On Mon, Aug 13, 2018 at 1:36 PM Peter Stephenson wrote: > So you're saying the commit you say is causing the problem isn't > fixing the problem it was supposed to fix? That implies something > else changed, since dana tried it out at the time and said it worked. > Sorry, I was unclear. The commit is indeed fixing the problem it was supposed to fix. I meant that on all terminals I tested, I saw consistent behaviour: the commit fixes the original problem but introduces a new one (mine). So it looks like it should be possible to fix both problems. -- Guillaume