Re: Bug in nested DataContext?

From: Andrus Adamchik (andru..bjectstyle.org)
Date: Fri Oct 05 2007 - 09:30:45 EDT

  • Next message: Andrus Adamchik: "Preparing 1.2.4/2.0.4"

    > So just changing the order of operations was the trick? Perhaps it
    > should just be documented that you remove from relationships before
    > you delete from the context?

    I'd say Cayenne should handle that. When I encounter quirks like that
    in other frameworks, I tend to jump to quick conclusions about the
    quality of a given framework. So this is one of those small things we
    should fix to ensure that the user impression of Cayenne is always
    "it just works" :-)

    Andrus

    On Oct 5, 2007, at 3:58 PM, Michael Gentry wrote:
    > So just changing the order of operations was the trick? Perhaps it
    > should just be documented that you remove from relationships before
    > you delete from the context?
    >
    > I think the main reason I did it in the order I did was because, in my
    > test, right after that I had:
    >
    > childContext.deleteObject((Detail) childM.getDetails().get(0));
    > childM.removeFromDetails((Detail) childM.getDetails().get(0));
    >
    > If I had reversed the order, I'd have needed a variable to store
    > childM.getDetails().get(0). How is that for lazy? :-)
    >
    > /dev/mrg
    >
    >
    > On 10/5/07, Andrus Adamchik <andru..bjectstyle.org> wrote:
    >> I committed a fix. Per my Jira comment there is a "correct" way to go
    >> about it that requires a bit of object lifecycle redesign, but my
    >> fairly straightforward workaround should work for this particular
    >> case.
    >>
    >> Andrus
    >>
    >>
    >> On Oct 4, 2007, at 9:57 PM, Michael Gentry wrote:
    >>
    >>> I'm not sure if it matters, but in the test program I only pulled
    >>> the
    >>> Master object into the child DC. Is it important to pull the Detail
    >>> objects into the child DC as well? For some reason, I was under the
    >>> impression they would get pulled into the child DC automatically,
    >>> but
    >>> perhaps I am mistaken?
    >>>
    >>> Thanks,
    >>>
    >>> /dev/mrg
    >>>
    >>
    >>
    >



    This archive was generated by hypermail 2.0.0 : Fri Oct 05 2007 - 09:31:09 EDT