X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FLeiXSearch.pm;h=6f8770191ee6a28b36b1f7f631cbb9ebc7b26e86;hb=6313d8bbc5f3816eb4936443fca57b411f4f6286;hp=29df07e0c8a82e3a97ad620eaa59d4dccb2ee4c2;hpb=8fbb1e5d9430b9403679c5701e109d43ad979f3f;p=public-inbox.git
diff --git a/lib/PublicInbox/LeiXSearch.pm b/lib/PublicInbox/LeiXSearch.pm
index 29df07e0..6f877019 100644
--- a/lib/PublicInbox/LeiXSearch.pm
+++ b/lib/PublicInbox/LeiXSearch.pm
@@ -1,4 +1,4 @@
-# Copyright (C) 2020-2021 all contributors
+# Copyright (C) all contributors
# License: AGPL-3.0+
# Combine any combination of PublicInbox::Search,
@@ -163,8 +163,9 @@ sub mset_progress {
}
sub l2m_progress {
- my ($lei, $nr) = @_;
- $lei->{-nr_write} += $nr;
+ my ($lei, $nr_write, $nr_seen) = @_;
+ $lei->{-nr_write} += $nr_write;
+ $lei->{-nr_seen} += $nr_seen;
}
sub query_one_mset { # for --threads and l2m w/o sort
@@ -285,7 +286,7 @@ sub each_remote_eml { # callback for MboxReader->mboxrd
my ($res, $kw) = $self->{import_sto}->wq_do('add_eml', $eml);
if (ref($res) eq ref($smsg)) { # totally new message
$smsg = $res;
- $self->{-imported} = 1;
+ $self->{-sto_imported} = 1;
}
$smsg->{kw} = $kw; # short-circuit xsmsg_vmd
}
@@ -342,7 +343,7 @@ sub query_remote_mboxrd {
local $SIG{TERM} = sub { exit(0) }; # for DESTROY (File::Temp, $reap)
my $lei = $self->{lei};
my $opt = $lei->{opt};
- my $qstr = $lei->{mset_opt}->{qstr};
+ chomp(my $qstr = $lei->{mset_opt}->{qstr});
$qstr =~ s/[ \n\t]+/ /sg; # make URLs less ugly
my @qform = (x => 'm');
push(@qform, t => 1) if $opt->{threads};
@@ -376,7 +377,7 @@ sub query_remote_mboxrd {
$fh = IO::Uncompress::Gunzip->new($fh, MultiStream => 1);
PublicInbox::MboxReader->mboxrd($fh, \&each_remote_eml, $self,
$lei, $each_smsg);
- if ($self->{import_sto} && delete($self->{-imported})) {
+ if (delete($self->{-sto_imported})) {
my $wait = $self->{import_sto}->wq_do('done');
}
$reap_curl->join;
@@ -447,13 +448,16 @@ Error closing $lei->{ovv}->{dst}: \$!=$! \$?=$?
}
if ($lei->{-progress}) {
my $tot = $lei->{-mset_total} // 0;
- my $nr = $lei->{-nr_write} // 0;
+ my $nr_w = $lei->{-nr_write} // 0;
+ my $d = ($lei->{-nr_seen} // 0) - $nr_w;
+ my $x = "$tot matches";
+ $x .= ", $d duplicates" if $d;
if ($l2m) {
- my $m = "# $nr written to " .
- "$lei->{ovv}->{dst} ($tot matches)";
- $nr ? $lei->qfin($m) : $lei->qerr($m);
+ my $m = "# $nr_w written to " .
+ "$lei->{ovv}->{dst} ($x)";
+ $nr_w ? $lei->qfin($m) : $lei->qerr($m);
} else {
- $lei->qerr("# $tot matches");
+ $lei->qerr("# $x");
}
}
$lei->start_mua if $l2m && !$l2m->lock_free;