X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FWWW.pm;h=4ddc187b7b46e9e6b928a200d1fb1c7109a1d2ce;hb=3d41aa23f35501ca92aab8aa42980fa73f7fa74f;hp=b602206d1cfa3a683667060125081e35b94a59d0;hpb=8d479224468ad9c8053ed80cd3532b8d9343e653;p=public-inbox.git

diff --git a/lib/PublicInbox/WWW.pm b/lib/PublicInbox/WWW.pm
index b602206d..4ddc187b 100644
--- a/lib/PublicInbox/WWW.pm
+++ b/lib/PublicInbox/WWW.pm
@@ -1,5 +1,5 @@
-# Copyright (C) 2014-2015 all contributors <meta@public-inbox.org>
-# License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt)
+# Copyright (C) 2014-2018 all contributors <meta@public-inbox.org>
+# License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
 #
 # Main web interface for mailing list archives
 #
@@ -15,7 +15,8 @@ use strict;
 use warnings;
 use PublicInbox::Config;
 use PublicInbox::Hval;
-use URI::Escape qw(uri_escape_utf8 uri_unescape);
+use URI::Escape qw(uri_unescape);
+use PublicInbox::MID qw(mid_escape);
 require PublicInbox::Git;
 use PublicInbox::GitHTTPBackend;
 our $INBOX_RE = qr!\A/([\w\.\-]+)!;
@@ -37,25 +38,29 @@ sub run {
 
 sub call {
 	my ($self, $env) = @_;
-	my $ctx = { env => $env, www => $self, pi_config => $self->{pi_config} };
+	my $ctx = { env => $env, www => $self };
 
 	# we don't care about multi-value
 	my %qp = map {
-		my ($k, $v) = split('=', $_, 2);
+		utf8::decode($_);
+		my ($k, $v) = split('=', uri_unescape($_), 2);
 		$v = '' unless defined $v;
 		$v =~ tr/+/ /;
 		($k, $v)
-	} split(/[&;]/, uri_unescape($env->{QUERY_STRING}));
+	} split(/[&;]+/, $env->{QUERY_STRING});
 	$ctx->{qp} = \%qp;
 
 	my $path_info = $env->{PATH_INFO};
 	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/(git-upload-pack)\z!) {
+			my $path = $2;
+			return invalid_inbox($ctx, $1) ||
+				serve_git($ctx, $path);
+		} elsif ($path_info =~ m!$INBOX_RE/!o) {
+			return invalid_inbox($ctx, $1) || mbox_results($ctx);
+		}
 	}
 	elsif ($method !~ /\AGET|HEAD\z/) {
 		return r(405, 'Method Not Allowed');
@@ -65,27 +70,25 @@ sub call {
 	if ($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);
+		invalid_inbox($ctx, $1) || get_atom($ctx);
 	} elsif ($path_info =~ m!$INBOX_RE/new\.html\z!o) {
-		invalid_inbox($self, $ctx, $1) || get_new($ctx);
+		invalid_inbox($ctx, $1) || get_new($ctx);
 	} elsif ($path_info =~ m!$INBOX_RE/
 				($PublicInbox::GitHTTPBackend::ANY)\z!ox) {
 		my $path = $2;
-		invalid_inbox($self, $ctx, $1) ||
-			serve_git($env, $ctx->{git}, $path);
+		invalid_inbox($ctx, $1) || serve_git($ctx, $path);
 	} elsif ($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)\z!o) {
 		my ($inbox, $mid, $suffix) = ($1, $2, $3);
@@ -98,13 +101,15 @@ sub call {
 
 	} 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);
 
 	# 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);
 	}
 }
 
@@ -112,8 +117,8 @@ sub call {
 sub preload {
 	require PublicInbox::Feed;
 	require PublicInbox::View;
-	require PublicInbox::Thread;
-	require Email::MIME;
+	require PublicInbox::SearchThread;
+	require PublicInbox::MIME;
 	require Digest::SHA;
 	require POSIX;
 
@@ -140,9 +145,10 @@ sub r404 {
 sub r { [ $_[0], ['Content-Type' => 'text/plain'], [ join(' ', @_, "\n") ] ] }
 
