diff options
| author | Eric Wong <e@80x24.org> | 2023-08-24 01:22:32 +0000 |
|---|---|---|
| committer | Eric Wong <e@80x24.org> | 2023-08-24 07:47:50 +0000 |
| commit | cf96412eb8f193ebd334fae340b2d91b6b7f2afe (patch) | |
| tree | e71b3af27231e3315aa23e47b2e907cedc337dad | |
| parent | 2f41101df13a82befc24edefe6946b2d52d2c96c (diff) | |
| download | public-inbox-cf96412eb8f193ebd334fae340b2d91b6b7f2afe.tar.gz | |
This aids in development, but I'm not sure it's going to stay or be moved into another interface.
| -rw-r--r-- | lib/PublicInbox/CodeSearchIdx.pm | 32 | ||||
| -rwxr-xr-x | script/public-inbox-cindex | 2 |
2 files changed, 33 insertions, 1 deletions
diff --git a/lib/PublicInbox/CodeSearchIdx.pm b/lib/PublicInbox/CodeSearchIdx.pm index 2480dbd2..e795c2b3 100644 --- a/lib/PublicInbox/CodeSearchIdx.pm +++ b/lib/PublicInbox/CodeSearchIdx.pm @@ -1058,6 +1058,37 @@ sub _prep_ibx { # each_inbox callback push @{$self->{IBX}}, $ibx; } +sub show_roots { # for diagnostics + my ($self) = @_; + local $self->{xdb}; + my $cur = $self->xdb->allterms_begin('G'); + my $end = $self->{xdb}->allterms_end('G'); + my $qrepo = $PublicInbox::Search::X{Query}->new('T'.'r'); + my $enq = $PublicInbox::Search::X{Enquire}->new($self->{xdb}); + $enq->set_weighting_scheme($PublicInbox::Search::X{BoolWeight}->new); + $enq->set_docid_order($PublicInbox::Search::ENQ_ASCENDING); + for (; $cur != $end; $cur++) { + my $G_oidhex = $cur->get_termname; + my $qry = $PublicInbox::Search::X{Query}->new( + PublicInbox::Search::OP_FILTER(), + $qrepo, $G_oidhex); + $enq->set_query($qry); + my ($off, $lim) = (0, 10000); + say 'commit ',substr($G_oidhex, 1), ' appears in:'; + while (1) { + my $mset = $enq->get_mset($off, $lim); + my $size = $mset->size or last; + for my $x ($mset->items) { + my $doc = $x->get_document; + for (xap_terms('P', $x->get_document)) { + say '- /', substr($_, 1); + } + } + $off += $size; + } + } +} + sub cidx_run { # main entry point my ($self) = @_; my $restore_umask = prep_umask($self); @@ -1150,6 +1181,7 @@ sub cidx_run { # main entry point PublicInbox::DS::event_loop($MY_SIG, $SIGSET) if shards_active(); PublicInbox::DS->Reset; $self->lock_release(!!$NCHANGE); + show_roots($self) if $self->{-opt}->{'show-roots'} # for diagnostics } sub ipc_atfork_child { # @IDX_SHARDS diff --git a/script/public-inbox-cindex b/script/public-inbox-cindex index 888c8b10..0526434c 100755 --- a/script/public-inbox-cindex +++ b/script/public-inbox-cindex @@ -29,7 +29,7 @@ GetOptions($opt, qw(quiet|q verbose|v+ reindex jobs|j=i fsync|sync! dangerous indexlevel|index-level|L=s associate associate-max=i associate-date-range=s associate-prefixes=s@ batch_size|batch-size=s max_size|max-size=s - include|I=s@ only=s@ all + include|I=s@ only=s@ all show-roots project-list=s exclude=s@ sort-parallel=s sort-compress-program=s sort-buffer-size=s d=s update|u scan! prune dry-run|n C=s@ help|h)) |