Using an Alias in a WHERE clause

Question

I have a query which is meant to show me any rows in table A which have not been updated recently enough. (Each row should be updated within 2 months after "month_no".):

SELECT A.identifier
     , A.name
     , TO_NUMBER(DECODE( A.month_no
             , 1, 200803 
             , 2, 200804 
             , 3, 200805 
             , 4, 200806 
             , 5, 200807 
             , 6, 200808 
             , 7, 200809 
             , 8, 200810 
             , 9, 200811 
             , 10, 200812 
             , 11, 200701 
             , 12, 200702
             , NULL)) as MONTH_NO
     , TO_NUMBER(TO_CHAR(B.last_update_date, 'YYYYMM')) as UPD_DATE
  FROM table_a A
     , table_b B
 WHERE A.identifier = B.identifier
   AND MONTH_NO > UPD_DATE

The last line in the WHERE clause causes an "ORA-00904 Invalid Identifier" error. Needless to say, I don't want to repeat the entire DECODE function in my WHERE clause. Any thoughts? (Both fixes and workarounds accepted...)

score 118 · Accepted Answer · edited May 23 '17 at 12:17

This is not possible directly, because chronologically, WHERE happens before SELECT, which always is the last step in the execution chain.

You can do a sub-select and filter on it:

SELECT * FROM
(
  SELECT A.identifier
    , A.name
    , TO_NUMBER(DECODE( A.month_no
      , 1, 200803 
      , 2, 200804 
      , 3, 200805 
      , 4, 200806 
      , 5, 200807 
      , 6, 200808 
      , 7, 200809 
      , 8, 200810 
      , 9, 200811 
      , 10, 200812 
      , 11, 200701 
      , 12, 200702
      , NULL)) as MONTH_NO
    , TO_NUMBER(TO_CHAR(B.last_update_date, 'YYYYMM')) as UPD_DATE
  FROM table_a A
    , table_b B
  WHERE A.identifier = B.identifier
) AS inner_table
WHERE 
  MONTH_NO > UPD_DATE

Interesting bit of info moved up from the comments:

There should be no performance hit. Oracle does not need to materialize inner queries before applying outer conditions -- Oracle will consider transforming this query internally and push the predicate down into the inner query and will do so if it is cost effective. – Justin Cave

score 16 · Answer 2 · edited Mar 21 '12 at 19:58

16

 SELECT A.identifier
 , A.name
 , TO_NUMBER(DECODE( A.month_no
         , 1, 200803 
         , 2, 200804 
         , 3, 200805 
         , 4, 200806 
         , 5, 200807 
         , 6, 200808 
         , 7, 200809 
         , 8, 200810 
         , 9, 200811 
         , 10, 200812 
         , 11, 200701 
         , 12, 200702
         , NULL)) as MONTH_NO
 , TO_NUMBER(TO_CHAR(B.last_update_date, 'YYYYMM')) as UPD_DATE
FROM table_a A, table_b B
WHERE .identifier = B.identifier
HAVING MONTH_NO > UPD_DATE

edited Mar 21 '12 at 19:58

wallyk

56,922
16
83
148

answered Jan 06 '12 at 06:48

me_an

489
4
11

1

`HAVING` is the real answer. `HAVING` is the clause checker for calculated columns, like `COUNT`, `MAX` and other expressions in the `SELECT` query, because it filters the final fetched data. – George Dimitriadis Dec 30 '19 at 16:02

score 11 · Answer 3 · edited Feb 24 '12 at 18:12

11

Or you can have your alias in a HAVING clause

edited Feb 24 '12 at 18:12

Taryn

242,637
56
362
405

answered Oct 15 '09 at 18:57

James

119
1
2

4

This would be an interesting approach, can you give any code? – rob5408 Nov 15 '10 at 02:16
the same rule as to where applies, so this is not a solution. – Alexey Nov 13 '15 at 16:44
I'm stuck on MySQL (5.5), don't know if this applies to Oracle. BUT: `SELECT CONCAT(names, surname) AS x FROM clients HAVING x LIKE '%a%'` works, while `SELECT CONCAT(names, surname) AS x FROM clients WHERE x LIKE '%a%'` fails ("Unknown column 'x' in 'where clause'") – frIT Dec 03 '15 at 11:23

score 2 · Answer 4 · answered Feb 14 '17 at 05:49

Just as an alternative approach to you can do:

WITH inner_table AS
(SELECT A.identifier
    , A.name
    , TO_NUMBER(DECODE( A.month_no
      , 1, 200803 
      , 2, 200804 
      , 3, 200805 
      , 4, 200806 
      , 5, 200807 
      , 6, 200808 
      , 7, 200809 
      , 8, 200810 
      , 9, 200811 
      , 10, 200812 
      , 11, 200701 
      , 12, 200702
      , NULL)) as MONTH_NO
    , TO_NUMBER(TO_CHAR(B.last_update_date, 'YYYYMM')) as UPD_DATE
  FROM table_a A
    , table_b B
  WHERE A.identifier = B.identifier)

    SELECT * FROM inner_table 
    WHERE MONTH_NO > UPD_DATE

Also you can create a permanent view for your queue and select from view.

CREATE OR REPLACE VIEW_1 AS (SELECT ...);
SELECT * FROM VIEW_1;

score 1 · Answer 5 · answered Apr 17 '18 at 13:22

It's possible to effectively define a variable that can be used in both the SELECT, WHERE and other clauses.

A subquery doesn't necessarily allow for appropriate binding to the referenced table columns, however OUTER APPLY does.

SELECT A.identifier
     , A.name
     , vars.MONTH_NO
     , TO_NUMBER(TO_CHAR(B.last_update_date, 'YYYYMM')) as UPD_DATE
FROM table_a A
     , table_b B ON A.identifier = B.identifier
OUTER APPLY (
   SELECT
        -- variables
        MONTH_NO = TO_NUMBER(DECODE( A.month_no
                     , 1, 200803 
                     , 2, 200804 
                     , 3, 200805 
                     , 4, 200806 
                     , 5, 200807 
                     , 6, 200808 
                     , 7, 200809 
                     , 8, 200810 
                     , 9, 200811 
                     , 10, 200812 
                     , 11, 200701 
                     , 12, 200702
                     , NULL))
) vars
WHERE vars.MONTH_NO > UPD_DATE

Kudos to Syed Mehroz Alam.

Using an Alias in a WHERE clause

5 Answers5

Linked

Related