[issue4104] Namespace inconsistency

Terry J. Reedy <[EMAIL PROTECTED]> added the comment: I consider the request invalid. Continue the example with B=A.B B The repr of an object cannot now and should not depend on the access path. =or= class C: pass class D: pass D C.D = D C.D Same comment. -- nosy: +tjreedy

[issue4104] Namespace inconsistency

Changes by Martin v. Löwis <[EMAIL PROTECTED]>: -- status: open -> closed ___ Python tracker <[EMAIL PROTECTED]> ___ ___ Python-bugs-lis

[issue4104] Namespace inconsistency

Georg Brandl <[EMAIL PROTECTED]> added the comment: Me too. There are too many cases where class statements can be executed to get this right. -- nosy: +georg.brandl ___ Python tracker <[EMAIL PROTECTED]> ___

[issue4104] Namespace inconsistency

Martin v. Löwis <[EMAIL PROTECTED]> added the comment: I'm tempted to close this as "won't fix". -- nosy: +loewis resolution: -> wont fix ___ Python tracker <[EMAIL PROTECTED]> __

[issue4104] Namespace inconsistency

Benjamin Peterson <[EMAIL PROTECTED]> added the comment: This is because type_repr looks at the __module__ attribute to determine the location. Fixing this would probably require adding a __nestedclass__ attribute. However, nested classes aren't used very much, so it's not urgent by any means. -

[issue4104] Namespace inconsistency

New submission from Antoine d'Otreppe <[EMAIL PROTECTED]>: Hello See the following code and comments for explanation ;) (Try it with interactive mode) --- >>> class A: ... class B: ... pass >>> A >>> A.B >>> B NameError: B is not defined --- This seems to be inconsistent, as Pyt