Instead of shifting the destination to the left, if was moving the source to
the right, which had the effect of shifting the dirty rect. So part of the
surface that should have been copied to the screen was not, and instead it
was copying part of the surface that did not need to be copied.
In Director, palette IDs in the palette channel apply to all subsequent
frames. If we seek to an arbitrary frame, the correct palette
should be applied.
Fixes the introductory cutscene in Majestic: Part 1.
Partially rolls back a change introduced in
2caa338dbddd9429c766a281231c8fb98701d968; Director relies on the notion
of updateScreen() being near-zero cost if there are no changes.
For example, starting director:henachoco05 went from taking a few
milliseconds to taking 11 seconds, and opening the top menu in
director:easternmind only works on 10% of the screens.
The CLUT in the resource fork can contain palettes unused by Director;
the key is that all of the ones used for cast members have a palette ID
of the cast ID + 1024. This is effectively what the D2 branch was
relying on, so we can clean up the logic and use it for D3 as well.
Fixes the Invictus logo in Over-Ring-Under, the colour cycling in The 7
Colors, and countless others.
SpaceMgr::m_parseText can be called multiple times, and assume the context
from the previous call exists. For example, ml.dir in The Dark Eye will
call it twice; the first time with a full hierarchy starting with a
SPACECOLLECTION, the second time with a hierarchy starting with a SPACE
and assuming _curSpaceCollection was set by the previous call.
We should be able to reuse the object's global "_cur" values for this,
as the game will override them soon after loading the model with a
SpaceMgr::m_setCurData call.
Fixes loading the scene graph data in The Dark Eye.
Previously, pathMakeRelative didn't make a distinction between relative
and absolute paths. The input would always be converted to a relative
path, then checked against a number of places (e.g. the current
working directory, all the locations in the search path...).
This causes an issue for functions such as getNthFileNameInFolder. Quite
often, this is used for absolute paths constructed with "the pathName"
as a base. ScummVM assumes the game root is also the root of the volume
(e.g. "C:\"). As such, when you would call getNthFileNameInFolder("C:\",
1), this would be translated to a relative path of "", and then match
with the current directory provided from getCurrentPath; clearly not the
same thing.
To fix this, we detect if the input to pathMakeRelative is an absolute path,
and if so avoid checking every location in the search path and prefixing
the result with getCurrentPath().
Fixes "I am unable to save or retrieve games" error message in The Dark Eye.
pathMakeRelative will attempt to look for files in locations listed in
the search path. This can include directories that don't end with a
separator. As such, ensure that paths being tested have a separator so
they don't get accidentally merged with the target filename.
Fixes the loading sequence in The Dark Eye.
They mostly come from OpenBSD 7.2 and Debian 11 manual pages, and they
shouldn't hurt since backends/networking/enet/source/unix.cpp already
includes them.
