Loading data-on-demand gives ERROR: value too long for type character varying(255)

[Chris Pudney]

G'day,

I'm using Spotfire Analyst v12.x

I can successfully load data from an information link (Postgres) into a data table. However, if I attempt to load the same information using data-on-demand I get the following error:

com.spotfire.ws.dat.DatException: com.spotfire.ws.dat.OperatorException: com.spotfire.ws.im.IMException: Failed to execute query: Batch entry 93 INSERT INTO SFTMP75520432991 VALUES (' Multi-pass membrane protein <LOTS OF TEXT SNIPPED> These foci are often found next to mtDNA nucleoids.') was aborted: ERROR: value too long for type character varying(255)  

It seems a temporary table is used to handle on-demand loading but the column in the temporary table has width 255.

Is this a known problem and if so is there a work-around.

Thanks,

Chris.

[Jose Leviaguirre]

Please raise a ticket at support.tibco.com

[Chris Pudney]

We resolved this issue by modifying the java-to-sql-type-conversions block of our PostgreSQL data source template as follows:

<java-to-sql-type-conversions> <type-mapping> <from>String</from> <to>TEXT</to> </type-mapping> </java-to-sql-type-conversions></jdbc-type-settings>

This was followed by a restart of all nodes.

Ref. https://docs.tibco.com/pub/spotfire_server/7.9.0/doc/html/TIB_sfire_server_tsas_admin_help/GUID-A286669E-F94A-4A65-9041-55F4137EDD5D.html