Skip to content

Plan, apply and roll back changes to a production ontology, with a blast radius report and a proof an auditor can re-check.

The Unreleased section has one Added and one Fixed heading

FABIOTESSc0416cf49d03
diff --git a/CHANGELOG.md b/CHANGELOG.md
index b2b3e861..69f6a354 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -4,6 +4,24 @@ All notable changes to Open Ontologies are documented here.
## [Unreleased]
+### Added
+
+- **`onto_shacl` and `shacl` write a W3C `sh:ValidationReport` graph**
+ (#318, reported by @noahgift). `report_format: "turtle" | "ntriples"`
+ (`--report-format` on the CLI) returns the report as RDF, carrying
+ `sh:conforms` and, per result, `sh:focusNode`, `sh:resultPath`, `sh:value`,
+ `sh:sourceShape`, `sh:sourceConstraintComponent`, `sh:resultSeverity` and
+ `sh:resultMessage`; a custom severity IRI and a language-tagged `sh:message`
+ are written as declared. An undetermined run (`conforms: null`) is written
+ `sh:conforms false` plus `oo:undetermined true` and its reason, never
+ `sh:conforms true`, and keeps the results it found. The CLI exit code is
+ still taken from the verdict (0, 1, 2) in every format. JSON stays the
+ default and is unchanged apart from `value`. Measured against the W3C suite:
+ 48 of the 56 `core/` cases that pass today produce a report matching the
+ expected one after RDFC-1.0 canonicalisation; the 8 that differ are named in
+ `tests/shacl_validation_report_w3c_test.rs` with their causes, which include
+ an `sh:or` false negative on IRI and ill-typed values.
+
### Fixed
- **`diff` and `drift` no longer hang on a symmetric blank-node graph** (#293).
@@ -104,25 +122,6 @@ All notable changes to Open Ontologies are documented here.
holds each as FAIL. `tools/fol_differential.py`
(every goal undetermined) and `tools/horn_differential.py` (every case not
run) had the same exit-0-on-nothing and now exit 2.
-### Added
-
-- **`onto_shacl` and `shacl` write a W3C `sh:ValidationReport` graph**
- (#318, reported by @noahgift). `report_format: "turtle" | "ntriples"`
- (`--report-format` on the CLI) returns the report as RDF, carrying
- `sh:conforms` and, per result, `sh:focusNode`, `sh:resultPath`, `sh:value`,
- `sh:sourceShape`, `sh:sourceConstraintComponent`, `sh:resultSeverity` and
- `sh:resultMessage`; a custom severity IRI and a language-tagged `sh:message`
- are written as declared. An undetermined run (`conforms: null`) is written
- `sh:conforms false` plus `oo:undetermined true` and its reason, never
- `sh:conforms true`, and keeps the results it found. The CLI exit code is
- still taken from the verdict (0, 1, 2) in every format. JSON stays the
- default and is unchanged apart from `value`. Measured against the W3C suite:
- 48 of the 56 `core/` cases that pass today produce a report matching the
- expected one after RDFC-1.0 canonicalisation; the 8 that differ are named in
- `tests/shacl_validation_report_w3c_test.rs` with their causes, which include
- an `sh:or` false negative on IRI and ill-typed values.
-
-### Fixed
- **Default-path SHACL results now carry the value node** (#318, reported by
@noahgift). The `sh:class`, `sh:datatype`, `sh:nodeKind`, `sh:pattern`,