From: Dimitrie O. Paun Sent: Monday, January 05, 2004 05:33
On January 2, 2004 01:12 pm, Steven Edwards wrote:
http://cvs.reactos.com/cvsweb.cgi/reactos/lib/user32/controls/ Is a good example of a place where we have been able to import a large chunck of WINE user32 code without a lot of nasty changes.
BTW, did any of the fixes from your tree been ported back to Wine? I guess not, and that's just not nice. You guys need to adopt a decent system of keeping track of what sources where imported from Wine, what got merged back to Wine, etc. I'd be willing to help, if you guys are interested.
One of the first controls to be ported was the static control. I submitted the fixes back to Wine, but the patch was dropped, 'cause someone said one of the flags worked a little bit differently on Win95/98. Well, guess what, I really don't care about 95/98 and I'm certainly not going to install it to figure it out. I think Casper Hornstrup had a similar experience recently with a LockResource16 call in ole32.dll.
Ofcourse, I do realize that patches are dropped for good reasons (I submitted a patch to comctl32 recently where someone pointed at a better solution), but in general I get the feeling that I should be thankful if a patch gets accepted.
Another case is the mega-shell32 patch that Martin Fuchs submitted. Serious work on shell32 began around 17 dec, just as Alexandre was about to go on vacation. So we (actually, 95% of the work was done by Martin) used the ReactOS CVS repository. I just counted, there were 107 separate commits to shell32 between 17 Dec and 02 Jan (I can't count the commits after 02 Jan easily). Do you guys really believe that a pile of 107 patch mails while Alexandre was unavailable would have been a better idea?
I am more than willing to do a little bit of extra work to keep our trees in synch but the simple fact remains that if the only way to get something accepted into Wine is to create a new Win95 installation and see how things work there I'm probably not going to bother.
Gé van Geldorp.