OverflowError: can't convert negative value to size_t

[northwarks]

Hi Guys -I'm having issues with SetImage(), I can display text with no issues at all, as soon as I try to add an image I get the following error:

  File "matrix_v2.py", line 134, in DisplayUpdate
  mycanvas.SetImage(image, 0, 0)
  File "core.pyx", line 22, in rgbmatrix.core.Canvas.SetImage
  File "core.pyx", line 46, in rgbmatrix.core.Canvas.SetPixelsPillow
  OverflowError: can't convert negative value to size_t

I'm calling DisplayUpdate() from a thread every second, no other thread or function is writing to the matrix. I'm a little unsure if I'm actually understanding the correct way of implementing images but I've tried for hours and it always boils down to the OverflowError: can't convert negative value to size_t error despite PIL reporting the image as 50x50 RGB.

  def DisplayUpdate(self):
    mycanvas = self.matrix.CreateFrameCanvas()
    #local_weather={'mylocal_temp': None, 'mylocal_humid': None, 'mylocal_status': None, 'mylocal_icon': None}
    #local_time={'curr_day': None, 'curr_time': None, 'curr_sec': None}
    _sec = local_time.get('curr_sec')
    _day = local_time.get('curr_day')
    _time = local_time.get('curr_time')
    _temp = local_weather.get('mylocal_temp')
    _humid = local_weather.get('mylocal_humid')
    _status = local_weather.get('mylocal_status')
    _icon = local_weather.get('mylocal_icon')
    mycanvas.Clear()
    graphics.DrawText(mycanvas, self.font1, 10, 20, self.blue, _sec)
    graphics.DrawText(mycanvas, self.font3, 34, 40, self.blue, _day)
    graphics.DrawText(mycanvas, self.font1, 38, 30, self.blue, _time)
    graphics.DrawText(mycanvas, self.font3, 3, 60, self.blue, _temp)
    graphics.DrawText(mycanvas, self.font3, 55, 60, self.blue, _humid)
    graphics.DrawText(mycanvas, self.font3, 100, 60, self.blue, _status)
    image = Image.open(_icon)
    image = image.convert('RGB')
    image.load()
    mycanvas.SetImage(image, 1, 1)

Any pointers or skeleton code would be really appreciated :)

[hzeller]

it looks like you are createing a FrameCanvas every time you call DisplayUpdate ? This will allocate a new frame every time so you will run out of memory.

You should only create a frame canvas once, then swap on vsync.

As with the size_t error: can you figure out where that message is coming from ? It looks like somewhere from the inside of the generated Python-wrapper.