X-Git-Url: http://git.vanrenterghem.biz/git.ikiwiki.info.git/blobdiff_plain/457d6bbbbfbb0474749a0d68e959613c86facf72..1c5a9c0cc8265a6a552b1500eb3a3d2aa17b9720:/IkiWiki/Render.pm?ds=inline

diff --git a/IkiWiki/Render.pm b/IkiWiki/Render.pm
index 3006b64d4..246c2260d 100644
--- a/IkiWiki/Render.pm
+++ b/IkiWiki/Render.pm
@@ -4,410 +4,251 @@ package IkiWiki;
 
 use warnings;
 use strict;
-use File::Spec;
-
-sub linkify ($$) { #{{{
-	my $content=shift;
-	my $page=shift;
-
-	$content =~ s{(\\?)$config{wiki_link_regexp}}{
-		$2 ? ( $1 ? "[[$2|$3]]" : htmllink($page, titlepage($3), 0, 0, pagetitle($2)))
-		   : ( $1 ? "[[$3]]" :    htmllink($page, titlepage($3)))
-	}eg;
-	
-	return $content;
-} #}}}
-
-my $_scrubber;
-sub scrubber { #{{{
-	return $_scrubber if defined $_scrubber;
-	
-	eval q{use HTML::Scrubber};
-	# Lists based on http://feedparser.org/docs/html-sanitization.html
-	$_scrubber = HTML::Scrubber->new(
-		allow => [qw{
-			a abbr acronym address area b big blockquote br
-			button caption center cite code col colgroup dd del
-			dfn dir div dl dt em fieldset font form h1 h2 h3 h4
-			h5 h6 hr i img input ins kbd label legend li map
-			menu ol optgroup option p pre q s samp select small
-			span strike strong sub sup table tbody td textarea
-			tfoot th thead tr tt u ul var
-		}],
-		default => [undef, { map { $_ => 1 } qw{
-			abbr accept accept-charset accesskey action
-			align alt axis border cellpadding cellspacing
-			char charoff charset checked cite class
-			clear cols colspan color compact coords
-			datetime dir disabled enctype for frame
-			headers height href hreflang hspace id ismap
-			label lang longdesc maxlength media method
-			multiple name nohref noshade nowrap prompt
-			readonly rel rev rows rowspan rules scope
-			selected shape size span src start summary
-			tabindex target title type usemap valign
-			value vspace width
-		}}],
-	);
-	return $_scrubber;
-} # }}}
-
-sub htmlize ($$) { #{{{
-	my $type=shift;
-	my $content=shift;
-	
-	if (! $INC{"/usr/bin/markdown"}) {
-		no warnings 'once';
-		$blosxom::version="is a proper perl module too much to ask?";
-		use warnings 'all';
-		do "/usr/bin/markdown";
-	}
-	
-	if ($type eq '.mdwn') {
-		$content=Markdown::Markdown($content);
-	}
-	else {
-		error("htmlization of $type not supported");
-	}
-
-	if ($config{sanitize}) {
-		$content=scrubber()->scrub($content);
+use IkiWiki;
+use Encode;
+
+my %backlinks;
+our %brokenlinks;
+my $links_calculated=0;
+
+sub calculate_links () {
+	return if $links_calculated;
+	%backlinks=%brokenlinks=();
+	foreach my $page (keys %links) {
+		foreach my $link (@{$links{$page}}) {
+			my $bestlink=bestlink($page, $link);
+			if (length $bestlink) {
+				$backlinks{$bestlink}{$page}=1
+					if $bestlink ne $page;
+			}
+			else {
+				push @{$brokenlinks{$link}}, $page;
+			}
+		}
 	}
-	
-	return $content;
-} #}}}
+	$links_calculated=1;
+}
 
-sub backlinks ($) { #{{{
+sub backlink_pages ($) {
 	my $page=shift;
 
