X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FSearch.pm;h=5c9dccb53af327282d10df960f7f112801986e58;hb=55b707d788ce13696e4411389583e720ea6dab01;hp=6f5e0624a4a5986369f12de95d81f6fe3a01fbf4;hpb=417612b7b7bb9750f3c91135e60aa8bbdd461496;p=public-inbox.git

diff --git a/lib/PublicInbox/Search.pm b/lib/PublicInbox/Search.pm
index 6f5e0624..5c9dccb5 100644
--- a/lib/PublicInbox/Search.pm
+++ b/lib/PublicInbox/Search.pm
@@ -1,4 +1,4 @@
-# Copyright (C) 2015-2018 all contributors <meta@public-inbox.org>
+# Copyright (C) 2015-2019 all contributors <meta@public-inbox.org>
 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
 # based on notmuch, but with no concept of folders, files or flags
 #
@@ -8,21 +8,52 @@ use strict;
 use warnings;
 
 # values for searching
-use constant DS => 0; # Date: header in Unix time
-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 TS => 4;  # Received: header in Unix time
-use constant YYYYMMDD => 5; # for searching in the WWW UI
-
-use Search::Xapian qw/:standard/;
+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 PublicInbox::SearchMsg;
-use PublicInbox::MIME;
-use PublicInbox::MID qw/id_compress/;
+use PublicInbox::Over;
+my $QP_FLAGS;
+our %X = map { $_ => 0 } qw(BoolWeight Database Enquire
+			NumberValueRangeProcessor QueryParser Stem);
+our $Xap; # 'Search::Xapian' or 'Xapian'
+my $ENQ_ASCENDING;
+
+sub load_xapian () {
+	return 1 if defined $Xap;
+	for my $x (qw(Search::Xapian Xapian)) {
+		eval "require $x";
+		next if $@;
+
+		$x->import(qw(:standard));
+		$Xap = $x;
+		$X{$_} = $Xap.'::'.$_ for (keys %X);
+
+		# ENQ_ASCENDING doesn't seem exported by SWIG Xapian.pm,
+		# so lets hope this part of the ABI is stable because it's
+		# just an integer:
+		$ENQ_ASCENDING = $x eq 'Xapian' ?
+				1 : Search::Xapian::ENQ_ASCENDING();
+
+		# for SearchMsg:
+		*PublicInbox::SearchMsg::sortable_unserialise =
+						$Xap.'::sortable_unserialise';
+		# n.b. FLAG_PURE_NOT is expensive not suitable for a public
+		# website as it could become a denial-of-service vector
+		# FLAG_PHRASE also seems to cause performance problems chert
+		# (and probably earlier Xapian DBs).  glass seems fine...
+		# TODO: make this an option, maybe?
+		# or make indexlevel=medium as default
+		$QP_FLAGS = FLAG_PHRASE() | FLAG_BOOLEAN() | FLAG_LOVEHATE() |
+				FLAG_WILDCARD();
+		return 1;
+	}
+	undef;
+}
 
 # This is English-only, everything else is non-standard and may be confused as
 # a prefix common in patch emails
-our $REPLY_RE = qr/^re:\s+/i;
 our $LANG = 'english';
 
 use constant {
@@ -43,21 +74,17 @@ use constant {
 	# 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
-	QP_FLAGS => FLAG_PHRASE|FLAG_BOOLEAN|FLAG_LOVEHATE|FLAG_WILDCARD,
+	# 15 - see public-inbox-v2-format(5)
+	#      further bumps likely unnecessary, we'll suggest in-place
+	#      "--reindex" use for further fixes and tweaks
+	SCHEMA_VERSION => 15,
 };
 
-# 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 => 'Q', # Message-ID (full/exact), this is mostly uniQue
+	dfpre => 'XDFPRE',
+	dfpost => 'XDFPOST',
+	dfblob => 'XDFPRE XDFPOST',
 );
 
 my $non_quoted_body = 'XNQ XDFN XDFA XDFB XDFHH XDFCTX XDFPRE XDFPOST';
@@ -82,9 +109,6 @@ my %prob_prefix = (
 	dfb => 'XDFB',
 	dfhh => 'XDFHH',
 	dfctx => 'XDFCTX',
-	dfpre => 'XDFPRE',
-	dfpost => 'XDFPOST',
-	dfblob => 'XDFPRE XDFPOST',
 
