[RFC 10/11] implement /s/$SUBJECT_PATH.html lookups

[Eric Wong <e@80x24.org>]

Quick-and-dirty wiring up of to Subject: paths.
This may prove more memorizable and easier-to-share than
/t/$MESSAGE_ID.html links, but less strict.

This changes our schema version to 1, since we now
use lower-case subject paths.
---
 lib/PublicInbox/Search.pm | 15 ++++++++--
 lib/PublicInbox/View.pm   | 72 ++++++++++++++++++++++++++++++++++++++++-------
 lib/PublicInbox/WWW.pm    | 20 ++++++++++++-
 t/search.t                | 16 +++++------
 4 files changed, 101 insertions(+), 22 deletions(-)

diff --git a/lib/PublicInbox/Search.pm b/lib/PublicInbox/Search.pm
index 39b06b0..f4f00b2 100644
--- a/lib/PublicInbox/Search.pm
+++ b/lib/PublicInbox/Search.pm
@@ -12,7 +12,10 @@ use PublicInbox::MID qw/mid_clean mid_compressed/;
 
 use constant {
 	TS => 0,
-	SCHEMA_VERSION => 0,
+	# SCHEMA_VERSION history
+	# 0 - initial
+	# 1 - subject_path is lower-cased
+	SCHEMA_VERSION => 1,
 	LANG => 'english',
 	QP_FLAGS => FLAG_PHRASE|FLAG_BOOLEAN|FLAG_LOVEHATE|FLAG_WILDCARD,
 };
@@ -209,6 +212,12 @@ sub query {
 	$self->do_enquire($query, $opts);
 }
 
+sub get_subject_path {
+	my ($self, $path, $opts) = @_;
+	my $query = $self->qp->parse_query("path:$path", 0);
+	$self->do_enquire($query);
+}
+
 # given a message ID, get replies to a message
 sub get_replies {
 	my ($self, $mid, $opts) = @_;
@@ -461,13 +470,13 @@ sub merge_threads {
 
 # normalize subjects so they are suitable as pathnames for URLs
 sub subject_path {
-	my ($subj) = @_;
+	my $subj = pop;
 
 	$subj =~ s/\A\s+//;
 	$subj =~ s/\s+\z//;
 	$subj =~ s/^(?:re|aw):\s*//i; # remove reply prefix (aw: German)
 	$subj =~ s![^a-zA-Z0-9_\.~/\-]+!_!g;
-	$subj;
+	lc($subj);
 }
 
 sub do_cat_mail {
diff --git a/lib/PublicInbox/View.pm b/lib/PublicInbox/View.pm
index c40a2a7..696d7d5 100644
--- a/lib/PublicInbox/View.pm
+++ b/lib/PublicInbox/View.pm
@@ -31,7 +31,7 @@ sub msg_html {
 	} else {
 		$footer = '';
 	}
-	headers_to_html_header($mime, $full_pfx) .
+	headers_to_html_header($mime, $full_pfx, $srch) .
 		multipart_text_as_html($mime, $full_pfx) .
 		'</pre><hr />' . PRE_WRAP .
 		html_footer($mime, 1, $full_pfx, $srch) . $footer .
@@ -179,6 +179,52 @@ sub thread_html {
 	$rv .= "</pre><hr />" . PRE_WRAP . $next . $foot . "</pre>";
 }
 
+sub subject_path_html {
+	my (undef, $ctx, $foot, $srch) = @_;
+	my $path = $ctx->{subject_path};
+	my $res = $srch->get_subject_path($path);
+	my $rv = '';
+	require PublicInbox::GitCatFile;
+	my $git = PublicInbox::GitCatFile->new($ctx->{git_dir});
+	my $nr = scalar @{$res->{msgs}};
+	return $rv if $nr == 0;
+	my @msgs;
+	while (my $smsg = shift @{$res->{msgs}}) {
+		my $m = $smsg->mid;
+
+		# Duplicated from WWW.pm
+		my ($x2, $x38) = ($m =~ /\A([a-f0-9]{2})([a-f0-9]{38})\z/);
+
+		unless (defined $x38) {
+			require Digest::SHA;
+			$m = Digest::SHA::sha1_hex($m);
+			($x2, $x38) = ($m =~ /\A([a-f0-9]{2})([a-f0-9]{38})\z/);
+		}
+
+		# FIXME: duplicated code from Feed.pm
+		my $mime = eval {
+			my $str = $git->cat_file("HEAD:$x2/$x38");
+			Email::MIME->new($str);
+		};
+		unless ($@) {
+			my $t = eval { str2time($mime->header('Date')) };
+			defined($t) or $t = 0;
+			$mime->header_set('X-PI-TS', $t);
+			push @msgs, $mime;
+		}
+	}
+	require PublicInbox::Thread;
+	my $th = PublicInbox::Thread->new(@msgs);
+	$th->thread;
+	$th->order(*PublicInbox::Thread::sort_ts);
+	my $state = [ undef, { root_anchor => 'dummy' }, undef, 0 ];
+	thread_entry(\$rv, $state, $_, 0) for $th->rootset;
+	my $final_anchor = $state->[3];
+	my $next = "<a\nid=\"s$final_anchor\">end of thread</a>\n";
+
+	$rv .= "</pre><hr />" . PRE_WRAP . $next . $foot . "</pre>";
+}
+
 # only private functions below.
 
