IPCapture v0.3.1 is out

As the title suggests, a new release of the IPCapture library is available for download at the following URL:

https://github.com/singintime/ipcapture/releases/tag/0.3.1

To install, simply download the IPCapture_0.3.1.zip archive and unpack it into your Processing libraries folder.

This new release gave me the chance to migrate the project from Google Code (shut down quite a long ago) to GitHub. The project repository is now available at http://github.com/singintime/ipcapture.

Please let me know if and how it works, any feedback is greatly appreciated.

Cheers! Stefano

Comments

  • Quick update: the library is also available at

    http://www.stefanobaldan.com/projects/ipcapture

    I'm also trying to comply with the Processing guidelines on writing libraries in order to have IPCapture listed on the Processing website and in the content manager.

  • Hi Stefano,

    Thanks for sharing your library! I've been using it for the last couple days and it has saved me a great deal of time. Here are two things I've observed while using it:

    • ArrayIndexOutofBounds : I think you are already aware of this problem but when leaving the IPcapture instance running over several hours the read() method eventually crashes with the message above. It happened to me after about 4 hours running 6 IPcapture instances simultaneously. I'm currently trying to catch the exception and restart the capture as a workaround but it takes a while to happen again so I don't know if this will work or not...

    • IPcapture.pixelWidth and IPcapture.pixelHeight vs. IPcapture.width and IPcapture.height: I noted that when I try copying a frame to a buffer by using buffer[x] = cam.get() the frame doesn't get copied properly when the screen resolution and the MJPEG stream resolution are not the same. When starting a new instance of IPcapture, pixelWidth and pixelHeight get initialized with the screen resolution while the width and height parameters get initialized with the MJPEG's resolution. Forcing pixelWidth and pixelHeight to the same value fixes the problem.

    Again, many thanks!

Sign In or Register to comment.