On Sat, May 5, 2012 at 3:52 PM, Joachim Durchholz <[email protected]> wrote:
> Am 05.05.2012 12:18, schrieb [email protected]:
>
>> is a two line change in functions.py and does not cause any test failures
>> at least for ./bin/test core and ./bin/doctest core.
>
>
> Run all the tests.
> Such a change might have effects further down the class hierarchy.
>
> If I understood what these parts of the class hierarchy are doing, I'd be
> able to say something more substantial.
>
> In general, if people have to ask what the purpose of a class is, then the
> class isn't well-documented enough. In particular if it turns out that
> different people have, obviously, NOT asked and everybody has just been
> assuming some interpretation - that's the point at which different people
> start making different assumptions, everybody coming to different
> conclusions about where some new feature belongs, and in the end, "not
> well-documented" has turned into "not well-defined".
>
> So while I can't judge whether Stefan's idea is good or not, but I think it
> would be easier to judge if we knew what the purpose of each core class is.
> And the best way to distribute that knowledge is that each class states that
> in a comment. If that's there, everybody knows where a class should be
> heading even if it's currently not there.
>

I agree, except "in a comment" should be replaced with "in the
docstring".  Documentation in the core in general is rather poor.  It
was Ronan's idea to split Basic and Expr in the first place, so
perhaps he can shed some light on what he thinks Expr should and
shouldn't be.

Personally, I think it all boils down to the discussion from
https://github.com/sympy/sympy/wiki/canonicalization.  If we want to
say two Exprs can be put together in an Add, for example, then it
matters if we let Add let its args define their own behavior on it or
not.

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.

Reply via email to