<<< Date Index >>>     <<< Thread Index >>>

Re: sscanf (x, "%u", &(uint64_t))

On Sun, Mar 20, 2005 at 05:24:54PM -0800, Brendan Cully wrote:
> On Monday, 21 March 2005 at 10:21, Tamotsu Takahashi wrote:
> > On Sun, Mar 20, 2005 at 02:43:45PM +0100, Thomas Glanzmann wrote:
> > > > I misread the documenation. UIDVALIDITY is 32bit not 64bit. I will send
> > > > a patch shortly.
> > > 
> > > here is the patch. Some people should test it before it is reported
> > > upstream. For me it works.
> > 
> > Thank you.
> > 
> > I have only 32bit machines, so, uint32_t is unsigned int.
> > Now there is no problem for _me_.
> > 
> > But, wouldn't some environment need this attached patch?
> > I'm afraid "%u" is not for uint32_t on some systems.
> 
> wouldn't it be simpler just to use 'unsigned int' instead?

Yes, it's better if we assume that 'unsigned int' has
always equal to or more than 32 bits. And I think it's
a safe assumption.

But, didn't you put '#elif SIZEOF_LONG==4' into m4/types.m4?

-- 
tamo