I want Pivot multi column. What use oracle pivot table?
SQL:
SELECT * FROM
(
SELECT *
FROM IRO_SIM A
WHERE A.COM_CODE = 'AAQ'
AND A.PCODE = 'AKIOP'
)
PIVOT
(
LISTAGG(SIMTYPE,',')
WITHIN GROUP (ORDER BY SIMTYPE)
FOR SIMTYPE IN ('H','V')
)
Sample Data:
COM_CODE | PCODE | L_VALUE | A_SIM | AMT_SIM | SIMTYPE
A | AKIOP | 1700 | TOTAL | 50 | H
A | AKIOP | 500 | EACH | 100 | V
A | BHUIO | 200 | TOTAL | 500 | H
A | BHUIO | 600 | TOTAL | 400 | V
i need Result:
COM_CODE | PCODE | H_VALUE | H_ASIM | H_AMTSIM | V_VALUE | V_ASIM | V_AMTSIM
A | AKIOP | 1700 | TOTAL | 50 | 500 | EACH | 100
A | BHUIO | 200 | TOTAL | 500 | 600 | TOTAL | 400
thanks advance :)
Just list the multiple columns. Every expression in your PIVOT clause will be matched with every value in the FOR clause. So, what you want is this:
SELECT * FROM d
PIVOT ( sum(l_value) as value, max(a_sim) as asim, sum(amt_sim) as amtsim
FOR simtype in ('H' AS "H", 'V' AS "V") )
With data...
with d as (
SELECT 'A' com_code, 'AKIOP' pcode, 1700 l_value, 'TOTAL' a_sim, 50 amt_sim, 'H' simtype FROM DUAL UNION ALL
SELECT 'A' com_code, 'AKIOP' pcode, 500 l_value, 'EACH' a_sim, 100 amt_sim, 'V' simtype FROM DUAL UNION ALL
SELECT 'A' com_code, 'BHUIO' pcode, 200 l_value, 'TOTAL' a_sim, 500 amt_sim, 'H' simtype FROM DUAL UNION ALL
SELECT 'A' com_code, 'BHUIO' pcode, 600 l_value, 'TOTAL' a_sim, 400 amt_sim, 'V' simtype FROM DUAL)
SELECT * FROM d
PIVOT ( sum(l_value) as value, max(a_sim) as asim, sum(amt_sim) as amtsim
FOR simtype in ('H' AS "H", 'V' AS "V") )
Related
I have a 2 table like this:
first table
+------------+---------------+--------+
| pk | user_one |user_two|
+------------+---------------+--------+
second table
+------------+---------------+--------+----------------+----------------+
| pk | sender |receiver|fk of firsttable|content |
+------------+---------------+--------+----------------+----------------+
First and second table have one to many(1:N) relations.
There are many records in second table:
| pk | sender|receiver|fk of firsttable|content |
|120 |car224 |car223 |1 |test message1 to 223
|121 |car224 |car223 |1 |test message2 to 223
|122 |car224 |car225 |21 |test message1 to 225
|123 |car224 |car225 |21 |test message2 to 225
|124 |car224 |car225 |21 |test message3 to 225
|125 |car224 |car225 |21 |test message4 to 225
I need to find if fk has the same value and I want the row with the largest pk.
I've changed the above column name to make it easier to understand.
Here is the actual sql I've tried so far:
select *
from (select rownum rn,
mr.mrno,
mr.user_one,
mr.user_two,
m.mno,
m.content
from tbl_messagerelation mr,
tbl_message m
where (mr.user_one = 'car224' or
mr.user_two='car224') and
m.rowid in (select max(rowid)
from tbl_message
group by m.mno) and
rownum <= 1*20)
where rn > (1-1) * 20
And this is the result:
+---------+-------+----------+----------+-------------------------+----------------------+
| rn | mrno | user_one | user_two | mno(pk of second table) | content |
+---------+-------+----------+----------+-------------------------+----------------------+
| 1 | 1 | car224 | car223 | 125 | test message4 to 225 |
| 2 | 21 | car224 | car225 | 125 | test message4 to 225 |
+---------+-------+----------+----------+-------------------------+----------------------+
My desired result is something like this:
+---------+---------+----------+--------------------+----------------------+
| fk | sender | receiver | pk of second table | content |
+---------+---------+----------+--------------------+----------------------+
| 1 | car224 | car223 | 121 | test message2 to 223 |
| 21 | car224 | car223 | 125 | test message4 to 225 |
+---------+---------+----------+--------------------+----------------------+
Your table description when compared to your query is confusing me. However, what I could understand was that you are probably looking for row_number().
