SQL & PL/SQL :: Query To Return Not Null Column Values With Priority
Dec 22, 2011
I have a table with multiple rows for the KEY attribute(its not a primary key) and a Rank for each row.
I want a query which fetches one row per KEY attribute.The row with lesser Rank should be considered. But in-case if the value is null for any column the value for next Rank should be considered.
WITH TMP_TBL AS
(
SELECT * FROM (
SELECT 'A' DUN,'1' RNK,'A21' col1,NULL col2,'A41' col3,NULL col4 FROM dual
UNION ALL
SELECT 'A','2','A122','A23',NULL,NULL FROM dual
UNION ALL
SELECT 'A','3','A32','A33',NULL,'A35' FROM dual
[code].......
DUN is the KEY attribute . RNK is the Rank for each Row. COL1... COL4 are data attributes
I want this to be done with SQL only. So I tried various ways but none were successful.Finally I created a Multi Row function row_nvl and it worked.
SELECT DUN,
row_nvl(rownvl_param_type(RNK,col1)),
row_nvl(rownvl_param_type(RNK,col2)),
row_nvl(rownvl_param_type(RNK,col3)),
row_nvl(rownvl_param_type(RNK,col4))
FROM TMP_TBL
GROUP BY DUN
But I don't think my manager will allow me to deploy a Multi Row function .
I dont want to print the repeated value(NAME) of C1 multiple times as below.
C1C2C3C4 NAMEJOHN10ABC SMITH30DEF ROBERT60XYZ
I could do it using the below query using union with the rownum.
select * from ( select rownum rn, c1,c2,c3,c4 from table_new ) where rn =1 union select * from ( select rownum rn, decode(c1,null,null),c2,c3,c4 from table_new ) where rn between 2 and 3
Is there any other way of displaying using a single sql query.
Our organization is attempting to learn more about the partitioning features of Oracle 11g. I've been reading the partitioning manuals, and I have not found a clear answer on this topic, but I suspect I know the answer.
If you create a range partitioned table; using interval partitioning, say something like this:
CREATE table range_parti ( CODE NUMBER(5), DESCRIPTION VARCHAR2(50), CREATED_DATE DATE) PARTITION BY RANGE (created_date) INTERVAL (NUMTOYMINTERVAL(1,'MONTH')) ( PARTITION my_parti VALUES LESS THAN (TO_DATE('01-NOV-2007','DD-MON-YYYY')) );
but you try to insert a null value as the partition key, you get the following error:
SQL> INSERT INTO range_parti VALUES (1,'one',NULL); INSERT INTO range_parti VALUES (1,'one',NULL) * ERROR at line 1: ORA-14400: inserted partition key does not map to any partition Elapsed: 00:00:00.07
Is there no way to tell it to use a default partition for NULL values? Or specifically designate a partition for NULL values WITHOUT having to manually list out each partition? It seems it works if you don't use the INTERVAL keyword, list out your partitions, and use MAXVALUE. However, our hope to avoid having that as it creates monstrously huge DDL statements for tables that have lots of date ranges, and we will be forced to manually add new partitions each month as data is added/time passes.
It appears from my experience so far, if your column can allow nulls, you cannot use interval range partitioning on that column.
previously i set null constraint to the column and creating some rows and need to change new entering values as not null constraint to the column in oracle without disturbing the old records. how can I do that.
I need a generic query to generate total # of records for each table in a schema, total # of records that are not null for each column in the table, and total # of records that are null for each of those columns in those tables.
ex:
the output should look like this.
owner schema table_name total# recs in the table, column_name, ------ ------ ---------- ------------------------- -----------
# of records not null # of records null ---------------------- --------------------
Attached query is running fine if inline view B (Marked in Comments) returning value. If inline view B returns NULL then it fails to return result. I want if Inline view B is returning NULL then it should pass ZERO to main query.