 	# default:
 	'' => 'XM S A XQUOT XFN ' . $non_quoted_body,
@@ -97,6 +121,9 @@ 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:' => <<EOF,
+date-time range as YYYYMMDDhhmmss (e.g. dt:19931002011000..19931002011200)
 EOF
 	'b:' => 'match within message body, including text attachments',
 	'nq:' => 'match non-quoted text within message body',
@@ -119,60 +146,69 @@ EOF
 );
 chomp @HELP;
 
-my $mail_query = Search::Xapian::Query->new('T' . 'mail');
-
-sub xdir {
-	my ($self) = @_;
+sub xdir ($;$) {
+	my ($self, $rdonly) = @_;
 	if ($self->{version} == 1) {
-		"$self->{mainrepo}/public-inbox/xapian" . SCHEMA_VERSION;
+		"$self->{inboxdir}/public-inbox/xapian" . SCHEMA_VERSION;
 	} else {
-		my $dir = "$self->{mainrepo}/xap" . SCHEMA_VERSION;
-		my $part = $self->{partition};
-		defined $part or die "partition not given";
-		$dir .= "/$part";
+		my $dir = "$self->{inboxdir}/xap" . SCHEMA_VERSION;
+		return $dir if $rdonly;
+
+		my $shard = $self->{shard};
+		defined $shard or die "shard not given";
+		$dir .= "/$shard";
 	}
 }
 
-sub new {
-	my ($class, $mainrepo, $altid) = @_;
-	my $version = 1;
-	my $ibx = $mainrepo;
-	if (ref $ibx) {
-		$version = $ibx->{version} || 1;
-		$mainrepo = $ibx->{mainrepo};
-	}
-	my $self = bless {
-		mainrepo => $mainrepo,
-		altid => $altid,
-		version => $version,
-	}, $class;
-	if ($version >= 2) {
-		my $dir = "$self->{mainrepo}/xap" . SCHEMA_VERSION;
-		my $xdb;
-		my $parts = 0;
-		foreach my $part (<$dir/*>) {
-			-d $part && $part =~ m!/\d+\z! or next;
-			$parts++;
-			my $sub = Search::Xapian::Database->new($part);
+sub _xdb ($) {
+	my ($self) = @_;
+	my $dir = xdir($self, 1);
+	my ($xdb, $slow_phrase);
+	my $qpf = \($self->{qp_flags} ||= $QP_FLAGS);
+	if ($self->{version} >= 2) {
+		foreach my $shard (<$dir/*>) {
+			-d $shard && $shard =~ m!/[0-9]+\z! or next;
+			my $sub = $X{Database}->new($shard);
 			if ($xdb) {
 				$xdb->add_database($sub);
 			} else {
 				$xdb = $sub;
 			}
+			$slow_phrase ||= -f "$shard/iamchert";
 		}
-		$self->{xdb} = $xdb;
-		$self->{skel} = Search::Xapian::Database->new("$dir/skel");
 	} else {
-		$self->{xdb} = Search::Xapian::Database->new($self->xdir);
+		$slow_phrase = -f "$dir/iamchert";
+		$xdb = $X{Database}->new($dir);
 	}
+	$$qpf |= FLAG_PHRASE() unless $slow_phrase;
+	$xdb;
+}
+
+sub xdb ($) {
+	my ($self) = @_;
+	$self->{xdb} ||= do {
+		load_xapian();
+		_xdb($self);
+	};
+}
+
+sub new {
+	my ($class, $ibx) = @_;
+	ref $ibx or die "BUG: expected PublicInbox::Inbox object: $ibx";
+	my $self = bless {
+		inboxdir => $ibx->{inboxdir},
+		altid => $ibx->{altid},
+		version => $ibx->{version} // 1,
+	}, $class;
+	my $dir = xdir($self, 1);
+	$self->{over_ro} = PublicInbox::Over->new("$dir/over.sqlite3");
 	$self;
 }
 
