[SciPy-dev] Namespaces in documentation

Fernando Perez fperez.net@gmail....
Tue Jun 3 15:52:47 CDT 2008

On Tue, Jun 3, 2008 at 11:40 AM, Robert Kern <robert.kern@gmail.com> wrote:
> Stéfan, Pauli,
> Make a decision on what you want. Stick with it. Don't let it be
> brought up for discussion. This is pointless bikeshedding that doesn't
> help improve the documentation effort in the slightest.
> Everybody else, please just drop it. If you feel like worked up enough
> to write more emails, *write a docstring instead*.

For the record, I've just added to the trunk of ipython a 'numpy'
profile that exposes np, the 'scipy' one now exposes np/sp, and the
-pylab option exposes np/plt (but *not* scipy, since many matplotlib
users do not have/want scipy all the time).  I'm sorry I did it
without voting, but I'm fed up with this discussion.

People who use ipython are free to benefit from this
convention/standard/edict or free to ignore it.  The changes should be
visible a bit later today when I push my private tree upstream.



ps - The unqualified names are also always available
(numpy/scipy/pylab/pyplot are always imported anyway).

More information about the Scipy-dev mailing list