One obvious way of handling blogs with categories is to create appropriate directory hierarchies for each category, then hardlink a page's file into all of the appropriate 'category' directories. However, this raises a problem: DWiki's idea of a page's identity is its path.

This results in several visible problems:

Using redirects (symlinks, say) isn't a really great solution, because it's hard to have the code work out when a redirect should still be included in the result set of a view. Right now the answer is 'always', which is the right answer when redirects are considered not so important aliases instead of important structural pieces.

As a starting point, DWiki now has code that excludes subsequent copies of hardlinked files from atom and blogdir views. The 'first' copy is the copy with the path that sorts lexically first. Of course there are some problems with this:

The Atom feed issue is a general one that comes up any time a DWiki page is moved around, and can't be solved short of a fundamental change to DWiki's idea of a page's identity. (Which is not going to happen.) The comment issue is similar (and actually worse for renamed pages, since then the old comments just disappear).

I am willing to call the blogdir problem a 'don't do that' issue and brush it under the carpet.

It's possible that pages with more than one hardlinks to them should be marked in some visible way. It's also possible that they shouldn't be; it's hard to see what exactly users could do with the information. (To present something sensible, like 'where are other copies of the page', would really require keeping a database of DWiki pages. This is likely to be a 'no!' for the foreseeable future.)

Page tools: View Source, Add Comment.
Login: Password:
Atom Syndication: Recent Comments.

Last modified: Thu Jun 9 01:38:35 2005
This dinky wiki is brought to you by the Insane Hackers Guild, Python sub-branch.