'channel', 0);
|
||||||
|
$_[0]->{'items'} = $_[0]->{'item'};
|
||||||
|
}
|
||||||
|
|
||||||
|
sub _parseRDF {
|
||||||
|
my ($rr, $cref) = @_;
|
||||||
|
|
||||||
|
$rr->{'items'} = [];
|
||||||
|
my $item;
|
||||||
|
|
||||||
|
parseXML($_[0], $_[1], 'rdf:RDF', 0);
|
||||||
|
|
||||||
|
# Alias RDF to RSS
|
||||||
|
if( exists($rr->{'item'}) ){
|
||||||
|
$rr->{'items'} = $rr->{'item'};
|
||||||
|
}
|
||||||
|
else{
|
||||||
|
my $li = $_[0]->{'rdf:li'} || $_[0]->{'rdf:Seq'}->{'rdf:li'};
|
||||||
|
foreach $item ( @{$li} ){
|
||||||
|
my %ia;
|
||||||
|
if (exists $item->{'dc:description'}) {
|
||||||
|
$ia{'description'} = $item->{'dc:description'};
|
||||||
|
}
|
||||||
|
if (exists $item->{'dc:title'}) {
|
||||||
|
$ia{'title'} = $item->{'dc:title'};
|
||||||
|
}
|
||||||
|
if (exists $item->{'dc:identifier'}) {
|
||||||
|
$ia{'link'} = delete($item->{'dc:identifier'});
|
||||||
|
}
|
||||||
|
|
||||||
|
push(@{$rr->{'items'}}, \%ia);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
sub _parseSN {
|
||||||
|
my ($rr, $cref) = @_;
|
||||||
|
|
||||||
|
$rr->{'items'} = ();
|
||||||
|
my $item;
|
||||||
|
|
||||||
|
parseXML($rr, $cref, 'channel', 0);
|
||||||
|
|
||||||
|
# Alias SN to RSS terms
|
||||||
|
foreach $item ( @{$_[0]->{'rdf:li'}} ){
|
||||||
|
my %ia;
|
||||||
|
if (exists $item->{'text'}) {
|
||||||
|
$ia{'description'} = $item->{'text'};
|
||||||
|
}
|
||||||
|
if (exists $item->{'linetext'}) {
|
||||||
|
$ia{'title'} = $item->{'linetext'};
|
||||||
|
}
|
||||||
|
if (exists $item->{'url'}) {
|
||||||
|
$ia{'link'} = $item->{'url'};
|
||||||
|
}
|
||||||
|
|
||||||
|
push(@{$rr->{'items'}}, \%ia);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
|
||||||
|
sub _parseWL {
|
||||||
|
my ($rr, $cref) = @_;
|
||||||
|
|
||||||
|
$rr->{'items'} = ();
|
||||||
|
my $item;
|
||||||
|
|
||||||
|
#XXX is this the right tag to parse for?
|
||||||
|
parseXML($rr, $cref, 'channel', 0);
|
||||||
|
|
||||||
|
# Alias WL to RSS
|
||||||
|
foreach $item ( @{$_[0]->{'rdf:li'}} ){
|
||||||
|
my %ia;
|
||||||
|
if (exists $item->{'url'}) {
|
||||||
|
$ia{'link'} = delete($item->{'url'});
|
||||||
|
}
|
||||||
|
|
||||||
|
push(@{$rr->{'items'}}, \%ia);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
|
||||||
sub postprocess {
|
sub postprocess {
|
||||||
my $rr = shift;
|
my $rr = shift;
|
||||||
|
|
||||||
$rr->{'link'} =~ s/&/&/gi;
|
#XXX Not much to do, what about un-munging URL's in source, etc.?!
