[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: cvs exit status

From: Jerry Nairn
Subject: RE: cvs exit status
Date: Fri, 12 Oct 2001 10:50:28 -0700

> From: address@hidden [mailto:address@hidden
> Sent: Friday, October 12, 2001 9:11 AM
> [ On Thursday, October 11, 2001 at 20:06:18 (-0700), Paul 
> Sander wrote: ]
> > And your point is what?  That it's okay to sometimes let 
> nondeterministic
> > errors go undetected?  Sorry, that's not good enough.
> I think in real life the chance for this overflow having any adverse
> affect would be about nil....  You're making a mountain out of a tiny
> mole hill.

C'mon. You flame other people to hell and back for ignoring mole hills like
that. You were trying to flame Paul just for saying that the mole hill
exists, not for exaggerating the significance of it.
(The definition of a flame may be imprecise. I consider almost any mail
which has multiple exclamation points and question marks, "!?!?!?!?", to be
a flame. Any mail which suggests that something someone else has said is
"bogus" is also usually a flame.)
Basically, if incrementing the exit code can produce a successful exit code
then it is the wrong way to indicate an error. And if the result of
incrementing the exit code, rather than setting it to a value, does not
reliably provide information about any errors which occurred, then there's
no reason to do it.
So incrementing the exit code can often give incorrect results in the form
of the wrong number, and seldom, but sometimes, indicate no error when some
errors occurred. That's bad.
You just want to say that it's not very bad. Swell. Whatever.

reply via email to

[Prev in Thread] Current Thread [Next in Thread]