As I am doing a outer join, if the resource is not available on a particular day the resource_id is coming as NULL as it is not available. Is there any way to populate this NULL resource_id with the original resource_id as the resource_id is same for all the result set.
I have a table and data as below.I need to select all the records if value have both 'M' and 'D'.If there is only 'M' or 'D' then select should not pull any records.
WITH data as ( Select '1' id, 'M' value from dual union all Select '1' id, 'M' value from dual union all Select '1' id, 'D' value from dual union all Select '1' id, 'D' value from dual ) select value from data group by value
I tried below query but it is not working.
WITH data as ( Select '1' id, 'M' value from dual union all Select '1' id, 'M' value from dual union all Select '1' id, 'D' value from dual union all Select '1' id, 'D' value from dual ) select * from data group by value having sum(count(distinct(value))) > 1
SQL> Describe Stu_Table Name Null? Type ----------------------------------------- -------- ---------------------------- STU_ID VARCHAR2(2) STU_NAME VARCHAR2(10) STU_CLASS VARCHAR2(10)
now when i try to modify this Stu_id column to not null its give me error.
SQL>ALTER TABLE Stu_Table MODIFY Stu_Id int(3)not null; ALTER TABLE Stu_Table MODIFY Stu_Id int(3)not null * ERROR at line 1: ORA-01735: invalid ALTER TABLE option
and when i try to add new column with not null its also gives me error
SQL> ALTER TABLE Stu_Table add C1_TEMP integer NOT NULL; ALTER TABLE Stu_Table add C1_TEMP integer NOT NULL * ERROR at line 1: ORA-01758: table must be empty to add mandatory (NOT NULL) column
the following procedure successfully compiled and executed. How to access the values returning from this procedure in another procedure or PLSQL block.
CREATE OR REPLACE PROCEDURE test IS
cursor c1 is SELECT a.idnumber, a.idp FROM holdings a; r1 c1%rowtype;
BEGIN
open c1; loop fetch c1 into r1; exit when c1%notfound; DBMS_OUTPUT.put_line (r1.idnumber||'--'||r1.idp); end loop; close c1; END;
I am trying write a script that will return all values (based on the minimum tarif) from the Germany table for any duplicate values. Duplicate values are any values with the same UFI, ZC,limitid,depot. The German table also contains the fields tarif, city, supplier, etc.
Below is the script I have previously used to sort out duplicates. I have tried 50 different ways get it to return just lines for the minimum tariff but haven't been successful.
select * from Germany t where (ufi,zc,limitid,depot) in ( select ufi,zc,limitid,depot from ( select ufi,zc,limitid,depot, count(*) n from Germany t group by ufi,zc,limitid,depot) where n<>1 )
I have a stored procedure which has varchar2 as IN and sys_recursor has OUT parameters.
CREATE OR REPLACE PROCEDURE check_values ( my_values IN emp.dept_no%TYPE, p_cursor OUT sys_refcursor )
[Code]....
/The problem I am facing is in where condition, if I give quoteValues it doesn't fetch me any records when I execute the procedure from sqlplus, but if I am giving my_values it does fetch me records. I am receiving IN parameters like 9856,9712,8723, so first I put single quote around the emp_no and pass that to where condition.
I have a table second_table which has a username and code field. A username (not the primary key) may be entered multiple times with different codes, or a single code 'ALL', in which case, the codes have to be fetched from 'third_table'. I am unable to form a 'case' clause to handle the fact that a list has to returned when the 'code' field from the second_table returns 'ALL'.
e.g.
Second_table username code A ALL B 23 B 56 B 33
Third_Table code 67 78
So, when the user asks the codes for user A, he should get 67 and 78 and when he asks for the user B, he should get 23,56 and 33
I am using functions to return multiple values of two rows or multiple rows.
For example emp id = 100 and i need to return the value for this(empid) input and output of this first_name and salary.
