Re: "set -e" handling is broken with zsh 5.3.1 and 5.4.1

[Peter Stephenson <p.w.stephenson@xxxxxxxxxxxx>]

On Sun, 27 Aug 2017 23:45:20 -0700
Bart Schaefer <schaefer@xxxxxxxxxxxxxxxx> wrote:
> Background is in workers/34055 and 34059, and this looks to have
> something to do with noerrexit not propagating into the first sublist
> in a function body.  It's entirely possible that the change from
> having a few numeric values for the flags to have a set of bit values
> has made that test unnecessary, or it may be needed in case that
> hasn't been tested yet -- the original report in 34055 involves
> calling a function within an if/else block.

I committed the chnage.  The regression tests in C03ttraps.ztst are
still passing, and I couldn't see any simple code involving functions
that caused a failure.  If someone finds one we will need to add it to
the tests.

pws