Damn, I'd promised it but never did it. I didn't want anyone to see the messy experimental code.... very sloppy and I made up functions that could have been handled by native code - I didn't know they existed, like BinaryDecode.
I discovered that later when I was checking out (late) Mark Smith's code for his audio waveform stack. I was too ashamed of my short cuts and didn't want to re-write the thing. It was also about "protecting my secret" at first, but I don't feel that way now... anybody could have done this. It just took heaps of trial and error, I had no help from anyone. Also it was incomplete, I could only get AIF conversion to work , WAV gave me problems, so I stopped when I proved audio could be exported. The specs for AIF and WAV are out there (although parts are vague). The hardest thing was to try to understand the way Rev handled audio files of different types and sample rates and the function of the numbers at the end of the file. Audio in LiveRev is VERY basic to begin with. Sample rates limited up to 48k. Sample depth limited to 16 bit. Stereo is interleaved. No compressed format can be stored, except the really old .au format, and I failed to export those, although I spent an inordinate time investigating and trying. LiveRev stores an exact, full size binary of the audio file after the headers are stripped no matter what the source was, so to recreate the file, the appropriate headers and jump tables need to be re-created. The audio type doesn't seem to be part of the spec that is stored so that must be derived from the suffix of the name of the stored audio clip. Perhaps these are the reasons why the creators of the engine gave up on the idea of exporting. I also found you don't want to create custom properties imbedded in an audio clip, even though the engine doesn't try to prevent that; it just bungs up the audio file, sometimes disastrously! I was just about to share with the late Mark Smith, and he was excited about the prospect. We were going to collaborate when I found out he had died. Lastly there was talk about improved AV in Livecode, so I shelved the project, and the client that brought up the possibility of wanting it disappeared, so I moved on. Now that the engine is open sourced, perhaps some of the mysteries are now in the open, but how can I use knowledge of the audio system to make a commercial application using this ? Gives me a headache. I think I promised Jerry a copy but forgot if I did or not; apologies if I didn't. He was really the only other person that asked besides you, Al... I need to see if these stacks even work any more in the latest version, I haven't checked since my mom passed. I will report back. sqb On Sun, Jun 9, 2013 at 12:38 PM, Alejandro Tejada <capellan2...@gmail.com>wrote: > Hi Stephen, > > > Stephen Barncard-4 wrote > > the copy command can obtain the binary image of any object, which then > > resides in clipboardData[ objects] and can be stored as a custom > > property > > or a binary 'blob' for your db > > This is what I did to create a way to export sound clips from a stack to > > see if it could be done. > > Did you released the script that allows to export > embedded sounds from stacks? > > Al > > > Stephen Barncard San Francisco Ca. USA more about sqb <http://www.google.com/profiles/sbarncar> _______________________________________________ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
