lei q: auto-memoize remote messages into lei/store

This lets users avoid network traffic on subsequent searches at
the expense of local disk space.  --no-import-remote may be
specified to reverse this trade-off for users with little
storage.

3 files changed, 10 insertions, 3 deletions

diff --git a/lib/PublicInbox/LEI.pm b/lib/PublicInbox/LEI.pm
index 8cd95ac2..50665b3e 100644
--- a/lib/PublicInbox/LEI.pm
+++ b/lib/PublicInbox/LEI.pm
@@ -112,6 +112,7 @@ our %CMD = ( # sorted in order of importance/use:
         save-as=s output|mfolder|o=s format|f=s dedupe|d=s threads|t augment|a
         sort|s=s reverse|r offset=i remote! local! external! pretty
         include|I=s@ exclude=s@ only=s@ jobs|j=s globoff|g stdin|
+        import-remote!
         alert=s@ mua=s no-torsocks torsocks=s verbose|v+ quiet|q C=s@),
         PublicInbox::LeiQuery::curl_opt(), opt_dash('limit|n=i', '[0-9]+') ],
 
@@ -225,6 +226,7 @@ my %OPTDESC = (
                 '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'],
+'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 743fa3f7..b57d1cc5 100644
--- a/lib/PublicInbox/LeiQuery.pm
+++ b/lib/PublicInbox/LeiQuery.pm
@@ -51,6 +51,7 @@ sub lei_q {
         # we'll allow "--only $LOCATION --local"
         my $sto = $self->_lei_store(1);
         my $lse = $sto->search;
+        $sto->write_prepare($self) if $opt->{'import-remote'} //= 1;
         if ($opt->{'local'} //= scalar(@only) ? 0 : 1) {
                 $lxs->prepare_external($lse);
         }
diff --git a/lib/PublicInbox/LeiXSearch.pm b/lib/PublicInbox/LeiXSearch.pm
index c46aba3b..2d399653 100644
--- a/lib/PublicInbox/LeiXSearch.pm
+++ b/lib/PublicInbox/LeiXSearch.pm
@@ -189,8 +189,9 @@ sub query_mset { # non-parallel for non-"--threads" users
         $lei->{ovv}->ovv_atexit_child($lei);
 }
 
-sub each_eml { # callback for MboxReader->mboxrd
+sub each_remote_eml { # callback for MboxReader->mboxrd
         my ($eml, $self, $lei, $each_smsg) = @_;
+        $lei->{sto}->ipc_do('set_eml', $eml) if $lei->{sto}; # --import-remote
         my $smsg = bless {}, 'PublicInbox::Smsg';
         $smsg->populate($eml);
         $smsg->parse_references($eml, mids($eml));
@@ -244,14 +245,17 @@ sub query_remote_mboxrd {
                 my ($fh, $pid) = popen_rd($cmd, undef, $rdr);
                 $reap_curl = PublicInbox::OnDestroy->new($sigint_reap, $pid);
                 $fh = IO::Uncompress::Gunzip->new($fh);
-                PublicInbox::MboxReader->mboxrd($fh, \&each_eml, $self,
+                PublicInbox::MboxReader->mboxrd($fh, \&each_remote_eml, $self,
                                                 $lei, $each_smsg);
                 my $err = waitpid($pid, 0) == $pid ? undef
                                                 : "BUG: waitpid($cmd): $!";
                 @$reap_curl = (); # cancel OnDestroy
                 die $err if $err;
+                my $nr = $lei->{-nr_remote_eml};
+                if ($nr && $lei->{sto}) {
+                        my $wait = $lei->{sto}->ipc_do('done');
+                }
                 if ($? == 0) {
-                        my $nr = $lei->{-nr_remote_eml};
                         mset_progress($lei, $lei->{-current_url}, $nr, $nr);
                         next;
                 }