I'm having trouble getting the FETCH statement to work correctly in the code below. It's actually not putting any data into the variable registerName. registerName has the same value as it has before the FETCH statement. Thanks!
-- Declare variables/cursors needed for building pivot query
DECLARE qry VARCHAR(8000);
DECLARE registerName VARCHAR(128) DEFAULT '';
DECLARE done BOOLEAN DEFAULT 0;
DECLARE registers CURSOR
FOR
SELECT RegisterName
FROM Register r
INNER JOIN EgaugeDevice ed ON ed.id = r.EgaugeDeviceId
INNER JOIN Site s ON s.id = ed.SiteId
INNER JOIN Facility f ON f.id = s.FacilityId
WHERE ShowInSite = 1 AND FacilityName = FACILITY;
DECLARE CONTINUE HANDLER FOR SQLSTATE '02000' SET done=1;
-- Use temporary table to get results from instantaneous view
CREATE TEMPORARY TABLE IF NOT EXISTS instData (
id INT,
RegisterId INT,
InstantaneousValue BIGINT,
Date_Time DATETIME,
Time_Stamp BIGINT
);
TRUNCATE TABLE instData;
INSERT INTO instData(id, RegisterId, InstantaneousValue, Date_Time, Time_Stamp)
SELECT id, RegisterId, InstantaneousValue, Date_Time, Time_Stamp
FROM vRegisterDataInstantaneous
WHERE Date_Time >= now() - INTERVAL 1 DAY
ORDER BY Time_Stamp DESC;
-- build pivot query from registers listed in Register table
OPEN registers;
FETCH registers INTO registerName;
select registerName AS Test;
CLOSE registers;
Is it possible to not have the same name for the column and the variable RegisterName? This could be causing a conflict, Local Variable Scope and Resolution
Related
I have a query that works outside of a stored procedure (populate dejan_output table), but inside it doesn't. Why?
CREATE DEFINER=`admin`#`%` PROCEDURE `test_sp`()
BEGIN
declare avg_all double;
declare Zipcode text(10);
declare from1, to1, from2, to2, from3, to3, from4, to4 int;
declare Year smallint;
#######################################################################
#create output table here
DROP TABLE IF EXISTS dejan_output;
CREATE TABLE dejan_output(
id INT NOT NULL AUTO_INCREMENT PRIMARY KEY,
Zipcode text(10),
year smallint,
month tinyint,
RevenueIn1000USD decimal(15,4),
calendarData tinyint,
processed bit
);
#######################################################################
#insert into output table here
insert into dejan_output(Zipcode, year, month, RevenueIn1000USD, processed)
select f.Zipcode, year(ReportingMonth) year, month(ReportingMonth) month, round(sum(RevenueUSD) / 1000, 4) value, 0
from monthly_performance_data f join
(
select Zipcode, year(ReportingMonth) year, count(distinct month(ReportingMonth)) count
from monthly_performance_data
group by Zipcode, year(ReportingMonth)
) d on f.Zipcode = d.Zipcode and year(ReportingMonth) = d.year and d.count = 12
group by f.Zipcode, year(ReportingMonth) , month(ReportingMonth)
order by f.Zipcode, year(ReportingMonth) , month(ReportingMonth);
END
The table is created successfully but without any data. I am running it from MySQL Workbench 8.0.
I've found what was causing this but I still don't know why? Still, I will publish my solution so someone else can follow. Just remove any declarations:
declare avg_all double;
declare Zipcode text(10);
declare from1, to1, from2, to2, from3, to3, from4, to4 int;
declare Year smallint;
Because you can use
set #Zipcode = 'some value'
to set that variable.
I store in my DB the demands some users can do. The demands can have differents status (stored as events), such as in progress, finished, waiting, and so on (there's 30ish differents status). The demands have differents deadlines corresponding of differents steps of the treatment.
I need to "freeze" some deadlines of the demands, if their current status belongs to a list of pre-defined ones.
In example :
If a demand has the status "A", I have to "freeze" the deadline 2 to 5.
If the status is "B" or "C", I have to "freeze" the deadline 3 to 5.
