PLSQL UPDATE statement in LOOP takes forever - oracle

If I execute
UPDATE person
SET CUSTOM2 = TIMESTAMP_DIFF(SYSTIMESTAMP,TO_TIMESTAMP (CUSTOM1, 'YYYY-MM-DD HH24:MI:SS.FF'))
WHERE person.id = 'p01';
it updates fine but if I execute update statement inside the loop it hangs forever
DECLARE
TYPE person_ids_t IS table of person.id%type index by PLS_INTEGER;
ids_collection person_ids_t;
CURSOR cur IS select id from person where CUSTOM1 is not null;
BEGIN
OPEN cur;
-- LOOP
FETCH cur BULK COLLECT INTO ids_collection LIMIT 10;
-- EXIT WHEN ids_collection.COUNT = 0;
FORALL idx IN 1 .. ids_collection.COUNT
UPDATE person
SET CUSTOM2 = TIMESTAMP_DIFF(SYSTIMESTAMP,TO_TIMESTAMP (CUSTOM1, 'YYYY-MM-DD HH24:MI:SS.FF'))
WHERE person.id = ids_collection(idx);
COMMIT;
-- END LOOP;
CLOSE cur;
END;
When I execute above block, ScriptRunner hangs forever and I have to force quit.
Not sure why.
even changing the script to
set SERVEROUTPUT ON;
DECLARE
c_id person.id%type;
CURSOR cur IS select id from person where CUSTOM1 is not nul;
vcount integer :=0;
BEGIN
OPEN cur;
LOOP
FETCH cur into c_id;
EXIT WHEN cur%notfound;
-- UPDATE person
-- SET CUSTOM2 = TIMESTAMP_DIFF(SYSTIMESTAMP,TO_TIMESTAMP (CUSTOM1, 'YYYY-MM-DD HH24:MI:SS.FF'))
-- WHERE person.id = c_id;
dbms_output.put_line('c_id: ' || c_id);
vcount := vcount + 1;
IF vcount = 1 THEN
EXIT;
END IF;
END LOOP;
CLOSE cur;
dbms_output.put_line('vcount: ' || vcount);
END;
/
If I execute this it prints
c_id: p01
vcount: 1
but when I uncomment the update statement it hangs forever

My guess is that previous update wasn't committed and it keeps those rows locked so you're just waiting for commit (or rollback).
Anyway: why would you use an option which is suboptimal? As you've already seen, an ordinary UPDATE does the job just fine. Doing it in a loop is updating row-by-row which promises to be slow-by-slow. If I were you, I wouldn't bother.
If you want to find who's blocking who, there are Oracle dictionary views you may query. There are also numerous scripts available on the Internet, have a look. Or, if you use some GUI - like TOAD, which offers the "Schema Browser" that lets you easily see that info - use it.
Meanwhile, just for example, there's the TEST table in Scott's schema; I updated it in one session as:
SQL> update test set sal = 2000 where empno = 7369;
1 row updated.
SQL>
Then I connected to another session (also as Scott) and ran
SQL> update test set sal = 3000;
and - nothing happened. It hangs. So I connected as SYS (which has access to all dictionary views; if you have some other user with appropriate privileges, use it) and queried the database as:
SQL> select
2 (select username from v$session where sid = a.sid) blocker,
3 --
4 a.sid,
5 ' is blocking ',
6 --
7 (select username from v$session where sid = b.sid) blockee,
8 --
9 b.sid
10 from v$lock a join v$lock b on a.id1 = b.id1 and a.id2 = b.id2
11 where a.block = 1
12 and b.request > 0;
BLOCKER SID 'ISBLOCKING' BLOCKEE SID
--------------- ---------- ------------- --------------- ----------
SCOTT 141 is blocking SCOTT 92
SQL>
So, yes - my session 141 is blocking my another session 92. I should commit (or rollback) in session 141 to let session 92 proceed.