-	my @links;
-	foreach my $p (keys %links) {
-		next if bestlink($page, $p) eq $page;
-		if (grep { length $_ && bestlink($p, $_) eq $page } @{$links{$p}}) {
-			my $href=File::Spec->abs2rel(htmlpage($p), dirname($page));
-			
-			# Trim common dir prefixes from both pages.
-			my $p_trimmed=$p;
-			my $page_trimmed=$page;
-			my $dir;
-			1 while (($dir)=$page_trimmed=~m!^([^/]+/)!) &&
-			        defined $dir &&
-			        $p_trimmed=~s/^\Q$dir\E// &&
-			        $page_trimmed=~s/^\Q$dir\E//;
-				       
-			push @links, { url => $href, page => $p_trimmed };
-		}
-	}
-
-	return sort { $a->{page} cmp $b->{page} } @links;
-} #}}}
+	calculate_links();
 
-sub parentlinks ($) { #{{{
-	my $page=shift;
-	
-	my @ret;
-	my $pagelink="";
-	my $path="";
-	my $skip=1;
-	foreach my $dir (reverse split("/", $page)) {
-		if (! $skip) {
-			$path.="../";
-			unshift @ret, { url => "$path$dir.html", page => $dir };
-		}
-		else {
-			$skip=0;
-		}
-	}
-	unshift @ret, { url => length $path ? $path : ".", page => $config{wikiname} };
-	return @ret;
-} #}}}
+	return keys %{$backlinks{$page}};
+}
 
-sub rsspage ($) { #{{{
+sub backlinks ($) {
 	my $page=shift;
 
-	return $page.".rss";
-} #}}}
-
-sub preprocess ($$) { #{{{
+	my @links;
+	foreach my $p (backlink_pages($page)) {
+		my $href=urlto($p, $page);
+                
+		# Trim common dir prefixes from both pages.
+		my $p_trimmed=$p;
+		my $page_trimmed=$page;
+		my $dir;
+		1 while (($dir)=$page_trimmed=~m!^([^/]+/)!) &&
+		        defined $dir &&
+		        $p_trimmed=~s/^\Q$dir\E// &&
+		        $page_trimmed=~s/^\Q$dir\E//;
+			       
+		push @links, { url => $href, page => pagetitle($p_trimmed) };
+	}
+	return @links;
+}
+
+sub genpage ($$) {
 	my $page=shift;
 	my $content=shift;
 
-	my %commands=(inline => \&preprocess_inline);
-	
-	my $handle=sub {
-		my $escape=shift;
-		my $command=shift;
-		my $params=shift;
-		if (length $escape) {
-			return "[[$command $params]]";
+	my $templatefile;
+	run_hooks(templatefile => sub {
+		return if defined $templatefile;
+		my $file=shift->(page => $page);
+		if (defined $file && defined template_file($file)) {
+			$templatefile=$file;
 		}
-		elsif (exists $commands{$command}) {
-			my %params;
-			while ($params =~ /(\w+)=\"([^"]+)"(\s+|$)/g) {
-				$params{$1}=$2;
-			}
-			return $commands{$command}->(page => $page, %params);
-		}
-		else {
-			return "[[bad directive $command]]";
-		}
-	};
-	
-	$content =~ s{(\\?)$config{wiki_processor_regexp}}{$handle->($1, $2, $3)}eg;
-	return $content;
-} #}}}
-
-sub blog_list ($$) { #{{{
-	my $globlist=shift;
-	my $maxitems=shift;
-	
-	my @list;
-	foreach my $page (keys %pagesources) {
-		if (globlist_match($page, $globlist)) {
-			push @list, $page;
-		}
-	}
-
-	@list=sort { $pagectime{$b} <=> $pagectime{$a} } @list;
-	return @list if ! $maxitems || @list <= $maxitems;
-	return @list[0..$maxitems - 1];
-} #}}}
-
-sub get_inline_content ($$) { #{{{
-	my $parentpage=shift;
-	my $page=shift;
-	
-	my $file=$pagesources{$page};
-	my $type=pagetype($file);
-	if ($type ne 'unknown') {
-		return htmlize($type, linkify(readfile(srcfile($file)), $parentpage));
-	}
-	else {
-		return "";
-	}
-} #}}}
+	});
+	my $template=template(defined $templatefile ? $templatefile : 'page.tmpl', blind_cache => 1);
+	my $actions=0;
 
