* aggregate: Revert use of forking to not save state, that was not the right

[git.ikiwiki.info.git] / IkiWiki / Plugin / aggregate.pm

diff --git a/IkiWiki/Plugin/aggregate.pm b/IkiWiki/Plugin/aggregate.pm
index d6592fef30f96fdfe20c5d812ef38fc77c373fd5..0f50fab06c6795154bff70786459a3ff474464d8 100644 (file)
--- a/IkiWiki/Plugin/aggregate.pm
+++ b/IkiWiki/Plugin/aggregate.pm
@@ -4,54 +4,64 @@ package IkiWiki::Plugin::aggregate;
 
 use warnings;
 use strict;
 
 use warnings;
 use strict;

-use IkiWiki;
+use IkiWiki 2.00;

 use HTML::Entities;
 use HTML::Parser;
 use HTML::Tagset;
 use URI;
 use HTML::Entities;
 use HTML::Parser;
 use HTML::Tagset;
 use URI;

+use open qw{:utf8 :std};

 
 my %feeds;
 my %guids;
 
 sub import { #{{{
 
 my %feeds;
 my %guids;
 
 sub import { #{{{

-       IkiWiki::hook(type => "getopt", id => "aggregate", 
-               call => \&getopt);
-       IkiWiki::hook(type => "checkconfig", id => "aggregate",
-               call => \&checkconfig);
-       IkiWiki::hook(type => "filter", id => "aggregate", 
-               call => \&filter);
-       IkiWiki::hook(type => "preprocess", id => "aggregate",
-               call => \&preprocess);
-        IkiWiki::hook(type => "delete", id => "aggregate",
-                call => \&delete);
-       IkiWiki::hook(type => "savestate", id => "aggregate",
-               call => \&savestate);
+       hook(type => "getopt", id => "aggregate", call => \&getopt);
+       hook(type => "checkconfig", id => "aggregate", call => \&checkconfig);
+       hook(type => "needsbuild", id => "aggregate", call => \&needsbuild);
+       hook(type => "preprocess", id => "aggregate", call => \&preprocess);
+        hook(type => "delete", id => "aggregate", call => \&delete);
+       hook(type => "savestate", id => "aggregate", call => \&savestate);

 } # }}}
 
 sub getopt () { #{{{
         eval q{use Getopt::Long};
 } # }}}
 
 sub getopt () { #{{{
         eval q{use Getopt::Long};

+       error($@) if $@;

         Getopt::Long::Configure('pass_through');
         Getopt::Long::Configure('pass_through');

-        GetOptions("aggregate" => \$IkiWiki::config{aggregate});
+        GetOptions("aggregate" => \$config{aggregate});

 } #}}}
 
 sub checkconfig () { #{{{
 } #}}}
 
 sub checkconfig () { #{{{

-       loadstate();
-       if ($IkiWiki::config{aggregate}) {
+       if ($config{aggregate} && ! ($config{post_commit} && 
+                                    IkiWiki::commit_hook_enabled())) {
+               if (! IkiWiki::lockwiki(0)) {
+                       debug("wiki is locked by another process, not aggregating");
+                       exit 1;
+               }
+       
+               loadstate();

                IkiWiki::loadindex();
                aggregate();
                IkiWiki::loadindex();
                aggregate();

+               expire();

                savestate();
                savestate();

+               clearstate();
+
+               IkiWiki::unlockwiki();

        }
 } #}}}
 
        }
 } #}}}
 

-sub filter (@) { #{{{
-       my %params=@_;
-       my $page=$params{page};
-
-       # Mark all feeds originating on this page as removable;
-       # preprocess will unmark those that still exist.
-       remove_feeds($page);
+sub needsbuild (@) { #{{{
+       my $needsbuild=shift;
+       
+       loadstate(); # if not already loaded

 
 

-       return $params{content};
+       foreach my $feed (values %feeds) {
+               if (exists $pagesources{$feed->{sourcepage}} && 
+                   grep { $_ eq $pagesources{$feed->{sourcepage}} } @$needsbuild) {
+                       # Mark all feeds originating on this page as removable;
+                       # preprocess will unmark those that still exist.
+                       remove_feeds($feed->{sourcepage});
+               }
+       }

