diff options
| author | Martin Kiewitz | 2016-02-19 02:00:03 +0100 |
|---|---|---|
| committer | Martin Kiewitz | 2016-02-19 02:00:35 +0100 |
| commit | cca9fc918f9d9c89b5526cce43ea42a9a003735e (patch) | |
| tree | 1ff38ebf242cec9003c9e31f37834f8a680eda39 /engines/agi/checks.cpp | |
| parent | c1dacbe65d4fa72128f7f521326beb0838784ac2 (diff) | |
| download | scummvm-rg350-cca9fc918f9d9c89b5526cce43ea42a9a003735e.tar.gz scummvm-rg350-cca9fc918f9d9c89b5526cce43ea42a9a003735e.tar.bz2 scummvm-rg350-cca9fc918f9d9c89b5526cce43ea42a9a003735e.zip | |
AGI: Save screen object loop_flag
loop_flag was previously vt.parm1, which was shared for multiple
uses. Was split up during graphics rewrite in commit
8a595e7771aa89d06876e13d7ab6751e26da8982
Is indirectly part of bug #7046. Saving, restarting ScummVM and
restoring right after grabbing the eagle resulted in the glitch
not happening (which was of course an inaccuracy anyway). This
was caused by AGI currently not saving/restoring the loop_flag.
Needs to get further figured out what's exactly happening
internally and if this issue was just hidden by the shared
vt.parm1 in previous versions. If triggered, it would have
just set another pseudo-random flag on end-of-loop.
Diffstat (limited to 'engines/agi/checks.cpp')
| -rw-r--r-- | engines/agi/checks.cpp | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/engines/agi/checks.cpp b/engines/agi/checks.cpp index 1e1670f674..c67b6a5810 100644 --- a/engines/agi/checks.cpp +++ b/engines/agi/checks.cpp @@ -129,7 +129,7 @@ bool AgiEngine::checkPriority(ScreenObjEntry *screenObj) { screenPriority = _gfx->getPriority(curX, curY); if (screenPriority == 0) { // unconditional black. no go at all! - touchedControl = 0; + touchedControl = false; break; } |