问题
I have a table and in this table i have data is not properly loaded data integrity issue ,since this is a dimension table we need to maintain the effective_dt_from and effective_dt_to and version correctly ,below is the table and sample data
create table TEST (
LOC_SID NUMBER(38,0),
CITY VARCHAR2(180 BYTE),
POSTAL_CD VARCHAR2(15 BYTE),
EFFECTIVE_DT_FROM DATE,
EFFECTIVE_DT_TO DATE,
VERSION NUMBER(38,0)
);
Insert into TEST values (25101,'Assam',1153,to_date('01.01.00 00:00:00','DD.MM.YY HH24:MI:SS'),to_date('01.01.17 23:59:59','DD.MM.YY HH24:MI:SS'),1);
Insert into TEST values (25102,'Assam',1153,to_date('01.01.00 00:00:00','DD.MM.YY HH24:MI:SS'),to_date('31.12.99 23:59:59','DD.MM.YY HH24:MI:SS'),1);
Insert into TEST values (25103,'Assam',1290,to_date('01.01.00 00:00:00','DD.MM.YY HH24:MI:SS'),to_date('01.01.17 23:59:59','DD.MM.YY HH24:MI:SS'),1);
Insert into TEST values (25104,'Assam',1290,to_date('01.01.00 00:00:00','DD.MM.YY HH24:MI:SS'),to_date('31.12.99 23:59:59','DD.MM.YY HH24:MI:SS'),1);
Insert into TEST values (25105,'Assam',1310,to_date('01.01.00 00:00:00','DD.MM.YY HH24:MI:SS'),to_date('01.01.17 23:59:59','DD.MM.YY HH24:MI:SS'),1);
Insert into TEST values (25106,'Assam',1310,to_date('01.01.00 00:00:00','DD.MM.YY HH24:MI:SS'),to_date('31.12.99 23:59:59','DD.MM.YY HH24:MI:SS'),1);
Insert into TEST values (25107,'Assam',1781,to_date('01.01.00 00:00:00','DD.MM.YY HH24:MI:SS'),to_date('01.01.17 23:59:59','DD.MM.YY HH24:MI:SS'),1);
Insert into TEST values (25108,'Assam',1781,to_date('01.01.00 00:00:00','DD.MM.YY HH24:MI:SS'),to_date('31.12.99 23:59:59','DD.MM.YY HH24:MI:SS'),1);
Insert into TEST values (25109,'Assam',1982,to_date('01.01.00 00:00:00','DD.MM.YY HH24:MI:SS'),to_date('01.01.17 23:59:59','DD.MM.YY HH24:MI:SS'),1);
Insert into TEST values (25110,'Assam',1982,to_date('01.01.00 00:00:00','DD.MM.YY HH24:MI:SS'),to_date('31.12.99 23:59:59','DD.MM.YY HH24:MI:SS'),1);
Data integrity check query
SELECT count(*) AS RowAffected
FROM
(SELECT LOC_SID,
VERSION,
EFFECTIVE_DT_FROM,
EFFECTIVE_DT_TO,
CITY,
POSTAL_CD
FROM
(SELECT t.*,
LEAD(EFFECTIVE_DT_FROM, 1) OVER(PARTITION BY CITY, POSTAL_CD
ORDER BY EFFECTIVE_DT_FROM) AS next_date,
LEAD(VERSION, 1) OVER(PARTITION BY CTY, POSTAL_CD
ORDER BY EFFECTIVE_DT_FROM) AS next_version
FROM TEST t)
WHERE valid_to != next_date
OR VERSION = next_version)
existing data set
LOC_SID CITY POSTAL_CD EFFECTIVE_DT_FROM EFFECTIVE_DT_TO VERSION
25101 ASSAM 1153 01.01.1900 31.12.2199 1
25102 ASSAM 1153 01.01.1900 31.12.2199 1
25103 ASSAM 1290 01.01.1900 31.12.2199 1
25104 ASSAM 1290 01.01.1900 31.12.2199 1
