X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FWWW.pm;h=e87e559428ffc55400b9942d4c1561c7197dd44e;hb=65ff86f333a1f8276943248efef7fa2a64afe9e3;hp=9ae7f7be40296ce11f68f9c3416cffb68ab901a8;hpb=5f3cef806f38dfd7f44d203923051f2ce5f4499a;p=public-inbox.git

diff --git a/lib/PublicInbox/WWW.pm b/lib/PublicInbox/WWW.pm
index 9ae7f7be..e87e5594 100644
--- a/lib/PublicInbox/WWW.pm
+++ b/lib/PublicInbox/WWW.pm
@@ -1,6 +1,8 @@
-# Copyright (C) 2014, Eric Wong <normalperson@yhbt.net> and all contributors
+# Copyright (C) 2014-2015 all contributors <meta@public-inbox.org>
 # License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt)
 #
+# Main web interface for mailing list archives
+#
 # We focus on the lowest common denominators here:
 # - targeted at text-only console browsers (w3m, links, etc..)
 # - Only basic HTML, CSS only for line-wrapping <pre> text content for GUIs
@@ -11,61 +13,74 @@ package PublicInbox::WWW;
 use 5.008;
 use strict;
 use warnings;
-use PublicInbox::Config;
+use PublicInbox::Config qw(try_cat);
 use URI::Escape qw(uri_escape_utf8 uri_unescape);
 use constant SSOMA_URL => 'http://ssoma.public-inbox.org/';
 use constant PI_URL => 'http://public-inbox.org/';
+require PublicInbox::Git;
+use PublicInbox::GitHTTPBackend;
 our $LISTNAME_RE = qr!\A/([\w\.\-]+)!;
 our $MID_RE = qr!([^/]+)!;
-our $pi_config;
+our $END_RE = qr!(f/|T/|t/|R/|t\.mbox(?:\.gz)?|t\.atom|raw|)!;
 
-sub run {
-	my ($cgi, $method) = @_;
+sub new {
+	my ($class, $pi_config) = @_;
 	$pi_config ||= PublicInbox::Config->new;
-	my %ctx = (cgi => $cgi, pi_config => $pi_config);
-	if ($method !~ /\AGET|HEAD\z/) {
+	bless { pi_config => $pi_config }, $class;
+}
+
+# backwards compatibility, do not use
+sub run {
+	my ($req, $method) = @_;
+	PublicInbox::WWW->new->call($req->env);
+}
+
+sub call {
+	my ($self, $env) = @_;
+	my $cgi = Plack::Request->new($env);
+	my $ctx = { cgi => $cgi, pi_config => $self->{pi_config} };
+	my $path_info = $cgi->path_info;
+
+	my $method = $cgi->method;
+	if ($method eq 'POST' &&
+		 $path_info =~ m!$LISTNAME_RE/(git-upload-pack)\z!) {
+		my $path = $2;
+		return (invalid_list($ctx, $1) ||
+			serve_git($cgi, $ctx->{git}, $path));
+	}
+	elsif ($method !~ /\AGET|HEAD\z/) {
 		return r(405, 'Method Not Allowed');
 	}
-	my $path_info = $cgi->path_info;
 
 	# top-level indices and feeds
 	if ($path_info eq '/') {
 		r404();
 	} elsif ($path_info =~ m!$LISTNAME_RE\z!o) {
-		invalid_list(\%ctx, $1) || r301(\%ctx, $1);
+		invalid_list($ctx, $1) || r301($ctx, $1);
 	} elsif ($path_info =~ m!$LISTNAME_RE(?:/|/index\.html)?\z!o) {
-		invalid_list(\%ctx, $1) || get_index(\%ctx);
+		invalid_list($ctx, $1) || get_index($ctx);
 	} elsif ($path_info =~ m!$LISTNAME_RE/(?:atom\.xml|new\.atom)\z!o) {
-		invalid_list(\%ctx, $1) || get_atom(\%ctx);
+		invalid_list($ctx, $1) || get_atom($ctx);
 
