Engage in relaxation with an AUTONOMOUS transaction

Similar Questions

• autonomous transaction block of exception

  Hi friends,
  
  Can you tell me if it is possible to use an autonomous transaction block exception?
  
  If so, how? Please provide me with a guide line
  
  thnx

  ROY wrote:
  Hi friends,

  Can you tell me if it is possible to use an autonomous transaction block exception?

  If so, how? Please provide me with a guide line

  You cannot use an autonomous transaction directly in the exception block:

  SQL> ed
  Wrote file afiedt.buf
  
    1  declare
    2    v_dummy number;
    3  begin
    4    select a into v_dummy from testtab where a = -1;
    5  exception
    6    when no_data_found then
    7      declare
    8        pragma autonomous_transaction;
    9      begin
   10        insert into testtab (a) values (-1);
   11      end;
   12      raise;
   13* end;
  SQL> /
        pragma autonomous_transaction;
               *
  ERROR at line 8:
  ORA-06550: line 8, column 14:
  PLS-00710: PRAGMA AUTONOMOUS_TRANSACTION cannot be specified here
  

  but, if it is declared in a procedure you can call it the exceptional block...

  SQL> ed
  Wrote file afiedt.buf
  
    1  declare
    2    v_dummy number;
    3    procedure insertit is
    4      pragma autonomous_transaction;
    5    begin
    6      insert into testtab (a) values (-1);
    7      commit;
    8    end;
    9  begin
   10    select a into v_dummy from testtab where a = -1;
   11  exception
   12    when no_data_found then
   13      insertit;
   14      raise;
   15* end;
  SQL> /
  declare
  *
  ERROR at line 1:
  ORA-01403: no data found
  ORA-06512: at line 14
  

  or

  SQL> ed
  Wrote file afiedt.buf
  
    1  declare
    2    v_dummy number;
    3  begin
    4    select a into v_dummy from testtab where a = -1;
    5  exception
    6    when no_data_found then
    7      declare
    8        procedure insertit is
    9          pragma autonomous_transaction;
   10        begin
   11          insert into testtab (a) values (-1);
   12          commit;
   13        end;
   14      begin
   15        insertit;
   16      end;
   17      raise;
   18* end;
  SQL> /
  declare
  *
  ERROR at line 1:
  ORA-01403: no data found
  ORA-06512: at line 17
  

• Pipeline raised ORA-06519 function: active autonomous transaction detected

  Hi all
  
  My name is John and I have a problem I need to share with you the guru and the experts. I created the following function of pipeline under user Oracle ABC:
  
  CREATE or replace FUNCTION SomeFunction(p_from_date DATE, p_to_date DATE) T_TAB_A RETURN pipelined
  IS
  PRAGMA autonomous_transaction;
  BEGIN
  DELETE FROM temp_rcm;
  
  INSERT INTO temp_rcm
  SELECT * FROM int.facility fd.
  int. Capacity co
  WHERE co.resource_name = fd.resource_name
  AND co.trade_date = fd.trade_date
  AND co.trade_date BETWEEN p_from_date AND p_to_date;
  
  COMMIT;
  
  FOR rec IN (SELECT co.*
  OF temp_rcm co
  o left join int.outage
  WE (o.flag = 'Y')
  AND o.reason_flag = 'F'
  AND o.INTERVAL = co. INTERVAL OF
  AND co.resource_name = o.resource_name)
  ORDER OF co. MEANTIME,
  Co.Name) LOOP
  pipe ROW (T_A (CRE. INTERVAL, rec.trade_date,
  Rec.resource_name, Rec.day_of_week_long, rec.working_day, rec.peak));
  END LOOP;
  
  RETURN;
  END SomeFunction;
  
  
  I could compile and create the function SomeFunction successfully, but when I executed the following command:
  
  Select * from table (SomeFunction (to_date ('01 / 01/2010 ', to_date('01/01/2010')));)
  
  I went with the Oracle error: ORA-06519: active autonomous transaction detected and restored
  
  I searched on the web, such Oracle error occurs when the function has a missing "COMMIT" or "ROLLBACK" command inside an autonomous_transaction. But the fact is that I have already included the "COMMIT"; in the function. I suspected that the error was caused by the paintings that I did against (such as int.facility and int.capacity) were all the views which belonged to another schema called int. Or is it something that miss me in the service? Thank you for your time and your help.
  
