Re: misc datatypes

[Ladislav Michl]

On Tue, Apr 29, 2003 at 10:44:59AM +0200, Pawel Kot wrote:
[snip]
> But at the moment I'm considering Ladis idea. unsigned int type is always
> at least as wide as uint8_t.

it is always wider...

> So using unsigned int should cause no harm and the compiler should do the
> work. Can anyone see the weakness of the idea? I have not yet made any tests.

you already tested it :) we had this construction in cfgreader.c (killed
in cvs revision 1.42)
gn_config_global.rfcomm_cn = atoi(val);
and since atoi is function returning int it is basically the same situation
as defining rfcomm_cn as unsigned int (only typecasting is done in another
place).

        ladis