diff options
Diffstat (limited to 'lib/PublicInbox/ExtMsg.pm')
| -rw-r--r-- | lib/PublicInbox/ExtMsg.pm | 73 |
1 files changed, 55 insertions, 18 deletions
diff --git a/lib/PublicInbox/ExtMsg.pm b/lib/PublicInbox/ExtMsg.pm index 03faf3a1..95feb885 100644 --- a/lib/PublicInbox/ExtMsg.pm +++ b/lib/PublicInbox/ExtMsg.pm @@ -1,4 +1,4 @@ -# Copyright (C) 2015-2020 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 @@ -11,7 +11,7 @@ use warnings; use PublicInbox::Hval qw(ascii_html prurl mid_href); use PublicInbox::WwwStream qw(html_oneshot); use PublicInbox::Smsg; -our $MIN_PARTIAL_LEN = 16; +our $MIN_PARTIAL_LEN = 14; # for 'XXXXXXXXXX.fsf' msgids gnus generates # TODO: user-configurable our @EXT_URL = map { ascii_html($_) } ( @@ -32,8 +32,8 @@ sub PARTIAL_MAX () { 100 } sub search_partial ($$) { my ($ibx, $mid) = @_; return if length($mid) < $MIN_PARTIAL_LEN; - my $srch = $ibx->search or return; - my $opt = { limit => PARTIAL_MAX, mset => 2 }; + my $srch = $ibx->isrch or return; + my $opt = { limit => PARTIAL_MAX, relevance => -1 }; my @try = ("m:$mid*"); my $chop = $mid; if ($chop =~ s/(\W+)(\w*)\z//) { @@ -76,7 +76,7 @@ sub search_partial ($$) { sub ext_msg_i { my ($other, $ctx) = @_; - return if $other->{name} eq $ctx->{-inbox}->{name} || !$other->base_url; + return if $other->{name} eq $ctx->{ibx}->{name} || !$other->base_url; my $mm = $other->mm or return; @@ -103,25 +103,54 @@ sub ext_msg_step { } } +sub ext_msg_ALL ($) { + 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 + 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 ($ctx) = @_; - sub { + ext_msg_ALL($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_config}, + $ctx->{www}->{pi_cfg}, \&ext_msg_step, $ctx); $iter->event_step; } else { - $ctx->{www}->{pi_config}->each_inbox(\&ext_msg_i, $ctx); + $ctx->{www}->{pi_cfg}->each_inbox(\&ext_msg_i, $ctx); finalize_exact($ctx); } }; } -# called via PublicInbox::DS->EventLoop +# called via PublicInbox::DS::event_loop sub event_step { my ($ctx, $sync) = @_; # can't find a partial match in current inbox, try the others: @@ -141,7 +170,7 @@ sub finalize_exact { # fall back to partial MID matching my $mid = $ctx->{mid}; - my $cur = $ctx->{-inbox}; + my $cur = $ctx->{ibx}; my $mids = search_partial($cur, $mid); if ($mids) { $ctx->{n_partial} = scalar(@$mids); @@ -159,7 +188,13 @@ sub finalize_exact { finalize_partial($ctx); } -sub finalize_partial { +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; @@ -167,16 +202,17 @@ sub finalize_partial { my $html = ascii_html($mid); my $title = "<$html> not found"; my $s = "<pre>Message-ID <$html>\nnot found\n"; + $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 = $ctx->{-inbox}->{name}; + 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 $href = mid_href($m); my $html = ascii_html($m); @@ -191,10 +227,11 @@ sub finalize_partial { } $ctx->{-html_tip} = $s .= '</pre>'; $ctx->{-title_html} = $title; - $ctx->{-upfx} = '../'; - $ctx->{-wcb}->(html_oneshot($ctx, $code)); + html_oneshot($ctx, $code); } +sub finalize_partial ($) { $_[0]->{-wcb}->(partial_response($_[0])) } + sub ext_urls { my ($ctx, $mid, $href, $html) = @_; @@ -222,13 +259,13 @@ sub exact { my $title = "<$html> 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; $ctx->{-html_tip} = join('', "<pre>Message-ID: <$html>\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>'); |