 sub reopen {
 	my ($self) = @_;
-	$self->{xdb}->reopen;
-	if (my $skel = $self->{skel}) {
-		$skel->reopen;
+	if (my $xdb = $self->{xdb}) {
+		$xdb->reopen;
 	}
 	$self; # make chaining easier
 }
@@ -180,75 +216,62 @@ sub reopen {
 # read-only
 sub query {
 	my ($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 $qp = qp($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, $opts);
 	}
-
-	_do_enquire($self, $query, $opts);
-}
-
-sub get_thread {
-	my ($self, $mid, $opts) = @_;
-	my $smsg = first_smsg_by_mid($self, $mid) or
-			return { total => 0, msgs => [] };
-	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);
 }
 
 sub retry_reopen {
-	my ($self, $cb) = @_;
-	my $ret;
-	for (1..10) {
-		eval { $ret = $cb->() };
-		return $ret unless $@;
+	my ($self, $cb, $arg) = @_;
+	for my $i (1..10) {
+		if (wantarray) {
+			my @ret;
+			eval { @ret = $cb->($arg) };
+			return @ret unless $@;
+		} else {
+			my $ret;
+			eval { $ret = $cb->($arg) };
+			return $ret unless $@;
+		}
 		# Exception: The revision being read has been discarded -
 		# you should call Xapian::Database::reopen()
-		if (ref($@) eq 'Search::Xapian::DatabaseModifiedError') {
+		if (ref($@) =~ /\bDatabaseModifiedError\b/) {
+			warn "reopen try #$i on $@\n";
 			reopen($self);
 		} else {
-			warn "ref: ", ref($@), "\n";
+			# let caller decide how to spew, because ExtMsg queries
+			# get wonky and trigger:
+			# "something terrible happened at .../Xapian/Enquire.pm"
 			die;
 		}
 	}
+	die "Too many Xapian database modifications in progress\n";
 }
 
 sub _do_enquire {
 	my ($self, $query, $opts) = @_;
-	retry_reopen($self, sub { _enquire_once($self, $query, $opts) });
+	retry_reopen($self, \&_enquire_once, [ $self, $query, $opts ]);
 }
 
-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 = $mail_query;
-	}
+sub _enquire_once { # retry_reopen callback
+	my ($self, $query, $opts) = @{$_[0]};
+	my $xdb = xdb($self);
+	my $enquire = $X{Enquire}->new($xdb);
 	$enquire->set_query($query);
 	$opts ||= {};
         my $desc = !$opts->{asc};
-	if ($opts->{relevance}) {
+	if (($opts->{mset} || 0) == 2) {
+		$enquire->set_docid_order($ENQ_ASCENDING);
+		$enquire->set_weighting_scheme($X{BoolWeight}->new);
+	} 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);
 	}
@@ -256,15 +279,14 @@ sub _enquire_once {
 	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);
-	} $mset->items;
+	my @msgs = map { PublicInbox::SearchMsg::from_mitem($_) } $mset->items;
+	return \@msgs unless wantarray;
 
-	{ total => $mset->get_matches_estimated, msgs => \@msgs }
+	($mset->get_matches_estimated, \@msgs)
 }
 
 # read-write
-sub stemmer { Search::Xapian::Stem->new($LANG) }
+sub stemmer { $X{Stem}->new($LANG) }
 
 # read-only
 sub qp {
@@ -272,18 +294,20 @@ sub qp {
 
 	my $qp = $self->{query_parser};
 	return $qp if $qp;
-
+	my $xdb = xdb($self);
 	# new parser
-	$qp = Search::Xapian::QueryParser->new;
-	$qp->set_default_op(OP_AND);
-	$qp->set_database($self->{xdb});
+	$qp = $X{QueryParser}->new;
+	$qp->set_default_op(OP_AND());
+	$qp->set_database($xdb);
 	$qp->set_stemmer($self->stemmer);
-	$qp->set_stemming_strategy(STEM_SOME);
-	$qp->add_valuerangeprocessor(
-		Search::Xapian::NumberValueRangeProcessor->new(YYYYMMDD, 'd:'));
+	$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:'));
 
 	while (my ($name, $prefix) = each %bool_pfx_external) {
-		$qp->add_boolean_prefix($name, $prefix);
+		$qp->add_boolean_prefix($name, $_) foreach split(/ /, $prefix);
 	}
 
