another Must declare the scalar variable Error - tsql

I have been trying to modify data in a temporary table using cursor and substring function. All works fine when I do all the separate steps via selects, but as soon as I put them together and start it it gives me the error message: Must declare the scalar variable "#tempvar1". These doesn't seem as two or more batches and I can't figure out why it isn't working.
declare #temptabulka table
(
ID_Studenta uniqueidentifier,
NAME varchar(120)
)
Insert into #temptabulka (ID_Studenta, NAME) SELECT ID_Studenta,NAME FROM STUDENTS
DECLARE #tempvarA varchar(120)
DECLARE #tampvar1 varchar(120)
DECLARE prvykurzor CURSOR FOR
SELECT NAME
FROM #temptabulka
OPEN prvykurzor
FETCH NEXT FROM prvykurzor INTO #tempvar1
WHILE ##FETCH_STATUS = 0
BEGIN
SET #tempvarA = #tampvar1
SET #tempvarA = substring(#tempvarA,0,
CASE
WHEN patindex('%_Nez. Priez %',#tempvarA)=0
THEN len(#tempvarA)+1
ELSE patindex('%_Nez. Priez %',#tempvarA)
END
)
UPDATE #temptabulka SET NAME = #tempvarA
WHERE NAME = #tampvar1
FETCH NEXT FROM prvykurzor INTO #tempvar1
END
CLOSE prvykurzor
DEALLOCATE prvy_cursor

You have declared the variable as tampvar1 not tempvar1 and also use it with the wrong name. Replace all tampvar1 with tempvar1 and you should be fine.
here: DECLARE #tampvar1 varchar(120)
here: SET #tempvarA = #tampvar1
here: WHERE NAME = #tampvar1

Related

Return result set using Weakly-typed cursor Variable

CREATE or replace PROCEDURE return_result_set ( )
LANGUAGE SQL
SPECIFIC return_result_set
DYNAMIC RESULT SETS 1
rrs: BEGIN
DECLARE rs_cur CURSOR WITH RETURN
FOR SELECT *
FROM dummytable;
OPEN rs_cur;
END rrs
I can return result set using cursor by using above stored procedure, but I want to use cursor variable (Weakly-typed) in my stored procedure as the select query and table are going to vary based on a condition. Sample code here seems to be for different use case..
How to return a result set using a Cursor Variable?
Try this as is:
--#SET TERMINATOR #
SET SERVEROUTPUT ON#
BEGIN
DECLARE V_C1 CURSOR;
DECLARE V_I INT;
DECLARE PROCEDURE L_PROC(OUT LP_C1 CURSOR)
BEGIN
-- Dynamic cursor
--DECLARE V_STMT VARCHAR(128) DEFAULT 'SELECT * FROM (VALUES 1) T(I)';
--PREPARE V_S1 FROM V_STMT;
--SET LP_C1 = CURSOR FOR V_S1;
-- Static cursor
SET LP_C1 = CURSOR FOR SELECT * FROM (VALUES 1) T(I);
OPEN LP_C1;
END;
CALL L_PROC(V_C1);
FETCH V_C1 INTO V_I;
CALL DBMS_OUTPUT.PUT_LINE('I: ' || V_I);
CLOSE V_C1;
END#
SET SERVEROUTPUT OFF#

PostgreSQL - Declaring a variable that can store multiple values

I need a way to declare a variable that can store multiple values. My first attempt was to declare a variable using the TABLE type:
DECLARE __id TABLE(results_id integer);
However this didnt go as planned, giving me type-declaration errors. My next attempt was to make an integer[] type
DECLARE __id integer[];
but it ended up giving me an error of that values needs to be inserted using curly braces whenever i attempted to insert them with a select function.
SELECT p.id FROM files.main p
WHERE p.reference = __reference
AND p.platform = __platform_id
INTO __id;
I wonder if there is any way to solve this problem?
If you have a table name t you can declare a variable of the type t
create or replace function tf1() returns int as
$BODY$
DECLARE
var public.t;
BEGIN
select * from public.t into var limit 1;
return var.id;
END;
$BODY$
LANGUAGE plpgsql ;
select * from tf1();
i need an array to use them afterwards, where every result will be run in a function and the results will be inserted into a table object of same type – Crated
While you can do this with arrays and variables, it's simpler and faster to do it in a single query using an insert into select.
INSERT INTO some_other_table (some_column)
SELECT your_function(p.id)
FROM files.main p
WHERE p.reference = __reference
AND p.platform = __platform_id
This will run each matching p.id through your_function and insert the results as some_column in some_other_table.

