Thu 28 Dec 2006 01:45:41 AM UTC, comment #12:
A new testcase from Zou helped fixing all existing testcases we have so far.. there's probably something else that broke so we'd need new testcases for that. Namely, is there any reason why a PLACEOBJECT should replace an existing character at the same depth ?
|
Tue 26 Dec 2006 03:15:55 PM UTC, comment #11:
Sorry, I'm making much confusion here.
The root_stop_test.swf has always worked since I reverted
the patch fixing loop_test.swf, it's loop_test.swf failing.
I think making both tests use the Deajagnu.swf file will
help closely tracking expected behaviour.
For example, for root_stop_test.swf, the 'counter' variable
is never reset with Gnash, is this expected ?
|
Tue 26 Dec 2006 03:03:05 PM UTC, comment #10:
it seems the root_stop_test.swf file produced with Ming actually works fine, at least the "visual" effect is the expected one. Can you confirm ?
|
Tue 26 Dec 2006 02:32:45 PM UTC, comment #9:
I've committed the new files, please check the runner.
|
Tue 26 Dec 2006 07:07:14 AM UTC, comment #8:
>>The current CVS could not even find a named movieclip, no >>good.
I can't repeat the test, forget it. sorry.
|
Tue 26 Dec 2006 02:00:22 AM UTC, comment #7:
The current CVS could not even find a named movieclip, no good.
Merry Chirstmas!
|
Sun 24 Dec 2006 08:12:21 PM UTC, comment #6:
patch #5620 broke this, but fixed the loop_test.swf AND bombgame.swf.
In current CVS the patch was reverted so you can play elvis,
but bombgame and loop_test are broken... more after christmas :)
|
Fri 22 Dec 2006 09:35:07 AM UTC, comment #5:
I committed your test runner. It fails, of course.
I'm starting to think the fix for your loop_test was not correct,
anyway it's not easy to tell as the code in sprite_instance::advance_sprite is really complex, most likely unecessarly complex...
|
Fri 22 Dec 2006 06:56:51 AM UTC, comment #4:
Update the root_stop_test.c a little.
Attached files:
root_stop_test_update.c
root_stop_testrunner.cpp
I don't check everything in the testrunner. Add more checks if you think it's neccessary. I just want to keep it simple.
(file #11579, file #11580)
|
Wed 20 Dec 2006 04:21:28 PM UTC, comment #3:
I can work on a testrunner now, let me know if you already have some progress.
|
Wed 20 Dec 2006 09:59:19 AM UTC, comment #2:
The attached "root_stop_test.c" may take place of the elvis.swf. It is much smaller. I need more time to prepare a testrunner for it.
(file #11568)
|
Mon 18 Dec 2006 08:11:51 PM UTC, comment #1:
Reverting the patch that fixed loop_test.swf testcase fixes elvis.swf... now we'll need to find what makes these two movies different, and possibly adding a small testcase reproducing current elvis.swf bug..
|
Mon 18 Dec 2006 06:52:02 PM UTC, original submission:
Dunno since when (have to dig it, but most likely due to one of my own commits) elvis.swf starts in STOP mode and needs an explicit kick to start. This is a regression as it was working before.
|