-sub preprocess_inline ($@) { #{{{
-	my %params=@_;
-	
-	if (! exists $params{pages}) {
-		return "";
-	}
-	if (! exists $params{archive}) {
-		$params{archive}="no";
-	}
-	if (! exists $params{show} && $params{archive} eq "no") {
-		$params{show}=10;
-	}
-	if (! exists $depends{$params{page}}) {
-		$depends{$params{page}}=$params{pages};
-	}
-	else {
-		$depends{$params{page}}.=" ".$params{pages};
+	if (length $config{cgiurl}) {
+		$template->param(editurl => cgiurl(do => "edit", page => $page))
+			if IkiWiki->can("cgi_editpage");
+		$template->param(prefsurl => cgiurl(do => "prefs"))
+			if exists $hooks{auth};
+		$actions++;
 	}
-
-	my $ret="";
-	
-	if (exists $params{rootpage}) {
-		# Add a blog post form, with a rss link button.
-		my $formtemplate=HTML::Template->new(blind_cache => 1,
-			filename => "$config{templatedir}/blogpost.tmpl");
-		$formtemplate->param(cgiurl => $config{cgiurl});
-		$formtemplate->param(rootpage => $params{rootpage});
-		if ($config{rss}) {
-			$formtemplate->param(rssurl => rsspage(basename($params{page})));
+		
+	if (defined $config{historyurl} && length $config{historyurl}) {
+		my $u=$config{historyurl};
+		$u=~s/\[\[file\]\]/$pagesources{$page}/g;
+		$template->param(historyurl => $u);
+		$actions++;
+	}
+	if ($config{discussion}) {
+		if ($page !~ /.*\/\Q$config{discussionpage}\E$/ &&
+		   (length $config{cgiurl} ||
+		    exists $links{$page."/".$config{discussionpage}})) {
+			$template->param(discussionlink => htmllink($page, $page, $config{discussionpage}, noimageinline => 1, forcesubpage => 1));
+			$actions++;
 		}
-		$ret.=$formtemplate->output;
-	}
-	elsif ($config{rss}) {
-		# Add a rss link button.
-		my $linktemplate=HTML::Template->new(blind_cache => 1,
-			filename => "$config{templatedir}/rsslink.tmpl");
-		$linktemplate->param(rssurl => rsspage(basename($params{page})));
-		$ret.=$linktemplate->output;
-	}
-	
-	my $template=HTML::Template->new(blind_cache => 1,
-		filename => (($params{archive} eq "no") 
-				? "$config{templatedir}/inlinepage.tmpl"
-				: "$config{templatedir}/inlinepagetitle.tmpl"));
-	
-	my @pages;
-	foreach my $page (blog_list($params{pages}, $params{show})) {
-		next if $page eq $params{page};
-		push @pages, $page;
-		$template->param(pagelink => htmllink($params{page}, $page));
-		$template->param(content => get_inline_content($params{page}, $page))
-			if $params{archive} eq "no";
-		$template->param(ctime => scalar(gmtime($pagectime{$page})));
-		$ret.=$template->output;
-	}
-	
-	# TODO: should really add this to renderedfiles and call
-	# check_overwrite, but currently renderedfiles
-	# only supports listing one file per page.
-	if ($config{rss}) {
-		writefile(rsspage($params{page}), $config{destdir},
-			genrss($params{page}, @pages));
 	}
-	
-	return $ret;
-} #}}}
-
-sub genpage ($$$) { #{{{
-	my $content=shift;
-	my $page=shift;
-	my $mtime=shift;
 
-	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 => cgiurl(do => "edit", page => $page));
-		$template->param(prefsurl => cgiurl(do => "prefs"));
-		if ($config{rcs}) {
-			$template->param(recentchangesurl => cgiurl(do => "recentchanges"));
-		}
+	if ($actions) {
+		$template->param(have_actions => 1);
 	}
 
-	if (length $config{historyurl}) {
-		my $u=$config{historyurl};
-		$u=~s/\[\[file\]\]/$pagesources{$page}/g;
-		$template->param(historyurl => $u);
+	my @backlinks=sort { $a->{page} cmp $b->{page} } backlinks($page);
+	my ($backlinks, $more_backlinks);
+	if (@backlinks <= $config{numbacklinks} || ! $config{numbacklinks}) {
+		$backlinks=\@backlinks;
+		$more_backlinks=[];
 	}
-	if ($config{hyperestraier}) {
-		$template->param(hyperestraierurl => cgiurl());
+	else {
+		$backlinks=[@backlinks[0..$config{numbacklinks}-1]];
+		$more_backlinks=[@backlinks[$config{numbacklinks}..$#backlinks]];
 	}
 