Related

For loop not printing last record in a cursor

So,I am working with a cursor,the cursor was initially operating with the initial for loop,now i needed to perform operation if the no of records in the cursor is more than one,so i fetched the no of records first and stored in a variable and used an if-condition based on that.Now the problem is when I run the whole process,the procedure does its job,but only for the first record in the cursor and entirely skips the second record.Please suggest or help me identify the mistake.
Adding code snippet.
for m in get_m_p(a,b)--main cursor
loop
fetch get_m_p into c_m;
exit when g_m_p%notfound;
end loop;
tempcount := g_m_p%ROWCOUNT:
statements---
if(tempcount>1) then
statements----
end if;
end loop;
for the two records the main curosr is returning in first line,operations are only done for the first one,and the second record is being skipped entirely.
This is a superfluous line:
fetch get_m_p into c_m;
You don't explicitly fetch within a cursor FOR loop, it is implicitly done in each loop iteration. Remove that line.
How to get number of rows returned by a cursor? Lucky you, it seems that you don't care whether how many rows - exactly - it returns. All you want to know is whether it returned more than 1 row. So, count them, literally; exit the loop if counter exceeds 1.
SQL> DECLARE
2 CURSOR get_m_p IS
3 SELECT *
4 FROM emp
5 WHERE deptno = 10;
6
7 l_cnt NUMBER := 0;
8 BEGIN
9 FOR m IN get_m_p
10 LOOP
11 l_cnt := l_cnt + 1;
12 EXIT WHEN l_cnt > 1;
13 END LOOP;
14
15 DBMS_OUTPUT.put_line ('Cursor returned at least ' || l_cnt || ' row(s)');
16
17 IF l_cnt > 1
18 THEN
19 NULL;
20 -- the rest of statements go here
21 END IF;
22 END;
23 /
Cursor returned at least 2 row(s)
PL/SQL procedure successfully completed.
SQL>
As there's no way to know how many rows will cursor return, unfortunately, you'll have to check that first, and then decide what to do with the result.
DECLARE
CURSOR get_m_p IS
SELECT *
FROM emp
WHERE deptno = 10;
l_cnt NUMBER := 0;
BEGIN
SELECT COUNT (*)
INTO l_cnt
FROM (-- this is cursor's SELECT statement
SELECT *
FROM emp
WHERE deptno = 10);
FOR m IN get_m_p
LOOP
-- some statements here
IF l_cnt > 1
THEN
NULL;
-- statements to be executed if cursor return more than 1 row
END IF;
END LOOP;
END;
/
Cursor:
Oracle creates memory area to process SQL statement which is called context area and the cursor is pointer to the context area. A cursor holds the rows (one or more) returned by a SQL statement. The set of rows the cursor holds is referred to as the active set.
There are two type of cursor
1. Implicit cursor
2. Explicit cursor
Implicit Cursors :
Implicit cursors are automatically created by Oracle whenever an SQL statement is executed. Any SQL cursor attribute will be accessed as sql%attribute_name as shown below in the example. Use the SQL%ROWCOUNT attribute to determine the number of rows affected
DECLARE
no_of_records number(2);
BEGIN
select * from records;
IF sql%notfound THEN
dbms_output.put_line('no records present');
ELSIF sql%found THEN
no_of_records := sql%rowcount;
IF no_of_records > 1 THEN
dbms_output.put_line('no of records ' || no_of_records);
END IF
END IF;
END;
Explicit Cursors :
Explicit cursors are programmer-defined cursors for gaining more control over the context area. An explicit cursor should be defined in the declaration section of the PL/SQL Block. It is created on a SELECT Statement which returns more than one row.
Please see below example:
DECLARE
r_id records.id%type;
CURSOR c_records is
SELECT id FROM records;
BEGIN
OPEN c_records;
LOOP
FETCH c_records into r_id;
EXIT WHEN c_records%notfound;
dbms_output.put_line('Record id ' || r_id );
END LOOP;
CLOSE c_records;
END;
Reference :
https://www.tutorialspoint.com/plsql/plsql_cursors.htm
As an alternative you can cache every row and process after.
Example using sample schema "HR" on Oracle 11g Express Edition:
DECLARE
CURSOR get_m_p
IS
SELECT *
FROM hr.employees
WHERE department_id = 60
order by employee_id;
--
rcEmp_last get_m_p%rowtype;
l_cnt NUMBER;
BEGIN
FOR rcM IN get_m_p LOOP
l_cnt := get_m_p%rowcount;
Dbms_Output.Put_Line('l_cnt='||l_cnt);
if l_cnt=1 then
rcEmp_last:=rcM;
Else
Dbms_Output.Put_Line('Process='||to_char(l_cnt-1));
Dbms_Output.Put_Line('rcEmp_last.employee_id='||rcEmp_last.employee_id);
--
rcEmp_last:=rcM;
END IF;
End loop;
--
Dbms_Output.Put_Line('Exited FOR-LOOP');
Dbms_Output.Put_Line('l_cnt='||l_cnt);
--
if l_cnt>1 then
Dbms_Output.Put_Line('rcEmp_last.employee_id='||rcEmp_last.employee_id);
End if;
END;
Output:
Text
PL/SQL block, executed in 1 ms
l_cnt=1
l_cnt=2
Process=1
rcEmp_last.employee_id=103
l_cnt=3
Process=2
rcEmp_last.employee_id=104
l_cnt=4
Process=3
rcEmp_last.employee_id=105
l_cnt=5
Process=4
rcEmp_last.employee_id=106
Exited FOR-LOOP
l_cnt=5
rcEmp_last.employee_id=107
Total execution time 35 ms

