I try to use bulk collect for one of my requirements. I defined a procedure and using bulk collect inside.
The procedure is the compilation of the warnings.
The error message is "PLS-00487: invalid reference to the variable 'I'".
Here is the procedure I use. ---------------------------------- CREATE OR REPLACE procedure APPS.XXX_MRP_PAY4 (RETCODE OUT VARCHAR2, ERRBUF OUT VARCHAR2 -P_SERVICE_CALC_DATE IN DATE, -P_VARIABLE IN NUMBERS ) is -L_VARIABLE NUMBER: = P_VARIABLE; cursor c1 is SELECT EMPLOYEE_NUMBER FULL_NAME OF XX_MRP_SALARY_BREAKUP SB, sit_date sd where = SB.employee_number sd.emp_num (+);
TYPE MRP_RESULTS IS TABLE OF c1% ROWTYPE INDEX BY PLS_INTEGER; MRP MRP_RESULTS;
CLOB L_TEXT;
Start
FND_FILE.put_line (FND_FILE.output, 'EMPLOYEE_NUMBER' |) CHR (9) | "FULL_NAME"); -I'm looping C1 OPEN c1; LOOP FETCH c1 BULK COLLECT INTO MRP LIMIT 1500; WHEN THE MRP OUTPUT. COUNT = 0; because me in 1... MRP. COUNTY loop L_TEXT: = I.EMPLOYEE_NUMBER | Chr (9) | I.FULL_NAME;
FND_FILE.put_line (FND_FILE.output, L_TEXT); end loop; END LOOP; CLOSE c1; end XXX_MRP_PAY4;
-------------------------------
You kindly could guide me on this plss
Seems you placed a CURSOR for... LOOP in a COLLECTION in BULK.
The variable 'i' used to be a record, but now it's a number.
Then the PL/SQL block below do not load and print whatever it is:
-Database 11 g: declare Type (vr_Rec) is rendered DESCRIPTION VARCHAR2 (100) QTDE NUMBER (2)); Type vt_Rec index vr_Rec table of Pls_Integer; va_rec vt_Rec;
vc_Cursor SYS_REFCURSOR;
Start Vc_Cursor open for 'SELECT "x" DESCRIPTION, 1 QTDE FROM DUAL UNION ALL' | "SELECT the"y"DESCRIPTION, 2 QTDE FROM DUAL; Loop Get the vc_Cursor collect in bulk va_rec limit 20000; When the output va_rec. Count = 0; End loop; Close Vc_Cursor;
If va_rec > 0 Then Count. For / * everything * / i in va_rec. first... va_rec. last loop dbms_output.put_line (va_rec (i). DESCRIPTION | ' - ' || va_rec (i). QTDE); End loop; End If; end;
lucitah wrote: Then the PL/SQL block below do not load and print whatever it is:
Yes, this isn't. Look at the code. Each extraction replaces data read by the previous fetch. And since he finished when va_rec. Count = 0, we are left with empty va_rec. You need something like:
declare
Type vr_Rec Is Record (
DESCRICAO VARCHAR2(100)
, QTDE NUMBER(2) );
Type vt_Rec Is table of vr_Rec Index By Pls_Integer;
va_rec vt_Rec;
vc_Cursor SYS_REFCURSOR;
begin
Open vc_Cursor For 'SELECT ''x'' DESCRICAO, 1 QTDE FROM DUAL UNION ALL '||
'SELECT ''y'' DESCRICAO, 2 QTDE FROM DUAL';
Loop
Fetch vc_Cursor Bulk Collect Into va_rec Limit 20000;
Exit When va_rec.Count = 0;
For i In va_rec.First..va_rec.Last Loop
dbms_output.put_line( va_rec(i).DESCRICAO || ' - ' || va_rec(i).QTDE );
End Loop;
End Loop;
Close vc_Cursor;
I was wondering if someone can look at my PL/SQL code and tell me where I'm wrong.