  Kind regards
  John

  This isn't how a ref cursor is used in general. See this thread: {: identifier of the thread = 886365}

  I can't really answer your question without knowing what kind of client you are using (Java, c#, etc.).

  However, the thread should shed a lot of light on the ref Cursor.

• ORA-06519: active autonomous transaction detected and restored

  Hello
  
  I have a trigger and stored procedure of the trigger call. This stored procedure contains instructions COMMIT. So I used PRAGMA AUTONOMOUS_TRANSACTION in my trigger. I created the trigger to fire after insertion in the table. When I try to insert the data, I get the error below.
  
  ORA-06519: active autonomous transaction detected and restored
  
  Can anyone clear or correct me.
  
  Thank you

  Before returning in a PL/SQL block, autonomous transactions started in the block must be met (to the either committed or rolled back at the back). If this is not the case, the active autonomous transaction implicitly rolls back and this error.

  Ensure that, prior to return from an autonomous PL/SQL block, autonomous, all active transactions are explicitly committed or canceled.

• "An autonomous transaction sees all changes by hand transact.

  Hello
  I'm trying to reproduce designs "a stand-alone transaction sees all changes made by the main transaction" on:
  

  Oracle® Database Application Developer's Guide - Fundamentals
  
  10g Release 2 (10.2)
  
  Part Number B14251-01
  
  chapter 2 SQL Processing for Application Developers
  
  
  
  Paragraph : Autonomous Transactions

  I implemented a simple case...

  create table emp_ as select * from emp
  
  
  
  begin
  
    update emp_ set hiredate=hiredate+100 where empno=7934;
  
  end;
  
  
  
  create or replace trigger trg_emp_
  
   after insert or update on emp_
  
   for each row
  
  
  
   declare
  
      pragma autonomous_transaction;
  
      emp_var emp.hiredate%type;
  
  
  
    begin
  
      select hiredate
  
        into emp_var
  
        from emp_
  
      where empno=:new.empno;
  
      dbms_output.put_line('empno: '||:new.empno);
  
      dbms_output.put_line('old hiredate: '||:old.hiredate);
  
      dbms_output.put_line('new hiredate: '||:new.hiredate);
  
    end;

  Before any change...

  SQL> select empno,hiredate from emp_;
  
   
  
  EMPNO HIREDATE
  
  ----- -----------
  
   5498 21/4/1982
  
   5499 11/10/1981
  
   5411 10/10/1981
  
   5410 10/10/1982
  
   7369 17/12/1980
  
   7499 20/2/1981
  
   7521 22/2/1981
  
   7566 2/4/1981
  
   7654 28/9/1981
  
   7698 1/5/1981
  
   7782 9/6/1981
  
   7788 19/4/1987
  
   7839 17/11/1981
  
   7844 8/9/1981
  
   7876 23/5/1987
  
   7900 3/12/1981
  
   7902 3/12/1981
  
   7934 23/1/1982

  After the change...

  SQL> begin
  
    2    update emp_ set hiredate=hiredate+100 where empno=7934;
  
    3  end;
  
    4  /
  
   
  
  empno: 7934
  
  old hiredate: 23/01/82
  
  new hiredate: 03/05/82
  
   
  
  PL/SQL procedure successfully completed

  According to the doc select Oracle of the autonomous transaction should not see the change to column hiredate in table in the main transaction (in the anonymous block)...
  
  Matter what I've done wrong... ???
  
