Messages in this thread | | | Date | Fri, 07 Dec 2007 16:41:31 +0100 | From | Richard Knutsson <> | Subject | Re: [Patch] net/xfrm/xfrm_policy.c: Some small improvements |
| |
David Miller wrote: > From: Richard Knutsson <ricknu-0@student.ltu.se> > Date: Thu, 06 Dec 2007 15:37:46 +0100 > > >> David Miller wrote: >> >>> But this time I'll just let you know up front that I >>> don't see much value in this patch. It is not a clear >>> improvement to replace int's with bool's in my mind and >>> the other changes are just whitespace changes. >>> >>> >> Is it not an improvement to distinct booleans from actual values? Do you >> use integers for ASCII characters too? It can also avoid some potential >> bugs like the 'if (i == TRUE)'... >> What is wrong with 'size_t' (since it is unsigned, compared to (some) >> 'int')? >> > > When you say "int found;" is there any doubt in your mind that > this integer is going to hold a 1 or a 0 depending upon whether > we "found" something? > > That's the problem I have with these kinds of patches, they do > not increase clarity, it's just pure mindless edits. > But is there not a good thing if also the compiler knows + names are sometime not as clear as that one? > In new code, fine, use booleans if you want. > > I would even accept that it helps to change to boolean for > arguments to functions that are global in scope. > > But not for function local variables in cases like this. > Oh, I see your point now. Believed it to be yet another 'booleans is not C idiom'.
Sorry about the noise Richard Knutsson
| |