Question

I am able to successfully call a function with ctypes in Python. I now have a buffer that is filled with Structures of data I want to extract. What is the best strategy for this? Anything else I should post?

Function:

class list():
    def __init__(self):
        #[...]

    def getdirentries(self, path):
        self.load_c()
        self.fd = os.open(path, os.O_RDONLY)
        self.statinfo = os.fstat(self.fd)
        self.buffer = ctypes.create_string_buffer(self.statinfo.st_size)
        nbytes = self.statinfo.st_size

        transferred_bytes = self.libc.getdirentries(
                        self.fd,
                        ctypes.byref(self.buffer),
                        nbytes,
                        ctypes.byref(self.basep)  )

    #[...]

Structure:

class dirent(ctypes.Structure):
    _fields_ = [ ("d_fileno", ctypes.c_uint32), # /* file number of entry */
                 ("d_reclen", ctypes.c_uint16), # /* length of this record */
                 ("d_type", ctypes.c_uint8), # /* file type */
                 ("d_namlen", ctypes.c_uint8), # /* length of string in d_name */
                 ("d_name", ctypes.c_char * (MAXNAMELEN + 1) ) ]

Some Output:
Transferred bytes: 156
sizeof buffer: 272
Buffer: <ctypes.c_char_Array_272 object at 0x8c3f0>

Answer 1

I wonder why you are using os.stat() instead of calling statinfo and os.path.walk() instead of calling getdirentries?

Normally, when you have buffers of data that you want to pass in and out of C, you would use the struct modules pack and unpack methods to do this.