Roger Binns
2004-12-31 08:48:39 UTC
It turns out that the VX4500 needs 16bit BMPs. The wxWidgets source code
explicitly mentions how writing 16 BMPs is disabled because wxQuantize
only quantizes to 8 bit colour.
We basically have two choices with how to deal the increasing number of
image issues (eg the prior work Stephen did do try and make pngs be
a certain size).
One way is to continue adding external binaries in the helper directory
as we encounter the various issues.
The other is to use an imaging library. The main relevant one is PIL:
http://www.pythonware.com/products/pil/
It interacts just fine with wxPython:
http://wiki.wxpython.org/index.cgi/WorkingWithImages
There is also GraphicsMagick, the successor of ImageMagick:
http://www.graphicsmagick.org/
Unfortunately its wrappers appear incomplete or outdated, as do
those for ImageMagick.
I am leaning towards just adding more binaries to the helper
directory and adding functions to call them in conversions.py.
Roger
explicitly mentions how writing 16 BMPs is disabled because wxQuantize
only quantizes to 8 bit colour.
We basically have two choices with how to deal the increasing number of
image issues (eg the prior work Stephen did do try and make pngs be
a certain size).
One way is to continue adding external binaries in the helper directory
as we encounter the various issues.
The other is to use an imaging library. The main relevant one is PIL:
http://www.pythonware.com/products/pil/
It interacts just fine with wxPython:
http://wiki.wxpython.org/index.cgi/WorkingWithImages
There is also GraphicsMagick, the successor of ImageMagick:
http://www.graphicsmagick.org/
Unfortunately its wrappers appear incomplete or outdated, as do
those for ImageMagick.
I am leaning towards just adding more binaries to the helper
directory and adding functions to call them in conversions.py.
Roger