Malthe Borch-2 wrote > > On 31 August 2012 12:44, Rok Garbas <rok.garbas@> wrote: >> we've been using to on pretty big vocabularies. but those vocabularies we >> managed outside of plone (eg: italian provinces, regions, ...). i was >> thinking >> that it would be nice to persist this in plone which would also made it >> possible to change it (you can then have any kind of permission over it) >> but >> this never happened since there was no usecase for it. > > I think based on this conversation and further consideration, I'd like > to make some changes in the design: > > 1. Only the "root nodes" (or vocabulary) will be a content object. > 2. We'll still attempt to use Plone's user interface to add, edit and > manage the contained (hierarchical) terms, but these objects won't be > content types (and won't be indexed). > 3. There will be no security on the terms. > 4. It might make sense to model the contained terms as "virtual > objects". This allows us to support both the requirement of "many > terms" and "manageable terms". >
This sounds like a very sensible modification to me. I'd consider having them live in the registry rather than the content space to benefit from easy import/export (although this might not be appropriate for very large taxonomies.) I suspect you will almost always want to edit in a batch orientated manner rather than as single items, so reusing Plone's editing interface is likely to be more trouble than its worth (you can create simple z3c.forms to edit stuff but folder_contents does not work without the catalogue.) The easiest UI approach might be to build the whole structure client side and simply post it back as json when editing (though andling conflicts with multiple editors might become a problem then.) > Extending on the original proposal, we'd also like to dynamically > associate a behavior to each vocabulary. This behavior simply adds a > choice or list field (configurable) which pulls values from the > vocabulary. I haven't looked into an implementation strategy for this > yet. > Could you perhaps give an example? Isn't this just an alternative way of presenting the leaf nodes of a taxonomy when tagging a content item? Laurence -- View this message in context: http://plone.293351.n2.nabble.com/Taxonomy-support-in-a-modern-Plone-tp7558448p7558635.html Sent from the Product Developers mailing list archive at Nabble.com. _______________________________________________ Product-Developers mailing list [email protected] https://lists.plone.org/mailman/listinfo/plone-product-developers