An important advice is to use standard explicit JOIN syntax rather than outdated a,b syntax for joins. Join keys were not clear to me and you may replace it appropriately in your final query.
select * from
(
select mr.*, m.*, row_number() over ( partition by m.fk order by m.pk desc ) as rn
from tbl_messagerelation mr join tbl_message m on mr.? = m.?
) where rn =1
Or perhaps you don't need that join at all
select * from
(
select m.*, row_number() over ( partition by m.fk order by m.pk desc ) as rn
from tbl_message m
) where rn =1
I have query with bind variables which comming from outer application.
The optimizer use the the unwanted index and I want to force it use another plan.
So I generate the good plan using index hint and then created the baseline with the plans
and connect the wanted plan to the query sql_id, and change the fixed attribute to 'YES'.
I executed the DBMS_XPLAN.DISPLAY_SQL_PLAN_BASELINE function
and the output shows that the wanted plan marked as fixed=yes.
So why when I'm running the query it still with the bad plan??
The code:
-- Query
SELECT DISTINCT t_01.puid
FROM PWORKSPACEOBJECT t_01 , PPOM_APPLICATION_OBJECT t_02
WHERE ( ( UPPER(t_01.pobject_type) IN ( UPPER( :1 ) , UPPER( :2 ) )
AND ( t_02.pcreation_date >= :3 ) ) AND ( t_01.puid = t_02.puid ) )
-- get the text
select sql_fulltext
from v$sqlarea
where sql_id = '21pts328r2nb7' and rownum = 1;
-- prepare the explain plan
explain plan for
SELECT DISTINCT t_01.puid
FROM PWORKSPACEOBJECT t_01 , PPOM_APPLICATION_OBJECT t_02
WHERE ( ( UPPER(t_01.pobject_type) IN ( UPPER( :1 ) , UPPER( :2 ) )
AND ( t_02.pcreation_date >= :3 ) ) AND ( t_01.puid = t_02.puid ) ) ;
-- we can see that there is no use of index - PIPIPWORKSPACEO_2
select * from table(dbms_xplan.display);
------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost |
------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 10382 | 517K| 61553 |
| 1 | HASH UNIQUE | | 10382 | 517K| 61553 |
| 2 | HASH JOIN | | 158K| 7885K| 61549 |
| 3 | INLIST ITERATOR | | | | |
| 4 | TABLE ACCESS BY INDEX ROWID| PWORKSPACEOBJECT | 158K| 4329K| 52689 |
| 5 | INDEX RANGE SCAN | PIPIPWORKSPACEO_3 | 158K| | 534 |
| 6 | INDEX RANGE SCAN | DBTAO_IX1_PPOM | 3402K| 74M| 2911 |
------------------------------------------------------------------------------------
Note
-----
- 'PLAN_TABLE' is old version
-- generate plan with the wanted index
explain plan for
select /*+ index(t_01 PIPIPWORKSPACEO_2)*/ distinct t_01.puid
from pworkspaceobject t_01 , ppom_application_object t_02
where ( ( upper(t_01.pobject_type) in ( upper( :1 ) , upper( :2 ) )
and ( t_02.pcreation_date >= :3 ) ) and ( t_01.puid = t_02.puid ) ) ;
-- the index working - the index used
select * from table(dbms_xplan.display);
-----------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost |
-----------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 10382 | 517K| 223K|
| 1 | HASH UNIQUE | | 10382 | 517K| 223K|
| 2 | HASH JOIN | | 158K| 7885K| 223K|
| 3 | TABLE ACCESS BY INDEX ROWID| PWORKSPACEOBJECT | 158K| 4329K| 214K|
| 4 | INDEX FULL SCAN | PIPIPWORKSPACEO_2 | 158K| | 162K|
| 5 | INDEX RANGE SCAN | DBTAO_IX1_PPOM | 3402K| 74M| 2911 |
-----------------------------------------------------------------------------------
Note
-----
- 'PLAN_TABLE' is old version
-- get the sql_id of the query with the good index
-- 7t72qvghr0zqh
select sql_id from v$sqlarea where sql_text like 'select /*+ index(t_01 PIPIPWORKSPACEO_2)%';
-- get the plan hash value of the good plan by the sql_id
--4040955653