 	$template->param(
-		title => $title,
+		title => $page eq 'index' 
+			? $config{wikiname} 
+			: pagetitle(basename($page)),
 		wikiname => $config{wikiname},
-		parentlinks => [parentlinks($page)],
 		content => $content,
-		backlinks => [backlinks($page)],
-		discussionlink => htmllink($page, "Discussion", 1, 1),
-		mtime => scalar(gmtime($mtime)),
-		styleurl => styleurl($page),
+		backlinks => $backlinks,
+		more_backlinks => $more_backlinks,
+		mtime => displaytime($pagemtime{$page}),
+		ctime => displaytime($pagectime{$page}),
+		baseurl => baseurl($page),
 	);
-	
-	return $template->output;
-} #}}}
 
-sub date_822 ($) { #{{{
-	my $time=shift;
-
-	eval q{use POSIX};
-	return POSIX::strftime("%a, %d %b %Y %H:%M:%S %z", localtime($time));
-} #}}}
+	run_hooks(pagetemplate => sub {
+		shift->(page => $page, destpage => $page, template => $template);
+	});
+	
+	$content=$template->output;
+	
+	run_hooks(postscan => sub {
+		shift->(page => $page, content => $content);
+	});
 
-sub absolute_urls ($$) { #{{{
-	# sucky sub because rss sucks
-	my $content=shift;
-	my $url=shift;
+	run_hooks(format => sub {
+		$content=shift->(
+			page => $page,
+			content => $content,
+		);
+	});
 
-	$url=~s/[^\/]+$//;
-	
-	$content=~s/<a\s+href="(?!http:\/\/)([^"]+)"/<a href="$url$1"/ig;
-	$content=~s/<img\s+src="(?!http:\/\/)([^"]+)"/<img src="$url$1"/ig;
 	return $content;
-} #}}}
+}
 
-sub genrss ($@) { #{{{
-	my $page=shift;
-	my @pages=@_;
-	
-	my $url="$config{url}/".htmlpage($page);
-	
-	my $template=HTML::Template->new(blind_cache => 1,
-		filename => "$config{templatedir}/rsspage.tmpl");
-	
-	my @items;
-	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};
-	}
+sub scan ($) {
+	my $file=shift;
 
-	$template->param(
-		title => $config{wikiname},
-		pageurl => $url,
-		items => \@items,
-	);
-	
-	return $template->output;
-} #}}}
-
-sub check_overwrite ($$) { #{{{
-	# Important security check. Make sure to call this before saving
-	# any files to the source directory.
-	my $dest=shift;
-	my $src=shift;
-	
-	if (! exists $renderedfiles{$src} && -e $dest && ! $config{rebuild}) {
-		error("$dest already exists and was rendered from ".
-			join(" ",(grep { $renderedfiles{$_} eq $dest } keys
-				%renderedfiles)).
-			", before, so not rendering from $src");
-	}
-} #}}}
+	my $type=pagetype($file);
+	if (defined $type) {
+		my $srcfile=srcfile($file);
+		my $content=readfile($srcfile);
+		my $page=pagename($file);
+		will_render($page, htmlpage($page), 1);
 
-sub mtime ($) { #{{{
-	my $file=shift;
-	
-	return (stat($file))[9];
-} #}}}
+		if ($config{discussion}) {
+			# Discussion links are a special case since they're
+			# not in the text of the page, but on its template.
+			$links{$page}=[ $page."/".lc($config{discussionpage}) ];
+		}
+		else {
+			$links{$page}=[];
+		}
 
-sub findlinks ($$) { #{{{
-	my $content=shift;
-	my $page=shift;
+		run_hooks(scan => sub {
+			shift->(
+				page => $page,
+				content => $content,
+			);
+		});
 
