| Checking duplicates with checksum before actual inserts into a table [message #93280] |
Mon, 03 May 2004 14:27 |
Swamy
Messages: 78
Registered: June 2002
|
Member |
|
|
We have a data loading issue related to duplicates and non-duplicates when there are no definite fields to be included in a unique constraint. I am using a surrogate key, which acts as a primary key with a database generated sequence number. Some how I heard about a routine called “checksum” that can be used for checking the database. Some one states as follows.
You would need to store the checksum for each record received (using whatever algorithm you decide to use). This field is considered part of the 'key' for the record. When a true duplicate is received, the checksum as well as the other fields in the key will be the same and the record should be deleted. You would just have to calculate the check sum for each record before you try to store it. You don't have to search the entire database. if the compound key value already exists, this is a duplicate.
Anyone heard about using checksum on Oracle databases? If so could you please help me, how can I use. I appreciate your help in advance.
Thanks
Swamy.
|
|
|
|
] 
Blog
Search
Help
Members
Register
Login
Home
