[stdlib-sig] proposed removal: the types module

[Barry Warsaw]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Apr 4, 2008, at 5:23 AM, Christian Heimes wrote:
> Barry Warsaw schrieb:
>> On Apr 3, 2008, at 11:09 AM, M.-A. Lemburg wrote:
>>
>>> The idea to spread type objects across several different modules
>>> is just not practical.
>>
>>> If you want to test for built-in types that are used by the  
>>> interpreter,
>>> then you want to look in *one central place* for the type object -  
>>> not
>>> in a gazillion different modules.
>>
>> Exactly right.
>
> Which is exactly what my patch http://bugs.python.org/issue1605
> implements, isn't it? And what was rejected by Guido.

Oh well.

BTW, I happen to agree in part with Guido's decision in the bug  
commentary, specifically that abc's not be all grouped together.  But  
I don't agree that the same logic implies that types should be strewn  
throughout the stdlib.  Types and abcs will  be used in completely  
different ways, and it's all about discoverability.

Guido says he's "100% against" this, so I guess another solution has  
to be found.

- -Barry

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.8 (Darwin)

iQCVAwUBR/YWu3EjvBPtnXfVAQKrqgP/ex8IvWscF8pgYV/nqeaCWFGuhsfZFpO7
swivsR6rGIQg6n0tZDyvxtt61YPbURSVL1lbO/rehmeKYPLxNZ3ubEU0ZMpv7pJK
blYBmFgJ/MpPs1aQwZ4iGpydWTG0uGn+7N84if4d6UhspZ9FAfhv/O+oO5s8LWLw
SZv5ZIWqhxw=
=O+hj
-----END PGP SIGNATURE-----