On 23-04-18 10:10AM, Stefan Sperling wrote:
> On Tue, Apr 18, 2023 at 12:33:39PM +1000, Mark Jamsek wrote:
> > The below diff removes the screendump defines and writes | and -
> > unconditionally when capturing screen dumps. We define TERM=vt220
> > in the regress setup script so this can more readily be adapted
> > rather than hardcoding it in the newterm() call.
> 
> Shouldn't TERM only be overridden for the tog process, instead of
> all the processes started by the test harness?
> 
> I am not sure what the consequences of this are in practice. But I am
> a bit worried about this causing some issue for people when running the
> tests for -portable.

I'm unaware of any consequences in this case because of the nature of
the tog tests but we can either define this when we call newterm() or
wrap tog invocations in the script if you prefer? I don't mind either
way.

> > diff 989289cabaf1c0a5d169b5bc196d19d68cad2bf9 192c3791a14d64b25b81cee0c8139f442d7a67f1
> > commit - 989289cabaf1c0a5d169b5bc196d19d68cad2bf9
> > commit + 192c3791a14d64b25b81cee0c8139f442d7a67f1
> > blob - e03be4c0dcc342c5c4bee7432c3f25219682d085
> > blob + ecee2678155e9ca794a20e40f84cf1d8c0be6de1
> > --- regress/tog/common.sh
> > +++ regress/tog/common.sh
> > @@ -19,6 +19,7 @@ export LC_ALL=C.UTF-8
> >  
> >  unset TOG_VIEW_SPLIT_MODE
> >  unset LC_ALL
> > +export TERM=vt220
> >  export LC_ALL=C.UTF-8
> >  export COLUMNS=80
> >  export LINES=24
> 

-- 
Mark Jamsek <fnc.bsdbox.org|got.bsdbox.org>
GPG: F2FF 13DE 6A06 C471 CA80  E6E2 2930 DC66 86EE CF68