#
# 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 text content for GUIs
+# and diff/syntax-highlighting (optional)
# - 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
package PublicInbox::WWW;
-use 5.008;
+use 5.010_001;
use strict;
use warnings;
-use Plack::Request;
+use bytes (); # only for bytes::length
+use Plack::Util;
use PublicInbox::Config;
-use URI::Escape qw(uri_escape_utf8 uri_unescape);
-use constant SSOMA_URL => '//ssoma.public-inbox.org/';
-use constant PI_URL => '//public-inbox.org/';
+use PublicInbox::Hval;
+use URI::Escape qw(uri_unescape);
+use PublicInbox::MID qw(mid_escape);
require PublicInbox::Git;
use PublicInbox::GitHTTPBackend;
-our $INBOX_RE = qr!\A/([\w\.\-]+)!;
+use PublicInbox::UserContent;
+
+# TODO: consider a routing tree now that we have more endpoints:
+our $INBOX_RE = qr!\A/([\w\-][\w\.\-]*)!;
our $MID_RE = qr!([^/]+)!;
-our $END_RE = qr!(T/|t/|R/|t\.mbox(?:\.gz)?|t\.atom|raw|)!;
+our $END_RE = qr!(T/|t/|t\.mbox(?:\.gz)?|t\.atom|raw|)!;
our $ATTACH_RE = qr!(\d[\.\d]*)-([[:alnum:]][\w\.-]+[[:alnum:]])!i;
+our $OID_RE = qr![a-f0-9]{7,40}!;
sub new {
my ($class, $pi_config) = @_;
@@ -37,18 +43,43 @@ sub run {
PublicInbox::WWW->new->call($req->env);
}
+my %path_re_cache;
+
+sub path_re ($) {
+ my $sn = $_[0]->{SCRIPT_NAME};
+ $path_re_cache{$sn} ||= do {
+ $sn = '/'.$sn unless index($sn, '/') == 0;
+ $sn =~ s!/\z!!;
+ qr!\A(?:https?://[^/]+)?\Q$sn\E(/[^\?\#]+)!;
+ };
+}
+
sub call {
my ($self, $env) = @_;
- my $cgi = Plack::Request->new($env);
- my $ctx = {cgi => $cgi, pi_config => $self->{pi_config}, www => $self};
- my $path_info = $env->{PATH_INFO};
+ my $ctx = { env => $env, www => $self };
+
+ # we don't care about multi-value
+ my %qp = map {
+ utf8::decode($_);
+ my ($k, $v) = split('=', uri_unescape($_), 2);
+ $v = '' unless defined $v;
+ $v =~ tr/+/ /;
+ ($k, $v)
+ } split(/[&;]+/, $env->{QUERY_STRING});
+ $ctx->{qp} = \%qp;
+
+ # not using $env->{PATH_INFO} here since that's already decoded
+ my ($path_info) = ($env->{REQUEST_URI} =~ path_re($env));
my $method = $env->{REQUEST_METHOD};
- if ($method eq 'POST' &&
- $path_info =~ m!$INBOX_RE/(git-upload-pack)\z!) {
- my $path = $2;
- return (invalid_inbox($self, $ctx, $1) ||
- serve_git($env, $ctx->{git}, $path));
+ if ($method eq 'POST') {
+ if ($path_info =~ m!$INBOX_RE/(?:(\d+)/)?(git-upload-pack)\z!) {
+ my ($part, $path) = ($2, $3);
+ return invalid_inbox($ctx, $1) ||
+ serve_git($ctx, $part, $path);
+ } elsif ($path_info =~ m!$INBOX_RE/!o) {
+ return invalid_inbox($ctx, $1) || mbox_results($ctx);
+ }
}
elsif ($method !~ /\AGET|HEAD\z/) {
return r(405, 'Method Not Allowed');
@@ -58,50 +89,63 @@ sub call {
if ($path_info eq '/') {
r404();
} elsif ($path_info =~ m!$INBOX_RE\z!o) {
- invalid_inbox($self, $ctx, $1) || r301($ctx, $1);
+ invalid_inbox($ctx, $1) || r301($ctx, $1);
} elsif ($path_info =~ m!$INBOX_RE(?:/|/index\.html)?\z!o) {
- invalid_inbox($self, $ctx, $1) || get_index($ctx);
+ invalid_inbox($ctx, $1) || get_index($ctx);
} elsif ($path_info =~ m!$INBOX_RE/(?:atom\.xml|new\.atom)\z!o) {
- invalid_inbox($self, $ctx, $1) || get_atom($ctx);
-
- } elsif ($path_info =~ m!$INBOX_RE/
+ invalid_inbox($ctx, $1) || get_atom($ctx);
+ } elsif ($path_info =~ m!$INBOX_RE/new\.html\z!o) {
+ invalid_inbox($ctx, $1) || get_new($ctx);
+ } elsif ($path_info =~ m!$INBOX_RE/(?:(\d+)/)?