|
||||||
|
return unless defined($rr->{'items'});
|
||||||
|
$rr->{'items'} = [$rr->{'items'}] unless ref($rr->{'items'}) eq 'ARRAY';
|
||||||
|
|
||||||
if (defined($rr->{'items'})) {
|
foreach my $i (@{$rr->{'items'}}) {
|
||||||
my $i;
|
$i->{description} = $i->{description}->{'<>'} if ref($i->{description});
|
||||||
|
|
||||||
foreach $i (@{$rr->{'items'}}) {
|
|
||||||
$i->{'link'} = trim($i->{'link'});
|
|
||||||
|
|
||||||
# Put stuff into the right name if necessary
|
# Put stuff into the right name if necessary
|
||||||
if (defined($i->{'url'}) and not $i->{'link'}) {
|
|
||||||
$i->{'link'} = $i->{'url'};
|
|
||||||
}
|
|
||||||
|
|
||||||
# Fix pre-process munging
|
|
||||||
$i->{'link'} =~ s/&/&/gi;
|
|
||||||
|
|
||||||
# See if you can use misplaced url in title for empty links
|
|
||||||
if( not $i->{'link'} ){
|
if( not $i->{'link'} ){
|
||||||
|
if( defined($i->{'url'}) ){
|
||||||
|
$i->{'link'} = delete($i->{'url'}); }
|
||||||
|
# See if you can use misplaced url in title for empty links
|
||||||
|
elsif( exists($i->{'title'}) ){
|
||||||
|
# The next case would trap this, but try to short-circuit the gathering
|
||||||
if ($i->{'title'} =~ /^(?:ht)|ftp:/) {
|
if ($i->{'title'} =~ /^(?:ht)|ftp:/) {
|
||||||
$i->{'link'} = $i->{'title'};
|
$i->{'link'} = $i->{'title'};
|
||||||
} elsif ($i->{'title'} =~ /"((?:ht)|ftp.*?)"/) {
|
}
|
||||||
|
elsif ($i->{'title'} =~ /"((?:ht)|ftp.*?)"/) {
|
||||||
$i->{'link'} = $1;
|
$i->{'link'} = $1;
|
||||||
$i->{'title'} =~ s/<.*?>//;
|
$i->{'title'} =~ s/<.*?>//;
|
||||||
} else {
|
}
|
||||||
|
else {
|
||||||
next;
|
next;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
}
|
||||||
|
|
||||||
# Make sure you've got an http/ftp link
|
# Make sure you've got an http/ftp link
|
||||||
if ($i->{'link'} !~ m{^(http|ftp)://}i) {
|
if( exists( $i->{'link'}) && $i->{'link'} !~ m{^(http|ftp)://}i) {
|
||||||
## Rip link out of anchor tag
|
## Rip link out of anchor tag
|
||||||
$i->{'link'} =~ m{a\s+href=("|")?(.*?)("|>|"|>)?}i;
|
$i->{'link'} =~ m{a\s+href=("|")?(.*?)("|>|"|>)?}i;
|
||||||
if( $2 ){
|
if( $2 ){
|
||||||
$i->{'link'} = $2;
|
$i->{'link'} = $2;
|
||||||
|
}
|
||||||
} elsif ($i->{'link'} =~ m{[\.#/]}i and
|
elsif( $i->{'link'} =~ m{[\.#/]}i and $rr->{'link'} =~ m{^http://} ){
|
||||||
$rr->{'link'} =~ m{^http://}) {
|
|
||||||
## Smells like a relative url
|
## Smells like a relative url
|
||||||
if (substr($i->{'link'}, 0, 1) ne '/') {
|
if (substr($i->{'link'}, 0, 1) ne '/') {
|
||||||
$i->{'link'} = '/' . $i->{'link'};
|
$i->{'link'} = '/' . $i->{'link'};
|
||||||
}
|
}
|
||||||
$i->{'link'} = $rr->{'link'} . $i->{'link'};
|
$i->{'link'} = $rr->{'link'} . $i->{'link'};
|
||||||
|
}
|
||||||
} else {
|
else {
|
||||||
next;
|
next;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
#If we don't have a title, use the link
|
#If we don't have a title, use the link
|
||||||
if (not $i->{'title'}) {
|
unless( defined($i->{'title'}) ){
|
||||||
$i->{'title'} = $i->{'link'};
|
$i->{'title'} = $i->{'link'};
|
||||||
}
|
}
|
||||||
$i->{'link'} =~ s/ //g;
|
|
||||||
|
if( exists($i->{'link'}) ){
|
||||||
|
#XXX # Fix pre-process munging
|
||||||
|
# $i->{'link'} =~ s/&/&/gi;
|
||||||
|
$i->{'link'} =~ s/ /%20/g;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
sub parseRSS {
|
sub parseXML{
|
||||||
my ($rr, $cr) = @_;
|
my($hash, $xml, $tag, $comments) = @_;
|
||||||
|
my($begin, $end, @comments);
|
||||||
|
local $_;
|
||||||
|
|
||||||
my $channel = xml_content_string('channel', $cr);
|
#Kill comments
|
||||||
$channel =~ s|<item.*?</item>||gis;
|
while( ($begin = index(${$xml}, '<!--')) > -1 &&
|
||||||
clean(\$channel);
|
${$xml} =~ m%<!--.*?--(>)%sg ){
|
||||||
|
my $str = substr(${$xml}, $begin, pos(${$xml})-$begin, '');
|
||||||
|
|
||||||
my $ca;
|
#Save them if requested
|
||||||
my @channel_attrs = ($channel =~ m|(<.*?>.*?</.*?>)|gi);
|
do{ unshift @comments, [$begin, substr($str, 4, length($str)-7)] }
|
||||||
foreach $ca (@channel_attrs) {
|
if $comments;
|
||||||
$ca =~ m|^<(.*?)>(.*?)</.*?>$|;
|
|
||||||
$rr->{$1} = trim($2);
|
|
||||||
}
|
}
|
||||||
|
|
||||||
$rr->{'items'} = ();
|
_parseXML($hash, $xml, $tag);
|
||||||
my $item;
