Re: [Success] XEmacs 21.4.10 "Military Intelligence (RC3)", i586-pc-win32
Adrian Aichner wrote:
>>>>> "Stephen" == Stephen J Turnbull
<stephen(a)xemacs.org> writes:
>>>>> "APA" == Adrian Aichner <adrian(a)xemacs.org> writes:
APA> lisp-tests.el: 3588 of 3604 (99%) tests successful.
APA> I think this are long-known problems on native Windows:
Stephen> You're right, I remember these being reported before.
Stephen> I'll check the archives and probably (ISTR this is hard
Stephen> to fix) add some code to say "expect these tests to fail
Stephen> because...." And add that information to PROBLEMS and
Stephen> docstrings and so on.
The way I read
http://www.opengroup.org/onlinepubs/007908799/xsh/fprintf.html
the exponent can be two digits or more.
What's the spec we are testing against?
Presumably the ANSI C spec, which says that:
The exponent always contains at least two digits, and only as
many more digits as necessary to represent the exponent.
Unfortunately, Microsoft decided to ignore the second half of that
sentence; their *printf() functions always use three digits for the
exponent.
--
Glynn Clements <glynn.clements(a)virgin.net>