Python VideoFilter Plugin. Buffer None in do_transform_frame

[hrojas at lacuatro.com.ar]

Hello,
         I am trying to develop a custom video filter plugin but I am 
getting an empty buffer in VideoFrame parameter.

The code is as follows:

     def do_transform_frame_ip(self, f):
         try:
             data = f.buffer.extract_dup(0, f.buffer.get_size())
             bgr = numpy.ndarray((1080, 1920, 3), buffer=data, 
dtype=numpy.uint8)
             print bgr.shape
         except Exception as e:
             print e

         return Gst.FlowReturn.OK

VideoFrame metada is fine, but Buffer is None


Testing with the following pipeline:
(mysink is the example provided in gst-plugin source code)

GST_DEBUG=python:6 gst-launch-1.0 filesrc 
location=/media/storage/videos/Mirch2.avi ! decodebin ! videoconvert ! 
miplugin ! mysink
Estableciendo el conducto a PAUSA ?
El conducto está PREPARÁNDOSE ?
libva info: VA-API version 0.38.1
libva info: va_getDriverName() returns 0
libva info: Trying to open /usr/X11/lib/dri/i965_drv_video.so
libva info: Found init function __vaDriverInit_0_38
libva info: va_openDriver() returns 0
Got context from element 'vaapidecode': gst.vaapi.Display=context, 
gst.vaapi.Display=(GstVaapiDisplay)NULL;
Redistribuir latencia?
0:00:00.421544040 28703 0x7f0464005a80 ERROR                  libav :0:: 
Context scratch buffers could not be allocated due to unknown size.


Buffer is reaching OK at the method do_render() in mysink which 
implements BaseSink()

Any help will be much appreciated.

Hernando