Problem Statement:
Batch processing in Showcase will only generate HEX or GRAPHIC fields in the output table. This is causing issues with applications that are expecting certain fields to be CHAR.
Running a batch query connecting to a ShowCase server writing the results to a batch table only returns the data types of graphics or hex. As a result a column defined as character (10) will not be returned as character (10) but as data type GRAPHICS when querying the result table via ShowCase Query.
Error seen in Batch Query:
Job State: The batch job failed due to a SQL error. Character conversion between CCSID 13488 and CCSID 65535 not valid
Solution:
Configuration change within the showcase.cfg file is now required. The column CCSIDs in the batch query results table will then retain the CCSID of the original table and a data type used within valid CAST functions will now be honored in the batch results tables.
Steps:
- Configuration change required, add the following line below to the Showcase.cfg file within your Showcase Server library:
showcasecorp.batchQuery.useOriginalCCSIDs=true
- Use the command below to edit the Showcase.cfg file and insert the line mentioned above in Step 1 before the first stanza (e.g. [session] ):
edtf '/scserver/proddata/config/showcase.cfg'
(replace 'scserver' with the name of your ShowCase server library)
- The Showcase server must be restarted after this file is edited to make the change effective.
Notes:
- Enabling this function will not permit the use of Unicode source columns in a batch query.
- If the original columns in the batch query are Unicode (CCSID=13488), then this function must be disabled by either removing or commenting out the line, that was added above and then restarting the server.
- Alternately, a CAST function can be applied to the original Unicode columns to change them to a valid non-unicode CCSID.
Ref#: 1478835
Last Modified On:
No, open a new Support Case