Mail Archives: cygwin-developers/2001/10/09/14:57:57
On Tue, Oct 09, 2001 at 10:43:55PM +0400, egor duda wrote:
>Hi!
>
>Tuesday, 09 October, 2001 Christopher Faylor cgf AT redhat DOT com wrote:
>
>CF> I keep seeing people who have set the CYGWIN environment variable from the bash
>CF> shell. This causes strange problems.
>
>hmm. on the first glance only 'ntsec' and 'tty' options may cause
>trouble when changed on-the-fly. OTOH, they're probably most important
>options in CYGWIN env. variable. if we can pinpoint those options that
>are really prohibited from changing during "cygwin session" we can
>ignore only them. but 1) there may still be "legitimate" ways to
>change such options; 2) changing other options may still produce
>unexpected results to the people who don't know what they're doing.
>
>CF> I wonder if it would be worthwhile to ignore the CYGWIN environment variable
>CF> if it is set in a process that is a child of a cygwin process.
>
>CF> The downside is that there are probably some situations where it
>CF> actually works ok to do this and that we'll end up screwing up people
>CF> who know what they're doing.
>
>it'd definitely screw some of my scripts. and what's most
>disappointing, i can't see any workarounds if such change is made.
>
>CF> We could just issue a warning but I hate for cygwin to be noisy in these kinds
>CF> of situations.
>
>CF> Any thoughts?
>
>I hate to sound non-constructive, but i think we'd better leave things
>as is. current situation is not good, but i don't see any better
>alternatives.
That's ok. I wasn't exactly sure about the idea myself. I do like the
idea of ignoring only the problematic settings but this might be too
much work for too little gain.
cgf
- Raw text -