Some Eclipse Foundation services are deprecated, or will be soon. Please ensure you've read this important communication.

Bug 335760

Summary:	Spurious Possible incompatible type between ...
Product:	[Modeling] Acceleo	Reporter:	Ed Willink <ed>
Component:	Core	Assignee:	Project Inbox <acceleo-inbox>
Status:	CLOSED FIXED	QA Contact:
Severity:	normal
Priority:	P3	CC:	laurent.goubet, stephane.begaudeau
Version:	3.0.0
Target Milestone:	---
Hardware:	PC
OS:	Windows Vista
Whiteboard:

Description Ed Willink

2011-01-29 07:20:55 EST

3.1.0M4a

Just about every for loop reports a spurious warning such as

Possible incompatible type between 'Type' and 'OrderedSet(Type)'.

Comment 1 Laurent Goubet

2011-01-29 13:54:18 EST

If I remember correctly, we've already fixed this since. Could you check with the latest promoted Nightly of the 3.1.0? (http://download.eclipse.org/modeling/m2t/acceleo/updates/nightly/3.1)

Comment 2 Ed Willink

2011-02-02 13:36:33 EST

Problem persists in M5.

Comment 3 Ed Willink

2011-02-02 13:38:52 EST

(Since in 336119 I have a problem with an unresolved proxy (EcorePackage), perhaps there's an unresolved proxy behind this problem too. Suggest explicit diagnostics for unresolved proxies.)

Comment 4 Stephane Begaudeau

2011-02-09 08:39:24 EST

For me the problem has been fixed, and it should work since M5. I'll leave the bug open until you can confirm that everything works for you. I'll investigate this anyway in the case of a proxy for M6.

Comment 5 Ed Willink

2011-03-15 16:19:47 EDT

Still a problem in M6.

For instance see CVS for

org.eclipse.mdt/org.eclipse.ocl/examples/org.eclipse.ocl.examples.build/src/org/eclipse/ocl/examples/build/acceleo/generateOclCommon.mtl

Comment 6 Stephane Begaudeau

2011-03-19 10:46:05 EDT

I was working on a small generator at home and this problem re-appeared when I was working with a dynamic model. When I switched to a new Eclipse instance, the problem disappeared. I'll look into it Monday to see if it's really link to dynamic models. The comparison is done between an OCL expression and an "Acceleo variable declaration", maybe we have two versions of the same metamodel in memory, one in the OCL parser and one in the Acceleo parser.

Comment 7 Stephane Begaudeau

2011-03-21 12:30:07 EDT

I have identified and fixed two problems with dynamic models. It is now working with all my examples, yet I'm pretty sure that there are still some problem with that algorithm. I'll have to test this more extensively tomorrow.