Display of zero-length spectrograms in viewer fails on Windows.

[HaroldMills]

Bill Evans recently showed me that the viewer clips window becomes nonfunctional if one looks at CLC noise clips for the night of 2014-09-02 in an example data set he is preparing. I was unable to reproduce this problem on Mac OS X, but I do see it on Windows 8 (Bill uses some version of Windows).

The problem appears to be related to zero-length spectrograms. The CLC clip with time 2014-09-03 01:02:08 has only 113 samples (the .wav file has 270 bytes), and so its spectrogram has zero spectra for the default window size of 128 samples. When I try to display such a spectrogram in the viewer (on Windows, but not on Mac OS X), an exception is raised from within Matplotlib and I see the following stack trace:

Traceback (most recent call last):
  File "C:\Users\Harold\Anaconda\lib\site-packages\matplotlib\backends\backend_qt5.py", line 338, in resizeEvent
    self.draw()
  File "C:\Users\Harold\Anaconda\lib\site-packages\matplotlib\backends\backend_qt5agg.py", line 148, in draw
    FigureCanvasAgg.draw(self)
  File "C:\Users\Harold\Anaconda\lib\site-packages\matplotlib\backends\backend_agg.py", line 456, in draw
    self.renderer = self.get_renderer(cleared=True)
  File "C:\Users\Harold\Anaconda\lib\site-packages\matplotlib\backends\backend_agg.py", line 473, in get_renderer
    self.renderer = RendererAgg(w, h, self.figure.dpi)
  File "C:\Users\Harold\Anaconda\lib\site-packages\matplotlib\backends\backend_agg.py", line 94, in __init__
    self._renderer = _RendererAgg(int(width), int(height), dpi, debug=False)
ValueError: width and height must each be below 32768

[HaroldMills]

In the PyDev debugger I see that the width of the figure canvas for the zero-length spectrogram (width in the call to _RenderAgg in the stack trace of the first issue comment) is -3. The particular value of the width may depend on the size of the clips window, but it's a problem that it's negative. The size that is assigned to the figure by the viewer (method _lay_out_clips of class _FiguresFrameWithFlowLayout_) is 3 pixels.

Regardless of this problem, it might be best for a clip figure to have a minimum (positive!) width (in pixels) to ensure that it is minimally noticeable. I'll need to think more about how best to accomplish this. This will incidentally function as a workaround to the current problem as long as we make the minimum width large enough.

[HaroldMills]

I've been planning to switch from PySide to PyQt, so in case that would help with this problem I went ahead and did that today. It did not help.