Why isn't patch code written into releases??

MYGRA1N · Post by **MYGRA1N** » Wed Mar 25, 2009 4:24 am

This has kind of annoyed me lately. The FFXI patch code has been around for ages now, but it still isn't included in development releases (currently 1.1.17). Nor probably more usefully is the Fallout 3 patch which benifits others games with the VideoDriver ability. Is there actually any reason why they are not included??

jeffz · Post by **jeffz** » Wed Mar 25, 2009 6:04 am

MYGRA1N wrote:This has kind of annoyed me lately. The FFXI patch code has been around for ages now, but it still isn't included in development releases (currently 1.1.17). Nor probably more usefully is the Fallout 3 patch which benifits others games with the VideoDriver ability. Is there actually any reason why they are not included??

patches have to be sent by their respective author to the wine-patches mailing list, if they do not take the correct approach or have other problems they will not be included.

austin987 · Post by **austin987** » Wed Mar 25, 2009 10:16 am

On Wed, Mar 25, 2009 at 4:24 AM, MYGRA1N <[email protected]> wrote:

This has kind of annoyed me lately. The FFXI patch code has been around for ages now, but it still isn't included in development releases (currently 1.1.17). Nor priobably more usefully is the Fallout 3 patch which benifits others games with the VideoDriver ability?? Is there actually any reason why they are not included??

If it's the patches I'm thinking of, they're hacks and won't be
accepted. Ask the author of the patch to write proper patches with
testcases, and you'd likely see them committed.

--
-Austin

MYGRA1N · Post by **MYGRA1N** » Thu Mar 26, 2009 2:49 am

Ugh that is just stupid. It's there. It works. It's just retarded not to include it >,<

Das Letzte Einhorn · Post by **Das Letzte Einhorn** » Thu Mar 26, 2009 7:29 am

Well, my guess is that they want some uniformity in the way the patches are written. Otherwise everybody could just write a patch and send it, and tracking progress / regressions would be much more difficult. It makes sense to me to request such criteria prior to acceptation.

austin987 · Post by **austin987** » Thu Mar 26, 2009 8:17 am

On Thu, Mar 26, 2009 at 2:49 AM, MYGRA1N <[email protected]> wrote:

Ug that is just stupid, it's there. It works. It's just retarded not to include it >,<

You're free to patch your own version. But if we start including tons
of hacks, random applications break and it becomes impossible to
maintain.

--
-Austin

felix · Post by **felix** » Thu Mar 26, 2009 9:43 am

On Thu, Mar 26, 2009 at 07:29:35AM -0500, Das Letzte Einhorn wrote:

Well, my guess is that they want some uniformity in the way the patches are written. Otherwise everybody could just write a patch and send it, and tracking progress / regressions would be much more difficult. It makes sense to me to request such criteria prior to acceptation.

Actually a closer look would tell you that it is most likely the wrong
place to change the behaviour. The correct location appears to be
dlls/winex11.drv/ime.c -> ImeSetCompositionString

Also the ImeInquire contains the following code:
577 /* Tell App we cannot accept ImeSetCompositionString calls */
578 lpIMEInfo->fdwSCSCaps = 0;

Which to my untrained eyes means that wine doesn't fully support the
functionality required by FFIX, or that FFIX is assuming certain
functionality is present and is not checking for it, the end result
being that it goes into a loop trying to call something and not getting
the expected result.

The correct fix would most likely be to finish the implementation, and I
don't understand enough of x11 and ImeSetCompositionString to tell you
if that is even possible with the current x11 capabilites.

All of this suggests to me that said patch that the user wants, is
really something that hides the real bug rather than fixing the problem.
Hence as mentioned before, it's considered a hack since it doesn't
really fix the issue, it just hides it for a particular game. Adding
suck hacks into place will only result in breakage for other apps.

As mentioned above, my understanding of what's going on in the code is
limited, so take the above details with an appropriately sized pinch of
salt.

--
Darragh

"Nothing is foolproof to a sufficiently talented fool."