- From: Alexandre Rafalovitch <alex@access.com.au>
- Date: Fri, 14 Jun 1996 13:10:13 +1000
- To: www-jigsaw@w3.org
Anselm Baird-Smith wrote: >Alexandre Rafalovitch writes: <snip myselft> > > Ps. Or am I barking on the wrong tree and there is never a reason to > > reindex non file based resource and this command should just do nothing for > > them? > >Well that's the difficult part of it. The only solution I can see is >to have another atribute that says wether the resource has been >manually edited or not. If manually edited, the reindex command would >skip them, otherwise, it would reindex them. (The point is to *never* >have exceptional treatment of resource based on their class). > >Anselm. I agree about never bit. There just no way knowing which hand written classes can be reindexed safely. I also agree, that there might have to be an attribute influencing reindexing strategy. Otherwise the next day somebody WILL come with recursive reindex solution and would reindex his root...... Here goes all management functions and no way to get back. Sort of a hard way to learn your mistakes. So, I think this problem has to be solved before or at the same time with multiple reindexing( recursive or set based does not matter). Alex. alex@access.com.au
Received on Thursday, 13 June 1996 23:12:10 UTC