  Thank you
  SIM

  Simon:

  As Tubby has pointed out, your dbms_output orders do not display the selected value in the trigger. Your trigger based demonstration should look like:

  SQL> SELECT * FROM t;
  
          ID DT
  ---------- -----------
           1 05-SEP-2009
           2 17-JUL-2009
  
  SQL> CREATE TRIGGER t_ai
    2     AFTER INSERT OR UPDATE ON t
    3     FOR EACH ROW
    4  DECLARE
    5     PRAGMA AUTONOMOUS_TRANSACTION;
    6     l_dt t.dt%TYPE;
    7  BEGIN
    8     SELECT dt INTO l_dt
    9     FROM t
   10     WHERE id = :new.id;
   11     DBMS_OUTPUT.Put_Line ('ID: '||:new.id);
   12     DBMS_OUTPUT.Put_Line ('Old dt: '||:old.dt);
   13     DBMS_OUTPUT.Put_Line ('New dt: '||:new.dt);
   14     DBMS_OUTPUT.Put_Line ('Aut dt: '||l_dt);
   15  END;
   16  /
  
  Trigger created.
  
  SQL> UPDATE t SET dt = sysdate WHERE id = 2;
  ID: 2
  Old dt: 17-JUL-2009
  New dt: 25-OCT-2009
  Aut dt: 17-JUL-2009
  
  1 row updated.
  

  Thus, the transaction, select automomous doesn't see the modified value of dt.

  I know you are just trying to understand transactions automomous here and would never sometihg like that in the production of right? :-)

  Your trigger, as written, has some interesting side effects because of the automomous transaction. For example:

  SQL> INSERT INTO t VALUES(3, sysdate - 25);
  INSERT INTO t VALUES(3, sysdate - 25)
              *
  ERROR at line 1:
  ORA-01403: no data found
  ORA-06512: at "OPS$ORACLE.T_AI", line 5
  ORA-04088: error during execution of trigger 'OPS$ORACLE.T_AI'
  
  SQL> UPDATE t SET id = 3 where trunc(dt) = TO_DATE('05-Sep-2009', 'dd-mon-yyyy');
  UPDATE t SET id = 3 where trunc(dt) = TO_DATE('05-Sep-2009', 'dd-mon-yyyy')
         *
  ERROR at line 1:
  ORA-01403: no data found
  ORA-06512: at "OPS$ORACLE.T_AI", line 5
  ORA-04088: error during execution of trigger 'OPS$ORACLE.T_AI'
  

  John

• How to stop the acquisition with a relaxation with the NI PCIe-6323

  Hi all

  I wonder if it is possible to stop data acquisition or pcie-6323 with a trigger pulse the same way that I begin to acquire samples with a finished sample mode trigger pulse.

  Thanks in advance

  M.

  The samplesPerChannel that you can show what the sample clock configuration defines the total number of samples for the acquisition of finishes.  In your case, you will read the data previously triggered so permanently in force this setting really only sets the size of the buffer.  If you want to just be large enough to avoid overtaking (although...) If the window you buy is potentially very short, you might want to explicitly configure the size of the buffer to something bigger and maintain the value of samplesPerChannel down so that the reference trigger can be accepted earlier).

  The numberOfSamples you specify when you start the player defines the number of samples for the next call for reading.  If you can read the small windows of streaming data to avoid having a blocking with a large timeout call.  If you do not want to change the time-out period, it is a property of the DAQStream class.

  The "continuous" examples (e.g. this one) should show how you can read back data asynchronous as it is acquired.  Your configuration looks more like a "continuous" example

  Since you want to start and stop using the same line as the trigger, perhaps an alternative to the evolution of the default read pointer would be to set up a central task of analog input with a relaxing break.  The caveat to this is that the break does not stop at the task and as soon as the line goes back to you will begin to acquire the data again - I would say using a meter to separate groups of samples in the buffer zone continued.  If you wish to purchase multiple windows of data in short succession well, then I would go with that instead to avoid having to restart the task (and potentially Miss samples during the restart of the task).

  Best regards

• BEFORE the UPDATE of relaxation with time stamp does not work as expected

  We have a scenario where I check update operations on a table.