 } # }}}
 
 sub preprocess (@) { #{{{
 } # }}}
 
 sub preprocess (@) { #{{{


@@ -59,7 +69,7 @@ sub preprocess (@) { #{{{
 
        foreach my $required (qw{name url}) {
                if (! exists $params{$required}) {
 
        foreach my $required (qw{name url}) {
                if (! exists $params{$required}) {

-                       return "[[aggregate plugin missing $required parameter]]";
+                       return "[[aggregate ".sprintf(gettext("missing %s parameter"), $required)."]]";

                }
        }
 
                }
        }
 


@@ -76,17 +86,19 @@ sub preprocess (@) { #{{{
        $feed->{url}=$params{url};
        my $dir=exists $params{dir} ? $params{dir} : $params{page}."/".IkiWiki::titlepage($params{name});
        $dir=~s/^\/+//;
        $feed->{url}=$params{url};
        my $dir=exists $params{dir} ? $params{dir} : $params{page}."/".IkiWiki::titlepage($params{name});
        $dir=~s/^\/+//;

-       ($dir)=$dir=~/$IkiWiki::config{wiki_file_regexp}/;
+       ($dir)=$dir=~/$config{wiki_file_regexp}/;

        $feed->{dir}=$dir;
        $feed->{feedurl}=defined $params{feedurl} ? $params{feedurl} : "";
        $feed->{updateinterval}=defined $params{updateinterval} ? $params{updateinterval} * 60 : 15 * 60;
        $feed->{expireage}=defined $params{expireage} ? $params{expireage} : 0;
        $feed->{expirecount}=defined $params{expirecount} ? $params{expirecount} : 0;
        delete $feed->{remove};
        $feed->{dir}=$dir;
        $feed->{feedurl}=defined $params{feedurl} ? $params{feedurl} : "";
        $feed->{updateinterval}=defined $params{updateinterval} ? $params{updateinterval} * 60 : 15 * 60;
        $feed->{expireage}=defined $params{expireage} ? $params{expireage} : 0;
        $feed->{expirecount}=defined $params{expirecount} ? $params{expirecount} : 0;
        delete $feed->{remove};

+       delete $feed->{expired};

        $feed->{lastupdate}=0 unless defined $feed->{lastupdate};
        $feed->{numposts}=0 unless defined $feed->{numposts};
        $feed->{newposts}=0 unless defined $feed->{newposts};
        $feed->{lastupdate}=0 unless defined $feed->{lastupdate};
        $feed->{numposts}=0 unless defined $feed->{numposts};
        $feed->{newposts}=0 unless defined $feed->{newposts};

-       $feed->{message}="new feed" unless defined $feed->{message};
+       $feed->{message}=gettext("new feed") unless defined $feed->{message};
+       $feed->{error}=0 unless defined $feed->{error};

        $feed->{tags}=[];
        while (@_) {
                my $key=shift;
        $feed->{tags}=[];
        while (@_) {
                my $key=shift;


@@ -97,8 +109,12 @@ sub preprocess (@) { #{{{
        }
 
        return "<a href=\"".$feed->{url}."\">".$feed->{name}."</a>: ".
        }
 
        return "<a href=\"".$feed->{url}."\">".$feed->{name}."</a>: ".

-              "<i>".$feed->{message}."</i> (".$feed->{numposts}.
-              " stored posts; ".$feed->{newposts}." new)<br />";
+              ($feed->{error} ? "<em>" : "").$feed->{message}.
+              ($feed->{error} ? "</em>" : "").
+              " (".$feed->{numposts}." ".gettext("posts").
+              ($feed->{newposts} ? "; ".$feed->{newposts}.
+                                   " ".gettext("new") : "").
+              ")";

