]> sipb.mit.edu Git - ikiwiki.git/blobdiff - IkiWiki/Render.pm
web commit by joey
[ikiwiki.git] / IkiWiki / Render.pm
index 4e2caa6cef31435bc026bf70f9b7573d406e46c1..d0d28e80276db70650a196323fea9054d79967cc 100644 (file)
@@ -1,3 +1,5 @@
+#!/usr/bin/perl
+
 package IkiWiki;
 
 use warnings;
@@ -9,12 +11,47 @@ sub linkify ($$) { #{{{
        my $page=shift;
 
        $content =~ s{(\\?)$config{wiki_link_regexp}}{
-               $1 ? "[[$2]]" : htmllink($page, $2)
+               $2 ? ( $1 ? "[[$2|$3]]" : htmllink($page, titlepage($3), 0, 0, pagetitle($2)))
+                  : ( $1 ? "[[$3]]" :    htmllink($page, titlepage($3)))
        }eg;
        
        return $content;
 } #}}}
 
+my $_scrubber;
+sub scrubber { #{{{
+       return $_scrubber if defined $_scrubber;
+       
+       eval q{use HTML::Scrubber};
+       # Lists based on http://feedparser.org/docs/html-sanitization.html
+       $_scrubber = HTML::Scrubber->new(
+               allow => [qw{
+                       a abbr acronym address area b big blockquote br
+                       button caption center cite code col colgroup dd del
+                       dfn dir div dl dt em fieldset font form h1 h2 h3 h4
+                       h5 h6 hr i img input ins kbd label legend li map
+                       menu ol optgroup option p pre q s samp select small
+                       span strike strong sub sup table tbody td textarea
+                       tfoot th thead tr tt u ul var
+               }],
+               default => [undef, { map { $_ => 1 } qw{
+                       abbr accept accept-charset accesskey action
+                       align alt axis border cellpadding cellspacing
+                       char charoff charset checked cite class
+                       clear cols colspan color compact coords
+                       datetime dir disabled enctype for frame
+                       headers height href hreflang hspace id ismap
+                       label lang longdesc maxlength media method
+                       multiple name nohref noshade nowrap prompt
+                       readonly rel rev rows rowspan rules scope
+                       selected shape size span src start summary
+                       tabindex target title type usemap valign
+                       value vspace width
+               }}],
+       );
+       return $_scrubber;
+} # }}}
+
 sub htmlize ($$) { #{{{
        my $type=shift;
        my $content=shift;
@@ -27,11 +64,17 @@ sub htmlize ($$) { #{{{
        }
        
        if ($type eq '.mdwn') {
-               return Markdown::Markdown($content);
+               $content=Markdown::Markdown($content);
        }
        else {
                error("htmlization of $type not supported");
        }
+
+       if ($config{sanitize}) {
+               $content=scrubber()->scrub($content);
+       }
+       
+       return $content;
 } #}}}
 
 sub backlinks ($) { #{{{
@@ -85,12 +128,11 @@ sub rsspage ($) { #{{{
        return $page.".rss";
 } #}}}
 
-sub postprocess { #{{{
-       # Takes content to postprocess followed by a list of postprocessor
-       # commands and subroutine references to run for the commands.
+sub preprocess ($$) { #{{{
        my $page=shift;
        my $content=shift;
-       my %commands=@_;
+
+       my %commands=(inline => \&preprocess_inline);
        
        my $handle=sub {
                my $escape=shift;
@@ -138,14 +180,14 @@ sub get_inline_content ($$) { #{{{
        my $file=$pagesources{$page};
        my $type=pagetype($file);
        if ($type ne 'unknown') {
-               return htmlize($type, linkify(readfile("$config{srcdir}/$file"), $parentpage));
+               return htmlize($type, linkify(readfile(srcfile($file)), $parentpage));
        }
        else {
                return "";
        }
 } #}}}
 
-sub postprocess_html_inline { #{{{
+sub preprocess_inline ($@) { #{{{
        my $parentpage=shift;
        my %params=@_;
        
