X-Git-Url: http://www.git.stargrave.org/?p=public-inbox.git;a=blobdiff_plain;f=lib%2FPublicInbox%2FExtMsg.pm;h=be29974900792f2023b7b7b97b5144b61ee6538c;hp=167dc5e5d37fe0e2185d2eccbc47ff6f94263131;hb=HEAD;hpb=9bd675d33ad1e49bd2ebe12a1d216216e61380de

diff --git a/lib/PublicInbox/ExtMsg.pm b/lib/PublicInbox/ExtMsg.pm
index 167dc5e5..be299749 100644
--- a/lib/PublicInbox/ExtMsg.pm
+++ b/lib/PublicInbox/ExtMsg.pm
@@ -1,4 +1,4 @@
-# Copyright (C) 2015-2019 all contributors <meta@public-inbox.org>
+# Copyright (C) all contributors <meta@public-inbox.org>
 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
 #
 # Used by the web interface to link to messages outside of the our
@@ -8,17 +8,17 @@
 package PublicInbox::ExtMsg;
 use strict;
 use warnings;
-use PublicInbox::Hval qw/ascii_html/;
-use PublicInbox::MID qw/mid2path/;
-use PublicInbox::WwwStream;
-our $MIN_PARTIAL_LEN = 16;
+use PublicInbox::Hval qw(ascii_html prurl mid_href);
+use PublicInbox::WwwStream qw(html_oneshot);
+use PublicInbox::Smsg;
+our $MIN_PARTIAL_LEN = 14; # for 'XXXXXXXXXX.fsf' msgids gnus generates
 
 # TODO: user-configurable
 our @EXT_URL = map { ascii_html($_) } (
 	# leading "//" denotes protocol-relative (http:// or https://)
 	'//marc.info/?i=%s',
 	'//www.mail-archive.com/search?l=mid&q=%s',
-	'nntp://news.gmane.org/%s',
+	'nntp://news.gmane.io/%s',
 	'https://lists.debian.org/msgid-search/%s',
 	'//docs.FreeBSD.org/cgi/mid.cgi?db=mid&id=%s',
 	'https://www.w3.org/mid/%s',
@@ -30,9 +30,10 @@ our @EXT_URL = map { ascii_html($_) } (
 sub PARTIAL_MAX () { 100 }
 
 sub search_partial ($$) {
-	my ($srch, $mid) = @_;
+	my ($ibx, $mid) = @_;
 	return if length($mid) < $MIN_PARTIAL_LEN;
-	my $opt = { limit => PARTIAL_MAX, mset => 2 };
+	my $srch = $ibx->isrch or return;
+	my $opt = { limit => PARTIAL_MAX, relevance => -1 };
 	my @try = ("m:$mid*");
 	my $chop = $mid;
 	if ($chop =~ s/(\W+)(\w*)\z//) {
@@ -64,87 +65,158 @@ sub search_partial ($$) {
 		# has too many results.  $@ can be
 		# Search::Xapian::QueryParserError or even:
 		# "something terrible happened at ../Search/Xapian/Enquire.pm"
-		my $mset = eval { $srch->query($m, $opt) } or next;
-
+		my $mset = eval { $srch->mset($m, $opt) } or next;
 		my @mids = map {
-			my $doc = $_->get_document;
-			PublicInbox::SearchMsg->load_doc($doc)->mid;
-		} $mset->items;
+			$_->{mid}
+		} @{$srch->mset_to_smsg($ibx, $mset)};
 		return \@mids if scalar(@mids);
 	}
 }
 
-sub ext_msg {
-	my ($ctx) = @_;
-	my $cur = $ctx->{-inbox};
-	my $mid = $ctx->{mid};
+sub ext_msg_i {
+	my ($other, $ctx) = @_;
 
-	eval { require PublicInbox::Msgmap };
-	my $have_mm = $@ ? 0 : 1;
-	my (@ibx, @found);
+	return if $other->{name} eq $ctx->{ibx}->{name} || !$other->base_url;
 
-	$ctx->{www}->{pi_config}->each_inbox(sub {
-		my ($other) = @_;
-		return if $other->{name} eq $cur->{name} || !$other->base_url;
+	my $mm = $other->mm or return;
 
-		my $mm = $other->mm or return;
+	# try to find the URL with Msgmap to avoid forking
+	my $num = $mm->num_for($ctx->{mid});
+	if (defined $num) {
+		push @{$ctx->{found}}, $other;
+	} else {
+		# no point in trying the fork fallback if we
+		# know Xapian is up-to-date but missing the
+		# message in the current repo
+		push @{$ctx->{again}}, $other;
+	}
+}
 
