From 0cc9b0a513e784597bc6e35f883fefa9e2d3210b Mon Sep 17 00:00:00 2001 From: Tim Saucer Date: Wed, 28 May 2025 14:34:19 -0400 Subject: [PATCH 001/220] Release DataFusion 47.0.0 (#1130) * Update cargo lock for ffi example * Running cargo update * Bump tokio to 1.45 * Bump object store * Bump astral * Prepare for 47 release * Minor: clippy update warning --- .github/workflows/build.yml | 8 +- .github/workflows/docs.yaml | 2 +- .github/workflows/test.yaml | 2 +- Cargo.lock | 890 ++++++++++++--------- Cargo.toml | 6 +- dev/changelog/47.0.0.md | 64 ++ dev/release/README.md | 20 +- examples/datafusion-ffi-example/Cargo.lock | 796 +++++++++--------- src/dataframe.rs | 4 +- 9 files changed, 994 insertions(+), 798 deletions(-) create mode 100644 dev/changelog/47.0.0.md diff --git a/.github/workflows/build.yml b/.github/workflows/build.yml index acabad3ca..61896e43d 100644 --- a/.github/workflows/build.yml +++ b/.github/workflows/build.yml @@ -33,7 +33,7 @@ jobs: with: python-version: "3.12" - - uses: astral-sh/setup-uv@v5 + - uses: astral-sh/setup-uv@v6 with: enable-cache: true @@ -52,7 +52,7 @@ jobs: runs-on: ubuntu-latest steps: - uses: actions/checkout@v4 - - uses: astral-sh/setup-uv@v5 + - uses: astral-sh/setup-uv@v6 with: enable-cache: true @@ -94,7 +94,7 @@ jobs: version: "27.4" repo-token: ${{ secrets.GITHUB_TOKEN }} - - uses: astral-sh/setup-uv@v5 + - uses: astral-sh/setup-uv@v6 with: enable-cache: true @@ -150,7 +150,7 @@ jobs: version: "27.4" repo-token: ${{ secrets.GITHUB_TOKEN }} - - uses: astral-sh/setup-uv@v5 + - uses: astral-sh/setup-uv@v6 with: enable-cache: true diff --git a/.github/workflows/docs.yaml b/.github/workflows/docs.yaml index 9037e0a5c..9341488a0 100644 --- a/.github/workflows/docs.yaml +++ b/.github/workflows/docs.yaml @@ -58,7 +58,7 @@ jobs: repo-token: ${{ secrets.GITHUB_TOKEN }} - name: Install dependencies and build - uses: astral-sh/setup-uv@v5 + uses: astral-sh/setup-uv@v6 with: enable-cache: true diff --git a/.github/workflows/test.yaml b/.github/workflows/test.yaml index eac765c4d..31c2febea 100644 --- a/.github/workflows/test.yaml +++ b/.github/workflows/test.yaml @@ -76,7 +76,7 @@ jobs: run: cargo clippy --all-targets --all-features -- -D clippy::all -D warnings -A clippy::redundant_closure - name: Install dependencies and build - uses: astral-sh/setup-uv@v5 + uses: astral-sh/setup-uv@v6 with: enable-cache: true diff --git a/Cargo.lock b/Cargo.lock index b32d19d4d..39489ed94 100644 --- a/Cargo.lock +++ b/Cargo.lock @@ -73,13 +73,13 @@ checksum = "aae1277d39aeec15cb388266ecc24b11c80469deae6067e17a1a7aa9e5c1f234" [[package]] name = "ahash" -version = "0.8.11" +version = "0.8.12" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e89da841a80418a9b391ebaea17f5c112ffaaa96f621d2c285b5174da76b9011" +checksum = "5a15f179cd60c4584b8a8c596927aadc462e27f2ca70c04e0071964a73ba7a75" dependencies = [ "cfg-if", "const-random", - "getrandom 0.2.15", + "getrandom 0.3.3", "once_cell", "version_check", "zerocopy", @@ -132,9 +132,9 @@ dependencies = [ [[package]] name = "anyhow" -version = "1.0.95" +version = "1.0.98" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "34ac096ce696dc2fcabef30516bb13c0a68a11d30131d3df6f04711467681b04" +checksum = "e16d2d3311acee920a9eb8d33b8cbc1787ce4a264e85f964c2404b969bdcd487" [[package]] name = "apache-avro" @@ -150,7 +150,7 @@ dependencies = [ "log", "num-bigint", "quad-rand", - "rand", + "rand 0.8.5", "regex-lite", "serde", "serde_bytes", @@ -179,9 +179,9 @@ checksum = "7c02d123df017efcdfbd739ef81735b36c5ba83ec3c59c80a9d7ecc718f92e50" [[package]] name = "arrow" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "3095aaf545942ff5abd46654534f15b03a90fba78299d661e045e5d587222f0d" +checksum = "b1bb018b6960c87fd9d025009820406f74e83281185a8bdcb44880d2aa5c9a87" dependencies = [ "arrow-arith", "arrow-array", @@ -201,9 +201,9 @@ dependencies = [ [[package]] name = "arrow-arith" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "00752064ff47cee746e816ddb8450520c3a52cbad1e256f6fa861a35f86c45e7" +checksum = "44de76b51473aa888ecd6ad93ceb262fb8d40d1f1154a4df2f069b3590aa7575" dependencies = [ "arrow-array", "arrow-buffer", @@ -215,9 +215,9 @@ dependencies = [ [[package]] name = "arrow-array" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "cebfe926794fbc1f49ddd0cdaf898956ca9f6e79541efce62dabccfd81380472" +checksum = "29ed77e22744475a9a53d00026cf8e166fe73cf42d89c4c4ae63607ee1cfcc3f" dependencies = [ "ahash", "arrow-buffer", @@ -226,15 +226,15 @@ dependencies = [ "chrono", "chrono-tz", "half", - "hashbrown 0.15.2", + "hashbrown 0.15.3", "num", ] [[package]] name = "arrow-buffer" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "0303c7ec4cf1a2c60310fc4d6bbc3350cd051a17bf9e9c0a8e47b4db79277824" +checksum = "b0391c96eb58bf7389171d1e103112d3fc3e5625ca6b372d606f2688f1ea4cce" dependencies = [ "bytes", "half", @@ -243,9 +243,9 @@ dependencies = [ [[package]] name = "arrow-cast" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "335f769c5a218ea823d3760a743feba1ef7857cba114c01399a891c2fff34285" +checksum = "f39e1d774ece9292697fcbe06b5584401b26bd34be1bec25c33edae65c2420ff" dependencies = [ "arrow-array", "arrow-buffer", @@ -264,9 +264,9 @@ dependencies = [ [[package]] name = "arrow-csv" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "510db7dfbb4d5761826516cc611d97b3a68835d0ece95b034a052601109c0b1b" +checksum = "9055c972a07bf12c2a827debfd34f88d3b93da1941d36e1d9fee85eebe38a12a" dependencies = [ "arrow-array", "arrow-cast", @@ -280,9 +280,9 @@ dependencies = [ [[package]] name = "arrow-data" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e8affacf3351a24039ea24adab06f316ded523b6f8c3dbe28fbac5f18743451b" +checksum = "cf75ac27a08c7f48b88e5c923f267e980f27070147ab74615ad85b5c5f90473d" dependencies = [ "arrow-buffer", "arrow-schema", @@ -292,9 +292,9 @@ dependencies = [ [[package]] name = "arrow-ipc" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "69880a9e6934d9cba2b8630dd08a3463a91db8693b16b499d54026b6137af284" +checksum = "a222f0d93772bd058d1268f4c28ea421a603d66f7979479048c429292fac7b2e" dependencies = [ "arrow-array", "arrow-buffer", @@ -306,9 +306,9 @@ dependencies = [ [[package]] name = "arrow-json" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "d8dafd17a05449e31e0114d740530e0ada7379d7cb9c338fd65b09a8130960b0" +checksum = "9085342bbca0f75e8cb70513c0807cc7351f1fbf5cb98192a67d5e3044acb033" dependencies = [ "arrow-array", "arrow-buffer", @@ -328,9 +328,9 @@ dependencies = [ [[package]] name = "arrow-ord" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "895644523af4e17502d42c3cb6b27cb820f0cb77954c22d75c23a85247c849e1" +checksum = "ab2f1065a5cad7b9efa9e22ce5747ce826aa3855766755d4904535123ef431e7" dependencies = [ "arrow-array", "arrow-buffer", @@ -341,9 +341,9 @@ dependencies = [ [[package]] name = "arrow-row" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "9be8a2a4e5e7d9c822b2b8095ecd77010576d824f654d347817640acfc97d229" +checksum = "3703a0e3e92d23c3f756df73d2dc9476873f873a76ae63ef9d3de17fda83b2d8" dependencies = [ "arrow-array", "arrow-buffer", @@ -354,18 +354,18 @@ dependencies = [ [[package]] name = "arrow-schema" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "7450c76ab7c5a6805be3440dc2e2096010da58f7cab301fdc996a4ee3ee74e49" +checksum = "73a47aa0c771b5381de2b7f16998d351a6f4eb839f1e13d48353e17e873d969b" dependencies = [ "bitflags", ] [[package]] name = "arrow-select" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "aa5f5a93c75f46ef48e4001535e7b6c922eeb0aa20b73cf58d09e13d057490d8" +checksum = "24b7b85575702b23b85272b01bc1c25a01c9b9852305e5d0078c79ba25d995d4" dependencies = [ "ahash", "arrow-array", @@ -377,9 +377,9 @@ dependencies = [ [[package]] name = "arrow-string" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "6e7005d858d84b56428ba2a98a107fe88c0132c61793cf6b8232a1f9bfc0452b" +checksum = "9260fddf1cdf2799ace2b4c2fc0356a9789fa7551e0953e35435536fecefebbd" dependencies = [ "arrow-array", "arrow-buffer", @@ -438,7 +438,7 @@ checksum = "3b43422f69d8ff38f95f1b2bb76517c91589a924d1559a0e935d7c8ce0274c11" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -449,7 +449,7 @@ checksum = "e539d3fca749fcee5236ab05e93a52867dd549cc157c8cb7f99595f3cedffdb5" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -475,9 +475,9 @@ checksum = "ace50bade8e6234aa140d9a2f552bbee1db4d353f69b8217bc503490fc1a9f26" [[package]] name = "backtrace" -version = "0.3.74" +version = "0.3.75" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "8d82cb332cdfaed17ae235a638438ac4d4839913cc2af585c3c6746e8f8bee1a" +checksum = "6806a6321ec58106fea15becdad98371e28d92ccbc7c8f1b3b6dd724fe8f1002" dependencies = [ "addr2line", "cfg-if", @@ -485,7 +485,7 @@ dependencies = [ "miniz_oxide", "object", "rustc-demangle", - "windows-targets", + "windows-targets 0.52.6", ] [[package]] @@ -516,9 +516,9 @@ dependencies = [ [[package]] name = "bitflags" -version = "2.8.0" +version = "2.9.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "8f68f53c83ab957f72c32642f3868eec03eb974d1fb82e453128456482613d36" +checksum = "1b8e56985ec62d17e9c1001dc89c88ecd7dc08e47eba5ec7c29c7b5eeecde967" [[package]] name = "blake2" @@ -531,9 +531,9 @@ dependencies = [ [[package]] name = "blake3" -version = "1.8.1" +version = "1.8.2" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "389a099b34312839e16420d499a9cad9650541715937ffbdd40d36f49e77eeb3" +checksum = "3888aaa89e4b2a40fca9848e400f6a658a5a3978de7be858e209cafa8be9a4a0" dependencies = [ "arrayref", "arrayvec", @@ -553,9 +553,9 @@ dependencies = [ [[package]] name = "brotli" -version = "7.0.0" +version = "8.0.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "cc97b8f16f944bba54f0433f07e30be199b6dc2bd25937444bbad560bcea29bd" +checksum = "9991eea70ea4f293524138648e41ee89b0b2b12ddef3b255effa43c8056e0e0d" dependencies = [ "alloc-no-stdlib", "alloc-stdlib", @@ -564,9 +564,9 @@ dependencies = [ [[package]] name = "brotli-decompressor" -version = "4.0.2" +version = "5.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "74fa05ad7d803d413eb8380983b092cbbaf9a85f151b871360e7b00cd7060b37" +checksum = "874bb8112abecc98cbd6d81ea4fa7e94fb9449648c93cc89aa40c81c24d7de03" dependencies = [ "alloc-no-stdlib", "alloc-stdlib", @@ -586,9 +586,9 @@ checksum = "1fd0f2584146f6f2ef48085050886acf353beff7305ebd1ae69500e27c67f64b" [[package]] name = "bytes" -version = "1.10.0" +version = "1.10.