@@ -159,7 +201,7 @@ sub postprocess_html_inline { #{{{
                $params{show}=10;
        }
        $inlinepages{$parentpage}=$params{pages};
-       
+
        my $ret="";
        
        if (exists $params{rootpage}) {
@@ -176,8 +218,10 @@ sub postprocess_html_inline { #{{{
                                ? "$config{templatedir}/inlinepage.tmpl"
                                : "$config{templatedir}/inlinepagetitle.tmpl"));
        
+       my @pages;
        foreach my $page (blog_list($params{pages}, $params{show})) {
                next if $page eq $parentpage;
+               push @pages, $page;
                $template->param(pagelink => htmllink($parentpage, $page));
                $template->param(content => get_inline_content($parentpage, $page))
                        if $params{archive} eq "no";
@@ -185,6 +229,14 @@ sub postprocess_html_inline { #{{{
                $ret.=$template->output;
        }
        
+       # TODO: should really add this to renderedfiles and call
+       # check_overwrite, but currently renderedfiles
+       # only supports listing one file per page.
+       if ($config{rss}) {
+               writefile(rsspage($parentpage), $config{destdir},
+                       genrss($parentpage, @pages));
+       }
+       
        return $ret;
 } #}}}
 
@@ -193,18 +245,16 @@ sub genpage ($$$) { #{{{
        my $page=shift;
        my $mtime=shift;
 
-       $content = postprocess($page, $content, inline => \&postprocess_html_inline);
-       
        my $title=pagetitle(basename($page));
        
        my $template=HTML::Template->new(blind_cache => 1,
                filename => "$config{templatedir}/page.tmpl");
        
        if (length $config{cgiurl}) {
-               $template->param(editurl => "$config{cgiurl}?do=edit&page=$page");
-               $template->param(prefsurl => "$config{cgiurl}?do=prefs");
+               $template->param(editurl => cgiurl(do => "edit", page => $page));
+               $template->param(prefsurl => cgiurl(do => "prefs"));
                if ($config{rcs}) {
-                       $template->param(recentchangesurl => "$config{cgiurl}?do=recentchanges");
+                       $template->param(recentchangesurl => cgiurl(do => "recentchanges"));
                }
        }
 
@@ -213,9 +263,12 @@ sub genpage ($$$) { #{{{
                $u=~s/\[\[file\]\]/$pagesources{$page}/g;
                $template->param(historyurl => $u);
        }
+       if ($config{hyperestraier}) {
+               $template->param(hyperestraierurl => cgiurl());
+       }
 
        if ($config{rss} && $inlinepages{$page}) {
-               $template->param(rssurl => rsspage($page));
+               $template->param(rssurl => rsspage(basename($page)));
        }
        
        $template->param(
@@ -226,6 +279,7 @@ sub genpage ($$$) { #{{{
                backlinks => [backlinks($page)],
                discussionlink => htmllink($page, "Discussion", 1, 1),
                mtime => scalar(gmtime($mtime)),
+               styleurl => styleurl($page),
        );
        
        return $template->output;
@@ -250,10 +304,9 @@ sub absolute_urls ($$) { #{{{
        return $content;
 } #}}}
 
-sub genrss ($$$) { #{{{
-       my $content=shift;
+sub genrss ($@) { #{{{
        my $page=shift;
-       my $mtime=shift;
+       my @pages=@_;
        
        my $url="$config{url}/".htmlpage($page);
        
@@ -261,35 +314,14 @@ sub genrss ($$$) { #{{{
                filename => "$config{templatedir}/rsspage.tmpl");
        
        my @items;
-       my $isblog=0;
-       my $gen_blog=sub {
-               my $parentpage=shift;
-               my %params=@_;
-               
-               return "" if exists $params{archive} && $params{archive} eq 'yes';
-               
-               if (! exists $params{show}) {
-                       $params{show}=10;
-               }
-               if (! exists $params{pages}) {
-                       return "";
-               }
-               
-               $isblog=1;
-               foreach my $page (blog_list($params{pages}, $params{show})) {
-                       next if $page eq $parentpage;
-                       push @items, {
-                               itemtitle => pagetitle(basename($page)),
-                               itemurl => "$config{url}/$renderedfiles{$page}",
-                               itempubdate => date_822($pagectime{$page}),
-                               itemcontent => absolute_urls(get_inline_content($parentpage, $page), $url),
-                       } if exists $renderedfiles{$page};
-               }
-               
-               return "";
-       };
-       
-       $content = postprocess($page, $content, inline => $gen_blog);
+       foreach my $p (@pages) {
+               push @items, {
+                       itemtitle => pagetitle(basename($p)),
+                       itemurl => "$config{url}/$renderedfiles{$p}",
+                       itempubdate => date_822($pagectime{$p}),
+                       itemcontent => absolute_urls(get_inline_content($page, $p), $url),
+               } if exists $renderedfiles{$p};
+       }
 
