PostgreSQL: One condition for multiple tables - performance

I need to retrieve data from tables by checking a single condition for multiple tables. Here is the sample example as shown below:
create table tb1 (
slno int,
name text,
address text
);
create table tb2 (
slno int,
fname text,
faddress text
);
create table tb3 (
slno int,
mname text,
maddress text
);
create table tb4 (
slno int,
lname text,
laddress text
);
insert into tb1 values(1,'aaa','bbb');
insert into tb1 values(2,'2aaa','2bbb');
insert into tb1 values(3,'3aaa','3bbb');
insert into tb1 values(4,'4aaa','4bbb');
insert into tb1 values(5,'5aaa','5bbb');
insert into tb2 values(1,'faaa','fbbb');
insert into tb2 values(2,'f2aaa','f2bbb');
insert into tb2 values(3,'f3aaa','f3bbb');
insert into tb2 values(4,'f4aaa','f4bbb');
insert into tb2 values(5,'f5aaa','f5bbb');
insert into tb3 values(1,'maaa','mbbb');
insert into tb3 values(2,'m2aaa','m2bbb');
insert into tb3 values(3,'m3aaa','m3bbb');
insert into tb3 values(4,'m4aaa','m4bbb');
insert into tb3 values(5,'m5aaa','m5bbb');
insert into tb4 values(1,'laaa','lbbb');
insert into tb4 values(2,'l2aaa','l2bbb');
insert into tb4 values(3,'l3aaa','l3bbb');
insert into tb4 values(4,'l4aaa','l4bbb');
insert into tb4 values(5,'l5aaa','l5bbb');
The query:
select distinct t2.slno
from
tb1,
tb2 as t2,
tb3 as t3,
tb4 as t4
where
tb1.slno = t2.slno or
t2.slno = t3.slno or
t3.slno = t4.slno;
Note: The above query makes my task complete but taking too more time for execution for huge data.
Questions:
1.How to reduce time complexity?
2. Is there any better way then this?

You can greatly reduce the cost by doing a distinct before the join
select distinct t2.slno
from
(select distinct slno from tb1) t1
cross join
(select distinct slno from tb2) t2
cross join
(select distinct slno from tb3) t3
cross join
(select distinct slno from tb4) t4
where
t1.slno = t2.slno or
t2.slno = t3.slno or
t3.slno = t4.slno
http://sqlfiddle.com/#!15/184dd/3
But if you explain why you have four identical tables you can have a better answer.

