| OCL expression updates [message #1295463] |
Mon, 14 April 2014 05:37  |
Eclipse User |
|
|
|
Hey,
Maybe I've not work enough with OCL API but....I've a "problem".
Note that I use Rational Software Modeler.
OCL is stored as a String so when I change the name of a property in a class (for example), when I load an OCL expression that use this property, my expression becomes invalid.
The reason is that OCL store the name of the property and not its internal ID.
So, if I'm right, don't you think that it could have been a good idea to store the ID of any meta-model element instead of its name. The OCL expression with the name will be built by the editor when necessary.
Note that I've created a basic SQL expression editor in for an UML profile and I store the ID so that the expressions becomes "renaming agnostic".
[Updated on: Mon, 14 April 2014 05:37] by Moderator |
|
|
|
| Re: OCL expression updates [message #1295503 is a reply to message #1295498] |
Mon, 14 April 2014 06:18  |
| Eclipse User |
|
|
|
Hi
https://bugs.eclipse.org/bugs/show_bug.cgi?id=432720
is the corresponding bug report for Papyrus.
Regards
Ed Willink
On 14/04/2014 11:10, Ed Willink wrote:
> Hi
>
> You are reporting a bug in the refactoring operation.
>
> As you say a String representation of OCL is blind to refactoring.
>
> However a compiled representation such as an ExpressionInOCL contains
> metamodel references as you suggest and will track many refactorings.
>
> So you should report a bug against RSM for failing to
> compile/pretty-print all OCL around refactoring operations.
>
> Regards
>
> Ed Willink
>
>
> On 14/04/2014 10:37, Missing name Mising name wrote:
>> Hey,
>>
>> Maybe I've not work enough with OCL API but....I've a "problem".
>>
>> Note that I use Rational Software Modeler.
>>
>> OCL is stored as a String so when I change the name of a property in
>> a class (for example), when I load an OCL expression that use this
>> property, my expression becomes invalid.
>> The reason is that OCL store the name of the property and not its
>> internal ID.
>>
>> So, if I'm right, don't you think that it could have been a good idea
>> to store the ID of any meta-model element instead of its name. The
>> OCL expression with the name will be built by the editor when necessary.
>>
>> Note that I've created a basic SQL expression editor in for an UML
>> profile and I store the ID so that the expressions becomes "renaming
>> agnostic".
>
|
|
|
Powered by
FUDForum. Page generated in 0.04715 seconds
Home