I'm not exactly sure why it happened.  And I wasn't suggesting it was
a priority, either.  If I have time, I'll see if I can isolate it
(since I obviously have the EOModel that triggered it).  I just
double-checked the 25 or so EOFetchSpecifications that got imported
and the only other anomaly I saw was where "coreUserTypeId <> '0'"
became "coreUserTypeId != null" on a varchar2(1) column.  Again, if I
have time, I can try to figure it out later.

Thanks!

/dev/mrg


On Mon, Mar 16, 2009 at 3:30 AM, Andrus Adamchik <[email protected]> wrote:
> Not sure why it didn't work. Certainly worth fixing, although it will be
> very low priority on my own list (it may be different for others).
>
> Andrus
>
>
> On Mar 13, 2009, at 9:56 PM, Michael Gentry wrote:
>
>> I imported an EOModel into Cayenne Modeler and it seemed to go
>> smoothly until at runtime it blew up trying to parse a qualifier.  I
>> looked in the Map XML and found (as part of a <query>...</query>:
>>
>> <qualifier><![CDATA[db: = "OSIS "]]></qualifier>
>> <ordering><![CDATA[contractNum]]></ordering>
>>
>> This looked very strange and I went back to the EOModel and it had:
>>
>> (contractOwner = 'OSIS ')
>>
>> (Yes, trailing white space and all.)  Clearly the import messed this
>> particular one up, but I don't know why.  Other EOFetchSpecifications
>> that were very similar worked fine and eye-balling the other
>> qualifiers looked OK, too.  I changed the Cayenne XML to:
>>
>> <qualifier><![CDATA[contractOwner = "OSIS "]]></qualifier>
>> <ordering><![CDATA[contractNum]]></ordering>
>>
>> And have moved on.
>>
>> Any thoughts as to what might have caused this and if it is worth
>> trying to isolate and fix?
>>
>> Thanks,
>>
>> /dev/mrg
>>
>
>

Reply via email to