 # returns undef if valid, array ref response if invalid
-sub invalid_inbox {
-	my ($self, $ctx, $inbox) = @_;
-	my $obj = $ctx->{pi_config}->lookup_name($inbox);
+sub invalid_inbox ($$) {
+	my ($ctx, $inbox) = @_;
+	my $www = $ctx->{www};
+	my $obj = $www->{pi_config}->lookup_name($inbox);
 	if (defined $obj) {
 		$ctx->{git_dir} = $obj->{mainrepo};
 		$ctx->{git} = $obj->git;
@@ -155,16 +161,16 @@ 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->{env});
+	$www->news_www->call($ctx->{env});
 }
 
 # 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) = @_;
+	my $ret = invalid_inbox($ctx, $inbox);
 	return $ret if $ret;
 
-	$ctx->{mid} = $mid = uri_unescape($mid);
+	$ctx->{mid} = $mid;
 	if ($mid =~ /\A[a-f0-9]{40}\z/) {
 		# this is horiffically wasteful for legacy URLs:
 		if ($mid = mid2blob($ctx)) {
@@ -195,7 +201,7 @@ sub get_new {
 sub get_index {
 	my ($ctx) = @_;
 	require PublicInbox::Feed;
-	my $srch = searcher($ctx);
+	searcher($ctx);
 	if ($ctx->{env}->{QUERY_STRING} =~ /(?:\A|[&;])q=/) {
 		require PublicInbox::SearchView;
 		PublicInbox::SearchView::sres_top_html($ctx);
@@ -224,8 +230,8 @@ sub get_mid_html {
 	my $x = mid2blob($ctx) or return r404($ctx);
 
 	require PublicInbox::View;
-	require Email::MIME;
-	my $mime = Email::MIME->new($x);
+	require PublicInbox::MIME;
+	my $mime = PublicInbox::MIME->new($x);
 	searcher($ctx);
 	PublicInbox::View::msg_html($ctx, $mime);
 }
@@ -239,6 +245,18 @@ sub get_thread {
 	PublicInbox::View::thread_html($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, $key);
+}
+
 sub ctx_get {
 	my ($ctx, $key) = @_;
 	my $val = $ctx->{$key};
@@ -288,7 +306,7 @@ sub get_thread_atom {
 }
 
 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) {
@@ -333,7 +351,7 @@ 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) {
@@ -341,7 +359,7 @@ sub legacy_redirects {
 	} elsif ($path_info =~ m!$INBOX_RE/(\S+/\S+)/f\z!o) {
 		r301($ctx, $1, $2);
 	} else {
-		$self->news_www->call($ctx->{env});
+		$ctx->{www}->news_www->call($ctx->{env});
 	}
 }
 
@@ -349,13 +367,13 @@ sub r301 {
 	my ($ctx, $inbox, $mid, $suffix) = @_;
 	my $obj = $ctx->{-inbox};
 	unless ($obj) {
-		my $r404 = invalid_inbox($ctx->{www}, $ctx, $inbox);
+		my $r404 = invalid_inbox($ctx, $inbox);
 		return $r404 if $r404;
 		$obj = $ctx->{-inbox};
 	}
 	my $url = $obj->base_url($ctx->{env});
 	my $qs = $ctx->{env}->{QUERY_STRING};
-	$url .= (uri_escape_utf8($mid) . '/') if (defined $mid);
+	$url .= (mid_escape($mid) . '/') if (defined $mid);
 	$url .= $suffix if (defined $suffix);
 	$url .= "?$qs" if $qs ne '';
 
@@ -365,9 +383,9 @@ sub r301 {
 }
 
 sub msg_page {
-	my ($self, $ctx, $inbox, $mid, $e) = @_;
+	my ($ctx, $inbox, $mid, $e) = @_;
 	my $ret;
-	$ret = invalid_inbox_mid($self, $ctx, $inbox, $mid) and return $ret;
+	$ret = invalid_inbox_mid($ctx, $inbox, $mid) 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);
@@ -382,13 +400,23 @@ sub msg_page {
 }
 
 sub serve_git {
-	my ($env, $git, $path) = @_;
-	PublicInbox::GitHTTPBackend::serve($env, $git, $path);
+	my ($ctx, $path) = @_;
+	PublicInbox::GitHTTPBackend::serve($ctx->{env}, $ctx->{git}, $path);
+}
+
+sub mbox_results {
+	my ($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);
@@ -397,10 +425,10 @@ 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 get_attach {