Because for built-in iterator types the next() method is implemented in C and we didn't want to invalidate existing C code by giving it an optional argument.
Ah, makes sense.
But why do you care?
Curiosity more than anything.
It would be elegant if .send() had identical behavior to .next() without an argument, instead of throwing an exception, but it's no biggie.
Masklinn <email@example.com> wrote:
You'd have to check the ML discussions on PEP 342 (Coroutines via Enhanced Generators).
Sweet, thanks for the pointer. Sorry for not searching harder.
On Fri, Feb 19, 2010 at 11:32 AM, Masklinn <firstname.lastname@example.org>
On 19 Feb 2010, at 17:12 , Andrey Fedorov wrote:You'd have to check the ML discussions on PEP 342 (Coroutines via Enhanced Generators). The PEP explicitly states that send(None) is equivalent to next()
> Since .next() seems to be equivalent to .send(None), why wasn't .next() just
> given an optional parameter which defaults to None?
> - Andrey
> A new method for generator-iterators is proposed, called send(). It
takes exactly one argument, which is the value that should be "sent
in" to the generator. Calling send(None) is exactly equivalent to
calling a generator's next() method. Calling send() with any other
value is the same, except that the value produced by the generator's
current yield expression will be different.
A new method was probably created because it would be quite unclear that `next(value)` sends that value into the generator for consumption by a yield expression. Whereas the purpose of `send(value)` looks pretty obvious.