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "f61dac84819c6588b558454b194026eb1f09c293b9036ae9b159e74e73ab6cf9" +checksum = "d71b6127be86fdcfddb610f7182ac57211d4b18a3e9c82eb2d17662f2227ad6a" [[package]] name = "bzip2" @@ -621,9 +621,9 @@ dependencies = [ [[package]] name = "cc" -version = "1.2.14" +version = "1.2.23" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "0c3d1b2e905a3a7b00a6141adb0e4c0bb941d11caf55349d863942a1cc44e3c9" +checksum = "5f4ac86a9e5bc1e2b3449ab9d7d3a6a405e3d1bb28d7b9be8614f55846ae3766" dependencies = [ "jobserver", "libc", @@ -644,9 +644,9 @@ checksum = "613afe47fcd5fac7ccf1db93babcb082c5994d996f20b8b159f2ad1658eb5724" [[package]] name = "chrono" -version = "0.4.40" +version = "0.4.41" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "1a7964611d71df112cb1730f2ee67324fcf4d0fc6606acbbe9bfe06df124637c" +checksum = "c469d952047f47f91b68d1cba3f10d63c11d73e4636f24f08daf0278abf01c4d" dependencies = [ "android-tzdata", "iana-time-zone", @@ -657,9 +657,9 @@ dependencies = [ [[package]] name = "chrono-tz" -version = "0.10.1" +version = "0.10.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "9c6ac4f2c0bf0f44e9161aec9675e1050aa4a530663c4a9e37e108fa948bca9f" +checksum = "efdce149c370f133a071ca8ef6ea340b7b88748ab0810097a9e2976eaa34b4f3" dependencies = [ "chrono", "chrono-tz-build", @@ -668,9 +668,9 @@ dependencies = [ [[package]] name = "chrono-tz-build" -version = "0.4.0" +version = "0.4.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e94fea34d77a245229e7746bd2beb786cd2a896f306ff491fb8cecb3074b10a7" +checksum = "8f10f8c9340e31fc120ff885fcdb54a0b48e474bbd77cab557f0c30a3e569402" dependencies = [ "parse-zoneinfo", "phf_codegen", @@ -710,7 +710,7 @@ version = "0.1.16" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "f9d839f2a20b0aee515dc581a6172f2321f96cab76c1a38a4c584a194955390e" dependencies = [ - "getrandom 0.2.15", + "getrandom 0.2.16", "once_cell", "tiny-keccak", ] @@ -787,9 +787,9 @@ dependencies = [ [[package]] name = "crossbeam-channel" -version = "0.5.14" +version = "0.5.15" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "06ba6d68e24814cb8de6bb986db8222d3a027d15872cabc0d18817bc3c0e4471" +checksum = "82b8f8f868b36967f9606790d1903570de9ceaf870a7bf9fbbd3016d636a2cb2" dependencies = [ "crossbeam-utils", ] @@ -902,7 +902,7 @@ dependencies = [ "object_store", "parking_lot", "parquet", - "rand", + "rand 0.8.5", "regex", "sqlparser", "tempfile", @@ -1025,7 +1025,7 @@ dependencies = [ "log", "object_store", "parquet", - "rand", + "rand 0.8.5", "tempfile", "tokio", "tokio-util", @@ -1136,7 +1136,7 @@ dependencies = [ "object_store", "parking_lot", "parquet", - "rand", + "rand 0.8.5", "tokio", ] @@ -1160,7 +1160,7 @@ dependencies = [ "log", "object_store", "parking_lot", - "rand", + "rand 0.8.5", "tempfile", "url", ] @@ -1241,7 +1241,7 @@ dependencies = [ "itertools 0.14.0", "log", "md-5", - "rand", + "rand 0.8.5", "regex", "sha2", "unicode-segmentation", @@ -1354,7 +1354,7 @@ checksum = "df6f88d7ee27daf8b108ba910f9015176b36fbc72902b1ca5c2a5f1d1717e1a1" dependencies = [ "datafusion-expr", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -1490,7 +1490,7 @@ dependencies = [ [[package]] name = "datafusion-python" -version = "46.0.0" +version = "47.0.0" dependencies = [ "arrow", "async-trait", @@ -1590,20 +1590,20 @@ checksum = "97369cbbc041bc366949bc74d34658d6cda5621039731c6310521892a3a20ae0" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] name = "dyn-clone" -version = "1.0.18" +version = "1.0.19" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "feeef44e73baff3a26d371801df019877a9866a8c493d315ab00177843314f35" +checksum = "1c7a8fb8a9fbf66c1f703fe16184d10ca0ee9d23be5b4436400408ba54a95005" [[package]] name = "either" -version = "1.13.0" +version = "1.15.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "60b1af1c220855b6ceac025d3f6ecdd2b7c4894bfe9cd9bda4fbb4bc7c0d4cf0" +checksum = "48c757948c5ede0e46177b7add2e67155f70e33c07fea8284df6576da70b3719" [[package]] name = "equivalent" @@ -1613,9 +1613,9 @@ checksum = "877a4ace8713b0bcf2a4e7eec82529c029f1d0619886d18145fea96c3ffe5c0f" [[package]] name = "errno" -version = "0.3.10" +version = "0.3.12" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "33d852cb9b869c2a9b3df2f71a3074817f01e1844f839a144f5fcef059a4eb5d" +checksum = "cea14ef9355e3beab063703aa9dab15afd25f0667c341310c1e5274bb1d0da18" dependencies = [ "libc", "windows-sys 0.59.0", @@ -1662,9 +1662,9 @@ checksum = "3f9eec918d3f24069decb9af1554cad7c880e2da24a9afd88aca000531ab82c1" [[package]] name = "foldhash" -version = "0.1.4" +version = "0.1.5" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "a0d2fde1f7b3d48b8395d5f2de76c18a528bd6a9cdde438df747bfcba3e05d6f" +checksum = "d9c4f5dac5e15c24eb999c26181a6ca40b39fe946cbe4c263c7209467bc83af2" [[package]] name = "form_urlencoded" @@ -1731,7 +1731,7 @@ checksum = "162ee34ebcb7c64a8abebc059ce0fee27c2262618d7b60ed8faf72fef13c3650" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -1785,9 +1785,9 @@ dependencies = [ [[package]] name = "getrandom" -version = "0.2.15" +version = "0.2.16" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "c4567c8db10ae91089c99af84c68c38da3ec2f087c3f82960bcdbf3656b6f4d7" +checksum = "335ff9f135e4384c8150d6f27c6daed433577f86b4750418338c01a1a2528592" dependencies = [ "cfg-if", "js-sys", @@ -1798,14 +1798,16 @@ dependencies = [ [[package]] name = "getrandom" -version = "0.3.1" +version = "0.3.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "43a49c392881ce6d5c3b8cb70f98717b7c07aabbdff06687b9030dbfbe2725f8" +checksum = "26145e563e54f2cadc477553f1ec5ee650b00862f0a58bcd12cbdc5f0ea2d2f4" dependencies = [ "cfg-if", + "js-sys", "libc", - "wasi 0.13.3+wasi-0.2.2", - "windows-targets", + "r-efi", + "wasi 0.14.2+wasi-0.2.4", + "wasm-bindgen", ] [[package]] @@ -1822,9 +1824,9 @@ checksum = "a8d1add55171497b4705a648c6b583acafb01d58050a51727785f0b2c8e0a2b2" [[package]] name = "h2" -version = "0.4.7" +version = "0.4.10" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "ccae279728d634d083c00f6099cb58f01cc99c145b84b8be2f6c74618d79922e" +checksum = "a9421a676d1b147b16b82c9225157dc629087ef8ec4d5e2960f9437a90dac0a5" dependencies = [ "atomic-waker", "bytes", @@ -1862,9 +1864,9 @@ dependencies = [ [[package]] name = "hashbrown" -version = "0.15.2" +version = "0.15.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "bf151400ff0baff5465007dd2f3e717f3fe502074ca563069ce3a6629d07b289" +checksum = "84b26c544d002229e640969970a2e74021aadf6e2f96372b9c58eff97de08eb3" dependencies = [ "allocator-api2", "equivalent", @@ -1885,9 +1887,9 @@ checksum = "7f24254aa9a54b5c858eaee2f5bccdb46aaf0e486a595ed5fd8f86ba55232a70" [[package]] name = "http" -version = "1.2.0" +version = "1.3.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "f16ca2af56261c99fba8bac40a10251ce8188205a4c448fbb745a2e4daa76fea" +checksum = "f4a85d31aea989eead29a3aaf9e1115a180df8282431156e533de47660892565" dependencies = [ "bytes", "fnv", @@ -1906,12 +1908,12 @@ dependencies = [ [[package]] name = "http-body-util" -version = "0.1.2" +version = "0.1.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "793429d76616a256bcb62c2a2ec2bed781c8307e797e2598c50010f2bee2544f" +checksum = "b021d93e26becf5dc7e1b75b1bed1fd93124b374ceb73f43d4d4eafec896a64a" dependencies = [ "bytes", - "futures-util", + "futures-core", "http", "http-body", "pin-project-lite", @@ -1919,15 +1921,15 @@ dependencies = [ [[package]] name = "httparse" -version = "1.10.0" +version = "1.10.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "f2d708df4e7140240a16cd6ab0ab65c972d7433ab77819ea693fde9c43811e2a" +checksum = "6dbf3de79e51f3d586ab4cb9d5c3e2c14aa28ed23d180cf89b4df0454a69cc87" [[package]] name = "humantime" -version = "2.1.0" +version = "2.2.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "9a3a5bfb195931eeb336b2a7b4d761daec841b97f947d34394601737a7bba5e4" +checksum = "9b112acc8b3adf4b107a8ec20977da0273a8c386765a3ec0229bd500a1443f9f" [[package]] name = "hyper" @@ -1969,9 +1971,9 @@ dependencies = [ [[package]] name = "hyper-util" -version = "0.1.10" +version = "0.1.11" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "df2dcfbe0677734ab2f3ffa7fa7bfd4706bfdc1ef393f2ee30184aed67e631b4" +checksum = "497bbc33a26fdd4af9ed9c70d63f61cf56a938375fbb32df34db9b1cd6d643f2" dependencies = [ "bytes", "futures-channel", @@ -1979,6 +1981,7 @@ dependencies = [ "http", "http-body", "hyper", + "libc", "pin-project-lite", "socket2", "tokio", @@ -1988,14 +1991,15 @@ dependencies = [ [[package]] name = "iana-time-zone" -version = "0.1.61" +version = "0.1.63" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "235e081f3925a06703c2d0117ea8b91f042756fd6e7a6e5d901e8ca1a996b220" +checksum = "b0c919e5debc312ad217002b8048a17b7d83f80703865bbfcfebb0458b0b27d8" dependencies = [ "android_system_properties", "core-foundation-sys", "iana-time-zone-haiku", "js-sys", + "log", "wasm-bindgen", "windows-core", ] @@ -2011,21 +2015,22 @@ dependencies = [ [[package]] name = "icu_collections" -version = "1.5.0" +version = "2.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "db2fa452206ebee18c4b5c2274dbf1de17008e874b4dc4f0aea9d01ca79e4526" +checksum = "200072f5d0e3614556f94a9930d5dc3e0662a652823904c3a75dc3b0af7fee47" dependencies = [ "displaydoc", + "potential_utf", "yoke", "zerofrom", "zerovec", ] [[package]] -name = "icu_locid" -version = "1.5.0" +name = "icu_locale_core" +version = "2.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "13acbb8371917fc971be86fc8057c41a64b521c184808a698c02acc242dbf637" +checksum = "0cde2700ccaed3872079a65fb1a78f6c0a36c91570f28755dda67bc8f7d9f00a" dependencies = [ "displaydoc", "litemap", @@ -2034,31 +2039,11 @@ dependencies = [ "zerovec", ] -[[package]] -name = "icu_locid_transform" -version = "1.5.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "01d11ac35de8e40fdeda00d9e1e9d92525f3f9d887cdd7aa81d727596788b54e" -dependencies = [ - "displaydoc", - "icu_locid", - "icu_locid_transform_data", - "icu_provider", - "tinystr", - "zerovec", -] - -[[package]] -name = "icu_locid_transform_data" -version = "1.5.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "fdc8ff3388f852bede6b579ad4e978ab004f139284d7b28715f773507b946f6e" - [[package]] name = "icu_normalizer" -version = "1.5.0" +version = "2.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "19ce3e0da2ec68599d193c93d088142efd7f9c5d6fc9b803774855747dc6a84f" +checksum = "436880e8e18df4d7bbc06d58432329d6458cc84531f7ac5f024e93deadb37979" dependencies = [ "displaydoc", "icu_collections", @@ -2066,67 +2051,54 @@ dependencies = [ "icu_properties", "icu_provider", "smallvec", - "utf16_iter", - "utf8_iter", - "write16", "zerovec", ] [[package]] name = "icu_normalizer_data" -version = "1.5.0" +version = "2.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "f8cafbf7aa791e9b22bec55a167906f9e1215fd475cd22adfcf660e03e989516" +checksum = "00210d6893afc98edb752b664b8890f0ef174c8adbb8d0be9710fa66fbbf72d3" [[package]] name = "icu_properties" -version = "1.5.1" +version = "2.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "93d6020766cfc6302c15dbbc9c8778c37e62c14427cb7f6e601d849e092aeef5" +checksum = "2549ca8c7241c82f59c80ba2a6f415d931c5b58d24fb8412caa1a1f02c49139a" dependencies = [ "displaydoc", "icu_collections", - "icu_locid_transform", + "icu_locale_core", "icu_properties_data", "icu_provider", - "tinystr", + "potential_utf", + "zerotrie", "zerovec", ] [[package]] name = "icu_properties_data" -version = "1.5.0" +version = "2.