I can't substitute the driver, because it is in the common/lib directory of
Tomcat on the testing machine.
To pass the option oracle.jdbc.V8Compatible="true" in cayenne, how do I do
it? I have already try to write it
in context.xml, but don't load it. DataContext have not methods to accessing
to properties of JDBC ......
Thanks!
2009/1/28 Joe Baldwin <jfbaldwi..arthlink.net>
> Is it possible that the problem is with the JDBC adapter version? I
> googled Oracle Date and found this brain-numbing essay on
> Oracle-JDBC-DATE-Date-Timestamp and the the version complications as well.
>
>
> http://www.oracle.com/technology/tech/java/sqlj_jdbc/htdocs/jdbc_faq.html#08_01
>
> (I am even more impressed with the Cayenne team after reading this FAQ.)
>
> Given that I have only a small clue as to what your error message means, I
> would recommend that you start by verifying the versions of your Oracle
> component configuration against this FAQ from Oracle. If your versions are
> in fact the ones you anticipate (and are not the cause of the Date/Timestamp
> controversy), only then I would go back to concentrating on the cayenne
> configuration.
>
> HTH,
> Joe
>
>
>
>
> On Jan 28, 2009, at 12:31 PM, Alessio Giovanni Baroni wrote:
>
> Nothing, returns an exception ("getTimestamp is not valid for getT4CNumber
>> .....").
>> If I use "TO_CHAR", there is another problem, because when I do get(...)
>> from DataObject, Cayenne execute
>> a query, not equal as the query in ctxt.performQuery(Bar.class,
>> "SELEC........"); Cayenne not save the TO_CHAR!
>>
>> Help?
>>
>> Thanks.
>>
>>
>> 2009/1/28 Andrus Adamchik <andru..bjectstyle.org>
>>
>> java.sql.Timestamp extends java.util.Date, so it is ok.
>>>
>>> (and in any event doublecheck that the column in a table for the Bar
>>> entity
>>> is mapped as TIMESTAMP).
>>>
>>> Andrus
>>>
>>>
>>> On Jan 28, 2009, at 5:34 PM, Alessio Giovanni Baroni wrote:
>>>
>>> But if I have SQLTemplate(Bar.class, .....), and in class Bar I set the
>>>
>>>> field Foo as Date, isn't an error cast it to java.lang.Timestamp??
>>>>
>>>> Thanks.
>>>>
>>>>
>>>> 2009/1/28 Andrus Adamchik <andru..bjectstyle.org>
>>>>
>>>> My (admittedly novice) solution was to map it via Modeler to a
>>>> TIMESTAMP.
>>>>
>>>>>
>>>>>
>>>>>>
>>>>>>
>>>>> This is actually a correct solution. Only if I am not mistaken it
>>>>> doesn't
>>>>> work 100% on 2.0.x (IIRC it works for SelectQuery, but not
>>>>> SQLTemplate).
>>>>> In
>>>>> 3.0 it should work with SQLTemplate as well.
>>>>>
>>>>> Andrus
>>>>>
>>>>>
>>>>>
>>>>> On Jan 28, 2009, at 5:12 PM, Joe Baldwin wrote:
>>>>>
>>>>> Andrus,
>>>>>
>>>>>
>>>>>> I looked up the Oracle Date definition and it does in fact keep time
>>>>>> info
>>>>>> in the database. My (admittedly novice) solution was to map it via
>>>>>> Modeler
>>>>>> to a TIMESTAMP. Does this reveal the same problem with the Oracle
>>>>>> adapter
>>>>>> as mapping it via Modeler-DATE?
>>>>>>
>>>>>> Joe
>>>>>>
>>>>>>
>>>>>> On Jan 28, 2009, at 9:58 AM, Andrus Adamchik wrote:
>>>>>>
>>>>>> I guess that's how Oracle driver returns the metadata for the DATE
>>>>>> column
>>>>>>
>>>>>> in the ResultSet. The workaround is to use #result() directive in the
>>>>>>> SQL:
>>>>>>>
>>>>>>> http://cayenne.apache.org/doc20/scripting-sqltemplate.html
>>>>>>>
>>>>>>> Hope this helps,
>>>>>>>
>>>>>>> Andrus
>>>>>>>
>>>>>>> On Jan 28, 2009, at 4:48 PM, Alessio Giovanni Baroni wrote:
>>>>>>>
>>>>>>> Hi to all,
>>>>>>>
>>>>>>> I have the following code:
>>>>>>>>
>>>>>>>> .............
>>>>>>>> query = new SQLTemplate("SELECT FOO FROM BAR");
>>>>>>>> results = ctxt.performQuery(Bar.class, query);
>>>>>>>> for(int i = 0; i < results.size() - 1; ++i)
>>>>>>>> {
>>>>>>>> Bar f = (Bar)results.get(i);
>>>>>>>> ...... (f.getFoo());
>>>>>>>> ....................
>>>>>>>> }
>>>>>>>>
>>>>>>>> In the Oracle DB the field FOO is a DATE and I map it to
>>>>>>>> java.util.Date.
>>>>>>>> When I do .getFoo(), the date that I have is without the
>>>>>>>> hours/minutes/seconds.
>>>>>>>> Why? How do I resolve it?
>>>>>>>>
>>>>>>>> Thanks.
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>
>
This archive was generated by hypermail 2.0.0 : Thu Jan 29 2009 - 04:17:24 EST