| Age | Commit message (Collapse) | Author |
|---|
|
|
|
Thanks Greencis for reporting it
|
|
|
|
|
|
|
|
(instead the numerous classes using it)
SmackerPlayer now creates a surface in the constructor and deletes it in the destructor. There will be only this surface during the lifetime of the player, so there shouldn't be any issues any more when the player is reused for different Smacker files. This is imo nicer than to remove/add the surface.
I kept the several openSmacker since it wraps some code which is nice.
|
|
|
|
|
|
Thanks Greencis for reporting it
|
|
|
|
Thanks Tomaso for reporting the problem
|
|
It didn't always reach va_end(), but rather than adding a call it
makes more sense to just drop the varargs from it, since it didn't
use it anyway.
|
|
|
|
These are shown on each scene change, but should no longer be needed
to be enabled by default, since we can now use the debugger to check
and change the current scene
|
|
|
|
The crash was caused by stale pointers, exactly the same as in commit
2e4f64066da7c08edcd17b5c66c6463887b836b4
|
|
|
|
|
|
|
|
|
|
|
|
I'm almost certain these are intentional fallthroughs. It makes
sense to handle the little-endian cases like the big-endian ones,
with the added FLAG_LITTLE_ENDIAN.
|
|
|
|
|
|
This commit changes the comment to use the same variable names as we do in the
code. Furthermore, it also makes the comment a bit easier to grasp.
|
|
|
|
The tables only contain (2^bitPrecision)/2 entries. The code allocated twice
as many entries previously.
|
|
This is a regression from f4ba8a6485b097a8ef1e2004d1af127243f379f1. The
commit replaced the static cosine and sine tables with dynamically created
ones. In the process of that a copy&paste error happened which made the sine
table use the layout of the cosine table. This commit now changes the
dynamically created sine tables to conform to the layout of the previous
static tables.
|
|
VIDEO: Wrap 8-bit smacker audio properly
|
|
Removed a call to BaseFileManager::registerPackages() in
WintermuteEngine::getGameInfo(), as the constructor for BaseFileManager
already calls registerPackages() once; another call should thus be
superfluous.
|
|
FSNode::getChildren() was called assuming that it returns 0 iff success.
Actually, it's the other way around.
|
|
|
|
|
|
From: Petr Salinger <Petr.Salinger@seznam.cz>
Bug-Debian: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=711320
Bug-ScummVM: https://sourceforge.net/tracker/?func=detail&atid=418820&aid=3614268&group_id=37116
|
|
Thanks to 'onlyjob' on pull request #337 for pointing out an instance of this.
|
|
This reverts the changes done to the DW2 entries in commit d4a354c1.
We only distinguish between floppy and CD versions if a game had both a
CD and a floppy version released. DW2 was only released as a CD version,
so the extra flagging of the DW2 entries as CD was redundant, and caused
confusion regarding previous saved games to users that readded the DW2
detection entries after this change
|
|
Thanks greencis for reporting it
|
|
|
|
The memory was allocated with Memory::alloc(), so it should be
freed with Memory::dealloc(), not delete.
|
|
|
|
Verify that 'bits' really is one of 8, 12 or 16 before decoding
the data. It's probably always the case (unless the data files are
damaged) but if it isn't we'll either try to queue NULL to the
audio stream, or queue the same buffer more than once, or free the
buffer more than once. All of which are bad, though Coverity only
noticed the last of these cases.
|
|
Though shouldn't it also alert the user to the failure?
|
|
Unfortunately, I'm not sure where and when this function is called.
But this should be correct. CID 1003575
|
|
Leaving the panel in scene 1105 only worked if the panel was open.
If the panel was closed, nothing would happen on the first click,
and on the second it would crash. If I understood this correctly,
it was because trying to leave while the panel was closed would
(incorrectly) mark it as open, and then it would crash when trying
to remove the objects behind the panel from the scene.
|
|
CID 1003018
|
|
SCUMM: iMuse - Clean-up of Roland GS code (updated)
|
|
|
|
CID 1002106
|
|
It seems pretty clear to me that these fallthroughs were not
intentional, seeing how one of them could cause ITE to fall through
to IHNM-specific code.
CID 1003744, 1003745
|
|
The effect of the removed code was to bypass the potion-mixing
puzzle while shrunken. The puzzle works fine for me, so there's
no reason to keep this bypass. Also, it caused a dead code
warning, CID 1022345.
|
