Tabular form on multiple tables

Question

Hi there,
I'm hoping to create a Tabular Form which is based on either a view or a join, whichever is easier.
We have two tables - Division and Targets. Division references Targets on target_id. The tables share a one-to-one relationship. I was originally told to simply merge the 36 fields in Targets to the Division table, however, we're likely going to have other tables (such as Department and Section) which also share a one-to-one relationship with Targets.
Is there a way to accomplish this with the Wizard, and if not, does anyone have any tutorials/advice on how to create PL/SQL to handle this?
Thanks,
Andrew

Answer 1

Mike
"Just for querying the data and making it appear in a single tabular form"
... hum so you would have to create a custom process anyway for the update - in which case, why have the view in the first place, there's no point.
>
What are the odds of moving the rows :P works for me anyways =P.
>
I'm not saying it's likely, but it can happen. As a developer you should take this into consideration when designing applications.
So
CREATE TABLE rowid_test
  pk_col             NUMBER PRIMARY KEY,
  col1               NUMBER,
  col2               VARCHAR2(6 CHAR)
TABLESPACE DAT_DATA;
CREATE SEQUENCE rowid_test_seq
  START WITH 1
  MAXVALUE 999999999999999999999999999
  MINVALUE 1
  NOCYCLE
  CACHE 20
  NOORDER;
BEGIN
  FOR i IN 1 .. 150
  LOOP
    INSERT INTO rowid_test
    (pk_col,
     col1,
     col2)
    VALUES
    (rowid_test_seq.nextval,
     ROUND(DBMS_RANDOM.VALUE(0, 100000), 2),
     DBMS_RANDOM.STRING('x', 6));
  END LOOP;
  COMMIT;
END;
CREATE OR REPLACE FORCE VIEW rowid_view AS
(SELECT rowid the_id, t.*
FROM rowid_test t);
SELECT *
FROM (SELECT /*+ FIRST_ROWS(n) */
             a.*, ROWNUM rnum
      FROM(SELECT *
           FROM rowid_view
           ORDER BY pk_col) a
      WHERE ROWNUM <= 30)
WHERE rnum  >= 15;
DELETE FROM rowid_test
WHERE pk_col = 17;
COMMIT;
INSERT INTO rowid_test
           (pk_col,
            col1,
            col2)
     VALUES
           (17,
            ROUND(DBMS_RANDOM.VALUE(0, 100000), 2),
            DBMS_RANDOM.STRING('x', 6));
COMMIT;
SELECT *
FROM (SELECT /*+ FIRST_ROWS(n) */
             a.*, ROWNUM rnum
      FROM(SELECT *
           FROM rowid_view
           ORDER BY pk_col) a
      WHERE ROWNUM <= 30)
WHERE rnum  >= 15;So row 17 could have a different rowid in the tabular form to the table itself.
If the rowid is used to update then it could either not find the row or find a different row (ludicriously unlikely but you never know).
So the next option would be to combine it with a primary key - very safe but you may not always find your row. So why not just use the primary key? Well, in this case we have two tables involved so we do not have a single PK on which to base the update... So we'll have to write a custom update procedure as APEX will base it upon the primary key... so not having a primary key is no longer an issue.
Either way, your query is ambiguos
CREATE OR REPLACE VIEW "HIHI" AS
SELECT ROWID unique_rowid, d.Dep "Department", t.goal "Sales Goal"
FROM Division d, Target t
WHERE d.target_id = t.target_idWhich table is the rowid coming from??? This couldn't work.
Cheers
Ben
http://www.munkyben.wordpress.com
Don't forget to mark replies helpful or correct ;)
Edited by: Munky on Aug 28, 2009 10:53 AM