If the status is "D", I have to "freeze" the deadline 4 and 5.
I plan to use an EVENT that runs every day, at 19:00 to update (add 1 day) the differents deadlines of the concerned demands.
Table structures :
Table demand
id | someDatas | deadline1 | deadline2 | deadline3 | deadline4 | deadline5
---+-----------+-----------+-----------+-----------+-----------+-----------
| | | | | |
Table status
id | name
---+-----
|
Table events
id | id_demand | someOthersDatas | id_status
---+-----------+-----------------+----------
| | |
I wrote a query to get the demands corresponding of a list of status :
SELECT dem.*, st.`name` as 'statusName'
FROM `status` st
INNER JOIN `events` eve
ON eve.id_status = st.id
INNER JOIN `demand` dem
ON eve.id_demand = dem.id
WHERE st.`name` IN ('A', 'B', 'C', 'D')
AND eve.id IN
(
SELECT MAX(even.id) ev
FROM `demand` de
INNER JOIN `events` even
ON even.id_demand = de.id
GROUP BY de.id
);
This query works perfectly and I can get the desired informations for my treatment, I have the id of the demands, its deadlines and the name of the current status.
I don't mind storing this result in a temporary table such as :
DROP TEMPORARY TABLE IF EXISTS pendingDemands;
CREATE TEMPORARY TABLE IF NOT EXISTS pendingDemands
SELECT /* the query shown above */
To make sure the day I want to add to the deadline is valid (= not a day off) I wrote a function that calculate the next valid day :
DELIMITER //
DROP FUNCTION IF EXISTS `get_next_valid_date`;
CREATE FUNCTION `get_next_valid_date`(MyDate DATETIME) RETURNS DATETIME
BEGIN
REPEAT
SET MyDate = (DATE_ADD(MyDate, INTERVAL 1 DAY));
SET #someCondition = (select isDayOff(MyDate));
UNTIL (#someCondition = 0) END REPEAT;
RETURN MyDate;
END//
This function works perfectly and I get the expected results, and isDayOff() don't need to be detailed.
My problem is that I don't know how to use them (the temporary table pendingDemands and the function get_next_valid_date) together to update the table demand, I'm not skilled enough in SQL to build such pretty UPDATE query.
Any direction I could take?
I finally found a work around based on this answer
I created a stored procedure in which I'm using a cursor storing the query I was using to feed the pendingDemands temporary table.
Then, I looped over that cursor and used a CASE WHEN statement to determine the values to modify :
DELIMITER $$
DROP PROCEDURE IF EXISTS `freezePendingDeadlines` $$
CREATE PROCEDURE `freezePendingDeadlines`()
BEGIN
-- from http://stackoverflow.com/questions/35858541/call-a-stored-procedure-from-the-declare-statement-when-using-cursors-in-mysql
-- declare the program variables where we'll hold the values we're sending into the procedure;
-- declare as many of them as there are input arguments to the second procedure,
-- with appropriate data types.
DECLARE p_id INT DEFAULT 0;
DECLARE pT2P DATETIME DEFAULT NULL;
DECLARE pT3P DATETIME DEFAULT NULL;
DECLARE pT4P DATETIME DEFAULT NULL;
DECLARE pT5P DATETIME DEFAULT NULL;
DECLARE pstatusName VARCHAR(255) DEFAULT NULL;
-- we need a boolean variable to tell us when the cursor is out of data
DECLARE done TINYINT DEFAULT FALSE;
-- declare a cursor to select the desired columns from the desired source table1
-- the input argument (which you might or might not need) is used in this example for row selection
DECLARE demandCursor
CURSOR FOR
SELECT p.id,
p.T2P,
p.T3P,
p.T4P,
p.T5P,
P.statusName
FROM
(
SELECT dem.*, st.`name` as 'statusName'
FROM `status` st
INNER JOIN `events` eve
ON eve.id_status = st.id
INNER JOIN `demand` dem
ON eve.id_demand = dem.id
WHERE st.`name` IN ('A', 'B', 'C', 'D')
AND eve.id IN
(
SELECT MAX(even.id) ev
FROM `demand` de
INNER JOIN `events` even
ON even.id_demand = de.id
GROUP BY de.id
)
) AS p;
-- a cursor that runs out of data throws an exception; we need to catch this.
