On Wednesday, November 13, 2013 10:39:48 PM UTC+1, Ondřej Čertík wrote:
>
>
> I wonder how difficult would be to write just a very simple core in 
> Julia, just so that we can experiment with this. 
> See how extensible it is, and so on. 
>

The problem is how to call it from Python. For what I was able to 
understand, Python can currently call Julia from IPython or through Julia's 
C API. It's not that easy.

Another option would be to create a code generator to translate Julia back 
to Python, and gradually switch the development to Julia. This isn't that 
complicated, as Julia is homoiconic, i.e. it can load and represent its own 
code in its data structures, and also provides macros to alter its own code.

By the way, in Julia there are two possible approaches:

   - Types and multiple dispatch.
   - Assumptions and pattern matching (minimal use of types).

The former is natively supported, the latter requires a pattern matching 
library (there are 3 under development in Julia).

I think that using types properly would allow to avoid pattern matching.

One important thing concerning multiple dispatch: remove all multiple 
inheritances! With multiple inheritance there can be matching ambiguities, 
such as


class A
  pass

class B
  pass

class C(A, B)
  pass

@dispatch(A)
def func(x):
  pass

@dispatch(B)
def func(x)
  pass

c = C()
func(c)  # which one of the two dispatched methods should it match???

-- 
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.
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to