I'm trying to dynamically count the number of null values held in a table. So far I have this:
BEGIN
DECLARE STMT VARCHAR(2000);
FOR v AS CRS CURSOR FOR
SELECT NAME
FROM SESSION.TT1
DO
SET STMT = 'UPDATE SESSION.TT1 TT1
SET NULL_COUNT = (
SELECT COUNT(*) - COUNT('''||v.NAME||''') NULL_COUNT
FROM Table1
)
WHERE TT1.COLUMN_NAME = '''||v.NAME||'''';
EXECUTE IMMEDIATE STMT;
END FOR;
END
This runs fine, and does populate the temp table with data, but it updates every row with the same value (which is obviously incorrect).
Where have I gone wrong?
Note:
The temporary table, TT1, was generated by another procedure, it contains two columns; NAME, and NULL_COUNT. All values in the NULL_COUNT column are undefined at this point, and the NAME column contains column names retrieved from syscolumns.
I also tried removing the row and then inserting a new row rather than updating on the matched column name but this provides the same results.
You need double quotes (or no quotes) around the column name, otherwise you are are simply counting a literal value, rather than the column...
DROP TABLE TABLE1#
CREATE TABLE TABLE1(C INT)#
INSERT INTO TABLE1 VALUES(NULL),( NULL)#
DECLARE GLOBAL TEMPORARY TABLE TT1(NAME VARCHAR(128), NULL_COUNT BIGINT)#
INSERT INTO SESSION.TT1 VALUES ('C',null)#
BEGIN
DECLARE STMT VARCHAR(2000);
FOR v AS CRS CURSOR FOR
SELECT NAME
FROM SESSION.TT1
DO
SET STMT = 'UPDATE SESSION.TT1 TT1
SET NULL_COUNT = (
SELECT COUNT(*) - COUNT("'||v.NAME||'") NULL_COUNT
FROM TABLE1
)
WHERE TT1.NAME = '''||v.NAME||'''';
EXECUTE IMMEDIATE STMT;
END FOR;
END
#
SELECT * FROM SESSION.TT1
#
NAME NULL_COUNT
---- ----------
C 2
Related
Current table1:
col1
-------------
schema.table1
schema.table2
schema.table3
Desired table1:
col1 col2
------------------------------------------------------------
schema.table1 value of (select count(*) from schema.table1)
schema.table2 value of (select count(*) from schema.table1)
schema.table3 value of (select count(*) from schema.table1)
It is not working, I tried using function too, but function doesn't allow to use 'FROM'
select col1, (select count(*) from col1)
from table1
I am trying to create this query in redshift. Can anyone please help me out?
To perform this task you will need a stored procedure AND a defined cursor. The stored procedure allows for looping and the cursor provides the ability to execute a newly created statement (dynamic querying).
For example:
Create the starting materials, 3 tables and a table that references these tables.
create table foo as (select 1 as A);
create table goo as (select 2 as A);
create table hoo as (select 3 as A);
create table tabs as (select 'foo' as tab union all select 'goo' union all select 'hoo');
Next define the stored procedure the will create the dynamic SQL
CREATE OR REPLACE procedure count_tabs(curs1 INOUT refcursor)
AS
$$
DECLARE
row record;
statement varchar := '';
union_needed BOOL := false;
BEGIN
for row in select tab from tabs LOOP
IF union_needed THEN
statement := statement || ' UNION ALL ';
END IF;
statement := statement || 'select \'' || row.tab || '\' as table_name, count(*) as table_count from ' || row.tab ;
union_needed := true;
END LOOP;
RAISE NOTICE 'sql to execute: %',statement;
open curs1 for execute statement;
END;
$$ LANGUAGE plpgsql;
Lastly we need to call the procedure and execute the cursor
call count_tabs('mycursor');
fetch 1000 from mycursor;
A few notes on this:
This assumes you want the results as output on your bench. If you want to create a table with the results this is doable in the same structure
Since the FROM clause value(s) is unknown at compile time this needs to be done in 2 steps - create the query and then execute the query.
I believe you can have the procedure walk this same cursor itself but doing this is exceptionally slow
I have a requirement to create or replace a table only if certain variable value is 1
SET variable = (select statement) -- $variable will be 1 or 0 depending on value returned from select statment
create or replace table2 as select * from table1 (only if $variable =1)
Is there any way to do it? If the variable value is 0, the create statement should be skipped
Using Snowflake Scripting block:
-- CREATE TABLE table1 AS SELECT 1 AS col;
SET variable = (SELECT 1);
BEGIN
IF ($variable = 1) THEN
create or replace table table2 as select * from table1;
END IF;
END;
Variable could be defined also at Snowflake Scripting block level:
DECLARE
variable INTEGER := (SELECT 1);
BEGIN
IF (:variable = 1) THEN
create or replace table table2 as select * from table1;
END IF;
END;
I need to create an update using sql dynamic and all the updated rows have to be sent in a log table.