-- when the NOT FOUND condition fires, "done" -- which defaults to FALSE -- will be set to true,
-- and since this is a CONTINUE handler, execution continues with the next statement.
DECLARE CONTINUE HANDLER FOR NOT FOUND SET done = TRUE;
DROP TEMPORARY TABLE IF EXISTS days_off;
CREATE TEMPORARY TABLE IF NOT EXISTS days_off
(
date_off VARCHAR(5)
);
INSERT INTO days_off VALUES('01-01'),
('05-01'),
('05-08'),
('07-14'),
('08-15'),
('11-01'),
('11-11'),
('12-25');
-- open the cursor
OPEN demandCursor;
my_loop: -- loops have to have an arbitrary label; it's used to leave the loop
LOOP
-- read the values from the next row that is available in the cursor
FETCH demandCursor INTO p_id, pT2P, pT3P, pT4P, pT5P, pstatusName;
IF done THEN -- this will be true when we are out of rows to read, so we go to the statement after END LOOP.
LEAVE my_loop;
ELSE
CASE pstatusName
WHEN 'A' THEN
SET pT2P=get_next_valid_date(pT2P);
SET pT3P=get_next_valid_date(pT3P);
SET pT4P=get_next_valid_date(pT4P);
SET pT5P=get_next_valid_date(pT5P);
WHEN 'B' THEN
SET pT3P=get_next_valid_date(pT3P);
SET pT4P=get_next_valid_date(pT4P);
SET pT5P=get_next_valid_date(pT5P);
WHEN 'C' THEN
SET pT3P=get_next_valid_date(pT3P);
SET pT4P=get_next_valid_date(pT4P);
SET pT5P=get_next_valid_date(pT5P);
WHEN 'D' THEN
SET pT4P=get_next_valid_date(pT4P);
SET pT5P=get_next_valid_date(pT5P);
END CASE;
UPDATE `demand`
SET T2P=pT2P,
T3P=pT3P,
T4P=pT4P,
T5P=pT5P
WHERE id=p_id;
END IF;
END LOOP;
CLOSE demandCursor;
DROP TEMPORARY TABLE IF EXISTS days_off;
END$$
could you please give me an advice how to CALL prcd with SELECT results? Or advice me pls better solution.. I am open minded to all working solution
I have a procedure to control inserting data ...
CREATE PROCEDURE control_insert (
)
And I need to pass data from SELECT results to procedure ...
SELECT t.c1, t.c2
FROM table t1
LEFT JOIN other_table t2
ON t1.id = t2.id
WHERE 1=1
The point is, I need to get some data via SELECT (around 6 tables joined to the base table) and I need to do control for each row before insert.. each row should meet some conditions .. if it doesn't meet them, it should just skip it and process next one ...
