Community
Participate
Working Groups
Some "view shortcuts" are not working if an E4 view is used in an Eclipse 3.x based application. The following shortcuts are not working: Ctrl+M (maximize view) Ctrl+F10 (show "view menu") Alt+- (Show the system menu of the view) F12 (change from view to current editor) Ctrl+F7 (next view) Ctrl+Shift+F7 (previous view)
Created attachment 239221 [details] Fix for bug
Thanks David for the patch. Could you please sign your CLA? You can click on the CLA icon next to your name. PW
I signed the CLA.
We discovered that in Luna M5 the class "E4PartWrapper" is deleted. So I want to ask what's happening with our patch now? Will it be processed in some way and merged into the current implementation or is the fix done in another way? Thanks, Alex
(In reply to Alexander Fichtinger from comment #4) > We discovered that in Luna M5 the class "E4PartWrapper" is deleted. ? There's an inner class in WorkbenchPartReference, but that's still there. > So I want to ask what's happening with our patch now? > Will it be processed in some way and merged into the current implementation > or is the fix done in another way? We hope to review this patch in about 2 weeks. Then we'll know if we can support it. PW
(In reply to Paul Webster from comment #5) > We hope to review this patch in about 2 weeks. Then we'll know if we can > support it. > > PW Ok, thanks! =)
(In reply to Alexander Fichtinger from comment #6) > (In reply to Paul Webster from comment #5) > > We hope to review this patch in about 2 weeks. Then we'll know if we can > > support it. > > > > PW > > Ok, thanks! =) Alex, please refactor your work so that it works with M5. In addition, put it in Gerrit [3]. If you do this, your patch has a much greater chance of getting in before Luna ships. We also have some discussion in e4-dev [4]. I have been working on [1] (see [2] for the Gerrit patch) to load the IDE with E4 fragments and processors. What is does NOT do is make the E4 parts First Class Citizens. I believe your work starts this. Which is excellent timing. [1] https://bugs.eclipse.org/bugs/show_bug.cgi?id=376486 [2] https://git.eclipse.org/r/#/c/21111/ [3] http://wiki.eclipse.org/Platform_UI/How_to_Contribute#Creating_a_Gerrit_review_or_a_patch [4] https://dev.eclipse.org/mailman/listinfo/e4-dev
(In reply to Wim Jongman from comment #7) > (In reply to Alexander Fichtinger from comment #6) > > (In reply to Paul Webster from comment #5) > > Alex, please refactor your work so that it works with M5. In addition, put > it in Gerrit [3]. If you do this, your patch has a much greater chance of > getting in before Luna ships. We also have some discussion in e4-dev [4]. > We will refactor our patch today =). And thank you for the hint with Gerrit and for the links. > I have been working on [1] (see [2] for the Gerrit patch) to load the IDE > with E4 fragments and processors. What is does NOT do is make the E4 parts > First Class Citizens. I believe your work starts this. Which is excellent > timing. This sounds good to me! Because we are working in full mixed mode =). (3.x Workbench with E4 parts) > [1] https://bugs.eclipse.org/bugs/show_bug.cgi?id=376486 > [2] https://git.eclipse.org/r/#/c/21111/ > [3] > http://wiki.eclipse.org/Platform_UI/ > How_to_Contribute#Creating_a_Gerrit_review_or_a_patch > [4] https://dev.eclipse.org/mailman/listinfo/e4-dev
We created a gerrit commit which works with 4.4 M5 https://git.eclipse.org/r/#/c/21817/
Committed: http://git.eclipse.org/c/platform/eclipse.platform.ui.git/commit/?id=b07e1de7fb6b4522463e119c303dac877dfeea20 You guys rock ! This is a great patch, thanks. Let me know if you come up with other snags...
(In reply to Eric Moffatt from comment #10) > Committed: See? ;)
(In reply to Wim Jongman from comment #11) If I read your patch correctly then the following is happening: If running with the compatibility layer you will create an E4PartWrapper on the fly if there is no CompatibilityPart associated with the part. So this will now work for every pure e4 part that is put into the model? Why do you put the E4Wrapper in the parts' transient data and not in it's context to make this possible: @Inject ViewPart legacyViewPart What do you guys currently do to bridge the selection services (bug 403930)? I see that there is some work in the SelectionService. How have you solved bridging the selection service?
(In reply to Wim Jongman from comment #12) > (In reply to Wim Jongman from comment #11) > > If I read your patch correctly then the following is happening: > > If running with the compatibility layer you will create an E4PartWrapper on > the fly if there is no CompatibilityPart associated with the part. > > So this will now work for every pure e4 part that is put into the model? > Apparently it does =). Now we don't have any problems with our E4 parts concerning view shortcuts (like Ctrl+M). > Why do you put the E4Wrapper in the parts' transient data and not in it's > context to make this possible: > > @Inject > ViewPart legacyViewPart > Maybe this would be a good idea, but we were glad, that we could make it work as it is now. > What do you guys currently do to bridge the selection services (bug 403930)? > I see that there is some work in the SelectionService. How have you solved > bridging the selection service? The SelectionService just got called when we double clicked something in our E4 part. Then an exception occured. So we also added the "else if", which we already added to other classes. Then it just worked =).
Verified in 4.4.0.I20140303-2000. I've opened bug 429590 because of an NPE caused because my e4 view didn't have a menu defined but that's a different issue (it did try to run...;-).
