|[QVTo] How to load separated (externalized) profile applications in QVTo transformations ? [message #1722870]
||Tue, 09 February 2016 20:23
| Toni Siljamaki
Registered: November 2012
Thanks to the excellent work by Cristian Damus, and since January 2015, Papyrus now supports separating profile applications into separate files.
Thank you, Christian.
This is referred to as "externalized profile applications" in Papyrus.
A Papyrus UML model may have multiple candidate profile applications, which can be mutually exclusive, that can be loaded and unloaded by the user.
This new capability in Papyrus provides important possibilities for larger-scale SW product development, such as product variant handling via MDA Marking Models = deploying the same CM'ed version of a source UML model onto multiple SW/HW platforms, or versions thereof, such as onto different versions of a multicore ASIC, or for different market adaptations etc. (just some obvious examples)
1) How can one or several selected "externalized profile applications" (separate .uml files) be loaded onto a source UML model by a QVTo transformation ?
2) What should the .qvto transformation declaration itself and its "header" look like?
3) What should the transformation Configuration look like, such as when running and testing the transformation in the debugger ?
...that is: How to get the externalized profile application(s) loaded onto the source UML model before the actual model transformation starts ?
A working demo example + tutorial is appreciated. (+ updated QVTo Help)
Embedded/headless transformation is the next step. (= how to do it programmatically)
[Updated on: Thu, 11 February 2016 17:34]
Report message to a moderator
Powered by FUDForum
. Page generated in 0.01619 seconds