|
|
||||||
foreach $item (xml_content_array('item', $cr)) {
|
# #XXX Context of comment is lost!
|
||||||
clean(\$item);
|
# #Expose comments if requested
|
||||||
my @item_attrs = ($item =~ m|(<.*?>.*?</.*?>)|gi);
|
# do{ push(@$comments, $_->[1]) for @comments } if ref($comments) eq 'ARRAY';
|
||||||
my $ia;
|
if( $comments ){
|
||||||
my %ia;
|
#Restore comments if requested
|
||||||
foreach $ia (@item_attrs) {
|
substr(${$xml}, $_->[0], 0, '<!--'.$_->[1].'-->') for @comments;
|
||||||
$ia =~ m|^<(.*?)>(.*?)</.*?>$|;
|
|
||||||
$ia{$1} = trim($2);
|
#Expose comments if requested
|
||||||
}
|
do{ push(@$comments, $_->[1]) for @comments } if ref($comments) eq 'ARRAY';
|
||||||
push(@{$rr->{'items'}}, \%ia);
|
|
||||||
#XXX test blanks here
|
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
sub parseSN {
|
sub _parseXML{
|
||||||
my ($rr, $cr) = @_;
|
my($hash, $xml, $tag, $index) = @_;
|
||||||
|
my($begin, $end);
|
||||||
|
|
||||||
my $channel = xml_content_string('header', $cr);
|
#Find topTag and set pos to start matching from there
|
||||||
$channel =~ s|<item.*?</item>||gis;
|
${$xml} =~ /<$tag(?:>|\s)/g;
|
||||||
clean(\$channel);
|
($begin, $end) = (0, pos(${$xml})||0);
|
||||||
|
|
||||||
my $ca;
|
#Match either <foo></foo> or <bar />, optional attributes, stash tag name
|
||||||
my @channel_attrs = ($channel =~ m|(<.*?>.*?</.*?>)|gi);
|
while( ${$xml} =~ m%<([^\s>]+)(?:\s+[^>]*?)?(?:/|>.*?</\1)>%sg ){
|
||||||
foreach $ca (@channel_attrs) {
|
|
||||||
$ca =~ m|^<(.*?)>(.*?)</.*?>$|;
|
#Save the tag name, we'll need it
|
||||||
$rr->{$1} = trim($2);
|
$tag = $1 || $2;
|
||||||
|
|
||||||
|
#Save the new beginning and end
|
||||||
|
($begin, $end) = ($end, pos(${$xml}));
|
||||||
|
|
||||||
|
#Get the bit we just matched.
|
||||||
|
my $str = substr(${$xml}, $begin, $end-$begin);
|
||||||
|
|
||||||
|
#Extract the actual attributes and contents of the tag
|
||||||
|
$str =~ m%<\Q$tag\E\s*([^>]*?)?>(.*?)</\Q$tag\E>%s ||
|
||||||
|
#XXX pointed out by hv
|
||||||
|
# $str =~ s%^.*?<$tag\s*([^>]*?)?>(.*?)</$tag>%<$tag>$2</$tag>%s ||
|
||||||
|
$str =~ m%<\Q$tag\E\s*([^>]*?)?\s*/>%;
|
||||||
|
my($attr, $content) = ($1, $2);
|
||||||
|
|
||||||
|
#Did we get attributes? clean them up and chuck them in a hash.