select plan_hash_value from v$sql_plan where sql_id = '7t72qvghr0zqh';
-- get the plan hash value of the bad plan by the sql_id
--1044780890
select plan_hash_value from v$sql_plan where sql_id = '21pts328r2nb7';
-- load the source plan
begin
dbms_output.put_line(
dbms_spm.load_plans_from_cursor_cache
( sql_id => '21pts328r2nb7' )
);
END;
-- the new base line created with the bad plan
select * from dba_sql_plan_baselines;
-- load the good plan of the second sql_id (with the wanted index)
-- and bind it to the sql_handle of the source query
begin
dbms_output.put_line(
DBMS_SPM.LOAD_PLANS_FROM_CURSOR_CACHE
( sql_id => '7t72qvghr0zqh',
plan_hash_value => 4040955653,
sql_handle => 'SQL_4afac4211aa3317d' )
);
end;
-- new there are 2 plans bind to the same sql_handle and sql_text
select * from dba_sql_plan_baselines;
-- alter the good one to be fixed
begin
dbms_output.put_line(
dbms_spm.alter_sql_plan_baseline
( sql_handle =>
'SQL_4afac4211aa3317d',
PLAN_NAME => 'SQL_PLAN_4pyq444da6cbxf7c97cc7',
ATTRIBUTE_NAME => 'fixed',
ATTRIBUTE_VALUE => 'YES'
)) ;
end;
-- check the good plan - fixed = yes
select * from table(
dbms_xplan.display_sql_plan_baseline (
sql_handle => 'SQL_4afac4211aa3317d',
plan_name => 'SQL_PLAN_4pyq444da6cbxf7c97cc7',
format => 'ALL'));
--------------------------------------------------------------------------------
SQL handle: SQL_4afac4211aa3317d
SQL text: SELECT DISTINCT t_01.puid FROM PWORKSPACEOBJECT t_01 ,
PPOM_APPLICATION_OBJECT t_02 WHERE ( ( UPPER(t_01.pobject_type) IN (
UPPER( :1 ) , UPPER( :2 ) ) AND ( t_02.pcreation_date >= :3 ) ) AND (
t_01.puid = t_02.puid ) )
--------------------------------------------------------------------------------
--------------------------------------------------------------------------------
Plan name: SQL_PLAN_4pyq444da6cbxf7c97cc7 Plan id: 4157177031
Enabled: YES Fixed: YES Accepted: YES Origin: MANUAL-LOAD
--------------------------------------------------------------------------------
Plan hash value: 4040955653
-----------------------------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes |TempSpc| Cost (%CPU)| Time |
-----------------------------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 10382 | 517K| | 223K (1)| 00:44:37 |
| 1 | HASH UNIQUE | | 10382 | 517K| | 223K (1)| 00:44:37 |
|* 2 | HASH JOIN | | 158K| 7885K| 6192K| 223K (1)| 00:44:37 |
| 3 | TABLE ACCESS BY INDEX ROWID| PWORKSPACEOBJECT | 158K| 4329K| | 214K (1)| 00:42:50 |
|* 4 | INDEX FULL SCAN | PIPIPWORKSPACEO_2 | 158K| | | 162K (1)| 00:32:25 |
|* 5 | INDEX RANGE SCAN | DBTAO_IX1_PPOM | 3402K| 74M| | 2911 (1)| 00:00:35 |
-----------------------------------------------------------------------------------------------------------
Query Block Name / Object Alias (identified by operation id):
-------------------------------------------------------------
1 - SEL$1
3 - SEL$1 / T_01#SEL$1
4 - SEL$1 / T_01#SEL$1
5 - SEL$1 / T_02#SEL$1
Predicate Information (identified by operation id):
---------------------------------------------------
2 - access("T_01"."PUID"="T_02"."PUID")
4 - filter(UPPER("POBJECT_TYPE")=UPPER(:1) OR UPPER("POBJECT_TYPE")=UPPER(:2))
5 - access("T_02"."PCREATION_DATE">=:3)
Column Projection Information (identified by operation id):
-----------------------------------------------------------
1 - (#keys=1) "T_01"."PUID"[VARCHAR2,15]
2 - (#keys=1) "T_01"."PUID"[VARCHAR2,15]
3 - "T_01"."PUID"[VARCHAR2,15]
4 - "T_01".ROWID[ROWID,10]
5 - "T_02"."PUID"[VARCHAR2,15]
Note
-----
- 'PLAN_TABLE' is old version