-		# try to find the URL with Msgmap to avoid forking
-		my $num = $mm->num_for($mid);
-		if (defined $num) {
-			push @found, $other;
-		} else {
-			# no point in trying the fork fallback if we
-			# know Xapian is up-to-date but missing the
-			# message in the current repo
-			push @ibx, $other;
-		}
-	});
+sub ext_msg_step {
+	my ($pi_cfg, $section, $ctx) = @_;
+	if (defined($section)) {
+		return if $section !~ m!\Apublicinbox\.([^/]+)\z!;
+		my $ibx = $pi_cfg->lookup_name($1) or return;
+		ext_msg_i($ibx, $ctx);
+	} else { # undef == "EOF"
+		finalize_exact($ctx);
+	}
+}
 
-	return exact($ctx, \@found, $mid) if @found;
+sub ext_msg_ALL ($) {
+	my ($ctx) = @_;
+	my $ALL = $ctx->{www}->{pi_cfg}->ALL or return;
+	my $by_eidx_key = $ctx->{www}->{pi_cfg}->{-by_eidx_key};
+	my $cur_key = eval { $ctx->{ibx}->eidx_key } //
+			return partial_response($ctx); # $cur->{ibx} == $ALL
+	my %seen = ($cur_key => 1);
+	my ($id, $prev);
+	while (my $x = $ALL->over->next_by_mid($ctx->{mid}, \$id, \$prev)) {
+		my $xr3 = $ALL->over->get_xref3($x->{num});
+		for my $k (@$xr3) {
+			$k =~ s/:[0-9]+:$x->{blob}\z// or next;
+			next if $k eq $cur_key;
+			my $ibx = $by_eidx_key->{$k} // next;
+			$ibx->base_url or next;
+			push(@{$ctx->{found}}, $ibx) unless $seen{$k}++;
+		}
+	}
+	return exact($ctx) if $ctx->{found};
 
 	# fall back to partial MID matching
-	my @partial;
-	my $n_partial = 0;
-	my $srch = $cur->search;
-	my $mids = search_partial($srch, $mid) if $srch;
-	if ($mids) {
-		$n_partial = scalar(@$mids);
-		push @partial, [ $cur, $mids ];
+	for my $ibxish ($ctx->{ibx}, $ALL) {
+		my $mids = search_partial($ibxish, $ctx->{mid}) or next;
+		push @{$ctx->{partial}}, [ $ibxish, $mids ];
+		last if ($ctx->{n_partial} += scalar(@$mids)) >= PARTIAL_MAX;
 	}
+	partial_response($ctx);
+}
 
+sub ext_msg {
+	my ($ctx) = @_;
+	ext_msg_ALL($ctx) // sub {
+		$ctx->{-wcb} = $_[0]; # HTTP server write callback
+
+		if ($ctx->{env}->{'pi-httpd.async'}) {
+			require PublicInbox::ConfigIter;
+			my $iter = PublicInbox::ConfigIter->new(
+						$ctx->{www}->{pi_cfg},
+						\&ext_msg_step, $ctx);
+			$iter->event_step;
+		} else {
+			$ctx->{www}->{pi_cfg}->each_inbox(\&ext_msg_i, $ctx);
+			finalize_exact($ctx);
+		}
+	};
+}
+
+# called via PublicInbox::DS::event_loop
+sub event_step {
+	my ($ctx, $sync) = @_;
 	# can't find a partial match in current inbox, try the others:
-	if (!$n_partial && length($mid) >= $MIN_PARTIAL_LEN) {
-		foreach my $ibx (@ibx) {
-			$srch = $ibx->search or next;
-			$mids = search_partial($srch, $mid) or next;
-			$n_partial += scalar(@$mids);
-			push @partial, [ $ibx, $mids];
-			last if $n_partial >= PARTIAL_MAX;
+	my $ibx = shift @{$ctx->{again}} or return finalize_partial($ctx);
+	my $mids = search_partial($ibx, $ctx->{mid}) or
+			return ($sync ? undef : PublicInbox::DS::requeue($ctx));
+	$ctx->{n_partial} += scalar(@$mids);
+	push @{$ctx->{partial}}, [ $ibx, $mids ];
+	$ctx->{n_partial} >= PARTIAL_MAX ? finalize_partial($ctx)
+			: ($sync ? undef : PublicInbox::DS::requeue($ctx));
+}
+
+sub finalize_exact {
+	my ($ctx) = @_;
+
+	return $ctx->{-wcb}->(exact($ctx)) if $ctx->{found};
+
+	# fall back to partial MID matching
+	my $mid = $ctx->{mid};
+	my $cur = $ctx->{ibx};
+	my $mids = search_partial($cur, $mid);
+	if ($mids) {
+		$ctx->{n_partial} = scalar(@$mids);
+		push @{$ctx->{partial}}, [ $cur, $mids ];
+	} elsif ($ctx->{again} && length($mid) >= $MIN_PARTIAL_LEN) {
+		bless $ctx, __PACKAGE__;
+		if ($ctx->{env}->{'pi-httpd.async'}) {
+			$ctx->event_step;
+			return;
 		}
+
+		# synchronous fall-through
+		$ctx->event_step while @{$ctx->{again}};
 	}
+	finalize_partial($ctx);
+}
 
