>A simpler solution would be to return zero from same_file if both
>files have inode values of zero. I think such values are impossible
>on Posix systems, but if they are, then this could be a Windows
>specific definition of the macro (system.h in Diffutils is already
>prepared to support a build
>I'm afraid it doesn't suffice merely to say "here's some code; figure it
>out"; we need a patch against Gnulib, something that you've tested. I
>suggest using the format of "git format-patch" for the patch, and using
>"git send-email" to email the patch to bug-gnulib@gnu.org. You can use
>the
Hi,
I was trying to get diffutils 3.5 working while using Visual Studio. I.e. I
like to generate native Windows binaries, that do not need Cygwin.
Based on recommendation in the thread at
https://debbugs.gnu.org/cgi/bugreport.cgi?bug=25814, I'm sending a request to
you.
Can you please fix th
Hi Bruno,
Thanks for reply.
>> 1. frexp.c is not needed, as Visual Studio already provides frexp()
>> function via system libraries.
...
>Do you have data that shows that MSVC14's frexp() behaves better than the one
>in MSVC 9?
Do you have advice how I can check this quickly? Visual Stu
>> 1. frexp.c is not needed, as Visual Studio already provides frexp()
>> function via system libraries.
>But probably with bugs. And even if it is not needed on your platform,
>it is part of the tarball to replace broken frexp() on systems where it
>is buggy. Part of configure determines
Hi Bruno,
>Most of these troubles should go away if you use the 'compile' and 'ar-lib'
>auxiliary scripts, as described in section 2 of
>http://git.savannah.gnu.org/gitweb/?p=gperf.git;a=blob_plain;f=README.windows
>These instructions were tested with GNU gperf and a couple of other GNU
>packages
>gnulib has been updated to work with this platform on 2016-12-13, see
>http://git.savannah.gnu.org/gitweb/?p=gnulib.git;a=commit;h=5506db6b006274762bf5ea1d23feb7a9801529c8
>Eric Blake replied:
>> If you can help port gnulib to the newer visual studio 2015, I'm sure
>> patches are welcome.
>> ...
>Are the mentioned patches in the snapshot that was provided by Jim?
>In any case, I will check today. Thanks for prompt replies.
>If you like, I can share some additional changes I've made, e.g. Visual Studio
>does not like -g flag, a WIN32 define that should be _WIN32 (a CL.exe built-in
>macro