[PATCH 4/5] lei q: start wiring up saved search

[Eric Wong <e@80x24.org>]

This will have a over.sqlite3 for content-based deduplication.
It may exhibit ibxish methods, so serving a read-only (or even
R/W) IMAP or instance or displaying HTML isn't outside the realm
of possibility.
---
 MANIFEST                          |   3 +
 lib/PublicInbox/LEI.pm            |   4 +-
 lib/PublicInbox/LeiQuery.pm       |   2 +
 lib/PublicInbox/LeiSavedSearch.pm | 134 ++++++++++++++++++++++++++++++
 lib/PublicInbox/LeiToMail.pm      |  10 ++-
 lib/PublicInbox/LeiXSearch.pm     |  31 ++++++-
 t/lei-q-save.t                    |  12 +++
 t/lei.t                           |   2 +-
 t/lei_saved_search.t              |  10 +++
 9 files changed, 199 insertions(+), 9 deletions(-)
 create mode 100644 lib/PublicInbox/LeiSavedSearch.pm
 create mode 100644 t/lei-q-save.t
 create mode 100644 t/lei_saved_search.t

diff --git a/MANIFEST b/MANIFEST
index 12247ad2..20615abc 100644
--- a/MANIFEST
+++ b/MANIFEST
@@ -201,6 +201,7 @@ lib/PublicInbox/LeiOverview.pm
 lib/PublicInbox/LeiP2q.pm
 lib/PublicInbox/LeiQuery.pm
 lib/PublicInbox/LeiRemote.pm
+lib/PublicInbox/LeiSavedSearch.pm
 lib/PublicInbox/LeiSearch.pm
 lib/PublicInbox/LeiStore.pm
 lib/PublicInbox/LeiStoreErr.pm
@@ -393,12 +394,14 @@ t/lei-mirror.t
 t/lei-p2q.t
 t/lei-q-kw.t
 t/lei-q-remote-import.t
+t/lei-q-save.t
 t/lei-q-thread.t
 t/lei-tag.t
 t/lei.t
 t/lei_dedupe.t
 t/lei_external.t
 t/lei_overview.t
+t/lei_saved_search.t
 t/lei_store.t
 t/lei_to_mail.t
 t/lei_xsearch.t
diff --git a/lib/PublicInbox/LEI.pm b/lib/PublicInbox/LEI.pm
index 475af8f0..7292d0f2 100644
--- a/lib/PublicInbox/LEI.pm
+++ b/lib/PublicInbox/LEI.pm
@@ -131,7 +131,7 @@ our %CMD = ( # sorted in order of importance/use:
 'q' => [ '--stdin|SEARCH_TERMS...', 'search for messages matching terms',
 	'stdin|', # /|\z/ must be first for lone dash
 	@lxs_opt,
-	qw(save-as=s output|mfolder|o=s format|f=s dedupe|d=s threads|t+
+	qw(save output|mfolder|o=s format|f=s dedupe|d=s threads|t+
 	sort|s=s reverse|r offset=i pretty jobs|j=s globoff|g augment|a
 	import-before! lock=s@ rsyncable alert=s@ mua=s verbose|v+), @c_opt,
 	opt_dash('limit|n=i', '[0-9]+') ],
@@ -249,7 +249,7 @@ my %OPTDESC = (
 'torsocks=s' => ['VAL|auto|no|yes',
 		'whether or not to wrap git and curl commands with torsocks'],
 'no-torsocks' => 'alias for --torsocks=no',
-'save-as=s' => ['NAME', 'save a search terms by given name'],
+'save' =>  "save a search for `lei up'",
 'import-remote!' => 'do not memoize remote messages into local store',
 
