[Python-Dev] pyc/embedded c++ efficiency wrt binary data (original) (raw)

Patrick Stinson ajole-1 at gci.net
Sun Apr 11 17:59:14 EDT 2004

I'm using pyuic, a tool that comes as part of pyqt, that reads images and creates a python file with the data embedded in it for easy loading. This is based on a similar tool called qembed that embeds image data in C++ header files. Well, I have found that upon loading the py (9.2MB) file, it takes quite a while, but a pyc file is created, and is approx. 1/4 the size. This gives me the indication that the binary encoding is quite good, and importing the pyc file will takes considerably less time. Does anyone have any idea how the binary encoding of pyc files is compared to something like this done in C++? Of course, I'm not asking about the ultimate efficiency including processing the binary data as pixmaps, but just the loading from disk bit.

Cheers! -P

More information about the Python-Dev mailing list