Re: [bug] busyloop upon $=var with NULs when $IFS contains both NUL and a byte > 0x7f

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

> On 13/12/2022 11:40 Jun T <takimoto-j@xxxxxxxxxxxxxxxxx> wrote:
> 
>  
> > 2022/12/13 19:13, Peter Stephenson <p.w.stephenson@xxxxxxxxxxxx> wrote:
> > 
> >> On 13/12/2022 09:49 Jun T <takimoto-j@xxxxxxxxxxxxxxxxx> wrote:
> >> 
> >> How can we 'give up'?  Is it OK to call exit()?
> > 
> > By giving up, I meant we stop splitting at that point and leave the
> > rest of the string intact.  If splitting fails, stop splitting...
> 
> If we modify wordcount() in this way, then we also need to modify
> at least sepsplit() and spacesplit()?

Right, this would have to be a general policy.

I'm not necessarily hooked on this way of doing it; it seems to
me it's a little bit less surprising than resetting IFS but I
suppose errors are always surprising.

pws