        $template->param(
                title => $config{wikiname},
@@ -315,9 +347,9 @@ sub check_overwrite ($$) { #{{{
 } #}}}
 
 sub mtime ($) { #{{{
-       my $page=shift;
+       my $file=shift;
        
-       return (stat($page))[9];
+       return (stat($file))[9];
 } #}}}
 
 sub findlinks ($$) { #{{{
@@ -326,7 +358,7 @@ sub findlinks ($$) { #{{{
 
        my @links;
        while ($content =~ /(?<!\\)$config{wiki_link_regexp}/g) {
-               push @links, lc($1);
+               push @links, titlepage($2);
        }
        # Discussion links are a special case since they're not in the text
        # of the page, but on its template.
@@ -337,34 +369,29 @@ sub render ($) { #{{{
        my $file=shift;
        
        my $type=pagetype($file);
-       my $content=readfile("$config{srcdir}/$file");
+       my $srcfile=srcfile($file);
        if ($type ne 'unknown') {
+               my $content=readfile($srcfile);
                my $page=pagename($file);
                
                $links{$page}=[findlinks($content, $page)];
                delete $inlinepages{$page};
                
                $content=linkify($content, $page);
+               $content=preprocess($page, $content);
                $content=htmlize($type, $content);
                
                check_overwrite("$config{destdir}/".htmlpage($page), $page);
-               writefile("$config{destdir}/".htmlpage($page),
-                       genpage($content, $page, mtime("$config{srcdir}/$file")));
+               writefile(htmlpage($page), $config{destdir},
+                       genpage($content, $page, mtime($srcfile)));
                $oldpagemtime{$page}=time;
                $renderedfiles{$page}=htmlpage($page);
-
-               # TODO: should really add this to renderedfiles and call
-               # check_overwrite, as above, but currently renderedfiles
-               # only supports listing one file per page.
-               if ($config{rss} && exists $inlinepages{$page}) {
-                       writefile("$config{destdir}/".rsspage($page),
-                               genrss($content, $page, mtime("$config{srcdir}/$file")));
-               }
        }
        else {
+               my $content=readfile($srcfile, 1);
                $links{$file}=[];
                check_overwrite("$config{destdir}/$file", $file);
-               writefile("$config{destdir}/$file", $content);
+               writefile($file, $config{destdir}, $content, 1);
                $oldpagemtime{$file}=time;
                $renderedfiles{$file}=$file;
        }
@@ -380,6 +407,57 @@ sub prune ($) { #{{{
        }
 } #}}}
 