-	my @links;
-	while ($content =~ /(?<!\\)$config{wiki_link_regexp}/g) {
-		push @links, titlepage($2);
+		# Preprocess in scan-only mode.
+		preprocess($page, $page, $content, 1);
+	}
+	else {
+		will_render($file, $file, 1);
+	}
+}
+
+sub fast_file_copy (@) {
+	my $srcfile=shift;
+	my $destfile=shift;
+	my $srcfd=shift;
+	my $destfd=shift;
+	my $cleanup=shift;
+
+	my $blksize = 16384;
+	my ($len, $buf, $written);
+	while ($len = sysread $srcfd, $buf, $blksize) {
+		if (! defined $len) {
+			next if $! =~ /^Interrupted/;
+			error("failed to read $srcfile: $!", $cleanup);
+		}
+		my $offset = 0;
+		while ($len) {
+			defined($written = syswrite $destfd, $buf, $len, $offset)
+				or error("failed to write $destfile: $!", $cleanup);
+			$len -= $written;
+			$offset += $written;
+		}
 	}
-	# Discussion links are a special case since they're not in the text
-	# of the page, but on its template.
-	return @links, "$page/discussion";
-} #}}}
+}
 
-sub render ($) { #{{{
+sub render ($) {
 	my $file=shift;
 	
 	my $type=pagetype($file);
 	my $srcfile=srcfile($file);
-	if ($type ne 'unknown') {
-		my $content=readfile($srcfile);
+	if (defined $type) {
 		my $page=pagename($file);
-		
-		$links{$page}=[findlinks($content, $page)];
 		delete $depends{$page};
+		delete $depends_simple{$page};
+		will_render($page, htmlpage($page), 1);
+		return if $type=~/^_/;
 		
-		$content=linkify($content, $page);
-		$content=preprocess($page, $content);
-		$content=htmlize($type, $content);
+		my $content=htmlize($page, $page, $type,
+			linkify($page, $page,
+			preprocess($page, $page,
+			filter($page, $page,
+			readfile($srcfile)))));
 		
-		check_overwrite("$config{destdir}/".htmlpage($page), $page);
-		writefile(htmlpage($page), $config{destdir},
-			genpage($content, $page, mtime($srcfile)));
-		$oldpagemtime{$page}=time;
-		$renderedfiles{$page}=htmlpage($page);
+		my $output=htmlpage($page);
+		writefile($output, $config{destdir}, genpage($page, $content));
 	}
 	else {
-		my $content=readfile($srcfile, 1);
-		$links{$file}=[];
-		check_overwrite("$config{destdir}/$file", $file);
-		writefile($file, $config{destdir}, $content, 1);
-		$oldpagemtime{$file}=time;
-		$renderedfiles{$file}=$file;
+		delete $depends{$file};
+		delete $depends_simple{$file};
+		will_render($file, $file, 1);
+		
+		if ($config{hardlink}) {
+			# only hardlink if owned by same user
+			my @stat=stat($srcfile);
+			if ($stat[4] == $>) {
+				prep_writefile($file, $config{destdir});
+				unlink($config{destdir}."/".$file);
+				if (link($srcfile, $config{destdir}."/".$file)) {
+					return;
+				}
+			}
+			# if hardlink fails, fall back to copying
+		}
+		
+		my $srcfd=readfile($srcfile, 1, 1);
+		writefile($file, $config{destdir}, undef, 1, sub {
+			fast_file_copy($srcfile, $file, $srcfd, @_);
+		});
 	}
-} #}}}
+}
 
-sub prune ($) { #{{{
+sub prune ($) {
 	my $file=shift;
 
