On Tue, Feb 1, 2011 at 8:28 AM, William Ross <w...@spanner.org> wrote:

If I was trying to give it a sane IA I wouldn't publish the calendar using a
> blog-history extension in the first place :)

Yeah, well we do what we can with what we've been handed. :-\

I do think there's an architectural flaw in the Archive extension: I would
> be inclined to interpolate the date string in the slug method (rather than
> in child_path) so that at validation we see the path that is actually used.
> This is very old code, though, carried forward mostly for compatibility
> reasons. I doubt that people will be keen to make fundamental changes, but I
> could well be wrong.
> The slug is a more monkeypatchable change so you might be able to do it
> from the outside just by overriding child_path and slug.  It probably
> depends how much use you are making of ArchiveMonthPage and the like.

I'm not sure how that would work...wouldn't the database uniqueness
constraint on 'slug' imposed by Page supercede such a patch?

The other obvious change is just to make your slugs different. beijing2 or
> beijing_june, say. You can still have 'Beijing' in page title and
> breadcrumb.

Yeah, I was hoping to avoid that...putting date info "_month" is redundant
with the path, and affixing some other disambiguation (like a number) just
looks like a hack. Just because we're holding things together with duct tape
and baling wire doesn't mean the public needs to know that. ;-)

Thanks for the ideas. We've decided to go for the slap-a-number-on-the-end
approach and then revisit the issue when and if I ever get this upgrade


Reply via email to