DQ sql is different from the Report SQL

[dinakarmu]

Hello,
In reality, the Data Quality(DQ) SQL failure report is much more beyond the DQ SQL itself. Let me take a simple example to explain:
Table1 EMP(EMP_ID, ENAME, MGR_ID, EMP_START_DT, EMP_END_DT)
Table2 SALARY(EMP_ID, SALARY_NO, CHANGE_DT, BONUS)

DQ SQL: select count(1) from SALARY where BONUS > 100000;

If the count above is greater than 0, then the DQ Report email or Jira ticket should contain this report from multiple tables:
EMP_ID, BONUS, MGR_ID, MGR_ENAME, MGR_EMAIL, MGR_ADDRESS

How do we setup a report SQL that is different from its DQ SQL, does the Soda core engine support this functionality today?
The Report SQL will use the primary keys returned by the DQ SQL to go against the dependent tables to construct the data for the Report. Please advice.

[tools-soda]

SAS-3028

[benjamin-pirotte]

Hi! Do I get it correctly that you would want to have the ability to provide an SQL query to get the failed rows which is independent of the SQL Soda Core used to generate the metric?

If so, we have a plan to add this capability to the user-defined checks (https://docs.soda.io/soda-cl/user-defined.html).

[dinakarmu]

Yes, that’s correct. Using the failed rows, we need an ability to generate a report containing more attributes from tables not used in the initial Soda core Data Quality SQL. Thank you!