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "67a8effbc3dd3e4ba1afa8ad918d5684b8868b3b26500753effea8d2eed19569" +checksum = "8197e866e47b68f8f7d95249e172903bec06004b18b2937f1095d40a0c57de04" [[package]] name = "icu_provider" -version = "1.5.0" +version = "2.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "6ed421c8a8ef78d3e2dbc98a973be2f3770cb42b606e3ab18d6237c4dfde68d9" +checksum = "03c80da27b5f4187909049ee2d72f276f0d9f99a42c306bd0131ecfe04d8e5af" dependencies = [ "displaydoc", - "icu_locid", - "icu_provider_macros", + "icu_locale_core", "stable_deref_trait", "tinystr", "writeable", "yoke", "zerofrom", + "zerotrie", "zerovec", ] -[[package]] -name = "icu_provider_macros" -version = "1.5.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "1ec89e9337638ecdc08744df490b221a7399bf8d164eb52a665454e60e075ad6" -dependencies = [ - "proc-macro2", - "quote", - "syn 2.0.100", -] - [[package]] name = "idna" version = "1.0.3" @@ -2140,9 +2112,9 @@ dependencies = [ [[package]] name = "idna_adapter" -version = "1.2.0" +version = "1.2.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "daca1df1c957320b2cf139ac61e7bd64fed304c5040df000a745aa1de3b4ef71" +checksum = "3acae9609540aa318d1bc588455225fb2085b9ed0c4f6bd0d9d5bcd86f1a0344" dependencies = [ "icu_normalizer", "icu_properties", @@ -2155,14 +2127,14 @@ source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "cea70ddb795996207ad57735b50c5982d8844f38ba9ee5f1aedcfb708a2aa11e" dependencies = [ "equivalent", - "hashbrown 0.15.2", + "hashbrown 0.15.3", ] [[package]] name = "indoc" -version = "2.0.5" +version = "2.0.6" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "b248f5224d1d606005e02c97f5aa4e88eeb230488bcc03bc9ca4d7991399f2b5" +checksum = "f4c7245a08504955605670dbf141fceab975f15ca21570696aebe9d2e71576bd" [[package]] name = "integer-encoding" @@ -2196,16 +2168,17 @@ dependencies = [ [[package]] name = "itoa" -version = "1.0.14" +version = "1.0.15" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "d75a2a4b1b190afb6f5425f10f6a8f959d2ea0b9c2b1d79553551850539e4674" +checksum = "4a5f13b858c8d314ee3e8f639011f7ccefe71f97f96e50151fb991f267928e2c" [[package]] name = "jobserver" -version = "0.1.32" +version = "0.1.33" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "48d1dbcbbeb6a7fec7e059840aa538bd62aaccf972c7346c4d9d2059312853d0" +checksum = "38f262f097c174adebe41eb73d66ae9c06b2844fb0da69969647bbddd9b0538a" dependencies = [ + "getrandom 0.3.3", "libc", ] @@ -2291,9 +2264,9 @@ dependencies = [ [[package]] name = "libc" -version = "0.2.171" +version = "0.2.172" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "c19937216e9d3aa9956d9bb8dfc0b0c8beb6058fc4f7a4dc4d850edf86a237d6" +checksum = "d750af042f7ef4f724306de029d18836c26c1765a54a6a3f094cbd23a7267ffa" [[package]] name = "libflate" @@ -2331,15 +2304,15 @@ dependencies = [ [[package]] name = "libm" -version = "0.2.11" +version = "0.2.15" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "8355be11b20d696c8f18f6cc018c4e372165b1fa8126cef092399c9951984ffa" +checksum = "f9fbbcab51052fe104eb5e5d351cf728d30a5be1fe14d9be8a3b097481fb97de" [[package]] name = "libmimalloc-sys" -version = "0.1.39" +version = "0.1.42" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "23aa6811d3bd4deb8a84dde645f943476d13b248d818edcf8ce0b2f37f036b44" +checksum = "ec9d6fac27761dabcd4ee73571cdb06b7022dc99089acbe5435691edffaac0f4" dependencies = [ "cc", "libc", @@ -2356,15 +2329,15 @@ dependencies = [ [[package]] name = "linux-raw-sys" -version = "0.4.15" +version = "0.9.4" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "d26c52dbd32dccf2d10cac7725f8eae5296885fb5703b261f7d0a0739ec807ab" +checksum = "cd945864f07fe9f5371a27ad7b52a172b4b499999f1d97574c9fa68373937e12" [[package]] name = "litemap" -version = "0.7.4" +version = "0.8.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "4ee93343901ab17bd981295f2cf0026d4ad018c7c31ba84549a4ddbb47a45104" +checksum = "241eaef5fd12c88705a01fc1066c48c4b36e0dd4377dcdc7ec3942cea7a69956" [[package]] name = "lock_api" @@ -2378,9 +2351,15 @@ dependencies = [ [[package]] name = "log" -version = "0.4.25" +version = "0.4.27" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "04cbf5b083de1c7e0222a7a51dbfdba1cbe1c6ab0b15e29fff3f6c077fd9cd9f" +checksum = "13dc2df351e3202783a1fe0d44375f7295ffb4049267b0f3018346dc122a1d94" + +[[package]] +name = "lru-slab" +version = "0.1.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "112b39cec0b298b6c1999fee3e31427f74f676e4cb9879ed1a121b43661a4154" [[package]] name = "lz4_flex" @@ -2429,9 +2408,9 @@ dependencies = [ [[package]] name = "mimalloc" -version = "0.1.43" +version = "0.1.46" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "68914350ae34959d83f732418d51e2427a794055d0b9529f48259ac07af65633" +checksum = "995942f432bbb4822a7e9c3faa87a695185b0d09273ba85f097b54f4e458f2af" dependencies = [ "libmimalloc-sys", ] @@ -2464,9 +2443,9 @@ dependencies = [ [[package]] name = "multimap" -version = "0.10.0" +version = "0.10.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "defc4c55412d89136f966bbb339008b474350e5e6e78d2714439c386b3137a03" +checksum = "1d87ecb2933e8aeadb3e3a02b828fed80a7528047e68b4f424523a0981a3a084" [[package]] name = "num" @@ -2554,9 +2533,9 @@ dependencies = [ [[package]] name = "object_store" -version = "0.12.0" +version = "0.12.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e9ce831b09395f933addbc56d894d889e4b226eba304d4e7adbab591e26daf1e" +checksum = "d94ac16b433c0ccf75326388c893d2835ab7457ea35ab8ba5d745c053ef5fa16" dependencies = [ "async-trait", "base64 0.22.1", @@ -2574,25 +2553,27 @@ dependencies = [ "parking_lot", "percent-encoding", "quick-xml", - "rand", + "rand 0.9.1", "reqwest", "ring", "rustls-pemfile", "serde", "serde_json", "serde_urlencoded", - "thiserror 2.0.11", + "thiserror 2.0.12", "tokio", "tracing", "url", "walkdir", + "wasm-bindgen-futures", + "web-time", ] [[package]] name = "once_cell" -version = "1.20.3" +version = "1.21.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "945462a4b81e43c4e3ba96bd7b49d834c6f61198356aa858733bc4acf3cbe62e" +checksum = "42f5e15c9953c5e4ccceeb2e7382a716482c34515315f7b03532b8b4e8393d2d" [[package]] name = "openssl-probe" @@ -2629,14 +2610,14 @@ dependencies = [ "libc", "redox_syscall", "smallvec", - "windows-targets", + "windows-targets 0.52.6", ] [[package]] name = "parquet" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "cd31a8290ac5b19f09ad77ee7a1e6a541f1be7674ad410547d5f1eef6eef4a9c" +checksum = "be7b2d778f6b841d37083ebdf32e33a524acde1266b5884a8ca29bf00dfa1231" dependencies = [ "ahash", "arrow-array", @@ -2653,7 +2634,7 @@ dependencies = [ "flate2", "futures", "half", - "hashbrown 0.15.2", + "hashbrown 0.15.3", "lz4_flex", "num", "num-bigint", @@ -2762,7 +2743,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "3c80231409c20246a13fddb31776fb942c38553c51e871f8cbd687a4cfb5843d" dependencies = [ "phf_shared", - "rand", + "rand 0.8.5", ] [[package]] @@ -2788,21 +2769,30 @@ checksum = "8b870d8c151b6f2fb93e84a13146138f05d02ed11c7e7c54f8826aaaf7c9f184" [[package]] name = "pkg-config" -version = "0.3.31" +version = "0.3.32" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "953ec861398dccce10c670dfeaf3ec4911ca479e9c02154b3a215178c5f566f2" +checksum = "7edddbd0b52d732b21ad9a5fab5c704c14cd949e5e9a1ec5929a24fded1b904c" [[package]] name = "portable-atomic" -version = "1.10.0" +version = "1.11.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "280dc24453071f1b63954171985a0b0d30058d287960968b9b2aca264c8d4ee6" +checksum = "350e9b48cbc6b0e028b0473b114454c6316e57336ee184ceab6e53f72c178b3e" + +[[package]] +name = "potential_utf" +version = "0.1.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "e5a7c30837279ca13e7c867e9e40053bc68740f988cb07f7ca6df43cc734b585" +dependencies = [ + "zerovec", +] [[package]] name = "ppv-lite86" -version = "0.2.20" +version = "0.2.21" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "77957b295656769bb8ad2b6a6b09d897d94f05c41b069aede1fcdaa675eaea04" +checksum = "85eae3c4ed2f50dcfe72643da4befc30deadb458a9b590d720cde2f2b1e97da9" dependencies = [ "zerocopy", ] @@ -2814,14 +2804,14 @@ source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "664ec5419c51e34154eec046ebcba56312d5a2fc3b09a06da188e1ad21afadf6" dependencies = [ "proc-macro2", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] name = "proc-macro2" -version = "1.0.93" +version = "1.0.95" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "60946a68e5f9d28b0dc1c21bb8a97ee7d018a8b322fa57838ba31cc878e22d99" +checksum = "02b3e5e68a3a1a02aad3ec490a98007cbc13c37cbe84a3cd7b8e406d76e7f778" dependencies = [ "unicode-ident", ] @@ -2843,7 +2833,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "be769465445e8c1474e9c5dac2018218498557af32d9ed057325ec9a41ae81bf" dependencies = [ "heck", - "itertools 0.13.0", + "itertools 0.14.0", "log", "multimap", "once_cell", @@ -2852,7 +2842,7 @@ dependencies = [ "prost", "prost-types", "regex", - "syn 2.0.100", + "syn 2.0.101", "tempfile", ] @@ -2863,10 +2853,10 @@ source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "8a56d757972c98b346a9b766e3f02746cde6dd1cd1d1d563472929fdd74bec4d" dependencies = [ "anyhow", - "itertools 0.13.0", + "itertools 0.14.0", "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -2889,18 +2879,18 @@ dependencies = [ [[package]] name = "psm" -version = "0.1.25" +version = "0.1.26" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "f58e5423e24c18cc840e1c98370b3993c6649cd1678b4d24318bcf0a083cbe88" +checksum = "6e944464ec8536cd1beb0bbfd96987eb5e3b72f2ecdafdc5c769a37f1fa2ae1f" dependencies = [ "cc", ] [[package]] name = "pyo3" -version = "0.24.1" +version = "0.24.2" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "17da310086b068fbdcefbba30aeb3721d5bb9af8db4987d6735b2183ca567229" +checksum = "e5203598f366b11a02b13aa20cab591229ff0a89fd121a308a5df751d5fc9219" dependencies = [ "cfg-if", "indoc", @@ -2929,9 +2919,9 @@ dependencies = [ [[package]] name = "pyo3-build-config" -version = "0.24.1" +version = "0.24.2" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e27165889bd793000a098bb966adc4300c312497ea25cf7a690a9f0ac5aa5fc1" +checksum = "99636d423fa2ca130fa5acde3059308006d46f98caac629418e53f7ebb1e9999" dependencies = [ "once_cell", "target-lexicon", @@ -2939,9 +2929,9 @@ dependencies = [ [[package]] name = "pyo3-ffi" -version = "0.24.1" +version = "0.24.2" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "05280526e1dbf6b420062f3ef228b78c0c54ba94e157f5cb724a609d0f2faabc" +checksum = "78f9cf92ba9c409279bc3305b5409d90db2d2c22392d443a87df3a1adad59e33" dependencies = [ "libc", "pyo3-build-config", @@ -2949,27 +2939,27 @@ dependencies = [ [[package]] name = "pyo3-macros" -version = "0.24.1" +version = "0.24.2" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "5c3ce5686aa4d3f63359a5100c62a127c9f15e8398e5fdeb5deef1fed5cd5f44" +checksum = "0b999cb1a6ce21f9a6b147dcf1be9ffedf02e0043aec74dc390f3007047cecd9" dependencies = [ "proc-macro2", "pyo3-macros-backend", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] name = "pyo3-macros-backend" -version = "0.24.1" +version = "0.24.2" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "f4cf6faa0cbfb0ed08e89beb8103ae9724eb4750e3a78084ba4017cbe94f3855" +checksum = "822ece1c7e1012745607d5cf0bcb2874769f0f7cb34c4cde03b9358eb9ef911a" dependencies = [ "heck", "proc-macro2", "pyo3-build-config", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -2980,9 +2970,9 @@ checksum = "5a651516ddc9168ebd67b24afd085a718be02f8858fe406591b013d101ce2f40" [[package]] name = "quick-xml" -version = "0.37.2" +version = "0.37.5" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "165859e9e55f79d67b96c5d96f4e88b6f2695a1972849c15a6a3f5c59fc2c003" +checksum = "331e97a1af0bf59823e6eadffe373d7b27f485be8748f71471c662c1f269b7fb" dependencies = [ "memchr", "serde", @@ -2990,37 +2980,40 @@ dependencies = [ [[package]] name = "quinn" -version = "0.11.6" +version = "0.11.8" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "62e96808277ec6f97351a2380e6c25114bc9e67037775464979f3037c92d05ef" +checksum = "626214629cda6781b6dc1d316ba307189c85ba657213ce642d9c77670f8202c8" dependencies = [ "bytes", + "cfg_aliases", "pin-project-lite", "quinn-proto", "quinn-udp", "rustc-hash", "rustls", "socket2", - "thiserror 2.0.11", + "thiserror 2.0.12", "tokio", "tracing", + "web-time", ] [[package]] name = "quinn-proto" -version = "0.11.9" +version = "0.11.12" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "a2fe5ef3495d7d2e377ff17b1a8ce2ee2ec2a18cde8b6ad6619d65d0701c135d" +checksum = "49df843a9161c85bb8aae55f101bc0bac8bcafd637a620d9122fd7e0b2f7422e" dependencies = [ "bytes", - "getrandom 0.2.15", - "rand", + "getrandom 0.3.3", + "lru-slab", + "rand 0.9.1", "ring", "rustc-hash", "rustls", "rustls-pki-types", "slab", - "thiserror 2.0.11", + "thiserror 2.0.12", "tinyvec", "tracing", "web-time", @@ -3028,9 +3021,9 @@ dependencies = [ [[package]] name = "quinn-udp" -version = "0.5.10" +version = "0.5.12" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e46f3055866785f6b92bc6164b76be02ca8f2eb4b002c0354b28cf4c119e5944" +checksum = "ee4e529991f949c5e25755532370b8af5d114acae52326361d68d47af64aa842" dependencies = [ "cfg_aliases", "libc", @@ -3049,6 +3042,12 @@ dependencies = [ "proc-macro2", ] +[[package]] +name = "r-efi" +version = "5.2.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "74765f6d916ee2faa39bc8e68e4f3ed8949b48cccdac59983d287a7cb71ce9c5" + [[package]] name = "rand" version = "0.8.5" @@ -3056,8 +3055,18 @@ source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "34af8d1a0e25924bc5b7c43c079c942339d8f0a8b57c39049bef581b46327404" dependencies = [ "libc", - "rand_chacha", - "rand_core", + "rand_chacha 0.3.1", + "rand_core 0.6.4", +] + +[[package]] +name = "rand" +version = "0.9.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "9fbfd9d094a40bf3ae768db9361049ace4c0e04a4fd6b359518bd7b73a73dd97" +dependencies = [ + "rand_chacha 0.9.0", + "rand_core 0.9.3", ] [[package]] @@ -3067,7 +3076,17 @@ source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "e6c10a63a0fa32252be49d21e7709d4d4baf8d231c2dbce1eaa8141b9b127d88" dependencies = [ "ppv-lite86", - "rand_core", + "rand_core 0.6.4", +] + +[[package]] +name = "rand_chacha" +version = "0.9.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "d3022b5f1df60f26e1ffddd6c66e8aa15de382ae63b3a0c1bfc0e4d3e3f325cb" +dependencies = [ + "ppv-lite86", + "rand_core 0.9.3", ] [[package]] @@ -3076,7 +3095,16 @@ version = "0.6.4" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "ec0be4795e2f6a28069bec0b5ff3e2ac9bafc99e6a9a7dc3547996c5c816922c" dependencies = [ - "getrandom 0.2.15", + "getrandom 0.2.16", +] + +[[package]] +name = "rand_core" +version = "0.9.3" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "99d9a13982dcf210057a8a78572b2217b667c3beacbf3a0d8b454f6f82837d38" +dependencies = [ + "getrandom 0.3.3", ] [[package]] @@ -3096,14 +3124,14 @@ source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "76009fbe0614077fc1a2ce255e3a1881a2e3a3527097d5dc6d8212c585e7e38b" dependencies = [ "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] name = "redox_syscall" -version = "0.5.8" +version = "0.5.12" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "03a862b389f93e68874fbf580b9de08dd02facb9a788ebadaf4a3fd33cf58834" +checksum = "928fca9cf2aa042393a8325b9ead81d2f0df4cb12e1e24cef072922ccd99c5af" dependencies = [ "bitflags", ] @@ -3149,7 +3177,7 @@ version = "0.10.3" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "78ef7fa9ed0256d64a688a3747d0fef7a88851c18a5e1d57f115f38ec2e09366" dependencies = [ - "hashbrown 0.15.2", + "hashbrown 0.15.3", "memchr", ] @@ -3164,9 +3192,9 @@ dependencies = [ [[package]] name = "reqwest" -version = "0.12.12" +version = "0.12.15" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "43e734407157c3c2034e0258f5e4473ddb361b1e85f95a66690d67264d7cd1da" +checksum = "d19c46a6fdd48bc4dab94b6103fccc55d34c67cc0ad04653aad4ea2a07cd7bbb" dependencies = [ "base64 0.22.1", "bytes", @@ -3210,13 +3238,13 @@ dependencies = [ [[package]] name = "ring" -version = "0.17.9" +version = "0.17.14" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e75ec5e92c4d8aede845126adc388046234541629e76029599ed35a003c7ed24" +checksum = "a4689e6c2294d81e88dc6261c768b63bc4fcdb852be6d1352498b114f61383b7" dependencies = [ "cc", "cfg-if", - "getrandom 0.2.15", + "getrandom 0.2.16", "libc", "untrusted", "windows-sys 0.52.0", @@ -3251,9 +3279,9 @@ dependencies = [ [[package]] name = "rustix" -version = "0.38.44" +version = "1.0.7" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "fdb5bc1ae2baa591800df16c9ca78619bf65c0488b41b96ccec5d11220d8c154" +checksum = "c71e83d6afe7ff64890ec6b71d6a69bb8a610ab78ce364b3352876bb4c801266" dependencies = [ "bitflags", "errno", @@ -3264,9 +3292,9 @@ dependencies = [ [[package]] name = "rustls" -version = "0.23.23" +version = "0.23.27" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "47796c98c480fce5406ef69d1c76378375492c3b0a0de587be0c1d9feb12f395" +checksum = "730944ca083c1c233a75c09f199e973ca499344a2b7ba9e755c457e86fb4a321" dependencies = [ "once_cell", "ring", @@ -3299,18 +3327,19 @@ dependencies = [ [[package]] name = "rustls-pki-types" -version = "1.11.0" +version = "1.12.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "917ce264624a4b4db1c364dcc35bfca9ded014d0a958cd47ad3e960e988ea51c" +checksum = "229a4a4c221013e7e1f1a043678c5cc39fe5171437c88fb47151a21e6f5b5c79" dependencies = [ "web-time", + "zeroize", ] [[package]] name = "rustls-webpki" -version = "0.102.8" +version = "0.103.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "64ca1bc8749bd4cf37b5ce386cc146580777b4e8572c7b97baf22c83f444bee9" +checksum = "e4a72fe2bcf7a6ac6fd7d0b9e5cb68aeb7d4c0a0271730218b3e92d43b4eb435" dependencies = [ "ring", "rustls-pki-types", @@ -3319,15 +3348,15 @@ dependencies = [ [[package]] name = "rustversion" -version = "1.0.19" +version = "1.0.20" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "f7c45b9784283f1b2e7fb61b42047c2fd678ef0960d4f6f1eba131594cc369d4" +checksum = "eded382c5f5f786b989652c49544c4877d9f015cc22e145a5ea8ea66c2921cd2" [[package]] name = "ryu" -version = "1.0.19" +version = "1.0.20" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "6ea1a2d0a644769cc99faa24c3ad26b379b786fe7c36fd3c546254801650e6dd" +checksum = "28d3b2b1366ec20994f1fd18c3c594f05c5dd4bc44d8bb0c1c632c8d6829481f" [[package]] name = "same-file" @@ -3368,7 +3397,7 @@ dependencies = [ "proc-macro2", "quote", "serde_derive_internals", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -3411,9 +3440,9 @@ dependencies = [ [[package]] name = "seq-macro" -version = "0.3.5" +version = "0.3.6" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "a3f0bf26fd526d2a95683cd0f87bf103b8539e2ca1ef48ce002d67aad59aa0b4" +checksum = "1bc711410fbe7399f390ca1c3b60ad0f53f80e95c5eb935e52268a0e2cd49acc" [[package]] name = "serde" @@ -3426,9 +3455,9 @@ dependencies = [ [[package]] name = "serde_bytes" -version = "0.11.15" +version = "0.11.17" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "387cc504cb06bb40a96c8e04e951fe01854cf6bc921053c954e4a606d9675c6a" +checksum = "8437fd221bde2d4ca316d61b90e337e9e702b3820b87d63caa9ba6c02bd06d96" dependencies = [ "serde", ] @@ -3441,7 +3470,7 @@ checksum = "5b0276cf7f2c73365f7157c8123c21cd9a50fbbd844757af28ca1f5925fc2a00" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -3452,7 +3481,7 @@ checksum = "18d26a20a969b9e3fdf2fc2d9f21eda6c40e2de84c9408bb5d3b05d499aae711" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -3476,7 +3505,7 @@ dependencies = [ "proc-macro2", "quote", "serde", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -3506,9 +3535,9 @@ dependencies = [ [[package]] name = "sha2" -version = "0.10.8" +version = "0.10.9" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "793db75ad2bcafc3ffa7c68b215fee268f537982cd901d132f89c6343f3a3dc8" +checksum = "a7507d819769d01a365ab707794a4084392c824f54a7a6a7862f8c3d0892b283" dependencies = [ "cfg-if", "cpufeatures", @@ -3544,9 +3573,9 @@ dependencies = [ [[package]] name = "smallvec" -version = "1.14.0" +version = "1.15.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "7fcf8323ef1faaee30a44a340193b1ac6814fd9b7b4e88e9d4519a3e4abe1cfd" +checksum = "8917285742e9f3e1683f0a9c4e6b57960b7314d0b08d30d1ecd426713ee2eee9" [[package]] name = "snap" @@ -3556,9 +3585,9 @@ checksum = "1b6b67fb9a61334225b5b790716f609cd58395f895b3fe8b328786812a40bc3b" [[package]] name = "socket2" -version = "0.5.8" +version = "0.5.9" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "c970269d99b64e60ec3bd6ad27270092a5394c4e309314b18ae3fe575695fbe8" +checksum = "4f5fd57c80058a56cf5c777ab8a126398ece8e442983605d280a44ce79d0edef" dependencies = [ "libc", "windows-sys 0.52.0", @@ -3583,7 +3612,7 @@ checksum = "da5fc6819faabb412da764b99d3b713bb55083c11e7e0c00144d386cd6a1939c" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -3594,9 +3623,9 @@ checksum = "a8f112729512f8e442d81f95a8a7ddf2b7c6b8a1a6f509a95864142b30cab2d3" [[package]] name = "stacker" -version = "0.1.18" +version = "0.1.21" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "1d08feb8f695b465baed819b03c128dc23f57a694510ab1f06c77f763975685e" +checksum = "cddb07e32ddb770749da91081d8d0ac3a16f1a569a18b20348cd371f5dead06b" dependencies = [ "cc", "cfg-if", @@ -3627,7 +3656,7 @@ dependencies = [ "proc-macro2", "quote", "rustversion", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -3651,7 +3680,7 @@ dependencies = [ "serde", "serde_json", "serde_yaml", - "syn 2.0.100", + "syn 2.0.101", "typify", "walkdir", ] @@ -3675,9 +3704,9 @@ dependencies = [ [[package]] name = "syn" -version = "2.0.100" +version = "2.0.101" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "b09a44accad81e1ba1cd74a32461ba89dee89095ba17b32f5d03683b1b1fc2a0" +checksum = "8ce2b7fc941b3a24138a0a7cf8e858bfc6a992e7978a068a5c760deb0ed43caf" dependencies = [ "proc-macro2", "quote", @@ -3695,13 +3724,13 @@ dependencies = [ [[package]] name = "synstructure" -version = "0.13.1" +version = "0.13.2" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "c8af7666ab7b6390ab78131fb5b0fce11d6b7a6951602017c35fa82800708971" +checksum = "728a70f3dbaf5bab7f0c4b1ac8d7ae5ea60a4b5549c8a5914361c99147a709d2" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -3712,13 +3741,12 @@ checksum = "e502f78cdbb8ba4718f566c418c52bc729126ffd16baee5baa718cf25dd5a69a" [[package]] name = "tempfile" -version = "3.16.0" +version = "3.20.