Data constraints complete

The check is warning severity. It runs at verify. A counterexample reports fail at warning. A parse miss, a timeout, or an unrecognised solver kind reports warn. Neither blocks advancement. Read the class tag on the finding first. They have different owners. Do not treat them as one bucket.

The tags are [disproved] (the solver produced a counterexample: close the gap or the overlap), [not-parsed] (Proof could not read the expression, so no proof was attempted), [inconclusive] (timeout or resource limit), and [unclassified] (an unrecognised solver result, printed verbatim). A parse miss is a tool gap. See proof help spec_lint_data_constraint_parseable.

Authoring is a YAML block, not a column CHECK:

variables:
  - name: amount_low
    type: bool
    direction: input
    data_constraint:
      domain: integer
      condition: "amount <= 100"
      parameters:
        - name: amount
          type: int
  - name: amount_high
    type: bool
    direction: input
    data_constraint:
      domain: integer
      condition: "amount >= 100"
      parameters:
        - name: amount
          type: int

Both booleans are true at amount = 100. Completeness can still pass. Exclusivity cannot. The hop names that overlap. It does not rewrite the conditions. It does not inspect the Go. See variable drift for list vs sentence on one requirement. This hop is whether the authored cases are disjoint and complete. See Z3/Kind2 on one function for a lemma on the helper itself.

proof verify-properties specs/system
proof audit --check data_constraints_complete --verbose
proof help spec_lint_data_constraint_parseable

Tighten the conditions until the cases are disjoint and complete. Add or strengthen parameter constraints when the domain itself is underspecified. Do not drop a boolean so the hop goes green. A missing case is still a missing case.

The teaching graph is still one group next to one domain. Read the class tag. Then pick one of four resolutions. They are alternatives.

proof verify-properties specs/system
proof audit --check data_constraints_complete --verbose
proof audit --check data_constraint_z3_coverage --verbose
proof help spec_lint_data_constraint_parseable

Close a [disproved] by tightening conditions until the cases are disjoint and complete. Route a [not-parsed] to the parse lint; do not treat it as a domain gap. Retry or bound an [inconclusive]. Do not drop a skipped variable so the hop goes green. The skip list stays on data_constraint_z3_coverage, a different hop. Jama still authors. Proof vs Jama.

Warning severity either way. A counterexample fails at warning. A parse miss, a timeout, or an unrecognised kind warns. Neither blocks advancement unless you raise the fail level. The hop does not rewrite condition:. It does not inspect the implementation. It does not prove the helper clamps. Move that claim to a lemma. We have not scored this partition against a frozen Jama pack, Postgres CHECK, or a Kind2 model of the Go. The loss is named, not scored.

The list vs sentence hop stays on variable drift. The function proof stays on Z3/Kind2 on one function. The engagement stays on software correctness audit. Jama still authors.

What is data constraints complete? Same question. Same URL.

Is this a SQL CHECK constraint? No. That SERP is a column the database enforces at write. The command on this install is proof audit --check data_constraints_complete.

Is this Z3 on one function? No. That hop is a lemma on a helper. This hop is whether the authored boolean cases partition the domain. See Z3/Kind2 on one function.

Is this variable drift? No. Drift is list vs sentence. This hop is completeness and exclusivity of the cases you already named. See variable drift.

Does a disproved partition fail the merge? No. The check keeps warning severity. A counterexample reports fail at warning.

Is a parse miss a domain gap? No. [not-parsed] means Proof could not read the expression. Route it to proof help spec_lint_data_constraint_parseable.

Is Proof a Jama alternative for the partition? No. Jama still authors. Proof vs Jama.