  I created a before update TRIGGER, so that every time he goes an update on the main table statement, one before the image of the lines is captured in the table of audit with timestamp.

  Since it is before updating, ideally the audit table timestamp (TRG_INS_TMST) should be less main table timestamp (IBMSNAP_LOGMARKER) VALUE, I mean TRIGGER should happen before the update.

  (I could understand in a way that the UPDATE statement is formulated with SYSTIMESTAMP earlier before the TRIGGER is evaluated and so UPDATE is to have a time stamp prior to TRIGGER, but this isn't what we wanted. We want PRIOR update)

  'Table' IBM_SNAPOPERATION IBM_SNAPLOGMARKER            
  ---- ----------------- -------------------------------
  T1   U                 13-OCT-15 03.07.01.775236 AM   <<---------- This is the main table, This should have the latest timestamp
  T2   I                 13-OCT-15 03.07.01.775953 AM
  

  Here is my test case.

  DELETE FROM TEST_TRIGGER_1;
  
  DELETE FROM TEST_TRIGGER_2;
  
  SELECT 'T1', ibm_snapoperation, ibm_snaplogmarker FROM TEST_TRIGGER_1
  UNION
  SELECT 'T2', ibm_snapoperation, TRG_INS_TMST FROM TEST_TRIGGER_2;
  
  INSERT INTO TEST_TRIGGER_1 (ID,ibm_snapoperation, ibm_snaplogmarker)
       VALUES (1, 'I', SYSTIMESTAMP);
  
  COMMIT;
  
  SELECT 'T1', ibm_snapoperation, ibm_snaplogmarker FROM TEST_TRIGGER_1
  UNION
  SELECT 'T2', ibm_snapoperation, TRG_INS_TMST FROM TEST_TRIGGER_2;
  
  UPDATE TEST_TRIGGER_1
     SET IBM_SNAPOPERATION = 'U', ibm_snaplogmarker = SYSTIMESTAMP;
  
  COMMIT;
  
  SELECT 'T1', ibm_snapoperation, ibm_snaplogmarker FROM TEST_TRIGGER_1
  UNION
  SELECT 'T2', ibm_snapoperation, TRG_INS_TMST FROM TEST_TRIGGER_2;
  

  Def trigger:

  CREATE OR REPLACE TRIGGER etl_dbo.TEST_TRIGGER_1_TRG BEFORE UPDATE OF IBM_SNAPOPERATION
  ON TEST_TRIGGER_1 REFERENCING OLD AS OLD NEW AS NEW
  FOR EACH ROW
  WHEN (
  NEW.IBM_SNAPOPERATION= 'U'
        )
  DECLARE
  V_SQLCODE  VARCHAR2(3000);
  --PRAGMA AUTONOMOUS_TRANSACTION;
  BEGIN
  INSERT INTO etl_dbo.TEST_TRIGGER_2
  (ID,
  IBM_SNAPOPERATION,
  IBM_SNAPLOGMARKER,
  TRG_INS_TMST
  )
  VALUES (:OLD.ID,:OLD.IBM_SNAPOPERATION,:OLD.IBM_SNAPLOGMARKER,SYSTIMESTAMP)
  ;
  --COMMIT;
  END;
  /
  

  Output is something like this

  1 row deleted.
  1 row deleted.
  no rows selected.
  1 row created.
  Commit complete.
  
  'T1' IBM_SNAPOPERATION IBM_SNAPLOGMARKER            
  ---- ----------------- -------------------------------
  T1   I                 13-OCT-15 03.07.00.927546 AM 
  1 row selected.
  1 row updated.
  Commit complete.
  
  'T1' IBM_SNAPOPERATION IBM_SNAPLOGMARKER            
  ---- ----------------- -------------------------------
  T1   U                 13-OCT-15 03.07.01.775236 AM   <<---------- This is the main table, This should have the latest timestamp
  T2   I                 13-OCT-15 03.07.01.775953 AM 
  
  2 rows selected.
  

