gcl-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gcl-devel] Re: GCL produces wrong EOL on mingw


From: Vadim V. Zhytnikov
Subject: Re: [Gcl-devel] Re: GCL produces wrong EOL on mingw
Date: Sun, 21 Mar 2004 12:54:13 +0300
User-agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; ru-RU; rv:1.5) Gecko/20031006

Mike Thomas ?????:
Hi Camm.

As to whether it is related to CT/LF issues , who would know.  I would
expect it to be one off instead of two.


OK, just to be clear, you don't see filename corruption which using
(ignore-errors ... on non-existent files, right?

Both head and stable have no such problem with existent and non-existent files and parent directories including replacement of forward slashes with double back-slashes:

=======================================
 >(ignore-errors (file-namestring "c:/notadirectory/notafile"))

"notafile"

 >(ignore-errors (file-namestring "c:/msys/notafile"))

"notafile"

 >(ignore-errors (file-namestring "c:/msys/1.0"))

"1.0"
=======================================

Vadim, I'm reluctant to chase the problem for 2.6.2 as we already have a workaround, but let me know if you disagree strongly and I'll try to do so during the coming week.

Cheers

MIke Thomas.


You won't trigger the problem that way. I only can reproduce it during Maxima build and it occurs not with the very first file but after dozens of successful (ignore-errors (file-namestring ...)). Even more. If Maxima build is interrupted by this notorious path error you may start make once again and get successful Maxima build maybe after several make restarts. But sometimes make always stops at the same file and
and you can't go further.

I don't insist on fixing the problem before 2.6.2. To be honest, I'll be
pretty busy myself next week or two. So if Camm don't mind lets release 2.6.2. We have only to mention somewhere in readme or release note that
gcl 2.6.2 on mingw require binitils 20040222 and gcc 3.3.1
Not 3.3.3! See today's posting by Barton Willis on Maxima list:
====================================================================
It seems that cvs Maxima does not build correctly using gcc
3.3.3-20040217.
After installing gcc 3.3.1,  cvs maxima builds correctly and passes all
tests
(windows XP and GCL 2.7.0).   With gcc 3.3.3, Maxima built okay and ran
okay for simple problems  but gave a  segmentation fault when it ran
the test suite.
====================================================================
I'm not sure that this is gcl problem but problem exist.

In any case I feel that polishing mingw gcl build should be one
of the primer objectives for post 2.6.2 development.



--
     Vadim V. Zhytnikov

      <address@hidden>
     <address@hidden>





reply via email to

[Prev in Thread] Current Thread [Next in Thread]