From: "Jeff Trawick" <[EMAIL PROTECTED]>
Sent: Sunday, May 13, 2001 7:15 PM
> "William A. Rowe, Jr." <[EMAIL PROTECTED]> writes:
>
> > > trawick 01/05/13 08:34:18
> > >
> > > Modified: test test_apr.h testmd5.c testproc.c
> > > Log:
> > > fix some bad parms to printf/fprintf
> > >
> > > - printf("Error was %ld : %s\n", rv, strerror(rv)); \
> > > + printf("Error was %d : %s\n", rv, strerror(rv)); \
> >
> > Do I smell a need for an APR_STATUS_T_FMT ? It doesn't seem we can assume
> > this on 64 bit machines (?).
>
> apr_status_t is always int so %d is always correct (though I suppose
> the code in apr_errno.h could change).
No. It's always a 32 bit int ... I want my fully 64 bit machine (none of this
silly 32 bit numeric magic with 64 bit pointer cruft). And at least BEOS won't
accept %d for a 32bit int representation.
As I mentioned in an aside to David, the first bit is an APR_INT32_T_FMT, which
can be the simple basis for APR_STATUS_T_FMT on all platforms.
> I wouldn't mind seeing APR_STATUS_T_FMT though... Why should the user
> stop to wonder which APR types are always the same and which are not?
Yes, consistency is key.
