Wine developer frustration (was Re: ntdll: Improve stub of NtQueryEaFile.)

[André Hentschel]

Am 16.06.2015 um 23:49 schrieb Vincent Povirk:
>> While it seems Theodore didn't do his homework and didn't know about that source distribution, one of his point remains:
>> it's not a database, git, or something easily readable, just a blob
>>
>> Oh man, i didn't plan to jump into that "flamewar"...
> 
> Sadly, we don't have that internally. We don't rebase all our hacks
> onto Wine every few weeks and get a "clean" set of patches, Alexandre
> just does merges. There are pros and cons to this system, but I doubt
> Alexandre will be willing to change it.
> 
> We file bugs in our internal tracker documenting and explaining each
> hack, but there are still diffs left over from before we adopted that
> system, which we don't have documented, and may not understand.
> 
> Those old diffs can be more of a liability for us, as they're not
> being tested by the wider community, and just because they helped some
> application 10 years ago doesn't mean they're not hurting things now.

Ok. I also don't think they could help Wine and know you'd upstream every non-hack. So those hacks aren't really interesting.
Just wanted to try to explain what Theodore said.

>> Simply don't push them out of bugzilla, but sure, we also don't want a flood of new bugs which might not be correctly reported by users...
> 
> A "staging" keyword in bugzilla would be a strong indication that
> wine-staging bugs are desired, but we want to know if there's a
> difference in behavior between staging and winehq.

Maybe really something like the regressions page, but for difference between vanilla and staging, and a list of bugs which were only tested on staging and not on vanilla yet... Just an idea
I guess more interesting than "wow, it works in staging" would be "ha! staging breaks something, why is that?"...