  But for some reason, even after the creation of the 'AFTER' trigger for update, it works as expected. Sense - the main table is not having the last timestamp given

  It's OKAY - I told you in my reply earlier. Reread my answer.

  could understand somehow that the UPDATE statement is made with earlier

  SYSTIMESTAMP until the TRIGGER is assessed and updated so is to have

  time stamp prior to the trigger, but this isn't what we wanted. We want to

  BEFORE the update)

  As I told you before that your UPDATE statement occurs BEFORE the trigger is activated.

  Despite what the other speakers have said, it makes NO DIFFERENCE if you use a BEFORE UPDATE or an AFTER UPDATE trigger. Your UPDATE statement runs ALWAYS BEFORE the trigger.

  HE has TO - it's your update processing statement that causes the trigger to fire.

  Your update statement includes SYSTIMESTAMP. If during the processing of your return to update the value of SYSTIMESTAMP "at this exact time" is captured.

  Then your trigger is activated and starts to run. ANY reference to SYSTIMESTAMP that you use in your trigger cannot be earlier than the value of until the trigger was executed. It's IMPOSSIBLE.

  The trigger can use the SAME value by referencing: NEW and the column name you store the value. Or the trigger can get its own value that your code is doing.

  But the SYSTIMESTAMP value in the trigger will NEVER earlier than the value in your query.

  And none of these values can actually be used to tell when the changes are really ENGAGED since the trigger does not work and CAN NOT, to know when, or if, a validation occurs.

  Reread my first answer - he explains all this.

• Problem with cursor and Transactions

  Hi friends,
  
  I will carry out the sub set of orders where some records are removed as a transaction. If there is no problem with a set of transactions, this recordset is ignored and following Recordset is executed. With code below, if an error occurs in one of the statement, control passes to the exception block where it cancels the registration point and unwanted instructions are restored. So far so good. But the problem is execution stops here. I need to skip this record id and loop for remaining IDs. Help, please.
  
  
  create or replace PROCEDURE DeleteRecords
  AS
  SelectDate date: = SysDate () - 27;
  NRTSI varchar2 (100);
  CURSOR c_TrnInfoNull IS select id from trn_info where CREATED_TISTMP < SelectDate and trn_tistmp is set to null;
  BEGIN
  
  FOR myrows IN c_TrnInfoNull LOOP
  SAVEPOINT TrnSavePointNull;
  delete from TRN_ERROR_DETAIL where TRN_INFO_ID = myrows.id;
  Select code from NRTSI from TRN_PYMT_DETAIL where trn_info_id = myrows.id;
  delete from TRN_PYMT_ADDNL_DATA where TRN_PYMT_DETAIL_ID = NRTSI;
  delete from TRN_PYMT_DETAIL where ID = NRTSI;
  delete from trn_info where id = myrows.id;
  COMMIT;
  END LOOP;
  
  EXCEPTION, THEN, THEN go BACK to TrnSavePointNull.
  DBMS_OUTPUT. Put_line (dbms_utility. FORMAT_ERROR_BACKTRACE());
  END;

  create or replace PROCEDURE DeleteRecords
  AS
  SelectDate date:= SysDate()-27;
  TRNID varchar2(100);
  CURSOR c_TrnInfoNull IS select id from trn_info where CREATED_TISTMP
  

• Add box of relaxation with the text?

  When I add a relaxation area, I would like that he looks just like the three that appear by default. But add those I have no text to the breast area. Is it possible to add a trigger/text/box similar to the three that appear automatically?

  Hi Kimberly,

  When adding a new target with a rectangle, a new trigger is created. You can then add a text box or a picture in the rectangle.

  Kind regards

  Aish

• With the DDL transaction number

  Is it true you will get the error trying to execute DDL for some a table if there is an active transaction at this table? Take a look at this case:
  

  insert into t1 values(1,2,3)
  drop table t1

  as you can see, I don't have "commited yet, if the T1 transaction is still active.
  
  I tried to open a new session in SQL Developer just to confirm that the DML insert has not been committed yet
  

  select * from t1

  and Yes, it returns no rows.
  
