Re: [Plplot-devel] Comprehensive test results for MingW-w64/MSYS2

[Greg J. <gv...@gm...>]

I'm working on it, from my point of view (which includes examining the
"installed" perspective).
I've had now since I adopted Msys2, a well-working cmake system that
applies equally to mingw/msys,
mingw32/msys2, and mingw64/msys.  All with the standard cmake executable
and with only a few
module replacements.

On Tue, Jul 14, 2015 at 5:25 PM, Alan W. Irwin <ir...@be...>
wrote:

> On 2015-07-14 15:45-0700 Greg Jung wrote:
>
>  From the ..env_out:
>>>
>>
>> PATH=/d/cmake3.2.2/bin:/usr/local/bin:/usr/bin:/bin:/opt/bin:/c/ProgramData/Oracle/Java/javapath:/d/CAF/bin:/c/GTK/bin:/c/tcl/bin:/c/windows/system32
>>
>> and:
>> -- Check for working C compiler: C:/msys64/usr/bin/gcc.exe -- works
>>
>> You are running this with the MSYS environment which should only be used
>> for things that
>> will stay under /msys64.  You need to bring up the mingw64 window and run
>> from that.
>>
>
>
>> The /mingw64/bin/: goes to the front and perl puts itself at the rear.
>>
>> /mingw64/bin:/usr/local/bin:/usr/bin:/bin:/c/Windows:/c/Windows/system32:/c/Windows/system32/Wbem:/usr/bin/site_perl:/usr/bin/vendor_perl:/usr/bin/core_perl
>>
>> There is a mingw64/bin/cmake that I'm sure Alex would recommend,
>>
>
> Hi Greg:
>
> I wonder who that "Alex" guy might be?  :-)
>
>
>  but I
>> don't know how to work it.
>>
>
> To access the MSYS2 version of cmake, you have to do two steps, see my
> recent post with the subject line 'Testing the "vanilla"
> MinGW-w64/MSYS2 platform'. Actually, I think the advice in that post
> which is derived from the advice in
> <http://www.openwalnut.org/projects/openwalnut/wiki/InstallMSYS2> is
> basically consistent with the advice you gave above although it is
> stated in different terms.
>
>
>> CMAKE_SYSTEM_NAME: Windows
>> UNIX: 1
>> WIN32: 1
>> APPLE:
>> MSVC: (MSVC_VERSION: )
>> MINGW:
>> MSYS: 1
>> CYGWIN: 1
>> BORLAND:
>> WATCOM:
>>
>> SWIG_FOUND: FALSE
>> PERL_FOUND: TRUE
>> X11_FOUND: 0
>>
>> is this "MSYS Makefiles" generator?  How does UNIX get set?
>>
>
> I think the above most peculiar combination of flags with CYGWIN set
> and MINGW not set is all moot now since it was generated with "MSYS
> Makefiles" (see comprehensive_sh.out in Arjen's report) rather than
> the recommended "Unix MakeFiles" and probably a cmake version that is
> not the recommended (MSYS2) version of cmake was used as well. Even
> so, Arjen's first round of testing of limited PLplot components on
> MinGW-w64/MSYS2 was quite encouraging. So it begs for a follow up with
> using the recommended generator and cmake version.  However, my
> understanding is that Arjen will not be able to do that follow up for
> several more weeks, i.e., long past our release date of Saturday, July
> 25th.
>
> So, Greg, would you be willing to do that follow up now instead?  That
> would be a big help for this release. Note that request is for the
> "vanilla" version of that platform which you can obtain by following
> the <sf.net/projects/msys2> wiki instructions for using the official
> "pacman" installer for MSYS2.
>
>
> Alan
> __________________________
> Alan W. Irwin
>
> Astronomical research affiliation with Department of Physics and Astronomy,
> University of Victoria (astrowww.phys.uvic.ca).
>
> Programming affiliations with the FreeEOS equation-of-state
> implementation for stellar interiors (freeeos.sf.net); the Time
> Ephemerides project (timeephem.sf.net); PLplot scientific plotting
> software package (plplot.sf.net); the libLASi project
> (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
> and the Linux Brochure Project (lbproject.sf.net).
> __________________________
>
> Linux-powered Science
> __________________________
>