+sub estcfg () { #{{{
+       my $estdir="$config{wikistatedir}/hyperestraier";
+       my $cgi=basename($config{cgiurl});
+       $cgi=~s/\..*$//;
+       open(TEMPLATE, ">$estdir/$cgi.tmpl") ||
+               error("write $estdir/$cgi.tmpl: $!");
+       print TEMPLATE misctemplate("search", 
+               "<!--ESTFORM-->\n\n<!--ESTRESULT-->\n\n<!--ESTINFO-->\n\n");
+       close TEMPLATE;
+       open(TEMPLATE, ">$estdir/$cgi.conf") ||
+               error("write $estdir/$cgi.conf: $!");
+       my $template=HTML::Template->new(
+               filename => "$config{templatedir}/estseek.conf"
+       );
+       eval q{use Cwd 'abs_path'};
+       $template->param(
+               index => $estdir,
+               tmplfile => "$estdir/$cgi.tmpl",
+               destdir => abs_path($config{destdir}),
+               url => $config{url},
+       );
+       print TEMPLATE $template->output;
+       close TEMPLATE;
+       $cgi="$estdir/".basename($config{cgiurl});
+       unlink($cgi);
+       symlink("/usr/lib/estraier/estseek.cgi", $cgi) ||
+               error("symlink $cgi: $!");
+} # }}}
+
+sub estcmd ($;@) { #{{{
+       my @params=split(' ', shift);
+       push @params, "-cl", "$config{wikistatedir}/hyperestraier";
+       if (@_) {
+               push @params, "-";
+       }
+       
+       my $pid=open(CHILD, "|-");
+       if ($pid) {
+               # parent
+               foreach (@_) {
+                       print CHILD "$_\n";
+               }
+               close(CHILD) || error("estcmd @params exited nonzero: $?");
+       }
+       else {
+               # child
+               open(STDOUT, "/dev/null"); # shut it up (closing won't work)
+               exec("estcmd", @params) || error("can't run estcmd");
+       }
+} #}}}
+
 sub refresh () { #{{{
        # find existing pages
        my %exists;
@@ -389,9 +467,7 @@ sub refresh () { #{{{
                no_chdir => 1,
                wanted => sub {
                        if (/$config{wiki_file_prune_regexp}/) {
-                               no warnings 'once';
                                $File::Find::prune=1;
-                               use warnings "all";
                        }
                        elsif (! -d $_ && ! -l $_) {
                                my ($f)=/$config{wiki_file_regexp}/; # untaint
@@ -406,6 +482,30 @@ sub refresh () { #{{{
                        }
                },
        }, $config{srcdir});
+       find({
+               no_chdir => 1,
+               wanted => sub {
+                       if (/$config{wiki_file_prune_regexp}/) {
+                               $File::Find::prune=1;
+                       }
+                       elsif (! -d $_ && ! -l $_) {
+                               my ($f)=/$config{wiki_file_regexp}/; # untaint
+                               if (! defined $f) {
+                                       warn("skipping bad filename $_\n");
+                               }
+                               else {
+                                       # Don't add files that are in the
+                                       # srcdir.
+                                       $f=~s/^\Q$config{underlaydir}\E\/?//;
+                                       if (! -e "$config{srcdir}/$f" && 
+                                           ! -l "$config{srcdir}/$f") {
+                                               push @files, $f;
+                                               $exists{pagename($f)}=1;
+                                       }
+                               }
+                       }
+               },
+       }, $config{underlaydir});
 
        my %rendered;
 
@@ -418,7 +518,8 @@ sub refresh () { #{{{
                        push @add, $file;
                        $links{$page}=[];
                        $pagesources{$page}=$file;
-                       $pagectime{$page}=time unless exists $pagectime{$page};
+                       $pagectime{$page}=mtime(srcfile($file))
+                               unless exists $pagectime{$page};
                }
        }
        my @del;
@@ -438,7 +539,7 @@ sub refresh () { #{{{
                my $page=pagename($file);
                
                if (! exists $oldpagemtime{$page} ||
-                   mtime("$config{srcdir}/$file") > $oldpagemtime{$page}) {
+                   mtime(srcfile($file)) > $oldpagemtime{$page}) {
                        debug("rendering changed file $file");
                        render($file);
                        $rendered{$file}=1;
@@ -481,6 +582,7 @@ FILE:               foreach my $file (@files) {
                                        if (globlist_match($page, $inlinepages{$p})) {
                                                debug("rendering $f, which inlines $page");
                                                render($f);
+                                               $rendered{$f}=1;
                                                last;
                                        }
                                }
@@ -515,9 +617,25 @@ FILE:              foreach my $file (@files) {
                        if (defined $linkfile) {
                                debug("rendering $linkfile, to update its backlinks");
                                render($linkfile);
+                               $rendered{$linkfile}=1;
                        }
                }
        }
+
+       if ($config{hyperestraier} && (%rendered || @del)) {
+               debug("updating hyperestraier search index");
+               if (%rendered) {
+                       estcmd("gather -cm -bc -cl -sd", 
+                               map { $config{destdir}."/".$renderedfiles{pagename($_)} }
+                               keys %rendered);
+               }
+               if (@del) {
+                       estcmd("purge -cl");
+               }
+               
+               debug("generating hyperestraier cgi config");
+               estcfg();
+       }
 } #}}}
 
 1