 } # }}}
 
 sub delete (@) { #{{{
 } # }}}
 
 sub delete (@) { #{{{


@@ -106,15 +122,18 @@ sub delete (@) { #{{{
 
        # Remove feed data for removed pages.
        foreach my $file (@files) {
 
        # Remove feed data for removed pages.
        foreach my $file (@files) {

-               my $page=IkiWiki::pagename($file);
+               my $page=pagename($file);

                remove_feeds($page);
        }
 } #}}}
 
                remove_feeds($page);
        }
 } #}}}
 

+my $state_loaded=0;

 sub loadstate () { #{{{
 sub loadstate () { #{{{

-       if (-e "$IkiWiki::config{wikistatedir}/aggregate") {
-               open (IN, "$IkiWiki::config{wikistatedir}/aggregate" ||
-                       die "$IkiWiki::config{wikistatedir}/aggregate: $!");
+       return if $state_loaded;
+       $state_loaded=1;
+       if (-e "$config{wikistatedir}/aggregate") {
+               open(IN, "<", "$config{wikistatedir}/aggregate") ||
+                       die "$config{wikistatedir}/aggregate: $!";

                while (<IN>) {
                        $_=IkiWiki::possibly_foolish_untaint($_);
                        chomp;
                while (<IN>) {
                        $_=IkiWiki::possibly_foolish_untaint($_);
                        chomp;


@@ -146,10 +165,12 @@ sub loadstate () { #{{{
 } #}}}
 
 sub savestate () { #{{{
 } #}}}
 
 sub savestate () { #{{{

+       return unless $state_loaded;

        eval q{use HTML::Entities};
        eval q{use HTML::Entities};

-       die $@ if $@;
-       open (OUT, ">$IkiWiki::config{wikistatedir}/aggregate" ||
-               die "$IkiWiki::config{wikistatedir}/aggregate: $!");
+       error($@) if $@;
+       my $newfile="$config{wikistatedir}/aggregate.new";
+       my $cleanup = sub { unlink($newfile) };
+       open (OUT, ">", $newfile) || error("open $newfile: $!", $cleanup);

        foreach my $data (values %feeds, values %guids) {
                if ($data->{remove}) {
                        if ($data->{name}) {
        foreach my $data (values %feeds, values %guids) {
                if ($data->{remove}) {
                        if ($data->{name}) {


@@ -160,10 +181,16 @@ sub savestate () { #{{{
                                }
                        }
                        else {
                                }
                        }
                        else {

-                               unlink pagefile($data->{page});
+                               unlink pagefile($data->{page})
+                                       if exists $data->{page};

                        }
                        next;
                }
                        }
                        next;
                }

+               elsif ($data->{expired} && exists $data->{page}) {
+                       unlink pagefile($data->{page});
+                       delete $data->{page};
+                       delete $data->{md5};
+               }

 
                my @line;
                foreach my $field (keys %$data) {
 
                my @line;
                foreach my $field (keys %$data) {


@@ -178,62 +205,135 @@ sub savestate () { #{{{
                                push @line, "$field=".$data->{$field};
                        }
                }
                                push @line, "$field=".$data->{$field};
                        }
                }

-               print OUT join(" ", @line)."\n";
+               print OUT join(" ", @line)."\n" || error("write $newfile: $!", $cleanup);
+       }
+       close OUT || error("save $newfile: $!", $cleanup);
+       rename($newfile, "$config{wikistatedir}/aggregate") ||
+               error("rename $newfile: $!", $cleanup);
+} #}}}
+
+sub clearstate () { #{{{
+       %feeds=();
+       %guids=();
+       $state_loaded=0;
+} #}}}
+
+sub expire () { #{{{
+       foreach my $feed (values %feeds) {
+               next unless $feed->{expireage} || $feed->{expirecount};
+               my $count=0;
+               my %seen;
+               foreach my $item (sort { $IkiWiki::pagectime{$b->{page}} <=> $IkiWiki::pagectime{$a->{page}} }
+                                 grep { exists $_->{page} && $_->{feed} eq $feed->{name} && $IkiWiki::pagectime{$_->{page}} }
+                                 values %guids) {
+                       if ($feed->{expireage}) {
+                               my $days_old = (time - $IkiWiki::pagectime{$item->{page}}) / 60 / 60 / 24;
+                               if ($days_old > $feed->{expireage}) {
+                                       debug(sprintf(gettext("expiring %s (%s days old)"),
+                                               $item->{page}, int($days_old)));
+                                       $item->{expired}=1;
+                               }
+                       }
+                       elsif ($feed->{expirecount} &&
+                              $count >= $feed->{expirecount}) {
+                               debug(sprintf(gettext("expiring %s"), $item->{page}));
+                               $item->{expired}=1;
+                       }
+                       else {
+                               if (! $seen{$item->{page}}) {
+                                       $seen{$item->{page}}=1;
+                                       $count++;
+                               }
+                       }
+               }