 	# we do not actually create AltId objects,
@@ -310,132 +334,6 @@ 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->{skel} || $self->{xdb});
-	$qp->add_valuerangeprocessor($self->num_range_processor);
-	my $query = $qp->parse_query("$beg..$end", QP_FLAGS);
-
-	my $opts = {
-		enquire => enquire_skel($self),
-		num => 1,
-		limit => 200,
-		offset => $offset,
-	};
-	_do_enquire($self, $query, $opts);
-}
-
-sub query_ts {
-	my ($self, $ts, $opts) = @_;
-	my $qp = $self->{qp_ts} ||= eval {
-		my $q = Search::Xapian::QueryParser->new;
-		$q->set_database($self->{skel} || $self->{xdb});
-		$q->add_valuerangeprocessor(
-			Search::Xapian::NumberValueRangeProcessor->new(TS));
-		$q
-	};
-	my $query = $qp->parse_query($ts, QP_FLAGS);
-	$opts->{enquire} = enquire_skel($self);
-	_do_enquire($self, $query, $opts);
-}
-
-sub first_smsg_by_mid {
-	my ($self, $mid) = @_;
-	my $smsg;
-	each_smsg_by_mid($self, $mid, sub { $smsg = $_[0]; undef });
-	$smsg;
-}
-
-sub lookup_article {
-	my ($self, $num) = @_;
-	my $term = 'XNUM'.$num;
-	my $smsg;
-	eval {
-		retry_reopen($self, sub {
-			my $db = $self->{skel} || $self->{xdb};
-			my $head = $db->postlist_begin($term);
-			my $tail = $db->postlist_end($term);
-			return if $head->equal($tail);
-			my $doc_id = $head->get_docid;
-			return unless defined $doc_id;
-			$head->inc;
-			if ($head->nequal($tail)) {
-				my $loc= $self->{mainrepo} .
-					($self->{skel} ? 'skel' : 'xdb');
-				warn "article #$num is not unique in $loc\n";
-			}
-			# raises on error:
-			my $doc = $db->get_document($doc_id);
-			$smsg = PublicInbox::SearchMsg->wrap($doc);
-			$smsg->load_expand;
-			$smsg->{doc_id} = $doc_id;
-		});
-	};
-	$smsg;
-}
-
-sub each_smsg_by_mid {
-	my ($self, $mid, $cb) = @_;
-	# XXX retry_reopen isn't necessary for V2Writable, but the PSGI
-	# interface will need it...
-	my $db = $self->{skel} || $self->{xdb};
-	my $term = 'Q' . $mid;
-	my $head = $db->postlist_begin($term);
-	my $tail = $db->postlist_end($term);
-	if ($head == $tail) {
-		$db->reopen;
-		$head = $db->postlist_begin($term);
-		$tail = $db->postlist_end($term);
-	}
-	return ($head, $tail, $db) if wantarray;
-	for (; $head->nequal($tail); $head->inc) {
-		my $doc_id = $head->get_docid;
-		my $doc = $db->get_document($doc_id);
-		my $smsg = PublicInbox::SearchMsg->wrap($doc, $mid);
-		$smsg->{doc_id} = $doc_id;
-		$cb->($smsg) or return;
-	}
-}
-
-# normalize subjects so they are suitable as pathnames for URLs
-# XXX: consider for removal
-sub subject_path {
-	my $subj = pop;
-	$subj = subject_normalized($subj);
-	$subj =~ s![^a-zA-Z0-9_\.~/\-]+!_!g;
-	lc($subj);
-}
-
-sub subject_normalized {
-	my $subj = pop;
-	$subj =~ s/\A\s+//s; # no leading space
-	$subj =~ s/\s+\z//s; # no trailing space
-	$subj =~ s/\s+/ /gs; # no redundant spaces
-	$subj =~ s/\.+\z//; # no trailing '.'
-	$subj =~ s/$REPLY_RE//igo; # remove reply prefix
-	$subj;
-}
-
-sub enquire {
-	my ($self) = @_;
-	$self->{enquire} ||= Search::Xapian::Enquire->new($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