  I was expecting error, but the

  drop table t1

  did not produce an error. He successfully executed. Am I missing something? Thank you

  You need to do the INSERTION and the FALL in two different sessions to generate an error (ORA-00054: resource busy and acquire with NOWAIT specified).

  DDL performs an implicit validation before and after that the DDL is executed. If your DROP statement commits first INSERT then drops the table and finally delivers a further validation. That does not generate errors because by the time wherever the table is deleted, there is no active transaction. In addition, even if there is an active transaction, the you'll never erred waits for a lock that already has your session. Because locks are held by the session, various statements in the same session cannot block the other.

  Justin

• JMS sender with container managed transaction ejb3

  Hi all
  
  I refer to the following link http://download.oracle.com/docs/cd/E11035_01/wls100/jms/trans.html#wp1035937
  
  I found that JTA support JMS.
  
  But I have no ' to use JTA explicitly, I want to use the container to manage the transaction. for example. within the ejb3 stateless session bean.
  
  support JMS?
  
  With respect,
  WP

  Hello

  Yes, WebLogic JMS supports JTA (alias XA, a.k.a global) transactions such as container managed transactions. There are two conditions for CMTs on SSB:

  (1) use a parameter descriptor XML or EJB annotation to turn on CMT for SSB

  (2) use a WebLogic JMS connection factory that is configured to have 'global transactions (XA) enabled.

  And I usually recommend also:

  (3) consider using a JEE 'res-ref' for the connection factory to allow reunification of the JMS resources. See "Enhanced Support for using WebLogic JMS with EJB and servlet ' (http://download.oracle.com/docs/cd/E14571_01/web.1111/e13727/j2ee.htm#g1329180) and the 'integration JMS remote suppliers' FAQ (http://download.oracle.com/docs/cd/E14571_01/web.1111/e13727/interop.htm#JMSPG553).

  (4) avoid using a BSN to receive messages. BMD is designed to process incoming messages.

  It is also possible to have WebLogic automatically registered foreign suppliers (non-WebLogic) in WebLogic transactions. See the 'integration JMS remote suppliers' FAQ for more details (http://download.oracle.com/docs/cd/E14571_01/web.1111/e13727/interop.htm#JMSPG553).

  Kind regards

  Tom

• L2 roaming with 2 autonomous access points?

  Hello

  Is it possible for 2 1130AG in stand-alone mode to quick support layer 2 roaming?

  My clients are a mix of clients of compatible and not cisco cisco ccx.

  (Authentication could be open or WPA etc.)

  If so, do 1 AP do acts as a WDS?

  I fell on the docs that say I need a WLSM and that an acting as a WDS AP don't work...

  If I do not use WDS and simply have 2 AP with the same SSID: -.

  If I only use open authentication, then the roam would be a little faster?

  The AP would update switches upstream MAC tables when a customer changes AP?

  Thank you

  Oh, I see now. You get absolutely L2 roaming out of the box. It's all part of the 802.11 Protocol - the customer will inform current accreditation which he left, which AP it will and then he will wander. With open authentication or PSK, it will be almost transparent to the client.

  Cisco Secure fast roaming is an entirely different topic. In the above case, if the customer is using the PEAP Protocol, it will need to authenticate to the RADIUS server whenever he roams to a new AP. Cisco FSR puts cached credentials in the WDS AP for fast authentication, which takes the RADIUS server outside the process. This significantly speeds up a course for users of the EAP, and it is especially noticeable (and necessary) for wireless IP phone users.

  Again, I hope it helps!

• Engage the creativity with the Cloud?

  I would like to know how to access the new range of Adobe applications. However, this is not prove easy and a "glich" seems to happen at all stages: being passed from one Department to the other.

  With a growing workload, it is essential that I am able to work faster. See the creative cloud, I thought to report without delay: this is the product for me!

  I'm looking for help to install PhotoShop CS6 currently a subscription cloud creative member. £46:88 per month. Order No. AD001224841UK. Order date: 30 October 2012.

