Build-Breaking MingW 3.3 Stuff

Mon Apr 14 05:18:00 GMT 2003

Hi Andrew,
>At this stage in the 3.3 release process we need to be careful about
>what we do.

A couple more questions concerning this stuff:
- What about the case of the _Jv_select() problem, where I believe that
 the logic is incorrect on all platforms, but the bug only manifests itself
 on Win32? Would a patch for this be considered a Windows-specific
 patch?
- Does the "be careful" part preclude any sort of cross-configury patch
 going into 3.3? (As of Friday night, I think I have a pretty decent one.)
What would be your attitude about your personally committing MingW-specific
code which is guaranteed not to mess up the Posix build, such as changes to win32.cc
or nat*Win32*.cc? Would you be inclined to let these slide or would we
need to find a maintainer with a Windows box in all cases? And what happens
if a maintainer doesn't step up to bat for a non-trivial change? (Like I said,
I'm new to this, so I don't know how many people have a
3.3 tree + Windows box + commit access.)
-- Mohan
http://www.thisiscool.com/
http://www.animalsong.org/