Releases: ray-project/deltacat
Releases · ray-project/deltacat
DeltaCAT 2.0.0.post4
Release Notes
Post-release feature enhancements to improve write performance for DeltaCAT 2.0 tables that don't require compaction.
New Features:
- Support
TableReadOptimizationLevel.NONEfor tables without Merge Keys (#576)
PyPi: https://pypi.org/project/deltacat/2.0.0.post4/
Full Changelog: 2.0.0.post3...2.0.0.post4
DeltaCAT 2.0.0.post3
Release Notes
Post-release bug fixes and feature enhancements for DeltaCAT 2.0.
New Features:
- Unordered Add Write Mode Support for Conflict-Free Concurrent Table Writes (#575)
- Portable Manifests for Catalog Sharing (#575)
- Partitioned Transaction Logs and Metafile Revisions to Improve Cloud Metadata Performance at Scale (#575)
- DeltaCAT Catalog Writer Version Tracking via Version Files (#575)
- Catalog Config File Support (#569)
Bug Fixes:
- Concurrent Namespace Creation Conflicts During Table Writes (#575)
- Catalog Re-Initialization in Every DeltaCAT Catalog API Call (#500, #575)
- Iceberg Equality Delete Converter Concurrent Write Conflict Resolution (#568)
PyPi: https://pypi.org/project/deltacat/2.0.0.post3/
Full Changelog: 2.0.0.post2...2.0.0.post3
DeltaCAT 2.0.0.post2
Release Notes
Post-release bug fixes for DeltaCAT 2.0.
Bug Fixes:
- Fixed Ray Data and Daft Infinite Retry Loop Issue when Writing to Cloud Storage (#574)
- Fixed Bug Causing Ray Data, Pandas, NumPy, and Polars to return None/NaN values for new fields following schema evolution in some write modes (#573)
- Improved support for external multimodal URL processing (#573)
Full Changelog: 2.0.0.post1...2.0.0.post2
DeltaCAT 2.0
Release Notes
Initial implementation of core DeltaCAT 2.0 catalog APIs for Daft, Ray Data, Pandas, PyArrow, NumPy, and Polars.
Among other features, it provides:
- Inline copy-on-write table compaction and table properties to control automated compaction.
- Automatic/manual schema evolution support, and table properties to control table schema evolution behavior.
- Support for writing/reading both schemaless tables and tables with schemas.
- Full cross-catalog, recursive metadata copy and backfill support (e.g., to support easily backfilling major revisions to catalog metadata storage specification).
- Frontpage "overview"/"quickstart" documentation and more detailed Storage, Table, and Schema README doc pages.
- Multi-table/namespace/etc. transaction support (i.e., transactions that can operate over any number of objects within the bounds of a single catalog).
- Comprehensive, auto-generated (via new make type-mappings makefile target) reader/writer support matrix in reader_compatibility_mapping.py across all Arrow data types, supported dataset types (PyArrow, Pandas, Polars, NumPy, Daft, Ray Data), and supported content types with inline schema (Parquet, Avro, Orc, Feather). This allows us to quickly detect and short-circuit any write that would break a declared supported reader before persisting data or doing any computationally expensive work.
- Transaction log queries and time travel.
Full Changelog: 2.0.0b11...2.0.0.post1
1.1.38
2.0.0b11
Changes
bump daft version to 0.4.15
2.0.0b10
bump deltacat version to 2.0.0b9 (#544)
Changes
- bump daft version to 0.4.13
2.0.0b8: chore: upgrade daft version (#543)
Changes
- bump daft version to 0.4.13 which is compatible with AL2
2.0.0b7: add missing `__init__` file under `deltacat.daft` module (#541)
- add missing
__init__file fordeltacat.daftmodule