Environment ========= 10g (Rel 2)
Algorithm of =======
A slider lets collect records The data block Bulk insert the data into a table
Implementation ==========
Set serveroutput on;
declare
cursor source_cur is Select * from from t1;
-Define the variables and collection types to be used by the BULK COLLECT clause
source_data_type type is the table of the source_cur % rowtype;
source_data_table source_data_type;
Start
Open source_cur; loop extraction source_cur bulk collect into source_data_table;
When the output source_data_table.count = 0;
-Now, we print the data collected -for i in source_data_table.first... source_data_table. Last -loop -dbms_output.put_line (' company name: ' | source_data_table (i).) Company); -dbms_output.put_line ('Count: ' | source_data_table.) (COUNT); -end loop; end loop; close source_cur;
Note: The statements above Print print the correct data and County
-Now, bulk insert us the data into the target table
Note: The print statement above indicates a number 0 =. Since there is no available record, FORALL loop below will not insert anything in the table
ForAll idx in source_data_table. FIRST... source_data_table. LAST Insert into T2 source_data_table (idx) values; end; /
Thanks in advance
rogers42
You'll want to do something like this
open source_cur;
loop
fetch source_cur
bulk collect into source_data_table;
exit when source_data_table.count = 0;
forall idx in source_data_table.FIRST .. source_data_table.LAST
insert into T2
values source_data_table(idx);
end loop;
close source_cur;
or even better:
insert into target
select col, col, col
from source
Im trying to run this piece of code, but get an error such as displayed below;
-COLLECTIONS IN BULK DECLARE CURSOR v_bulk IS SELECT employee_id Employees;
TYPE c_array IS TABLE OF v_bulk % ROWTYPE INDEX BY PLS_INTEGER; v_array c_array;
BEGIN OPEN v_bulk. LOOP Get the v_bulk COLLECT LOOSE v_array LIMIT 100; EXIT WHEN v_bulk % NOTFOUND; DBMS_OUTPUT. Put_line (v_array); END OF LOOP: CLOSE V_bulk; END; /
Error: Variable binding of the SP2-0552 'CLOSE' not stated
I'm trying to COLLECT in a PL/SQL table, but I get ORA-00947: not enough values error message, even if the table has 4 values and select 4 values. Am I missing something?
I have to add something to this?
I've included the types of database objects that I created on the database.
I have commented on the Original code and used the table DOUBLE just to make simple workout.
/*
CREATE or REPLACE TYPE Usage_Groups_for_coda_rec as
SELECT Usage_Groups_for_coda_rec(6874534,"It is a test - CAE",100, 50)
LOOSE COLLECTION t_uge_Grp_for_coda_tab
DOUBLE;
You try bulk collect into a collection of Usage_Groups_for_coda_recs, then you will need to make type compatible using the implicit of the type constructor.
I plsql record type with several tables with multiple columns. so when I used bulk collect with education for ALL. I want to insert records in multiple tables.
Please give me suggestions.
ForAll is designed to be used with a single DML statement, which may include dynamic SQL statements. However, I do not know what advantage this will give you your list iteration save several times, one for each table - especially since there is an air show with SQL dynamic.
Example 1 (dynamic SQL):
begin
...
forall i in vRecList.First..vRecList.Last
execute immediate '
begin
insert into Table1 (Col1, Col2, Col3) values (:1, :2, :3);
insert into Table2 (Col1, Col2, Col3) values (:1, :2, :3);
end;' using vRecList(i).Col1, vRecList(i).Col2, vRecList(i).Col3;
end;
Another approach that I should work (but not tested) is using to insert all the Scriptures and based record inserts, but you need to try on your version of Oracle forall has changed between the versions. In this case vRecList must be compatible with the Table % ROWTYPE and Table2% ROWTYPE type.
Example 2 (insert all):
begin
...
forall i in vRecList.First..vRecList.Last
insert all
into Table1 values vRecList(i)
into Table2 values vRecList(i)
select 1 from dual; end;
SELECT TYPE(A, B, C, D)
BULK COLLECT INTO ARRAY_TYPE_VAR
FROM table
WHERE ROWNUM <= number;
And after execution of this I need to update the selected lines to mark as processed.
UPDATE table
SET is_proc = 'Y'
WHERE ...
-I want to just select a predefined number of lines (ROWNUM < = number)
-Selected data comes from an intermediate table that I can't control and there is no PKs.
Is it possible that I can use to identify the selected lines and use them in an UPDATE statement?
Thank you
Hugo
I need to select the data to perform certain operations and validations and insert in the paintings of application.
Since I have a large number of files, I found myself with performance while loading problems all in memory.
I am trying to create a kind of limiting process:
Until I have 0 not dealt with the lines:
Select lines of high x-> perform data operations-> lines of the selected mark as processed
The sql code I posted is inside a loop.
I'm probably complicate an easy problem, but I'm kinda stuck with it.
And Yes, there is a column that can be used to identify the data.
There is NO NEED to 'Butterfly' lines of process or as treated brand.
Assuming that you need to use the collections and PL/SQL just treat ALL lines but only a limited number at the time.
