Bug 506810

Summary: Transition kind should be to valid kind by default
Product: [Modeling] Papyrus-rt Reporter: Young-Soo Roh <ysroh>
Component: toolAssignee: Young-Soo Roh <ysroh>
Status: CLOSED FIXED QA Contact: Peter Cigehn <peter.cigehn>
Severity: normal    
Priority: P3 CC: charles, peter.cigehn
Version: 0.7.2Keywords: test
Target Milestone: 0.9.0   
Hardware: PC   
OS: Windows NT   
See Also: https://git.eclipse.org/r/84996
https://git.eclipse.org/c/papyrus-rt/org.eclipse.papyrus-rt.git/commit/?id=73fdfda5d4e129d877772e8b3bb622b8169282d8
https://git.eclipse.org/r/85519
https://git.eclipse.org/c/papyrus-rt/org.eclipse.papyrus-rt.git/commit/?id=5ea2f34498756e99b5b12ec948270aca6bce2621
https://git.eclipse.org/r/85707
https://git.eclipse.org/c/papyrus-rt/org.eclipse.papyrus-rt.git/commit/?id=3435de0374e51dfe7789dd01b1f2310459e5dde7
Whiteboard:
Bug Depends on:    
Bug Blocks: 494284    
Attachments:
Description Flags
snapshot none

Description Young-Soo Roh CLA 2016-10-31 15:37:10 EDT
Created attachment 265125 [details]
snapshot

Transition kind is set to external by default but in some situation external does not make sense. 
For example, transition from entry point of composite state to a sub-state should be set to local by default. See snapshot.
Comment 1 Peter Cigehn CLA 2016-11-07 07:16:14 EST
Since the work with the transition tool have not started yet, I suggest that the work of ensuring the correct kind of transition to be part of Bug 494284 which is used for tracking the work needed for the transition tool. Either close this one and add a comment about it on that Bugzilla. Or make this Bugzilla block Bug 494284.

Please see the enumerated list of valid kinds of transitions in Bug 494284 Comment 0 and what kind they should be assigned, i.e external or local.

Keep in mind that the creation of internal transitions are handled completely different and are tracked by Bug 494287.
Comment 2 Eclipse Genie CLA 2016-11-14 14:18:24 EST
New Gerrit change created: https://git.eclipse.org/r/84996
Comment 5 Peter Cigehn CLA 2016-11-22 04:48:10 EST
I've tested the latest Papyrus-RT build and the transition kind seem to be set correctly now, at least for the cases of external vs. local (since internal shall be handled rather differently) according to what is enumerated in Bug 494284 Comment 0.

There are some inconsistent cases though with respect to the how the radio buttons of the kind property in the properties view is enabled/disabled, but I guess that should be handled within the scope of Bug 494291. I will provide more detailed comments there instead.
Comment 6 Eclipse Genie CLA 2016-11-22 12:43:14 EST
New Gerrit change created: https://git.eclipse.org/r/85519
Comment 8 Eclipse Genie CLA 2016-11-24 11:50:33 EST
New Gerrit change created: https://git.eclipse.org/r/85707
Comment 10 Peter Cigehn CLA 2016-11-28 08:37:01 EST
I tested this in the latest Papyrus-RT build. It is a bit hard to test since we don't have Bug 494284 in place. From what I can see the case with a self-transition on the inside of a composite state, which should be a local self-transition, becomes external. Only self-transitions drawn on the outside of a composite state should be external.

The rendering of the transition also i very confusing, since the transition gets drawn outside of the frame of the composite state. It is more natural that it is drawn inside the frame to indicate that it is local. This also how the legacy tooling behaves. So I am not sure what to do with this bug, and what the scope of it actually was (if it was limited to single case of a local transition to a sub-state of a composite state).
Comment 11 Peter Cigehn CLA 2017-03-10 05:10:54 EST
Verified to be fixed in the latest Papyrus-RT build. The transition kind are now set correctly in all cases as identified in Bug 494284.

Putting into verified fixed, and Young-Soo (or Charles) can close, or reopen if needed.
Comment 12 Young-Soo Roh CLA 2017-03-10 09:18:26 EST
Closing the bug