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;
}
$self->{mm}->mid_delete(mid_mime($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);
} 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
}
$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) {