You could:
INSERT INTO REGION VALUES (33, 'New Dar');
UPDATE DISTRICT SET region_id = 33 WHERE region_id = 99;
DELETE FROM REGION WHERE region_id = 99;
UPDATE REGION SET region_name = 'Dar es Salaam' WHERE region_id = 33;
Of course, if there is no uniqueness constraint on region_name then
yo
The idea of id's is that they are meaningless, so saying "this row was supposed
to be 33" is senseless.
If you want Dar es Salaam to be 33 because eg it's the postal code, then add a
column postal_code to your region table
but keep the id to make the reference.
>>> James Kitambara <[EMAIL PROTEC
James Kitambara wrote:
> For this UPDATE I wanted, when I change the region _id from '99' to
> '33' of the last ROW in REGION table AUTOMATICALLY to change the
> last three ROWS of the DISTRICT table which reference to '99', 'Dar
> es Salaam'.
>
> If I do this, I will get the error message "You
Thank you !
But I think that there is a solution.
If it happens that you have the following data in your tables
REGION
--
region_id | region_name
--
11| Dodoma
22| Tabora
99 | Dar es Salaa
I think (one of) the point(s) of id fields is not to change them. You can
update the region_name field (eg a correct a misspelling), but the id stays the
same.
That way the district stays connected to the same region.
>>> James Kitambara <[EMAIL PROTECTED]> 2008-09-17 8:50 >>>
Hello Mambers o
Hello Mambers of PGSQL-SQL,
I have two tables namely:
REGION (region_id, region_name)
DISTRICT (dist_id, dist_name, region_id (FK))
I would like to have the CREATE TABLE Command which will create these tables in
such a way that when REGION table is UPDATED automatical the FOREGN KE
