X-Git-Url: http://www.git.stargrave.org/?p=public-inbox.git;a=blobdiff_plain;f=lib%2FPublicInbox%2FSearch.pm;h=7175ddc5a2e06c078ef10f1d7875e3eebd62d709;hp=0f102dac0b72414d87c52233030e81f3ddfd5bd5;hb=fad9acd35e56a289ade90a62d056b2a6663d448c;hpb=5abd4b9a1013f5cf3859dbd6e02044757fb98489 diff --git a/lib/PublicInbox/Search.pm b/lib/PublicInbox/Search.pm index 0f102dac..7175ddc5 100644 --- a/lib/PublicInbox/Search.pm +++ b/lib/PublicInbox/Search.pm @@ -8,16 +8,15 @@ use strict; use warnings; # values for searching -use constant TS => 0; # timestamp -use constant NUM => 1; # NNTP article number -use constant BYTES => 2; # :bytes as defined in RFC 3977 -use constant LINES => 3; # :lines as defined in RFC 3977 -use constant YYYYMMDD => 4; # for searching in the WWW UI +use constant TS => 0; # Received: header in Unix time +use constant YYYYMMDD => 1; # Date: header for searching in the WWW UI +use constant DT => 2; # Date: YYYYMMDDHHMMSS use Search::Xapian qw/:standard/; use PublicInbox::SearchMsg; use PublicInbox::MIME; -use PublicInbox::MID qw/mid_clean id_compress/; +use PublicInbox::MID qw/id_compress/; +use PublicInbox::Over; # This is English-only, everything else is non-standard and may be confused as # a prefix common in patch emails @@ -42,23 +41,18 @@ use constant { # 13 - fix threading for empty References/In-Reply-To # (commit 83425ef12e4b65cdcecd11ddcb38175d4a91d5a0) # 14 - fix ghost root vivification - SCHEMA_VERSION => 14, + SCHEMA_VERSION => 15, # n.b. FLAG_PURE_NOT is expensive not suitable for a public website # as it could become a denial-of-service vector QP_FLAGS => FLAG_PHRASE|FLAG_BOOLEAN|FLAG_LOVEHATE|FLAG_WILDCARD, }; -# setup prefixes -my %bool_pfx_internal = ( - type => 'T', # "mail" or "ghost" - thread => 'G', # newsGroup (or similar entity - e.g. a web forum name) -); - my %bool_pfx_external = ( - mid => 'XMID', # Message-ID (full/exact) + mid => 'Q', # Message-ID (full/exact), this is mostly uniQue ); +my $non_quoted_body = 'XNQ XDFN XDFA XDFB XDFHH XDFCTX XDFPRE XDFPOST'; my %prob_prefix = ( # for mairix compatibility s => 'S', @@ -69,12 +63,12 @@ my %prob_prefix = ( c => 'XCC', tcf => 'XTO XCC A', a => 'XTO XCC A', - b => 'XNQ XQUOT', - bs => 'XNQ XQUOT S', + b => $non_quoted_body . ' XQUOT', + bs => $non_quoted_body . ' XQUOT S', n => 'XFN', q => 'XQUOT', - nq => 'XNQ', + nq => $non_quoted_body, dfn => 'XDFN', dfa => 'XDFA', dfb => 'XDFB', @@ -85,7 +79,7 @@ my %prob_prefix = ( dfblob => 'XDFPRE XDFPOST', # default: - '' => 'XM S A XNQ XQUOT XFN', + '' => 'XM S A XQUOT XFN ' . $non_quoted_body, ); # not documenting m: and mid: for now, the using the URLs works w/o Xapian @@ -95,6 +89,9 @@ our @HELP = ( date range as YYYYMMDD e.g. d:19931002..20101002 Open-ended ranges such as d:19931002.. and d:..20101002 are also supported +EOF + 'dt:' => < 'match within message body, including text attachments', 'nq:' => 'match non-quoted text within message body', @@ -117,8 +114,6 @@ EOF ); chomp @HELP; -my $mail_query = Search::Xapian::Query->new('T' . 