Re: tests

Date view Thread view Subject view Author view

From: Godmar Back (gback@cs.utah.edu)
Date: Sat Feb 06 1999 - 23:08:44 EST


>
> On Feb 7, 1999, Godmar Back <gback@cs.utah.edu> wrote:
>
> >> We use uint8, not uint8_t, and the BeOS patch seems to be attempting
> >> to use the latter.
>
> > No it doesn't. In fact, if I do a rgrep uint8_t, I don't see anything.
>
> Ah! Now I see it. When I first read the configure.in code that
> ``tested'' for those defines, I had assumed they had something to do
> with the *int*_t types, not *int*. If BeOS defines types without the
> _t suffix, I see some reason for the tests, and some other platforms
> may do it as well.
>
> > It was really all about the types kaffe defines and uses, which
> > clashed with beos definitions.
>
> Shouldn't we stop using those potentially conflicting type names, and
> use only the j* types with equivalent meaning?
>
> Most of the native code could also benefit (in terms of portability)
> from not using standard C types directly, using the j* types instead.

 I'm all against any unnecessary typedefs. There's reasons that Java does
not allow different names for the same type.

Hold off with it, though, since I've got lots of outstanding changes at
the moment.

        - Godmar


Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Sat Sep 23 2000 - 19:57:59 EDT