v0.16.4

🧘 QoL Improvements

• Add new configurable env var AIRBYTE_TEMP_FILE_CLEANUP (#336)
• Process cache streams using deterministic (alpha-sorted) ordering (#335)

v0.16.3

🐛 Fixes

• Improve log file handling for connectors (#333)
• Fix spammy logs from grpcio (temporarily forces downgrade of grpcio) (#334)

v0.16.2

🐛 Fixes

• Add handling for new "generations" feature: default sync IDs, generation IDs, and min generation IDs (#330)
• Resolve issue where stream success statuses were not sent by PyAirbyte when syncing from the cache to a destination (#330)

🛠️ Internal Tooling

• Add script for PyAirbyte performance profiling option without caching (#323)

v0.16.1

This release makes destinations easier to use by defaulting to Docker-based executors when no other execution or installation method is provided.

▶️ Changes

• Default to docker-based destination executors (#322)

v0.16.0

✨ Feature Improvements

• Add destinations support (#257)
• More readable and more informative progress display during sync operations (#302)

⚡ Performance Improvements

• Multiple performance improvements including faster Snowflake loads and faster per-record processing (#298, #301)

📖 Documentation Improvements

• Clean up readme and module docs (#316)
• Expose config classes in cache modules (#315)
• Add destinations API reference docs (#314)
• Improve docstrings and automated API ref docs (#313)

⚙️ Under the Hood

• Feat: Replace ULID raw record ID with UUID v7 (#300)
• Added performance benchmarking script and viztracer performance profiling for contributors (#298)
• Misc code hygiene and cleanup (#317)

v0.15.0

✨ New Features

• Sources based on Docker have been promoted out of experimental status. 🎉 (#294)
• Sources based on No-Code YAML Manifests have been promoted out of experimental status. 🎉 (#294)
• The get_source() args now supports docker_image and source_manifest without importing from the airbyte.experimental module. (#294)

v0.14.0

PyAirbyte v0.14.0 - The Hackathon Release! 🤖 🚀

This is a very special release of PyAirbyte - all contributions were provided by our community Hackathon participants! A very big thanks to everyone who participated in the Hackathon! 🙏

✨ New Feature

• Adds Apache Arrow interface with new methods Dataset.to_arrow() and Cache.get_arrow_dataset() (#281) - Thanks, @avirajsingh7!

💪 Quality of Life Improvements

• Print message when syncing incremental streams, provides better context to "0 records" output message when re-running incremental streams in Source.read() (#282) - Thanks, @avirajsingh7!
• Adds log message whenever a new stream starts in Source.read() (#288) - Thanks, @vspanxcode!
• Instead of failing, raise a warning if stream selection is performed before config is set (#285) - Thanks, @Suraj-Vishwakarma70!

v0.13.0

Changed

• Upgrade to Pydantic 2.0 and Airbyte CDK 2.0 (#291)

v0.12.1

✨ New Features

• Added use_host_network param to get_source(), allowing Docker containers to use host's network (#277) - Thanks, @niyasrad!

⚙️ Under the Hood

• Update low-code registry overrides to reflex latest connectors lists (#278)

v0.12.0

✨ New Features

• Add install_root argument for get_source(), allowing users to override their virtualenv base path (#268) - Thanks, @tqtensor!
• Add experimental implementation for running no-code sources (declarative YAML definitions) using get_source(). The YAML source definitions can be provided inline or automatically downloaded, with no need for a separate installation of the source connector. (#175)
• Add capability to disable the Rich "live view" progress updates with NO_LIVE_PROGRESS env variable (#271)

🐛 Fixes

• Add back default schema name 'airbyte_raw' in PyAirbyte caches (#259)

😮 Other Changes

• Update DuckDB engine from 0.9 to new (stable) v1.x version (#269)