treewide: replace confess with croak

[public-inbox.git] / lib / PublicInbox / OverIdx.pm

diff --git a/lib/PublicInbox/OverIdx.pm b/lib/PublicInbox/OverIdx.pm
index 635aa314373ac6ca3c3da231a7237efbc909ec66..9013ae23e5349486a9913bc8593cc4ade791bed2 100644 (file)
--- a/lib/PublicInbox/OverIdx.pm
+++ b/lib/PublicInbox/OverIdx.pm
@@ -1,4 +1,4 @@
-# Copyright (C) 2018-2020 all contributors <meta@public-inbox.org>
+# Copyright (C) 2018-2021 all contributors <meta@public-inbox.org>
 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
 
 # for XOVER, OVER in NNTP, and feeds/homepage/threads in PSGI
@@ -79,6 +79,11 @@ SELECT $id_col FROM $tbl WHERE $val_col = ? LIMIT 1
        }
 }
 
+sub ibx_id {
+       my ($self, $eidx_key) = @_;
+       id_for($self, 'inboxes', 'ibx_id', eidx_key => $eidx_key);
+}
+
 sub sid {
        my ($self, $path) = @_;
        return unless defined $path && $path ne '';
@@ -238,26 +243,6 @@ sub link_refs {
        $tid;
 }
 
-sub parse_references ($$$) {
-       my ($smsg, $hdr, $mids) = @_;
-       my $refs = references($hdr);
-       push(@$refs, @$mids) if scalar(@$mids) > 1;
-       return $refs if scalar(@$refs) == 0;
-
-       # prevent circular references here:
-       my %seen = ( $smsg->{mid} => 1 );
-       my @keep;
-       foreach my $ref (@$refs) {
-               if (length($ref) > PublicInbox::MID::MAX_MID_SIZE) {
-                       warn "References: <$ref> too long, ignoring\n";
-                       next;
-               }
-               push(@keep, $ref) unless $seen{$ref}++;
-       }
-       $smsg->{references} = '<'.join('> <', @keep).'>' if @keep;
-       \@keep;
-}
-
 # normalize subjects so they are suitable as pathnames for URLs
 # XXX: consider for removal
 sub subject_path ($) {
@@ -278,7 +263,9 @@ sub add_overview {
        my ($self, $eml, $smsg) = @_;
        $smsg->{lines} = $eml->body_raw =~ tr!\n!\n!;
        my $mids = mids_for_index($eml);
-       my $refs = parse_references($smsg, $eml, $mids);
+       my $refs = $smsg->parse_references($eml, $mids);
+       $mids->[0] //= $smsg->{mid} //= $eml->{-lei_fake_mid};
+       $smsg->{mid} //= '';
        my $subj = $smsg->{subject};
        my $xpath;
        if ($subj ne '') {
@@ -468,10 +455,14 @@ sub dbh_close {
 
 sub create {
        my ($self) = @_;
-       unless (-r $self->{filename}) {
+       my $fn = $self->{filename} // do {
+               croak('BUG: no {filename}') unless $self->{dbh};
+               return;
+       };
+       unless (-r $fn) {
                require File::Path;
                require File::Basename;
-               File::Path::mkpath(File::Basename::dirname($self->{filename}));
+               File::Path::mkpath(File::Basename::dirname($fn));
        }
        # create the DB:
        PublicInbox::Over::dbh($self);
@@ -526,7 +517,7 @@ sub eidx_prep ($) {
        my ($self) = @_;
        $self->{-eidx_prep} //= do {
                my $dbh = $self->dbh;
-               $dbh->do(<<"");
+               $dbh->do(<<'');
 INSERT OR IGNORE INTO counter (key) VALUES ('eidx_docid')
 
                $dbh->do(<<'');
@@ -558,7 +549,14 @@ CREATE TABLE IF NOT EXISTS eidx_meta (
        val VARCHAR(255) NOT NULL
 )
 
-               $dbh;
+               # A queue of current docids which need reindexing.
+               # eidxq persists across aborted -extindex invocations
+               # Currently used for "-extindex --reindex" for Xapian
+               # data, but may be used in more places down the line.
+               $dbh->do(<<'');
+CREATE TABLE IF NOT EXISTS eidxq (docid INTEGER PRIMARY KEY NOT NULL)
+
+               1;
        };
 }
 
@@ -588,7 +586,7 @@ sub eidx_max {
 sub add_xref3 {
        my ($self, $docid, $xnum, $oidhex, $eidx_key) = @_;
        begin_lazy($self);
-       my $ibx_id = id_for($self, 'inboxes', 'ibx_id', eidx_key => $eidx_key);
+       my $ibx_id = ibx_id($self, $eidx_key);
        my $oidbin = pack('H*', $oidhex);
        my $sth = $self->{dbh}->prepare_cached(<<'');
 INSERT OR IGNORE INTO xref3 (docid, ibx_id, xnum, oidbin) VALUES (?, ?, ?, ?)
@@ -607,8 +605,7 @@ sub remove_xref3 {
        my $oidbin = pack('H*', $oidhex);
        my ($sth, $ibx_id);
        if (defined $eidx_key) {
-               $ibx_id = id_for($self, 'inboxes', 'ibx_id',
-                                       eidx_key => $eidx_key);
+               $ibx_id = ibx_id($self, $eidx_key);
                $sth = $self->{dbh}->prepare_cached(<<'');
 DELETE FROM xref3 WHERE docid = ? AND ibx_id = ? AND oidbin = ?
 
@@ -657,4 +654,28 @@ UPDATE over SET ddd = ? WHERE num = ?
        $sth->execute;
 }
 
+sub eidxq_add {
+       my ($self, $docid) = @_;
+       $self->dbh->prepare_cached(<<'')->execute($docid);
+INSERT OR IGNORE INTO eidxq (docid) VALUES (?)
+
+}
+
+sub eidxq_del {
+       my ($self, $docid) = @_;
+       $self->dbh->prepare_cached(<<'')->execute($docid);
+DELETE FROM eidxq WHERE docid = ?
+
+}
+
+sub blob_exists {
+       my ($self, $oidhex) = @_;
+       my $sth = $self->dbh->prepare_cached(<<'', undef, 1);
+SELECT COUNT(*) FROM xref3 WHERE oidbin = ?
+
+       $sth->bind_param(1, pack('H*', $oidhex), SQL_BLOB);
+       $sth->execute;
+       $sth->fetchrow_array;
+}
+
 1;