Configuration¶
Most of the time the default configuration of PyCLibrary should be sufficient. However it may not always be so. Here are some ways to tweak it to your needs.
Specifying headers and libraries locations¶
When parsing numerous headers or using PyCLibrary in an application, it might prove tedious to always specifies the full path to headers files (the same can apply to libraries if their not located in a standard location). PyCLibrary allows you to add folder in which to look into for headers and libraries.
Consider a case in which you store the headers in a ‘headers’ folder by your script, and the library into a ‘lib’ folder :
import os
from pyclibrary import add_header_locations, add_library_locations
curr_dir = os.path.dirname(__file__)
add_header_locations([os.path.join(curr_dir, 'headers')])
add_library_locations([os.path.join(curr_dir, 'lib')])
parser = CParser('my_lib_header.h')
clib = CLibrary('my_lib.so', parser)
Note
PyCLibrary does not explore sub-folders when looking for headers and library (it might in the future).
Manual initialization¶
The first time you create a CParser or CLibrary, PyCLibrary does some initialization based on your OS. It basically defines the standard known ctypes and the specific modifiers supported by the compiler (things like __stdcall for example).
You can manually initialize the CParser and the CLibrary by calling the initialization function found in pyclibrary and specifies your own types and modifiers. You can also use the auto_init function to add things on top of your OS specific stuff.