Related

Data purging in oracle

[ I have two tables TABLE5 & TABLE6
CREATE TABLE TABLE5 (NAME VARCHAR2(1), CKPY VARCHAR2(3), DT TIMESTAMP(3));
INSERT INTO TABLE5 VALUES (‘A’, ‘100’, TO_TIMESTAMP('2021-02-18 00:00:00.000000000','YYYY-MM-DD HH24:MI:SS.FF'));
INSERT INTO TABLE5 VALUES (‘B’, ‘200’, TO_TIMESTAMP('2020-02-18 00:00:00.000000000','YYYY-MM-DD HH24:MI:SS.FF'));
INSERT INTO TABLE5 VALUES (‘C', ‘300’, TO_TIMESTAMP('2019-02-18 00:00:00.000000000','YYYY-MM-DD HH24:MI:SS.FF'));
INSERT INTO TABLE5 VALUES (‘D’, ‘400’, TO_TIMESTAMP('2018-02-18 00:00:00.000000000','YYYY-MM-DD HH24:MI:SS.FF'));
INSERT INTO TABLE5 VALUES (‘E’, ‘600’, TO_TIMESTAMP('2017-02-18 00:00:00.000000000','YYYY-MM-DD HH24:MI:SS.FF'));
INSERT INTO TABLE5 VALUES (‘F', ‘SKI’, TO_TIMESTAMP('2016-02-18 00:00:00.000000000','YYYY-MM-DD HH24:MI:SS.FF'));
INSERT INTO TABLE5 VALUES (‘G’, ‘SKI’, TO_TIMESTAMP('2015-02-18 00:00:00.000000000','YYYY-MM-DD HH24:MI:SS.FF'));
INSERT INTO TABLE5 VALUES (‘H’, ‘SKI’, TO_TIMESTAMP('2015-02-18 00:00:00.000000000','YYYY-MM-DD HH24:MI:SS.FF'));
INSERT INTO TABLE5 VALUES (‘I’, ‘500’, TO_TIMESTAMP('2014-02-04 12:00:00.000000000','YYYY-MM-DD HH24:MI:SS.FF'));
CREATE TABLE TABLE6 (CKPY VARCHAR2(3), STS VARCHAR2(2));
INSERT INTO TABLE6 VALUES ('100','03');
INSERT INTO TABLE6 VALUES ('200','04');
INSERT INTO TABLE6 VALUES ('300','03');
INSERT INTO TABLE6 VALUES ('400','04');
INSERT INTO TABLE6 VALUES ('500','01')
I want to store the historical records in new table TABLE7(which has same table structure as TABLE5) and populate the records which are over 6 months old from now
We need to find out the records not having STS of '03' &'01' and join table5 and table6 with CKPY common values and store in historical table TABLE7 which are over 6 months old from now ...Whatever records are stored in TABLE7, should get deleted from TABLE5.I was able to move the historical records in TABLE7 but not able to delete those records from TABLE5...The query I have used is
CREATE TABLE TABLE7(NAME VARCHAR2(1), CKPY VARCHAR2(3), DT TIMESTAMP(3));
INSERT INTO TABLE7
( NAME,
CKPY,
DT
)
SELECT T5.NAME,
T5.CKPY,
T5.DT
FROM TABLE5 T5 JOIN TABLE6 T6
ON T5.CKPY = T6.CKPY
WHERE T6.STS NOT IN ('03','01')
AND T5.DT < ADD_MONTHS(CURRENT_TIMESTAMP, -6);
I was able to move historical data into TABLE7 but not able to delete those historical records from TABLE5 after storing it in TABLE7]1
You could give your delete similar logic, joining table5 and table6 again, but you can't do that directly so you need a subquery; and as there would a small but non-zero gap between the two statements you'd probably want to trunc() the date you're comparing with in both the insert and delete, so something like:
DELETE TABLE5
WHERE (NAME, CKPY, DT) IN (
SELECT T5.NAME,
T5.CKPY,
T5.DT
FROM TABLE5 T5 JOIN TABLE6 T6
ON T5.CKPY = T6.CKPY
WHERE T6.STS NOT IN ('03','01')
AND T5.DT < ADD_MONTHS(TRUNC(CURRENT_DATE), -6)
);
You might even want to use TRUNC(CURRENT_DATE, 'MM') to keep the archived/purged data to whole months. Either way, use the same filter for the insert.
