treewide: use run_qx where appropriate

This saves us some code, and is a small step towards getting
ProcessIO working with stat, fcntl and other perlops that don't
work with tied handles.

1 files changed, 3 insertions, 4 deletions

diff --git a/lib/PublicInbox/LeiMirror.pm b/lib/PublicInbox/LeiMirror.pm
index 43e59e6c..c2f7cbed 100644
--- a/lib/PublicInbox/LeiMirror.pm
+++ b/lib/PublicInbox/LeiMirror.pm
@@ -7,7 +7,7 @@ use v5.12;
 use parent qw(PublicInbox::IPC);
 use IO::Uncompress::Gunzip qw(gunzip $GunzipError);
 use IO::Compress::Gzip qw(gzip $GzipError);
-use PublicInbox::Spawn qw(popen_rd spawn run_wait run_die);
+use PublicInbox::Spawn qw(spawn run_wait run_die run_qx);
 use File::Path ();
 use File::Temp ();
 use File::Spec ();
@@ -57,9 +57,8 @@ sub try_scrape {
         my $curl = $self->{curl} //= PublicInbox::LeiCurl->new($lei) or return;
         my $cmd = $curl->for_uri($lei, $uri, '--compressed');
         my $opt = { 0 => $lei->{0}, 2 => $lei->{2} };
-        my $fh = popen_rd($cmd, undef, $opt);
-        my $html = do { local $/; <$fh> } // die "read(curl $uri): $!";
-        CORE::close($fh) or return $lei->child_error($?, "@$cmd failed");
+        my $html = run_qx($cmd, undef, $opt);
+        return $lei->child_error($?, "@$cmd failed") if $?;
 
         # we grep with URL below, we don't want Subject/From headers
         # making us clone random URLs.  This assumes remote instances