Re: popt with msvc?

2010-10-28 Thread Jeff Johnson

On Oct 28, 2010, at 12:50 AM, Matt Arsenault wrote:

 
 Other than those failed tests, I do now have it working without the need to 
 revert to an older version.
 

If you send alon a patch for VS, I'll try to carry that in popt (I don't do VS 
or Windoze but ..)
There's a patch for Xcode in popt already ... popt changes little enough that 
VS might be
maintainable in spite of bit rot.

The gnulib approach used by Mingw is way way overkill for a teensy library like 
POPT imho.

73 de Jeff


__
POPT Library   http://rpm5.org
Developer Communication List   popt-devel@rpm5.org


Re: popt with msvc?

2010-10-27 Thread Jeff Johnson

On Oct 27, 2010, at 12:25 AM, Matt Arsenault wrote:

 
 On Oct 26, 2010, at 11:27 PM, Jeff Johnson wrote:
 What test are failing? Like long long and double, the tests are quite 
 complete
 over data types and It Really Doesn't Matter: hardly ant program needs/uses
 long long and double CLI options.
 
 test1 - 19, 20, 21, 22, 23, 60, 61

19-23 are testing alias/exec functionality from test-poptrc.
If you can live without alias/exec (and you likely can,
only rpm makes heavy/serious use of popt alias/exec afaik).

60-61 are just cosmetic --help comparisons. If popt alias/exec
isn't functional, then the --help display will fail too.

 test3 - 51, 52, 53
 
 I just sort of haphazardly threw things together until it built, so I may 
 have broken something. I quickly added cmake to try building it since I have 
 no idea how to get autotools to work with VS. 

Yes the AutoFu chains are getting rather daunting. The basic trick
with autotools, is use configure, and avoid using the autotools, on
tricky platforms.

I haven't a clue how to get AutoFu to work with VS either.

You might try older POPT tarballs. The basic functionality of POPT hasn't 
changed since
forever, nor has the API, nor are there significant bugs. Older tarballs might
have less AutoFu.

73 de Jeff
__
POPT Library   http://rpm5.org
Developer Communication List   popt-devel@rpm5.org