Hi Guys,
The foreign key convention is supported and is demonstrated in this
test: ImpliedRelationshipTests.testAddNewOrder()
The convention is lightly documented here:
http://wiki.apache.org/ws/ConventionOverConfiguration
Thanks,
--Kevin
Luciano Resende wrote:
Have you setup the relationship on your DAS config file ? Should look
something like the example below :
<Relationship name="BDATA" primaryKeyTable="A" foreignKeyTable="B"
many="true">
<KeyPair primaryKeyColumn="ID" foreignKeyColumn="B_ID"/>
</Relationship>
See more info around relationships on [1] and [2].
Now, if what you want is to have the relationship done via Convention
over
Configuration, I don't think we have that implemented, our set of
implemented conventions are limited at the moment.
[1] - http://wiki.apache.org/ws/ForeignKeyRepresentationAndManagement
[2] - http://wiki.apache.org/ws/WorkingWithRelationships
On 4/7/07, Adriano Crestani <[EMAIL PROTECTED]> wrote:
Another doubt:
I have the following tables on my database:
create table B (
ID INTEGER NOT NULL PRIMARY KEY
);
create table A (
ID INTEGER NOT NULL PRIMARY KEY,
B_ID INTEGER,
FOREIGN KEY (B_ID) REFERENCES B(ID)
);
OK, then I was analyzing the generated sdo graph when I execute the
query
"select * from A, B;", I got something like this:
Graph1:
http://cwiki.apache.org/confluence/download/attachments/45093/das_graph.JPG
I understand the column B_ID would be considered only an integer column
when
of table A when there is no B table returned on the query. However,
the B
table is also returned on the query and as the B_ID column is a FK to
the
table B, shouldn't the graph be like this?:
Graph2:
http://cwiki.apache.org/confluence/download/attachments/45093/das+
graph+as+I+suppose+it+shoud+be.JPG
<
http://cwiki.apache.org/confluence/download/attachments/45093/das+graph+as+I+suppose+it+shoud+be.JPG
>
This way I think the data returned by the query would be better
represented
on the graph. Maybe there is a reason for the graph to be generated as
Graph1 instead of Graph2.
What do you say guys?
Adriano Crestani
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