The procedure should look like:
CREATE PROCEDURE control_insert (
IN v_c1 INT,
IN v_c2 INT
)
BEGIN
IF v_c1 > 1 THEN
INSERT INTO controlled_table (id, type) VALUES (v_c1, v_c2);
ELSE
/* do nothing */
END IF;
END;
CALL control_insert ( SELECT .... );
Could you help me with that? Is there any possibility to do this via MySQL? I can write a PERL skript, but I want to avoid this type of solution ... I just one to do it only in MySQL way
Thank you
EDIT1: I need to check if ID of the SELECT result and LABEL is already in this table for specific date ... this code above is only an example to demonstrate the situation
SOLUTION
I've found the solution ... so for the other visitors:
calling procedure:
CALL controlInsert();
procedure body:
CREATE PROCEDURE controlInsert()
BEGIN
DECLARE done INT DEFAULT FALSE;
DECLARE v_id INT;
DECLARE v_id_dupl INT;
DECLARE v_label INT;
DECLARE v_date DATE;
DECLARE v_type VARCHAR(100);
DECLARE v_category VARCHAR(255);
DECLARE v_user VARCHAR(255);
DECLARE v_country VARCHAR(255);
DECLARE c1 CURSOR FOR SELECT id, label, date, type, category, user, country FROM t1 LEFT JOIN ... /* whole select with 6 joins ended by ; */
DECLARE CONTINUE HANDLER FOR NOT FOUND SET done = TRUE;
## open cursor
OPEN c1;
## loop through the cursor
read_loop: LOOP
## fetch cursor into variables
FETCH c1 INTO v_id , v_label, v_date, v_type, v_category, v_user, v_country;
## check if there is any record
IF done THEN
LEAVE read_loop;
END IF;
## get count of existing records
SELECT count(*) INTO v_id_dupl
FROM
WHERE 1=1
AND id = v_id
AND label= v_label
AND date = v_date;
## if v_id_dupl = 0 => no rows found (ok to load)
IF (v_id_dupl = 0) THEN
INSERT INTO target_table (id, label, date, type, category, user, country)
VALUES (v_id , v_label, v_date, v_type, v_category, v_user, v_country);
END IF;
END LOOP;
CLOSE c1;
END
If that is all your stored procedure is doing, then you don't actually need it. You can do the whole thing in a single statement:
INSERT INTO controlled_table (id, type)
SELECT t.c1, t.c2
FROM table t1
LEFT JOIN other_table t2 ON t1.id = t2.id
WHERE something = somethingElse
AND t.c1 > 1
Essentially, I've just combined your original query with the INSERT statement in your procedure.
If your procedure is more complex and needs to do multiple operations on each row, then you should look into using a cursor.
I have a sql stored proc where I use a cursor that contains a set of id's from a select statement and I use these id's going over one by one using the cursor to get values into other variables and use those variables to do sql joins .My problem is when I execute this I get many tables returned whereas I need just one table returned.
SET NOCOUNT ON
declare #BSVal as int
declare #GSVal as int
declare #mID as int
declare #qID as int
DECLARE M_Cursor cursor for
select
ms.MID,ms.QID
from vM As ms join QS as qs
ON ms.QSID=qs.QIDjoin
Mar as mar on mar.MarID=qs.MarID
where (ms.Cid='Web')
open M_Cursor
FETCH NEXT FROM M_Cursor
INTO #mID, #qID
--Get values
WHILE ##FETCH_STATUS = 0
BEGIN
set #BSVal= (select top 1 SCID from vSC where (EnID in
(select EnID from En where EnName='BAIDU')
and QTID=1 and MID=#mIDand QSID=#qID)order by ITime desc);
set #GSVal= (select top 1 SCID from vSC where ( EnID in
(select EnID from En where EnName='GRAPHIC') and QTID=1
and MID=#mIDand QSID=#QSID) order by ITime desc);
select * from
vM m
join vw5TABLE BNDCG on (m.QSid=BNDCG.QID And BNDCG.Position=1)
join vw5TABLE GNDCG on (m.QSid=GNDCG.QID And GNDCG.Position=1)
where
BNDCG.SCid=#BSVal
and GNDCG.SCid=#GSVal
and BNDCG.QSID=# qID
and GNDCG.QSID=# qID
and m.MID=#mID
FETCH NEXT FROM M_Cursor
INTO #MID, #QSID
END
CLOSE M_Cursor;
DEALLOCATE M_Cursor;
That code will run a select for each iteration of the cursor, which makes it look like 'many tables'. It sounds like you need to insert the results of that select into a temp table or table variable inside the cursor, then once the cursor is complete, select once from that temp table. I have not gone over your code in detail but I'm guessing a cursor may not be required for this.
Here's a rough sample using a table variable.
DECLARE #temptable TABLE (col1 INT, Col2 VARCHAR(3), Col3 INT)
insert into #temptable (col1,col2,col3)
select (col1,col2,col3) from
vM m
join vw5TABLE BNDCG on (m.QSid=BNDCG.QID And BNDCG.Position=1)
join vw5TABLE GNDCG on (m.QSid=GNDCG.QID And GNDCG.Position=1)
where
BNDCG.SCid=#BSVal
and GNDCG.SCid=#GSVal
and BNDCG.QSID=# qID
and GNDCG.QSID=# qID
and m.MID=#mID
....
