SQL & PL/SQL :: Distinct With Analytical Query And IN Statement?
Nov 18, 2011
We've got a query which returns one row, but uses an IN statement. The IN statement links to more than one row in the subquery. When we use a combination of DISTINCT and an ANALYTICAL sum, the sum total is multiplied by the number of rows in the sub query. Remove the DISTINCT and we get a single value.
A simplified example of the problem is below.
I can't see how a query which returns a single row then returns multiple values with the addition of a DISTINCT. Removing the analytical sum also provides a single row, but we need this in the actual query we're running. So it seems to be some combination of DISTINCT, ANALYTICAL SUM and IN query is causing multiple values to be returned.
CREATE TABLE go_test_distinct1
(gtd_value NUMBER);
-- Three identical values
-- To replicate the three identical values returned by
I need to calcaulate the salary avarage for three days prior, leaving the current row. That should happen to every row moving back words.I have given all the details.
create table Employee( ID VARCHAR2(4 BYTE) NOT NULL, name varchar(20), Start_Date DATE, Salary Number(8,2), mv_avg number(8,2) [code]....
My table has two date columns EFF_DT which is the start date and TERM_DT is the end date. The EFF_DT of the next record should be the next date of the TERM_DT record.
In the fourth record, the effective date should be 1-Oct-13 which is the next date to the last TERM_DT 30-Sep-13.As the is the break in the date, the output should show 15-Oct-13 sa the second start date.
Note: Refer to the PI_ID columns, there is a break in the date for the sale PI_ID 'ABC'.
Here I am trying to generate a pseudo column, so that the table with the pseudo column looks like as shown below. and I can use first_value and LAST_value by partitioning on the pseudo column to get the desired output.
1) CNT_VAL is the pseudo column: ----------------------------- CK_IDPI_IDEFF_DT TERM_DT CNT_VAL Mem1ABC1-Jan-1331-Mar-131 Mem1ABC1-Apr-1331-May-131 Mem1ABC1-Jun-1330-Sep-131
[code].....
My Query : ----------
I not getting the desired output here as the value in pseudo column is 3.
select CK_ID, PI_ID,EFF_DT,TERM_DT, (case when case_CONT - LAG(case_CONT,1) over (ORDER BY EFF_DT) = 0 then to_char(case_CONT) when case_CONT - LAG(case_CONT,1) over (ORDER BY EFF_DT) <> 0 then to_char(LAG(case_CONT,1) over (ORDER BY EFF_DT) + 1) else to_char(nvl(case_CONT,0))
I am having a table with 5 lakhs transactions. I want to fetch the last balance for a particular date. So i have have returned a query like below.
SELECT curr_balance FROM transaction_details WHERE acct_num = '10'
[Code]...
This has to be executed for incrementing of 12 months to find the last balance for each particular month. But this query is having more cpu cost, 12 times it is taking huge time. how to remodify athe above query to get the results in faster way using analytical query. Whether this can be broken into two part in PL/SQL to achive the performance. ?
I'm posting below test case in which I'm not able to understand output for LAST_VALUE function. I'm expecting maximum value for the salary in a department. Because I'm partitioning by department and ordering a partition as assending so being last value it should give me maximum value within a partition i.e. department in this case.
I was reading a tutorial for analytical function and i found something like this
sum(princial) keep(dense_rank first order by d_date) over partition by (userid, alias, sec_id, flow, p_date)
How to translate this into simple queries / subquery? i am aware that analytical function are faster but i would like to know how this can translate to using query without analytical function.
Is there any way to apply the restriction on analytical functions, just like WHERE and HAVING .AS we know that we can apply the restriction on table by using WHERE and grouping functions by using HAVING clause .
For Ex: Departments wise count including all employees record :
SQL> select count(*) over(partition by deptno) dept_Count, ce.* 2 from scott.emp ce 3 order by deptno, job;
DEPT_COUNT EMPNO ENAME JOB MGR HIREDATE SAL COMM DEPTNO ---------- ----- ---------- --------- ----- ----------- --------- --------- ------ 3 7934 MILLER CLERK 7782 1/23/1982 1300.00 10 3 7782 CLARK MANAGER 7839 6/9/1981 2450.00 10 3 7839 KING PRESIDENT 11/17/1981 5000.00 10 5 7788 SCOTT ANALYST 7566 4/19/1987 3000.00 20 [code]....
CREATE TABLE TEST1 (AGG_DATE DATE, COL1 NUMBER(9), COL2 NUMBER(9), COL3 NUMBER(9)); Here is the test-data population script: insert into TEST1 (AGG_DATE, COL1, COL2, COL3) values (to_date('01-01-2012', 'dd-mm-yyyy'), 1, 1, 1); [code]....
The problem is when I wrote an analytical query, it is giving the BEGIN_DATE and END_DATE by taking all the partition values together and so instead of the values above, it is creating an answer as follows:
Now i am trying to get the max(Value_2011) keep (dense_rank Last order by Month_ID) but i get a NULL. I can understand its because the Month_ID accomodates all years but i only need it to look at Month_ID for 2011 and return me the last dense_rank value, how can i achieve this?
I tried a couple of different methods like Last_Value() but i have group by in my original statement and i think analytical functions dont like GROUP by if they are not part of it. How can i achieve this?
