Re: SingleObjectQuery & ObjectID

From: Andrus Adamchik (andru..bjectstyle.org)
Date: Sat Feb 25 2006 - 00:14:50 EST

  • Next message: Andrus Adamchik: "Re: HSQLDB"

    For the full object refresh, including relationships, do
    DataContext.invlidateObjects(..).

    Andrus

    On Feb 24, 2006, at 12:33 PM, Dave Merrin wrote:
    > Hi Andrus,
    >
    > cheers for that. I've now run into the problem that my to-many
    > relationships
    > (from the newly retrieved object)(and possibly to-one
    > relationships) are
    > picking data up from the cache. How can I stop this from happening?
    >
    > Cheers,
    >
    > Dave
    >
    >> -----Original Message-----
    >> From: Andrus Adamchik [mailto:andru..bjectstyle.org]
    >> Sent: 23 February 2006 17:11
    >> To: cayenne-use..bjectstyle.org
    >> Subject: Re: SingleObjectQuery & ObjectID
    >>
    >>
    >>
    >> On Feb 23, 2006, at 12:00 PM, Dave Merrin wrote:
    >>
    >>> Hi,
    >>>
    >>> I've just come across the caching feature of objectForPK. This is
    >>> causing me
    >>> problems so I've had a look at the SingleObjectQuery. This looks
    >>> much more
    >>> like what I need.
    >>
    >> Yep, that's the way to go if you want to control cache behavior
    >> manually.
    >>
    >>
    >>> Now I'd like to modify my template to provide a method
    >>> which creates an ObjectId given a pk value. Something a bit like:
    >>>
    >>> createObjectID(int value)
    >>> {
    >>> ObjectId result = new ObjectId(this.tablename,
    >> this.pkColumn, value);
    >>> return result;
    >>> }
    >>>
    >>> This should be fine but what about the tables that have compound
    >>> primary
    >>> keys? Can I create an ObjectId given multiple columns?
    >>
    >> Absolutely. Look at other ObjectId constructors - ObjectId can take a
    >> map of values.
    >>
    >> Andrus
    >>
    >
    >



    This archive was generated by hypermail 2.0.0 : Sat Feb 25 2006 - 00:14:54 EST