0.3.1

Changed

• Minimum cognite-pygen-spark version raised to 0.3.1 (aligns with pygen-spark security patch release)
• pyasn1 minimum raised to 0.6.3 (CVE-2026-23490)
• uv.lock security updates: cryptography 46.0.7, protobuf 6.33.6, pyjwt 2.12.1, urllib3 2.7.0, idna 3.15, and dev tooling (pytest, black, pygments, filelock, virtualenv)

Improved

• README notes that patch releases may update locked dependencies without API changes

0.3.0

Added

• generate_udtf_notebook passes a Databricks-specific CDF audit deployment tail into SparkUDTFGenerator (module constant _DATABRICKS_AUDIT_TAIL)
• Dependency floor cognite-pygen-spark>=0.3.0 for generated UDTF HTTP client audit headers (x-cdp-app, x-cdp-sdk, User-Agent)

Changed

• README: Restructured Latest Release as a bullet list; removed duplicate Notebook-Friendly API entry under high-level APIs
• Pre-commit / lockfile: mypy hook and uv.lock aligned with cognite-pygen-spark 0.3.0

0.2.4

Changed

• Require cognite-pygen-spark>=0.2.4 so consumers resolve TypeConverter-based UDTF field typing, SparkValueKind / value_kind normalization, and Arrow-friendly timestamp handling from generated UDTFs

Improved

• README links to GitHub Releases for full version history

0.2.3

Fixed

• View SQL and _parse_udtf_params now use the same safe field names as pygen-spark UDTFField for properties that clash with Python reserved words, while preserving CDF property keys for filters and foreign-key handling.

Changed

• cognite-pygen-spark dependency raised to >=0.2.3.
• Mypy pre-commit hook installs cognite-pygen-spark>=0.2.3 so type-checking matches the updated UDTFField.from_property(..., view_id) API.

0.2.2

Fixed

• PyPI Documentation URL now points to https://cognite-cognite-databricks.readthedocs-hosted.com/en/latest/ instead of cognite-databricks.readthedocs.io

0.2.1

Added

• SQL-native time series UDTF support with predicate pushdown hints (time_series_sql_udtf)
• SQL query analyzer (SQLQueryAnalyzer) for extracting pushdown hints from SQL queries
• Support for aggregate hints (avg, sum, min, max, count) in SQL-native UDTF
• Granularity hints for time-based grouping in SQL-native UDTF
• generate_time_series_sql_view() function for creating SQL views with pushdown hints
• Documentation for SQL-native time series UDTF usage

Improved

• Updated dependency on cognite-pygen-spark to >=0.2.2
• Enhanced time series UDTF support with SQL-native template integration
• Better predicate pushdown support for time series queries

0.2.0

Improved

• Unified UDTF registration to use file-based parsing for all UDTF types (Data Model and Time Series UDTFs)
• Improved consistency in UDTF registration workflow across different UDTF types

0.1.0

Added

• Initial release of cognite-databricks
• Unity Catalog UDTF and View registration (register_udtfs_and_views()) with DBR 18.1+ requirement
• Session-scoped UDTF registration (register_session_scoped_udtfs()) for all DBR versions
• Secret Manager integration for OAuth2 credentials
• View generation with Secret injection
• DBR 18.1+ version detection and validation with multiple fallback methods
• Template-generated time series UDTFs (datapoints, latest datapoints, long format)
• Comprehensive error handling and troubleshooting
• Full type hints and IDE support
• Type conversion utilities using PySpark as source of truth
• Connection configuration management from TOML/YAML files

Improved

• Robust DBR version detection using multiple fallback methods (Spark conf, SQL query, environment variable)
• Enhanced error messages for pre-DBR 18.1 compatibility
• Comprehensive documentation and examples
• Clear guidance for pre-DBR 18.1 users to use session-scoped registration