Need help on Triggers

Question

Hi,
I need to write a trigger on the prr_no so that whenever it is inserted or updated the prr_ind (which is again in the same table) is automatically set to 'Y' if prr_no <> null and prr_ind set to 'N' if prr_no = null
I am using Oracle 11g, and following are the triggers I wrote, not sure which one is correct(as per my requirement):
CREATE OR REPLACE TRIGGER CMISTST1.update_PRR_IND
AFTER INSERT OR UPDATE OF PRR_NO
ON CMISTST1.T_ET_CCR_HDR
FOR EACH ROW WHEN (new.PRR_NO IS NOT NULL)
BEGIN
UPDATE T_ET_CCR_HDR
SET T_ET_CCR_HDR.PRR_IND = 'Y';
END;
*****OR*****
CREATE OR REPLACE TRIGGER CMISTST1.update_PRR_IND
AFTER INSERT OR UPDATE OF PRR_NO
ON CMISTST1.T_ET_CCR_HDR
FOR EACH ROW WHEN (new.PRR_NO IS NULL)
BEGIN
UPDATE T_ET_CCR_HDR
SET T_ET_CCR_HDR.PRR_IND = 'Y'
END;
but neither works! Here CMISTST1 is the schema, t_et_ccr_hdr is the table.
Thanks in advance!

Answer 1

Elaborating on Justin's final remark and the fact you're on 11G, you could consider replacing the trigger by a virtual column:
SQL> create table t (prr_no number, prr_ind as (decode(prr_no, null, 'N', 'Y')));
Table created.
SQL> insert into t (prr_no) values (1);
1 row created.
SQL> insert into t (prr_no) values (null);
1 row created.
SQL> insert into t (prr_no) values (2);
1 row created.
SQL> insert into t (prr_no) values (null);
1 row created.
SQL> commit;
Commit complete.
SQL> exec dbms_stats.gather_table_stats(user, 'T', cascade => true);
PL/SQL procedure successfully completed.
SQL> select /*+ gather_plan_statistics */
  2         *
  3  from   t
  4  where  prr_ind = 'Y';
    PRR_NO P
         1 Y
         2 Y
2 rows selected.
SQL> select * from table(dbms_xplan.display_cursor(null,null,'ALLSTATS LAST'));
PLAN_TABLE_OUTPUT
SQL_ID  6sz2rzn1dwf6z, child number 0
select /*+ gather_plan_statistics */        * from   t where  prr_ind =
'Y'
Plan hash value: 1601196873
| Id  | Operation         | Name | Starts | E-Rows | A-Rows |   A-Time   | Buffers |
|   0 | SELECT STATEMENT  |      |      1 |        |      2 |00:00:00.01 |       8 |
|*  1 |  TABLE ACCESS FULL| T    |      1 |      2 |      2 |00:00:00.01 |       8 |
Predicate Information (identified by operation id):
   1 - filter(DECODE(TO_CHAR("PRR_NO"),NULL,'N','Y')='Y')
19 rows selected.
SQL> create index t_i on t(prr_ind);
Index created.
SQL> exec dbms_stats.gather_table_stats(user, 'T', cascade => true);
PL/SQL procedure successfully completed.
SQL> select /*+ gather_plan_statistics */
  2         *
  3  from   t
  4  where  prr_ind = 'Y';
    PRR_NO P
         1 Y
         2 Y
2 rows selected.
SQL> select * from table(dbms_xplan.display_cursor(null,null,'ALLSTATS LAST'));
PLAN_TABLE_OUTPUT
SQL_ID  6sz2rzn1dwf6z, child number 0
select /*+ gather_plan_statistics */        * from   t where  prr_ind =
'Y'
Plan hash value: 1658221841
| Id  | Operation                   | Name | Starts | E-Rows | A-Rows |   A-Time   | Buffers |
|   0 | SELECT STATEMENT            |      |      1 |        |      2 |00:00:00.01 |       4 |
|   1 |  TABLE ACCESS BY INDEX ROWID| T    |      1 |      2 |      2 |00:00:00.01 |       4 |
|*  2 |   INDEX RANGE SCAN          | T_I  |      1 |      2 |      2 |00:00:00.01 |       2 |
Predicate Information (identified by operation id):
   2 - access("PRR_IND"='Y')
20 rows selected.So, you can index a virtual column as well (or make it a key-column and so on...), as showed, just in case you were wondering.
Changing column values through triggers can be considered as 'doing stuff automagically'.
A virtual column would be more self-documenting as well, since it's part of the CREATE TABLE script.