($PublicInbox::GitHTTPBackend::ANY)\z!ox) {
- my $path = $2;
- invalid_inbox($self, $ctx, $1) ||
- serve_git($env, $ctx->{git}, $path);
+ my ($part, $path) = ($2, $3);
+ invalid_inbox($ctx, $1) || serve_git($ctx, $part, $path);
} elsif ($path_info =~ m!$INBOX_RE/([\w-]+).mbox\.gz\z!o) {
- serve_mbox_range($self, $ctx, $1, $2);
+ serve_mbox_range($ctx, $1, $2);
} elsif ($path_info =~ m!$INBOX_RE/$MID_RE/$END_RE\z!o) {
- msg_page($self, $ctx, $1, $2, $3);
+ msg_page($ctx, $1, $2, $3);
} elsif ($path_info =~ m!$INBOX_RE/$MID_RE/$ATTACH_RE\z!o) {
my ($idx, $fn) = ($3, $4);
- invalid_inbox_mid($self, $ctx, $1, $2) ||
- get_attach($ctx, $idx, $fn);
+ invalid_inbox_mid($ctx, $1, $2) || get_attach($ctx, $idx, $fn);
# in case people leave off the trailing slash:
- } elsif ($path_info =~ m!$INBOX_RE/$MID_RE/(T|t|R)\z!o) {
- my ($inbox, $mid, $suffix) = ($1, $2, $3);
+ } elsif ($path_info =~ m!$INBOX_RE/$MID_RE/(T|t)\z!o) {
+ my ($inbox, $mid_ue, $suffix) = ($1, $2, $3);
$suffix .= $suffix =~ /\A[tT]\z/ ? '/#u' : '/';
- r301($ctx, $inbox, $mid, $suffix);
+ r301($ctx, $inbox, $mid_ue, $suffix);
+
+ } elsif ($path_info =~ m!$INBOX_RE/$MID_RE/R/?\z!o) {
+ my ($inbox, $mid_ue) = ($1, $2);
+ r301($ctx, $inbox, $mid_ue, '#R');
} elsif ($path_info =~ m!$INBOX_RE/$MID_RE/f/?\z!o) {
r301($ctx, $1, $2);
-
+ } elsif ($path_info =~ m!$INBOX_RE/_/text(?:/(.*))?\z!o) {
+ get_text($ctx, $1, $2);
+ } elsif ($path_info =~ m!$INBOX_RE/([\w\-\.]+)\.css\z!o) {
+ get_css($ctx, $1, $2);
+ } elsif ($path_info =~ m!$INBOX_RE/($OID_RE)/s/\z!o) {
+ get_vcs_object($ctx, $1, $2);
+ } elsif ($path_info =~ m!$INBOX_RE/($OID_RE)/s/([\w\.\-]+)\z!o) {
+ get_vcs_object($ctx, $1, $2, $3);
+ } elsif ($path_info =~ m!$INBOX_RE/($OID_RE)/s\z!o) {
+ r301($ctx, $1, $2, 's/');
# convenience redirects order matters
} elsif ($path_info =~ m!$INBOX_RE/([^/]{2,})\z!o) {
r301($ctx, $1, $2);
} else {
- legacy_redirects($self, $ctx, $path_info);
+ legacy_redirects($ctx, $path_info);