|
||||||
|
if( $attr ){
|
||||||
|
($_, $attr) = ($attr, {});
|
||||||
|
$attr->{$1} = $3 while m/([^\s=]+)\s*=\s*(['"])(.*?)\2/g;
|
||||||
}
|
}
|
||||||
|
|
||||||
##
|
my $inhash;
|
||||||
## Alias SN to RSS terms
|
#Recurse if contents has more tags, replace contents with reference we get
|
||||||
##
|
if( $content && index($content, '<') > -1 ){
|
||||||
if (exists $rr->{'channelDescription'}) {
|
_parseXML($inhash={}, \$str, $tag);
|
||||||
$rr->{'description'} = $rr->{'channelDescription'};
|
#Was there any data in the contents? We should extract that...
|
||||||
}
|
if( $str =~ />[^><]+</ ){
|
||||||
if (exists $rr->{'channelTitle'}) {
|
#The odd RE above shortcircuits unnecessary entry
|
||||||
$rr->{'title'} = $rr->{'channelTitle'};
|
|
||||||
}
|
|
||||||
if (exists $rr->{'channelLink'}) {
|
|
||||||
$rr->{'link'} = $rr->{'channelLink'};
|
|
||||||
}
|
|
||||||
|
|
||||||
$rr->{'items'} = ();
|
#Clean whitespace between tags
|
||||||
my $item;
|
#$str =~ s%(?<=>)?\s*(?=<)%%g; #XXX ~same speed, wacko warning
|
||||||
foreach $item (xml_content_array('item', $cr)) {
|
#$str =~ s%(>?)\s*<%$1<%g;
|
||||||
clean(\$item);
|
#XXX #$str =~ s%(?:^|(?<=>))\s*(?:(?=<)|\z)%%g
|
||||||
my @item_attrs = ($item =~ m|(<.*?>.*?</.*?>)|gi);
|
|
||||||
my $ia;
|
|
||||||
my %ia;
|
|
||||||
foreach $ia (@item_attrs) {
|
|
||||||
$ia =~ m|^<(.*?)>(.*?)</.*?>$|;
|
|
||||||
$ia{$1} = trim($2);
|
|
||||||
}
|
|
||||||
|
|
||||||
# Links are nested, kill prev {'link'} and rebuild attrs inside it
|
my $qr = qr{@{[join('|', keys %{$inhash})]}};
|
||||||
undef $ia{'link'};
|
$content =~ s%<($qr)\s*(?:[^>]*?)?(?:/|>.*?</\1)>%%sg;
|
||||||
my @linkitems = xml_content_array('link', \$item)
|
|
||||||
or next;
|
|
||||||
|
|
||||||
my $linkitem = $linkitems[0]; ## Usually first one is most relevant
|
$inhash->{'<>'} = $content if $content =~ /\S/;
|
||||||
@item_attrs = ($linkitem =~ m|(<.*?>.*?</.*?>)|gi);
|
|
||||||
foreach $ia (@item_attrs) {
|
|
||||||
$ia =~ m|^<(.*?)>(.*?)</.*?>$|;
|
|
||||||
$ia{$1} = trim($2);
|
|
||||||
}
|
|
||||||
|
|
||||||
# Alias SN to RSS
|
|
||||||
if (exists $ia{'text'}) {
|
|
||||||
$ia{'description'} = $ia{'text'};
|
|
||||||
}
|
|
||||||
if (exists $ia{'linetext'}) {
|
|
||||||
$ia{'title'} = $ia{'linetext'};
|
|
||||||
}
|
|
||||||
if (exists $ia{'url'}) {
|
|
||||||
$ia{'link'} = $ia{'url'};
|
|
||||||
}
|
|
||||||
push(@{$rr->{'items'}}, \%ia);
|
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
|
if( ref($inhash) ){
|
||||||
sub parseWL {
|
#We have attributes? Then we should merge them.
|
||||||
my ($rr, $cr) = @_;
|
if( ref($attr) ){
|
||||||
|
for( keys %{$attr} ){
|
||||||
# my $channel = xml_content_string('header', $cr);
|
$inhash->{$_} = exists($inhash->{$_}) ?
|
||||||
# $channel =~ s|<item.*?</item>||gis;
|
(ref($inhash->{$_}) eq 'ARRAY' ?