 sub index_walk {
@@ -235,7 +281,7 @@ sub enc_for {
 }
 
 sub multipart_text_as_html {
-	my ($mime, $full_pfx) = @_;
+	my ($mime, $full_pfx, $srch) = @_;
 	my $rv = "";
 	my $part_nr = 0;
 	my $enc_msg = enc_for($mime->header("Content-Type"));
@@ -339,7 +385,7 @@ sub add_text_body_full {
 }
 
 sub headers_to_html_header {
-	my ($mime, $full_pfx) = @_;
+	my ($mime, $full_pfx, $srch) = @_;
 
 	my $rv = "";
 	my @title;
@@ -347,18 +393,21 @@ sub headers_to_html_header {
 		my $v = $mime->header($h);
 		defined($v) && length($v) or next;
 		$v = PublicInbox::Hval->new_oneline($v);
-		$rv .= "$h: " . $v->as_html . "\n";
 
 		if ($h eq 'From') {
 			my @from = Email::Address->parse($v->raw);
-			$v = $from[0]->name;
-			unless (defined($v) && length($v)) {
-				$v = '<' . $from[0]->address . '>';
-			}
-			$title[1] = ascii_html($v);
+			$title[1] = ascii_html($from[0]->name);
 		} elsif ($h eq 'Subject') {
 			$title[0] = $v->as_html;
+			if ($srch) {
+				my $path = $srch->subject_path($v->raw);
+				$rv .= "$h: <a\nhref=\"../s/$path.html\">";
+				$rv .= $v->as_html . "</a>\n";
+				next;
+			}
 		}
+		$rv .= "$h: " . $v->as_html . "\n";
+
 	}
 
 	my $header_obj = $mime->header_obj;
@@ -510,6 +559,9 @@ sub hash_subj {
 sub thread_replies {
 	my ($dst, $root, $res) = @_;
 	my @msgs = map { $_->mini_mime } @{$res->{msgs}};
+	foreach (@{$res->{msgs}}) {
+		print STDERR "smsg->path: <", $_->path, ">\n";
+	}
 	require PublicInbox::Thread;
 	$root->header_set('X-PI-TS', '0');
 	my $th = PublicInbox::Thread->new($root, @msgs);
@@ -532,7 +584,7 @@ sub thread_entry {
 	my ($dst, $state, $node, $level) = @_;
 	# $state = [ $search_res, $seen, undef, 0 (msg_nr) ];
 	# $seen is overloaded with 3 types of fields:
-	#	1) "root" => Message-ID,
+	#	1) "root_anchor" => anchor_for(Message-ID),
 	#	2) seen subject hashes: sha1(subject) => 1
 	#	3) anchors hashes: "#$sha1_hex" (same as $seen in index_entry)
 	if (my $mime = $node->message) {
diff --git a/lib/PublicInbox/WWW.pm b/lib/PublicInbox/WWW.pm
index 52e51c4..7fe9b85 100644
--- a/lib/PublicInbox/WWW.pm
+++ b/lib/PublicInbox/WWW.pm
@@ -53,8 +53,13 @@ sub run {
 	} elsif ($path_info =~ m!$LISTNAME_RE/t/(\S+)\.html\z!o) {
 		invalid_list_mid(\%ctx, $1, $2) || get_thread(\%ctx, $cgi);
 
