Bug 458452 - Incorrect package names in sysml.ecore
Summary: Incorrect package names in sysml.ecore
Status: NEW
Alias: None
Product: Papyrus
Classification: Modeling
Component: SysML (show other bugs)
Version: 1.0.0   Edit
Hardware: All All
: P3 normal (vote)
Target Milestone: ---   Edit
Assignee: Project Inbox CLA
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on: 478893
Blocks: 440082
  Show dependency tree
 
Reported: 2015-01-26 12:44 EST by Ed Willink CLA
Modified: 2016-10-04 05:09 EDT (History)
3 users (show)

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Ed Willink CLA 2015-01-26 12:44:33 EST
Package names in sysml.ecore are all lowercase while those in sysml.profile.uml follow the UML capitalized convention.

sysml.ecore should have 'originalName' EAnnotations to indicate that it has corrupted names otherwise OCL tooling that works on UML names gets confused.
Comment 1 Ed Willink CLA 2015-01-26 15:23:25 EST
(In reply to Ed Willink from comment #0)
> sysml.ecore should have 'originalName' EAnnotations to indicate that it has
> corrupted names otherwise OCL tooling that works on UML names gets confused.

Fortunately the erroneous nested packages all have URIs, so OCL tooling can be adjusted to prefer a defined URI for both nested and root packages over the nested name.
Comment 2 Camille Letavernier CLA 2015-01-27 05:10:08 EST
The SysML 1.4 profile is properly formatted (Upper case for UML Packages, lower case for Ecore Package, with originalName EAnnotation), so I think this will be fixed as soon as SysML 1.4 is promoted in Papyrus
Comment 3 Ed Willink CLA 2015-01-27 06:18:39 EST
Since the originalName EAnnotation was introduced in Luna, there is a case for the Luna maintenance release exploiting it.

But changing models in maintenance releases is risky.

My understanding is that SysML 1.4 will be an option, not a replacement in Mars, so perhaps the Mars SysML 1.2 model could be corrected.
Comment 4 Camille Letavernier CLA 2015-01-27 07:18:19 EST
> My understanding is that SysML 1.4 will be an option, not a replacement in Mars, so perhaps the Mars SysML 1.2 model could be corrected.

It's not decided yet. *If* we can provide a 100% SysML 1.1 to SysML 1.4 automated translation, then we will only ship SysML 1.4. Otherwise, we will ship two independent versions

Otherwise, we may simply regenerate SysML 1.1 to provide the originalName EAnnotation
Comment 5 Klaas Gadeyne CLA 2015-12-04 08:17:33 EST
(In reply to Camille Letavernier from comment #4)
> > My understanding is that SysML 1.4 will be an option, not a replacement in Mars, so perhaps the Mars SysML 1.2 model could be corrected.
> 
> It's not decided yet. *If* we can provide a 100% SysML 1.1 to SysML 1.4
> automated translation, then we will only ship SysML 1.4. Otherwise, we will
> ship two independent versions
> 
> Otherwise, we may simply regenerate SysML 1.1 to provide the originalName
> EAnnotation

What's the status on this one?
Comment 6 Benoit Maggi CLA 2015-12-04 10:55:35 EST
I want to make the ocl constraint works before adding profiled informations
Comment 7 Ed Willink CLA 2015-12-04 12:24:46 EST
(In reply to Benoit Maggi from comment #6)
> I want to make the ocl constraint works before adding profiled informations

What does this mean? This bug is about a bug that prevents OCL working so surely it must be fixed to "make the ocl constraint works".