Hi again, after receiving a lot of help from Holger on how to set up the IntelliJ environment, I am still working on this and hope to push my results soon. Could you please grant me push permissions for Gerrit? While implementing this feature, I have encountered two small unrelated issues that I would like to push separately.
Regarding the feature, I can successfully show the remote caret's position in IntelliJ using text marks for now - I will have a look on the "real" carets later, when this works correctly. However, there is an issue I would appreciate some advice on: When the caret is moved around, everything is fine. When text is entered however, both activities are triggered - first the one for text editing, then the one for text selection, as the caret has also been moved by the entered character. This means that when displaying the text marks in the editor, it is unknown whether the caret had been moved with or without entering text. Thus, existing text marks can't be cleared or left in place correctly. I am wondering where the behavior is faulty in this case. Should I suppress sending an selection activity when entering text? Best regards, Jacob -----Original Message----- From: Holger Schmeisky [mailto:holge...@fu-berlin.de] Sent: Sonntag, 8. November 2015 20:57 To: dpp-devel@lists.sourceforge.net Subject: Re: [DPP-Devel] #116 Multiple Carets in IntelliJ Hi Jacob, great to hear you're fixing this! > 2. Display the caret's position upon moving by the user through > the TextSelectionActivity. This would be the preferred implementation, because it provides more awareness to the user. > > We can read the carets position with the CaretListener, but there > seems to be no specific activity to transmit the updated caret. Have you tested the listener yet? I have not used it so I am not sure what kind of input you get. We could use > the TextSelectionActivity with a length parameter of 0 to signal a > moved caret That's how Eclipse does it I think. and handle it in LocalEditorManipulator.selectText? > sounds like a good approach. Have you seen [1]? I am not sure how difficult it is to implement that, so if it turns out to be too difficult with a real second caret, you can use just a simple text highlight instead. (but a real second caret is preferred of course) > > There will probably be even other approaches that we didn't consider > yet, and we would value your input on this. > Sounds good! > > > In another matter, we couldn't find any test cases related to the > editor classes. Could you maybe kindly point us in the right > direction? Thank you in advance. > I'm sorry to say there are none for the high-level Saros/I classes. You will have to do manual testing [2] [1] http://www.jetbrains.org/intellij/sdk/docs/reference_guide/multiple_carets.h tml [2] http://www.saros-project.org/testing-sarosi-locally > > > Best regards, > > Jacob Schramm > > > > > ---------------------------------------------------------------------- > -------- > > > > _______________________________________________ > DPP-Devel mailing list > DPP-Devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/dpp-devel > ---------------------------------------------------------------------------- -- _______________________________________________ DPP-Devel mailing list DPP-Devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dpp-devel ------------------------------------------------------------------------------ _______________________________________________ DPP-Devel mailing list DPP-Devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dpp-devel
