On 12.04.2013 23:02, Aaron Meurer wrote:
We could add OrderedDict to compatibility. A pure Python version
probably won't be as fast as the one in 2.7, but it will be
functional.
For class decorators, I agree. We also have the same issue with the
recent pull request that lets you decorate functions and classes to
say that their doctests depend on external dependencies, and then
automatically run them only when those dependencies are installed. For
classes, we add an attribute to the class, but then we have to use
some descriptor magic to ensure that the attribute isn't inherited by
subclasses, because subclasses will not have the same docstring (see
http://asmeurersympy.wordpress.com/2013/04/06/how-to-make-attributes-un-inheritable-in-python-using-descriptors/).
This is a little clumsy unless you have class decorators.
Does anyone want to volunteer to do a release, so that we can get one
more release out with Python 2.5 support? I think we should also
definitely do a release before we attempt to remove the old
assumptions. Tom, would you be willing to donate some time to this? I
personally don't have the resources to do a release at least until May
when classes end.
Willing yes, able rather not. Well actually I do not know how much work
this is, but since you have done it many times and try to avoid it at
all costs, I guess a lot.
The point is, in all honesty, investing more than two hours/day on sympy
isn't going to be possible before for me my exams end (and two hours is
already quite a stretch), i.e. beginning of june. And currently all of
these two hours goes into investigating stuff for my proposal (and
writing it).
I could probably try to do some work on this once the application phase
is over, but expect it to be slow ...
And I also don't want to do one until
https://code.google.com/p/sympy/issues/detail?id=3445 is fixed. If
anyone out there is interested in working on that idea for GSoC,
helping to do a release now would look great for your application.
Otherwise, let's just drop support right away.
Aaron Meurer
On Fri, Apr 12, 2013 at 12:31 PM, Mateusz Paprocki<[email protected]> wrote:
Hi,
On 21 January 2013 09:15, Aaron Meurer<[email protected]> wrote:
I was looking at what benefits we would get from dropping Python 2.5
support (other than the obvious no longer having to test on it). If
we dropped support, we would be able to use any feature that was
introduced in Python 2.6. Here's a page that lists all of them:
http://docs.python.org/2/whatsnew/2.6.html.
I don't see anything particularly compelling. There are some syntax
changes that are annoying to work around (such as f(*args,
keyword_argument=True, **kwargs), "except Exception as e", or binary
literals). Most of the stuff seems mainly useful to people who want
to use the same code base for Python 2 and Python 3. I did notice the
following features:
- Abstract base classes
- New-style string formatting (like "The number is {x}".format(x=1))
- Class decorators
- Standard library fractions module (meaning we can use this directly
in the polys again, assuming it isn't slower)
Can anyone think of good uses we might make of these or other Python
2.6+ only features in SymPy?
A little late in the discussion, but yes please, lets drop 2.5 as soon as
possible. I have a few serious applications for class decorators (e.g.
automatic pollution of __all__ without repeating yourself). Also those
syntax changes changes will help reducing confusion and dirty workarounds.
We don't have to use `fractions` because we have our own PythonRational and
`fractions` is pure Python with ABCs, so there will be no gain (in terms of
speed). I could take advantage of OrderedDict, but that's 2.7.
Aaron Meurer
--
You received this message because you are subscribed to the Google Groups
"sympy" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/sympy?hl=en.
Mateusz
--
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 [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sympy?hl=en-US.
For more options, visit https://groups.google.com/groups/opt_out.
--
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 [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sympy?hl=en-US.
For more options, visit https://groups.google.com/groups/opt_out.
