-# Copyright (C) 2015-2018 all contributors <meta@public-inbox.org>
+# Copyright (C) 2015-2020 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
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',
- 'http://mid.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',
'doc-url=/lurker&format=en.html&query=id:%s'
);
+sub PARTIAL_MAX () { 100 }
+
+sub search_partial ($$) {
+ my ($ibx, $mid) = @_;
+ return if length($mid) < $MIN_PARTIAL_LEN;
+ my $srch = $ibx->search or return;
+ my $opt = { limit => PARTIAL_MAX, mset => 2 };
+ my @try = ("m:$mid*");
+ my $chop = $mid;
+ if ($chop =~ s/(\W+)(\w*)\z//) {
+ my ($delim, $word) = ($1, $2);
+ if (length($word)) {
+ push @try, "m:$chop$delim";
+ push @try, "m:$chop$delim*";
+ }
+ push @try, "m:$chop";
+ push @try, "m:$chop*";
+ }
+
+ # break out long words individually to search for, because
+ # too many messages begin with "Pine.LNX." (or "alpine" or "nycvar")
+ if ($mid =~ /\w{9,}/) {
+ my @long = ($mid =~ m!(\w{3,})!g);
+ push(@try, join(' ', map { "m:$_" } @long));
+
+ # is the last element long enough to not trigger excessive
+ # wildcard matches?
+ if (length($long[-1]) > 8) {
+ $long[-1] .= '*';
+ push(@try, join(' ', map { "m:$_" } @long));
+ }
+ }
+
+ foreach my $m (@try) {
+ # 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 = map {
+ $_->{mid}
+ } @{$ibx->over->get_all(@{$srch->mset_to_artnums($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, $other;
+ }
+}
+
sub ext_msg {
my ($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 = [];
+ 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 $n_partial = 0;
my @partial;
-
- if (my $mm = $cur->mm) {
- my $tmp_mid = $mid;
- my $res = $mm->mid_prefixes($tmp_mid, 100);
- if ($res && scalar(@$res)) {
- $n_partial += scalar(@$res);
- push @partial, [ $cur, $res ];
- # fixup common errors:
- } elsif ($tmp_mid =~ s,/[tTf],,) {
- $res = $mm->mid_prefixes($tmp_mid, 100);
- if ($res && scalar(@$res)) {
- $n_partial += scalar(@$res);
- push @partial, [ $cur, $res ];
- }
- }
+ my $n_partial = 0;
+ 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) {
- my $tmp_mid = $mid;
-again:
- foreach my $ibx (@ibx) {
- my $mm = $ibx->mm or next;
- my $res = $mm->mid_prefixes($tmp_mid, 100);
- if ($res && scalar(@$res)) {
- $n_partial += scalar(@$res);
- push @partial, [ $ibx, $res ];
- last if $n_partial >= 100;
- }
- }
- # fixup common errors:
- if (!$n_partial && $tmp_mid =~ s,/[tTf],,) {
- goto again;
+ if (!$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;
}
}
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 $s = "<pre>Message-ID <$html>\nnot found\n";
if ($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 $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;
- $s .= qq{<a\nhref="$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};
}
}
}
$ctx->{-html_tip} = $s .= '</pre>';
$ctx->{-title_html} = $title;
$ctx->{-upfx} = '../';
- PublicInbox::WwwStream->response($ctx, $code);
+ html_oneshot($ctx, $code);
}
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">$t</a>\n};
sub exact {
my ($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 . $end;
qq(<a\nhref="$u$href/">$u$html/</a>\n)
} @$found),
$ext_urls, '</pre>');
- PublicInbox::WwwStream->response($ctx, $code);
+ html_oneshot($ctx, $code);
}
1;