What about letting x.( ... ? ... ) be equivalent to ( ... x ... )

[al]

Fredrik Lundh a écrit :
> if you have a fear of introducing new local variables, you have problems
> that cannot be solved by syntax.

Dear Fredrik,

I have read the original messages on fr.comp.lang.python, and I don't 
understand your answer.

It is not about a fear of introducing new local variables, but for me it 
is an elegant solution to a common problem, to avoid creation of useless 
variables (what in french we call "variables muettes", like indexes in 
loops who are just there because some langages level is too low).

It also avoid the increase of parenthesis depth, and so the readability 
is enhanced.

And it solve a problem that in all object oriented langages, a method 
that process 2 or more different classes of objets belongs just to one 
of those classes.

All this kind of problems appears often to me (and in different 
langages), and contrarily to you, I'm very impressed by the compacity 
and elegance of the solution. I think it would be nice if implemented in 
different langages (because it breaks nothing), and firstly Python.

Best regards,
Al

PS : sorry for my approximative english, but my natural langage is french.