In microsoft, i can use OUTPUT clause and it inserts the updated rows in a table, but how can i do this in db2, using sql dynamic?
I have the following tables:
AllCustomers - contains all customers from a db
Id
Name
1
John
2
Test
gdpr_id. - contains all customers which should be updated
Id
Name
1
John
gdpr_log - should contain the output of the update stmt
Id
Name
1
John
I found the below syntax , but it just displays the results.
SELECT fields FROM FINAL TABLE
(update table set field = 'value' where id ='xyz')
I tried to create another dynamic stmt as
INSERT INTO
SELECT fields FROM FINAL TABLE
(update table set field = 'value' where id ='xyz')
and the syntax is not recognized.
How can i replace it to insert all the updated values in a log table?
I have to use sql dynamic because the tables which need to be updated are stored in a metadata table and with a cursor, i create the update script for each line from the metadata table.
UPDATE:
Metadata table looks like this:
table
column
AllCustom
Name
AllCustom
Lastname
CREATE OR REPLACE PROCEDURE sp_test ()
DYNAMIC RESULT SETS 1
P1: BEGIN
--*****************VARIABLES *****************
DECLARE EOF INT DEFAULT 0;
declare v_table nvarchar(50);
declare v_column nvarchar(50);
declare v_rowid nvarchar(50);
declare v_stmt nvarchar(8000);
declare s1 statement;
--*****************UPDATE STEP *****************
-- Declare cursor
DECLARE cursor1 CURSOR WITH HOLD WITH RETURN FOR
SELECT table,column FROM metadata_tbl;
declare c1 cursor for s1;
DECLARE CONTINUE HANDLER FOR NOT FOUND
SET EOF = 1;
OPEN cursor1;
WHILE EOF = 0 DO
FETCH FROM cursor1 INTO v_table,v_column;
SET v_stmt = 'WITH A AS
(
SELECT name
FROM FINAL TABLE
(
UPDATE ' || v_table || ' set ' || v_column || ' = ''some name'' where id in (select ID from gdpr_id )
)
)
SELECT COUNT (1) as tst
FROM FINAL TABLE
(
INSERT INTO GDPR_LOG (table,name, LOGDATE)
SELECT ''' || v_table || ''', name, current_timestamp from A
) B';
PREPARE s1 FROM v_stmt ;
open c1 using v_table,v_column;
close c1;
END WHILE;
CLOSE cursor1;
END P1
Update step works fine, insert step duplicates the rows inserted.
What should I do to have the insert step ok?
You have to use SELECT as an outermost statement and keep inner SELECTs in distinct CTEs, if you have a number of them.
Try this:
WITH A AS
(
SELECT ID, NAME
FROM FINAL TABLE
(
UPDATE GDPR
SET NAME = 'Some name'
WHERE ID = 1
)
)
SELECT COUNT (1)
FROM FINAL TABLE
(
INSERT INTO GDPR_LOG (ID, NAME)
SELECT * FROM A
) B
Update:
Using dynamic SQL.
You must enclose the whole statement with some statement termination character (say, #) different from the default one (;) if you use some tool to run this compound statement and specify this statement terminator correctly there.
BEGIN
DECLARE C1 CURSOR FOR S1;
PREPARE S1 FROM
'
WITH A AS
(
SELECT ID, NAME
FROM FINAL TABLE
(
UPDATE GDPR
SET NAME = ?
WHERE ID = ?
)
)
SELECT COUNT (1)
FROM FINAL TABLE
(
INSERT INTO GDPR_LOG (ID, NAME)
SELECT * FROM A
) B
';
OPEN C1 USING 'Name', 1;
CLOSE C1;
END
For example a I have two schemas: SCHEMA_1 and SCHEMA_2. In SHEMA_1 I have a table named TABLE. This table includes two fields: FIELD_1, FIELD_2. In TABLE in FIELD_1 I have some letters: A, B, C. FIELD_2 has tables' names of SCHEMA 2: TABLE_10, TABLE_20, TABLE_30.