'mail'); - sub xdir { my ($self) = @_; if ($self->{version} == 1) { @@ -144,8 +139,9 @@ sub new { altid => $altid, version => $version, }, $class; + my $dir; if ($version >= 2) { - my $dir = "$self->{mainrepo}/xap" . SCHEMA_VERSION; + $dir = "$self->{mainrepo}/xap" . SCHEMA_VERSION; my $xdb; my $parts = 0; foreach my $part (<$dir/*>) { @@ -158,74 +154,62 @@ sub new { $xdb = $sub; } } - warn "v2 repo with $parts found in $dir\n"; $self->{xdb} = $xdb; - $self->{skel} = Search::Xapian::Database->new("$dir/all"); } else { - $self->{xdb} = Search::Xapian::Database->new($self->xdir); + $dir = $self->xdir; + $self->{xdb} = Search::Xapian::Database->new($dir); } + $self->{over_ro} = PublicInbox::Over->new("$dir/over.sqlite3"); $self; } sub reopen { my ($self) = @_; $self->{xdb}->reopen; - if (my $skel = $self->{skel}) { - $skel->reopen; - } + $self; # make chaining easier } # read-only sub query { my ($self, $query_string, $opts) = @_; - my $query; - $opts ||= {}; - unless ($query_string eq '') { - $query = $self->qp->parse_query($query_string, QP_FLAGS); + if ($query_string eq '' && !$opts->{mset}) { + $self->{over_ro}->recent($opts); + } else { + my $query = $self->qp->parse_query($query_string, QP_FLAGS); $opts->{relevance} = 1 unless exists $opts->{relevance}; + _do_enquire($self, $query, $opts); } - - _do_enquire($self, $query, $opts); } sub get_thread { - my ($self, $mid, $opts) = @_; - my $smsg = retry_reopen($self, sub { lookup_skeleton($self, $mid) }); - - return { total => 0, msgs => [] } unless $smsg; - my $qtid = Search::Xapian::Query->new('G' . $smsg->thread_id); - my $path = $smsg->path; - if (defined $path && $path ne '') { - my $path = id_compress($smsg->path); - my $qsub = Search::Xapian::Query->new('XPATH' . $path); - $qtid = Search::Xapian::Query->new(OP_OR, $qtid, $qsub); - } - $opts ||= {}; - $opts->{limit} ||= 1000; - - # always sort threads by timestamp, this makes life easier - # for the threading algorithm (in SearchThread.pm) - $opts->{asc} = 1; - $opts->{enquire} = enquire_skel($self); - _do_enquire($self, $qtid, $opts); + my ($self, $mid, $prev) = @_; + $self->{over_ro}->get_thread($mid, $prev); } sub retry_reopen { my ($self, $cb) = @_; - my $ret; - for (1..10) { - eval { $ret = $cb->() }; - return $ret unless $@; + for my $i (1..10) { + if (wantarray) { + my @ret; + eval { @ret = $cb->() }; + return @ret unless $@; + } else { + my $ret; + eval { $ret = $cb->() }; + return $ret unless $@; + } # Exception: The revision being read has been discarded - # you should call Xapian::Database::reopen() if (ref($@) eq 'Search::Xapian::DatabaseModifiedError') { + warn "reopen try #$i on $@\n"; reopen($self); } else { warn "ref: ", ref($@), "\n"; die; } } + die "Too many Xapian database modifications in progress\n"; } sub _do_enquire { @@ -235,19 +219,16 @@ sub _do_enquire { sub _enquire_once { my ($self, $query, $opts) = @_; - my $enquire = $opts->{enquire} || enquire($self); - if (defined $query) { - $query = Search::Xapian::Query->new(OP_AND,$query,$mail_query); - } else { - $query = $mail_query; - } + my $enquire = enquire($self); $enquire->set_query($query); $opts ||= {}; my $desc = !$opts->{asc}; - if ($opts->{relevance}) { + if (($opts->{mset} || 0) == 2) { + $enquire->set_docid_order(Search::Xapian::ENQ_ASCENDING()); + $enquire->set_weighting_scheme(Search::Xapian::BoolWeight->new); + delete $self->{enquire}; + } elsif ($opts->{relevance}) { $enquire->set_sort_by_relevance_then_value(TS, $desc); - } elsif ($opts->{num}) { - $enquire->set_sort_by_value(NUM, 0); } else { $enquire->set_sort_by_value_then_relevance(TS, $desc); } @@ -258,8 +239,9 @@ sub _enquire_once { my @msgs = map { PublicInbox::SearchMsg->load_doc($_->get_document); } $mset->items; + return \@msgs unless