What's Changed 🚀 - build(docs): Adds make docs phony target @rchowell (#3693) ✨ Features - feat: Add a new dashboard UI to Daft @raunakab (#3738) - feat(shuffles): Determination logic for pre shuffle merge @colin-ho (#3674) - feat: Limit number of sources in merged scan task @colin-ho (#3695) - feat: Expose parquet chunk size to swordfish reads @colin-ho (#3714) - feat: add LiteralValue::Int8 and Int16 @ugoa (#3736) - feat: with\_column(s)\_renamed expression for DataFrame @jessie-young (#3732) - feat: Explain for swordfish @colin-ho (#3667) - feat: Adds .describe() to DataFrame and DESCRIBE to SQL @rchowell (#3720) - feat: Add column format option to iter rows @colin-ho (#3681) - feat(core): make micropartition streamable over tables @universalmind303 (#3709) - feat(iceberg): Adds support for read\_iceberg with metadata\_location to Daft-SQL @rchowell (#3701) - feat: Overwrite partitions mode @colin-ho (#3687) - feat(docs): Adds copy-to-clipboard to code samples @rchowell (#3702) - feat(connect): sql @universalmind303 (#3696) - feat: add binary string operations (length and concatenation) @f4t4nt (#3646) - feat(sql): Adds url\_download and url\_upload to daft-sql @rchowell (#3690) - feat(connect): distinct + sort @universalmind303 (#3677) - feat(core): Implement null-safe equality operator @f4t4nt (#3663) - feat(sql): Adds JsonScanBuilder to daft-scan and read\_json to daft-sql @rchowell (#3683) - feat: support using S3Config.credentials\_provider for writes @kevinzwang (#3648) - feat(sql): Adds FROM source check for string paths @rchowell (#3679) - feat(connect): Rust ray exec @universalmind303 (#3666) 🐛 Bug Fixes - fix: Set filter selectivity estimate lower bound @colin-ho (#3694) - fix(join): joining on different types @kevinzwang (#3716) - fix: to\_cnf and to\_dnf functions @kevinzwang (#3728) - fix: pushdowns for unpivot @universalmind303 (#3724) - fix(optimizer): Fix issues with join graph construction @desmondcheongzx (#3668) - fix: Run filter null join key optimization once @colin-ho (#3657) 🚀 Performance - perf: Track accumulated selectivity in logical plan to improve probe side decisions @desmondcheongzx (#3734) - perf: simplify boolean expression rules @kevinzwang (#3731) - perf(shuffles): Incrementally retrieve metadata in reduce @colin-ho (#3545) - perf: Improve stats for join side determination @colin-ho (#3655) ♻️ Refactor - refactor: remove eyre from daft-connect @universalmind303 (#3719) - refactor(execution): NativeExecutor refactor @universalmind303 (#3689) - refactor: logical op constructor+builder boundary @kevinzwang (#3684) - refactor(connect): internal refactoring to make connect code more organized \& extensible @universalmind303 (#3680) 📖 Documentation - docs: linked mkdocs \& api docs @ccmao1130 (#3703) - docs: higher quality daft diagram for readme @ccmao1130 (#3697) - docs: add daft launcher docs to docs v2 @ccmao1130 (#3678) 👷 CI - ci: Allow upstream git refs to be used for benchmarking @desmondcheongzx (#3730) - ci: Remove daft tracing @raunakab (#3692) - ci: Add new benchmarking cluster profile @raunakab (#3665) 🔧 Maintenance - chore: Pin sql server version in docker compose @colin-ho (#3715) - chore(connect): better error propagation \& handling @universalmind303 (#3675) - chore(connect): consolidate multiple files in tests/connect @universalmind303 (#3676) Full Changelog: https://github.com/Eventual-Inc/Daft/compare/v0.4.2...v0.4.3 Release Notes: https://github.com/Eventual-Inc/Daft/releases/tag/v0.4.3