I believe there is a function call now available for getting the size of widechar, however that will not work if there is an old version of liblouis installed. Regarding converting from 16-bit and 32-bit unicode, in jlouis I used the utf-16 and utf-32 encodings. However I do caution the use of that system as liblouis uses ucs-2 not utf-16 (ucs-2 is a fixed 2 byte encoding where as utf-16 is variable length so can accept characters which require 4 bytes to represent, I don't know what would happen if such a character were passed to liblouis). On 11 May 2011, at 22:39, liblouis@xxxxxxxxxxxxxx wrote: > > Comment #1 on issue 15 by ja...@xxxxxxxxxxxx: Python wchar_t and UTF-16 > mismatch > http://code.google.com/p/liblouis/issues/detail?id=15 > > I realise liblouis allows the widechar size to be configured separately, but > why would you want to compile for 16 bit widechars on a system which has a 32 > bit wchar_t? > > If we do indeed need to support this, we're going to need two things: > 1. A way to programmatically determine what widechar size liblouis was > compiled with; and > 2. A way to handle the alternative widechar type in ctypes, which (as far as > I can tell) only handles wchar_t and char out of the box. I'm not sure how to > do this or if it's even possible. ctypes does provide functionality to > convert between C and Python types, so it should be possible, but I'm not > certain. > > For a description of the software, to download it and links to > project pages go to http://www.abilitiessoft.com For a description of the software, to download it and links to project pages go to http://www.abilitiessoft.com