| Summary: | [Preferences] Regeneration of all diagrams | ||
|---|---|---|---|
| Product: | [Modeling] Papyrus | Reporter: | Patrick Tessier <Patrick.Tessier> |
| Component: | Core | Assignee: | Patrick Tessier <Patrick.Tessier> |
| Status: | CLOSED FIXED | QA Contact: | |
| Severity: | blocker | ||
| Priority: | P1 | CC: | thibault.landre, vincent.hemery |
| Version: | 0.7.0 | ||
| Target Milestone: | --- | ||
| Hardware: | PC | ||
| OS: | Windows XP | ||
| Whiteboard: | |||
| Bug Depends on: | 316688 | ||
| Bug Blocks: | |||
|
Description
Patrick Tessier
It seems that the new preference system modifies the way the default figure size is managed. Can you explain how it works because figure are now created without the correct size ? Have you regenerate the diagrams. The default size is now store in preferences. And each element can store its dimension. Yes, and I have tried again. Example : StateInvariant and DestructionEvent in the sequence diagram A specific size is specified through the GMFGen, but the preferenceStore doesn't return those size. The default value seems to be not adapted for all diagrams. I will research a solution in order to fixe it. I am reporting here problems discussed in private for traceability : A. Default size should be -1, -1. This size is set in method org.eclipse.papyrus.preferences.pages.PapyrusAllDiagramsPreferencePage.initDefaults(IPreferenceStore), with keys PreferenceConstantHelper.WIDTH and PreferenceConstantHelper.HEIGHT. The problem can be easily tested with TimeObservation in Sequence diagram (which are way too big with 100*100). B. In sequence diagram, the initialization method org.eclipse.papyrus.diagram.sequence.preferences.DiagramPreferenceInitializer.initializeDefaultPreferences() seems to never be called. The problem can be seen on DestructionEvent, which size is much too big (or invisible if default size is -1), despite it is correctly defined to 40*40 in method org.eclipse.papyrus.diagram.sequence.preferences.DestructionEventPreferencePage.initDefaults(IPreferenceStore) point B. solved : a block which had the generated NOT tag has been updated in the sequence diagram. all diagram has been regenerated 2 years ago |