  Is there any help Adobe out there please?
  

  You had not verified, your Adobe ID email address. You should have received an email providing a link to do this. I forced a verification of your email and so you should now be able to connect to the cloud of Creative.

• error with triggers

  Hi all
  I get the error message with trigger
  What is the meaning of the error message
  and how to fix it?
  

  Error at startup on line 63 of order:
  call xx_GL_Vacancies.delete_transaction (128540)
  Error report:

  SQL Error: ORA-06519: active transaction autonomous detected and restored

  ORA-06512: at "XXGLOBALLOGIC. XX_GL_VACANCIES', line 37
  06519 00000 - "active autonomous transaction detected and restored.
  * Cause: Before returning from a PL/SQL stand-alone, autonomous everything block

  > transactions started in the block must be completed (either
  (> validated or cancelled). If not, activate autonomous
  > transaction implicitly rolls back and this error.

  * Action: make sure that before you return from an autonomous PL/SQL block,

  > autonomous all active transactions are explicitly engaged
  or restored back.

  I'm not angry.
  "You wouldn't like me when I'm angry."
  http://en.Wikipedia.org/wiki/The_Incredible_Hulk _ (TV_series)

  Just keen you can do good things.

  Think about what's going on with this autonomous transaction.
  If I run a delete transaction that subsequently cancels for any reason, you have just updated your xx_per_all_vacancies table.
  And you did it in a standalone transaction.
  That must be validated.
  So it won't restore.
  Is that what you want?

  Unless you connect to something that should engage regardless of the success or failure of the main operation, you should not use autonomous transactions.

  None of the actions performed by your trigger must be commit the code to be directly in a trigger or procedure.

  You should also read up on statement restarts and the consequences for triggers to see why the use of the autonomous transaction is doubly bad.
  http://tkyte.blogspot.com/2010/04/that-old-restart-problem-again.html

  So, in summary, if you really need to do all this in a trigger (really?) have then the logic of the trigger or call to a procedure (preferable), do not use an autonomous transaction, not to commit. Simple.

  Also, if it doesn't "work" provide a bit more detail. Error messages, line numbers, they are all here to help. To say something is does not, is not useful.

  Published by: Dom Brooks on February 23, 2011 11:23

  Published by: Dom Brooks on February 23, 2011 11:26

• How to create and insert data with Execute Immediate?

  Hi guys

  Am stuck on a procedure of formatting in a package... script works okay however integrating a module turns out to be difficult!

  Am not used to oracle... I have the script runs, but not in the package... Well not all that... Drop Table worked

  CREATE or REPLACE PACKAGE BODY is

  PROCEDURE DropTable1 IS

  BEGIN

  run immediately ("DROP TABLE mytable1");

  END;

  PROCEDURE PopulateTable1 IS

  BEGIN

  immediately execute ('CREATE TABLE mytable1 )

  AS LONG AS)

  Select

  substr (T1.genarea, 3, 3) as M_Class,

  substr (T1.genarea, 6, 30) as M_Description,

  substr (T1.genarea, 36, 3) as M_Class,

  substr (T1.genarea, 39, 30) as M_Description,

  substr (T1. ItemItem, 1, 3) as product_code,.

  T3. CHANNEL_NUM as SALES_CHANNEL,

  to_date(''t2.time_id'',''dd-mon-yyyy'') as mis_date,

  Sum (T2.ap_cw_cfi_irp + T2.ap_cw_issues_irp) as ap_gross,

  sum (t2. Ap_Cw_Cfi_Irp + t2. Ap_Revivals_Irp) as ap_net,

  Sum (T2.sp_inc_irp + T2.sp_issues_irp) as sp_gross,

  Sum (T2.sp_dec_irp + T2.sp_fs_irp) as sp_net

  Of

  d_pr t1, t2 age_map t3 law

  where

  T1.pfx = "WE"and t1.coy ="1" and t1.tabl = "T81" and substr (t1.itemitem, 1, 3) = t2.product_id and t3. AGE_NUM = t2.age_id

  Group

  substr (T1.genarea, 3, 3),

  substr (T1.genarea, 6, 30),

  substr (T1.genarea, 36, 3),

  substr (T1.genarea, 39, 30),

  substr (T1. ItemItem, 1, 3).

