On Mon, 29 Jun 2020 at 15:54, Javier Arantegui
<javier.arante...@gmail.com> wrote:
>
> Hi!
Hi Javier,
> Most probably, this is a silly question...
I don't think it's a silly question
> What is the best way to import Sympy?
>
> I usually use:
>
> from sympy import *
>
> I know that using a wild import should be avoided. Is this a better way:
>
> import sympy as sp
This is more of a general Python question than a sympy-specific
question. My answer is that it really depends on the context but my
rules of thumb is that star-imports are fine if you're writing a
single script that uses a single star-import.
Are you writing a small script to do a short calculation or are you
editing a large codebase with many different modules and internal and
external imports?
In any significant codebase star-import is a bad idea. It makes it
hard to trace the origin of an imported name when looking at the code.
If I'm looking at the code and I see sp.cos then I expect that sp will
be defined or imported somewhere at the top of the file and I can just
go to the first occurrence of it in the file to see where it comes
from. Likewise if you use "from sympy import cos" and I see cos(2) I
expect that cos will be imported at the top and I can search for that.
There have been issues in sympy where __init__.py files used
star-imports making some objects importable by accident. The
conclusion from those examples is that it is bad to use star-imports
when writing modules that other code will import from. Within the
sympy codebase itself most imports are more explicit like
from sympy.polys.polymatrix import PolyMatrix
That's a good practice for internal imports in a large codebase. There
are a few places around the sympy code base where names are imported
from top level sympy like
from sympy import cos
Those are not fine *within* sympy because they lead to circular import
problems. For users of sympy though I think that this would be
considered good practice.
The above is all about large codebases though (e.g. sympy has 750000
lines of code). If you are writing a short script that just uses sympy
to compute something then I don't see anything wrong with using a
star-import for brevity. Of course there's a big range between a
10-line calculation and a 750K-LOC library and you'll have to decide
where your code fits in that range...
Also are you combining sympy with other libraries such as numpy and
are you star-importing from all of them?
Many problems of star-import come from things like multiple star
imports. Both numpy and sympy have a cos function so if you
star-import both modules which one will you get? The two cos functions
are deliberately different so it does matter. Because of that even in
a short script that mixes numpy and sympy I would go with
import sympy as sp
import numpy as np
or perhaps
from sympy import *
import numpy as np
but never
from sympy import *
from numpy import *
Both sympy and numpy have hundreds of names at top-level and many of
them are the same names.
For simple interactive calculations I use isympy which does this when it starts:
>>> from __future__ import division
>>> from sympy import *
>>> x, y, z, t = symbols('x y z t')
>>> k, m, n = symbols('k m n', integer=True)
>>> f, g, h = symbols('f g h', cls=Function)
>>> init_printing()
That saves a bunch of boiler-plate so you can quickly get started.
It's also convenient for pasting examples: I often paste code snippets
to show someone something with the implication that it works in isympy
with the above import and symbols. This is the original purpose of
star-import: to be convenient in an interactive context.
I see lots of examples around where people do different variants like
import sympy; sympy.cos(2)
import sympy as sp; sp.cos(2)
import sympy as sy; sy.cos(2)
and other less common examples. I wonder if it would be worthwhile for
sympy to endorse one of these explicitly in the way that numpy does
with np.
--
Oscar
--
You received this message because you are subscribed to the Google Groups
"sympy" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sympy+unsubscr...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/sympy/CAHVvXxRu1Y3CzHJOWJbATOsDvTnY7vZRUb5yYowZvqjQHJqVfw%40mail.gmail.com.
