X-Git-Url: http://www.git.stargrave.org/?p=public-inbox.git;a=blobdiff_plain;f=lib%2FPublicInbox%2FLeiSearch.pm;h=936c275178dfc6908577487ce97d3a83c58c5449;hp=568277a6f40443217494e383b609f83f9c5d571f;hb=961690bae47c90a4a6960952587c6f4463fb4b19;hpb=bf7fdfe4b939c3a97b20fd3e9b65c9cd493a692d diff --git a/lib/PublicInbox/LeiSearch.pm b/lib/PublicInbox/LeiSearch.pm index 568277a6..936c2751 100644 --- a/lib/PublicInbox/LeiSearch.pm +++ b/lib/PublicInbox/LeiSearch.pm @@ -1,4 +1,4 @@ -# Copyright (C) 2020-2021 all contributors +# Copyright (C) all contributors # License: AGPL-3.0+ # read-only counterpart for PublicInbox::LeiStore @@ -7,22 +7,14 @@ use strict; use v5.10.1; use parent qw(PublicInbox::ExtSearch); # PublicInbox::Search->reopen use PublicInbox::Search qw(xap_terms); -use PublicInbox::ContentHash qw(content_digest content_hash); +use PublicInbox::ContentHash qw(content_digest content_hash git_sha); use PublicInbox::MID qw(mids mids_for_index); use Carp qw(croak); -# get combined docid from over.num: -# (not generic Xapian, only works with our sharding scheme) -sub num2docid ($$) { - my ($self, $num) = @_; - my $nshard = $self->{nshard}; - ($num - 1) * $nshard + $num % $nshard + 1; -} - sub _msg_kw { # retry_reopen callback my ($self, $num) = @_; my $xdb = $self->xdb; # set {nshard} for num2docid; - xap_terms('K', $xdb, num2docid($self, $num)); + xap_terms('K', $xdb, $self->num2docid($num)); } sub msg_keywords { # array or hashref @@ -35,7 +27,7 @@ sub _oid_kw { # retry_reopen callback my $xdb = $self->xdb; # set {nshard}; my %kw; for my $num (@$nums) { # there should only be one... - my $doc = $xdb->get_document(num2docid($self, $num)); + my $doc = $xdb->get_document($self->num2docid($num)); my $x = xap_terms('K', $doc); %kw = (%kw, %$x); } @@ -56,7 +48,7 @@ sub _xsmsg_vmd { # retry_reopen $kw{flagged} = 1 if delete($smsg->{lei_q_tt_flagged}); my @num = $self->over->blob_exists($smsg->{blob}); for my $num (@num) { # there should only be one... - $doc = $xdb->get_document(num2docid($self, $num)); + $doc = $xdb->get_document($self->num2docid($num)); $x = xap_terms('K', $doc); %kw = (%kw, %$x); if ($want_label) { # JSON/JMAP only @@ -109,7 +101,7 @@ sub xoids_for { my $git = $self->git; my $xoids = {}; # no lms when used via {ale}: - my $lms = $self->{-lms_ro} //= lms($self) if defined($self->{topdir}); + my $lms = $self->{-lms_rw} //= lms($self) if defined($self->{topdir}); for my $mid (@$mids) { for my $o (@overs) { my ($id, $prev); @@ -119,13 +111,20 @@ sub xoids_for { $git->cat_async($cur->{blob}, \&_cmp_1st, [$chash, $xoids, $cur, $lms]); if ($min && scalar(keys %$xoids) >= $min) { - $git->cat_async_wait; + $git->async_wait_all; return $xoids; } } } } - $git->cat_async_wait; + $git->async_wait_all; + + # it could be an 'lei index'-ed file that just got renamed + if (scalar(keys %$xoids) < ($min // 1) && defined($self->{topdir})) { + my $hex = git_sha(1, $eml)->hexdigest; + my @n = $overs[0]->blob_exists($hex); + for (@n) { $xoids->{$hex} //= $_ } + } scalar(keys %$xoids) ? $xoids : undef; } @@ -137,6 +136,10 @@ sub kw_changed { my $xoids = xoids_for($self, $eml) // return; $docids //= []; @$docids = sort { $a <=> $b } values %$xoids; + if (!@$docids && $self->over) { + my $bin = git_sha(1, $eml)->digest; + @$docids = $self->over->oidbin_exists($bin); + } } for my $id (@$docids) { $cur_kw = eval { msg_keywords($self, $id) } and last; @@ -172,6 +175,8 @@ sub all_terms { sub qparse_new { my ($self) = @_; my $qp = $self->SUPER::qparse_new; # PublicInbox::Search + $self->{qp_flags} |= PublicInbox::Search::FLAG_PHRASE() | + PublicInbox::Search::FLAG_PURE_NOT(); $qp->add_boolean_prefix('kw', 'K'); $qp->add_boolean_prefix('L', 'L'); $qp