matplotlib-devel Mailing List for matplotlib

On Sat, Apr 12, 2008 at 2:19 PM, Eric Firing <ef...@ha...> wrote:
>
> referring to:
> http://news.gmane.org/find-root.php?message_id=%3cc7009a550804100055g6388b20ej520e85d8e679a55%40mail.gmail.com%3e
>
> There was a slightly jumbled set of threads on Numpy-discussion that
> included a discussion of standardizing imports, like this:
>
> import numpy as np
> import scipy as sp
>
> and then this was added, as something else that had been agreed at the
> same sprint:
>
> import pylab as plt
>
> I think this is a mistake, and it should have been
>
> import matplotlib.pyplot as plt
This is what we agreed to (import matplotlib.pyplot as plt) during the
numpy sprint (I was on google chat remotely but participated in the
discussion). I agree that pylab as plt would just add to the
confusion. We agreed on promoting (and I currently use)
import numpy as np
import scipy as sp
import matplotlib.pyplot as plt
JDH

referring to: 
http://news.gmane.org/find-root.php?message_id=%3cc7009a550804100055g6388b20ej520e85d8e679a55%40mail.gmail.com%3e
There was a slightly jumbled set of threads on Numpy-discussion that 
included a discussion of standardizing imports, like this:
import numpy as np
import scipy as sp
and then this was added, as something else that had been agreed at the 
same sprint:
import pylab as plt
I think this is a mistake, and it should have been
import matplotlib.pyplot as plt
because, for most production code, we don't want to be accessing numpy 
indirectly via pylab. I suspect it was just a slip-up by the poster, 
but since I was not involved in whatever discussion led to "plt", and 
the thread was going in several directions at once, I did not want to 
jump in.
We should get this straight, however. Note that the point is not to 
tell users that they must use a particular form, but to try to 
standardize the usage in the projects themselves, including examples and 
documentation, so as to reduce confusion among new users.
Eric