Thread: sizeof pointer-to-array -- &array

Originally Posted by josemariasola

OK, that's right for MS, but does that also goes for Borland?

Maybe I'm beeing a little naive, but I carefully configured both, Borland and MS, compilers to conform to ANSI C 90 and to emit all possible warnings.

Should I just accept it's a bug?

Joey.

Borland, from my experience, likes to play by their own rules. Some of the most complained about compilers in the history of the language come from Borland. In my opinion, if it isn't from GNU, it's not worth using. Microsoft is the closest exception to that, that I know of.

>But is it possible that both companies be wrong on this same subject for so long?
Yes. It's a subtle and rarely displayed bug in one of the rare exceptions to the rule. Because of that, 1) it's very easy to make that mistake, 2) it's unlikely to be noticed, and 3) it might not be important enough to justify the cost of fixing. So the bug could easily exist through several versions of different compilers.

Originally Posted by Prelude

3) it might not be important enough to justify the cost of fixing. So the bug could easily exist through several versions of different compilers.

I know my company has taken that stance with a couple bugs. Click a button a little to fast on one of the schematic loading pages and you get a nice error... so we just have a warning. We could go back and revise that section, but there are other more pressing maters for us to work on.

Originally Posted by josemariasola

Thanks for the link. What I'm failing to interpret is if discarding or not the & in sizeof &a is an implementation defined behavior.

With what others such as Lawrence Kirby said in that thread, and Prelude and others have said in this one, I'd go with, "No, it's broken".