 'type=s' => [ 'any|mid|git', 'disambiguate type' ],
diff --git a/lib/PublicInbox/LeiQuery.pm b/lib/PublicInbox/LeiQuery.pm
index 224eba69..8bca1020 100644
--- a/lib/PublicInbox/LeiQuery.pm
+++ b/lib/PublicInbox/LeiQuery.pm
@@ -53,6 +53,7 @@ sub qstr_add { # PublicInbox::InputPipe::consume callback for --stdin
 	my ($self) = @_; # $_[1] = $rbuf
 	if (defined($_[1])) {
 		$_[1] eq '' and return eval {
+			$self->{mset_opt}->{q_raw} = $self->{mset_opt}->{qstr};
 			$self->{lse}->query_approxidate($self->{lse}->git,
 						$self->{mset_opt}->{qstr});
 			_start_query($self);
@@ -142,6 +143,7 @@ no query allowed on command-line with --stdin
 		PublicInbox::InputPipe::consume($self->{0}, \&qstr_add, $self);
 		return;
 	}
+	$mset_opt{q_raw} = \@argv;
 	$mset_opt{qstr} =
 		$self->{lse}->query_argv_to_string($self->{lse}->git, \@argv);
 	_start_query($self);
diff --git a/lib/PublicInbox/LeiSavedSearch.pm b/lib/PublicInbox/LeiSavedSearch.pm
new file mode 100644
index 00000000..ab9f393b
--- /dev/null
+++ b/lib/PublicInbox/LeiSavedSearch.pm
@@ -0,0 +1,134 @@
+# Copyright (C) 2021 all contributors <meta@public-inbox.org>
+# License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
+
+# pretends to be like LeiDedupe and also PublicInbox::Inbox
+package PublicInbox::LeiSavedSearch;
+use strict;
+use v5.10.1;
+use parent qw(PublicInbox::Lock);
+use PublicInbox::OverIdx;
+use PublicInbox::LeiSearch;
+use PublicInbox::Config;
+use PublicInbox::Spawn qw(run_die);
+use PublicInbox::ContentHash qw(content_hash git_sha);
+use PublicInbox::Eml;
+use PublicInbox::Hval qw(to_filename);
+
+sub new {
+	my ($cls, $lei, $dir) = @_;
+	my $self = bless { ale => $lei->ale, -cfg => {} }, $cls;
+	if (defined $dir) { # updating existing saved search
+		my $f = $self->{'-f'} = "$dir/lei.saved-search";
+		-f $f && -r _ or
+			return $lei->fail("$f non-existent or unreadable");
+		$self->{-cfg} = PublicInbox::Config::git_config_dump($f);
+		my $q = $lei->{mset_opt}->{q_raw} = $self->{-cfg}->{'lei.q'} //
+					return $lei->fail("lei.q unset in $f");
+		my $lse = $lei->{lse} // die 'BUG: {lse} missing';
+		$lei->{mset_opt}->{qstr} = ref($q) ?
+				$lse->query_argv_to_string($lse->git, $q) :
+				$lse->query_approxidate($lse->git, $q);
+	} else { # new saved search
+		my $saved_dir = $lei->store_path . '/../saved-searches/';
+		my (@name) = ($lei->{ovv}->{dst} =~ m{([\w\-\.]+)/*\z});
+		push @name, to_filename($lei->{mset_opt}->{qstr});
+		$dir = $saved_dir . join('-', @name);
+		require File::Path;
+		File::Path::make_path($dir); # raises on error
+		$self->{'-f'} = "$dir/lei.saved-search";
+		my $q = $lei->{mset_opt}->{q_raw};
+		if (ref $q) {
+			cfg_set($self, '--add', 'lei.q', $_) for @$q;
+		} else {
+			cfg_set($self, 'lei.q', $q);
+		}
+	}
+	bless $self->{-cfg}, 'PublicInbox::Config';
+	$self->{lock_path} = "$self->{-f}.flock";
+	$self->{-ovf} = "$dir/over.sqlite3";
+	$self;
+}
+
+sub description { $_[0]->{qstr} } # for WWW
+
+sub cfg_set {