|
||||||
# clean(\$channel);
|
[@{$inhash->{$_}}, $attr->{$_}] :
|
||||||
|
[ $inhash->{$_}, $attr->{$_}] ) : $attr->{$_};
|
||||||
# my $ca;
|
|
||||||
# my @channel_attrs = ($channel =~ m|(<.*?>.*?</.*?>)|gi);
|
|
||||||
# foreach $ca (@channel_attrs) {
|
|
||||||
# $ca =~ m|^<(.*?)>(.*?)</.*?>$|;
|
|
||||||
# $rr->{$1} = trim($2);
|
|
||||||
# }
|
|
||||||
|
|
||||||
##
|
|
||||||
## Alias SN to RSS terms
|
|
||||||
##
|
|
||||||
# if (exists $rr->{'channelDescription'}) {
|
|
||||||
# $rr->{'description'} = $rr->{'channelDescription'};
|
|
||||||
# }
|
|
||||||
# if (exists $rr->{'channelTitle'}) {
|
|
||||||
# $rr->{'title'} = $rr->{'channelTitle'};
|
|
||||||
# }
|
|
||||||
# if (exists $rr->{'channelLink'}) {
|
|
||||||
# $rr->{'link'} = $rr->{'channelLink'};
|
|
||||||
# }
|
|
||||||
|
|
||||||
$rr->{'items'} = ();
|
|
||||||
my $item;
|
|
||||||
foreach $item (xml_content_array('link', $cr)) {
|
|
||||||
clean(\$item);
|
|
||||||
my @item_attrs = ($item =~ m|(<.*?>.*?</.*?>)|gi);
|
|
||||||
my $ia;
|
|
||||||
my %ia;
|
|
||||||
foreach $ia (@item_attrs) {
|
|
||||||
$ia =~ m|^<(.*?)>(.*?)</.*?>$|;
|
|
||||||
$ia{$1} = trim($2);
|
|
||||||
}
|
|
||||||
# Alias WL to RSS
|
|
||||||
if (exists $ia{'url'}) {
|
|
||||||
$ia{'link'} = $ia{'url'};
|
|
||||||
}
|
|
||||||
|
|
||||||
push(@{$rr->{'items'}}, \%ia);
|
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
}
|
||||||
sub usableXML {
|
elsif( ref($attr) ){
|
||||||
my $cref = shift;
|
$inhash = $attr;
|
||||||
my $content = $$cref; ## Don't change caller's content just for usability check
|
}
|
||||||
|
else{
|
||||||
preprocess(\$content);
|
#Otherwise save our content
|
||||||
|
$inhash = $content;
|
||||||
return 1 if isRSS(\$content);
|
|
||||||
return 2 if isRDF(\$content);
|
|
||||||
return 3 if isSN(\$content);
|
|
||||||
return 4 if isWL(\$content);
|
|
||||||
|
|
||||||
return 0;
|
|
||||||
}
|
}
|
||||||
|
|
||||||
sub isRSS {
|
$hash->{$tag} = exists($hash->{$tag}) ?
|
||||||
my $cref = shift;
|
(ref($hash->{$tag}) eq 'ARRAY' ?
|
||||||
return scalar($$cref =~ /<rss.*>.*<\/rss>/is);
|
[@{$hash->{$tag}}, $inhash] :
|
||||||
|
[ $hash->{$tag}, $inhash] ) : $inhash;
|
||||||
}
|
}
|
||||||
|
|
||||||
sub isRDF {
|
|
||||||
my $cref = shift;
|
|
||||||
return scalar($$cref =~ /<rdf:RDF.*>.*<\/rdf:RDF>/is);
|
|
||||||
}
|
|
||||||
|
|
||||||
sub isSN {
|
|
||||||
my $cref = shift;
|
|
||||||
return scalar($$cref =~ /<scriptingnews.*>.*<\/scriptingnews>/is);
|
|
||||||
}
|
|
||||||
|
|
||||||
sub isWL {
|
|
||||||
my $cref = shift;
|
|
||||||
return scalar($$cref =~ /<weblog.*>.*<\/weblog>/is);
|
|
||||||
}
|
|
||||||
|
|
||||||
sub xml_content_string {
|
|
||||||
my $tag = shift;
|
|
||||||
my $cref = shift;
|
|
||||||
|
|
||||||
$$cref =~ /<${tag}.*?>(.*)<\/${tag}>/is;
|
|
||||||
return $1;
|
|
||||||
}
|
|
||||||
|
|
||||||
sub xml_content_array {
|
|
||||||
my $tag = shift;
|
|
||||||
my $cref = shift;
|
|
||||||
my $keeptags = shift;
|
|
||||||
$keeptags = 0 unless $keeptags;
|
|
||||||
my @result;
|
|
||||||
|
|
||||||
if ($keeptags) {
|
|
||||||
@result = ($$cref =~ /(<${tag}.*?>.*?<\/${tag}>)/gis);
|
|
||||||
} else {
|
|
||||||
