update copyright headers and email addresses

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

diff --git a/lib/PublicInbox/Search.pm b/lib/PublicInbox/Search.pm
index d3faaebbccbca2cf53c09261b8816e417eae5e6c..2b33b395e6340544fc0fa1ff142e9587f89235e1 100644 (file)
--- a/lib/PublicInbox/Search.pm
+++ b/lib/PublicInbox/Search.pm
@@ -1,11 +1,12 @@
-# Copyright (C) 2015, all contributors <meta@public-inbox.org>
+# Copyright (C) 2015 all contributors <meta@public-inbox.org>
 # License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt)
 # based on notmuch, but with no concept of folders, files or flags
 package PublicInbox::Search;
 use strict;
 use warnings;
-use PublicInbox::SearchMsg;
+use constant TS => 0;
 use Search::Xapian qw/:standard/;
+use PublicInbox::SearchMsg;
 use Email::MIME;
 use PublicInbox::MID qw/mid_clean mid_compress/;
 
@@ -15,7 +16,6 @@ our $REPLY_RE = qr/^re:\s+/i;
 our $LANG = 'english';
 
 use constant {
-       TS => 0,
        # SCHEMA_VERSION history
        # 0 - initial
        # 1 - subject_path is lower-cased
@@ -25,23 +25,30 @@ use constant {
        # 5 - subject_path drops trailing '.'
        # 6 - preserve References: order in document data
        # 7 - remove references and inreplyto terms
-       SCHEMA_VERSION => 7,
+       # 8 - remove redundant/unneeded document data
+       # 9 - disable Message-ID compression
+       SCHEMA_VERSION => 9,
+
+       # 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"
-       mid => 'Q', # uniQue id (Message-ID or mid_compress)
+       thread => 'G', # newsGroup (or similar entity - e.g. a web forum name)
 );
 
 my %bool_pfx_external = (
        path => 'XPATH',
-       thread => 'G', # newsGroup (or similar entity - e.g. a web forum name)
+       mid => 'Q', # uniQue id (Message-ID)
 );
 
 my %prob_prefix = (
        subject => 'S',
+       s => 'S', # for mairix compatibility
+       m => 'Q', # 'mid' is exact, 'm' can do partial
 );
 
 my %all_pfx = (%bool_pfx_internal, %bool_pfx_external, %prob_prefix);
@@ -73,17 +80,21 @@ sub reopen { $_[0]->{xdb}->reopen }
 # read-only
 sub query {
        my ($self, $query_string, $opts) = @_;
-       my $query = $self->qp->parse_query($query_string, QP_FLAGS);
+       my $query;
 
        $opts ||= {};
-       $opts->{relevance} = 1;
+       unless ($query_string eq '') {
+               $query = $self->qp->parse_query($query_string, QP_FLAGS);
+               $opts->{relevance} = 1 unless exists $opts->{relevance};
+       }
+
        $self->do_enquire($query, $opts);
 }
 
 sub get_subject_path {
        my ($self, $path, $opts) = @_;
-       my $query = $self->qp->parse_query("path:".mid_compress($path), 0);
-       $self->do_enquire($query, $opts);
+       my $q = Search::Xapian::Query->new(xpfx("path").mid_compress($path));
+       $self->do_enquire($q, $opts);
 }
 
 sub get_thread {
@@ -91,9 +102,9 @@ sub get_thread {
        my $smsg = eval { $self->lookup_message($mid) };
 
        return { total => 0, msgs => [] } unless $smsg;
-       my $qp = $self->qp;
-       my $qtid = $qp->parse_query('thread:'.$smsg->thread_id, 0);
-       my $qsub = $qp->parse_query('path:'.mid_compress($smsg->path), 0);
+       my $qtid = Search::Xapian::Query->new(xpfx('thread').$smsg->thread_id);
+       my $path = mid_compress($smsg->path);
+       my $qsub = Search::Xapian::Query->new(xpfx('path').$path);
        my $query = Search::Xapian::Query->new(OP_OR, $qtid, $qsub);
        $self->do_enquire($query, $opts);
 }
@@ -103,18 +114,22 @@ sub get_thread {
 sub do_enquire {
        my ($self, $query, $opts) = @_;
        my $enquire = $self->enquire;
-
-       $query = Search::Xapian::Query->new(OP_AND, $query, $mail_query);
+       if (defined $query) {
+               $query = Search::Xapian::Query->new(OP_AND,$query,$mail_query);
+       } else {
+               $query = $mail_query;
+       }
        $enquire->set_query($query);
        if ($opts->{relevance}) {
-               $enquire->set_sort_by_relevance_then_value(TS, 0);
+               $enquire->set_sort_by_relevance_then_value(TS, 1);
        } else {
-               $enquire->set_sort_by_value(TS, 0);
+               $enquire->set_sort_by_value_then_relevance(TS, 1);
        }
        $opts ||= {};
        my $offset = $opts->{offset} || 0;
        my $limit = $opts->{limit} || 50;
        my $mset = $enquire->get_mset($offset, $limit);
+       return $mset if $opts->{mset};
        my @msgs = map {
                PublicInbox::SearchMsg->load_doc($_->get_document);
        } $mset->items;
@@ -163,7 +178,6 @@ sub date_range_processor {
 sub lookup_message {
        my ($self, $mid) = @_;
        $mid = mid_clean($mid);
-       $mid = mid_compress($mid);
 
        my $doc_id = $self->find_unique_doc_id('mid', $mid);
        my $smsg;
@@ -255,4 +269,13 @@ sub enquire {
        $self->{enquire} ||= Search::Xapian::Enquire->new($self->{xdb});
 }
 
+sub mid_prefix {
+       my ($self, $mpfx) = @_;
+       my $query = eval { $self->qp->parse_query("m:$mpfx", FLAG_PARTIAL) };
+       return if $@;
+       my $res = $self->do_enquire($query, { relevance => 1 });
+       return unless $res->{total};
+       [ map { $_->mid } @{$res->{msgs}} ];
+}
+
 1;