Write rules
Write rules determine whether data can be:
- created
- updated
- deleted
They are evaluated whenever a write operation is attempted.
How evaluation works
- Classification is evaluated
- Rule checks When True
- If applicable → produces feedback
| Classification | When True | Rule applies |
|---|---|---|
| true | checked | yes |
| true | unchecked | no |
| false | checked | no |
| false | unchecked | yes |
Severity levels
Write rules produce feedback with different severity levels.
The only functional difference is:
- Error → blocks saving
- All other levels → allow saving
| Severity | Effect |
|---|---|
| Error | Blocks saving |
| Warning | Allows saving (highlighted) |
| Info | Allows saving (informational) |
| Confirmation | Allows saving (requires acknowledgment) |
The non-error levels differ only in how the message is presented in the UI (e.g. color and emphasis), not in behavior.
Violation model
Write rules only block new violations introduced by the current change.
This ensures:
- existing invalid data does not block edits
- users can fix data incrementally
- no new invalid states are introduced
Import behavior
Write rules are not evaluated during imports, allowing data migration without operational constraints.
Related resources
Understand the concept