Hey everyone!
Just wanted to bring to your attention http://bugs.winehq.org/show_bug.cgi?id=1348 again, and ask what is the holdup? As far as I can see it appears a solution and working patch has been posted. I am interested in getting this into the main tree and wondering what kind of modifications have to be made to the patch supplied to make it happen.
------- Comment #8 From Robert Reif 2007-02-17 15:45:56 ------- The reason I was given by Alexandre is that the functionality should be moved into the server.
Why is it not possible to accept this patch and then start working on moving it into the server? With a quick look the only changes from the patch and the git, is the change from dll/user to dll/user32.
Cheers! James
On Thu, Jul 10, 2008 at 2:13 PM, James Keane james.keane@gmail.com wrote:
Hey everyone!
Just wanted to bring to your attention http://bugs.winehq.org/show_bug.cgi?id=1348 again, and ask what is the holdup? As far as I can see it appears a solution and working patch has been posted. I am interested in getting this into the main tree and wondering what kind of modifications have to be made to the patch supplied to make it happen.
------- Comment #8 From Robert Reif 2007-02-17 15:45:56 ------- The reason I was given by Alexandre is that the functionality should be moved into the server.
Why is it not possible to accept this patch and then start working on moving it into the server? With a quick look the only changes from the patch and the git, is the change from dll/user to dll/user32.
If the correct way to do it is X, and the current implementation does it by Y, which is not correct, then why would we accept the patch that implements Y? By definition it's a hack, and if we did accept it, there'd be no incentive to fix the hack, because everything would already work (just in the wrong way).
Yes that makes complete sense, but in this case how would the implementation be better if we moved it into the wineserver? Isn't the point of having seperate drivers for X11 and quartz etc, is to take advantage of exposed interfaces ie Hotkeys?
Also, how am I as a wine outsider suppose to implement X if I don't know what it is; this is why I asked.
Let me rephrase my original question then: why is it better to move this in the server?
James
On Thu, Jul 10, 2008 at 4:44 PM, James Hawkins truiken@gmail.com wrote:
On Thu, Jul 10, 2008 at 2:13 PM, James Keane james.keane@gmail.com wrote:
Hey everyone!
Just wanted to bring to your attention http://bugs.winehq.org/show_bug.cgi?id=1348 again, and ask what is the holdup? As far as I can see it appears a solution and working patch has been posted. I am interested in getting this into the main tree and wondering what kind of modifications have to be made to the patch supplied to make it happen.
------- Comment #8 From Robert Reif 2007-02-17 15:45:56 ------- The reason I was given by Alexandre is that the functionality should be moved into the server.
Why is it not possible to accept this patch and then start working on moving it into the server? With a quick look the only changes from the patch and the git, is the change from dll/user to dll/user32.
If the correct way to do it is X, and the current implementation does it by Y, which is not correct, then why would we accept the patch that implements Y? By definition it's a hack, and if we did accept it, there'd be no incentive to fix the hack, because everything would already work (just in the wrong way).
-- James Hawkins
On Thu, Jul 10, 2008 at 4:01 PM, James Keane james.keane@gmail.com wrote:
Yes that makes complete sense, but in this case how would the implementation be better if we moved it into the wineserver? Isn't the point of having seperate drivers for X11 and quartz etc, is to take advantage of exposed interfaces ie Hotkeys?
Also, how am I as a wine outsider suppose to implement X if I don't know what it is; this is why I asked.
Let me rephrase my original question then: why is it better to move this in the server?
If more people knew the answer to your question, it would probably be implemented by now :-) Usually the reason to move something into the server is to share common functionality, with each frontend (X11, quartz) augmenting that common backend with their own functionality.
2008/7/10 James Hawkins truiken@gmail.com:
If more people knew the answer to your question, it would probably be implemented by now :-) Usually the reason to move something into the server is to share common functionality, with each frontend (X11, quartz) augmenting that common backend with their own functionality.
I imagine that in the case of hotkeys making it work across processes is a consideration as well.
The supplied patch does work across process
On Thu, Jul 10, 2008 at 6:12 PM, H. Verbeet hverbeet@gmail.com wrote:
2008/7/10 James Hawkins truiken@gmail.com:
If more people knew the answer to your question, it would probably be implemented by now :-) Usually the reason to move something into the server is to share common functionality, with each frontend (X11, quartz) augmenting that common backend with their own functionality.
I imagine that in the case of hotkeys making it work across processes is a consideration as well.
-
H. Verbeet -
James Hawkins -
James Keane