 	unlink($file);
@@ -415,199 +256,261 @@ sub prune ($) { #{{{
 	while (rmdir($dir)) {
 		$dir=dirname($dir);
 	}
-} #}}}
-
-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";
+}
+
+sub srcdir_check () {
+	# security check, avoid following symlinks in the srcdir path by default
+	my $test=$config{srcdir};
+	while (length $test) {
+		if (-l $test && ! $config{allow_symlinks_before_srcdir}) {
+			error(sprintf(gettext("symlink found in srcdir path (%s) -- set allow_symlinks_before_srcdir to allow this"), $test));
+		}
+		unless ($test=~s/\/+$//) {
+			$test=dirname($test);
 		}
-		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;
-	my @files;
+sub find_src_files () {
+	my (@files, %pages);
 	eval q{use File::Find};
+	error($@) if $@;
 	find({
 		no_chdir => 1,
 		wanted => sub {
-			if (/$config{wiki_file_prune_regexp}/) {
+			$_=decode_utf8($_);
+			if (file_pruned($_, $config{srcdir})) {
 				$File::Find::prune=1;
 			}
-			elsif (! -d $_ && ! -l $_) {
+			elsif (! -l $_ && ! -d _) {
 				my ($f)=/$config{wiki_file_regexp}/; # untaint
 				if (! defined $f) {
-					warn("skipping bad filename $_\n");
+					warn(sprintf(gettext("skipping bad filename %s"), $_)."\n");
 				}
 				else {
 					$f=~s/^\Q$config{srcdir}\E\/?//;
 					push @files, $f;
-					$exists{pagename($f)}=1;
+					my $pagename = pagename($f);
+					if ($pages{$pagename}) {
+						debug(sprintf(gettext("%s has multiple possible source pages"), $pagename));
+					}
+					$pages{$pagename}=1;
 				}
 			}
 		},
 	}, $config{srcdir});
-	find({
-		no_chdir => 1,
-		wanted => sub {
-			if (/$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");
+	foreach my $dir (@{$config{underlaydirs}}, $config{underlaydir}) {
+		find({
+			no_chdir => 1,
+			wanted => sub {
+				$_=decode_utf8($_);
+				if (file_pruned($_, $dir)) {
+					$File::Find::prune=1;
 				}
-				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;
+				elsif (! -l $_ && ! -d _) {
+					my ($f)=/$config{wiki_file_regexp}/; # untaint
+					if (! defined $f) {
+						warn(sprintf(gettext("skipping bad filename %s"), $_)."\n");
+					}
+					else {
+						$f=~s/^\Q$dir\E\/?//;
+						# avoid underlaydir
+						# override attacks; see
+						# security.mdwn
+						if (! -l "$config{srcdir}/$f" && 
+						    ! -e _) {
+						    	my $page=pagename($f);
+							if (! $pages{$page}) {
+								push @files, $f;
+								$pages{$page}=1;
+							}
+						}
 					}
 				}
-			}
-		},
-	}, $config{underlaydir});
+			},
+		}, $dir);
+	};
 
-	my %rendered;
+	# Returns a list of all source files found, and a hash of 
+	# the corresponding page names.
+	return \@files, \%pages;
+}
 
+sub refresh () {
+	srcdir_check();
+	run_hooks(refresh => sub { shift->() });
+	my ($files, $exists)=find_src_files();
+
+	my (%rendered, @add, @del, @internal);
 	# check for added or removed pages
-	my @add;
-	foreach my $file (@files) {
+	foreach my $file (@$files) {
 		my $page=pagename($file);
-		if (! $oldpagemtime{$page}) {
-			debug("new page $page") unless exists $pagectime{$page};
-			push @add, $file;
-			$links{$page}=[];
-			$pagesources{$page}=$file;
-			$pagectime{$page}=mtime(srcfile($file))
-				unless exists $pagectime{$page};
+		if (exists $pagesources{$page} && $pagesources{$page} ne $file) {
+			# the page has changed its type
+			$forcerebuild{$page}=1;
+		}
+		$pagesources{$page}=$file;
+		if (! $pagemtime{$page}) {
+			if (isinternal($page)) {
+				push @internal, $file;
+			}
+			else {
+				push @add, $file;
+				if ($config{getctime} && -e "$config{srcdir}/$file") {
+					eval {
+						my $time=rcs_getctime("$config{srcdir}/$file");
+						$pagectime{$page}=$time;
+					};
+					if ($@) {
+						print STDERR $@;
+					}
+				}
+			}
+			$pagecase{lc $page}=$page;
+			if (! exists $pagectime{$page}) {
+				$pagectime{$page}=(srcfile_stat($file))[10];
+			}
 		}
 	}
-	my @del;
-	foreach my $page (keys %oldpagemtime) {
-		if (! $exists{$page}) {
-			debug("removing old page $page");
-			push @del, $pagesources{$page};
-			prune($config{destdir}."/".$renderedfiles{$page});
-			delete $renderedfiles{$page};
-			$oldpagemtime{$page}=0;
+	foreach my $page (keys %pagemtime) {
+		if (! $exists->{$page}) {
+			if (isinternal($page)) {
+				push @internal, $pagesources{$page};
+			}
+			else {
+				debug(sprintf(gettext("removing old page %s"), $page));
+				push @del, $pagesources{$page};
+			}
+			$links{$page}=[];
+			$renderedfiles{$page}=[];
+			$pagemtime{$page}=0;
+			foreach my $old (@{$oldrenderedfiles{$page}}) {
+				prune($config{destdir}."/".$old);
+			}
 			delete $pagesources{$page};
+			foreach my $source (keys %destsources) {
+				if ($destsources{$source} eq $page) {
+					delete $destsources{$source};
+				}
+			}
 		}
 	}
-	
-	# render any updated files
-	foreach my $file (@files) {
+
+	# find changed and new files
+	my @needsbuild;
+	foreach my $file (@$files) {
 		my $page=pagename($file);
-		
-		if (! exists $oldpagemtime{$page} ||
-		    mtime(srcfile($file)) > $oldpagemtime{$page}) {
-		    	debug("rendering changed file $file");
-			render($file);
-			$rendered{$file}=1;
+		my ($srcfile, @stat)=srcfile_stat($file);
+		if (! exists $pagemtime{$page} ||
+		    $stat[9] > $pagemtime{$page} ||
+	    	    $forcerebuild{$page}) {
+			$pagemtime{$page}=$stat[9];
+			if (isinternal($page)) {
+				push @internal, $file;
+				# Preprocess internal page in scan-only mode.
+				preprocess($page, $page, readfile($srcfile), 1);
+			}
+			else {
+				push @needsbuild, $file;
+			}
 		}
 	}
+	run_hooks(needsbuild => sub { shift->(\@needsbuild) });
+
+	# scan and render files
+	foreach my $file (@needsbuild) {
+		debug(sprintf(gettext("scanning %s"), $file));
+		scan($file);
+	}
+	calculate_links();
+	foreach my $file (@needsbuild) {
+		debug(sprintf(gettext("building %s"), $file));
+		render($file);
+		$rendered{$file}=1;
+	}
+	foreach my $file (@internal) {
+		# internal pages are not rendered
+		my $page=pagename($file);
+		delete $depends{$page};
+		delete $depends_simple{$page};
+		foreach my $old (@{$renderedfiles{$page}}) {
+			delete $destsources{$old};
+		}
+		$renderedfiles{$page}=[];
+	}
 	
