Dear Forum,
I'm using Cognos 11.1
I've written a query that generates some text. It works without problems.
I want to join this query with another one.
However, if I want to have a look at the contents of the new table, it results in an error-message:
QE-DEF-0478 Invalid coercion from 'value' to 'string' for (some expression in my up-till-then correct first query)

I can still generate the table of this supposedly errorsome query.

Does anyone know, what this could be and what I should change?

Thanks in advance

------------------------------
Ludwig Leijten
------------------------------

#CognosAnalyticswithWatson

Hi Ludwig,

what's the data source here? I assume it's a relational data source and not a cube. What kind of data items do you use for the join condition? Have you checked their aggregation settings? Cognos may interprete them as measure and therefore as value instead of a character. Attributes should have set their aggregation to "None".
Another attempt would to change the data format of the columns with a (vendor-specific) function like cast, to_char or whatever you want the data to be - if applicable.

------------------------------
Robert Dostal
Team Leader BI
GEMÜ
Ingelfingen
------------------------------

Hi Robert,
thanks for your answer. While trying to check your suggestions I noticed a very silly mistake due to copy-paste. This caused the problem.
So, please consider this question as not posted.
Regards

------------------------------
Ludwig Leijten
------------------------------

Original Message:
Sent: Fri January 21, 2022 09:36 AM
From: Robert Dostal
Subject: Query would have error ONLY after joining it with another query

Hi Ludwig,

what's the data source here? I assume it's a relational data source and not a cube. What kind of data items do you use for the join condition? Have you checked their aggregation settings? Cognos may interprete them as measure and therefore as value instead of a character. Attributes should have set their aggregation to "None".
Another attempt would to change the data format of the columns with a (vendor-specific) function like cast, to_char or whatever you want the data to be - if applicable.

------------------------------
Robert Dostal
Team Leader BI
GEMÜ
Ingelfingen
------------------------------