Japanese column names are not displayed correctly in virtualized data in Watson Query

Japanese column names are not displayed correctly in virtualized data

If Japanese column headers are not displayed properly when you virtualize tables in Watson Query, you can change the allownonalphanumeric setting to fix the display of the column headers.

Symptoms

In Watson Query on Cloud Pak for Data, when you virtualize JSON or CSV files that have Japanese data, on IBM® Cloud Object Storage, the Japanese column names might be translated to hex values.

Resolving the problem

When you virtualize JSON or CSV files that have Japanese data, on IBM Cloud Object Storage, and the Japanese column names are not displayed correctly, you can use the allownonalphanumeric option to view the virtualized Japanese column headers properly. This option is disabled by default and you must enable it.

To enable the allownonalphanumeric option, complete the following steps:

Access the head pod in the Watson Query instance (c-db2u-dv-db2u-0).

Run the following command to edit the configuration to include the allownonalphanumeric option:

db2uctl adm bigsql config --key bigsql.catalog.identifier.mappingrules --value allownonalphanumeric,allowleadingdigits,allowreservedwords

Run the following command to restart Big SQL:

su - db2inst1 
bigsql stop ; 
bigsql start

You can now preview or virtualize Japanese data on IBM Cloud Object Storage with the Japanese column names displayed correctly.