X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FSearch.pm;h=82a6e5416240e19876203fc61fcd5b25d289d420;hb=feb4a75affbe5f940116dc4f6bcc5ddb52a110b4;hp=24cb26670754b2f4a21b62e7d22c3cdc76098a5a;hpb=53e8cfbe4e06e5ae6ad61fb7e9bd45804c253a72;p=public-inbox.git diff --git a/lib/PublicInbox/Search.pm b/lib/PublicInbox/Search.pm index 24cb2667..82a6e541 100644 --- a/lib/PublicInbox/Search.pm +++ b/lib/PublicInbox/Search.pm @@ -16,7 +16,7 @@ use constant YYYYMMDD => 4; # for searching in the WWW UI use Search::Xapian qw/:standard/; use PublicInbox::SearchMsg; -use Email::MIME; +use PublicInbox::MIME; use PublicInbox::MID qw/mid_clean id_compress/; # This is English-only, everything else is non-standard and may be confused as @@ -39,7 +39,10 @@ use constant { # 10 - optimize doc for NNTP overviews # 11 - merge threads when vivifying ghosts # 12 - change YYYYMMDD value column to numeric - SCHEMA_VERSION => 12, + # 13 - fix threading for empty References/In-Reply-To + # (commit 83425ef12e4b65cdcecd11ddcb38175d4a91d5a0) + # 14 - fix ghost root vivification + SCHEMA_VERSION => 14, # n.b. FLAG_PURE_NOT is expensive not suitable for a public website # as it could become a denial-of-service vector @@ -89,7 +92,7 @@ are also supported EOF 'b:' => 'match within message body, including text attachments', 'nq:' => 'match non-quoted text within message body', - 'quot:' => 'match quoted text within message body', + 'q:' => 'match quoted text within message body', 'n:' => 'match filename of attachment(s)', 't:' => 'match within the To header', 'c:' => 'match within the Cc header', @@ -108,12 +111,6 @@ my %all_pfx = (%bool_pfx_internal, %bool_pfx_external, %prob_prefix); sub xpfx { $all_pfx{$_[0]} } -our %PFX2TERM_RMAP; -my %meta_pfx = (mid => 1, thread => 1, path => 1); -while (my ($k, $v) = each %all_pfx) { - $PFX2TERM_RMAP{$v} = $k if $meta_pfx{$k}; -} - my $mail_query = Search::Xapian::Query->new(xpfx('type') . 'mail'); sub xdir { @@ -288,15 +285,17 @@ sub lookup_message { # raises on error: my $doc = $self->{xdb}->get_document($doc_id); $smsg = PublicInbox::SearchMsg->wrap($doc, $mid); - $smsg->doc_id($doc_id); + $smsg->{doc_id} = $doc_id; } $smsg; } sub lookup_mail { # no ghosts! my ($self, $mid) = @_; - my $smsg = lookup_message($self, $mid) or return; - PublicInbox::SearchMsg->load_doc($smsg->{doc}); + retry_reopen($self, sub { + my $smsg = lookup_message($self, $mid) or return; + PublicInbox::SearchMsg->load_doc($smsg->{doc}); + }); } sub find_unique_doc_id { @@ -330,6 +329,7 @@ sub find_doc_ids_for_term { } # normalize subjects so they are suitable as pathnames for URLs +# XXX: consider for removal sub subject_path { my $subj = pop; $subj = subject_normalized($subj); @@ -347,32 +347,6 @@ sub subject_normalized { $subj; } -# for doc data -sub subject_summary { - my $subj = pop; - my $max = 68; - if (length($subj) > $max) { - my @subj = split(/\s+/, $subj); - $subj = ''; - my $l; - - while ($l = shift @subj) { - my $new = $subj . $l . ' '; - last if length($new) >= $max; - $subj = $new; - } - if ($subj ne '') { - my $r = scalar @subj ? ' ...' : ''; - $subj =~ s/ \z/$r/s; - } else { - # subject has one REALLY long word, and NOT spam? wtf - @subj = ($l =~ /\A(.{1,72})/); - $subj = $subj[0] . ' ...'; - } - } - $subj; -} - sub enquire { my ($self) = @_; $self->{enquire} ||= Search::Xapian::Enquire->new($self->{xdb});