}
}
# for CoW-friendliness, MOOOOO!
sub preload {
+ my ($self) = @_;
require PublicInbox::Feed;
require PublicInbox::View;
- require PublicInbox::Thread;
- require Email::MIME;
+ require PublicInbox::SearchThread;
+ require PublicInbox::MIME;
require Digest::SHA;
require POSIX;
@@ -110,6 +154,10 @@ sub preload {
PublicInbox::NewsWWW)) {
eval "require $_;";
}
+ if (ref($self)) {
+ $self->cgit;
+ $self->stylesheets_prepare($_) for ('', '../', '../../');
+ }
}
# private functions below
@@ -127,17 +175,20 @@ sub r404 {
# simple response for errors
sub r { [ $_[0], ['Content-Type' => 'text/plain'], [ join(' ', @_, "\n") ] ] }
+sub news_cgit_fallback ($) {
+ my ($ctx) = @_;
+ my $www = $ctx->{www};
+ my $env = $ctx->{env};
+ my $res = $www->news_www->call($env);
+ $res->[0] == 404 ? $www->cgit->call($env) : $res;
+}
+
# returns undef if valid, array ref response if invalid
-sub invalid_inbox {
- my ($self, $ctx, $inbox) = @_;
- my $obj = $ctx->{pi_config}->lookup_name($inbox);
- if (defined $obj) {
- $ctx->{git_dir} = $obj->{mainrepo};
- $ctx->{git} = $obj->git;
- # for PublicInbox::HTTP::weaken_task:
- $ctx->{cgi}->{env}->{'pi-httpd.inbox'} = $obj;
- $ctx->{-inbox} = $obj;
- $ctx->{inbox} = $inbox;
+sub invalid_inbox ($$) {
+ my ($ctx, $inbox) = @_;
+ my $ibx = $ctx->{www}->{pi_config}->lookup_name($inbox);
+ if (defined $ibx) {
+ $ctx->{-inbox} = $ibx;
return;
}
@@ -145,24 +196,25 @@ sub invalid_inbox {
# generation and link things intended for nntp:// to https?://,
# so try to infer links and redirect them to the appropriate
# list URL.
- $self->news_www->call($ctx->{cgi}->{env});
+ news_cgit_fallback($ctx);
}
# returns undef if valid, array ref response if invalid
sub invalid_inbox_mid {
- my ($self, $ctx, $inbox, $mid) = @_;
- my $ret = invalid_inbox($self, $ctx, $inbox);
+ my ($ctx, $inbox, $mid_ue) = @_;
+ my $ret = invalid_inbox($ctx, $inbox);
return $ret if $ret;
- $ctx->{mid} = $mid = uri_unescape($mid);
- if ($mid =~ /\A[a-f0-9]{40}\z/) {
- # this is horiffically wasteful for legacy URLs:
- if ($mid = mid2blob($ctx)) {
- require Email::Simple;
- use PublicInbox::MID qw/mid_clean/;
- my $s = Email::Simple->new($mid);
- $ctx->{mid} = mid_clean($s->header('Message-ID'));
- }
+ my $mid = $ctx->{mid} = uri_unescape($mid_ue);
+ my $ibx = $ctx->{-inbox};
+ if ($mid =~ m!\A([a-f0-9]{2})([a-f0-9]{38})\z!) {
+ my ($x2, $x38) = ($1, $2);
+ # this is horrifically wasteful for legacy URLs:
+ my $str = $ctx->{-inbox}->msg_by_path("$x2/$x38") or return;
+ require Email::Simple;
+ my $s = Email::Simple->new($str);
+ $mid = PublicInbox::MID::mid_clean($s->header('Message-ID'));
+ return r301($ctx, $inbox, mid_escape($mid));
}
undef;
}
@@ -174,13 +226,19 @@ sub get_atom {
PublicInbox::Feed::generate($ctx);
}
+# /$INBOX/new.html -> HTML only
+sub get_new {
+ my ($ctx) = @_;
+ require PublicInbox::Feed;
+ PublicInbox::Feed::new_html($ctx);
+}
+
# /$INBOX/?r=$GIT_COMMIT -> HTML only
sub get_index {
my ($ctx) = @_;
require PublicInbox::Feed;
- my $srch = searcher($ctx);
- footer($ctx);
- if (defined $ctx->{cgi}->param('q')) {
+ searcher($ctx);
+ if ($ctx->{env}->{QUERY_STRING} =~ /(?:\A|[&;])q=/) {
require PublicInbox::SearchView;
PublicInbox::SearchView::sres_top_html($ctx);
} else {
@@ -188,57 +246,52 @@ sub get_index {
}
}
-# just returns a string ref for the blob in the current ctx
-sub mid2blob {
- my ($ctx) = @_;
- require PublicInbox::MID;
- my $path = PublicInbox::MID::mid2path($ctx->{mid});
- $ctx->{git}->cat_file("HEAD:$path");
-}
-