..
DEALLOCATE M_Cursor;
SELECT Col1,Col2,Col3 FROM #temptable
DELIMITER $$
CREATE PROCEDURE Load_Fact_List()
BEGIN
DECLARE Project_Number_Temp INT;
DECLARE Panel_Id_Temp INT;
DECLARE Employee_Id_Temp INT;
DECLARE Zip_Temp VARCHAR(255);
DECLARE Created_Date_Temp DATE;
DECLARE Country_Temp VARCHAR(255);
DECLARE no_more_rows BOOLEAN;
DECLARE loop_cntr INT DEFAULT 0;
DECLARE num_rows INT DEFAULT 0;
DECLARE load_cur CURSOR FOR
SELECT Project_Id, Panel_Id, Employee_Id, Zip, Created_Date
FROM Fact_List;
DECLARE CONTINUE HANDLER FOR NOT FOUND
SET no_more_rows = TRUE;
OPEN load_cur;
select FOUND_ROWS() into num_rows;
the_loop: LOOP
FETCH load_cur
INTO Project_Number_Temp, Panel_Id_Temp, Employee_Id_Temp, Zip_Temp, Created_Date_Temp;
IF no_more_rows THEN
CLOSE load_cur;
LEAVE the_loop;
END IF;
SET Country_Temp= (select Country from Zip where Zip= Zip_Temp);
INSERT INTO Test_Fact
(
Project_Key,
Campaign_Key,
Respondents_Key,
Event_Key,
Employee_Key,
Geography_Key,
Date_Key
)
SELECT (SELECT Project_Key from Project_Dim where Project_Id= Project_Number_Temp AND Quota_Country= Country_Temp),0,(SELECT MAX(Respondents_Key) from Respondents_Dim WHERE Panel_Id= Panel_Id_Temp),1,(select MAX(Employee_Key) from Employee_Dim WHERE Employee_Id= Employee_Id_Temp),(Select Geography_Key from Geography_Dim where Zip= Zip_Temp), (Select Date_Key from Date_Dim where Full_Date= Created_Date_Temp);
SET loop_cntr = loop_cntr + 1;
END LOOP the_loop;
select num_rows, loop_cntr;
END $$
The above code is properly working but it is damn slow. For every 1 hour it is loading 1000 records. I got lacks of records to load into fact table. can anyone suggest me any optimization?
Requirement is to load fact table by looping through other table and gathering required key values from dimension tables.
The usual procedure is actually like this.
You have your dimensions built and you just gathered the data you want to insert into your fact table in a temporary table. Then you insert this data in another temporary table like this:
INSERT INTO tmp_fact_table
(
fact_key,
dim1_key,
dim2_key,
...
fact1,
fact2
...
)
SELECT
ISNULL (f.fact_key, 0),
ISNULL (d1.sid, 0) as whatever,
ISNULL (d2.sid, 0) as whatever2,
...
ISNULL (tt.fact1, 0),
ISNULL (tt.fact2, 0)
FROM
yourTempTable tt
LEFT JOIN Dim1 d1 ON tt.identifying_column = d1.identifying_column
...
LEFT JOIN fact_table f ON
f.dim1_key = d1.sid
AND f.dim2_key = d2.sid
where
fact_key is the identifying column in your fact table
dim1_key is the foreign key in your fact table to the dimensions
fact1 and so on are the facts you want in your fact table, clear
the ISNULL() function returns 0 when no entry is found. 0 is the id of your dummy row in each dimension for unknown data
Then you will have a table where you have the IDs of your dimensions linked to the data you want to import into your fact table with 0 as fact key when the entry in the fact table does not already exist and the ID of the fact table entry otherwise.
Then you update the fact table where tmp_fact_table.fact_key != 0
Then you insert into the fact table where tmp_fact_table.fact_key = 0
That's it.
I'm doing this with millions of rows and it takes about half an hour. 300,000 rows is peanuts.