Some examples of code (ONLY for the demonstration) which deals with the entire EMP table a few lines at a time:
EXTRACTION is a BULK of COLLECTING all the data in "v". It will get either all data, or none if there is none.
The construction of the LOOP would be used when you have a LIMIT clause while Oracle would be 'loop' return to Download the following set of lines. Run this example in the SCOTT schema and you will see how the LIMIT clause. I have 14 rows in my table EMP.
DECLARE CURSOR c1 IS (SELECT * FROM emp); TYPE typ_tbl IS TABLE c1% rowtype; v typ_tbl; BEGIN OPEN c1; LOOP --Loop added C1 FETCH BULK COLLECT IN v LIMIT 3; -treat 3 records at a time -treat the first 3 max records DBMS_OUTPUT. Put_line ('Treatment' | v.COUNT |) "documents."); I'm IN v.first... v.Last LOOP DBMS_OUTPUT. Put_line (v (i) .empno); END LOOP; OUTPUT WHEN c1% NOTFOUND; END LOOP; DBMS_OUTPUT. Put_line ('all done'); END;
In the loop, you would do any processing of the nested table, you want to make and could use a FORALL do one INSERT into another table.
Get the ext_table_cursor COLLECT LOOSE ext_table_array LIMIT 1000;
EXIT WHEN ext_table_cursor % NOTFOUND;
FORALL rec IN 1.ext_table_array. COUNTY
INSERT INTO ext_table_offline VALUES (rec) ext_table_array;
END LOOP;
ON THE OTHER
RAISE ext_table_empty;
END IF
CLOSE Ext_table_cursor;
EXCEPTION
WHEN ext_table_empty
THEN dbms_output.put_line ('outer Table is empty');
END populate_offline_table;
Can someone help me please
2726801 wrote:
I have some questions:
1. What is the disadvantages of using slider with BULK COLLECT and LIMIT?
This isn't a slow line according to the treatment, and after treatment. The term "bulk collect" is not a magic formula that transforms the transformation of the slow line in function in evolutionary and optimal treatment.
If you move/copy data from a database table to the other, and you use the PL/SQL engine and the memory of PL/SQL to transfer data from a target SQL to another, it is neither optimal or scalable.
Remove the PL/SQL engine called detour and DIRECTLY copy/move the target of SQL source data to the target SQL, optimal and scalable.
I wrote the following PL SQL anonymous block. However, the line dbms_output.put_line (total_tckt_col. FINALLY gives me) output as 366 (DBMS_OUTPUT is Developer SQL) which is correct when there is no limit. If the limit is set at 100 in the FETCH statement then dbms_output.put_line (total_tckt_col. Give me FINALLY) 66. What I'm doing wrong here?
DECLARE
CURSOR cur_total_tckt is select t.ticket_id ticket_id, t.created_date created_date, t.created_by created_by, t.ticket_status ticket_status, t.last_changed last_changed, h.created_date closed_date from n01.cc_ticket_info t innerjoin n01.cc_ticket_status_history h on(t.ticket_id = h.ticket_id) where t.last_changed >='6/28/2012 17:28:59'and t.last_changed <(sysdate + interval '1' day);
type total_tckt_colcn is TABLEOF cur_total_tckt%rowtype; total_tckt_col total_tckt_colcn; total_coach_col total_tckt_colcn; begin
total_tckt_col := total_tckt_colcn (); total_coach_col := total_tckt_colcn (); OPEN cur_total_tckt; loop fetch cur_total_tckt bulk collect into total_tckt_col; -- fetch cur_total_tckt bulk collect into total_tckt_col limit 100; EXIT WHEN(cur_total_tckt%NOTFOUND); END LOOP ; CLOSE cur_total_tckt;
dbms_output.put_line(total_tckt_col.LAST);
FOR i IN total_tckt_col.first..total_tckt_col.last LOOP
dbms_output.put_line(i);
END LOOP; end;
Jocelyn says:
This is a modified version of your code on the standard table EMP in schema scott.
Have you tested it? All you have is last batch has 4 rows. But you print outsite the loop. That way if the last batch is incomplete (within boundary lines) last batch does not have your loop. Suppose you want to print the enames:
DECLARE
CURSOR cur_total_tckt
IS
Select ename
EMP; -I have a total of 14 records in this table
type total_tckt_colcn
is
TABLE OF THE cur_total_tckt % rowtype;
total_tckt_col total_tckt_colcn;
BEGIN
total_tckt_col: = total_tckt_colcn ();
OPEN cur_total_tckt.
