Skip to main content
Announcements
July 15, NEW Customer Portal: Initial launch will improve how you submit Support Cases. IMPORTANT DETAILS
cancel
Showing results for 
Search instead for 
Did you mean: 
Shako
Partner - Contributor III
Partner - Contributor III

Incremental Load By String, NVARCHAR, Text Field

Hi,

Is there way to implement incremental load without having Date field and Key column is NVARCHAR. Database is SQL 

I want to Load only New Rows. No update or Delete is Required

The field example:

DocNo
PSI16_0000001
PSI16_0000002
PSI16_0001003
Labels (4)
1 Solution

Accepted Solutions
rubenmarin

At the end the question is, can you identify in some way which records are new based on the data previosly loaded?. If you can do it manually then there is an option to do it automatic,if there is no way you can do it only with the last DocNo then there is not possible.

If the initial character have a limited combimnations you can do a bucle to load each prefix separatedly. It it will be more dinamic, you can also recognize them while loading the data, do a list, a bucle based ont hat list... but maybe all this extra process makes the incremental load take more time than anormal load.

View solution in original post

3 Replies
rubenmarin

Hi, that field looks like a number, if it's always increasing that number you can use it to do the incremental load.

Loading from qvd the max value will be easy, using Right() or Subfield() to retrieve the higuer value loaded.

A Where condition should be applied on sql using that value, syntax will depend on the language used by the database, it could be something like: Where Right(DocNo,7)>=$(vHigherValueFromQvd)

Shako
Partner - Contributor III
Partner - Contributor III
Author

I Checked key field and it may change not only numerically , for example:

PMI16_0000001

PSI16_0000001

rubenmarin

At the end the question is, can you identify in some way which records are new based on the data previosly loaded?. If you can do it manually then there is an option to do it automatic,if there is no way you can do it only with the last DocNo then there is not possible.

If the initial character have a limited combimnations you can do a bucle to load each prefix separatedly. It it will be more dinamic, you can also recognize them while loading the data, do a list, a bucle based ont hat list... but maybe all this extra process makes the incremental load take more time than anormal load.