You could also, more simply, do:
DELETE TABLE5
WHERE (NAME, CKPY, DT) IN (
SELECT NAME, CKPY, DT
FROM TABLE7
);
but that might not scale once you've transferred a lot of data across.

Oracle delete from tableA where a duplicate row is in tableB

As the title says, I am looking for a way to remove all rows from TableA where there is a matching row in TableB.
the Tables A & B have about 30 columns in them so a WHERE A.col1 = B.col1 etc would be a little problematical. Ideally I was hoping for something like
DELETE FROM tableA WHERE IN TableB
(overly simplified by this type of thing)
IN clause can compare all columns returned from select
DELETE FROM tableA WHERE ( col1,col2,col3,.. ) IN ( select col1,col2,col3... FROM TableB );
The brute force way to establish if two records from each table are the same is to just compare every column:
DELETE
FROM tableA a
WHERE EXISTS (SELECT 1 FROM tableB b WHERE a.col1 = b.col1 AND a.col2 = b.col2 AND ...
a.col30 = b.col30);
You could create function which checks structures of tables and, if they are the same, creates string containing correct conditions to compare.
For example here are two tables:
create table t1 (id, name, age) as (
select 1, 'Tom', 67 from dual union all
select 2, 'Tia', 42 from dual union all
select 3, 'Bob', 16 from dual );
create table t2 (id, name, age) as (
select 1, 'Tom', 51 from dual union all
select 3, 'Bob', 16 from dual );
Now use function:
select generate_condition('T1', 'T2') from dual;
result:
T1.ID = T2.ID and T1.NAME = T2.NAME and T1.AGE = T2.AGE
Copy this, paste and run delete query:
delete from t1 where exists (select 1 from t2 where <<PASTE_HERE>>)
Here is the function, adjust it if needed. I used user_tab_columns so if tables are on different schemas you need all_tab_columns and compare owners too. If you have Oracle 11g you can replace loop with listagg(). Second table has to contain all columns of first table and they have to be same type and length.
create or replace function generate_condition(i_t1 in varchar2, i_t2 in varchar2)
return varchar2 is
v varchar2(1000) := '';
begin
for rec in (select column_name, u2.column_id
from user_tab_cols u1
left join (select * from user_tab_cols where table_name = i_t2) u2
using (column_name, data_type, data_length)
where u1.table_name = i_t1 order by u1.column_id)
loop
if rec.column_id is null then
v := 'ERR: incompatible structures';
goto end_loop;
end if;
v := v||' and '||i_t1||'.'||rec.column_name
||' = '||i_t2||'.'||rec.column_name;
end loop;
<< end_loop >>
return(ltrim(v, ' and '));
end;
If you want to avoid running process manually you need dynamic PL/SQL.
create table tableA (a NUMBER, b VARCHAR2(5), c INTEGER);
create table tableB (a NUMBER, b VARCHAR2(5), c INTEGER);
As you said
WHERE A.col1 = B.col1 etc would be a little problematical
you could intersect the tables and mention all columns from tableA one time, like this:
delete tableA
where (a,b,c) in (select * from tableA
intersect
select * from tableB);

