| Summary: | EObjectObservableMap should allow to observe none existant attributes | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Product: | [Modeling] EMF | Reporter: | Thomas Schindl <tom.schindl> | ||||||||
| Component: | Core | Assignee: | Ed Merks <Ed.Merks> | ||||||||
| Status: | VERIFIED FIXED | QA Contact: | |||||||||
| Severity: | enhancement | ||||||||||
| Priority: | P3 | CC: | Alfredo.Bencomo, arash.aghevli, eclipse-bugs | ||||||||
| Version: | unspecified | ||||||||||
| Target Milestone: | --- | ||||||||||
| Hardware: | PC | ||||||||||
| OS: | Mac OS X - Carbon (unsup.) | ||||||||||
| URL: | http://www.eclipsezone.com/eclipse/forums/t112822.html | ||||||||||
| Whiteboard: | |||||||||||
| Attachments: |
|
||||||||||
|
Description
Thomas Schindl
Created attachment 113086 [details]
Example project
Tom, I'll try to catch up with my backlog of committable things next week when I'm back from Germany... Created attachment 115671 [details]
Improved fix.
Tom, this is a more general approach that takes some of the gross XML Schema-isms into account. Note that I've guarded only to the doGet. I suspect that the doPut should fail since it doesn't make sense to try to modify something that doesn't exist and won't persist the state you are trying to set. Maybe that's a bad assumption and I need the same guard for the doPut case? Please advise.
You don't need to guard doPut, so your patch is perfect. The changes have been committed to CVS for 2.5. Fix available in HEAD: 2.5.0.I200810211800. Fix available in HEAD: 2.5.0 (R200906151043). |