  T3. CHANNEL_NUM,

  to_date(''t2.time_id'',''dd-mon-yyyy'')

  )');

  COMMIT;

  END PopulateTable1;

  END test;

  /

  Thank you

  Hello

  a few notes.

  1. to_date(t2.time_id,'dd-mon-yyyy')

  New York T2.Time_ID Cis the varchar2 data type that contains values to JJ-me-YYYY format?

  And if, in which language is used for the names of the months? Conversions are point impossible to solve in reasonable time limits without logging of dml errors and unique failure on charges of staging nightly!

  
  2 single quote escaping "alternative in string literals.

  You can use Q or q to escape single quotes in strings.

  http://docs.Oracle.com/database/121/SQLRF/sql_elements003.htm#SQLRF00218

  3. validation

  Not required because the DDL commands commit implied.

  In general I recommend you write commit clauses only in calling script rather than in the code unless it connects with an autonomous transaction.

  You end up with a lot of validation of code here and there and you don't know where it is and where it isn't. That is, your process may have more than one appeal process and commit the middle of the process, it is not atomic processes.

  Here's the demo although I recommend also using the static table and truncate to efficiency. ETG is good choice for the purpose of maintenance table.

  create or replace
  package testing is
    procedure staging_one;
  end;
  /
  create or replace
  package body testing is
    --
    procedure staging_one is
      --
      procedure drop_staging_one is
        table_does_not_exist exception;
        pragma exception_init(table_does_not_exist, -00942);
      begin
        execute immediate q'{
          drop table staging_one purge
        }';
      exception when table_does_not_exist then
        return; -- fine
      end;
      --
      procedure create_staging_one is
      begin
        execute immediate q'{
  
          -- remove >>>
          create table staging_one nologging
          as
          select * from dual
          -- <<< remove
  
          /* uncomment >>>
          create table staging_one nologging
          as
          select
            substr(t1.genarea,3,3)                    as m_class,
            substr(t1.genarea,6,30)                   as m_description,
            substr(t1.genarea,36,3)                   as m_class,
            substr(t1.genarea,39,30)                  as m_description,
            substr(t1.itemitem,1,3)                   as product_code,
            t3.channel_num                            as sales_channel,
            to_date(t2.time_id,'dd-mon-yyyy')         as mis_date,
            sum(t2.ap_cw_cfi_irp+t2.ap_cw_issues_irp) as ap_gross,
            sum(t2.ap_cw_cfi_irp+t2.ap_revivals_irp)  as ap_net,
            sum(t2.sp_inc_irp   +t2.sp_issues_irp)    as sp_gross,
            sum(t2.sp_dec_irp   +t2.sp_fs_irp)        as sp_net
          from
            d_pr t1,
            act t2,
            age_map t3
          where
            t1.pfx                      = 'IT'
            and t1.coy                  = '1'
            and t1.tabl                 = 'T81'
            and substr(t1.itemitem,1,3) = t2.product_id
            and t3.age_num              = t2.age_id
          group by
            substr(t1.genarea,3,3),
            substr(t1.genarea,6,30),
            substr(t1.genarea,36,3),
            substr(t1.genarea,39,30),
            substr(t1.itemitem,1,3),
            t3.channel_num,
            to_date(t2.time_id,'dd-mon-yyyy')
          <<< uncomment */
        }';
        --
      end;
    -- main
    begin
      drop_staging_one;
      create_staging_one;
    end;
  end;
  /
  
  set serveroutput on
  
  exec testing.staging_one;
  
  select * from staging_one
  ;
  commit
  ;
  
  PACKAGE TESTING compiled
  PACKAGE BODY TESTING compiled
  anonymous block completed
  DUMMY
  -----
  X 
  
  committed.