Avoid inserting in oracle [duplicate]

I need to be able to run an Oracle query which goes to insert a number of rows, but it also checks to see if a primary key exists and if it does, then it skips that insert. Something like:
INSERT ALL
IF NOT EXISTS( SELECT 1 WHERE fo.primary_key='bar' )
(
INSERT INTO
schema.myFoo fo ( primary_key, value1, value2 )
VALUES
('bar','baz','bat')
),
IF NOT EXISTS( SELECT 1 WHERE fo.primary_key='bar1' )
(
INSERT INTO
schema.myFoo fo ( primary_key, value1, value2 )
VALUES
('bar1','baz1','bat1')
)
SELECT * FROM schema.myFoo;
Is this at all possible with Oracle?
Bonus points if you can tell me how to do this in PostgreSQL or MySQL.
Coming late to the party, but...
With oracle 11.2.0.1 there is a semantic hint that can do this: IGNORE_ROW_ON_DUPKEY_INDEX
Example:
insert /*+ IGNORE_ROW_ON_DUPKEY_INDEX(customer_orders,pk_customer_orders) */
into customer_orders
(order_id, customer, product)
values ( 1234, 9876, 'K598')
;
UPDATE: Although this hint works (if you spell it correctly), there are better approaches which don't require Oracle 11R2:
First approach—direct translation of above semantic hint:
begin
insert into customer_orders
(order_id, customer, product)
values ( 1234, 9876, 'K698')
;
commit;
exception
when DUP_VAL_ON_INDEX
then ROLLBACK;
end;
Second aproach—a lot faster than both above hints when there's a lot of contention:
begin
select count (*)
into l_is_matching_row
from customer_orders
where order_id = 1234
;
if (l_is_matching_row = 0)
then
insert into customer_orders
(order_id, customer, product)
values ( 1234, 9876, 'K698')
;
commit;
end if;
exception
when DUP_VAL_ON_INDEX
then ROLLBACK;
end;
The statement is called MERGE. Look it up, I'm too lazy.
Beware, though, that MERGE is not atomic, which could cause the following effect (thanks, Marius):
SESS1:
create table t1 (pk int primary key, i int);
create table t11 (pk int primary key, i int);
insert into t1 values(1, 1);
insert into t11 values(2, 21);
insert into t11 values(3, 31);
commit;
SESS2: insert into t1 values(2, 2);
SESS1:
MERGE INTO t1 d
USING t11 s ON (d.pk = s.pk)
WHEN NOT MATCHED THEN INSERT (d.pk, d.i) VALUES (s.pk, s.i);
SESS2: commit;
SESS1: ORA-00001
We can combine the DUAL and NOT EXISTS to achieve your requirement:
INSERT INTO schema.myFoo (
primary_key, value1, value2
)
SELECT
'bar', 'baz', 'bat'
FROM DUAL
WHERE NOT EXISTS (
SELECT 1
FROM schema.myFoo
WHERE primary_key = 'bar'
);
This only inserts if the item to be inserted is not already present.
Works the same as:
if not exists (...) insert ...
in T-SQL
insert into destination (DESTINATIONABBREV)
select 'xyz' from dual
left outer join destination d on d.destinationabbrev = 'xyz'
where d.destinationid is null;
may not be pretty, but it's handy :)
If you do NOT want to merge in from an other table, but rather insert new data... I came up with this. Is there perhaps a better way to do this?
MERGE INTO TABLE1 a
USING DUAL
ON (a.C1_pk= 6)
WHEN NOT MATCHED THEN
INSERT(C1_pk, C2,C3,C4)
VALUES (6, 1,0,1);
It that code is on the client then you have many trips to the server so to eliminate that.
Insert all the data into a temportary table say T with the same structure as myFoo
Then
insert myFoo
select *
from t
where t.primary_key not in ( select primary_key from myFoo)
This should work on other databases as well - I have done this on Sybase
It is not the best if very few of the new data is to be inserted as you have copied all the data over the wire.
DECLARE
tmp NUMBER(3,1);
BEGIN
SELECT COUNT(content_id) INTO tmp FROM contents WHERE (condition);
if tmp != 0 then
INSERT INTO contents VALUES (...);
else
INSERT INTO contents VALUES (...);
end if;
END;
I used the code above. It is long, but, simple and worked for me. Similar, to Micheal's code.
If your table is "independent" from others (I mean, it will not trigger a cascade delete or will not set any foreign keys relations to null), a nice trick could be to first DELETE the row and then INSERT it again. It could go like this:
DELETE FROM MyTable WHERE prop1 = 'aaa'; //assuming it will select at most one row!
INSERT INTO MyTable (prop1, ...) VALUES ('aaa', ...);
If your are deleting something which does not exist, nothing will happen.
This is an answer to the comment posted by erikkallen:
You don't need a temp table. If you
only have a few rows, (SELECT 1 FROM
dual UNION SELECT 2 FROM dual) will
do. Why would your example give
ORA-0001? Wouldn't merge take the
update lock on the index key and not
continue until Sess1 has either
committed or rolled back? – erikkallen
Well, try it yourself and tell me whether you get the same error or not:
SESS1:
create table t1 (pk int primary key, i int);
create table t11 (pk int primary key, i int);
insert into t1 values(1, 1);
insert into t11 values(2, 21);
insert into t11 values(3, 31);
commit;
SESS2: insert into t1 values(2, 2);
SESS1:
MERGE INTO t1 d
USING t11 s ON (d.pk = s.pk)
WHEN NOT MATCHED THEN INSERT (d.pk, d.i) VALUES (s.pk, s.i);
SESS2: commit;
SESS1: ORA-00001
INSERT INTO schema.myFoo ( primary_key , value1 , value2 )
SELECT 'bar1' AS primary_key ,'baz1' AS value1 ,'bat1' AS value2 FROM DUAL WHERE (SELECT 1 AS value FROM schema.myFoo WHERE LOWER(primary_key) ='bar1' AND ROWNUM=1) is null;

