| Tuple Type problems [message #27977] |
Thu, 14 June 2007 06:24  |
Eclipse User |
|
|
|
Hi,
We tried to parse the Tuple types but for some cases the results were not as
it's expected. The example of OCL Input below results in the same Tuple Type
for both seqInTuple1 and seqInTuple2, which is the Tuple contained of
Sequence(Integer). When the order of the OCL input is changed, both of the
elements have the type of Tuple contained of Sequence(String). We are using
mdt-ocl-SDK-1.1M7. The parsing process is invoked using parse(OCL Input)
operation of org.eclipse.ocl.ecore.OCL.
package p
context CTupleTypes
def def9:
seqInTuple2 : Tuple(names:Sequence(Integer)) = OclInvalid
context CTupleTypes
def def8:
seqInTuple1 : Tuple(names:Sequence(String)) = OclInvalid
endpackage
Any help is appreciated.
Best regards,
Jibran
|
|
|
| Re: Tuple Type problems [message #28017 is a reply to message #27977] |
Thu, 14 June 2007 10:08  |
| Eclipse User |
|
|
|
Originally posted by: cdamus.ca.ibm.com
Hi, Jibran,
That's a fairly severe bug, and the only way that you can work around it (I
think) is by using a different name for the "names" part of each of the
tuple types. e.g.,
Tuple(names:Sequence(String)) and Tuple(intNames:Sequence(Integer))
Please raise a bug and we'll see whether we can get this fixed in time for
the release. The tuple types should recognize the difference between
Sequence(Integer) and Sequence(String).
https://bugs.eclipse.org/bugs/enter_bug.cgi?product=MDT& component=OCL&version=1.1.0&severity=major
Cheers,
Christian
A. Jibran Shidqie wrote:
> Hi,
>
> We tried to parse the Tuple types but for some cases the results were not
> as it's expected. The example of OCL Input below results in the same Tuple
> Type for both seqInTuple1 and seqInTuple2, which is the Tuple contained of
> Sequence(Integer). When the order of the OCL input is changed, both of the
> elements have the type of Tuple contained of Sequence(String). We are
> using mdt-ocl-SDK-1.1M7. The parsing process is invoked using parse(OCL
> Input) operation of org.eclipse.ocl.ecore.OCL.
>
> package p
> context CTupleTypes
> def def9:
> seqInTuple2 : Tuple(names:Sequence(Integer)) = OclInvalid
>
> context CTupleTypes
> def def8:
> seqInTuple1 : Tuple(names:Sequence(String)) = OclInvalid
> endpackage
>
> Any help is appreciated.
>
> Best regards,
> Jibran
|
|
|
| Re: Tuple Type problems [message #28056 is a reply to message #28017] |
Fri, 15 June 2007 05:59 |
| Eclipse User |
|
|
|
Hi Christian,
I've added to the Bugzilla, as in the link below:
https://bugs.eclipse.org/bugs/show_bug.cgi?id=192832
Thanks
Best regards,
Jibran
"Christian W. Damus" <cdamus@ca.ibm.com> wrote in message
news:f4ri5n$k3a$1@build.eclipse.org...
> Hi, Jibran,
>
> That's a fairly severe bug, and the only way that you can work around it
> (I
> think) is by using a different name for the "names" part of each of the
> tuple types. e.g.,
>
> Tuple(names:Sequence(String)) and Tuple(intNames:Sequence(Integer))
>
> Please raise a bug and we'll see whether we can get this fixed in time for
> the release. The tuple types should recognize the difference between
> Sequence(Integer) and Sequence(String).
>
> https://bugs.eclipse.org/bugs/enter_bug.cgi?product=MDT& component=OCL&version=1.1.0&severity=major
>
> Cheers,
>
> Christian
>
>
> A. Jibran Shidqie wrote:
>
>> Hi,
>>
>> We tried to parse the Tuple types but for some cases the results were not
>> as it's expected. The example of OCL Input below results in the same
>> Tuple
>> Type for both seqInTuple1 and seqInTuple2, which is the Tuple contained
>> of
>> Sequence(Integer). When the order of the OCL input is changed, both of
>> the
>> elements have the type of Tuple contained of Sequence(String). We are
>> using mdt-ocl-SDK-1.1M7. The parsing process is invoked using parse(OCL
>> Input) operation of org.eclipse.ocl.ecore.OCL.
>>
>> package p
>> context CTupleTypes
>> def def9:
>> seqInTuple2 : Tuple(names:Sequence(Integer)) = OclInvalid
>>
>> context CTupleTypes
>> def def8:
>> seqInTuple1 : Tuple(names:Sequence(String)) = OclInvalid
>> endpackage
>>
>> Any help is appreciated.
>>
>> Best regards,
>> Jibran
>
|
|
|
Powered by
FUDForum. Page generated in 0.03510 seconds
Home