LOOP
Cur_total_tckt fetch bulk collect within the limits of the total_tckt_col 5;
SMITH ALLEN WARD JONES MARTIN BLAKE CLARK SCOTT KING TURNER
PL/SQL procedure successfully completed.
SQL >
As you can see, it didn't print the last batch. Why? Because NOTFOUND is set to true if the number of lines that you asked to fetch was not recovered. This last batch has 4 rows while the claim code to get 5. Therefore, NOTFOUND has the value true and outputs code before dealing with this last batch. Therefore, repeat the treatment code again once outside the loop:
DECLARE
CURSOR cur_total_tckt
IS
Select ename
EMP; -I have a total of 14 records in this table
type total_tckt_colcn
is
TABLE OF THE cur_total_tckt % rowtype;
total_tckt_col total_tckt_colcn;
BEGIN
total_tckt_col: = total_tckt_colcn ();
OPEN cur_total_tckt.
LOOP
Cur_total_tckt fetch bulk collect within the limits of the total_tckt_col 5;
SMITH ALLEN WARD JONES MARTIN BLAKE CLARK SCOTT KING TURNER ADAMS JAMES FORD MILLER
PL/SQL procedure successfully completed.
SQL >
But you have to accept to repeat that twice the processing code is no better solution. When you use BULK COLLECT LIMIT we should not leave NOTFOUND but rather by collection.count = 0:
DECLARE
CURSOR cur_total_tckt
IS
Select ename
EMP; -I have a total of 14 records in this table
type total_tckt_colcn
is
TABLE OF THE cur_total_tckt % rowtype;
total_tckt_col total_tckt_colcn;
BEGIN
total_tckt_col: = total_tckt_colcn ();
OPEN cur_total_tckt.
LOOP
Cur_total_tckt fetch bulk collect within the limits of total_tckt_col 6;
I need to select distinct column values (for example: Col (c) .sourcecode) collection and the need to insert in the table if the value in this column does not exist in the table.
Say, Col (c) .sourcecode has the value XYZ, ABC. If the "XYZ" value does not exist in the table, so I need to insert in the table and if "ABC" exists in the tables then don't insert it. (something like fusion) after seecting to the distinct values in a column in the collection.
Thanks n regards Nana Akkivalli.
The collection must be inserted before you can use this topic.
In the Begin block, you have to collect a loose and insert inside: something like
SELECT *
BULK COLLECT INTO l_array
FROM ALL_USER
...CONDITIONS...
1 declare
2 type vcArray is table of varchar2(10);
3 type vdateArray is table of date;
4 l_group# vcArray;
5 l_status vcArray;
6 l_date vdateArray;
7 cursor c is select to_char(sysdate, 'YYYYMMDD_HH24MISS') data, GROUP# , status from v$log;
8 begin
9 open c;
10 loop
11 fetch c bulk collect into l_date, l_group# , l_status;
12 for i in 1..l_date.count
13 loop
14 dbms_output.put_line('l_date: ' || l_date(i));
15 end loop;
16 exit when c%notfound;
17 end loop;
18 close c;
19* end;
declare
*
ERROR at line 1:
ORA-01861: literal does not match format string
ORA-06512: at line 11
Just try to display loop a few output of v$ log view :). Concerning GregG
Hello
GregG says: Hello should work with 10.2 DB.
1 declare
2 type vcArray is table of varchar2(10);
3 type vdateArray is table of date;
4 l_group# vcArray;
5 l_status vcArray;
6 l_date vdateArray;
7 cursor c is select to_char(sysdate, 'YYYYMMDD_HH24MISS') data, GROUP# , status from v$log;
8 begin
9 open c;
10 loop
11 fetch c bulk collect into l_date, l_group# , l_status;
12 for i in 1..l_date.count
13 loop
14 dbms_output.put_line('l_date: ' || l_date(i));
15 end loop;
16 exit when c%notfound;
17 end loop;
18 close c;
19* end;
declare
*
ERROR at line 1:
ORA-01861: literal does not match format string
ORA-06512: at line 11
Just try to display loop a few output of v$ log view :). Concerning GregG
To_char, as the name suggests, returns a VARCHAR2, c.data is therefore a VARCHAR2, but you try to store it in an array of DATEs. Do not use TO_CHAR. In fact, there is no need to use a column for that at all, because SYSDATE will always be the same.
