[Cmb-users] python UCS2 or UCS4

Elijah DeLee deleeke at gmail.com
Sat Feb 11 15:40:51 EST 2017


Hello!

I am a student at UNC Chapel Hill and I have been working with Bob and John
over there at Kitware on creating a simulation template so IBAMR (
https://github.com/IBAMR/IBAMR/)  users can use Computational Model Builder
to set up their simulations.

I use IPython and Jupyter notebooks and would like to be able to import the
smtk module so I can explore and test out more of its functionality.

Unfortunately the python 2.7 packaged with CMB appears to have been
installed with 2-byte unicode characters:

$ /path/to/CMBpython

>>> import sysconfig
>>> sysconfig.get_config_var('Py_UNICODE_SIZE')
2


This is not compatible with my system Ipython kernel...I believe Ubuntu
14.04 and later have moved to 4-byte unicode characters.

Is there a way I can build CMB where python is compiled to use 4-byte
Unicode characters? I'm not sure how to set this option when building CMB
with CMake.

If I was building it from source myself I could install it like so:


./configure --enable-unicode=ucs4
make
make install./configure --enable-unicode=ucs4

My thought is that if it is installed with UCS4, then it will build/install
all the modules correctly and then I can access them with my system ipython
by appropriately using sys.path.append("path/to/smtk").

Thanks for your time,
Elijah
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/cmb-users/attachments/20170211/0a239f97/attachment.html>


More information about the Cmb-users mailing list