Peter,
On 2005 Feb 1 , at 13.26, Peter W. Draper wrote:
>> This has puzzled me, and I've spent a good part of today trying to
>> work
>> out what is going on here. I _think_ that m4 was not building
>> properly
>> because of a timestamps problem in the sources, which I think I've
>> fixed (though it could have been Peter's adjustments to libtool
>> yesterday which did the magic).
>
> Hi Norman,
>
> this little lot has turned out to be a bug in m4 (which the datestamp
> fixes just side-stepped). I've applied a simple-mind patch to output.c
> for
> it from the bug-m4 mailing list and CONVERT now passes bootstrap. Check
> the following URLs:
I saw your fix. It's rather reassuring that there was an actual bug in
here rather than CVS timestamp fragility.
> Which suggest we might be better upgrading to 1.4.2 or later, but let's
> see what happens first.
I'd be happy to import that version as soon as I get back.
Norman
--
----------------------------------------------------------------------
Norman Gray : Physics & Astronomy, Glasgow University, UK
http://www.astro.gla.ac.uk/users/norman/ : www.starlink.ac.uk
|