Hi all, This weekend I had fun with creating views for catalog results. In order to do this, I had to: - Create an ICatalogResult interface and use a five:implements directive to configure ZCatalog.CatalogBrains.AbstractCatalogBrain as implementing it. - Set Five.traversable.Traversable as a mixin brain for catalog results (Catalog.useBrains). With these changes, I could a) define views for catalog result objects, and b) traverse over them in a template to obtain the view. I'd like to add this to Zope trunk, with Traversable a base class for AbstractCatalogBrain and a implements(ICatalogResult) on that same class. The obvious advantage of the latter is that one doesn't need to use Catalog.useBrains any more to add functionality to result objects (just use adapters). The first change is a necessity if you want to be able to look up views on results through traversal (in a template for example). I'd like to have some feedback on this idea before I do this however, just to make sure that using Traversable here is a good idea or not. For the curious, my usecase involved creating workflow menus (a lá Plone) for a page showing up to 100s of content objects, all AJAX driven. This enables you to quickly workflow several related objects without waking up all the content objects just to see their available workflow transitions. The workflow actions menu is a view defined for both content objects and catalog results, with the view also supporting invoking the transition. There are different view classes for content objects and catalog results, but the template for the view is the same for both cases. The catalog result view looks up transitions by hacking up action information objects, and the only limitation is that transition guards can not use permission or roles guards (no meaningful context to look up on) and guard expressions need to be written taking into account that the context is either a real content object or a catalog result. These limitations were not a problem for the workflows involved. -- Martijn Pieters