@result = ($$cref =~ /<${tag}.*?>(.*?)<\/${tag}>/gis);
|
|
||||||
}
|
|
||||||
|
|
||||||
return @result;
|
|
||||||
}
|
|
||||||
|
|
||||||
sub clean {
|
|
||||||
my $cref = shift;
|
|
||||||
|
|
||||||
$$cref =~ s{(\n|<p>|</p>|<b>|</b>|<i>|</i>|<h\d>|</h\d>|<strong>|</strong>|<center>|</center>|<quote>|</quote>)}{ }gsi;
|
|
||||||
}
|
|
||||||
|
|
||||||
sub trim {
|
|
||||||
my $s = shift;
|
|
||||||
|
|
||||||
$s =~ s/^\s*(.*?)\s*$/$1/;
|
|
||||||
return $s;
|
|
||||||
}
|
}
|
||||||
|
|
||||||
1;
|
1;
|
||||||
|
|
@ -338,7 +303,7 @@ __END__
|
||||||
|
|
||||||
=head1 NAME
|
=head1 NAME
|
||||||
|
|
||||||
XML::RSSLite - Perl extension for "relaxed" RSS parsing
|
XML::RSSLite - lightweight, "relaxed" RSS (and XML-ish) parser
|
||||||
|
|
||||||
=head1 SYNOPSIS
|
=head1 SYNOPSIS
|
||||||
|
|
||||||
|
|
@ -346,14 +311,14 @@ XML::RSSLite - Perl extension for "relaxed" RSS parsing
|
||||||
|
|
||||||
. . .
|
. . .
|
||||||
|
|
||||||
parseXML(\%result, \$content);
|
parseRSS(\%result, \$content);
|
||||||
|
|
||||||
print "=== Channel ===\n",
|
print "=== Channel ===\n",
|
||||||
"Title: $result{'title'}\n",
|
"Title: $result{'title'}\n",
|
||||||
"Desc: $result{'description'}\n",
|
"Desc: $result{'description'}\n",
|
||||||
"Link: $result{'link'}\n\n";
|
"Link: $result{'link'}\n\n";
|
||||||
|
|
||||||
foreach $item (@{$result{'items'}}) {
|
foreach $item (@{$result{'item'}}) {
|
||||||
print " --- Item ---\n",
|
print " --- Item ---\n",
|
||||||
" Title: $item->{'title'}\n",
|
" Title: $item->{'title'}\n",
|
||||||
" Desc: $item->{'description'}\n",
|
" Desc: $item->{'description'}\n",
|
||||||
|
|
@ -369,7 +334,7 @@ old-fashioned Perl regular expressions. It stores the data in a simple
|
||||||
hash structure, and "aliases" certain tags so that when done, you can
|
hash structure, and "aliases" certain tags so that when done, you can
|
||||||
count on having the minimal data necessary for re-constructing a valid
|
count on having the minimal data necessary for re-constructing a valid
|
||||||
RSS file. This means you get the basic title, description, and link for a
|
RSS file. This means you get the basic title, description, and link for a
|
||||||
channel and its items. Anything else present in the hash is a bonus :)
|
channel and its items.
|
||||||
|
|
||||||
This module extracts more usable links by parsing "scriptingNews" and
|
This module extracts more usable links by parsing "scriptingNews" and
|
||||||
"weblog" formats in addition to RDF & RSS. It also "sanitizes" the
|
"weblog" formats in addition to RDF & RSS. It also "sanitizes" the
|
||||||
|
|
@ -397,60 +362,96 @@ output for best results. The munging includes:
|
||||||
|
|
||||||
=over
|
=over
|
||||||
|
|
||||||
=item parseXML($outHashRef, $inScalarRef)
|
=item parseRSS($outHashRef, $inScalarRef)
|
||||||
|
|
||||||
I<$inScalarRef> is a reference to a scalar containing the document to be
|
I<$inScalarRef> is a reference to a scalar containing the document to be
|
||||||
parsed, the contents will effectively be destroyed. I<$outHashRef> is a
|
parsed, the contents will effectively be destroyed. I<$outHashRef> is a
|
||||||
reference to the hash within which to store the parsed content.