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "38c246215d7d24f48ae091a2902398798e05d978b24315d6efbc00ede9a8bb91" +checksum = "e8a64e3985349f2441a1a9ef0b853f869006c3855f2cda6862a94d26ebb9d6a1" dependencies = [ - "cfg-if", "fastrand", - "getrandom 0.3.1", + "getrandom 0.3.3", "once_cell", "rustix", "windows-sys 0.59.0", @@ -3735,11 +3763,11 @@ dependencies = [ [[package]] name = "thiserror" -version = "2.0.11" +version = "2.0.12" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "d452f284b73e6d76dd36758a0c8684b1d5be31f92b89d07fd5822175732206fc" +checksum = "567b8a2dae586314f7be2a752ec7474332959c6460e02bde30d702a66d488708" dependencies = [ - "thiserror-impl 2.0.11", + "thiserror-impl 2.0.12", ] [[package]] @@ -3750,18 +3778,18 @@ checksum = "4fee6c4efc90059e10f81e6d42c60a18f76588c3d74cb83a0b242a2b6c7504c1" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] name = "thiserror-impl" -version = "2.0.11" +version = "2.0.12" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "26afc1baea8a989337eeb52b6e72a039780ce45c3edfcc9c5b9d112feeb173c2" +checksum = "7f7cf42b4507d8ea322120659672cf1b9dbb93f8f2d4ecfd6e51350ff5b17a1d" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -3786,9 +3814,9 @@ dependencies = [ [[package]] name = "tinystr" -version = "0.7.6" +version = "0.8.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "9117f5d4db391c1cf6927e7bea3db74b9a1c1add8f7eda9ffd5364f40f57b82f" +checksum = "5d4f6d1145dcb577acf783d4e601bc1d76a13337bb54e6233add580b07344c8b" dependencies = [ "displaydoc", "zerovec", @@ -3796,9 +3824,9 @@ dependencies = [ [[package]] name = "tinyvec" -version = "1.8.1" +version = "1.9.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "022db8904dfa342efe721985167e9fcd16c29b226db4397ed752a761cfce81e8" +checksum = "09b3661f17e86524eccd4371ab0429194e0d7c008abb45f7a7495b1719463c71" dependencies = [ "tinyvec_macros", ] @@ -3811,9 +3839,9 @@ checksum = "1f3ccbac311fea05f86f61904b462b55fb3df8837a366dfc601a0161d0532f20" [[package]] name = "tokio" -version = "1.44.2" +version = "1.45.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e6b88822cbe49de4185e3a4cbf8321dd487cf5fe0c5c65695fef6346371e9c48" +checksum = "2513ca694ef9ede0fb23fe71a4ee4107cb102b9dc1930f6d0fd77aae068ae165" dependencies = [ "backtrace", "bytes", @@ -3833,14 +3861,14 @@ checksum = "6e06d43f1345a3bcd39f6a56dbb7dcab2ba47e68e8ac134855e7e2bdbaf8cab8" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] name = "tokio-rustls" -version = "0.26.1" +version = "0.26.2" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "5f6d0975eaace0cf0fcadee4e4aaa5da15b5c079146f2cffb67c113be122bf37" +checksum = "8e727b36a1a0e8b74c376ac2211e40c2c8af09fb4013c60d910495810f008e9b" dependencies = [ "rustls", "tokio", @@ -3848,9 +3876,9 @@ dependencies = [ [[package]] name = "tokio-util" -version = "0.7.14" +version = "0.7.15" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "6b9590b93e6fcc1739458317cccd391ad3955e2bde8913edf6f95f9e65a8f034" +checksum = "66a539a9ad6d5d281510d5bd368c973d636c02dbf8a67300bfb6b950696ad7df" dependencies = [ "bytes", "futures-core", @@ -3905,7 +3933,7 @@ checksum = "395ae124c09f9e6918a2310af6038fba074bcf474ac352496d5910dd59a2226d" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -3977,14 +4005,14 @@ checksum = "f9534daa9fd3ed0bd911d462a37f172228077e7abf18c18a5f67199d959205f8" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] name = "typenum" -version = "1.17.0" +version = "1.18.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "42ff0bf0c66b8238c6f3b578df37d0b7848e55df8577b3f74f92a69acceeb825" +checksum = "1dccffe3ce07af9386bfd29e80c0ab1a8205a2fc34e4bcd40364df902cfa8f3f" [[package]] name = "typify" @@ -4011,8 +4039,8 @@ dependencies = [ "semver", "serde", "serde_json", - "syn 2.0.100", - "thiserror 2.0.11", + "syn 2.0.101", + "thiserror 2.0.12", "unicode-ident", ] @@ -4029,15 +4057,15 @@ dependencies = [ "serde", "serde_json", "serde_tokenstream", - "syn 2.0.100", + "syn 2.0.101", "typify-impl", ] [[package]] name = "unicode-ident" -version = "1.0.16" +version = "1.0.18" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "a210d160f08b701c8721ba1c726c11662f877ea6b7094007e1ca9a1041945034" +checksum = "5a5f39404a5da50712a4c1eecf25e90dd62b613502b7e925fd4e4d19b5c96512" [[package]] name = "unicode-segmentation" @@ -4053,9 +4081,9 @@ checksum = "1fc81956842c57dac11422a97c3b8195a1ff727f06e85c84ed2e8aa277c9a0fd" [[package]] name = "unindent" -version = "0.2.3" +version = "0.2.4" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "c7de7d73e1754487cb58364ee906a499937a0dfabd86bcb980fa99ec8c8fa2ce" +checksum = "7264e107f553ccae879d21fbea1d6724ac785e8c3bfc762137959b5802826ef3" [[package]] name = "unsafe-libyaml" @@ -4080,12 +4108,6 @@ dependencies = [ "percent-encoding", ] -[[package]] -name = "utf16_iter" -version = "1.0.5" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "c8232dd3cdaed5356e0f716d285e4b40b932ac434100fe9b7e0e8e935b9e6246" - [[package]] name = "utf8_iter" version = "1.0.4" @@ -4098,7 +4120,7 @@ version = "1.16.0" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "458f7a779bf54acc9f347480ac654f68407d3aab21269a6e3c9f922acd9e2da9" dependencies = [ - "getrandom 0.3.1", + "getrandom 0.3.3", "js-sys", "serde", "wasm-bindgen", @@ -4137,9 +4159,9 @@ checksum = "9c8d87e72b64a3b4db28d11ce29237c246188f4f51057d65a7eab63b7987e423" [[package]] name = "wasi" -version = "0.13.3+wasi-0.2.2" +version = "0.14.2+wasi-0.2.4" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "26816d2e1a4a36a2940b96c5296ce403917633dff8f3440e9b236ed6f6bacad2" +checksum = "9683f9a5a998d873c0d21fcbe3c083009670149a8fab228644b8bd36b2c48cb3" dependencies = [ "wit-bindgen-rt", ] @@ -4166,7 +4188,7 @@ dependencies = [ "log", "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", "wasm-bindgen-shared", ] @@ -4201,7 +4223,7 @@ checksum = "8ae87ea40c9f689fc23f209965b6fb8a99ad69aeeb0231408be24920604395de" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", "wasm-bindgen-backend", "wasm-bindgen-shared", ] @@ -4281,11 +4303,37 @@ checksum = "712e227841d057c1ee1cd2fb22fa7e5a5461ae8e48fa2ca79ec42cfc1931183f" [[package]] name = "windows-core" -version = "0.52.0" +version = "0.61.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "33ab640c8d7e35bf8ba19b884ba838ceb4fba93a4e8c65a9059d08afcfc683d9" +checksum = "46ec44dc15085cea82cf9c78f85a9114c463a369786585ad2882d1ff0b0acf40" dependencies = [ - "windows-targets", + "windows-implement", + "windows-interface", + "windows-link", + "windows-result", + "windows-strings 0.4.1", +] + +[[package]] +name = "windows-implement" +version = "0.60.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "a47fddd13af08290e67f4acabf4b459f647552718f683a7b415d290ac744a836" +dependencies = [ + "proc-macro2", + "quote", + "syn 2.0.101", +] + +[[package]] +name = "windows-interface" +version = "0.59.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "bd9211b69f8dcdfa817bfd14bf1c97c9188afa36f4750130fcdf3f400eca9fa8" +dependencies = [ + "proc-macro2", + "quote", + "syn 2.0.101", ] [[package]] @@ -4296,32 +4344,40 @@ checksum = "76840935b766e1b0a05c0066835fb9ec80071d4c09a16f6bd5f7e655e3c14c38" [[package]] name = "windows-registry" -version = "0.2.0" +version = "0.4.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e400001bb720a623c1c69032f8e3e4cf09984deec740f007dd2b03ec864804b0" +checksum = "4286ad90ddb45071efd1a66dfa43eb02dd0dfbae1545ad6cc3c51cf34d7e8ba3" dependencies = [ "windows-result", - "windows-strings", - "windows-targets", + "windows-strings 0.3.1", + "windows-targets 0.53.0", ] [[package]] name = "windows-result" -version = "0.2.0" +version = "0.3.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "1d1043d8214f791817bab27572aaa8af63732e11bf84aa21a45a78d6c317ae0e" +checksum = "4b895b5356fc36103d0f64dd1e94dfa7ac5633f1c9dd6e80fe9ec4adef69e09d" dependencies = [ - "windows-targets", + "windows-link", ] [[package]] name = "windows-strings" -version = "0.1.0" +version = "0.3.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "4cd9b125c486025df0eabcb585e62173c6c9eddcec5d117d3b6e8c30e2ee4d10" +checksum = "87fa48cc5d406560701792be122a10132491cff9d0aeb23583cc2dcafc847319" dependencies = [ - "windows-result", - "windows-targets", + "windows-link", +] + +[[package]] +name = "windows-strings" +version = "0.4.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "2a7ab927b2637c19b3dbe0965e75d8f2d30bdd697a1516191cad2ec4df8fb28a" +dependencies = [ + "windows-link", ] [[package]] @@ -4330,7 +4386,7 @@ version = "0.52.0" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "282be5f36a8ce781fad8c8ae18fa3f9beff57ec1b52cb3de0789201425d9a33d" dependencies = [ - "windows-targets", + "windows-targets 0.52.6", ] [[package]] @@ -4339,7 +4395,7 @@ version = "0.59.0" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "1e38bc4d79ed67fd075bcc251a1c39b32a1776bbe92e5bef1f0bf1f8c531853b" dependencies = [ - "windows-targets", + "windows-targets 0.52.6", ] [[package]] @@ -4348,14 +4404,30 @@ version = "0.52.6" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "9b724f72796e036ab90c1021d4780d4d3d648aca59e491e6b98e725b84e99973" dependencies = [ - "windows_aarch64_gnullvm", - "windows_aarch64_msvc", - "windows_i686_gnu", - "windows_i686_gnullvm", - "windows_i686_msvc", - "windows_x86_64_gnu", - "windows_x86_64_gnullvm", - "windows_x86_64_msvc", + "windows_aarch64_gnullvm 0.52.6", + "windows_aarch64_msvc 0.52.6", + "windows_i686_gnu 0.52.6", + "windows_i686_gnullvm 0.52.6", + "windows_i686_msvc 0.52.6", + "windows_x86_64_gnu 0.52.6", + "windows_x86_64_gnullvm 0.52.6", + "windows_x86_64_msvc 0.52.6", +] + +[[package]] +name = "windows-targets" +version = "0.53.