-	# if any files were added or removed, check to see if each page
-	# needs an update due to linking to them or inlining them.
-	# TODO: inefficient; pages may get rendered above and again here;
-	# problem is the bestlink may have changed and we won't know until
-	# now
+	# rebuild pages that link to added or removed pages
 	if (@add || @del) {
-FILE:		foreach my $file (@files) {
-			my $page=pagename($file);
-			foreach my $f (@add, @del) {
-				my $p=pagename($f);
-				foreach my $link (@{$links{$page}}) {
-					if (bestlink($page, $link) eq $p) {
-		   				debug("rendering $file, which links to $p");
-						render($file);
-						$rendered{$file}=1;
-						next FILE;
-					}
-				}
+		foreach my $f (@add, @del) {
+			my $p=pagename($f);
+			foreach my $page (keys %{$backlinks{$p}}) {
+				my $file=$pagesources{$page};
+				next if $rendered{$file};
+		   		debug(sprintf(gettext("building %s, which links to %s"), $file, $p));
+				render($file);
+				$rendered{$file}=1;
 			}
 		}
 	}
 
-	# Handle backlinks; if a page has added/removed links, update the
-	# pages it links to. Also handles rebuilding dependat pages.
-	# TODO: inefficient; pages may get rendered above and again here;
-	# problem is the backlinks could be wrong in the first pass render
-	# above
-	if (%rendered || @del) {
-		foreach my $f (@files) {
+	if (%rendered || @del || @internal) {
+		my @changed=(keys %rendered, @del);
+
+ 		my %lcchanged = map { lc(pagename($_)) => 1 } @changed;
+ 
+		# rebuild dependant pages
+		foreach my $f (@$files) {
+			next if $rendered{$f};
 			my $p=pagename($f);
-			if (exists $depends{$p}) {
-				foreach my $file (keys %rendered, @del) {
-					my $page=pagename($file);
-					if (globlist_match($page, $depends{$p})) {
-						debug("rendering $f, which depends on $page");
-						render($f);
-						$rendered{$f}=1;
+			my $reason = undef;
+
+			if (exists $depends_simple{$p}) {
+				foreach my $d (keys %{$depends_simple{$p}}) {
+					if (exists $lcchanged{$d}) {
+						$reason = $d;
 						last;
 					}
 				}
 			}
+
+			if (exists $depends{$p} && ! defined $reason) {
+				D: foreach my $d (keys %{$depends{$p}}) {
+					my $sub=pagespec_translate($d);
+					next if $@ || ! defined $sub;
+
+					# only consider internal files
+					# if the page explicitly depends
+					# on such files
+					foreach my $file (@changed, $d =~ /internal\(/ ? @internal : ()) {
+						next if $file eq $f;
+						my $page=pagename($file);
+						if ($sub->($page, location => $p)) {
+							$reason = $page;
+							last D;
+						}
+					}
+				}
+			}
+
+			if (defined $reason) {
+				debug(sprintf(gettext("building %s, which depends on %s"), $f, $reason));
+				render($f);
+				$rendered{$f}=1;
+			}
 		}
 		
