#!/usr/bin/perl -w
# Copyright (C) 2014, Eric Wong <normalperson@yhbt.net> and all contributors
# License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt)
-#
-# We focus on the lowest common denominators here:
-# - targeted at text-only console browsers (lynx, w3m, etc..)
-# - Only basic HTML, CSS only for line-wrapping <pre> text content for GUIs
-# - No JavaScript, graphics or icons allowed.
-# - Must not rely on static content
-# - UTF-8 is only for user-content, 7-bit US-ASCII for us
-
-use 5.008;
use strict;
use warnings;
-use CGI qw(:cgi -nosticky); # PSGI/FastCGI/mod_perl compat
-use Encode qw(find_encoding);
-use PublicInbox::Config;
-use URI::Escape qw(uri_escape uri_unescape);
-our $enc_utf8 = find_encoding('UTF-8');
-our $LISTNAME_RE = qr!\A/([\w\.\-]+)!;
-our $pi_config;
+require PublicInbox::WWW;
+use CGI qw/-nosticky/;
+our $NO_SCRIPT_NAME;
BEGIN {
- $pi_config = PublicInbox::Config->new;
- # TODO: detect and reload config as needed
- if ($ENV{MOD_PERL}) {
- CGI->compile;
- }
-}
-
-if ($ENV{PI_PLACKUP}) {
- psgi_app();
-} else {
- my $ret = main();
- binmode STDOUT;
- if (@ARGV && $ARGV[0] eq 'static') {
- print $ret->[2]->[0];
- } else { # CGI
- cgi_print($ret);
- }
-}
-
-# private functions below
-
-sub main {
- # some servers (Ruby webrick) include scheme://host[:port] here,
- # which confuses CGI.pm when generating self_url.
- # RFC 3875 does not mention REQUEST_URI at all,
- # so nuke it since CGI.pm functions without it.
- delete $ENV{REQUEST_URI};
-
- my $cgi = CGI->new;
- my %ctx;
- if ($cgi->request_method !~ /\AGET|HEAD\z/) {
- return r(405, 'Method Not Allowed');
- }
- my $path_info = $enc_utf8->decode($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) || redirect_list_index(\%ctx, $cgi);
- } elsif ($path_info =~ m!$LISTNAME_RE(?:/|/index\.html)?\z!o) {
- invalid_list(\%ctx, $1) || get_index(\%ctx, $cgi, 0);
- } elsif ($path_info =~ m!$LISTNAME_RE/atom\.xml\z!o) {
- invalid_list(\%ctx, $1) || get_atom(\%ctx, $cgi, 0);
-
- # single-message pages
- } elsif ($path_info =~ m!$LISTNAME_RE/m/(\S+)\.txt\z!o) {
- invalid_list_mid(\%ctx, $1, $2) || get_mid_txt(\%ctx, $cgi);
- } elsif ($path_info =~ m!$LISTNAME_RE/m/(\S+)\.html\z!o) {
- invalid_list_mid(\%ctx, $1, $2) || get_mid_html(\%ctx, $cgi);
-
- # full-message page
- } elsif ($path_info =~ m!$LISTNAME_RE/f/(\S+)\.html\z!o) {
- invalid_list_mid(\%ctx, $1, $2) || get_full_html(\%ctx, $cgi);
-
- # convenience redirects, order matters
- } elsif ($path_info =~ m!$LISTNAME_RE/(?:m|f)/(\S+)\z!o) {
- invalid_list_mid(\%ctx, $1, $2) || redirect_mid(\%ctx, $cgi);
-
- } else {
- r404();
- }
-}
-
-sub r404 { r(404, 'Not Found') }
-
-# simple response for errors
-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");
- if (defined $git_dir) {
- $ctx->{git_dir} = $git_dir;
- $ctx->{listname} = $listname;
- return;
- }
- r404();
-}
-
-# returns undef if valid, array ref response if invalid
-sub invalid_list_mid {
- my ($ctx, $listname, $mid) = @_;
- my $ret = invalid_list($ctx, $listname, $mid);
- $ctx->{mid} = uri_unescape($mid) unless $ret;
- $ret;
-}
-
-# /$LISTNAME/atom.xml -> Atom feed, includes replies
-sub get_atom {
- my ($ctx, $cgi, $top) = @_;
- require PublicInbox::Feed;
- [ 200, [ 'Content-Type' => 'application/xml' ],
