Re: flex beta 2.5.23 released

[W. L. Estes]

On Monday, 25 November 2002,12:30 -0500, Bruce Lilly wrote:

> Indeed it does; e.g. K&R declarations.  My implicit point is that
> if pre-C99 compatibility is to be abandoned (and therefore K&R
> compatibility as well), then there are other things, such as the
> K&R declarations, that should be revised; conversely if there is
> to be any semblance of portability to pre-C99 systems, use of
> C99-specific types isn't going to work.

K&R compatibility is a non-interest. ANSI compatibility is nice, but
not always practical. POSIX is where most of the flex development
happens. I personally think that expecting POSIX compliance is a
reasonable expectation.

> >There are some changes in the flexint.h header in flex beta
> >2.5.24. Could you have a look at it and tell us how it works for you?
> 
> Will do.

Good. I've committed a fix to flexint.h which is about 3 lines
long. Let me know if you want it and I'll send it to you off-list. It
includes a call to inttypes.h for C99 systems.

> Unless I'm mistaken, *exact* sizes aren't necessary; what is needed is
> assurance that a particular type is sufficiently large to hold some
> number (state number, source code line number, etc.).

Yes and what with the least types being the required ones, we'll
probably end up using them somehow.

> I suspect that part of the problem is the assumption that gcc is always
> right, which is not a valid assumption.  Solicitation of tests on non-gcc
> systems would probably reveal a number of issues.  In the specific case
> under discussion, gcc has the intN_t types defined in the wrong place.

On most systems that I've looked at, sys/types includes something like
inttypes.h (but it's sometimes called int_types.h or similar) so using
sys/types.h works in a lot of places. Including an unconditional call
on inttypes.h fails more often than not.

Generally, header files are provided by some sort of library, not by
the compiler, although gcc does provide some headers.