# /$INBOX/$MESSAGE_ID/raw -> raw mbox
sub get_mid_txt {
my ($ctx) = @_;
- my $x = mid2blob($ctx) or return r404($ctx);
require PublicInbox::Mbox;
- PublicInbox::Mbox::emit1($ctx, $x);
+ PublicInbox::Mbox::emit_raw($ctx) || r404($ctx);
}
# /$INBOX/$MESSAGE_ID/ -> HTML content (short quotes)
sub get_mid_html {
my ($ctx) = @_;
- my $x = mid2blob($ctx) or return r404($ctx);
-
require PublicInbox::View;
- my $foot = footer($ctx);
- require Email::MIME;
- my $mime = Email::MIME->new($x);
searcher($ctx);
- [ 200, [ 'Content-Type' => 'text/html; charset=UTF-8' ],
- [ PublicInbox::View::msg_html($ctx, $mime, $foot) ] ];
-}
-
-# /$INBOX/$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)] ];
+ PublicInbox::View::msg_page($ctx) || r404($ctx);
}
# /$INBOX/$MESSAGE_ID/t/
sub get_thread {
my ($ctx, $flat) = @_;
- my $srch = searcher($ctx) or return need_search($ctx);
- require PublicInbox::View;
- my $foot = footer($ctx);
+ searcher($ctx) or return need_search($ctx);
$ctx->{flat} = $flat;
- PublicInbox::View::thread_html($ctx, $foot, $srch);
+ require PublicInbox::View;
+ PublicInbox::View::thread_html($ctx);
+}
+
+# /$INBOX/_/text/$KEY/
+# /$INBOX/_/text/$KEY/raw
+# KEY may contain slashes
+sub get_text {
+ my ($ctx, $inbox, $key) = @_;
+ my $r404 = invalid_inbox($ctx, $inbox);
+ return $r404 if $r404;
+
+ require PublicInbox::WwwText;
+ PublicInbox::WwwText::get_text($ctx, $key);
+}
+
+# show git objects (blobs and commits)
+# /$INBOX/_/$OBJECT_ID/show
+# /$INBOX/_/${OBJECT_ID}_${FILENAME}
+# KEY may contain slashes
+sub get_vcs_object ($$$;$) {
+ my ($ctx, $inbox, $oid, $filename) = @_;
+ my $r404 = invalid_inbox($ctx, $inbox);
+ return $r404 if $r404;
+ require PublicInbox::ViewVCS;
+ PublicInbox::ViewVCS::show($ctx, $oid, $filename);
}
sub ctx_get {
@@ -248,42 +301,6 @@ sub ctx_get {
$val;
}
-sub footer {
- my ($ctx) = @_;
- return '' unless $ctx;
- my $obj = $ctx->{-inbox} or return '';
-
- # auto-generate a footer
- chomp(my $desc = $obj->description);
-
- my $urls;
- my @urls = @{$obj->cloneurl};
- my %seen = map { $_ => 1 } @urls;
- my $cgi = $ctx->{cgi};
- my $http = $cgi->base->as_string . $obj->{name};
- $seen{$http} or unshift @urls, $http;
- my $ssoma_url = PublicInbox::Hval::prurl($cgi->{env}, SSOMA_URL);
- if (scalar(@urls) == 1) {
- $urls = "URL for ssoma or git clone --mirror $urls[0]);
- } else {
- $urls = "URLs for ssoma or git clone --mirror\n) .
- join("\n", map { "\tgit clone --mirror $_" } @urls);
- }
-
- my $addr = $obj->{-primary_address};
- $ctx->{footer} = join("\n",
- '- ' . $desc,
- "A {cgi}->{env}, PI_URL) .
- '">public-inbox, ' .
- 'anybody may post in plain-text (not HTML):',
- $addr,
- $urls
- );
-}
-
# search support is optional, returns undef if Xapian is not installed
# or not configured for the given GIT_DIR
sub searcher {
@@ -321,13 +338,12 @@ sub get_thread_mbox {
sub get_thread_atom {
my ($ctx) = @_;
searcher($ctx) or return need_search($ctx);
- $ctx->{self_url} = $ctx->{cgi}->uri->as_string;
require PublicInbox::Feed;
PublicInbox::Feed::generate_thread_atom($ctx);
}
sub legacy_redirects {
- my ($self, $ctx, $path_info) = @_;
+ my ($ctx, $path_info) = @_;
# single-message pages
if ($path_info =~ m!$INBOX_RE/m/(\S+)/\z!o) {
@@ -372,7 +388,7 @@ sub legacy_redirects {
# some Message-IDs have slashes in them and the HTTP server
# may try to be clever and unescape them :<
} elsif ($path_info =~ m!$INBOX_RE/(\S+/\S+)/$END_RE\z!o) {
- msg_page($self, $ctx, $1, $2, $3);
+ msg_page($ctx, $1, $2, $3);