SCHEMA_2 includes three tables: TABLE_10, TABLE_20, TABLE_30 with some numbers.
enter image description here
I have to write query to get maximum number of each table in SCHEMA_2. How can I get this result>
enter image description here
Try this:
create table schema1.tableA
(field_1 char(1),
field_2 varchar(10));
insert into schema1.tableA
values ('A','table_10'),
('B','table_20'),
('C','table_30');
create table schema2.table_10
(field_1 dec(5,0));
insert into schema2.table_10
values (20), (30), (40);
create table schema2.table_20
(field_1 dec(5,0));
insert into schema2.table_20
values (6), (9), (12);
create table schema2.table_30
(field_1 dec(5,0));
insert into schema2.table_30
values (10), (15), (20);
with tmp (table_name, field_1) as (
select 'table_10', max(field_1) from schema2.table_10
union all
select 'table_20', max(field_1) from schema2.table_20
union all
select 'table_30', max(field_1) from schema2.table_30)
select a.field_1, b.field_1
from schema1.tableA a
join tmp b on b.table_name = a.field_2;
If you have too many tables to make the above work, you can use a user defined function like this:
create or replace function MaxNbr
(p_TableName varchar(128),
p_TableSchema varchar(128))
Returns dec(5,0)
language sql
not deterministic
no external action
reads sql data
returns null on null input
not fenced
begin
declare l_stmt varchar(1024);
declare l_table varchar(128);
declare l_schema varchar(128);
declare l_result dec(5,0);
set l_table = replace(upper(p_TableName),'"','');
set l_schema = replace(upper(p_TableSchema),'"','');
set l_stmt = 'values (select max(field_1) from "' || l_schema || '"."' ||
trim(l_table) || '") into ?';
prepare S1 from l_stmt;
allocate sql descriptor 'D1';
describe S1 using sql descriptor 'D1';
execute S1 into sql descriptor 'D1';
get sql descriptor 'D1' value 1 l_result = data;
deallocate sql descriptor 'D1';
return l_result;
end;
This is necessary because you cannot use a variable for an identifier like a table name or schema name.
To use the user defined function, you can do something like this:
set schema schema1;
set path = udf_schema;
select field_1, MaxNbr(field_2, 'schema2') as field_2
from tableA;
I am trying to write a script that will copy all the data in table a to table b if table a exists. Table b is the exact same structure as table a would be if it exists, though it may not. I am able to copy using the following statement: INSERT INTO 'B' SELECT * FROM 'A', but I don't know where to use IF EXISTS, or if I even can to determine if I an perform the insertion. I am trying to do this in SQL only as it will be run through as a .sql script from the command line.
MySQL only:
DROP PROCEDURE IF EXISTS myupdate;
DELIMITER //
CREATE PROCEDURE myupdate ()
BEGIN
DECLARE found VARCHAR(64);
SET found = (SELECT TABLE_NAME FROM INFORMATION_SCHEMA.TABLES WHERE TABLE_SCHEMA = Database() AND TABLE_NAME = 'A');
IF found = 'types' THEN
INSERT INTO B SELECT * FROM A;
SELECT 'A into B';
ELSE
SELECT 'A not found';
END IF;
END;//
DELIMITER ;
CALL myupdate();
DROP PROCEDURE myupdate;
Expand to you're liking comparing the column definition in INFORMATION_SCHEMA.COLUMNS for A & B if you need finer control.
I have accepted Wrikken's answer but am using this as my final code. I need to reuse the procedure he provided for multiple tables, so I modified it slightly. It makes the assumption that the backup table has already been created.
DROP PROCEDURE IF EXISTS tableUpdate;
DELIMITER //
CREATE PROCEDURE tableUpdate(name VARCHAR(32))
BEGIN
DECLARE cnt tinyint(1);
DECLARE btable VARCHAR(36);
SET btable = CONCAT(name,'BAK');
SET cnt = (SELECT COUNT(TABLE_NAME) FROM INFORMATION_SCHEMA.TABLES WHERE TABLE_SCHEMA = 'TAA' AND TABLE_NAME = name);
IF cnt > 0 THEN
SET #q:= CONCAT('INSERT INTO ',CONCAT(btable,CONCAT(' SELECT * FROM ',name)));
PREPARE stmt FROM #q;
EXECUTE stmt;
COMMIT;
ELSE
SELECT 'No update necessary.';
END IF;
END;//
DELIMITER ;
CALL tableUpdate('A');
DROP PROCEDURE tableUpdate;
You can do so by performing the following:
select count(*) from my_tables where table_name='b';
If count>0 then
update b...;
else
create table b;
insert into my_tables(table_name) values('b');
insert into b...;
end if;