[ On Thursday, September 20, 2001 at 14:08:51 (-0400), Andrew Brown wrote: ]
> Subject: Re: ignoring return values from functions.
>
> so gcc has no problems with that.  lint still doesn't like it.  this
> is another problem.  who's right?

well, I certainly can't tell without seeing the code -- send it off list
if you like....  :-)

>  are there things i should fix to
> make the warnings go away, should i just ignore them, or should i just
> tell lint to shut up?  a tool that won't shut up about meaningless
> things becomes, itself, meaningless.

It's the understanding of the meaning that's important....  Obviously if
you don't understand what lint is complaining about then it's output is
just noise to you.

> i expected it to have a non-zero exit status if it found something to
> complain about.

that would be very un-unix like of it, and also contrary to one's usual
goals when running 'lint' from make, etc.  Usually you want to collect
the entire set of warnings and nits and then deal with them on a
priority basis.  in this case you're only interested if the lint process
fails because of some system or hardware error.

> then his manner and terminology are wrong.  i, for one, would never
> have taken that to be anything but pedantic humor because of the way
> it was written.

I believe the "ten commandments" were originally an appendix to Henry's
ammended version of the Indian Hill Style Manual (they appear on it,
that much is certain).  They became quite popular in the late 1980's or
early 1990's though and circulated independently of the paper on Usenet
(perhaps before the copyright on the paper was clarified or something
like that -- I don't really remember).

-- 
							Greg A. Woods

+1 416 218-0098      VE3TCP      <gwoods@acm.org>     <woods@robohack.ca>
Planix, Inc. <woods@planix.com>;   Secrets of the Weird <woods@weird.com>