how to put stored procedure job for sleep for duration of 35 mins

I have below piece of code that is below is stored procedure to clean up the table based on certain conditions that is there is view and if it get refreshed , now please advise if we need to put the logic in a loop.. if the clean-up doesn’t happen, it should sleep for 15 mins and then re-attempt , please advise how to achieve this
create or replace procedure table_clean_up
is
v_refresh_date date;
v_table_count_m integer;
v_table_count_p integer;
begin
select count(*) into v_table_count_m
from all_mviews
where owner = 'M_TO'
and mview_name in ('DC_CASHFLOW_VIEW','DC_CASHFLOW_VIEW_ZERO')
and last_refresh_type = 'COMPLETE';
if v_table_count_m = 2 then
select cast(last_start_date as date) into v_refresh_date
from user_scheduler_jobs
where job_name = 'TABLE_CLEAN_UP_JOB';
select count(*) into v_table_count_p
from all_mviews m
where m.owner = 'P_SM_TO'
and m.mview_name in ('DC_CASHFLOW_VIEW', 'DC_CASHFLOW_VIEW_ZERO')
and m.last_refresh_type = 'COMPLETE'
and m.last_refresh_date > v_refresh_date;
if v_table_count_p = 2 then
delete dc_cashflow_delta;
end if;
end if;
end;
...
DBMS_LOCK.SLEEP(60*35);
...

Return refcursor after select for update

I am trying to return the data through refcursor that is used in Select for update but I am not able to find any way. Can any one please guide me.
CREATE OR REPLACE PROCEDURE SELECT_SCHEDULED_REPORTS
(o_scheduledreports_cursor OUT SYS_REFCURSOR)
IS
CURSOR report_ids
IS
SELECT *
FROM dwp_rep_scheduler_t
WHERE SCHEDULE_ID IN
(SELECT SCHEDULE_ID
FROM
(SELECT *
FROM dwp_rep_scheduler_t a
WHERE status = 1
AND schedule_type = 1
ORDER BY a.start_date
)
WHERE ROWNUM <= 5
) FOR UPDATE OF status;
BEGIN
FOR report_id IN report_ids
LOOP
UPDATE dwp_rep_scheduler_t SET status = 2 WHERE CURRENT OF report_ids;
END LOOP;
COMMIT;
-- can I do something like open o_scheduledreports_cursor for report_ids
END;
/
As suggested by #Lalit Kumar B, i tried following but now it compiles with error as "PLS-00221: 'O_SCHEDULEDREPORTS_CURSOR' is not a procedure or is undefined"
CREATE OR REPLACE PROCEDURE SELECT_SCHEDULED_REPORTS (
o_scheduledreports_cursor OUT SYS_REFCURSOR)
IS
begin
open o_scheduledreports_cursor for
SELECT *
FROM dwp_rep_scheduler_t
WHERE SCHEDULE_ID IN (SELECT SCHEDULE_ID
FROM ( SELECT *
FROM dwp_rep_scheduler_t a
WHERE status = 1 AND schedule_type = 1
ORDER BY a.start_date)
WHERE ROWNUM <= 5)
FOR UPDATE OF status;
BEGIN
FOR report_id IN o_scheduledreports_cursor
LOOP
UPDATE dwp_rep_scheduler_t
SET status = 2
WHERE CURRENT OF report_ids;
END LOOP;
COMMIT;
END;
END SELECT_SCHEDULED_REPORTS;
/
Select for update is a programming tool which you would use to ensure that no one else updates your data. Inside you PL/SQL, you would use the rows locked for update, and then do the required transaction. Commit your changes, Oracle would release the lock mode 3.
You could simply do,
Open cur for
select column_list from table where ....
Not that specific, but, this asktom link about select for ...update is a good read https://asktom.oracle.com/pls/asktom/f?p=100:11:0::::P11_QUESTION_ID:927629362932
You might also be interested to know few interesting things about the clause, here is my take on a myth about select..for update http://lalitkumarb.wordpress.com/2014/09/04/a-myth-about-row-exclusive-table-lock-on-select-for-update-with-no-rows/
Hope it helps!