-	# thread display
-	} elsif ($path_info =~ m!$LISTNAME_RE/$MID_RE/t/\z!o) {
-		invalid_list_mid(\%ctx, $1, $2) || get_thread(\%ctx);
-	} elsif ($path_info =~ m!$LISTNAME_RE/$MID_RE/t\.mbox(\.gz)?\z!o) {
-		my $sfx = $3;
-		invalid_list_mid(\%ctx, $1, $2) || get_thread_mbox(\%ctx, $sfx);
-	} elsif ($path_info =~ m!$LISTNAME_RE/$MID_RE/t\.atom\z!o) {
-		invalid_list_mid(\%ctx, $1, $2) || get_thread_atom(\%ctx);
-	} elsif ($path_info =~ m!$LISTNAME_RE/$MID_RE/T/\z!o) {
-		$ctx{flat} = 1;
-		invalid_list_mid(\%ctx, $1, $2) || get_thread(\%ctx);
+	} elsif ($path_info =~ m!$LISTNAME_RE/
+				($PublicInbox::GitHTTPBackend::ANY)\z!ox) {
+		my $path = $2;
+		invalid_list($ctx, $1) || serve_git($cgi, $ctx->{git}, $path);
+	} elsif ($path_info =~ m!$LISTNAME_RE/$MID_RE/$END_RE\z!o) {
+		msg_page($ctx, $1, $2, $3);
 
-	# single-message pages
-	} elsif ($path_info =~ m!$LISTNAME_RE/$MID_RE/\z!o) {
-		invalid_list_mid(\%ctx, $1, $2) || get_mid_html(\%ctx);
-	} elsif ($path_info =~ m!$LISTNAME_RE/$MID_RE/raw\z!o) {
-		invalid_list_mid(\%ctx, $1, $2) || get_mid_txt(\%ctx);
-
-	# full-message page
-	} elsif ($path_info =~ m!$LISTNAME_RE/$MID_RE/f/\z!o) {
-		invalid_list_mid(\%ctx, $1, $2) || get_full_html(\%ctx);
+	# in case people leave off the trailing slash:
+	} elsif ($path_info =~ m!$LISTNAME_RE/$MID_RE/(f|T|t|R)\z!o) {
+		my ($listname, $mid, $suffix) = ($1, $2, $3);
+		$suffix .= $suffix =~ /\A[tT]\z/ ? '/#u' : '/';
+		r301($ctx, $listname, $mid, $suffix);
 
 	# convenience redirects order matters
 	} elsif ($path_info =~ m!$LISTNAME_RE/([^/]{2,})\z!o) {
-		r301(\%ctx, $1, $2);
+		r301($ctx, $1, $2);
 
 	} else {
-		legacy_redirects(\%ctx, $path_info);
+		legacy_redirects($ctx, $path_info);
 	}
 }
 
@@ -74,13 +89,13 @@ sub preload {
 	require PublicInbox::Feed;
 	require PublicInbox::View;
 	require PublicInbox::Thread;
-	require PublicInbox::GitCatFile;
 	require Email::MIME;
 	require Digest::SHA;
 	require POSIX;
 
 	eval {
 		require PublicInbox::Search;
+		require PublicInbox::SearchView;
 		require PublicInbox::Mbox;
 		require IO::Compress::Gzip;
 	};
@@ -88,7 +103,15 @@ sub preload {
 
 # private functions below
 
-sub r404 { r(404, 'Not Found') }
+sub r404 {
+	my ($ctx) = @_;
+	if ($ctx && $ctx->{mid}) {
+		require PublicInbox::ExtMsg;
+		searcher($ctx);
+		return PublicInbox::ExtMsg::ext_msg($ctx);
+	}
+	r(404, 'Not Found');
+}
 
 # simple response for errors
 sub r { [ $_[0], ['Content-Type' => 'text/plain'], [ join(' ', @_, "\n") ] ] }
@@ -96,9 +119,10 @@ sub r { [ $_[0], ['Content-Type' => 'text/plain'], [ join(' ', @_, "\n") ] ] }
 # returns undef if valid, array ref response if invalid
 sub invalid_list {
 	my ($ctx, $listname) = @_;
-	my $git_dir = $pi_config->get($listname, "mainrepo");
+	my $git_dir = $ctx->{pi_config}->get($listname, "mainrepo");
 	if (defined $git_dir) {
 		$ctx->{git_dir} = $git_dir;
+		$ctx->{git} = PublicInbox::Git->new($git_dir);
 		$ctx->{listname} = $listname;
 		return;
 	}
@@ -109,8 +133,18 @@ sub invalid_list {
 sub invalid_list_mid {
 	my ($ctx, $listname, $mid) = @_;
 	my $ret = invalid_list($ctx, $listname, $mid);
-	$ctx->{mid} = uri_unescape($mid) unless $ret;
-	$ret;
+	return $ret if $ret;
+
+	$ctx->{mid} = $mid = uri_unescape($mid);
+	if ($mid =~ /\A[a-f0-9]{40}\z/) {
+		if ($mid = mid2blob($ctx)) {
+			require Email::Simple;
+			use PublicInbox::MID qw/mid_clean/;
+			$mid = Email::Simple->new($mid);
+			$ctx->{mid} = mid_clean($mid->header('Message-ID'));
+		}
+	}
+	undef;
 }
 