- [ PublicInbox::Feed->generate({
- git_dir => $ctx->{git_dir},
- listname => $ctx->{listname},
- pi_config => $pi_config,
- cgi => $cgi,
- top => $top,
- }) ]
- ];
-}
-
-# /$LISTNAME/?r=$GIT_COMMIT -> HTML only
-sub get_index {
- my ($ctx, $cgi, $top) = @_;
- require PublicInbox::Feed;
- [ 200, [ 'Content-Type' => 'text/html' ],
- [ PublicInbox::Feed->generate_html_index({
- git_dir => $ctx->{git_dir},
- listname => $ctx->{listname},
- pi_config => $pi_config,
- cgi => $cgi,
- top => $top,
- }) ]
- ];
-}
-
-# just returns a string ref for the blob in the current ctx
-sub mid2blob {
- my ($ctx) = @_;
- local $ENV{GIT_DIR} = $ctx->{git_dir};
- require Digest::SHA;
- my $hex = Digest::SHA::sha1_hex($ctx->{mid});
- $hex =~ /\A([a-f0-9]{2})([a-f0-9]{38})\z/i or
- die "BUG: not a SHA-1 hex: $hex";
- my $blob = `git cat-file blob HEAD:$1/$2 2>/dev/null`;
- $? == 0 ? \$blob : undef;
-}
-
-# /$LISTNAME/m/$MESSAGE_ID.txt -> raw original
-sub get_mid_txt {
- my ($ctx, $cgi) = @_;
- my $x = mid2blob($ctx);
- $x ? [ 200, [ 'Content-Type' => 'text/plain' ], [ $$x ] ] : r404();
-}
-
-# /$LISTNAME/m/$MESSAGE_ID.html -> HTML content (short quotes)
-sub get_mid_html {
- my ($ctx, $cgi) = @_;
- my $x = mid2blob($ctx);
- return r404() unless $x;
-
- require PublicInbox::View;
- my $mid_href = PublicInbox::Hval::ascii_html(uri_escape($ctx->{mid}));
- my $pfx = "../f/$mid_href.html";
- require Email::MIME;
- [ 200, [ 'Content-Type' => 'text/html' ],
- [ PublicInbox::View->as_html(Email::MIME->new($$x), $pfx) ] ];
-}
-
-# /$LISTNAME/f/$MESSAGE_ID.html -> HTML content (fullquotes)
-sub get_full_html {
- my ($ctx, $cgi) = @_;
- my $x = mid2blob($ctx);
- return r404() unless $x;
- require PublicInbox::View;
- require Email::MIME;
- [ 200, [ 'Content-Type' => 'text/html' ],
- [ PublicInbox::View->as_html(Email::MIME->new($$x))] ];
-}
-
-sub redirect_list_index {
- my ($ctx, $cgi) = @_;
- do_redirect($cgi->self_url . "/");
-}
-
-sub redirect_mid {
- my ($ctx, $cgi) = @_;
- my $url = $cgi->self_url;
- $url =~ s!/f/!/m/!;
- do_redirect($url . '.html');
-}
-
-sub do_redirect {
- my ($url) = @_;
- [ 301,
- [ Location => $url, 'Content-Type' => 'text/plain' ],
- [ "Redirecting to $url\n" ]
- ]
-}
-
-sub psgi_app {
- require CGI::Emulate::PSGI;
-
- # preload so we are CoW friendly
- require PublicInbox::Feed;
- require PublicInbox::View;
- require Mail::Thread;
- require Digest::SHA;
- require POSIX;
- require XML::Atom::SimpleFeed;
- eval { require Git };
- sub {
- my ($e) = @_;
- local %ENV = (%ENV, CGI::Emulate::PSGI->emulate_environment($e));
- main();
- }
-}
-
-sub cgi_print {
- my ($ret) = @_;
+ $NO_SCRIPT_NAME = 1 if $ENV{NO_SCRIPT_NAME};
+ CGI->compile if $ENV{MOD_PERL};
+}
+
+# some servers (Ruby webrick) include scheme://host[:port] here,
+# which confuses CGI.pm when generating self_url.
+# RFC 3875 does not mention REQUEST_URI at all,
+# so nuke it since CGI.pm functions without it.
+delete $ENV{REQUEST_URI};
+$ENV{SCRIPT_NAME} = '' if $NO_SCRIPT_NAME;
+my $req = CGI->new;
+my $ret = PublicInbox::WWW::run($req, $req->request_method);
+binmode STDOUT;
+if (@ARGV && $ARGV[0] eq 'static') {
+ print $ret->[2]->[0]; # only show the body
+} else { # CGI
my ($status, $headers, $body) = @$ret;
my %codes = (
200 => 'OK',