Hi Eliot, Thank you for your feedback. See more below:
1. Ability to control collapsing of entire tree, e.g. "collapse children/expand children). When you're dealing with a large schema with lots of top-level elements and types and you want to focus on just one part it's tedious to collapse everything. It would probably help to also have the default be to have all the children of the root initially collapsed. Epic Editor's XML tree view has a neat feature that I've never seen anywhere else: if you triple click on a tree item it will collapse the children of that item--this makes it really quick to collapse a long list of children.
Oxygen 5.1 has the same collapse children option for the folds (see the collapse child folds action from the editor or folds contextual menu). In <oXygen/> 6 there is even a better action added for managing the folds tree that is named "Collapse other folds" and has as effect collapsing all the children of the nodes on the path to the root, that is you can get into focus only the edited part. We will extend these actions to the schema diagrams, I just added a bugzilla entry with this (this will go in 6.1).
2. Option to view just the element/type hierarchy without all the details of reference, refinement, extension, etc. As it is, it can be a little difficult to navigate down through a hierarchy of element types because the tree can get pretty wide, so there's a lot of scrolling, especially if you've used long names for types (as I tend to do).
One of the nice things about Turbo's view is that it is pretty quick to navigate down the tree.
You have this in the Logical Model view. There it is presented the fully parsed schema and the references, extensions, etc. are resolved. This was exactly the purpose for the logical model, to show the parsed model.
3. Provide a "where used" view, that is, when you click on an element type, show what other element or type contexts it's allowed in, from which you can navigate to those types.
Yes, this sounds useful, we should have a Search references action that should present all the places a given element/type/attribute is referred.
4. In the logcal model view, also synchronize with the source, including providing the option to open the containing XSD file if it's not already open.
The logical model view renders the fully parsed schema and currently we do not have location information for that. For instance in the Relax NG case the parsed model is obtained after applying a number of simplification rules thus most of the content in the logical view is obtained in this process and does not have a corresponding schema location. In the XML Schema case we may be able to solve this with a Find declaration action that should select in the source the declaration of that schema component.
Things I really like:
- The list of types, elements, and groups. Very handy for getting a view over how the schema is organized from a coding standpoint.
- The synchronization of the tree view with the markup view. Really handy. Will save much time now spent doing searches for specific entries in the schema.
- The showing of enumerations for element content.
- For the logical model view, the grouping of types by their namespace.
Thanks! George --------------------------------------------------------------------- George Cristian Bina <oXygen/> XML Editor, Schema Editor and XSLT Editor/Debugger http://www.oxygenxml.com