X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FSearch.pm;h=cfa942b2f75a794f3ed356bf743b7799b86dca72;hb=80b887f29b2ec71d025b4c266a1c26314758994c;hp=01bbe73de2090b4bdbd539abb109edc33bb5fece;hpb=a367ec1b15a2458e532245f5308565dd84f8ca63;p=public-inbox.git diff --git a/lib/PublicInbox/Search.pm b/lib/PublicInbox/Search.pm index 01bbe73d..cfa942b2 100644 --- a/lib/PublicInbox/Search.pm +++ b/lib/PublicInbox/Search.pm @@ -54,19 +54,33 @@ use constant { use PublicInbox::Smsg; use PublicInbox::Over; my $QP_FLAGS; -our %X = map { $_ => 0 } qw(BoolWeight Database Enquire - NumberValueRangeProcessor QueryParser Stem); +our %X = map { $_ => 0 } qw(BoolWeight Database Enquire QueryParser Stem); our $Xap; # 'Search::Xapian' or 'Xapian' +my $NVRP; # '$Xap::'.('NumberValueRangeProcessor' or 'NumberRangeProcessor') my $ENQ_ASCENDING; sub load_xapian () { return 1 if defined $Xap; - for my $x (qw(Search::Xapian Xapian)) { + # n.b. PI_XAPIAN is intended for development use only. We still + # favor Search::Xapian since that's what's available in current + # Debian stable (10.x) and derived distros. + for my $x (($ENV{PI_XAPIAN} // 'Search::Xapian'), 'Xapian') { eval "require $x"; next if $@; $x->import(qw(:standard)); $Xap = $x; + + # `version_string' was added in Xapian 1.1 + my $xver = eval('v'.eval($x.'::version_string()')) // + eval('v'.eval($x.'::xapian_version_string()')); + + # NumberRangeProcessor was added in Xapian 1.3.6, + # NumberValueRangeProcessor was removed for 1.5.0+, + # favor the older /Value/ variant since that's what our + # (currently) preferred Search::Xapian supports + $NVRP = $x.'::'.($x eq 'Xapian' && $xver ge v1.5 ? + 'NumberRangeProcessor' : 'NumberValueRangeProcessor'); $X{$_} = $Xap.'::'.$_ for (keys %X); # ENQ_ASCENDING doesn't seem exported by SWIG Xapian.pm, @@ -265,18 +279,13 @@ sub reopen { } # read-only -sub query { +sub mset { my ($self, $query_string, $opts) = @_; $opts ||= {}; - if ($query_string eq '' && !$opts->{mset}) { - $self->{over_ro}->recent($opts); - } else { - my $qp = $self->{qp} //= qparse_new($self); - my $qp_flags = $self->{qp_flags}; - my $query = $qp->parse_query($query_string, $qp_flags); - $opts->{relevance} = 1 unless exists $opts->{relevance}; - _do_enquire($self, $query, $opts); - } + my $qp = $self->{qp} //= qparse_new($self); + my $query = $qp->parse_query($query_string, $self->{qp_flags}); + $opts->{relevance} = 1 unless exists $opts->{relevance}; + _do_enquire($self, $query, $opts); } sub retry_reopen { @@ -337,17 +346,17 @@ sub _enquire_once { # retry_reopen callback if ($opts->{thread} && has_threadid($self)) { $enquire->set_collapse_key(THREADID); } + $enquire->get_mset($opts->{offset} || 0, $opts->{limit} || 50); +} - my $offset = $opts->{offset} || 0; - my $limit = $opts->{limit} || 50; - my $mset = $enquire->get_mset($offset, $limit); - return $mset if $opts->{mset}; +sub mset_to_smsg { + my ($self, $ibx, $mset) = @_; my $nshard = $self->{nshard} // 1; my $i = 0; my %order = map { mdocid($nshard, $_) => ++$i } $mset->items; my @msgs = sort { $order{$a->{num}} <=> $order{$b->{num}} - } @{$self->{over_ro}->get_all(keys %order)}; + } @{$ibx->over->get_all(keys %order)}; wantarray ? ($mset->get_matches_estimated, \@msgs) : \@msgs; } @@ -364,15 +373,18 @@ sub qparse_new ($) { $qp->set_database($xdb); $qp->set_stemmer(stemmer($self)); $qp->set_stemming_strategy(STEM_SOME()); - $qp->set_max_wildcard_expansion(100); - my $nvrp = $X{NumberValueRangeProcessor}; - $qp->add_valuerangeprocessor($nvrp->new(YYYYMMDD, 'd:')); - $qp->add_valuerangeprocessor($nvrp->new(DT, 'dt:')); + my $cb = $qp->can('set_max_wildcard_expansion') // + $qp->can('set_max_expansion'); # Xapian 1.5.0+ + $cb->($qp, 100); + $cb = $qp->can('add_valuerangeprocessor') // + $qp->can('add_rangeprocessor'); # Xapian 1.5.0+ + $cb->($qp, $NVRP->new(YYYYMMDD, 'd:')); + $cb->($qp, $NVRP->new(DT, 'dt:')); # for IMAP, undocumented for WWW and may be split off go away - $qp->add_valuerangeprocessor($nvrp->new(BYTES, 'bytes:')); - $qp->add_valuerangeprocessor($nvrp->new(TS, 'ts:')); - $qp->add_valuerangeprocessor($nvrp->new(UID, 'uid:')); + $cb->($qp, $NVRP->new(BYTES, 'bytes:')); + $cb->($qp, $NVRP->new(TS, 'ts:')); + $cb->($qp, $NVRP->new(UID, 'uid:')); while (my ($name, $prefix) = each %bool_pfx_external) { $qp->add_boolean_prefix($name, $_) foreach split(/ /, $prefix);