Andrus Adamchik <andru..bjectstyle.org> wrote:
> > .... It doesn't appear to pick up existing Oracle sequences.
> Well, yeah, Cayenne does not reverse engineer sequences.
I certainly can see why that's not the case, but any possiblity of a future
methodology where we can give the modeler hints on how to look for a
possible Oracle Sequence file name pattern? Ie, "look for sequences
matching SEQ_* where * is the table name"
Definitely a low-priority item....
> > I'm also seeing some odd "inserted value is too large for column"
> > errors, but unfortunately DBUnit is horrible for reporting what data
> > it's trying to insert, so I don't know for sure it's a porting
> > problem.
>
> Also not sure sure what this is. If you manage to get to the cause,
> please let me know.
This is looking like user error. I was using DBUnit to set a "Y"/"N" char
field to Boolean.FALSE. Openbase didn't care and must have assigned some
value to it (probably 0 based on past experience) while Oracle just said
that 'false' doesn't fit in a CHAR(1).
The tables all see to be correct -- although Cayenne creates NUMBER(38) from
NUMBER types -- once the TIMESTAMP issue is dealt with. I'll know more if
all of the unit tests run through.
-Mike
This archive was generated by hypermail 2.0.0 : Wed Oct 06 2004 - 14:54:10 EDT