how to capture the error when RID value not in the table (pl/sql question)

From: Guang Mei <GMei_at_ph.com>
Date: Wed, 23 Feb 2005 13:22:57 -0500
Message-ID: <7E62D965D357694C993D4F0E13B3C39B04D37EFD@phexchange.ph.com>

I have the following code in a pl/sql procedure (oracle 9i), which is called a lot.

• RID is PK column on table Customers
• p_name is a passed-in parameter
• p_CustomerRID is a passed-in parameter

   SAVEPOINT sp123;

    UPDATE Customers
    SET Name = p_name
    WHERE RID = p_CustomerRID;

   IF SQL%ROWCOUNT > 1 THEN

      ROLLBACK WORK TO SAVEPOINT sp123;
      RaiseError.TooManyUpdates('Customers');
   ELSIF SQL%ROWCOUNT = 0 THEN
      RaiseError.NoRecordUpdated('Customers');

   END IF; The code works fine now. But I would to change the update sql to

      UPDATE Customers
      SET Name = p_name
      WHERE RID = p_CustomerRID
      AND Name != p_name;

to avoid unnecessary redo writing. But if I check SQL%ROWCOUNT for this new query and get 0, I would not know if it is due to a non-exist RID (Which I want to capture this situation) or simply p_name already exists for that RID. Is there a better way of doing this kind of thing, minimize redo writing and capture the error the same time with only one sql run?

TIA. Guang

---

PRIVILEGED AND CONFIDENTIAL:
This communication, including attachments, is for the exclusive use of addressee and may contain proprietary, confidential and/or privileged information. If you are not the intended recipient, any use, copying, disclosure, dissemination or distribution is strictly prohibited. If you are not the intended recipient, please notify the sender immediately by return e-mail, delete this communication and destroy all copies.

---

--
http://www.freelists.org/webpage/oracle-l

Received on Wed Feb 23 2005 - 13:22:17 CST