Re: Exiting from a function or procedure

[Robert Dewar <robert_dewar@my-deja.com>]

In article <ITRM4.955$7N2.399746@news.pacbell.net>,
  tmoran@bix.com wrote:
> If a situation "clearly" warrants a warning, then it warrants
> a warning.  But there are cases where the "clearly warrants"
> is disputable.

General claims like this are a bit unconvincing, an example
would be helpful. As I said earlier, I do not know of any
warnings in GNAT where there is a good argument for omitting
the warning (or they would not be there). On the other hand,
we get quite a bit of input from users asking for additional
warnings.

> One reason for deciding not to have the
> compiler issue a particular warning might be:
>
> >The trouble with leaving warnings in your code is that you
> >can easily get into the habit of ignoring warnings, and we
> > see many

Actually, our experience is most often that people ignore
warnings which ARE useful, and very often we get examples of
code that is full of warnings, ALL of which represent real
problems, and the programmer has just ignored them all on the
grounds that warnings are usually useless. My point here is
that, again in our experience, the warnings in GNAT are
definitely useful and should NOT be ignored. I am definitely
not making a point about warnings in general. I am quite aware
that some compilers generate a lot of useless warnings!


Sent via Deja.com http://www.deja.com/
Before you buy.