[gwt-contrib] Re: Changes in com.google.gwt.app land to catch up to the RequestFactory (issue946802)
Submitting. http://gwt-code-reviews.appspot.com/946802/diff/1/3 File user/src/com/google/gwt/app/place/AbstractProxyEditActivity.java (right): http://gwt-code-reviews.appspot.com/946802/diff/1/3#newcode83 user/src/com/google/gwt/app/place/AbstractProxyEditActivity.java:83: setWaiting(true); There was such an assert, but it's never fired. And it doesn't seem like it would be any more informative than the NPE that would happen on the next line. On 2010/10/03 23:31:12, cromwellian wrote: Is it possible for saveClicked() to be invoked when editorDriver is null? In theory it shouldn't be possible of onStop() is called after the widget that generates saveClicked() is disabled, but I wondering if it is possible to 'queue up' an event by double clicking the save button rapidly. I guess don't worry about it for now (could throw in an assert editorDriver != null), just looking to catch eisenbugsā„ http://gwt-code-reviews.appspot.com/946802/diff/1/12 File user/src/com/google/gwt/app/place/ProxyPlace.java (right): http://gwt-code-reviews.appspot.com/946802/diff/1/12#newcode46 user/src/com/google/gwt/app/place/ProxyPlace.java:46: String bits[] = token.split(!); On 2010/10/03 23:31:12, cromwellian wrote: Maybe make ! a constant? Done. http://gwt-code-reviews.appspot.com/946802/diff/1/20 File user/test/com/google/gwt/requestfactory/client/RequestFactoryTest.java (right): http://gwt-code-reviews.appspot.com/946802/diff/1/20#newcode250 user/test/com/google/gwt/requestfactory/client/RequestFactoryTest.java:250: public void disabled_testChangedEdit() { Removing this, since it will be added by http://gwt-code-reviews.appspot.com/931803 http://gwt-code-reviews.appspot.com/946802/show -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
[gwt-contrib] Re: Changes in com.google.gwt.app land to catch up to the RequestFactory (issue946802)
http://gwt-code-reviews.appspot.com/946802/show -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
[gwt-contrib] Re: Changes in com.google.gwt.app land to catch up to the RequestFactory (issue946802)
Thanks, will fix. On Mon, Oct 4, 2010 at 3:22 AM, t.bro...@gmail.com wrote: http://gwt-code-reviews.appspot.com/946802/diff/8001/9006 File user/src/com/google/gwt/app/place/FindAndEditProxy.java (right): http://gwt-code-reviews.appspot.com/946802/diff/8001/9006#newcode50 user/src/com/google/gwt/app/place/FindAndEditProxy.java:50: FindAndEditProxy.super.start(display, eventBus); This activity doesn't support startup cancellation: super.start() will view.setDelegate() (among other things) whether or not the activity was cancelled. It will be a real problem in the case you switch from a FindAndEditProxy (FAEP) to a CreateAndEditProxy (CAEP), if they share the same singleton view: 1. FAEP starts (async) 2. user changes his mind: FAEP is cancelled (a no-op actually) and CAEP is started (synchronous: sets itself as the view delegate) 3. FAEP async start response arrives: calls super.start which overwrites the view's delegate. From the ActivityManager POV, FAEP is dead and CAEP is active, but actually the view delegates its events to FAEP, and CAEP is blind (kind of a zombie activity). http://gwt-code-reviews.appspot.com/946802/show -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
[gwt-contrib] Re: Changes in com.google.gwt.app land to catch up to the RequestFactory (issue946802)
Caught that case, fix is in http://gwt-code-reviews.appspot.com/950801 On Mon, Oct 4, 2010 at 1:22 AM, t.bro...@gmail.com wrote: http://gwt-code-reviews.appspot.com/946802/diff/8001/9005 File user/src/com/google/gwt/app/place/CreateAndEditProxy.java (right): http://gwt-code-reviews.appspot.com/946802/diff/8001/9005#newcode54 user/src/com/google/gwt/app/place/CreateAndEditProxy.java:54: display.setWidget(null); Er, really fall through to super.exit when cancelling a creation? (it will goTo the details of the cancelled proxy!?) http://gwt-code-reviews.appspot.com/946802/show -- http://groups.google.com/group/Google-Web-Toolkit-Contributors
[gwt-contrib] Re: Changes in com.google.gwt.app land to catch up to the RequestFactory (issue946802)
http://gwt-code-reviews.appspot.com/946802/diff/8001/9005 File user/src/com/google/gwt/app/place/CreateAndEditProxy.java (right): http://gwt-code-reviews.appspot.com/946802/diff/8001/9005#newcode54 user/src/com/google/gwt/app/place/CreateAndEditProxy.java:54: display.setWidget(null); Er, really fall through to super.exit when cancelling a creation? (it will goTo the details of the cancelled proxy!?) http://gwt-code-reviews.appspot.com/946802/show -- http://groups.google.com/group/Google-Web-Toolkit-Contributors