I've been trying to extend sgefant's work on "sound in the sequence editor" by drawing the sequence box with the waveform drawn on it for cueing.
However, the routines in editsound.c are giving me fits. I can't seem to find a reliable way to determine if the bSample associated with any particular bSound is valid. Even if I call sound_load_sample(sound) everytime I try to access the data, I eventually get garbage back in the sample data (and I end up throwing a floating point exception as I try to use it).
I think part of the problem stems from the fact that sgefant's code doesn't treat a sound sequence as a sound (the ones listed in G.main->sounds) but I'm not sure.
We have added code to writefile.c and readfile.c that seem to be saving/loading the bSound struct. Do we need to save the bSample too? I don't think so, since it's transient, but I could be wrong.
Bob (aka Det. Thorn)