Repetive problem with indexed attribute
Chris Muller
asqueaker at gmail.com
Fri Feb 8 05:01:29 UTC 2008
Hilaire, thanks for the report. However, I can't make a determination
whether you've identified a bug or not since I can't see what you have
for your #functions getter. And there's a lot of other code I don't
have the benefit of seeing, so can only pretty much just speculate
about the possibilities..
It sounds like you've narrowed the problem down to a reproducible
test-case script.
If you can show me how to reproduce the problem, perhaps with a
workspace script that adds instances of Association indexed on #key
and/or #value, I'd be happy to work on it.
Thanks,
Chris
On Feb 7, 2008 9:00 AM, Hilaire Fernandes <hilaire at ofset.org> wrote:
> In my quest to write tests to control Magma behaviour, I have found the
> following problem:
>
> With a magma collection of users object defined as follow:
>
> | users |
> users := MagmaCollection new.
> users
> addIndex: (MaIntegerIndex attribute: #id);
> addIndex: (MaKeywordIndex attribute: #functions);
> addIndex: (MaSearchStringIndex attribute: #schoolYear).
> ^ users
>
> I have problem with the #functions key.
> functions is instanciated as an empty OrderedCollection.
> I can update function with #function: message
>
> functions: aCollection
> (self class functions includesAllOf: aCollection)
> ifTrue:
> [IFIDbSession commit:
> [IFIDbSession noteOldKeyFor: self.
> functions := aCollection]]
>
>
> In my test, I can do:
>
> adult functions: #(#teacher).
> self should: adult functions = #(#teacher).
>
> but the next one got the error (see SqueakLog enclosed)
>
> adult functions: #(#teacher #admin).
> self should: adult functions = #(#teacher #admin).
>
> If I change the previous test code to:
> adult functions: #(#director #admin ).
> self should: adult functions = #(#director #admin ).
>
> It is fine.
>
> It seems the problem appear when two successive changes to functions has
> common value, here #teacher was in both #(#teacher) and #(#teacher
> #admin)
>
> Please help!
>
>
> _______________________________________________
> Magma mailing list
> Magma at lists.squeakfoundation.org
> http://lists.squeakfoundation.org/mailman/listinfo/magma
>
>
More information about the Magma
mailing list