DB2. Select all columns in a table that contains a value

I would like to find all column names in a table that contains a value in any given record.
I.e All columns that contains a string in the record value.
'%ABC%' or '%QAW%' or '%IGH%'
If possible give me all the tables and columns in a DB schema, so I do not have to query ever table manually
2016-06-15
So I got a little further, I can now get all the values from each column in each row in each table. Now I need to see if that value ( v_value ) exist in a list of airport codes. i.e ['LAS','LAX','BIL']
I have all the airports in a table that I want to read into and array.
I am having trouble with creating that array and getting the data into it.
Here is what I have so far.
Look at the TODO's
CREATE OR REPLACE PROCEDURE "CMSDB"."TEST1"
()
LANGUAGE SQL
SPECIFIC SQL3
P1: BEGIN
DECLARE v_tabschema VARCHAR(255);
DECLARE v_tabname VARCHAR(255);
DECLARE v_colname VARCHAR(255);
DECLARE v_airport VARCHAR(255);
DECLARE v_stmt VARCHAR(3000);
DECLARE V_SQL VARCHAR(3000);
DECLARE v_value VARCHAR(255);
DECLARE SQLSTATE CHAR(5) DEFAULT '00000';
DECLARE v_stmt2 STATEMENT;
DECLARE v_value_cursor CURSOR FOR v_stmt2;
DECLARE v_airport_cursor CURSOR FOR select IDX from CMSDB.AIRPORTS;
DECLARE syscat_cursor CURSOR FOR select trim(tabschema), tabname, colname from cmsdb.syscat.columns where tabname = 'ACCTGROUP' and tabschema = 'CMSDB' and TYPENAME = 'VARCHAR' and colname not in ('CHGDATE','CHGPAGE','CHGPROG','CHGTYPE','CHGUSER','CREATEDATETIME','CREATEDBYID','REC_ID');
open v_airport_cursor;
FETCH FROM v_airport_cursor INTO v_airport;
WHILE (SQLSTATE = '00000') DO
call DBMS_OUTPUT.PUT_LINE(v_airport);
-- TODO Add each value to a list, arryalist that can be used to check if the v_value is in the list.
FETCH FROM v_airport_cursor INTO v_airport;
END WHILE;
close v_airport_cursor;
OPEN syscat_cursor;
FETCH FROM syscat_cursor INTO v_tabschema, v_tabname, v_colname;
WHILE (SQLSTATE = '00000') DO
--call DBMS_OUTPUT.PUT_LINE(v_tabschema || ' ' || v_tabname || ' ' || v_colname);
SET v_stmt = 'select ' || v_colname || ' from ' || v_tabschema || '.' || v_tabname;
--call DBMS_OUTPUT.PUT_LINE(v_stmt);
PREPARE v_stmt2 FROM v_stmt;
OPEN v_value_cursor;
FETCH FROM v_value_cursor INTO v_value;
WHILE (SQLSTATE = '00000') DO
-- TODO
--IF ( airportList contains v_value) THEN
--call DBMS_OUTPUT.PUT_LINE(v_value);
--END IF;
FETCH FROM v_value_cursor INTO v_value;
END WHILE;
CLOSE v_value_cursor;
FETCH FROM syscat_cursor INTO v_tabschema, v_tabname, v_colname;
END WHILE;
close syscat_cursor;
END P1
You can use sysibm.syscolumns:
select colname
from sysibm.syscolumns
where tbname = 'XX' and
(name like %ABC%' or name like '%QAW%' or name like '%IGH%');
You'll need to create a cursor over SYSTABLES that returns all the tables in the system. Then have another cursor that returns all the column names in a given table. Once you have those, you can build a dynamic statement that checks all the columns in a given table for the value you are looking for. Fetch the next table name and do it all over again.
Obviously, if you can narrow down your search to a particular schema or even limit the search to tables/columns with a particular naming pattern; you'd be better off.
Another technique, depending on your platform and version of DB2. You might be able to do some sort of a bulk export to a set of text files. Then use a tool that will serach the contents of those text files.

