wwwstream: subclass off GzipFilter

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

diff --git a/lib/PublicInbox/SearchIdx.pm b/lib/PublicInbox/SearchIdx.pm
index a790ac4076acefbd6b50b68ae706fc2fe0ae40fc..4caa66d3751d175f1a22e6685b173b66e800dec6 100644 (file)
--- a/lib/PublicInbox/SearchIdx.pm
+++ b/lib/PublicInbox/SearchIdx.pm
@@ -78,12 +78,12 @@ sub new {
 sub need_xapian ($) { $_[0]->{indexlevel} =~ $xapianlevels }
 
 sub _xdb_release {
-       my ($self) = @_;
+       my ($self, $wake) = @_;
        if (need_xapian($self)) {
                my $xdb = delete $self->{xdb} or croak 'not acquired';
                $xdb->close;
        }
-       $self->lock_release if $self->{creat};
+       $self->lock_release($wake) if $self->{creat};
        undef;
 }
 
@@ -549,11 +549,23 @@ sub unindex_mm {
        $self->{mm}->mid_delete(mid_mime($mime));
 }
 
+# returns the number of bytes to add if given a non-CRLF arg
+sub crlf_adjust ($) {
+       if (index($_[0], "\r\n") < 0) {
+               # common case is LF-only, every \n needs an \r;
+               # so favor a cheap tr// over an expensive m//g
+               $_[0] =~ tr/\n/\n/;
+       } else { # count number of '\n' w/o '\r', expensive:
+               scalar(my @n = ($_[0] =~ m/(?<!\r)\n/g));
+       }
+}
+
 sub index_both { # git->cat_async callback
        my ($bref, $oid, $type, $size, $sync) = @_;
        my ($nr, $max) = @$sync{qw(nr max)};
        ++$$nr;
        $$max -= $size;
+       $size += crlf_adjust($$bref);
        my $smsg = bless { bytes => $size, blob => $oid }, 'PublicInbox::Smsg';
        my $self = $sync->{sidx};
        my $eml = PublicInbox::Eml->new($bref);
@@ -682,6 +694,7 @@ sub _git_log {
                } else {
                        # normal regen is for for fresh data
                        $self->{regen_down} = $fcount;
+                       $self->{regen_down} += $high unless $opts->{reindex};
                }
        } else {
                # Give oldest messages the smallest numbers
@@ -787,7 +800,7 @@ sub _index_sync {
                }
                $self->commit_txn_lazy;
                $git->cleanup;
-               $xdb = _xdb_release($self);
+               $xdb = _xdb_release($self, $nr);
                # let another process do some work... <
                $pr->("indexed $nr/$self->{ntodo}\n") if $pr && $nr;
                if (!$newest) {