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "b1e4c7e8ceaaf9cb7d7507c974735728ab453b67ef8f18febdd7c11fe59dca8b" +dependencies = [ + "windows_aarch64_gnullvm 0.53.0", + "windows_aarch64_msvc 0.53.0", + "windows_i686_gnu 0.53.0", + "windows_i686_gnullvm 0.53.0", + "windows_i686_msvc 0.53.0", + "windows_x86_64_gnu 0.53.0", + "windows_x86_64_gnullvm 0.53.0", + "windows_x86_64_msvc 0.53.0", ] [[package]] @@ -4364,68 +4436,110 @@ version = "0.52.6" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "32a4622180e7a0ec044bb555404c800bc9fd9ec262ec147edd5989ccd0c02cd3" +[[package]] +name = "windows_aarch64_gnullvm" +version = "0.53.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "86b8d5f90ddd19cb4a147a5fa63ca848db3df085e25fee3cc10b39b6eebae764" + [[package]] name = "windows_aarch64_msvc" version = "0.52.6" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "09ec2a7bb152e2252b53fa7803150007879548bc709c039df7627cabbd05d469" +[[package]] +name = "windows_aarch64_msvc" +version = "0.53.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "c7651a1f62a11b8cbd5e0d42526e55f2c99886c77e007179efff86c2b137e66c" + [[package]] name = "windows_i686_gnu" version = "0.52.6" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "8e9b5ad5ab802e97eb8e295ac6720e509ee4c243f69d781394014ebfe8bbfa0b" +[[package]] +name = "windows_i686_gnu" +version = "0.53.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "c1dc67659d35f387f5f6c479dc4e28f1d4bb90ddd1a5d3da2e5d97b42d6272c3" + [[package]] name = "windows_i686_gnullvm" version = "0.52.6" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "0eee52d38c090b3caa76c563b86c3a4bd71ef1a819287c19d586d7334ae8ed66" +[[package]] +name = "windows_i686_gnullvm" +version = "0.53.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "9ce6ccbdedbf6d6354471319e781c0dfef054c81fbc7cf83f338a4296c0cae11" + [[package]] name = "windows_i686_msvc" version = "0.52.6" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "240948bc05c5e7c6dabba28bf89d89ffce3e303022809e73deaefe4f6ec56c66" +[[package]] +name = "windows_i686_msvc" +version = "0.53.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "581fee95406bb13382d2f65cd4a908ca7b1e4c2f1917f143ba16efe98a589b5d" + [[package]] name = "windows_x86_64_gnu" version = "0.52.6" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "147a5c80aabfbf0c7d901cb5895d1de30ef2907eb21fbbab29ca94c5b08b1a78" +[[package]] +name = "windows_x86_64_gnu" +version = "0.53.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "2e55b5ac9ea33f2fc1716d1742db15574fd6fc8dadc51caab1c16a3d3b4190ba" + [[package]] name = "windows_x86_64_gnullvm" version = "0.52.6" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "24d5b23dc417412679681396f2b49f3de8c1473deb516bd34410872eff51ed0d" +[[package]] +name = "windows_x86_64_gnullvm" +version = "0.53.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "0a6e035dd0599267ce1ee132e51c27dd29437f63325753051e71dd9e42406c57" + [[package]] name = "windows_x86_64_msvc" version = "0.52.6" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "589f6da84c646204747d1270a2a5661ea66ed1cced2631d546fdfb155959f9ec" +[[package]] +name = "windows_x86_64_msvc" +version = "0.53.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "271414315aff87387382ec3d271b52d7ae78726f5d44ac98b4f4030c91880486" + [[package]] name = "wit-bindgen-rt" -version = "0.33.0" +version = "0.39.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "3268f3d866458b787f390cf61f4bbb563b922d091359f9608842999eaee3943c" +checksum = "6f42320e61fe2cfd34354ecb597f86f413484a798ba44a8ca1165c58d42da6c1" dependencies = [ "bitflags", ] -[[package]] -name = "write16" -version = "1.0.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "d1890f4022759daae28ed4fe62859b1236caebfc61ede2f63ed4e695f3f6d936" - [[package]] name = "writeable" -version = "0.5.5" +version = "0.6.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "1e9df38ee2d2c3c5948ea468a8406ff0db0b29ae1ffde1bcf20ef305bcc95c51" +checksum = "ea2f10b9bb0928dfb1b42b65e1f9e36f7f54dbdf08457afefb38afcdec4fa2bb" [[package]] name = "xz2" @@ -4438,9 +4552,9 @@ dependencies = [ [[package]] name = "yoke" -version = "0.7.5" +version = "0.8.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "120e6aef9aa629e3d4f52dc8cc43a015c7724194c97dfaf45180d2daf2b77f40" +checksum = "5f41bb01b8226ef4bfd589436a297c53d118f65921786300e427be8d487695cc" dependencies = [ "serde", "stable_deref_trait", @@ -4450,55 +4564,54 @@ dependencies = [ [[package]] name = "yoke-derive" -version = "0.7.5" +version = "0.8.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "2380878cad4ac9aac1e2435f3eb4020e8374b5f13c296cb75b4620ff8e229154" +checksum = "38da3c9736e16c5d3c8c597a9aaa5d1fa565d0532ae05e27c24aa62fb32c0ab6" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", "synstructure", ] [[package]] name = "zerocopy" -version = "0.7.35" +version = "0.8.25" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "1b9b4fd18abc82b8136838da5d50bae7bdea537c574d8dc1a34ed098d6c166f0" +checksum = "a1702d9583232ddb9174e01bb7c15a2ab8fb1bc6f227aa1233858c351a3ba0cb" dependencies = [ - "byteorder", "zerocopy-derive", ] [[package]] name = "zerocopy-derive" -version = "0.7.35" +version = "0.8.25" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "fa4f8080344d4671fb4e831a13ad1e68092748387dfc4f55e356242fae12ce3e" +checksum = "28a6e20d751156648aa063f3800b706ee209a32c0b4d9f24be3d980b01be55ef" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] name = "zerofrom" -version = "0.1.5" +version = "0.1.6" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "cff3ee08c995dee1859d998dea82f7374f2826091dd9cd47def953cae446cd2e" +checksum = "50cc42e0333e05660c3587f3bf9d0478688e15d870fab3346451ce7f8c9fbea5" dependencies = [ "zerofrom-derive", ] [[package]] name = "zerofrom-derive" -version = "0.1.5" +version = "0.1.6" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "595eed982f7d355beb85837f651fa22e90b3c044842dc7f2c2842c086f295808" +checksum = "d71e5d6e06ab090c67b5e44993ec16b72dcbaabc526db883a360057678b48502" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", "synstructure", ] @@ -4508,11 +4621,22 @@ version = "1.8.1" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "ced3678a2879b30306d323f4542626697a464a97c0a07c9aebf7ebca65cd4dde" +[[package]] +name = "zerotrie" +version = "0.2.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "36f0bbd478583f79edad978b407914f61b2972f5af6fa089686016be8f9af595" +dependencies = [ + "displaydoc", + "yoke", + "zerofrom", +] + [[package]] name = "zerovec" -version = "0.10.4" +version = "0.11.2" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "aa2b893d79df23bfb12d5461018d408ea19dfafe76c2c7ef6d4eba614f8ff079" +checksum = "4a05eb080e015ba39cc9e23bbe5e7fb04d5fb040350f99f34e338d5fdd294428" dependencies = [ "yoke", "zerofrom", @@ -4521,13 +4645,13 @@ dependencies = [ [[package]] name = "zerovec-derive" -version = "0.10.3" +version = "0.11.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "6eafa6dfb17584ea3e2bd6e76e0cc15ad7af12b09abdd1ca55961bed9b1063c6" +checksum = "5b96237efa0c878c64bd89c436f661be4e46b2f3eff1ebb976f7ef2321d2f58f" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -4538,27 +4662,27 @@ checksum = "868b928d7949e09af2f6086dfc1e01936064cc7a819253bce650d4e2a2d63ba8" [[package]] name = "zstd" -version = "0.13.2" +version = "0.13.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "fcf2b778a664581e31e389454a7072dab1647606d44f7feea22cd5abb9c9f3f9" +checksum = "e91ee311a569c327171651566e07972200e76fcfe2242a4fa446149a3881c08a" dependencies = [ "zstd-safe", ] [[package]] name = "zstd-safe" -version = "7.2.1" +version = "7.2.4" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "54a3ab4db68cea366acc5c897c7b4d4d1b8994a9cd6e6f841f8964566a419059" +checksum = "8f49c4d5f0abb602a93fb8736af2a4f4dd9512e36f7f570d66e65ff867ed3b9d" dependencies = [ "zstd-sys", ] [[package]] name = "zstd-sys" -version = "2.0.13+zstd.1.5.6" +version = "2.0.15+zstd.1.5.7" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "38ff0f21cfee8f97d94cef41359e0c89aa6113028ab0291aa8ca0038995a95aa" +checksum = "eb81183ddd97d0c74cedf1d50d85c8d08c1b8b68ee863bdee9e706eedba1a237" dependencies = [ "cc", "pkg-config", diff --git a/Cargo.toml b/Cargo.toml index 2c4188bb0..8107d76d3 100644 --- a/Cargo.toml +++ b/Cargo.toml @@ -17,7 +17,7 @@ [package] name = "datafusion-python" -version = "46.0.0" +version = "47.0.0" homepage = "https://datafusion.apache.org/python" repository = "https://github.com/apache/datafusion-python" authors = ["Apache DataFusion "] @@ -34,7 +34,7 @@ protoc = [ "datafusion-substrait/protoc" ] substrait = ["dep:datafusion-substrait"] [dependencies] -tokio = { version = "1.44", features = ["macros", "rt", "rt-multi-thread", "sync"] } +tokio = { version = "1.45", features = ["macros", "rt", "rt-multi-thread", "sync"] } pyo3 = { version = "0.24", features = ["extension-module", "abi3", "abi3-py39"] } pyo3-async-runtimes = { version = "0.24", features = ["tokio-runtime"]} arrow = { version = "55.0.0", features = ["pyarrow"] } @@ -47,7 +47,7 @@ uuid = { version = "1.16", features = ["v4"] } mimalloc = { version = "0.1", optional = true, default-features = false, features = ["local_dynamic_tls"] } async-trait = "0.1.88" futures = "0.3" -object_store = { version = "0.12.0", features = ["aws", "gcp", "azure", "http"] } +object_store = { version = "0.12.1", features = ["aws", "gcp", "azure", "http"] } url = "2" [build-dependencies] diff --git a/dev/changelog/47.0.0.md b/dev/changelog/47.0.0.md new file mode 100644 index 000000000..a7ed90313 --- /dev/null +++ b/dev/changelog/47.0.0.md @@ -0,0 +1,64 @@ + + +# Apache DataFusion Python 47.0.0 Changelog + +This release consists of 23 commits from 5 contributors. See credits at the end of this changelog for more information. + +**Implemented enhancements:** + +- feat: support unparser [#1088](https://github.com/apache/datafusion-python/pull/1088) (chenkovsky) +- feat: update datafusion dependency 47 [#1107](https://github.com/apache/datafusion-python/pull/1107) (timsaucer) +- feat: alias with metadata [#1111](https://github.com/apache/datafusion-python/pull/1111) (chenkovsky) +- feat: add missing PyLogicalPlan to_variant [#1085](https://github.com/apache/datafusion-python/pull/1085) (chenkovsky) +- feat: add user defined table function support [#1113](https://github.com/apache/datafusion-python/pull/1113) (timsaucer) + +**Fixed bugs:** + +- fix: recursive import [#1117](https://github.com/apache/datafusion-python/pull/1117) (chenkovsky) + +**Other:** + +- Update changelog and version number [#1089](https://github.com/apache/datafusion-python/pull/1089) (timsaucer) +- Documentation updates: mention correct dataset on basics page [#1081](https://github.com/apache/datafusion-python/pull/1081) (floscha) +- Add Configurable HTML Table Formatter for DataFusion DataFrames in Python [#1100](https://github.com/apache/datafusion-python/pull/1100) (kosiew) +- Add DataFrame usage guide