|
reference to the hash within which to store the parsed content.
|
||||||
|
|
||||||
=item usableXML($inScalarRef)
|
|
||||||
|
|
||||||
Test whether or not B<XML::RSSLite> understands the content of the referenced
|
|
||||||
document.
|
|
||||||
|
|
||||||
=back
|
=back
|
||||||
|
|
||||||
=head2 EXPORTABLE
|
=head2 EXPORTABLE
|
||||||
|
|
||||||
=over
|
=over
|
||||||
|
|
||||||
=item isRDF($inScalarRef)
|
=item parseXML(\%parsedTree, \$parseThis, 'topTag', $comments);
|
||||||
|
|
||||||
Tests if a referenced document is RDF.
|
=over
|
||||||
|
|
||||||
=item isRSS($inScalarRef)
|
=item parsedTree - required
|
||||||
|
|
||||||
Tests if a referenced document is RSS.
|
Reference to hash to store the parsed document within.
|
||||||
|
|
||||||
=item isSN($inScalarRef)
|
=item parseThis - required
|
||||||
|
|
||||||
Tests if a referenced document is scriptingNews.
|
Reference to scalar containing the document to parse.
|
||||||
|
|
||||||
=item isWL($inScalarRef)
|
=item topTag - optional
|
||||||
|
|
||||||
Tests if a referenced document is weblog.
|
Tag to consider the root node, leaving this undefined is not recommended.
|
||||||
|
|
||||||
|
=item comments - optional
|
||||||
|
|
||||||
|
=over
|
||||||
|
|
||||||
|
=item false will remove contents from parseThis
|
||||||
|
|
||||||
|
=item true will not remove comments from parseThis
|
||||||
|
|
||||||
|
=item array reference is true, comments are stored here
|
||||||
|
|
||||||
=back
|
=back
|
||||||
|
|
||||||
=head1 BUGS
|
=back
|
||||||
|
|
||||||
Sometimes the title of an item will be missed,
|
=head2 CAVEATS
|
||||||
the condition will presist until additional items have been added
|
|
||||||
to the document. As a stop gap, when this happens the item title
|
|
||||||
is set equal to the item link.
|
|
||||||
|
|
||||||
It may take awhile for the tuits to fix this to accumulate.
|
This is not a conforming parser. It does not handle the following
|
||||||
feel free to submit a patch.
|
|
||||||
|
=over
|
||||||
|
|
||||||
|
=item
|
||||||
|
|
||||||
|
<foo bar=">">
|
||||||
|
|
||||||
|
=item
|
||||||
|
|
||||||
|
<foo><bar> <bar></bar> <bar></bar> </bar></foo>
|
||||||
|
|
||||||
|
=item
|
||||||
|
|
||||||
|
<![CDATA[ ]]>
|
||||||
|
|
||||||
|
=item
|
||||||
|
|
||||||
|
PI
|
||||||
|
|
||||||
|
=back
|
||||||
|
|
||||||
|
It's non-validating, without a DTD the following cannot be properly addressed
|
||||||
|
|
||||||
|
=over
|
||||||
|
|
||||||
|
=item entities
|
||||||
|
|
||||||
|
=item namespaces
|
||||||
|
|
||||||
|
This might be arriving in the next release.
|
||||||
|
|
||||||
|
=back
|
||||||
|
|
||||||
|
=back
|
||||||
|
|
||||||
=head1 SEE ALSO
|
=head1 SEE ALSO
|
||||||
|
|
||||||
perl(1), C<XML::RSS>
|
perl(1), C<XML::RSS>, C<XML::SAX::PurePerl>,
|
||||||
|
C<XML::Parser::Lite>, <XML::Parser>
|
||||||
|
|
||||||
=head1 AUTHOR
|
=head1 AUTHOR
|
||||||
|
|
||||||
Jerrad Pierce <jpierce@cpan.org>.
|
Jerrad Pierce <jpierce@cpan.org>.
|
||||||
|
|
||||||
Scott Thomason <scott@industrial-linux.org>
|
Scott Thomason <scott@thomasons.org>
|
||||||
|
|
||||||
=head1 LICENSE
|
=head1 LICENSE
|
||||||
|
|
||||||
|
|
|
||||||
Loading…
Add table
Add a link
Reference in a new issue