X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FExtMsg.pm;h=929737f1b9d1961f4c63f556c313494df0db90fb;hb=352e2799ecec328f71aa33219214a0e3fc3d5f10;hp=51e7799de8640929be9538713d3f2ee7e1b1081d;hpb=a46893a2b5dabfdbcf7b593ac19967daecfb1772;p=public-inbox.git diff --git a/lib/PublicInbox/ExtMsg.pm b/lib/PublicInbox/ExtMsg.pm index 51e7799d..929737f1 100644 --- a/lib/PublicInbox/ExtMsg.pm +++ b/lib/PublicInbox/ExtMsg.pm @@ -1,4 +1,4 @@ -# Copyright (C) 2015-2018 all contributors +# Copyright (C) 2015-2020 all contributors # License: AGPL-3.0+ # # Used by the web interface to link to messages outside of the our @@ -8,16 +8,17 @@ package PublicInbox::ExtMsg; use strict; use warnings; -use PublicInbox::Hval; -use PublicInbox::MID qw/mid2path/; -use PublicInbox::WwwStream; +use PublicInbox::Hval qw(ascii_html prurl mid_href); +use PublicInbox::WwwStream qw(html_oneshot); +use PublicInbox::Smsg; +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', - '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', @@ -29,7 +30,9 @@ our @EXT_URL = ( sub PARTIAL_MAX () { 100 } sub search_partial ($$) { - my ($srch, $mid) = @_; + my ($ibx, $mid) = @_; + return if length($mid) < $MIN_PARTIAL_LEN; + my $srch = $ibx->search or return; my $opt = { limit => PARTIAL_MAX, mset => 2 }; my @try = ("m:$mid*"); my $chop = $mid; @@ -58,64 +61,64 @@ sub search_partial ($$) { } 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; - } + # 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 { $srch->mset($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, $mset)}; return \@mids if scalar(@mids); } } +sub ext_msg_i { + my ($other, $cur, $mid, $ibxs, $found) = @_; + + 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, $other; + } +} + sub ext_msg { my ($ctx) = @_; my $cur = $ctx->{-inbox}; 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 = []; + + $ctx->{www}->{pi_config}->each_inbox(\&ext_msg_i, + $cur, $mid, $ibxs, $found); - return exact($ctx, \@found, $mid) if @found; + return exact($ctx, $found, $mid) if @$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; + my $mids = search_partial($cur, $mid); if ($mids) { $n_partial = scalar(@$mids); push @partial, [ $cur, $mids ]; } # can't find a partial match in current inbox, try the others: - if (!$n_partial && length($mid) >= 16) { - foreach my $ibx (@ibx) { - $srch = $ibx->search or next; - $mids = search_partial($srch, $mid) or next; + if (!$n_partial && length($mid) >= $MIN_PARTIAL_LEN) { + foreach my $ibx (@$ibxs) { + $mids = search_partial($ibx, $mid) or next; $n_partial += scalar(@$mids); push @partial, [ $ibx, $mids]; last if $n_partial >= PARTIAL_MAX; @@ -123,9 +126,8 @@ sub ext_msg { } 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 $title = "<$html> not found"; my $s = "
Message-ID <$html>\nnot found\n";
 	if ($n_partial) {
@@ -139,10 +141,9 @@ sub ext_msg {
 			my $env = $ctx->{env} if $ibx->{name} eq $cur_name;
 			my $u = $ibx->base_url($env) or next;
 			foreach my $m (@$res) {
-				my $p = PublicInbox::Hval->new_msgid($m);
-				my $r = $p->{href};
-				my $t = $p->as_html;
-				$s .= qq{$u$t/\n};
+				my $href = mid_href($m);
+				my $html = ascii_html($m);
+				$s .= qq{$u$html/\n};
 			}
 		}
 	}
@@ -154,7 +155,7 @@ sub ext_msg {
 	$ctx->{-html_tip} = $s .= '
'; $ctx->{-title_html} = $title; $ctx->{-upfx} = '../'; - PublicInbox::WwwStream->response($ctx, $code); + html_oneshot($ctx, $code); } sub ext_urls { @@ -165,7 +166,7 @@ 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); $e .= qq{$t\n}; @@ -177,9 +178,8 @@ 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 $href = mid_href($mid); + my $html = ascii_html($mid); my $title = "<$html> found in "; my $end = @$found == 1 ? 'another inbox' : 'other inboxes'; $ctx->{-title_html} = $title . $end; @@ -193,7 +193,7 @@ sub exact { qq($u$html/\n) } @$found), $ext_urls, ''); - PublicInbox::WwwStream->response($ctx, $code); + html_oneshot($ctx, $code); } 1;