2009/8/20 Philipp A. trueflyingsheep@googlemail.com:
what do you mean with “clean it up”? i don’t see anything not-clean about it. and why is it nonfunctional in your opinion? and for testing: that’s what svn-versions are for, aren’t they?
2009/8/20 James McKenzie jjmckenzie51@earthlink.net
Philipp A. wrote:
Well bug 8555 (http://bugs.winehq.org/show_bug.cgi?id=8555) is affected by that FIXME, so that can be a starting bug.
Well, i think, they don't like working apps. What on earth can i do to make somebody stop by and put the damn patch into the svn?
Clean it up and make it functional including tests, if needed.
James McKenzie
Phillip, firstly, please bottom-post on the mailing list.
If the patch was not accepted there would have been a reason. Perhaps there was an obvious error in the code, perhaps the approach was wrong, perhaps it won't work in all cases. It also is possible that it doesn't match the behaviour of Windows, or if it does, it needs to be proved with a conformance test. Conformance tests are essential to prove that a.) The fix works, b.) It matches the behaviour on Windows and c.) a code change won't cause a regression in the future (because the test would flag that up).
Just because it appears to give the correct result doesn't mean the patch isn't flawed in some other way, you can always hack a solution that would work in some cases but it won't be the right approach.
So, if you want this patch to be accepted, then someone needs to take it up, they need to consult Alexandre Julliard in #winehackers to find out why he rejected it in the first place, then they need to fix any issues he has spotted and likely write a conformance test to prove the patch is correct. You've said you don't want to get involved in working on the patch itself, so you can find/make a bug report, link to this patch and explain in the report that the patch seems to fix it for you.
Luke.