Excuse me, I did not write sooner. It was nessesary to do all thing
related to university. Now Im afraid I have no time. But anyway,

> OK.  What exactly would like to implement?
I red this issue https://github.com/sympy/sympy/wiki/Parsing
Well, the task is to implement a parser framework for sympy that will
support unambigious grammars. Unfortunately, I have never been in
touch with Earley algorithm. So I thought, final approach is to
implement flexible parse classes that can potentially describe
grammars of text formats which uses Mathematica6/7/8, Maxima, and
LaTeX style input, etc. Common pattern for this tasks that I know is
Haskell-like monadic parse combinators, but they can give low
perfomance in python.

Also, I choose this issue
http://code.google.com/p/sympy/issues/detail?id=1374&colspec=ID%20Type%20Status%20Priority%20Milestone%20Reporter%20Summary%20Stars&start=100
but have faced with a problem. There are several ways to fix it:
1) assume that Float(2.0) should return Integer(2), but its a
`horrible' way, because Float constructor expected to return Float
values. And tests show that many modules rely on behavior of Float. I
decided not to modify standard sympy type system.
2) fix Add.flatten function. It uses `terms' dictionary for reduction,
and hash(x**2) is not equal to hash(x**2.0) but x**2 == x**2.0 is
True. Thus, have tried to look similar values in terms.keys() directly
and its work pretty good, except that test_minimal_polynomial() in
sympy.polys.tests fails at string 92:
assert minimal_polynomial(sqrt(a) + sqrt(sqrt(a)), x) == f
it goes into infinite recursion. Have no idea how to consider this
case.

I will be glad to work outside of gsoc programm.

-- 
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