-- run explain plan for the query
-- need to use the new plan
declare
v_string clob;
begin
select sql_fulltext
into v_string
from v$sqlarea
where sql_id = '21pts328r2nb7' and rownum = 1;
execute immediate 'explain plan for ' || v_string using '1','1',sysdate;
end;
-- check the plan - still the unwanted index and plan
select * from table(dbms_xplan.display);
------------------------------------------------------------------------------------
| Id | Operation | Name | Rows | Bytes | Cost |
------------------------------------------------------------------------------------
| 0 | SELECT STATEMENT | | 10382 | 517K| 61553 |
| 1 | HASH UNIQUE | | 10382 | 517K| 61553 |
| 2 | HASH JOIN | | 158K| 7885K| 61549 |
| 3 | INLIST ITERATOR | | | | |
| 4 | TABLE ACCESS BY INDEX ROWID| PWORKSPACEOBJECT | 158K| 4329K| 52689 |
| 5 | INDEX RANGE SCAN | PIPIPWORKSPACEO_3 | 158K| | 534 |
| 6 | INDEX RANGE SCAN | DBTAO_IX1_PPOM | 3402K| 74M| 2911 |
------------------------------------------------------------------------------------
Note
-----
- 'PLAN_TABLE' is old version
From a read through of your test case, I suspect the problem is that you're interpreting the FIXED attribute incorrectly.
If you list all the plans for your baseline, you will probably find the original and the loaded cursor plan are both ENABLED and ACCEPTED at the moment. I think what you need to do (based on my own usage of these calls) is use the ENABLED attribute. Set ENABLED to NO for the unwanted plan.
Try:
exec dbms_spm.alter_sql_plan_baseline(
sql_handle=>'SQL_...' -- baseline to update
,plan_name=>'SQL_PLAN_...' -- unwanted plan signature to disable
,attribute_name=>'ENABLED',attribute_value=>'NO')
I want to fetch records from oracle DB where column value is NULL. Also i am using unpivot in my query. Somehow NULL values are not getting selected because of unpivot keyword. Can you please help me about how to get rows for the same when using unpivot.
EDIT:
SELECT a.emp_id, a.emp_dept, b.emp_location
FROM employee a,
location b UNPIVOT (emp_id
FOR dummy_id
IN (emp_id AS 'EMP_ID', last_date AS 'LAST_DATE'))
WHERE emp_id = 123 AND b.emp_loc_id = 'india' AND b.location IS NULL;
Use UNPIVOT INCLUDE NULLS:
SQL Fiddle
Oracle 11g R2 Schema Setup:
CREATE TABLE test ( id, a, b, c, d ) AS
SELECT 1, 1, 2, 3, 4 FROM DUAL UNION ALL
SELECT 2, 1, NULL, 3, NULL FROM DUAL;
Query 1:
SELECT *
FROM test
UNPIVOT INCLUDE NULLS ( value FOR name IN ( a, b, c, d ) )
Results:
| ID | NAME | VALUE |
|----|------|--------|
| 1 | A | 1 |
| 1 | B | 2 |
| 1 | C | 3 |
| 1 | D | 4 |
| 2 | A | 1 |
| 2 | B | (null) |
| 2 | C | 3 |
| 2 | D | (null) |
I have a table structure (that I did not design nor can I change) that uses a varchar field to store an attribute about the entity. I would like to write a SQL query to search for two attributes in particular and combine multiple result rows into single rows. To illustrate, my tables are similar to this:
company
=============
| id | name |
-------------
| 1 | co1 |
| 2 | co2 |
| 3 | co3 |
=============
agent
====================================
| id | name | company_id | type |
------------------------------------
| 1 | Tom | 1 | 'type1' |
| 2 | Bob | 1 | 'type2' |
| 3 | Bill | 2 | 'type1' |
| 4 | Jack | 2 | 'type2' |
| 5 | John | 3 | 'type1' |
| 6 | Joe | 3 | 'type2' |
====================================
type1 and type2 are hard-coded into the software as valid values (again, I didn't write it), so a search for these values should be successful (null is permitted). So, I must base my search off of these values.