How to find number of rows in cursor

I would like to find the number of rows in a cursor. Is there a keyword that can help? Using COUNT, we have to write a query. Any help will be greatly appreciated.
The cursor_variable.%ROWCOUNT is the solution. But its value will be 0 if you check it after opening. You need to loop through all the records, to get the total row count. Example below:
DECLARE
cur sys_refcursor;
cur_rec YOUR_TABLE%rowtype;
BEGIN
OPEN cur FOR
SELECT * FROM YOUR_TABLE;
dbms_output.put_line(cur%rowcount);--returning 0
LOOP
FETCH cur INTO cur_rec;
EXIT WHEN cur%notfound;
dbms_output.put_line(cur%rowcount);--will return row number beginning with 1
dbms_output.put_line(cur_rec.SOME_COLUMN);
END LOOP;
dbms_output.put_line('Total Rows: ' || cur%rowcount);--here you will get total row count
END;
/
You must open the cursor and then fetch and count every row. Nothing else will work.
You can also use BULK COLLECT so that a LOOP is not needed,
DECLARE
CURSOR c
IS SELECT *
FROM employee;
TYPE emp_tab IS TABLE OF c%ROWTYPE INDEX BY BINARY_INTEGER;
v_emp_tab emp_tab;
BEGIN
OPEN c;
FETCH c BULK COLLECT INTO v_emp_tab;
DBMS_OUTPUT.PUT_LINE(v_emp_tab.COUNT);
CLOSE c;
END;
/
Edit: changed employee%ROWTYPE to c%ROWTYPE
You can use following simple single line code to print cursor count
dbms_output.put_line(TO_CHAR(cur%rowcount));
This should work for you
DECLARE
CURSOR get_data_ IS
SELECT *
FROM table_abc_
WHERE owner = user_; -- your query
counter_ NUMBER:= 0;
BEGIN
FOR data_ IN get_data_ LOOP
counter_ := counter_ + 1;
END LOOP;
dbms_output.put_line (counter_);
END;
DECLARE #STRVALUE NVARCHAR(MAX),
#CREATEDDATE DATETIME,
#STANTANCEVALUE NVARCHAR(MAX),
#COUNT INT=0,
#JOBCODE NVARCHAR(50)='JOB00123654',
#DATE DATETIME=GETDATE(),
#NAME NVARCHAR(50)='Ramkumar',
#JOBID INT;
CREATE TABLE #TempContentSplitValue (ITEMS NVARCHAR(200))
SELECT #JOBID = i.Id FROM JobHeader_TBL i WHERE Id=1201;
IF EXISTS (SELECT 1 FROM JobHeader_TBL WHERE Id=#JOBID)
BEGIN
SELECT #STRVALUE= Description from ContentTemplate_TBL where Id=1
INSERT INTO #TempContentSplitValue SELECT * FROM dbo.split(#STRVALUE, '_')
SET #STRVALUE=''
DECLARE db_contentcursor CURSOR FOR SELECT ITEMS FROM #TempContentSplitValue
OPEN db_contentcursor
FETCH NEXT FROM db_contentcursor
INTO #STANTANCEVALUE
WHILE (##FETCH_STATUS = 0)
BEGIN
SET #STRVALUE += #STANTANCEVALUE + 'JOB00123654'
SET #COUNT += 1
SELECT #COUNT
FETCH NEXT FROM db_contentcursor INTO #STANTANCEVALUE
END
CLOSE db_contentcursor
DEALLOCATE db_contentcursor
DROP TABLE #TempContentSplitValue
SELECT #STRVALUE
END