wantarray; - { total => $mset->get_matches_estimated, msgs => \@msgs } + ($mset->get_matches_estimated, \@msgs) } # read-write @@ -280,6 +262,8 @@ sub qp { $qp->set_stemming_strategy(STEM_SOME); $qp->add_valuerangeprocessor( Search::Xapian::NumberValueRangeProcessor->new(YYYYMMDD, 'd:')); + $qp->add_valuerangeprocessor( + Search::Xapian::NumberValueRangeProcessor->new(DT, 'dt:')); while (my ($name, $prefix) = each %bool_pfx_external) { $qp->add_boolean_prefix($name, $prefix); @@ -309,94 +293,25 @@ EOF $self->{query_parser} = $qp; } -sub num_range_processor { - $_[0]->{nrp} ||= Search::Xapian::NumberValueRangeProcessor->new(NUM); -} - # only used for NNTP server sub query_xover { my ($self, $beg, $end, $offset) = @_; - my $qp = Search::Xapian::QueryParser->new; - $qp->set_database($self->{xdb}); - $qp->add_valuerangeprocessor($self->num_range_processor); - my $query = $qp->parse_query("$beg..$end", QP_FLAGS); - - _do_enquire($self, $query, {num => 1, limit => 200, offset => $offset}); + $self->{over_ro}->query_xover($beg, $end, $offset); } -sub lookup_skeleton { - my ($self, $mid) = @_; - my $skel = $self->{skel} or return lookup_message($self, $mid); - $mid = mid_clean($mid); - my $term = 'XMID' . $mid; - my $smsg; - my $beg = $skel->postlist_begin($term); - if ($beg != $skel->postlist_end($term)) { - my $doc_id = $beg->get_docid; - if (defined $doc_id) { - # raises on error: - my $doc = $skel->get_document($doc_id); - $smsg = PublicInbox::SearchMsg->wrap($doc, $mid); - $smsg->{doc_id} = $doc_id; - } - } - $smsg; -} - -sub lookup_message { - my ($self, $mid) = @_; - $mid = mid_clean($mid); - - my $doc_id = $self->find_first_doc_id('XMID' . $mid); - my $smsg; - if (defined $doc_id) { - # raises on error: - my $doc = $self->{xdb}->get_document($doc_id); - $smsg = PublicInbox::SearchMsg->wrap($doc, $mid); - $smsg->{doc_id} = $doc_id; - } - $smsg; +sub query_ts { + my ($self, $ts, $prev) = @_; + $self->{over_ro}->query_ts($ts, $prev); } -sub lookup_mail { # no ghosts! - my ($self, $mid) = @_; - retry_reopen($self, sub { - my $smsg = lookup_message($self, $mid) or return; - $smsg->load_expand; - }); +sub lookup_article { + my ($self, $num) = @_; + $self->{over_ro}->get_art($num); } -sub find_unique_doc_id { - my ($self, $termval) = @_; - - my ($begin, $end) = $self->find_doc_ids($termval); - - return undef if $begin->equal($end); # not found - - my $rv = $begin->get_docid; - - # sanity check - $begin->inc; - $begin->equal($end) or die "Term '$termval' is not unique\n"; - $rv; -} - -# returns begin and end PostingIterator -sub find_doc_ids { - my ($self, $termval) = @_; - my $db = $self->{xdb}; - - ($db->postlist_begin($termval), $db->postlist_end($termval)); -} - -sub find_first_doc_id { - my ($self, $termval) = @_; - - my ($begin, $end) = $self->find_doc_ids($termval); - - return undef if $begin->equal($end); # not found - - $begin->get_docid; +sub next_by_mid { + my ($self, $mid, $id, $prev) = @_; + $self->{over_ro}->next_by_mid($mid, $id, $prev); } # normalize subjects so they are suitable as pathnames for URLs @@ -423,15 +338,6 @@ sub enquire { $self->{enquire} ||= Search::Xapian::Enquire->new($self->{xdb}); } -sub enquire_skel { - my ($self) = @_; - if (my $skel = $self->{skel}) { - $self->{enquire_skel} ||= Search::Xapian::Enquire->new($skel); - } else { - enquire($self); - } -} - sub help { my ($self) = @_; $self->qp; # parse altids