Aaron Lehmann schrieb:

On Feb 11, 2008, at 10:06 AM, Christian Theune wrote:


Aaron Lehmann schrieb:
What if people are relying on those error messages to do different things? Possibly they already have normalization code they expect to come into play in this situation, which is different than yours

I see that people might be doing that, however, it seems like a bad idea to code that way. ;)

Maybe. But enforcing this on others seems to cut against "consenting adults." After all, how do you know that your desired resolution is the resolution for everyone?

Sure, that's why I'm discussing it here. ;)

On the other hand, having to do this seems annoying, so as someone writing new code I'd consider your proposed patch a feature, rather than a bugfix.

I'd be happy to declare it a feature, but I consider the existing state to not be ideal and thus it should be changed.

Possibly instead of returning an error, it might resolve the name, and take an optional callback that performs said resolution? That way, people who have their own resolution code can easily factor it in.

Hmm. Sounds like a little too much engineering and not very componentish to me.

Up to this point my thinking went this way: We have an interface that allows us to a) check whether a name is ok b) create a name that is ok for us, probably considering a name that we give him.

The second method feels like it should never raise an error - that's what the `check` method is for.

If you want to customize the behaviour, you can always create a different name chooser implementation. It doesn't have to be 100% correct for everyone, but it should avoid the 80% pitfalls and it certainly should implement the interface the way it is intended.

Unfortunately the interface is a bit unspecific whether chooseName may raise an error if the given name does not conform to checkName or whether it is obligued to always pick a good one.


gocept gmbh & co. kg - forsterstrasse 29 - 06112 halle (saale) - germany
www.gocept.com - [EMAIL PROTECTED] - phone +49 345 122 9889 7 -
fax +49 345 122 9889 1 - zope and plone consulting and development
Zope-Dev maillist  -  Zope-Dev@zope.org
**  No cross posts or HTML encoding!  **
(Related lists - http://mail.zope.org/mailman/listinfo/zope-announce
http://mail.zope.org/mailman/listinfo/zope )

Reply via email to