# in case people leave off the trailing slash:
} elsif ($path_info =~ m!$INBOX_RE/(\S+/\S+)/(T|t)\z!o) {
@@ -380,22 +396,25 @@ sub legacy_redirects {
} elsif ($path_info =~ m!$INBOX_RE/(\S+/\S+)/f\z!o) {
r301($ctx, $1, $2);
} else {
- $self->news_www->call($ctx->{cgi}->{env});
+ news_cgit_fallback($ctx);
}
}
sub r301 {
- my ($ctx, $inbox, $mid, $suffix) = @_;
- my $cgi = $ctx->{cgi};
- my $obj = $ctx->{-inbox};
- unless ($obj) {
- my $r404 = invalid_inbox($ctx->{www}, $ctx, $inbox);
+ my ($ctx, $inbox, $mid_ue, $suffix) = @_;
+ my $ibx = $ctx->{-inbox};
+ unless ($ibx) {
+ my $r404 = invalid_inbox($ctx, $inbox);
return $r404 if $r404;
- $obj = $ctx->{-inbox};
+ $ibx = $ctx->{-inbox};
+ }
+ my $url = $ibx->base_url($ctx->{env});
+ my $qs = $ctx->{env}->{QUERY_STRING};
+ if (defined $mid_ue) {
+ # common, and much nicer as '@' than '%40':
+ $mid_ue =~ s/%40/@/g;
+ $url .= $mid_ue . '/';
}
- my $url = $obj->base_url($cgi);
- my $qs = $cgi->env->{QUERY_STRING};
- $url .= (uri_escape_utf8($mid) . '/') if (defined $mid);
$url .= $suffix if (defined $suffix);
$url .= "?$qs" if $qs ne '';
@@ -405,32 +424,43 @@ sub r301 {
}
sub msg_page {
- my ($self, $ctx, $inbox, $mid, $e) = @_;
+ my ($ctx, $inbox, $mid_ue, $e) = @_;
my $ret;
- $ret = invalid_inbox_mid($self, $ctx, $inbox, $mid) and return $ret;
+ $ret = invalid_inbox_mid($ctx, $inbox, $mid_ue) and return $ret;
'' eq $e and return get_mid_html($ctx);
+ 'T/' eq $e and return get_thread($ctx, 1);
'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);
# legacy, but no redirect for compatibility:
'f/' eq $e and return get_mid_html($ctx);
-
- 'R/' eq $e and return get_reply_html($ctx);
r404($ctx);
}
sub serve_git {
- my ($env, $git, $path) = @_;
- PublicInbox::GitHTTPBackend::serve($env, $git, $path);
+ my ($ctx, $part, $path) = @_;
+ my $env = $ctx->{env};
+ my $ibx = $ctx->{-inbox};
+ my $git = defined $part ? $ibx->git_part($part) : $ibx->git;
+ $git ? PublicInbox::GitHTTPBackend::serve($env, $git, $path) : r404();
+}
+
+sub mbox_results {
+ my ($ctx) = @_;
+ if ($ctx->{env}->{QUERY_STRING} =~ /(?:\A|[&;])q=/) {
+ searcher($ctx) or return need_search($ctx);
+ require PublicInbox::SearchView;
+ return PublicInbox::SearchView::mbox_results($ctx);
+ }
+ r404();
}
sub serve_mbox_range {
- my ($self, $ctx, $inbox, $range) = @_;
- invalid_inbox($self, $ctx, $inbox) || eval {
+ my ($ctx, $inbox, $range) = @_;
+ invalid_inbox($ctx, $inbox) || eval {
require PublicInbox::Mbox;
searcher($ctx);
PublicInbox::Mbox::emit_range($ctx, $range);
@@ -439,10 +469,24 @@ sub serve_mbox_range {
sub news_www {
my ($self) = @_;
- my $nw = $self->{news_www};
- return $nw if $nw;
- require PublicInbox::NewsWWW;
- $self->{news_www} = PublicInbox::NewsWWW->new($self->{pi_config});
+ $self->{news_www} ||= do {
+ require PublicInbox::NewsWWW;
+ PublicInbox::NewsWWW->new($self->{pi_config});
+ }
+}
+
+sub cgit {
+ my ($self) = @_;
+ $self->{cgit} ||= do {
+ my $pi_config = $self->{pi_config};
+
+ if (defined($pi_config->{'publicinbox.cgitrc'})) {
+ require PublicInbox::Cgit;
+ PublicInbox::Cgit->new($pi_config);
+ } else {
+ Plack::Util::inline_object(call => sub { r404() });
+ }
+ }
}
sub get_attach {
@@ -451,4 +495,136 @@ sub get_attach {
PublicInbox::WwwAttach::get_attach($ctx, $idx, $fn);
}
+# User-generated content (UGC) may have excessively long lines
+# and screw up rendering on some browsers, so we use pre-wrap.
+#
+# We also force everything to the same scaled font-size because GUI
+# browsers (tested both Firefox and surf (webkit)) uses a larger font
+# for the Search