+	my ($self, @args) = @_;
+	my $lk = $self->lock_for_scope; # git-config doesn't wait
+	run_die([qw(git config -f), $self->{'-f'}, @args]);
+}
+
+# drop-in for LeiDedupe API
+sub is_dup {
+	my ($self, $eml, $smsg) = @_;
+	my $oidx = $self->{oidx} // die 'BUG: no {oidx}';
+	my $blob = $smsg ? $smsg->{blob} : undef;
+	return 1 if $blob && $oidx->blob_exists($blob);
+	my $lk = $self->lock_for_scope_fast;
+	if (my $xoids = PublicInbox::LeiSearch::xoids_for($self, $eml, 1)) {
+		for my $docid (values %$xoids) {
+			$oidx->add_xref3($docid, -1, $blob, '.');
+		}
+		$oidx->commit_lazy;
+		1;
+	} else {
+		# n.b. above xoids_for fills out eml->{-lei_fake_mid} if needed
+		unless ($smsg) {
+			$smsg = bless {}, 'PublicInbox::Smsg';
+			$smsg->{bytes} = 0;
+			$smsg->populate($eml);
+		}
+		$oidx->begin_lazy;
+		$smsg->{num} = $oidx->adj_counter('eidx_docid', '+');
+		$smsg->{blob} //= git_sha(1, $eml)->hexdigest;
+		$oidx->add_overview($eml, $smsg);
+		$oidx->add_xref3($smsg->{num}, -1, $smsg->{blob}, '.');
+		$oidx->commit_lazy;
+		undef;
+	}
+}
+
+sub prepare_dedupe {
+	my ($self) = @_;
+	$self->{oidx} //= do {
+		my $creat = !-f $self->{-ovf};
+		my $lk = $self->lock_for_scope; # git-config doesn't wait
+		my $oidx = PublicInbox::OverIdx->new($self->{-ovf});
+		$oidx->{-no_fsync} = 1;
+		$oidx->dbh;
+		if ($creat) {
+			$oidx->{dbh}->do('PRAGMA journal_mode = WAL');
+			$oidx->eidx_prep; # for xref3
+		}
+		$oidx
+	};
+}
+
+sub over { $_[0]->{oidx} } # for xoids_for
+
+sub git { $_[0]->{ale}->git }
+
+sub pause_dedupe {
+	my ($self) = @_;
+	$self->{ale}->git->cleanup;
+	my $oidx = delete($self->{oidx}) // return;
+	$oidx->commit_lazy;
+}
+
+sub mm { undef }
+
+sub altid_map { {} }
+
+sub cloneurl { [] }
+no warnings 'once';
+*nntp_url = \&cloneurl;
+*base_url = \&PublicInbox::Inbox::base_url;
+*smsg_eml = \&PublicInbox::Inbox::smsg_eml;
+*smsg_by_mid = \&PublicInbox::Inbox::smsg_by_mid;
+*msg_by_mid = \&PublicInbox::Inbox::msg_by_mid;
+*modified = \&PublicInbox::Inbox::modified;
+*recent = \&PublicInbox::Inbox::recent;
+*max_git_epoch = *nntp_usable = *msg_by_path = \&mm; # undef
+*isrch = *search = \&mm; # TODO
+*DESTROY = \&pause_dedupe;
+
+1;
diff --git a/lib/PublicInbox/LeiToMail.pm b/lib/PublicInbox/LeiToMail.pm
index 7adbffe7..bd2b714a 100644
--- a/lib/PublicInbox/LeiToMail.pm
+++ b/lib/PublicInbox/LeiToMail.pm
@@ -9,7 +9,6 @@ use parent qw(PublicInbox::IPC);
 use PublicInbox::Eml;
 use PublicInbox::ProcessPipe;
 use PublicInbox::Spawn qw(spawn);
-use PublicInbox::LeiDedupe;
 use PublicInbox::PktOp qw(pkt_do);
 use Symbol qw(gensym);
 use IO::Handle; # ->autoflush
@@ -350,7 +349,11 @@ sub new {
 		die "bad mail --format=$fmt\n";
 	}
 	$self->{dst} = $dst;
-	$lei->{dedupe} = PublicInbox::LeiDedupe->new($lei);
+	my $dd_cls = 'PublicInbox::'.
+		($lei->{opt}->{save} ? 'LeiSavedSearch' : 'LeiDedupe');
+	eval "require $dd_cls";
+	die "$dd_cls: $@" if $@;
+	$lei->{dedupe} = $dd_cls->new($lei);
 	$self;
 }
 