with HTML rendering customization options [#1108](https://github.com/apache/datafusion-python/pull/1108) (kosiew) +- 1075/enhancement/Make col class with __getattr__ [#1076](https://github.com/apache/datafusion-python/pull/1076) (deanm0000) +- 1064/enhancement/add functions to Expr class [#1074](https://github.com/apache/datafusion-python/pull/1074) (deanm0000) +- ci: require approving review [#1122](https://github.com/apache/datafusion-python/pull/1122) (timsaucer) +- Partial fix for 1078: Enhance DataFrame Formatter Configuration with Memory and Display Controls [#1119](https://github.com/apache/datafusion-python/pull/1119) (kosiew) +- Add fill_null method to DataFrame API for handling missing values [#1019](https://github.com/apache/datafusion-python/pull/1019) (kosiew) +- minor: reduce error size [#1126](https://github.com/apache/datafusion-python/pull/1126) (timsaucer) +- Move the udf module to user_defined [#1112](https://github.com/apache/datafusion-python/pull/1112) (timsaucer) +- add unit tests for expression functions [#1121](https://github.com/apache/datafusion-python/pull/1121) (timsaucer) + +## Credits + +Thank you to everyone who contributed to this release. Here is a breakdown of commits (PRs merged) per contributor. + +``` + 12 Tim Saucer + 4 Chen Chongchen + 4 kosiew + 2 deanm0000 + 1 Florian Schäfer +``` + +Thank you also to everyone who contributed in other ways such as filing issues, reviewing PRs, and providing feedback on this release. + diff --git a/dev/release/README.md b/dev/release/README.md index f0b333999..692473930 100644 --- a/dev/release/README.md +++ b/dev/release/README.md @@ -56,6 +56,8 @@ Before creating a new release: - a PR should be created and merged to update the major version number of the project - A new release branch should be created, such as `branch-0.8` +## Preparing a Release Candidate + ### Change Log We maintain a `CHANGELOG.md` so our users know what has been changed between releases. @@ -76,21 +78,17 @@ Categorizing pull requests Generating changelog content ``` -This process is not fully automated, so there are some additional manual steps: - -- Add the ASF header to the generated file -- Add a link to this changelog from the top-level `/datafusion/CHANGELOG.md` -- Add the following content (copy from the previous version's changelog and update as appropriate: +### Update the version number -``` -## [24.0.0](https://github.com/apache/datafusion-python/tree/24.0.0) (2023-05-06) +The only place you should need to update the version is in the root `Cargo.toml`. -[Full Changelog](https://github.com/apache/datafusion-python/compare/23.0.0...24.0.0) -``` +### Tag the Repository -### Preparing a Release Candidate +Commit the changes to the changelog and version. -### Tag the Repository +Assuming you have set up a remote to the `apache` repository rather than your personal fork, +you need to push a tag to start the CI process for release candidates. The following assumes +the upstream repository is called `apache`. ```bash git tag 0.8.0-rc1 diff --git a/examples/datafusion-ffi-example/Cargo.lock b/examples/datafusion-ffi-example/Cargo.lock index 11096ac69..075ebd5a1 100644 --- a/examples/datafusion-ffi-example/Cargo.lock +++ b/examples/datafusion-ffi-example/Cargo.lock @@ -67,13 +67,13 @@ checksum = "512761e0bb2578dd7380c6baaa0f4ce03e84f95e960231d1dec8bf4d7d6e2627" [[package]] name = "ahash" -version = "0.8.11" +version = "0.8.12" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e89da841a80418a9b391ebaea17f5c112ffaaa96f621d2c285b5174da76b9011" +checksum = "5a15f179cd60c4584b8a8c596927aadc462e27f2ca70c04e0071964a73ba7a75" dependencies = [ "cfg-if", "const-random", - "getrandom 0.2.15", + "getrandom 0.3.3", "once_cell", "version_check", "zerocopy", @@ -105,9 +105,9 @@ dependencies = [ [[package]] name = "allocator-api2" -version = "0.2.20" +version = "0.2.21" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "45862d1c77f2228b9e10bc609d5bc203d86ebc9b87ad8d5d5167a6c9abf739d9" +checksum = "683d7910e743518b0e34f1186f92494becacb047c7b6bf616c96772180fef923" [[package]] name = "android-tzdata" @@ -126,9 +126,9 @@ dependencies = [ [[package]] name = "anyhow" -version = "1.0.93" +version = "1.0.98" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "4c95c10ba0b00a02636238b814946408b1322d5ac4760326e6fb8ec956d85775" +checksum = "e16d2d3311acee920a9eb8d33b8cbc1787ce4a264e85f964c2404b969bdcd487" [[package]] name = "arrayref" @@ -144,9 +144,9 @@ checksum = "7c02d123df017efcdfbd739ef81735b36c5ba83ec3c59c80a9d7ecc718f92e50" [[package]] name = "arrow" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "3095aaf545942ff5abd46654534f15b03a90fba78299d661e045e5d587222f0d" +checksum = "b1bb018b6960c87fd9d025009820406f74e83281185a8bdcb44880d2aa5c9a87" dependencies = [ "arrow-arith", "arrow-array", @@ -165,9 +165,9 @@ dependencies = [ [[package]] name = "arrow-arith" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "00752064ff47cee746e816ddb8450520c3a52cbad1e256f6fa861a35f86c45e7" +checksum = "44de76b51473aa888ecd6ad93ceb262fb8d40d1f1154a4df2f069b3590aa7575" dependencies = [ "arrow-array", "arrow-buffer", @@ -179,9 +179,9 @@ dependencies = [ [[package]] name = "arrow-array" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "cebfe926794fbc1f49ddd0cdaf898956ca9f6e79541efce62dabccfd81380472" +checksum = "29ed77e22744475a9a53d00026cf8e166fe73cf42d89c4c4ae63607ee1cfcc3f" dependencies = [ "ahash", "arrow-buffer", @@ -190,15 +190,15 @@ dependencies = [ "chrono", "chrono-tz", "half", - "hashbrown 0.15.1", + "hashbrown 0.15.3", "num", ] [[package]] name = "arrow-buffer" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "0303c7ec4cf1a2c60310fc4d6bbc3350cd051a17bf9e9c0a8e47b4db79277824" +checksum = "b0391c96eb58bf7389171d1e103112d3fc3e5625ca6b372d606f2688f1ea4cce" dependencies = [ "bytes", "half", @@ -207,9 +207,9 @@ dependencies = [ [[package]] name = "arrow-cast" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "335f769c5a218ea823d3760a743feba1ef7857cba114c01399a891c2fff34285" +checksum = "f39e1d774ece9292697fcbe06b5584401b26bd34be1bec25c33edae65c2420ff" dependencies = [ "arrow-array", "arrow-buffer", @@ -228,9 +228,9 @@ dependencies = [ [[package]] name = "arrow-csv" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "510db7dfbb4d5761826516cc611d97b3a68835d0ece95b034a052601109c0b1b" +checksum = "9055c972a07bf12c2a827debfd34f88d3b93da1941d36e1d9fee85eebe38a12a" dependencies = [ "arrow-array", "arrow-cast", @@ -244,9 +244,9 @@ dependencies = [ [[package]] name = "arrow-data" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e8affacf3351a24039ea24adab06f316ded523b6f8c3dbe28fbac5f18743451b" +checksum = "cf75ac27a08c7f48b88e5c923f267e980f27070147ab74615ad85b5c5f90473d" dependencies = [ "arrow-buffer", "arrow-schema", @@ -256,9 +256,9 @@ dependencies = [ [[package]] name = "arrow-ipc" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "69880a9e6934d9cba2b8630dd08a3463a91db8693b16b499d54026b6137af284" +checksum = "a222f0d93772bd058d1268f4c28ea421a603d66f7979479048c429292fac7b2e" dependencies = [ "arrow-array", "arrow-buffer", @@ -270,9 +270,9 @@ dependencies = [ [[package]] name = "arrow-json" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "d8dafd17a05449e31e0114d740530e0ada7379d7cb9c338fd65b09a8130960b0" +checksum = "9085342bbca0f75e8cb70513c0807cc7351f1fbf5cb98192a67d5e3044acb033" dependencies = [ "arrow-array", "arrow-buffer", @@ -292,9 +292,9 @@ dependencies = [ [[package]] name = "arrow-ord" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "895644523af4e17502d42c3cb6b27cb820f0cb77954c22d75c23a85247c849e1" +checksum = "ab2f1065a5cad7b9efa9e22ce5747ce826aa3855766755d4904535123ef431e7" dependencies = [ "arrow-array", "arrow-buffer", @@ -305,9 +305,9 @@ dependencies = [ [[package]] name = "arrow-row" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "9be8a2a4e5e7d9c822b2b8095ecd77010576d824f654d347817640acfc97d229" +checksum = "3703a0e3e92d23c3f756df73d2dc9476873f873a76ae63ef9d3de17fda83b2d8" dependencies = [ "arrow-array", "arrow-buffer", @@ -318,18 +318,18 @@ dependencies = [ [[package]] name = "arrow-schema" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "7450c76ab7c5a6805be3440dc2e2096010da58f7cab301fdc996a4ee3ee74e49" +checksum = "73a47aa0c771b5381de2b7f16998d351a6f4eb839f1e13d48353e17e873d969b" dependencies = [ "bitflags", ] [[package]] name = "arrow-select" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "aa5f5a93c75f46ef48e4001535e7b6c922eeb0aa20b73cf58d09e13d057490d8" +checksum = "24b7b85575702b23b85272b01bc1c25a01c9b9852305e5d0078c79ba25d995d4" dependencies = [ "ahash", "arrow-array", @@ -341,9 +341,9 @@ dependencies = [ [[package]] name = "arrow-string" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "6e7005d858d84b56428ba2a98a107fe88c0132c61793cf6b8232a1f9bfc0452b" +checksum = "9260fddf1cdf2799ace2b4c2fc0356a9789fa7551e0953e35435536fecefebbd" dependencies = [ "arrow-array", "arrow-buffer", @@ -402,7 +402,7 @@ checksum = "e539d3fca749fcee5236ab05e93a52867dd549cc157c8cb7f99595f3cedffdb5" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -422,9 +422,9 @@ checksum = "ace50bade8e6234aa140d9a2f552bbee1db4d353f69b8217bc503490fc1a9f26" [[package]] name = "backtrace" -version = "0.3.74" +version = "0.3.75" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "8d82cb332cdfaed17ae235a638438ac4d4839913cc2af585c3c6746e8f8bee1a" +checksum = "6806a6321ec58106fea15becdad98371e28d92ccbc7c8f1b3b6dd724fe8f1002" dependencies = [ "addr2line", "cfg-if", @@ -456,9 +456,9 @@ dependencies = [ [[package]] name = "bitflags" -version = "2.9.0" +version = "2.9.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "5c8214115b7bf84099f1309324e63141d4c5d7cc26862f97a0a857dbefe165bd" +checksum = "1b8e56985ec62d17e9c1001dc89c88ecd7dc08e47eba5ec7c29c7b5eeecde967" [[package]] name = "blake2" @@ -493,9 +493,9 @@ dependencies = [ [[package]] name = "brotli" -version = "7.0.0" +version = "8.0.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "cc97b8f16f944bba54f0433f07e30be199b6dc2bd25937444bbad560bcea29bd" +checksum = "9991eea70ea4f293524138648e41ee89b0b2b12ddef3b255effa43c8056e0e0d" dependencies = [ "alloc-no-stdlib", "alloc-stdlib", @@ -504,9 +504,9 @@ dependencies = [ [[package]] name = "brotli-decompressor" -version = "4.0.1" +version = "5.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "9a45bd2e4095a8b518033b128020dd4a55aab1c0a381ba4404a472630f4bc362" +checksum = "874bb8112abecc98cbd6d81ea4fa7e94fb9449648c93cc89aa40c81c24d7de03" dependencies = [ "alloc-no-stdlib", "alloc-stdlib", @@ -514,9 +514,9 @@ dependencies = [ [[package]] name = "bumpalo" -version = "3.16.0" +version = "3.17.