You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Problem Statement: When a column with a NUMBER data type (without scale) is involved in a join, mismatched records (both data and record count) are not displayed on the dashboard.
Root Cause: In Oracle, a NUMBER without scale is treated as a decimal, which causes the join criteria to fail, resulting in empty outputs.
Note: By commenting the transformation rule to handle nulls in joins, we were able to display the total number of mismatch records. However, mismatch data is not shown in the detail window.
This is really to not generate false negatives during hash comparisons
select a ,b, a=b, hash(a), hash(b), hash(a)=hash(b)
from (select cast(21.00 AS decimal(38,0))as a, cast(21 AS int) as b);
The text was updated successfully, but these errors were encountered:
sundarshankar89
changed the title
[BUG]: [DSG] Number without scale in Oracle vs bigint in Databricks - Reconcilation is treating all records as "mismatch"
[BUG]: Number without scale in Oracle vs bigint in Databricks - Reconcilation is treating all records as "mismatch"
Nov 14, 2024
Is there an existing issue for this?
Category of Bug / Issue
ReconcileError
Current Behavior
Problem Statement: When a column with a NUMBER data type (without scale) is involved in a join, mismatched records (both data and record count) are not displayed on the dashboard.
Root Cause: In Oracle, a NUMBER without scale is treated as a decimal, which causes the join criteria to fail, resulting in empty outputs.
Note: By commenting the transformation rule to handle nulls in joins, we were able to display the total number of mismatch records. However, mismatch data is not shown in the detail window.
This is really to not generate false negatives during hash comparisons
The text was updated successfully, but these errors were encountered: