ansaurus

Question

Answer 1

+1 A:

You mean, something like...:

>>> import ctypes
>>> x = ctypes.create_string_buffer('howdy!')
>>> x.value
'howdy!'
>>> x[0] = 'C'
>>> x.value
'Cowdy!'

...?

Alex Martelli 2010-03-19 20:20:16

Yes, like that, thanks. But how can I put an integer value in there (i.e. without having to look up the ascii symbol)? For example, if I want to set a byte to 0xff, how could I do that?

Chris 2010-03-19 20:37:12

Oh wait, I figured it out:>>> x[0] = "\x43"I was trying this with "xff", but the printout was throwing me off by actually showing "xff". It makes more sense when you print it after using a code for a letter.

Chris 2010-03-19 20:42:59

@Chris, if you have a small integer in a variable `i`, `chr(i)` gives you the corresponding 1-byte character; the `struct` and `array` modules offer other ways to convert various types to byte strings.

Alex Martelli 2010-03-19 20:55:05

Answer 2

A:

You may find that Cython is a lot nicer then the ctypes module for melding C libraries with Python code.

Mike Graham 2010-03-19 22:55:44

Hmm... that does sound useful. We have a lot of python already written, though, and I don't necessarily want to worry about Cython/Python incompatibilities. But I could build Python-callable modules with Cython, right? Might be an option, and might be a little less crazy when it comes to wrapping these functions.

Chris 2010-03-22 15:45:25

Cython modules can be imported from Python; the main use of Cython is to make Python modules that wrap C libraries. Typically you would use Cython to do the wrapping, making a native-feeling API and write the main part of your code using the wrapper you wrote.

Mike Graham 2010-03-22 21:46:46

ansaurus

tags:

views:

answers:

ctypes buffer modification

related questions