Here I am trying to count the total number of customers with age greater than 25. So store the result in the cursor first. Then count the size of the cursor inside the function or in the main begin itself.
DECLARE
cname customer24.cust_name%type;
count1 integer :=0;
CURSOR MORETHAN is
SELECT cust_name
FROM customer24
where age>25;
BEGIN
OPEN MORETHAN;
LOOP
FETCH MORETHAN into cname;
count1:=count1+1;
EXIT WHEN MORETHAN%notfound;
END LOOP;
-- dbms_output.put_line(count1);
dbms_output.put_line(MORETHAN%ROWCOUNT);
CLOSE MORETHAN;
END;
There is a possible work around that may be useful/needed because of the overhead of accessing a database server over a network (e.g., when using Ajax calls)
Consider this:
CURSOR c_data IS
SELECT per_first_name , null my_person_count
FROM person
UNION
SELECT null as per_first_name , count( distinct per_id ) as my_person_count
FROM person
order by my_person_count ;
The first row fetched has the count of records. One MUST add specific columns fetched (the use of the * does not work), and one can add additional filters.
Try this:
print(len(list(cursor)))
I always read that people loop through results. Why not using a count(*)?
An example from my production code:
PROCEDURE DeleteStuff___(paras_ IN Parameters_Type_Rec)
IS
CURSOR findEntries_ IS
select * from MyTable
where order_no = paras_.order_no;
counter_ NUMBER;
CURSOR findEntries_count_ IS
SELECT COUNT(*) from MyTable
where order_no = paras_.order_no;
BEGIN
OPEN findEntries_count_;
FETCH findEntries_count_ INTO counter_;
CLOSE findEntries_count_;
dbms_output.put_line('total records found: '||counter_);
IF (counter_ = 0) THEN
-- log and leave procedure
RETURN;
END IF;
FOR order_rec_ IN findEntries_ LOOP
EXIT WHEN findEntries_%NOTFOUND OR findEntries_%NOTFOUND IS NULL;
-- do stuff - i.e. delete a record.
API_Package.Delete(order_rec_);
END LOOP;
END DeleteStuff___;
If the query is small, that is my prefered way.
In this example, I just want to know (and log) how many entries I'll delete.
p.s. Ignore the three underlines. In IFS, this is used when you want private procedures or functions.
You can’t have cursor count at start. For that you need to fetch complete cursor; that is the way get cursor count.
declare
cursor c2 is select * from dept;
var c2%rowtype;
i number :=0;
begin
open c2;
loop
fetch c2 into var;
exit when c2%NOTFOUND;
i: = i+1;
end loop;
close c2;
dbms_output.put_line('total records in cursor'||i);
end;
You can use %ROWCOUNT attribute of a cursor.
e.g:
DECLARE
CURSOR lcCursor IS
SELECT *
FROM DUAL;
BEGIN
OPEN lcCursor ;
DBMS_OUTPUT.PUT_LINE(lcCursor%ROWCOUNT);
CLOSE lcCursor ;
END;