subquery inside INSERT statement

I just recently found out that subqueries are not allowed in INSERT statements that are inside stored procedures. This is my script:
begin
execute immediate 'truncate table itcustadm.GL_DTPJ_TEST2';
insert into GL_DTPJ_TEST2
(rule_no,
posted_by_user_id,
transaction_id,
transaction_sr_no,
dr_amount,
cr_amount,
tran_crncy_code,
bkdt_tran_flg,
bank_desc
)
select
tq.rule_no,
tq.posted_by_user_id,
tq.transaction_id,
tq.transaction_sr_no,
tq.dr_amount,
tq.cr_amount,
tq.tran_crncy_code,
tq.bkdt_tran_flg,
(select ent.bank_desc from crmuser.end ent where ent.bank_id = gam.bank_id);
But since the (select ent.bank_desc from crmuser.end ent where ent.bank_id = gam.bank_id) at the bottom of the SELECT statement is not allowed by Oracle, what's the best way to accomplish this?
I actually have this code right before the INSERT statement, but I don't know how to exactly use it:
get_bank_desc := '(select ent.bank_desc from crmuser.end ent ' ||
'where ent.bank_id = gam.bank_id)';
I am not sure what you are exactly trying for, but below code may be useful for you, you can achieve inserting a SubQuery output into a table using below query sample, but make sure output of the SubQuery is a single row o/p, so that you can escape from "ORA-01427: single-row SubQuery returns more than one row" ERROR.
insert into test_ins1
values(1,(SELECT COL2 FROM TEST_INS WHERE COL1=1 ));
Even then you can use rownum in where condition and take the single value.
Please let me know in case of any doubts
declare
bank_desc_temp bank_desk_type; /* the type defined in crmuser.ent for bank_desc*/
begin
select ent.bank_desc into bank_desc_temp from crmuser.end ent where ent.bank_id = gam.bank_id;
execute immediate 'truncate table itcustadm.GL_DTPJ_TEST2';
insert into GL_DTPJ_TEST2
(rule_no,
posted_by_user_id,
transaction_id,
transaction_sr_no,
dr_amount,
cr_amount,
tran_crncy_code,
bkdt_tran_flg,
bank_desc
)
select
tq.rule_no,
tq.posted_by_user_id,
tq.transaction_id,
tq.transaction_sr_no,
tq.dr_amount,
tq.cr_amount,
tq.tran_crncy_code,
tq.bkdt_tran_flg,
bank_desc_temp;
end;
When you say "not allowed" what do you mean? Did you get an error?
I ask, because subqueries are definitely allowed inside an insert as select statement, providing you have the syntax correct (and the subquery returns at most one row), e.g.:
create table test_tab (col1 number, col2 varchar2(10));
begin
insert into test_tab
select 1,
(select 'Yes' from dual d2 where d.dummy = d2.dummy)
from dual d;
commit;
end;
/
select * from test_tab;
COL1 COL2
---------- ----------
1 Yes
There are some syntax issues with the code you provided - where is the from clause, and where are the tq and gam aliases defined?
There are two syntax you can use in your insert statement:
(I)
INSERT INTO table_name( column1, column2....columnN)
VALUES ( value1, value2....valueN);
(II)
INSERT INTO table (column1, column2, ... )
SELECT expression1, expression2, ...
FROM source_table(s)
WHERE conditions;
In your example, you should choose the second approach:
insert into GL_DTPJ_TEST2 (rule_no,
posted_by_user_id,
transaction_id,
transaction_sr_no,
dr_amount,
cr_amount,
tran_crncy_code,
bkdt_tran_flg,
bank_desc
)
select tq.rule_no,
tq.posted_by_user_id,
tq.transaction_id,
tq.transaction_sr_no,
tq.dr_amount,
tq.cr_amount,
tq.tran_crncy_code,
tq.bkdt_tran_flg,
ent.bank_desc
from crmuser.gam
join crmuser.end ent
on ent.bank_id = gam.bank_id
;
basically, if you want to add records using an insert statement, you should use a full select statement first. Here is how I would do it:
(1)
select *
from table1;
(2)
select column1
,column2
,column3
from table1;
(3)
select t1.column1
,t1.column2
,t1.column3
,t2.column4
,t2.column5
from table1 t1
join table2 t2
on t2.id = t1.id
;
(4)
insert into table3 (col1
,col2
,col3
,col4
,col5)
select t1.column1
,t1.column2
,t1.column3
,t2.column4
,t2.column5
from table1 t1
join table2 t2
on t2.id = t1.id
;

Insert into select query with cursor value

I want to write a SQK script with insert query in Oracle where one of the value will be fetched from cursor and rest all will be retrieved from table.
For example, consider Employee table:
Emp_No | Emp_Name
1 | AAA
...........
I am reading the table into a cursor.
Cursor c1 is select emp_no, emp_name from employee;
I am iterating the cursor and adding to a table along with information from another table.
for empCur in c1
loop
insert into employee_info(emp_no, emp_name, address, age, ... ) values (empCur.emp_no, empCur.emp_name, select t.address, t.age, ... from employee_temp_table t where t.emp_no=empCur.emp_no)
end loop;
Is my script valid? If not is there any other way to achieve it? Since few values are in cursor and few are in another table I am not sure how to handle this.
Your script isn't correct because this
select t.address, t.age, ... from employee_temp_table t where t.emp_no=empCur.emp_no
is not a valid expression. You may use a scalar subquery (one-row, one-column subquery) only, like this:
insert into t1(col1, col2, col3) values (1, (select col1 from t2), (select col2 from t2));
Or you may try insert from select:
for empCur in c1 loop
insert into employee_info(emp_no, emp_name, address, age, ... )
select empCur.emp_no, empCur.emp_name, t.address, t.age, ... from employee_temp_table t where t.emp_no=empCur.emp_no;
end loop;
If you want to use the cursor, why not just join the tables inside the cursor?
for empCur in ( select e.emp_no, e.emp_name, t.address, t.age ...
from employee e join employee_temp_table t on ( t.emp_no = e.emp_no )
) loop
insert into employee_info(...) values ( empCur.emp_no, ...);
end loop;
Or with a sql insert: (if you can choose sql over pl/sql - T Kyte says do it)
insert into employee_info
select e.emp_no, e.emp_name, t.address, t.age ...
from employee e join employee_temp_table t on ( t.emp_no = e.emp_no );

Resources