I had a look at the git patch of Wim and reflecting on it, I had a similar approach, though a bit more hacky.
If I get it right, what is needed to finilize the solution of Wim is to decouple E4 logic from compat layer logic. 2017-01-09 7:55 GMT+01:00 Lars Vogel <[email protected]>: > Hi Erdal, > > Extending e3 elements via e4 elements would be a great feature. Can you > create a bug report and provide your model assembler as Gerrit review? > > Best regards, Lars > > Am 06.01.2017 12:08 nachm. schrieb "Erdal Karaca" < > [email protected]>: > >> Hi devs, >> The ModelAssembler only merges fragments into the app model if it can >> find the parent to extend. If we have an E4 app with compat layer AND the >> fragments are extending elements generated within E3 workbench >> intitialization phase, the fragments are not merged. >> >> I have experimented a bit with the internals of >> E4Application/ModelAssembler to load E4 fragments into the application >> model after the compat layer has loaded the E3 workbench. >> The result is that it seems to be possible to merge E4 fragments into E3 >> API driven products, i.e. E4 with compat layer. >> >> Now, I am not sure if I am missing some obvious points using this >> approach as seemed to work quite smoothly. >> As the various renderes listen to model changes, I expect them to handle >> dynamically added elements properly. >> Do you have or know of any concerns? >> >> _______________________________________________ >> e4-dev mailing list >> [email protected] >> To change your delivery options, retrieve your password, or unsubscribe >> from this list, visit >> https://dev.eclipse.org/mailman/listinfo/e4-dev >> > > _______________________________________________ > e4-dev mailing list > [email protected] > To change your delivery options, retrieve your password, or unsubscribe > from this list, visit > https://dev.eclipse.org/mailman/listinfo/e4-dev >
_______________________________________________ e4-dev mailing list [email protected] To change your delivery options, retrieve your password, or unsubscribe from this list, visit https://dev.eclipse.org/mailman/listinfo/e4-dev