+	# subject_path display
+	} elsif ($path_info =~ m!$LISTNAME_RE/s/(\S+)\.html\z!o) {
+		my $sp = $2;
+		invalid_list(\%ctx, $1) || get_subject_path(\%ctx, $cgi, $sp);
+
 	# convenience redirects, order matters
-	} elsif ($path_info =~ m!$LISTNAME_RE/(m|f|t)/(\S+)\z!o) {
+	} elsif ($path_info =~ m!$LISTNAME_RE/(m|f|t|s)/(\S+)\z!o) {
 		my $pfx = $2;
 		invalid_list_mid(\%ctx, $1, $3) ||
 			redirect_mid(\%ctx, $cgi, $2);
@@ -211,6 +216,19 @@ sub get_thread {
 	  [ $body ] ];
 }
 
+# /$LISTNAME/s/$SUBJECT_PATH.html
+sub get_subject_path {
+	my ($ctx, $cgi, $sp) = @_;
+	$ctx->{subject_path} = $sp;
+	my $srch = searcher($ctx) or return need_search($ctx);
+	require PublicInbox::View;
+	my $foot = footer($ctx);
+	my $body = PublicInbox::View->subject_path_html($ctx, $foot, $srch) or
+		return r404();
+	[ 200, [ 'Content-Type' => 'text/html; charset=UTF-8' ],
+	  [ $body ] ];
+}
+
 sub self_url {
 	my ($cgi) = @_;
 	ref($cgi) eq 'CGI' ? $cgi->self_url : $cgi->uri->as_string;
diff --git a/t/search.t b/t/search.t
index 9de6d28..9bdd3ce 100644
--- a/t/search.t
+++ b/t/search.t
@@ -22,7 +22,7 @@ my $ro = PublicInbox::Search->new($git_dir);
 	my $root = Email::MIME->create(
 		header_str => [
 			Date => 'Fri, 02 Oct 1993 00:00:00 +0000',
-			Subject => 'hello world',
+			Subject => 'Hello world',
 			'Message-ID' => '<root@s>',
 			From => 'John Smith <js@example.com>',
 			To => 'list@example.com',
@@ -31,7 +31,7 @@ my $ro = PublicInbox::Search->new($git_dir);
 	my $last = Email::MIME->create(
 		header_str => [
 			Date => 'Sat, 02 Oct 2010 00:00:00 +0000',
-			Subject => 'Re: hello world',
+			Subject => 'Re: Hello world',
 			'In-Reply-To' => '<root@s>',
 			'Message-ID' => '<last@s>',
 			From => 'John Smith <js@example.com>',
@@ -70,19 +70,19 @@ sub filter_mids {
 		is($res->{count}, 0, "path variant `$p' does not match");
 	}
 
-	$res = $ro->query('subject:(hello world)');
+	$res = $ro->query('subject:(Hello world)');
 	@res = filter_mids($res);
 	is_deeply(\@res, \@exp, 'got expected results for subject:() match');
 
-	$res = $ro->query('subject:"hello world"');
+	$res = $ro->query('subject:"Hello world"');
 	@res = filter_mids($res);
 	is_deeply(\@res, \@exp, 'got expected results for subject:"" match');
 
-	$res = $ro->query('subject:"hello world"', {limit => 1});
+	$res = $ro->query('subject:"Hello world"', {limit => 1});
 	is(scalar @{$res->{msgs}}, 1, "limit works");
 	my $first = $res->{msgs}->[0];
 
-	$res = $ro->query('subject:"hello world"', {offset => 1});
+	$res = $ro->query('subject:"Hello world"', {offset => 1});
 	is(scalar @{$res->{msgs}}, 1, "offset works");
 	my $second = $res->{msgs}->[0];
 
@@ -207,7 +207,7 @@ sub filter_mids {
 	$rw->add_message(Email::MIME->create(
 		header_str => [
 			Date => 'Sat, 02 Oct 2010 00:00:01 +0000',
-			Subject => 'hello',
+			Subject => 'Hello',
 			'Message-ID' => '<quote@a>',
 			From => 'Quoter <quoter@example.com>',
 			To => 'list@example.com',
@@ -217,7 +217,7 @@ sub filter_mids {
 	$rw->add_message(Email::MIME->create(
 		header_str => [
 			Date => 'Sat, 02 Oct 2010 00:00:02 +0000',
-			Subject => 'hello',
+			Subject => 'Hello',
 			'Message-ID' => '<nquote@a>',
 			From => 'Non-Quoter<non-quoter@example.com>',
 			To => 'list@example.com',
-- 
EW