+		# handle backlinks; if a page has added/removed links,
+		# update the pages it links to
 		my %linkchanged;
-		foreach my $file (keys %rendered, @del) {
+		foreach my $file (@changed) {
 			my $page=pagename($file);
 			
 			if (exists $links{$page}) {
 				foreach my $link (map { bestlink($page, $_) } @{$links{$page}}) {
 					if (length $link &&
-					    ! exists $oldlinks{$page} ||
-					    ! grep { $_ eq $link } @{$oldlinks{$page}}) {
+					    (! exists $oldlinks{$page} ||
+					     ! grep { bestlink($page, $_) eq $link } @{$oldlinks{$page}})) {
 						$linkchanged{$link}=1;
 					}
 				}
@@ -615,37 +518,67 @@ FILE:		foreach my $file (@files) {
 			if (exists $oldlinks{$page}) {
 				foreach my $link (map { bestlink($page, $_) } @{$oldlinks{$page}}) {
 					if (length $link &&
-					    ! exists $links{$page} ||
-					    ! grep { $_ eq $link } @{$links{$page}}) {
+					    (! exists $links{$page} || 
+					     ! grep { bestlink($page, $_) eq $link } @{$links{$page}})) {
 						$linkchanged{$link}=1;
 					}
 				}
 			}
 		}
+
 		foreach my $link (keys %linkchanged) {
 		    	my $linkfile=$pagesources{$link};
 			if (defined $linkfile) {
-				debug("rendering $linkfile, to update its backlinks");
+				next if $rendered{$linkfile};
+				debug(sprintf(gettext("building %s, to update its backlinks"), $linkfile));
 				render($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");
+	# remove no longer rendered files
+	foreach my $src (keys %rendered) {
+		my $page=pagename($src);
+		foreach my $file (@{$oldrenderedfiles{$page}}) {
+			if (! grep { $_ eq $file } @{$renderedfiles{$page}}) {
+				debug(sprintf(gettext("removing %s, no longer built by %s"), $file, $page));
+				prune($config{destdir}."/".$file);
+			}
 		}
-		
-		debug("generating hyperestraier cgi config");
-		estcfg();
 	}
-} #}}}
+
+	if (@del) {
+		run_hooks(delete => sub { shift->(@del) });
+	}
+	if (%rendered) {
+		run_hooks(change => sub { shift->(keys %rendered) });
+	}
+}
+
+sub commandline_render () {
+	lockwiki();
+	loadindex();
+	unlockwiki();
+
+	my $srcfile=possibly_foolish_untaint($config{render});
+	my $file=$srcfile;
+	$file=~s/\Q$config{srcdir}\E\/?//;
+
+	my $type=pagetype($file);
+	die sprintf(gettext("ikiwiki: cannot build %s"), $srcfile)."\n" unless defined $type;
+	my $content=readfile($srcfile);
+	my $page=pagename($file);
+	$pagesources{$page}=$file;
+	$content=filter($page, $page, $content);
+	$content=preprocess($page, $page, $content);
+	$content=linkify($page, $page, $content);
+	$content=htmlize($page, $page, $type, $content);
+	$pagemtime{$page}=(stat($srcfile))[9];
+	$pagectime{$page}=$pagemtime{$page} if ! exists $pagectime{$page};
+
+	print genpage($page, $content);
+	exit 0;
+}
 
 1