I have the following SQL query:
select
ID, COLUMN1, COLUMN2
from
(select ID, COLUMN1, COLUMN2, row_number() over (order by 2 DESC) NO from A_TABLE)
whe
Why don't you use
Select top 100 ID, Column1, Column2
From A_Table
where Column1 like '%value%'
order by Column2 desc
Check out this Oracle FAQ. In particular this part:
SELECT *
FROM (SELECT a.*, rownum RN
FROM (SELECT *
FROM t1 ORDER BY key_column) a
WHERE rownum <=7)
WHERE rn >=5
Here you will get the limited record from the oracle database without the usage of rownum
select * from
( select ,column1,column2,row_number() over (order by columnName) as rnum
from table_name)
where rnum between 5 and 10;
rownum is a pseudo column that counts rows in the result set after the where clause has been applied.
Is this what you're trying to get?
SELECT *
FROM (
SELECT id, column1, column2
FROM atable ORDER BY 2 DESC
)
WHERE ROWNUM < 100;
Because it's a pseudo column that is strictly a counter of rows resulting from the where clause it will not allow you to do pagination (i.e. between 200 & 300).
This is probably what you're looking for:
SELECT *
FROM
(SELECT a.*, rownum rnum FROM
(SELECT id, column1, column2 FROM atable ORDER BY 2 DESC) a WHERE rownum <= 300)
WHERE rnum >= 200;
To answer your first question: Don't use a column number in your order by clause, but use the column name. I don't fully understand your second question, because adding a WHERE in your most inner SELECT should do the trick:
select ID
, COLUMN1
, COLUMN2
from (select ID
, COLUMN1
, COLUMN2
, row_number() over (order by COLUMN1 DESC) NO
from A_TABLE
where COLUMNX LIKE '%SOME VALUE%'
)
where NO between 0 and 100
P.S. (to willcodejavaforfood) I think using row_number() is better when you want the rows to be ordered. It saves an inner view (big win for readability).