 # /$LISTNAME/new.atom                     -> Atom feed, includes replies
@@ -126,7 +160,12 @@ sub get_index {
 	require PublicInbox::Feed;
 	my $srch = searcher($ctx);
 	footer($ctx);
-	PublicInbox::Feed::generate_html_index($ctx);
+	if (defined $ctx->{cgi}->param('q')) {
+		require PublicInbox::SearchView;
+		PublicInbox::SearchView::sres_top_html($ctx);
+	} else {
+		PublicInbox::Feed::generate_html_index($ctx);
+	}
 }
 
 # just returns a string ref for the blob in the current ctx
@@ -134,32 +173,21 @@ sub mid2blob {
 	my ($ctx) = @_;
 	require PublicInbox::MID;
 	my $path = PublicInbox::MID::mid2path($ctx->{mid});
-	my @cmd = ('git', "--git-dir=$ctx->{git_dir}",
-			qw(cat-file blob), "HEAD:$path");
-	my $pid = open my $fh, '-|';
-	defined $pid or die "fork failed: $!\n";
-	if ($pid == 0) {
-		open STDERR, '>', '/dev/null'; # ignore errors
-		exec @cmd or die "exec failed: $!\n";
-	} else {
-		my $blob = eval { local $/; <$fh> };
-		close $fh;
-		$? == 0 ? \$blob : undef;
-	}
+	$ctx->{git}->cat_file("HEAD:$path");
 }
 
 # /$LISTNAME/$MESSAGE_ID/raw                    -> raw mbox
 sub get_mid_txt {
 	my ($ctx) = @_;
-	my $x = mid2blob($ctx) or return r404();
+	my $x = mid2blob($ctx) or return r404($ctx);
 	require PublicInbox::Mbox;
-	PublicInbox::Mbox::emit1($x);
+	PublicInbox::Mbox::emit1($ctx, $x);
 }
 
 # /$LISTNAME/$MESSAGE_ID/                   -> HTML content (short quotes)
 sub get_mid_html {
 	my ($ctx) = @_;
-	my $x = mid2blob($ctx) or return r404();
+	my $x = mid2blob($ctx) or return r404($ctx);
 
 	require PublicInbox::View;
 	my $foot = footer($ctx);
@@ -173,7 +201,7 @@ sub get_mid_html {
 # /$LISTNAME/$MESSAGE_ID/f/                   -> HTML content (fullquotes)
 sub get_full_html {
 	my ($ctx) = @_;
-	my $x = mid2blob($ctx) or return r404();
+	my $x = mid2blob($ctx) or return r404($ctx);
 
 	require PublicInbox::View;
 	my $foot = footer($ctx);
@@ -184,20 +212,29 @@ sub get_full_html {
 	  [ PublicInbox::View::msg_html($ctx, $mime, undef, $foot)] ];
 }
 
+# /$LISTNAME/$MESSAGE_ID/R/                   -> HTML content (fullquotes)
+sub get_reply_html {
+	my ($ctx) = @_;
+	my $x = mid2blob($ctx) or return r404($ctx);
+
+	require PublicInbox::View;
+	my $foot = footer($ctx);
+	require Email::MIME;
+	my $hdr = Email::MIME->new($x)->header_obj;
+	[ 200, [ 'Content-Type' => 'text/html; charset=UTF-8' ],
+	  [ PublicInbox::View::msg_reply($ctx, $hdr, $foot)] ];
+}
+
 # /$LISTNAME/$MESSAGE_ID/t/
 sub get_thread {
-	my ($ctx) = @_;
+	my ($ctx, $flat) = @_;
 	my $srch = searcher($ctx) or return need_search($ctx);
 	require PublicInbox::View;
 	my $foot = footer($ctx);
+	$ctx->{flat} = $flat;
 	PublicInbox::View::thread_html($ctx, $foot, $srch);
 }
 
-sub self_url {
-	my ($cgi) = @_;
-	ref($cgi) eq 'CGI' ? $cgi->self_url : $cgi->uri->as_string;
-}
-
 sub ctx_get {
 	my ($ctx, $key) = @_;
 	my $val = $ctx->{$key};
@@ -205,17 +242,6 @@ sub ctx_get {
 	$val;
 }
 