25105 ASSAM 1310 01.01.1900 31.12.2199 1
25106 ASSAM 1310 01.01.1900 31.12.2199 1
25107 ASSAM 1781 01.01.1900 31.12.2199 1
25108 ASSAM 1781 01.01.1900 31.12.2199 1
25109 ASSAM 1982 01.01.1900 31.12.2199 1
25110 ASSAM 1982 01.01.1900 31.12.2199 1
expected data set
LOC_SID CITY POSTAL_CD EFFECTIVE_DT_FROM EFFECTIVE_DT_TO VERSION
25101 ASSAM 1153 01.01.1900 01.01.2017 1
25102 ASSAM 1153 01.01.2017 31.12.2199 2
25103 ASSAM 1290 01.01.1900 01.01.2017 1
25104 ASSAM 1290 01.01.2017 31.12.2199 2
25105 ASSAM 1310 01.01.1900 01.01.2017 1
25106 ASSAM 1310 01.01.2017 31.12.2199 2
25107 ASSAM 1781 01.01.1900 01.01.2017 1
25108 ASSAM 1781 01.01.2017 31.12.2199 2
25109 ASSAM 1982 01.01.1900 01.01.2017 1
25110 ASSAM 1982 01.01.2017 31.12.2199 2
i tried the below query ,but this will only update the only version ,but doesn't fix the issue
merge into test t
using (
select
t.*,
row_number() over(order by loc_sid asc) rn,
count(*) over() cnt
from test t
where city = 'Assam'
) t1
on (t1.loc_sid = t.loc_id)
when matched the update set
t.version = t1.rn,
t.effective_dt_from =
case
when rn = 1 then t.effective_dt_from
else date '2017-01-01' + rn - 2
end,
t.effective_dt_to =
case
when rn = cnt then t.effective_dt_to
else date '2017-01-01' + rn - 1
end
回答1:
The logic is fine for the dates, however you need row_number() over postal_cd partitions to manage the version:
merge into test t
using (
select
loc_sid,
row_number() over(order by loc_sid asc) rn,
count(*) over() cnt,
row_number() over(partition by postal_cd order by loc_sid) version
from test t
where city = 'Assam'
) t1
on (t1.loc_sid = t.loc_sid)
when matched then update set
t.version = t1.version,
t.effective_dt_from =
case
when rn = 1 then t.effective_dt_from
else date '2017-01-01' + rn - 2
end,
t.effective_dt_to =
case
when rn = cnt then t.effective_dt_to
else date '2017-01-01' + rn - 1
end
Demo on DB Fiddle:
LOC_SID | CITY | POSTAL_CD | EFFECTIVE_DT_FROM | EFFECTIVE_DT_TO | VERSION ------: | :---- | :-------- | :---------------- | :-------------- | ------: 25101 | Assam | 1153 | 01-JAN-00 | 01-JAN-17 | 1 25102 | Assam | 1153 | 01-JAN-17 | 02-JAN-17 | 2 25103 | Assam | 1290 | 02-JAN-17 | 03-JAN-17 | 1 25104 | Assam | 1290 | 03-JAN-17 | 04-JAN-17 | 2 25105 | Assam | 1310 | 04-JAN-17 | 05-JAN-17 | 1 25106 | Assam | 1310 | 05-JAN-17 | 06-JAN-17 | 2 25107 | Assam | 1781 | 06-JAN-17 | 07-JAN-17 | 1 25108 | Assam | 1781 | 07-JAN-17 | 08-JAN-17 | 2 25109 | Assam | 1982 | 08-JAN-17 | 09-JAN-17 | 1 25110 | Assam | 1982 | 09-JAN-17 | 31-DEC-99 | 2
来源：https://stackoverflow.com/questions/61581527/data-integrity-issue-query-fix-logic-oracle-sql