Terms
The terms
These are the words that the rest of the book uses, with the one meaning each has.
| Term | Meaning |
|---|---|
| Normative content | The fields that define what a requirement means and how it is verified. The canonical form and hashes lists them. |
| Gated requirement | A requirement with test in verification and a status of approved or required. Only gated requirements take part in the test gate. |
bytes_hash |
The hash of the bytes of a file as read in this invocation. It is used in the inputs fingerprint, in run records and by affected. |
effective_hash |
The hash of the normative fields after inheritance and defaults are resolved. It is used for verification, for staleness and in the baseline. |
| Snapshot | The frozen view of requirements, sources, annotations, selectors and configuration that one command works on. |
| Suite | A configured test command and the report it writes. |
SourceTestId |
The canonical identity of a test declaration: target::module::fn for Rust, or a namespaced opaque id for a test of another runner. See Test identities. |
| Selector | A SourceTestId written in the tests list of a requirement. |
| Logical case | One invocation group of a test in a report: its suite, class, name and parameterization, without its attempts. |
| Attempt | One execution of a logical case, with an order, an outcome and a duration. |
| Run | One invocation of rqm test. It writes a run record. |
| Run record | The file under .rqm/runs/ that says what a run did and what it proved. It is written once and never changed, and it is not meant to be committed. |
| Baseline | The file that records what the evidence established, for each requirement uuid and each effective_hash. It is an ordinary file at the configured path. Normally CI writes it and keeps it outside the repository. |
| Baseline record | The entry for one requirement in the baseline. |
| Baseline evaluation | The comparison of a baseline record with the current snapshot: current, stale, absent, invalid or disabled. |
| Run evaluation | What the latest run proved about a requirement: satisfied, incomplete, unknown or unsatisfied, or none when no run is in scope. |
| Eligibility | The conjunction of conditions that decides whether the evidence of a requirement may be promoted. |
| Presentation | The one label shown for a requirement, computed from its run evaluation and its baseline evaluation. |
| Promotion | What rqm baseline update does: it turns the evidence of one run into baseline records, and removes the records that the evidence no longer supports. |
| Boundary statement | The sentence that ends every output that shows a label: rqm shows that the evidence meets the declared policy, and does not show that the tests are adequate. |