oracle-databasequery-tuninginline-view

Restructure Inline view


SELECT MAX(column1)
FROM table1 B , table2 A, table3 H
WHERE B.unit=A.unit
AND B.value=A.value
AND B.unit=H.unit
AND B.value=H.value
AND A.number=1234

Can someone help me to restructure this query in inline view?

SAMPLE

Table1
------
Value  Unit 
001    A1
002    B1
003    C2
002    A1

Table2
--------
Value  Unit  Number
001    B4        11
002    B1      1234
004    B1        22

TABLE3
-------
VALUE  UNIT  NUMBER  COLUMN1
001    B4        11      555
002    B1      1234      557
002    B1      1234      559

OUTPUT
------
MAX(C0LUMN1)
-----------
559

Solution

  • In your query there is no need for inlineview :- if that is rewritten in inlineview it will be like

    Select Max(Column1)
     From  (Select Value,Unit From Table1)B,
    (Select Value,Unit,Number From Table2)A,
     Table3 as H
     Where B.Unit=A.Unit
     And B.Value=A.Value
    AND B.unit=H.unit
    And B.Value=H.Value
    AND A.number=1234;
    

    Below is the example when to use inline view hope this help!!!

    An inline view is a SELECT statement in the FROM clause. As mentioned in the View section, a view is a virtual table that has the characteristics of a table yet does not hold any actual data. In an inline view construct, instead of specifying table name(s) after the FROM keyword, the source of the data actually comes from a view that is created within the SQL statement. The syntax for an inline view is,

    SELECT "column_name" FROM (Inline View);
    

    When should we use inline view? Below is an example:

    Assume we have two tables: The first table is User_Address, which maps each user to a ZIP code; the second table is User_Score, which records all the scores of each user. The question is, how to write a SQL query to find the number of users who scored higher than 200 for each ZIP code?

    Without using an inline view, we can accomplish this in two steps:

    Query 1

    CREATE TABLE User_Higher_Than_200
    SELECT User_ID, SUM(Score) FROM User_Score
    GROUP BY User_ID
    HAVING SUM(Score) > 200;
    

    Query 2

    SELECT a2.ZIP_CODE, COUNT(a1.User_ID)
    FROM User_Higher_Than_200 a1, User_Address a2
    WHERE a1.User_ID = a2.ZIP_CODE
    GROUP BY a2.ZIP_CODE;
    

    In the above code, we introduced a temporary table, User_Higher_Than_200, to store the list of users who scored higher than 200. User_Higher_Than_200 is then used to join to the User_Address table to get the final result.

    We can simplify the above SQL using the inline view construct as follows:
    

    Query 3

         SELECT a2.ZIP_CODE, COUNT(a1.User_ID)
            FROM
            (SELECT User_ID, SUM(Score) FROM 
            User_Score GROUP BY User_ID HAVING SUM(Score) > 200) a1,
            User_Address a2
            WHERE a1.User_ID = a2.ZIP_CODE
            GROUP BY a2.ZIP_CODE;
    

    There are two advantages on using inline view here:

    1. We do not need to create the temporary table. This prevents the database from having too many objects, which is a good thing as each additional object in the database costs resources to manage.

    2. We can use a single SQL query to accomplish what we want Notice that we treat the inline view exactly the same as we treat a table. Comparing Query 2 and Query 3, we see that the only difference is we replace the temporary table name in Query 2 with the inline view statement in Query 3. Everything else stays the same.

    Inline view is sometimes referred to as derived table. These two terms are used interchangeably.