Repository navigation
Releases
Releases · ludovicschmetz-stack/datavow
v0.4.0 — Airflow Operator
What's New
- DataVowOperator — run data contract validation as an Airflow task
- Install:
pip install datavow[airflow](requiresapache-airflow>=2.7) on_failuremodes:fail(raise AirflowException),warn(log warning),skip(AirflowSkipException)fail_onthresholds:strained(score < 95),broken(score < 80),shattered(score < 50)- Lazy imports in
execute()— scheduler node doesn't need datavow installed (K8s executor compatible) - XCom outputs:
vow_score,vow_verdict,violations_critical,violations_warning,violations_info,contract_name,report_path - 27 new tests (164 total), all passing
Refs: D026, D040 override for FundsDLT mission
v0.3.1
Remove cloud warehouse extras — use datavow dbt sync for Snowflake/BigQuery/Redshift/Databricks validation via your existing dbt adapter.
v0.3.0 — ODCS v3.1 native support
What's new
datavow odcs check— validate contracts against official ODCS v3.1.0 JSON Schemadatavow odcs convert— convert ODCS-native contracts to DataVow format- Auto-detection —
define,validate,ciaccept both DataVow and ODCS-native formats - Bundled official ODCS v3.1.0 JSON Schema
- New dependency:
jsonschema>=4.20 - 20 new tests (137 total, 0 regressions)
v0.2.3 — dbt 1.11 compat + dbt ci
Added
- datavow dbt ci — full pipeline: sync → dbt test → Vow Score
Fixed
- dbt 1.11 deprecation warning: generic test arguments now use nested format
v0.2.2 — dbt ci pipeline
Added
- datavow dbt ci — full pipeline: sync → dbt test → Vow Score
- Two modes: dbt-test (all warehouses) and direct (PostgreSQL/DuckDB)
- --report flag for HTML report generation
- dbt on-run-end hook (datavow-dbt package)
v0.2.1 — dbt sync
Added
- datavow dbt sync — generates dbt-native tests from DataVow contracts
- Works with ANY dbt adapter (Snowflake, BigQuery, Redshift, SQL Server, etc.)
- dbt test --select tag:datavow to run generated tests
- 19 new tests
See CHANGELOG for details.
v0.2.0 — dbt integration
Added
datavow dbt generate— auto-generate contracts from dbt manifest.jsondatavow dbt validate— validate dbt models against contracts via warehouse- PostgreSQL + DuckDB connectors (zero extra deps via DuckDB ATTACH)
- dbt tests (not_null, unique, accepted_values) auto-mapped to DataVow rules
- Optional extras:
pip install datavow[snowflake],[bigquery],[redshift],[sqlserver] - 21 new tests
See CHANGELOG for details.
v0.1.0 — A solemn vow on your data
Changelog
All notable changes to DataVow will be documented in this file.
The format is based on Keep a Changelog,
and this project adheres to Semantic Versioning.
0.1.0 - 2026-03-06
Added
- CLI commands:
init,define,validate,report,ci - Contract format: YAML-based, ODCS v3.1 superset with metadata, schema, quality rules, SLA
- Validation engine: DuckDB-based validation for schema, quality, and freshness checks
- Rule types:
not_null,unique,row_count,range,pattern,sql,freshness - Severity levels: CRITICAL (blocks), WARNING (alerts), INFO (logs)
- Vow Score: 0-100 scoring with verdicts (Kept / Strained / Broken / Shattered)
- Reports: HTML (self-contained, branded) and Markdown output
- CI mode:
datavow ciwith exit code 0/1 for pipeline integration - Data sources: CSV, Parquet, JSON, JSONL, TSV via DuckDB
- GitHub Actions CI: automated tests on Python 3.12 and 3.13