libretro/scummvm
1
0
Fork 0
mirror of https://github.com/libretro/scummvm.git synced 2024-12-11 19:54:03 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

COMMON: When flattening QuickTime edit lists, don't flatten non-silent audio chunks into silent chunks or vice versa.

Browse Source 
Fixes missing vidbot VO in Obsidian Japanese.
This commit is contained in:
elasota 2023-08-20 18:08:18 -04:00 committed by Filippos Karapetis
parent 93d8dcadcd
commit 92320d616d
1 changed files with 27 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

33
common/formats/quicktime.cpp
View File

@ -895,14 +895,35 @@ void QuickTimeParser::flattenEditLists() {
//
// Other players seem to just play the audio track chunks consecutively without the
// 30-sample skips, which produces the correct results, not sure why.
//
//
// We also need to account for mixed silent and non-silent tracks. In Obsidian's
// Japanese localization, the vidbot that you talk to at the end of the maze (asset
// 4375) has a brief silent edit followed by the actual audio track. If we
// collapse the audio track into the silent edit, then it causes the entire track
// to be silent.
for (Track *track : _tracks) {
if (track->editList.size()) {
EditListEntry &lastEntry = track->editList.back();
EditListEntry &firstEntry = track->editList.front();
firstEntry.trackDuration = (lastEntry.timeOffset + lastEntry.trackDuration);
if (track->editList.size() >= 2) {
Common::Array<EditListEntry> newEdits;
track->editList.resize(1);
for (const EditListEntry &curEdit : track->editList) {
if (newEdits.size() == 0) {
newEdits.push_back(curEdit);
continue;
}
EditListEntry &prevEdit = newEdits.back();
bool prevIsSilent = (prevEdit.mediaTime == -1);
bool currentIsSilent = (curEdit.mediaTime == -1);
if (prevIsSilent != currentIsSilent) {
newEdits.push_back(curEdit);
continue;
} else
prevEdit.trackDuration += curEdit.trackDuration;
}
track->editList = Common::move(newEdits);
}
}
}