        }
        }

-       close OUT;

 } #}}}
 
 sub aggregate () { #{{{
        eval q{use XML::Feed};
 } #}}}
 
 sub aggregate () { #{{{
        eval q{use XML::Feed};

-       die $@ if $@;
+       error($@) if $@;
+       eval q{use URI::Fetch};
+       error($@) if $@;

        eval q{use HTML::Entities};
        eval q{use HTML::Entities};

-       die $@ if $@;
+       error($@) if $@;

 
        foreach my $feed (values %feeds) {
 
        foreach my $feed (values %feeds) {

-               next unless $IkiWiki::config{rebuild} || 
+               next unless $config{rebuild} || 

                        time - $feed->{lastupdate} >= $feed->{updateinterval};
                $feed->{lastupdate}=time;
                $feed->{newposts}=0;
                        time - $feed->{lastupdate} >= $feed->{updateinterval};
                $feed->{lastupdate}=time;
                $feed->{newposts}=0;

+               $feed->{message}=sprintf(gettext("processed ok at %s"),
+                       displaytime($feed->{lastupdate}));
+               $feed->{error}=0;

                $IkiWiki::forcerebuild{$feed->{sourcepage}}=1;
 
                $IkiWiki::forcerebuild{$feed->{sourcepage}}=1;
 

-               IkiWiki::debug("checking feed ".$feed->{name}." ...");
+               debug(sprintf(gettext("checking feed %s ..."), $feed->{name}));

 
                if (! length $feed->{feedurl}) {
                        my @urls=XML::Feed->find_feeds($feed->{url});
                        if (! @urls) {
 
                if (! length $feed->{feedurl}) {
                        my @urls=XML::Feed->find_feeds($feed->{url});
                        if (! @urls) {

-                               $feed->{message}="could not find feed at ".$feed->{feedurl};
-                               IkiWiki::debug($feed->{message});
+                               $feed->{message}=sprintf(gettext("could not find feed at %s"), $feed->{url});
+                               $feed->{error}=1;
+                               debug($feed->{message});

                                next;
                        }
                        $feed->{feedurl}=pop @urls;
                }
                                next;
                        }
                        $feed->{feedurl}=pop @urls;
                }

-               my $f=eval{XML::Feed->parse(URI->new($feed->{feedurl}))};
+               my $res=URI::Fetch->fetch($feed->{feedurl});
+               if (! $res) {
+                       $feed->{message}=URI::Fetch->errstr;
+                       $feed->{error}=1;
+                       debug($feed->{message});
+                       next;
+               }
+               if ($res->status == URI::Fetch::URI_GONE()) {
+                       $feed->{message}=gettext("feed not found");
+                       $feed->{error}=1;
+                       debug($feed->{message});
+                       next;
+               }
+               my $content=$res->content;
+               my $f=eval{XML::Feed->parse(\$content)};

                if ($@) {
                if ($@) {

-                       $feed->{message}="feed crashed XML::Feed! $@";
-                       IkiWiki::debug($feed->{message});
+                       # One common cause of XML::Feed crashing is a feed
+                       # that contains invalid UTF-8 sequences. Convert
+                       # feed to ascii to try to work around.
+                       $feed->{message}.=" ".sprintf(gettext("(invalid UTF-8 stripped from feed)"));
+                       $content=Encode::decode_utf8($content);
+                       $f=eval{XML::Feed->parse(\$content)};
+               }
+               if ($@) {
+                       # Another possibility is badly escaped entities.
+                       $feed->{message}.=" ".sprintf(gettext("(feed entities escaped)"));
+                       $content=~s/\&(?!amp)(\w+);/&amp;$1;/g;
+                       $content=Encode::decode_utf8($content);
+                       $f=eval{XML::Feed->parse(\$content)};
+               }
+               if ($@) {
+                       $feed->{message}=gettext("feed crashed XML::Feed!")." ($@)";
+                       $feed->{error}=1;
+                       debug($feed->{message});