comma separated value in sql statement from a variable

I am getting comma separated value like this in a variable (let say variable name #listobj)
'abc' , 'xyz'
but when I am using below statement it is not giving me the correct result
SELECT * FROM someTable
Where column1 IN (#listobj)
but abc is present in the table.
Where I am doing it wrong?
create a function that split the string to
CREATE FUNCTION dbo.Split(#String varchar(8000), #Delimiter char(1))
returns #temptable TABLE (items varchar(8000))
as
begin
declare #idx int
declare #slice varchar(8000)
select #idx = 1
if len(#String)<1 or #String is null return
while #idx!= 0
begin
set #idx = charindex(#Delimiter,#String)
if #idx!=0
set #slice = left(#String,#idx - 1)
else
set #slice = #String
if(len(#slice)>0)
insert into #temptable(Items) values(#slice)
set #String = right(#String,len(#String) - #idx)
if len(#String) = 0 break
end
return
end
then make call to the function SELECT * FROM someTable
Where column1 IN (dbo.Split(#listobj))
enter link description here
SQLFiddle demo
select * from someTable
where ','+LTRIM(RTRIM(#listobj))+',' LIKE '%,'+LTRIM(RTRIM(column1))+',%'
A classic question and the answer is no, you cannot use a parameter in this way. There are several workarounds though
One of which is to parse the value inside the stored procedure and dynamically generate sql inside the procedure to be execute later. However, this is not a good practice.
Refer to this question
How to pass a comma separated list to a stored procedure?
and also some good discussion on it here
http://social.msdn.microsoft.com/Forums/en-US/sqlintegrationservices/thread/1ccdd39e-8d58-45b2-9c21-5c4dbd857f95/

Getting value from stored procedure in another stored procedure

Sorry, lots of code coming up..
I saw another question like this that used output parameters. I'm using the RETURN statement to return the value I want to use.
I have one stored procedure InsertMessage that looks like this:
ALTER PROCEDURE dbo.InsertNewMessage
(
#messageText text,
#dateTime DATETIME,
#byEmail bit,
#bySMS bit
)
AS
DECLARE #NewId int
BEGIN
BEGIN TRANSACTION
INSERT INTO MessageSet VALUES (#byEmail, #bySMS, #dateTime, #messageText)
SET #NewId = SCOPE_IDENTITY()
COMMIT
END
RETURN #NewId
which another stored procedure uses:
ALTER PROCEDURE dbo.InsertMessageFromUserToGroup
(
#userEmail nvarchar(256),
#groupId int,
#messageText text,
#bySMS bit,
#byEmail bit
)
AS
--Inserts a new message to a group
DECLARE #messageId int
DECLARE #dateTime DATETIME = GETDATE()
--First check if user is a part of the group
IF NOT EXISTS (SELECT userEmail FROM UserToGroupSet WHERE userEmail = #userEmail AND groupId = #groupId)
RETURN 'User not part of group'
ELSE --User is a part of the group, add message
BEGIN
BEGIN TRANSACTION
SET #messageId = [dbo].[InsertNewMessage](#messageText, #dateTime, #bySMS, #byEmail)
INSERT INTO MessageToUser VALUES(#userEmail, #messageId)
INSERT INTO MessageToGroup VALUES(#messageId, #groupId)
COMMIT
END
The row that causes the trouble and of which I'm unsure how to handle is this one:
SET #messageId = [dbo].[InsertNewMessage](#messageText, #dateTime, #bySMS, #byEmail)
The syntax seems ok because I can save it. When I run it I get the error message:
Running [dbo].[InsertMessageFromUserToGroup] ( #userEmail = test#test.com, #groupId = 5, #messageText = sdfsdf, #bySMS = false, #byEmail = true ).
Cannot find either column "dbo" or the user-defined function or aggregate "dbo.InsertNewMessage", or the name is ambiguous.
Transaction count after EXECUTE indicates a mismatching number of BEGIN and COMMIT statements. Previous count = 0, current count = 1.
No rows affected.
(0 row(s) returned)
#RETURN_VALUE =
Finished running [dbo].[InsertMessageFromUserToGroup].
It seems as if the other stored procedure can't be found. I've tried different ways of calling the procedure but everything else fails as well. Any suggestions?
Try changing
SET #messageId = [dbo].[InsertNewMessage](#messageText, #dateTime, #bySMS,
#byEmail)
to
EXEC #messageId = [dbo].[InsertNewMessage] #messageText, #dateTime, #bySMS,
#byEmail
Notice that SET has been changed to EXEC, and the parentheses have been removed from the parameters.
See the example in the MSDN documenation at the end of the article for more information.