@@ -368,6 +371,7 @@ sub _pre_augment_maildir {
 
 sub _do_augment_maildir {
 	my ($self, $lei) = @_;
+	return if defined($lei->{opt}->{save});
 	my $dst = $lei->{ovv}->{dst};
 	my $lse = $lei->{opt}->{'import-before'} ? $lei->{lse} : undef;
 	my $mdr = PublicInbox::MdirReader->new;
@@ -398,6 +402,7 @@ sub _imap_augment_or_delete { # PublicInbox::NetReader::imap_each cb
 
 sub _do_augment_imap {
 	my ($self, $lei) = @_;
+	return if defined($lei->{opt}->{save});
 	my $net = $lei->{net};
 	my $lse = $lei->{opt}->{'import-before'} ? $lei->{lse} : undef;
 	if ($lei->{opt}->{augment}) {
@@ -468,6 +473,7 @@ sub _do_augment_mbox {
 	my ($self, $lei) = @_;
 	return unless $self->{seekable};
 	my $opt = $lei->{opt};
+	return if defined($opt->{save});
 	my $out = $lei->{1};
 	my ($fmt, $dst) = @{$lei->{ovv}}{qw(fmt dst)};
 	return unless -s $out;
diff --git a/lib/PublicInbox/LeiXSearch.pm b/lib/PublicInbox/LeiXSearch.pm
index 9d367977..7c540c1c 100644
--- a/lib/PublicInbox/LeiXSearch.pm
+++ b/lib/PublicInbox/LeiXSearch.pm
@@ -149,22 +149,38 @@ sub query_one_mset { # for --threads and l2m w/o sort
 	local $0 = "$0 query_one_mset";
 	my $lei = $self->{lei};
 	my ($srch, $over) = ($ibxish->search, $ibxish->over);
-	my $desc = $ibxish->{inboxdir} // $ibxish->{topdir};
-	return warn("$desc not indexed by Xapian\n") unless ($srch && $over);
-	my $mo = { %{$lei->{mset_opt}} };
+	my $dir = $ibxish->{inboxdir} // $ibxish->{topdir};
+	return warn("$dir not indexed by Xapian\n") unless ($srch && $over);
+	my $mo = { %{$lei->{mset_opt}} }; # copy
 	my $mset;
 	my $each_smsg = $lei->{ovv}->ovv_each_smsg_cb($lei);
 	my $can_kw = !!$ibxish->can('msg_keywords');
 	my $threads = $lei->{opt}->{threads} // 0;
 	my $fl = $threads > 1 ? 1 : undef;
+	my $lss = $lei->{dedupe};
+	$lss = undef unless $lss && $lss->can('cfg_set'); # saved search
+	my $maxk = "external.$dir.maxnum";
+	my $stop_at = $lss ? $lss->{-cfg}->{$maxk} : undef;
+	if (defined $stop_at) {
+		die "$maxk=$stop_at has multiple values" if ref $stop_at;
+		my @e;
+		local $SIG{__WARN__} = sub { push @e, @_ };
+		$stop_at += 0;
+		return warn("$maxk=$stop_at: @e") if @e;
+	}
+	my $first_ids;
 	do {
 		$mset = $srch->mset($mo->{qstr}, $mo);
-		mset_progress($lei, $desc, $mset->size,
+		mset_progress($lei, $dir, $mset->size,
 				$mset->get_matches_estimated);
 		wait_startq($lei); # wait for keyword updates
 		my $ids = $srch->mset_to_artnums($mset, $mo);
+		@$ids = grep { $_ > $stop_at } @$ids if defined($stop_at);
 		my $i = 0;
 		if ($threads) {
+			# copy $ids if $lss since over->expand_thread
+			# shifts @{$ctx->{ids}}
+			$first_ids = [ @$ids ] if $lss;
 			my $ctx = { ids => $ids };
 			my %n2item = map { ($ids->[$i++], $_) } $mset->items;
 			while ($over->expand_thread($ctx)) {
@@ -183,6 +199,7 @@ sub query_one_mset { # for --threads and l2m w/o sort
 				@{$ctx->{xids}} = ();
 			}
 		} else {
+			$first_ids = $ids;
 			my @items = $mset->items;
 			for my $n (@$ids) {
 				my $mitem = $items[$i++];
@@ -193,6 +210,12 @@ sub query_one_mset { # for --threads and l2m w/o sort
 			}
 		}
 	} while (_mset_more($mset, $mo));
+	if ($lss && scalar(@$first_ids)) {
+		undef $stop_at;
+		my $max = $first_ids->[0];
+		$lss->cfg_set($maxk, $max);
+		undef $lss;
+	}
 	undef $each_smsg; # may commit
 	$lei->{ovv}->ovv_atexit_child($lei);
 }
diff --git a/t/lei-q-save.t b/t/lei-q-save.t
new file mode 100644
index 00000000..56f7cb37
--- /dev/null
+++ b/t/lei-q-save.t
@@ -0,0 +1,12 @@
+#!perl -w
+# Copyright (C) 2021 all contributors <meta@public-inbox.org>
+# License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
+use strict; use v5.10.1; use PublicInbox::TestCommon;
+test_lei(sub {
+	my $home = $ENV{HOME};
+	lei_ok qw(import t/plack-qp.eml);
+	lei_ok qw(q --save z:0..), '-o', "$home/md/";
+	my @s = glob("$home/.local/share/lei/saved-searches/md-*");
+	is(scalar(@s), 1, 'got one saved search');
+});
+done_testing;
diff --git a/t/lei.t b/t/lei.t
index 2be9b4e8..6ade2f18 100644
--- a/t/lei.t
+++ b/t/lei.t
@@ -114,7 +114,7 @@ my $test_completion = sub {
 	%out = map { $_ => 1 } split(/\s+/s, $lei_out);
 	for my $sw (qw(-f --format -o --output --mfolder --augment -a
 			--mua --no-local --local --verbose -v
-			--save-as --no-remote --remote --torsocks
+			--save --no-remote --remote --torsocks
 			--reverse -r )) {
 		ok($out{$sw}, "$sw offered as `lei q' completion");
 	}
diff --git a/t/lei_saved_search.t b/t/lei_saved_search.t
new file mode 100644
index 00000000..6d26cd2b
--- /dev/null
+++ b/t/lei_saved_search.t
@@ -0,0 +1,10 @@
+#!perl -w
+# Copyright (C) 2021 all contributors <meta@public-inbox.org>
+# License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
+use strict;
+use v5.10.1;
+use PublicInbox::TestCommon;
+require_mods(qw(DBD::SQLite));
+use_ok 'PublicInbox::LeiSavedSearch';
+
+done_testing;