-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Wichert Akkerman wrote:

Previously Nikolay Kim wrote:

...

...

...

Probably this is not much important enough for that *import* usecase, but if we start with that, can you tell me wheres the border with mixin code into Zope3 packages and what is Ok and where do we stop? I agree it should rarely be used, but it was just one import so then I thought it was ok. Conditional import are ugly and many of us doesn't use zope2 at all.

And many of us do use Zope2 and want to be able to use all the nice Zope3 technology. A single conditional import is an extremely small price to pay to give a large group of people access to new technology.

I thought the componentized Zope 3 was all about making it easier to reuse modules. Did that change?

I don't know. I think the objection to the change was that it introduced a dependency inversion, as well as making it impossible to test the code properly (you would need to run the tests in both an environment with and without Five present). In general, bridging gaps between the "depended on" Zope 3 stuff and the "dependent" Zope2 stuff is not supposed to involve introducing ugliness into the Zope 3 stuff: the Five package has been the "ugly magnet" since it was first created, for exactly this reason. (Followups reset to zope-dev: the checkins list isn't supposed to be fore discussion). Tres. - -- =================================================================== Tres Seaver +1 540-429-0999 tseaver@palladion.com Palladion Software "Excellence by Design" http://palladion.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFH7Wyk+gerLs4ltQ4RAugcAJ49/v6V+4aIZvVJBHeHfON4JbVLnQCg3Vfq 6EKI3wvuVOzYTPwjQV6df+Y= =P4UJ -----END PGP SIGNATURE-----