                        next;
                }
                if (! $f) {
                        $feed->{message}=XML::Feed->errstr;
                        next;
                }
                if (! $f) {
                        $feed->{message}=XML::Feed->errstr;

-                       IkiWiki::debug($feed->{message});
+                       $feed->{error}=1;
+                       debug($feed->{message});

                        next;
                }
 
                foreach my $entry ($f->entries) {
                        add_page(
                                feed => $feed,
                        next;
                }
 
                foreach my $entry ($f->entries) {
                        add_page(
                                feed => $feed,

+                               copyright => $f->copyright,

                                title => defined $entry->title ? decode_entities($entry->title) : "untitled",
                                link => $entry->link,
                                title => defined $entry->title ? decode_entities($entry->title) : "untitled",
                                link => $entry->link,

-                               content => $entry->content->body,
+                               content => defined $entry->content->body ? $entry->content->body : "",

                                guid => defined $entry->id ? $entry->id : time."_".$feed->name,
                                ctime => $entry->issued ? ($entry->issued->epoch || time) : time,
                        );
                }
                                guid => defined $entry->id ? $entry->id : time."_".$feed->name,
                                ctime => $entry->issued ? ($entry->issued->epoch || time) : time,
                        );
                }

-
-               $feed->{message}="processed ok";

        }
        }

-
-       # TODO: expiry

 } #}}}
 
 sub add_page (@) { #{{{
 } #}}}
 
 sub add_page (@) { #{{{


@@ -245,6 +345,7 @@ sub add_page (@) { #{{{
        if (exists $guids{$params{guid}}) {
                # updating an existing post
                $guid=$guids{$params{guid}};
        if (exists $guids{$params{guid}}) {
                # updating an existing post
                $guid=$guids{$params{guid}};

+               return if $guid->{expired};

        }
        else {
                # new post
        }
        else {
                # new post


@@ -260,18 +361,30 @@ sub add_page (@) { #{{{
                # directory name or trigger ".." disallowing code.
                $page=~s!([/.])!"__".ord($1)."__"!eg;
                $page=$feed->{dir}."/".$page;
                # directory name or trigger ".." disallowing code.
                $page=~s!([/.])!"__".ord($1)."__"!eg;
                $page=$feed->{dir}."/".$page;

-               $page=lc($page);
-               ($page)=$page=~/$IkiWiki::config{wiki_file_regexp}/;
+               ($page)=$page=~/$config{wiki_file_regexp}/;

                if (! defined $page || ! length $page) {
                        $page=$feed->{dir}."/item";
                }
                my $c="";
                if (! defined $page || ! length $page) {
                        $page=$feed->{dir}."/item";
                }
                my $c="";

-               while (exists $IkiWiki::pagesources{$page.$c} ||
+               while (exists $IkiWiki::pagecase{lc $page.$c} ||

                       -e pagefile($page.$c)) {
                        $c++
                }
                       -e pagefile($page.$c)) {
                        $c++
                }

+
+               # Make sure that the file name isn't too long. 
+               # NB: This doesn't check for path length limits.
+               my $max=POSIX::pathconf($config{srcdir}, &POSIX::_PC_NAME_MAX);
+               if (defined $max && length(htmlfn($page)) >= $max) {
+                       $c="";
+                       $page=$feed->{dir}."/item";
+                       while (exists $IkiWiki::pagecase{lc $page.$c} ||
+                              -e pagefile($page.$c)) {
+                               $c++
+                       }
+               }
+

                $guid->{page}=$page;
                $guid->{page}=$page;

-               IkiWiki::debug("creating new page $page");
+               debug(sprintf(gettext("creating new page %s"), $page));

        }
        $guid->{feed}=$feed->{name};
        