I am tried in this way below but got errors (ORA-00932: inconsistent datatypes: expected NUMBER got HR.EMP_TYPE)
create or replace type emp_type as object(first_name varchar2(20),salary number);
create or replace function f1(empid in number) return emp_type as emp_record emp_type; begin select first_name,salary into emp_record.first_name,emp_record.salary from employees where employee_id = empid ; return emp_record; end; /
I have a PL/SQL function where i call java class. In java i`m making some calculations and i need to return to body of PL/SQL function some values ( like 6 to 10, few numbers, string, 1 xml and 1 html).
I will show pl/sql function..
CREATE OR REPLACE FUNCTION GET_TAB(P_CURS SYS_REFCURSOR) RETURN TYP1_TAB PIPELINED IS OUT_REC TYP1:=TYP1(NULL,NULL,NULL,NULL); V_NAZWA VARCHAR2(5); V_NUMER NUMBER;
I have 2 tables, AFF_TEMP and COUNTY AFF_TEMP has the following columns FNAME, LNAME, EMAIL and COUNTY COUNTY has 2 columns COUNTY_ID and CNAME
Both tables have the following test data AFF_TEMP Joe, Bloggs, joe@gmail.com, '' Ann, Bloggs, anne@gmail.com,Donegal
and COUNTY column in AFF_TEMP can contain a NULL value
County table has the following Test data, 1, Dublin 2, Donegal 3, Tipperary, 4, Galway
I am trying to select the following from both tables FNAME, LNAME, EMAIL, COUNTY_ID.Tried the following queries select a.FNAME, a.LNAME,a.EMAIL, C.COUNTY_ID FROM temp_aff A LEFT OUTER JOIN COUNTY C ON A.COUNTY=C.CNAME OR (A.COUNTY IS NULL) select a.FNAME, a.LNAME,a.EMAIL, C.COUNTY_ID FROM temp_aff A, COUNTY C WHERE C.CNAME IN (SELECT UPPER(A.COUNTY) FROM TEMP_AFF A)
We are working on a migration project and we need to move 75 million rows from source system to target system.
Total number of columns in source system - 90 cols.
Out of the 90 columns 10 cols are system fields and rest 80 are properties for each record.
We are required to migrate all system cols and some required properties. In total we will migrate around 25 columns[10+15] for each record.
Before actaul migration , we need to do a data cleansing activity and hence we move the data to a staging table.
To create the staging table, we considered the below appraoches.
1. Create the staging table with around 30 coloumns so as to fit the data from source system[map the columns based on datatype]
2.Create the staging table with actual columns[90 columns] and import only the required properties. The rest all columns will remain NULL.
Do the data cleansing and move to target system.
My question here is, if we go with approach 2, We will not mix the data, as there will be a one-to-one mapping. But many columns will not have data and remain NULL. Will it affect the performance since we deal with 75 million rows.
I have main query..And i written a Query for Issued Quantity. The query is retrieving exact data. And when i add the Query in main query the data is coming with extra null values.
for ex. Query for Issued Qty fetch 16 records. and i run the main query it fetch 20 records...there r 4 records which issued Qty is null. How can i restrict that records.
FUNCTION normalize(str IN VARCHAR2) RETURN VARCHAR2 IS BEGIN RETURN TRANSLATE(LOWER(str), 'äàáâãăāåąæčçðďéèëêěĕėęğģġîĭïīìíłļľŀñńňņöóòôõσøőřśŝšşţüúùûǔųūůŵýÿżźžżαβßγδεζŋηικλμµνξπρσςτυφω', 'aaaaaaaaaaccddeeeeeeeegggiiiiiillllnnnnoooooooorsssstuuuuuuuuwyyzzzzassydeznniklmmnxprsstufo' ); END;
I'm tired to add missing characters in this list...
What I would like is that all characters with an accent or diacritics should be replace by their "base" letter (ë -> e) and Greek letters to be replace by their corresponding values (ω (omega) -> o)