How can I find the number of records in an Oracle PL/SQL cursor?

Here's my cursor:
CURSOR C1 IS SELECT * FROM MY_TABLE WHERE SALARY < 50000 FOR UPDATE;
I immediately open the cursor in order to lock these records for the duration of my procedure.
I want to raise an application error in the event that there are < 2 records in my cursor. Using the C1%ROWCOUNT property fails because it only counts the number which have been fetched thus far.
What is the best pattern for this use case? Do I need to create a dummy MY_TABLE%ROWTYPE variable and then loop through the cursor to fetch them out and keep a count, or is there a simpler way? If this is the way to do it, will fetching all rows in my cursor implicitly close it, thus unlocking those rows, or will it stay open until I explicitly close it even if I've fetched them all?
I need to make sure the cursor stays open for a variety of other tasks beyond this count.
NB: i just reread you question.. and you want to fail if there is ONLY 1 record..
i'll post a new update in a moment..
lets start here..
From Oracle® Database PL/SQL User's Guide and Reference
10g Release 2 (10.2)
Part Number B14261-01
reference
All rows are locked when you open the cursor, not as they are fetched. The rows are unlocked when you commit or roll back the transaction. Since the rows are no longer locked, you cannot fetch from a FOR UPDATE cursor after a commit.
so you do not need to worry about the records unlocking.
so try this..
declare
CURSOR mytable_cur IS SELECT * FROM MY_TABLE WHERE SALARY < 50000 FOR UPDATE;
TYPE mytable_tt IS TABLE OF mytable_cur %ROWTYPE
INDEX BY PLS_INTEGER;
l_my_table_recs mytable_tt;
l_totalcount NUMBER;
begin
OPEN mytable_cur ;
l_totalcount := 0;
LOOP
FETCH mytable_cur
BULK COLLECT INTO l_my_table_recs LIMIT 100;
l_totalcount := l_totalcount + NVL(l_my_table_recs.COUNT,0);
--this is the check for only 1 row..
EXIT WHEN l_totalcount < 2;
FOR indx IN 1 .. l_my_table_recs.COUNT
LOOP
--process each record.. via l_my_table_recs (indx)
END LOOP;
EXIT WHEN mytable_cur%NOTFOUND;
END LOOP;
CLOSE mytable_cur ;
end;
ALTERNATE ANSWER
I read you answer backwards to start and thought you wanted to exit if there was MORE then 1 row.. not exactly one.. so here is my previous answer.
2 simple ways to check for ONLY 1 record.
Option 1 - Explicit Fetchs
declare
CURSOR C1 IS SELECT * FROM MY_TABLE WHERE SALARY < 50000 FOR UPDATE;
l_my_table_rec C1%rowtype;
l_my_table_rec2 C1%rowtype;
begin
open C1;
fetch c1 into l_my_table_rec;
if c1%NOTFOUND then
--no data found
end if;
fetch c1 into l_my_table_rec2;
if c1%FOUND THEN
--i have more then 1 row
end if;
close c1;
-- processing logic
end;
I hope you get the idea.
Option 2 - Exception Catching
declare
CURSOR C1 IS SELECT * FROM MY_TABLE WHERE SALARY < 50000 FOR UPDATE;
l_my_table_rec C1%rowtype;
begin
begin
select *
from my_table
into l_my_table_rec
where salary < 50000
for update;
exception
when too_many_rows then
-- handle the exception where more than one row is returned
when no_data_found then
-- handle the exception where no rows are returned
when others then raise;
end;
-- processing logic
end;
Additionally
Remember: with an explicit cursor.. you can %TYPE your variable off the cursor record rather then the original table.
this is especially useful when you have joins in your query.
Also, rememebr you can update the rows in the table with an
UPDATE table_name
SET set_clause
WHERE CURRENT OF cursor_name;
type statement, but I that will only work if you haven't 'fetched' the 2nd row..
for some more information about cursor FOR loops.. try
Here
If you're looking to fail whenver you have more than 1 row returned, try this:
declare
l_my_table_rec my_table%rowtype;
begin
begin
select *
from my_table
into l_my_table_rec
where salary < 50000
for update;
exception
when too_many_rows then
-- handle the exception where more than one row is returned
when no_data_found then
-- handle the exception where no rows are returned
when others then raise;
end;
-- processing logic
end;
If this is the way to do it, will
fetching all rows in my cursor
implicitly close it, thus unlocking
those rows
The locks will be present for the duration of the transaction (ie until you do a commit or rollback) irrespective of when (or whether) you close the cursor.
I'd go for
declare
CURSOR C1 IS SELECT * FROM MY_TABLE WHERE SALARY < 50000 FOR UPDATE;;
v_1 c1%rowtype;
v_cnt number;
begin
open c_1;
select count(*) into v_cnt FROM MY_TABLE WHERE SALARY < 50000 and rownum < 3;
if v_cnt < 2 then
raise_application_error(-20001,'...');
end if;
--other processing
close c_1;
end;
There's a very small chance that, between the time the cursor is opened (locking rows) and the select count, someone inserts one or more rows into the table with a salary under 50000. In that case the application error would be raised but the cursor would only process the rows present when the cursor was opened. If that is a worry, at the end do another check on c_1%rowcount and, if that problem was experienced, you'd need to rollback to a savepoint.
Create a savepoint before you iterate through the cursor and then use a partial rollback when you find there are < 2 records returned.
You can start transaction and check if SELECT COUNT(*) MY_TABLE WHERE SALARY < 50000 greater than 1.

Resources