"GOT", but the "O" is a cute, smiling pufferfish. Index | Thread | Search

From:
Stefan Sperling <stsp@stsp.name>
Subject:
Re: two more dead stores in tog/tog.c
To:
Mark Jamsek <mark@jamsek.com>
Cc:
Florian Obser <florian@narrans.de>, gameoftrees <gameoftrees@openbsd.org>
Date:
Fri, 29 Jul 2022 12:54:08 +0200

Download raw body.

Thread
On Fri, Jul 22, 2022 at 01:18:55PM +1000, Mark Jamsek wrote:
> On 22-07-21 10:28pm, Stefan Sperling wrote:
> > On Thu, Jul 21, 2022 at 07:09:23PM +0200, Florian Obser wrote:
> > > scan-build reports two more dead stores in tog/tog.c where I don't know
> > > what to do about.
> > > 
> > > No idea what this does, should there be only one call to
> > > offset_selection_down()? or are there simply two
> > > 	if (err)
> > > 		return err;
> > > missing?
> > > 
> > >   1206		if (v->mode == TOG_VIEW_SPLIT_HRZN) {
> > >   1207			err = offset_selection_down(v);
> > >   1208			err = offset_selection_down(v->child);
> > >   1209		} else {
> > 
> > To me this looks like two missing error checks.
> > In the else clause, we run the corresponding _up functions for
> > both v and v->child as well.
> 
> Yes, Stefan is correct: it's two missing error checks.  I committed this
> mistake recently when changing the behaviour of the 'S' key map.  The
> second call is needed when switching to a hsplit because we need to
> scroll down to offset selections in the child view too.
> 
> Thanks, Florian!

Since Florian did not send a patch, I have fixed these problems now.
Thanks again for spotting these errors, Florian!