As a novice, I could write this SQL:
select c.name, a.name, a.type
from company c
inner join agent a on c.id = a.company_id
and sort through these results in my software (Java program):
===========================
| c.name | a.name | type |
---------------------------
| co1 | Tom | type1 |
| co1 | Bob | type2 |
| co2 | Bill | type1 |
| co2 | Jack | type2 |
| co3 | John | type1 |
| co3 | Joe | type2 |
===========================
But, I was hoping there would be a way to combine the rows into something more efficient:
-- my failed attempt at writing this query
select c.name, a.name as type_1_agent, a.name as type_2_agent
from company c
inner join agent a on c.id = a.company_id
group by c.id -- ?
where -- ?
results:
======================================
| name | type_1_agent | type_2_agent |
--------------------------------------
| co1 | Tom | Bob |
| co2 | Bill | Jack |
| co3 | John | Joe |
======================================
Is this possible?
Oracle version:
WITH company (id, name) AS (
SELECT 1, 'co1' FROM DUAL UNION ALL
SELECT 2, 'co2' FROM DUAL UNION ALL
SELECT 3, 'co3' FROM DUAL
),
agent (id, name, company_id, type) AS (
SELECT 1, 'Tom', 1, 'type1' FROM DUAL UNION ALL
SELECT 2, 'Bob', 1, 'type2' FROM DUAL UNION ALL
SELECT 3, 'Bill', 2, 'type1' FROM DUAL UNION ALL
SELECT 4, 'Jack', 2, 'type2' FROM DUAL UNION ALL
SELECT 5, 'John', 3, 'type1' FROM DUAL UNION ALL
SELECT 6, 'Joe', 3, 'type2' FROM DUAL
)
SELECT
company_name, type_1_agent, type_2_agent
FROM
(SELECT company.name company_name, agent.name agent_name, type FROM company JOIN agent ON company.id = agent.company_id)
PIVOT (
MAX(agent_name) agent
FOR type IN ('type1' type_1, 'type2' type_2)
)
ORDER BY
company_name
You can do this with PIVOT functionality, like so:
select company,type_1_agent,type_2_agent
from
(select a.name as agentname, a.type as agenttype,c.name as company
from agent a
inner join company c on c.id = a.company_id
) s
pivot
(max(agentname) for agenttype in ('type1' type_1_agent,'type2' type_2_agent)) p
order by company
Demo
Of course, in this case we hard coded the values for type. This can be made dynamic to accomodate an unknown number of these values.
I need to add a column to a table that check for input to be a max value of 999 to 999, like a soccer match score. How do I write this statement?
example:
| Score |
---------
| 1-2 |
| 10-1 |
|999-999|
| 99-99 |
SQL Fiddle
Oracle 11g R2 Schema Setup:
CREATE TABLE SCORES (Score ) AS
SELECT '1-2' FROM DUAL
UNION ALL SELECT '10-1' FROM DUAL
UNION ALL SELECT '999-999' FROM DUAL
UNION ALL SELECT '99-99' FROM DUAL
UNION ALL SELECT '1000-1000' FROM DUAL;
Query 1:
SELECT SCORE,
CASE WHEN REGEXP_LIKE( SCORE, '^\d{1,3}-\d{1,3}$' )
THEN 'Valid'
ELSE 'Invalid'
END AS Validity
FROM SCORES
Results:
| SCORE | VALIDITY |
|-----------|----------|
| 1-2 | Valid |
| 10-1 | Valid |
| 999-999 | Valid |
| 99-99 | Valid |
| 1000-1000 | Invalid |