I need to calculate the sum of values over a period of exactly one month (including the current row). Now if I use a windowing clause of "range between interval '1' month preceding and current row", the total period length is 1 month plus one day (being the day in the current record).
Basically, I want to sum over a period starting at "add_months(startdate, -1) + 1" up until startdate of each row.
drop table window_tst; create table window_tst ( id number primary key
[Code]....
So instead of having 01-feb going back to 01-jan, it should only include 02-jan till 01-feb
I could of course recalculate the period length back to a number of days for each row, but that is not really what I would prefer, as it would make the code rather unreadable.
I want to use Analytical function instead of group by clause for below query..
select CASE WHEN ADMT.SOURCESYSTEM ='CLU' THEN COUNT(ADMT.TOTAL_COUNT)*5 ELSE COUNT(ADMT.TOTAL_COUNT) END TOTAL_COUNT from ESMARTABC.ABC_DRVR_MFAILS_TMP ADMT group by ADMT.SOURCESYSTEM
how to write this query using all three tables in one query statement.
Result:
Input value of "GM" would return "GM@EMAIL.COM"
Input value of "GM" and "GMC" would return "GMC@EMAIL.COM"
If "GM" is supplied, then only return the "GM@EMAIL.COM" email.
If "GM" and "GMC" is supplied, then only return the "GMC@EMAIL.COM" email
Test Case : GM would return "GM@EMAIL.COM" email
Test Case : GM and GMC would return "GMC@EMAIL.COM" email
select email from email a, company b, sub_company c where (company = "GM" but no match found in sub_company) (company = "GM" and sub_company_name = "GMC" match is found in sub_company)
#select name from v$database; #select log_mode from v$database; #select count(*)"INVALID_OBJECTS" from dba_objects where status='INVALID'; #select count(*) "INVALID_N/A_INDEXES" from dba_indexes where status!='VALID'; #select count(*)"Invalid Triggers" from user_objects where OBJECT_NAME like '%TRIGGERS%' and status='VALID'; #select count(*) "Broken Jobs" from dba_jobs where broken!='Y'; #select count(*) "Block Corruption" from v$database_block_corruption;
i want a table which can be generated just by select cmd and it will list the result of all the above queires as follow:-
I mean to say i want multiple select queries into 1 table (note:- i m not saying to create a tables and then insert,update(using select from other tables), its just a sheel script that will fetch these record into a txt file)
The "HAVING" statement from the query below removes all rows where the "name" field starts with "Port" and it removes all rows where count is less than 11. What I want is to removes all rows where the "name" field starts with "Port" and the count for that row is less than 11.
SELECT deviceEventClassId "Event ID", deviceAddress "Device Address", name "Event Name", deviceHostName "Device Host Name", count( concat(deviceEventClassId, deviceHostName, name)) "Count" FROM events WHERE (deviceVendor = "Bilbo" and
I've seen this example numerous places, and tried to implement it, but I keep getting an "invalid identifier" error message, despite the fact that I've got the table and column specifically identified.For instance, my query reads like:
UPDATE tbl1 SET tbl1.EMPID = (SELECT tbl2.EMPIDA FROM tbl2 WHERE LOWER(tbl1.EMAILCOL) = LOWER(tbl2.EMAILCOL2) ) WHERE tbl2.EMPIDA IN ('Z1O435','S8M4722','M0D5156') AND EXISTS (SELECT tbl2.EMPIDA FROM tbl2 WHERE tbl1.EMAILCOL= tbl2.EMAILCOL2 );
But I'll keep getting flagged at the tbl2.EMPIDA column reference. I have not tried this in SQL Plus, just in TOAD, but it seems to repeatedly fail.I have had to dump records to standalone Access tables and link back to perform the updates.
how to delete duplicated records from a table without using row_id. I found the duplicated rows from a table using Analytical Function. But i could not use the Analytical function in the where condition.
I'm using pivot query feature of oracle 11g and came across a strange situation where i need to pass a "select statement" in a "in clause" of pivot query.
I have tried with pivot xml but it not giving desired output in sql*plus session.It is giving unreadable output.
select * from (select uin,testing_id,pfa_result from test1) pivot xml (max(pfa_result) as result for (testing_id) in (select distinct testing_id from test1));
[code]....
Here actually i want to use "select distinct id from test1" instead of "in (11,12,13,14,15)". Because i don't know how many id's will be there and of which values. e.g. 11 or 21 or 25.
DECLARE @MainTable TABLE (UniqueID INTEGER, Category VARCHAR(200), WeekDate DATETIME, VALUE INTEGER) INSERT INTO @MainTable VALUES(123, 'Shirts', '10/07/2011', 5000) INSERT INTO @MainTable VALUES(123, 'Shirts', '10/14/2011', 8000) INSERT INTO @MainTable VALUES(124, 'Pants', '10/07/2011', 4000) INSERT INTO @MainTable VALUES(125, 'Shorts', '10/14/2011', 8000) INSERT INTO @MainTable VALUES(126, 'Shoes', '10/21/2011', 9000); --select * from @MainTable; [code]...
The query works with all the CTEs up to the last select statement. Oracle does not support the OUTER APPLY statement, how should the last piece be written to make it work in Oracle?
