Classification filters
Classification filters are used inside relational classifications to remove rows that should not be considered during evaluation.
In simple terms:
A classification filter asks: Should this row continue in the evaluation, or should it be removed?
Why classification filters exist
When relational classifications follow relations between tables, they often collect many rows.
Sometimes only a subset of those rows should be counted.
Classification filters allow the system to keep only rows that satisfy a specific condition.
Example:
A booking might only count rooms that are active.
Relation path:
Filter:
Only rooms where Active room = true will remain in the evaluation.
How classification filters work
A classification filter is applied at a specific step in a relation path.
At that step the system:
- Follows the relation and collects rows
- Evaluates the referenced classification for each row
- Keeps or removes rows depending on the filter rule
- Passes the remaining rows to the next step
Example:
| Room | Active room |
|---|---|
| Room 101 | true |
| Room 102 | false |
If the filter is Include Active room, the result becomes:
| Room |
|---|
| Room 101 |
Include vs exclude
Classification filters support two modes.
Include
Only rows where the classification is true remain.
Example:
Result:
Only active rooms remain.
Exclude
Rows where the classification is true are removed.
Example:
Result:
Rooms marked as maintenance are removed from the set.
Multiple filters
Several classification filters can be applied at the same step.
In that case all filters must pass for a row to remain.
Example:
A room must satisfy both rules to stay in the result set.
Example
Tables
- Booking
- Room
Relation
Classification filter
Evaluation:
| Booking | Rooms found | Active rooms counted |
|---|---|---|
| B001 | Room101, Room102 | Room101 |
| B002 | Room103 | Room103 |
Only the rows that pass the filter are counted in the final classification result.
A simple way to think about classification filters
A classification filter answers the question:
Which of the related rows should actually be considered?
By reusing existing classifications, filters make relational logic easier to reuse and easier to maintain.
Related resources
Reference
Related concepts