It turns out that at least one video in Starship Titanic, for the
Lift Indicator, has only a single transparency frame in track 2.
The added code, therefore, when it doesn't find an index entry
for the desired frame number, works backwards until it finds a valid
frame (likely frame 0), and then scans forward. If it hits the end
of the video, then it simply uses whatever last frame it last decoded.
This is needed for Starship Titanic, where videos can have a secondary
video track. It was simpler to use the callback as a means to select
one video track each across two decoders than trying to make VideoDecoder
and/or AVIDecoder support decoding from multiple video tracks simultaneously
Starship Titanic in particular needs this, since some of the videos
have extra junk at the end of the file, such as ycursors.avi,
and parsing fails if we don't respect the filesize field
Relying on the videos to have 'initial frames' for audio tracks is not the best way to handle AVI videos. Now videos without initial frames (or broken interleaving) will buffer properly.
This is used in German release of Full Pipe. Unfortunately our
current MP3 decoder cannot work with streamed MP3s, and bails out
at the AVI header since there is no full MP3 header yet.
Truemotion uses its own demuxer and seems to follow its own AVI rules. Work around it by coercing the video's dimensions to use the codec's internal dimensions.
Reimplementation of 7a49802c01
This is based on a suggestion made by clone2727, so the original
idea belongs to him.
Engines can now override the common AVI audio track handler with a
custom one. This is needed for the Z-Engine AVI videos, since they
use a custom audio decoder that is only used in the two Z-Engine
games, and has its own fake AVI audio format (17). This clashes with
the MS IMA ADPCM format, and therefore shouldn't pollute the common
AVI video decoder code. The addition of this over-ridable method
allows the Z-Engine to add its own custom AVI decoder while avoiding
code duplication.