Hash: SHA1

Tres Seaver wrote:
> Hanno Schlichting wrote:
>> And then adjust exportComponentRegistry in a similar way. This is
>> untested code :)
> Why would a PAS plugin do such a thing?  PAS doesn't expect (or want)
> any local / persistent component registry.  This issue (steps showing up
> in inappropriate profiles) was the reason for the step registries being
> part of a profile, and why I argued (unsuccessfully) against their
> deprecation way back when:  I would have preferred fixing whatever
> issues existesd, over using a "global" stop registry.

That particular question, why the PAS import/export routine would try to
run those CMF steps, was exactly what I thought when I tried it all out.

I do remember that GS design discussion vaguely but never paid much
attention because I never used GS "in anger" and outside of the CMF.
It's not my goal to reopen that argument, but I'm left with the
impression that at least from the perspective of someone who uses the
setuptool ZMI to explore and use GS there are some expectation
mismatches and usability problems. Once you recognize the design
decision you can at least explain the behavior.

I'll continue down the path of trying to make the steps themselves more
resilient so I get somewhere.

At some later date it may be useful to review the current behavior. I
feel some decisions may have been taken with a too-narrow focus (CMF
only) because very few people use GS outside of the CMF or Plone.


Version: GnuPG v1.4.8 (Darwin)

Zope-CMF maillist  -  Zope-CMF@zope.org

See https://bugs.launchpad.net/zope-cmf/ for bug reports and feature requests

Reply via email to