-sub try_cat {
-	my ($path) = @_;
-	my $rv;
-	if (open(my $fh, '<', $path)) {
-		local $/;
-		$rv = <$fh>;
-		close $fh;
-	}
-	$rv;
-}
-
 sub footer {
 	my ($ctx) = @_;
 	return '' unless $ctx;
@@ -237,18 +263,20 @@ sub footer {
 
 	my $urls = try_cat("$git_dir/cloneurl");
 	my @urls = split(/\r?\n/, $urls || '');
-	my $nurls = scalar @urls;
-	if ($nurls == 0) {
-		$urls = '($GIT_DIR/cloneurl missing)';
-	} elsif ($nurls == 1) {
-		$urls = "git URL for <a\nhref=\"" . SSOMA_URL .
-			'">ssoma</a>: ' . $urls[0];
+	my %seen = map { $_ => 1 } @urls;
+	my $http = $ctx->{cgi}->base->as_string . $listname;
+	$seen{$http} or unshift @urls, $http;
+	if (scalar(@urls) == 1) {
+		$urls = "URL for <a\nhref=\"" . SSOMA_URL .
+			qq(">ssoma</a> or <b>git clone --mirror \$URL</b> :) .
+			$urls[0];
 	} else {
-		$urls = "git URLs for <a\nhref=\"" . SSOMA_URL .
-			"\">ssoma</a>:\n" . join("\n", map { "\t$_" } @urls);
+		$urls = "URLs for <a\nhref=\"" . SSOMA_URL .
+			qq(">ssoma</a> or <b>git clone --mirror \$URL</b>\n) .
+			join("\n", map { "\t$_" } @urls);
 	}
 
-	my $addr = $pi_config->get($listname, 'address');
+	my $addr = $ctx->{pi_config}->get($listname, 'address');
 	if (ref($addr) eq 'ARRAY') {
 		$addr = $addr->[0]; # first address is primary
 	}
@@ -301,7 +329,7 @@ sub get_thread_mbox {
 sub get_thread_atom {
 	my ($ctx) = @_;
 	searcher($ctx) or return need_search($ctx);
-	$ctx->{self_url} = self_url($ctx->{cgi});
+	$ctx->{self_url} = $ctx->{cgi}->uri->as_string;
 	require PublicInbox::Feed;
 	PublicInbox::Feed::generate_thread_atom($ctx);
 }
@@ -349,6 +377,14 @@ sub legacy_redirects {
 	} elsif ($path_info =~ m!$LISTNAME_RE/f/(\S+)\z!o) {
 		r301($ctx, $1, $2, 'f/');
 
+	# some Message-IDs have slashes in them and the HTTP server
+	# may try to be clever and unescape them :<
+	} elsif ($path_info =~ m!$LISTNAME_RE/(\S+/\S+)/$END_RE\z!o) {
+		msg_page($ctx, $1, $2, $3);
+
+	# in case people leave off the trailing slash:
+	} elsif ($path_info =~ m!$LISTNAME_RE/(\S+/\S+)/(f|T|t)\z!o) {
+		r301($ctx, $1, $2, $3 eq 't' ? 't/#u' : $3);
 	} else {
 		r404();
 	}
@@ -358,19 +394,36 @@ sub r301 {
 	my ($ctx, $listname, $mid, $suffix) = @_;
 	my $cgi = $ctx->{cgi};
 	my $url;
-	if (ref($cgi) eq 'CGI') {
-		$url = $cgi->url(-base) . '/';
-	} else {
-		$url = $cgi->base->as_string;
-	}
-
-	$url .= $listname . '/';
+	my $qs = $cgi->env->{QUERY_STRING};
+	$url = $cgi->base->as_string . $listname . '/';
 	$url .= (uri_escape_utf8($mid) . '/') if (defined $mid);
 	$url .= $suffix if (defined $suffix);
+	$url .= "?$qs" if $qs ne '';
 
 	[ 301,
 	  [ Location => $url, 'Content-Type' => 'text/plain' ],
 	  [ "Redirecting to $url\n" ] ]
 }
 
+sub msg_page {
+	my ($ctx, $list, $mid, $e) = @_;
+	unless (invalid_list_mid($ctx, $list, $mid)) {
+		'' eq $e and return get_mid_html($ctx);
+		't/' eq $e and return get_thread($ctx);
+		't.atom' eq $e and return get_thread_atom($ctx);
+		't.mbox' eq $e and return get_thread_mbox($ctx);
+		't.mbox.gz' eq $e and return get_thread_mbox($ctx, '.gz');
+		'T/' eq $e and return get_thread($ctx, 1);
+		'raw' eq $e and return get_mid_txt($ctx);
+		'f/' eq $e and return get_full_html($ctx);
+		'R/' eq $e and return get_reply_html($ctx);
+	}
+	r404($ctx);
+}
+
+sub serve_git {
+	my ($cgi, $git, $path) = @_;
+	PublicInbox::GitHTTPBackend::serve($cgi, $git, $path);
+}
+
 1;