X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FExtMsg.pm;h=49fc1ac5521c1506acb137afaf1f667f666f231d;hb=55b707d788ce13696e4411389583e720ea6dab01;hp=51e7799de8640929be9538713d3f2ee7e1b1081d;hpb=a46893a2b5dabfdbcf7b593ac19967daecfb1772;p=public-inbox.git

diff --git a/lib/PublicInbox/ExtMsg.pm b/lib/PublicInbox/ExtMsg.pm
index 51e7799d..49fc1ac5 100644
--- a/lib/PublicInbox/ExtMsg.pm
+++ b/lib/PublicInbox/ExtMsg.pm
@@ -1,4 +1,4 @@
-# Copyright (C) 2015-2018 all contributors <meta@public-inbox.org>
+# Copyright (C) 2015-2019 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
@@ -8,12 +8,12 @@
 package PublicInbox::ExtMsg;
 use strict;
 use warnings;
-use PublicInbox::Hval;
-use PublicInbox::MID qw/mid2path/;
+use PublicInbox::Hval qw(ascii_html prurl);
 use PublicInbox::WwwStream;
+our $MIN_PARTIAL_LEN = 16;
 
 # TODO: user-configurable
-our @EXT_URL = (
+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',
@@ -28,8 +28,13 @@ our @EXT_URL = (
 
 sub PARTIAL_MAX () { 100 }
 
+sub mids_from_mset { # Search::retry_reopen callback
+	[ map { PublicInbox::SearchMsg::from_mitem($_)->mid } $_[0]->items ];
+}
+
 sub search_partial ($$) {
 	my ($srch, $mid) = @_;
+	return if length($mid) < $MIN_PARTIAL_LEN;
 	my $opt = { limit => PARTIAL_MAX, mset => 2 };
 	my @try = ("m:$mid*");
 	my $chop = $mid;
@@ -58,17 +63,33 @@ sub search_partial ($$) {
 	}
 
 	foreach my $m (@try) {
-		my $mset = eval { $srch->query($m, $opt) };
-		if (ref($@) eq 'Search::Xapian::QueryParserError') {
-			# If Xapian can't handle the wildcard since it
-			# has too many results.
-			next;
-		}
-		my @mids = map {
-			my $doc = $_->get_document;
-			PublicInbox::SearchMsg->load_doc($doc)->mid;
-		} $mset->items;
-		return \@mids if scalar(@mids);
+		# If Xapian can't handle the wildcard since it
+		# 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 $mids = $srch->retry_reopen(\&mids_from_mset, $mset);
+		return $mids if scalar(@$mids);
+	}
+}
+
+sub ext_msg_i {
+	my ($other, $arg) = @_;
+	my ($cur, $mid, $ibxs, $found) = @$arg;
+
+	return if $other->{name} eq $cur->{name} || !$other->base_url;
+
+	my $mm = $other->mm or return;
+
+	# try to find the URL with Msgmap to avoid forking
+	my $num = $mm->num_for($mid);
+	if (defined $num) {
+		push @$found, $other;
+	} else {
+		# no point in trying the fork fallback if we
+		# know Xapian is up-to-date but missing the
+		# message in the current repo
+		push @$ibxs, $other;
 	}
 }
 
@@ -78,28 +99,13 @@ sub ext_msg {
 	my $mid = $ctx->{mid};
 
 	eval { require PublicInbox::Msgmap };
-	my $have_mm = $@ ? 0 : 1;
-	my (@ibx, @found);
-
-	$ctx->{www}->{pi_config}->each_inbox(sub {
-		my ($other) = @_;
-		return if $other->{name} eq $cur->{name} || !$other->base_url;
-
-		my $mm = $other->mm or return;
-
-		# 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;
-		}
-	});
+	my $ibxs = [];
+	my $found = [];
+	my $arg = [ $cur, $mid, $ibxs, $found ];
+
+	$ctx->{www}->{pi_config}->each_inbox(\&ext_msg_i, $arg);
 
-	return exact($ctx, \@found, $mid) if @found;
+	return exact($ctx, $found, $mid) if @$found;
 
 	# fall back to partial MID matching
 	my @partial;
@@ -112,8 +118,8 @@ sub ext_msg {
 	}
 
 	# can't find a partial match in current inbox, try the others:
-	if (!$n_partial && length($mid) >= 16) {
-		foreach my $ibx (@ibx) {
+	if (!$n_partial && length($mid) >= $MIN_PARTIAL_LEN) {
+		foreach my $ibx (@$ibxs) {
 			$srch = $ibx->search or next;
 			$mids = search_partial($srch, $mid) or next;
 			$n_partial += scalar(@$mids);
@@ -165,7 +171,7 @@ 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};