[ODE] Making it build again

Gary R. Van Sickle g.r.vansickle at worldnet.att.net
Fri Sep 24 20:42:34 MST 2004


> I wouldn't say "bothered by warnings." It's well-accepted 
> practice in the industry that WARNINGS ARE ERRORS.

No it isn't.

> If you're 
> not treating warnings as errors, well, you should! (I don't 
> want to argue about it, but you can't find anyone who's any 
> good who wouldn't agree).

I'm good.  I don't agree.  Errors are errors.  Warnings are not errors,
that's why they've been allocated a different name.  Warnings are to be
minimized and ideally eliminated if possible; they are not to stop the
production of object code.

While it sounds like your mind's made up, consider:

1.  What level of warnings do you have selected?  What's the well-accepted
warning level used by the industry?
2.  Every compiler has different warnings.
3.  Every compiler I've ever used issued warnings on perfectly valid code
which could not be eliminated.  Examples: VC6 (IIRC) issues a warning if you
pass a bool to an int parameter, even if you cast it.  Many if not most
compilers warn on a "while(1)", and there's nothing you can do about it.

Warnings-as-errors has one purpose and one purpose only: to aggravate
developers.

-- 
Gary R. Van Sickle
 



More information about the ODE mailing list