Windows > Software Development for Windows Client Forums > DirectShow Development > How to capture closed captions correctly?
Ask a questionAsk a question
 

QuestionHow to capture closed captions correctly?

  • Friday, November 06, 2009 9:47 AMdims12 Users MedalsUsers MedalsUsers MedalsUsers MedalsUsers Medals
     
    I want to capture (not to display) closed captions.

    I wrote my own filter, based on Dump example. The only thing I modified is mediatype of input pin.

    I wrote an application which does the following:

    1) Creates graph builders (normal and capture)
    2) Finds a capture device. The criterion is a category and a presense of audio, video and VBI output pins.
    3) Finds a Tee Sink device. The criterion is a category.
    4) Finds CC decoder device. The criterion is a category and a presense of pins.
    5) Finds other required devices by capture graph builder method

    I render audio output pin of capture device, then connect VBI output to Tee Sink input, then connect Tee Sink output to CC decoder, then connect CC decoder output to my filter's input.

    Also I use crossbar to turn sound on.

    No errors occure, I hear sound, I can switch channels, but my filter creates empty files.

    Is all above correct in general?

    How program should behave in the case of CCs absent? Should it write zeros or just do nothing? I see no breaks in debugger within CGrabberCCInputPin::Receive method.

    How my filter can send captured CCs back to application with the help of events?

    What is IMediaSample structure in the case of CCs?

    Thanks.

All Replies

  • Tuesday, November 10, 2009 10:02 AMdims12 Users MedalsUsers MedalsUsers MedalsUsers MedalsUsers Medals
     
    It looks like it is necessary to render video capture pin to capture CCs.

    I.e. modular architecture of DS is a fake, because it is imposible to use one part without another.