Try something like this:
declare
type vcArray is table of varchar2(10);
l_group# vcArray;
l_status vcArray;
cursor c is select GROUP# , status from v$log;
run_time DATE;
begin
open c;
loop
run_time := SYSDATE;
fetch c bulk collect into l_group# , l_status;
for i in 1..l_group#.count
loop
dbms_output.put_line('run_time: ' || run_time);
end loop;
exit when c%notfound;
end loop;
close c;
end;
/
Published by: Frank Kulash on 12 March 2012 10:26 Added example.
SELECT COL_NAME, COUNT BULK COLLECT INTO v_collections FROM
(
SELECT 'PROD_NAME' COL_NAME, COUNT(*) COUNT FROM TEST_TABLE
WHERE LENGTH(PROD_NAME)>50
UNION ALL
SELECT 'PROD_DESC' COL_NAME, COUNT(*) COUNT FROM TEST_TABLE
WHERE PROD_DESC IS NULL
.
.
.
..10 MORE UNION ALL
);
{code}
--does it lead to one context switch or multiple one for each select in the union all
--I do want to capture # of errors for each column
--although the code runs very fast, still I am worried that it may not be a good coding practice.
--is there any better way to do this?
--TEST_TABLE has 1 million records
Thanks,
RN
> -causes change of context or more for each selection in the union all the > A single switch since the entire SQL (i.e. all queries combined set) is sent. > is there a better way to do it? > It depends on. What you do with the data you are interviewing? The best way is to use SQL instead of PL/SQL whenever possible. But since we do not know what everything you do has no way to help us.
Please review the below excerpt as it errors with a PLS-00103: Encoountered synbol «END».
IF v_geospatial_coordinate_type = "None". THEN P_cursor OPEN; LOOP SEARCH p_cursor LOOSE COLLECTION v_search_results_basic LIMIT 100; BECAUSE me in 1.v_search_results_basic.count
WHEN the OUTPUT v_search_results_basic.count < 100; END LOOP;
CLOSE P_cursor;
Thanks for the help and of course your professionalism.
you missed "LOOP".
IF v_geospatial_coordinate_type = 'None'
THEN
OPEN p_cursor;
LOOP
FETCH p_cursor
BULK COLLECT INTO v_search_results_basic LIMIT 100;
FOR I in 1..v_search_results_basic.count
LOOP --<---
EXIT WHEN v_search_results_basic.count < 100;
END LOOP;
CLOSE p_cursor;
I have a cursor loop where I insert a record into < HEADERS table >.
Inside this loop, I do
SELECT * COLLECT LOOSE lines_table
...
FORALL x IN lines_table. FIRST... lines_table. LAST
INSERT INTO < table > VALUES)
< sequence row id >. NEXTVAL,
< need to insert the number of the line here >
So I need to have the line number for heading 1,: 1, number of line: 2,...
For heading 2, I need to have the line number: 1, number of line: 2 and so on.
Any ideas?
Thank you
Anatoliy
I have line_number in my collection. But here's my problem: if I add rownum in the selection list, for example
SELECT FAH_DATA_STG_ID
BATCH_ID
..........
ROWNUM as line_number
LOOSE COLLECTION lines_table
Of
....
I'm getting ORA-00947: not enough values.
Well, I can understand that because lines_table is replica, so when I add select ROWNUM list he complains that he has no place to put it in. I'm doing it wrong or what I need to add a field to the as well?
Well - it's confusing.
You say you "have line_number in my collection. But then, you ask, that if you add ROWNUM in the selection list he has no place to put it.
BOTH cannot be correct.
Your SCALAR type MUST HAVE an attribute for each value you want to include. If you must have an attribute for EACH column of the table (if you need all the columns) and then an attribute more to the line number.
Then, the SELECT list must have the SAME NUMBER of values.
After you install this update, I was surprised to hear a beep at each arrival of a new e-mail message. The related checkbox (General::Play sound) has no effect. I get the sound if the box is checked or unchecked. After you change the setting stop and
The last two times I turned on my iMac (27 ", late 2013; 10.11.3 bone), a window is open on the desktop with the question above" where is FinderSyncAPIExtension? The window has a comprehensive list of all applications on the computer (and includes ba
Hello I bought HP Pavilion dv4-2110tu laptop. Posses IT core i3 and 4 GB of RAM. Now, I want to upgrade the RAM. I want another 4 GB of RAM, as well as the machine will have 8 GB of RAM that will solv my goal. My questions are these. 1. this laptop m
OJ Pro 8000 A809n - equipped printer CONTINUE and the POWER of the lights. I press NEXT and it feeds a sheet, but top sheet passes through the print area and stops at halfway to the bottom of the sheet. I press CONTINUE and it ejects the sheet, but t