        }
        $guid->{feed}=$feed->{name};
        


@@ -279,29 +392,33 @@ sub add_page (@) { #{{{
        # to avoid unneccessary rebuilding. The mtime from rss cannot be
        # trusted; let's use a digest.
        eval q{use Digest::MD5 'md5_hex'};
        # to avoid unneccessary rebuilding. The mtime from rss cannot be
        # trusted; let's use a digest.
        eval q{use Digest::MD5 'md5_hex'};

+       error($@) if $@;

        require Encode;
        my $digest=md5_hex(Encode::encode_utf8($params{content}));
        require Encode;
        my $digest=md5_hex(Encode::encode_utf8($params{content}));

-       return unless ! exists $guid->{md5} || $guid->{md5} ne $digest || $IkiWiki::config{rebuild};
+       return unless ! exists $guid->{md5} || $guid->{md5} ne $digest || $config{rebuild};

        $guid->{md5}=$digest;
 
        # Create the page.
        $guid->{md5}=$digest;
 
        # Create the page.

-       my $template=IkiWiki::template("aggregatepost.tmpl", blind_cache => 1);
+       my $template=template("aggregatepost.tmpl", blind_cache => 1);

        $template->param(title => $params{title})
                if defined $params{title} && length($params{title});
        $template->param(content => htmlescape(htmlabs($params{content}, $feed->{feedurl})));
        $template->param(title => $params{title})
                if defined $params{title} && length($params{title});
        $template->param(content => htmlescape(htmlabs($params{content}, $feed->{feedurl})));

-       $template->param(url => $feed->{url});

        $template->param(name => $feed->{name});
        $template->param(name => $feed->{name});

-       $template->param(link => urlabs($params{link}, $feed->{feedurl}))
+       $template->param(url => $feed->{url});
+       $template->param(copyright => $params{copyright})
+               if defined $params{copyright} && length $params{copyright};
+       $template->param(permalink => urlabs($params{link}, $feed->{feedurl}))

                if defined $params{link};
        if (ref $feed->{tags}) {
                $template->param(tags => [map { tag => $_ }, @{$feed->{tags}}]);
        }
                if defined $params{link};
        if (ref $feed->{tags}) {
                $template->param(tags => [map { tag => $_ }, @{$feed->{tags}}]);
        }

-       IkiWiki::writefile($guid->{page}.".html", $IkiWiki::config{srcdir},
+       writefile(htmlfn($guid->{page}), $config{srcdir},

                $template->output);
 
        # Set the mtime, this lets the build process get the right creation
        # time on record for the new page.
                $template->output);
 
        # Set the mtime, this lets the build process get the right creation
        # time on record for the new page.

-       utime $mtime, $mtime, pagefile($guid->{page}) if defined $mtime;
+       utime $mtime, $mtime, pagefile($guid->{page})
+               if defined $mtime && $mtime <= time;

 } #}}}
 
 sub htmlescape ($) { #{{{
 } #}}}
 
 sub htmlescape ($) { #{{{


@@ -322,7 +439,7 @@ sub htmlabs ($$) { #{{{
        # Convert links in html from relative to absolute.
        # Note that this is a heuristic, which is not specified by the rss
        # spec and may not be right for all feeds. Also, see Debian
        # Convert links in html from relative to absolute.
        # Note that this is a heuristic, which is not specified by the rss
        # spec and may not be right for all feeds. Also, see Debian

-       # bug #XXXX TODO: get bug.
+       # bug #381359.

        my $html=shift;
        my $urlbase=shift;
 
        my $html=shift;
        my $urlbase=shift;
 


@@ -371,7 +488,11 @@ sub remove_feeds () { #{{{
 sub pagefile ($) { #{{{
        my $page=shift;
 
 sub pagefile ($) { #{{{
        my $page=shift;
 

-       return "$IkiWiki::config{srcdir}/$page.html";
+       return "$config{srcdir}/".htmlfn($page);
+} #}}}
+
+sub htmlfn ($) { #{{{
+       return shift().".".$config{htmlext};

 } #}}}
 
 1
 } #}}}
 
 1