+sub _url_pfx ($$) {
+	my ($ctx, $u) = @_;
+	(index($u, '://') < 0 && index($u, '/') != 0) ?
+		"$ctx->{-upfx}../$u" : $u;
+}
+
+sub partial_response ($) {
+	my ($ctx) = @_;
+	my $mid = $ctx->{mid};
 	my $code = 404;
-	my $h = PublicInbox::Hval->new_msgid($mid);
-	my $href = $h->{href};
-	my $html = $h->as_html;
+	my $href = mid_href($mid);
+	my $html = ascii_html($mid);
 	my $title = "&lt;$html&gt; not found";
 	my $s = "<pre>Message-ID &lt;$html&gt;\nnot found\n";
-	if ($n_partial) {
+	$ctx->{-upfx} //= '../';
+	if (my $n_partial = $ctx->{n_partial}) {
 		$code = 300;
 		my $es = $n_partial == 1 ? '' : 'es';
 		$n_partial .= '+' if ($n_partial == PARTIAL_MAX);
 		$s .= "\n$n_partial partial match$es found:\n\n";
-		my $cur_name = $cur->{name};
-		foreach my $pair (@partial) {
+		my $cur_name = $ctx->{ibx}->{name};
+		foreach my $pair (@{$ctx->{partial}}) {
 			my ($ibx, $res) = @$pair;
-			my $env = $ctx->{env} if $ibx->{name} eq $cur_name;
-			my $u = $ibx->base_url($env) or next;
+			my $e = $ibx->{name} eq $cur_name ? $ctx->{env} : undef;
+			my $u = _url_pfx($ctx, $ibx->base_url($e) // next);
 			foreach my $m (@$res) {
-				my $p = PublicInbox::Hval->new_msgid($m);
-				my $r = $p->{href};
-				my $t = $p->as_html;
-				$s .= qq{<a\nhref="$u$r/">$u$t/</a>\n};
+				my $href = mid_href($m);
+				my $html = ascii_html($m);
+				$s .= qq{<a\nhref="$u$href/">$u$html/</a>\n};
 			}
 		}
 	}
@@ -155,11 +227,13 @@ sub ext_msg {
 	}
 	$ctx->{-html_tip} = $s .= '</pre>';
 	$ctx->{-title_html} = $title;
-	$ctx->{-upfx} = '../';
-	PublicInbox::WwwStream->response($ctx, $code);
+	html_oneshot($ctx, $code);
 }
 
+sub finalize_partial ($) { $_[0]->{-wcb}->(partial_response($_[0])) }
+
 sub ext_urls {
+	return "";
 	my ($ctx, $mid, $href, $html) = @_;
 
 	# Fall back to external repos if configured
@@ -167,7 +241,7 @@ sub ext_urls {
 		my $env = $ctx->{env};
 		my $e = "\nPerhaps try an external site:\n\n";
 		foreach my $url (@EXT_URL) {
-			my $u = PublicInbox::Hval::prurl($env, $url);
+			my $u = prurl($env, $url);
 			my $r = sprintf($u, $href);
 			my $t = sprintf($u, $html);
 			$e .= qq{<a\nhref="$r">$t</a>\n};
@@ -178,24 +252,25 @@ sub ext_urls {
 }
 
 sub exact {
-	my ($ctx, $found, $mid) = @_;
-	my $h = PublicInbox::Hval->new_msgid($mid);
-	my $href = $h->{href};
-	my $html = $h->as_html;
+	my ($ctx) = @_;
+	my $mid = $ctx->{mid};
+	my $found = $ctx->{found};
+	my $href = mid_href($mid);
+	my $html = ascii_html($mid);
 	my $title = "&lt;$html&gt; found in ";
 	my $end = @$found == 1 ? 'another inbox' : 'other inboxes';
 	$ctx->{-title_html} = $title . $end;
-	$ctx->{-upfx} = '../';
+	$ctx->{-upfx} //= '../';
 	my $ext_urls = ext_urls($ctx, $mid, $href, $html);
 	my $code = (@$found == 1 && $ext_urls eq '') ? 200 : 300;
 	$ctx->{-html_tip} = join('',
 			"<pre>Message-ID: &lt;$html&gt;\nfound in $end:\n\n",
 				(map {
-					my $u = $_->base_url;
+					my $u = _url_pfx($ctx, $_->base_url);
 					qq(<a\nhref="$u$href/">$u$html/</a>\n)
 				} @$found),
 			$ext_urls, '</pre>');
-	PublicInbox::WwwStream->response($ctx, $code);
+	html_oneshot($ctx, $code);
 }
 
 1;