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "79296716171880943b8470b5f8d03aa55eb2e645a4874bdbb28adb49162e012c" +checksum = "1628fb46dfa0b37568d12e5edd512553eccf6a22a78e8bde00bb4aed84d5bdbf" [[package]] name = "byteorder" @@ -551,9 +551,9 @@ dependencies = [ [[package]] name = "cc" -version = "1.1.37" +version = "1.2.23" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "40545c26d092346d8a8dab71ee48e7685a7a9cba76e634790c215b41a4a7b4cf" +checksum = "5f4ac86a9e5bc1e2b3449ab9d7d3a6a405e3d1bb28d7b9be8614f55846ae3766" dependencies = [ "jobserver", "libc", @@ -568,9 +568,9 @@ checksum = "baf1de4339761588bc0619e3cbc0120ee582ebb74b53b4efbf79117bd2da40fd" [[package]] name = "chrono" -version = "0.4.40" +version = "0.4.41" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "1a7964611d71df112cb1730f2ee67324fcf4d0fc6606acbbe9bfe06df124637c" +checksum = "c469d952047f47f91b68d1cba3f10d63c11d73e4636f24f08daf0278abf01c4d" dependencies = [ "android-tzdata", "iana-time-zone", @@ -580,9 +580,9 @@ dependencies = [ [[package]] name = "chrono-tz" -version = "0.10.0" +version = "0.10.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "cd6dd8046d00723a59a2f8c5f295c515b9bb9a331ee4f8f3d4dd49e428acd3b6" +checksum = "efdce149c370f133a071ca8ef6ea340b7b88748ab0810097a9e2976eaa34b4f3" dependencies = [ "chrono", "chrono-tz-build", @@ -591,9 +591,9 @@ dependencies = [ [[package]] name = "chrono-tz-build" -version = "0.4.0" +version = "0.4.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e94fea34d77a245229e7746bd2beb786cd2a896f306ff491fb8cecb3074b10a7" +checksum = "8f10f8c9340e31fc120ff885fcdb54a0b48e474bbd77cab557f0c30a3e569402" dependencies = [ "parse-zoneinfo", "phf_codegen", @@ -601,12 +601,11 @@ dependencies = [ [[package]] name = "comfy-table" -version = "7.1.1" +version = "7.1.4" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "b34115915337defe99b2aff5c2ce6771e5fbc4079f4b506301f5cf394c8452f7" +checksum = "4a65ebfec4fb190b6f90e944a817d60499ee0744e582530e2c9900a22e591d9a" dependencies = [ - "strum", - "strum_macros", + "unicode-segmentation", "unicode-width", ] @@ -625,16 +624,16 @@ version = "0.1.16" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "f9d839f2a20b0aee515dc581a6172f2321f96cab76c1a38a4c584a194955390e" dependencies = [ - "getrandom 0.2.15", + "getrandom 0.2.16", "once_cell", "tiny-keccak", ] [[package]] name = "const_panic" -version = "0.2.10" +version = "0.2.12" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "013b6c2c3a14d678f38cd23994b02da3a1a1b6a5d1eedddfe63a5a5f11b13a81" +checksum = "2459fc9262a1aa204eb4b5764ad4f189caec88aea9634389c0a25f8be7f6265e" [[package]] name = "constant_time_eq" @@ -665,9 +664,9 @@ checksum = "69f3b219d28b6e3b4ac87bc1fc522e0803ab22e055da177bff0068c4150c61a6" [[package]] name = "cpufeatures" -version = "0.2.14" +version = "0.2.17" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "608697df725056feaccfa42cffdaeeec3fccc4ffc38358ecd19b243e716a78e0" +checksum = "59ed5838eebb26a2bb2e58f6d5b5316989ae9d08bab10e0e6d103e656d1b0280" dependencies = [ "libc", ] @@ -683,24 +682,24 @@ dependencies = [ [[package]] name = "crossbeam-channel" -version = "0.5.13" +version = "0.5.15" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "33480d6946193aa8033910124896ca395333cae7e2d1113d1fef6c3272217df2" +checksum = "82b8f8f868b36967f9606790d1903570de9ceaf870a7bf9fbbd3016d636a2cb2" dependencies = [ "crossbeam-utils", ] [[package]] name = "crossbeam-utils" -version = "0.8.20" +version = "0.8.21" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "22ec99545bb0ed0ea7bb9b8e1e9122ea386ff8a48c0922e43f36d45ab09e0e80" +checksum = "d0a5c400df2834b80a4c3327b3aad3a4c4cd4de0629063962b03235697506a28" [[package]] name = "crunchy" -version = "0.2.2" +version = "0.2.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "7a81dae078cea95a014a339291cec439d2f232ebe854a9d672b796c6afafa9b7" +checksum = "43da5946c66ffcc7745f48db692ffbb10a83bfe0afd96235c5c2a4fb23994929" [[package]] name = "crypto-common" @@ -726,9 +725,9 @@ dependencies = [ [[package]] name = "csv-core" -version = "0.1.11" +version = "0.1.12" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "5efa2b3d7902f4b634a20cae3c9c4e6209dc4779feb6863329607560143efa70" +checksum = "7d02f3b0da4c6504f86e9cd789d8dbafab48c2321be74e9987593de5a894d93d" dependencies = [ "memchr", ] @@ -786,7 +785,7 @@ dependencies = [ "datafusion-sql", "flate2", "futures", - "itertools 0.14.0", + "itertools", "log", "object_store", "parking_lot", @@ -821,7 +820,7 @@ dependencies = [ "datafusion-session", "datafusion-sql", "futures", - "itertools 0.14.0", + "itertools", "log", "object_store", "parking_lot", @@ -909,7 +908,7 @@ dependencies = [ "flate2", "futures", "glob", - "itertools 0.14.0", + "itertools", "log", "object_store", "parquet", @@ -994,7 +993,7 @@ dependencies = [ "datafusion-physical-plan", "datafusion-session", "futures", - "itertools 0.14.0", + "itertools", "log", "object_store", "parking_lot", @@ -1058,7 +1057,7 @@ dependencies = [ "arrow", "datafusion-common", "indexmap", - "itertools 0.14.0", + "itertools", "paste", ] @@ -1101,7 +1100,7 @@ dependencies = [ "datafusion-expr-common", "datafusion-macros", "hex", - "itertools 0.14.0", + "itertools", "log", "md-5", "rand", @@ -1161,7 +1160,7 @@ dependencies = [ "datafusion-functions-aggregate", "datafusion-macros", "datafusion-physical-expr-common", - "itertools 0.14.0", + "itertools", "log", "paste", ] @@ -1217,7 +1216,7 @@ checksum = "df6f88d7ee27daf8b108ba910f9015176b36fbc72902b1ca5c2a5f1d1717e1a1" dependencies = [ "datafusion-expr", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -1232,7 +1231,7 @@ dependencies = [ "datafusion-expr", "datafusion-physical-expr", "indexmap", - "itertools 0.14.0", + "itertools", "log", "recursive", "regex", @@ -1255,7 +1254,7 @@ dependencies = [ "half", "hashbrown 0.14.5", "indexmap", - "itertools 0.14.0", + "itertools", "log", "paste", "petgraph", @@ -1272,7 +1271,7 @@ dependencies = [ "datafusion-common", "datafusion-expr-common", "hashbrown 0.14.5", - "itertools 0.14.0", + "itertools", ] [[package]] @@ -1289,7 +1288,7 @@ dependencies = [ "datafusion-physical-expr", "datafusion-physical-expr-common", "datafusion-physical-plan", - "itertools 0.14.0", + "itertools", "log", "recursive", ] @@ -1317,7 +1316,7 @@ dependencies = [ "half", "hashbrown 0.14.5", "indexmap", - "itertools 0.14.0", + "itertools", "log", "parking_lot", "pin-project-lite", @@ -1368,7 +1367,7 @@ dependencies = [ "datafusion-physical-plan", "datafusion-sql", "futures", - "itertools 0.14.0", + "itertools", "log", "object_store", "parking_lot", @@ -1411,36 +1410,36 @@ checksum = "97369cbbc041bc366949bc74d34658d6cda5621039731c6310521892a3a20ae0" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] name = "either" -version = "1.13.0" +version = "1.15.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "60b1af1c220855b6ceac025d3f6ecdd2b7c4894bfe9cd9bda4fbb4bc7c0d4cf0" +checksum = "48c757948c5ede0e46177b7add2e67155f70e33c07fea8284df6576da70b3719" [[package]] name = "equivalent" -version = "1.0.1" +version = "1.0.2" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "5443807d6dff69373d433ab9ef5378ad8df50ca6298caf15de6e52e24aaf54d5" +checksum = "877a4ace8713b0bcf2a4e7eec82529c029f1d0619886d18145fea96c3ffe5c0f" [[package]] name = "errno" -version = "0.3.9" +version = "0.3.12" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "534c5cf6194dfab3db3242765c03bbe257cf92f22b38f6bc0c58d59108a820ba" +checksum = "cea14ef9355e3beab063703aa9dab15afd25f0667c341310c1e5274bb1d0da18" dependencies = [ "libc", - "windows-sys 0.52.0", + "windows-sys", ] [[package]] name = "fastrand" -version = "2.2.0" +version = "2.3.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "486f806e73c5707928240ddc295403b1b93c96a02038563881c4a2fd84b81ac4" +checksum = "37909eebbb50d72f9059c3b6d82c0463f2ff062c9e95845c43a6c9c0355411be" [[package]] name = "ffi-table-provider" @@ -1553,7 +1552,7 @@ checksum = "162ee34ebcb7c64a8abebc059ce0fee27c2262618d7b60ed8faf72fef13c3650" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -1607,9 +1606,9 @@ dependencies = [ [[package]] name = "getrandom" -version = "0.2.15" +version = "0.2.16" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "c4567c8db10ae91089c99af84c68c38da3ec2f087c3f82960bcdbf3656b6f4d7" +checksum = "335ff9f135e4384c8150d6f27c6daed433577f86b4750418338c01a1a2528592" dependencies = [ "cfg-if", "libc", @@ -1618,9 +1617,9 @@ dependencies = [ [[package]] name = "getrandom" -version = "0.3.2" +version = "0.3.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "73fea8450eea4bac3940448fb7ae50d91f034f941199fcd9d909a5a07aa455f0" +checksum = "26145e563e54f2cadc477553f1ec5ee650b00862f0a58bcd12cbdc5f0ea2d2f4" dependencies = [ "cfg-if", "libc", @@ -1636,9 +1635,9 @@ checksum = "07e28edb80900c19c28f1072f2e8aeca7fa06b23cd4169cefe1af5aa3260783f" [[package]] name = "glob" -version = "0.3.1" +version = "0.3.2" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "d2fabcfbdc87f4758337ca535fb41a6d701b65693ce38287d856d1674551ec9b" +checksum = "a8d1add55171497b4705a648c6b583acafb01d58050a51727785f0b2c8e0a2b2" [[package]] name = "half" @@ -1663,9 +1662,9 @@ dependencies = [ [[package]] name = "hashbrown" -version = "0.15.1" +version = "0.15.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "3a9bfc1af68b1726ea47d3d5109de126281def866b33970e10fbab11b5dafab3" +checksum = "84b26c544d002229e640969970a2e74021aadf6e2f96372b9c58eff97de08eb3" [[package]] name = "heck" @@ -1692,20 +1691,21 @@ dependencies = [ [[package]] name = "humantime" -version = "2.1.0" +version = "2.2.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "9a3a5bfb195931eeb336b2a7b4d761daec841b97f947d34394601737a7bba5e4" +checksum = "9b112acc8b3adf4b107a8ec20977da0273a8c386765a3ec0229bd500a1443f9f" [[package]] name = "iana-time-zone" -version = "0.1.61" +version = "0.1.63" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "235e081f3925a06703c2d0117ea8b91f042756fd6e7a6e5d901e8ca1a996b220" +checksum = "b0c919e5debc312ad217002b8048a17b7d83f80703865bbfcfebb0458b0b27d8" dependencies = [ "android_system_properties", "core-foundation-sys", "iana-time-zone-haiku", "js-sys", + "log", "wasm-bindgen", "windows-core", ] @@ -1721,21 +1721,22 @@ dependencies = [ [[package]] name = "icu_collections" -version = "1.5.0" +version = "2.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "db2fa452206ebee18c4b5c2274dbf1de17008e874b4dc4f0aea9d01ca79e4526" +checksum = "200072f5d0e3614556f94a9930d5dc3e0662a652823904c3a75dc3b0af7fee47" dependencies = [ "displaydoc", + "potential_utf", "yoke", "zerofrom", "zerovec", ] [[package]] -name = "icu_locid" -version = "1.5.0" +name = "icu_locale_core" +version = "2.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "13acbb8371917fc971be86fc8057c41a64b521c184808a698c02acc242dbf637" +checksum = "0cde2700ccaed3872079a65fb1a78f6c0a36c91570f28755dda67bc8f7d9f00a" dependencies = [ "displaydoc", "litemap", @@ -1744,31 +1745,11 @@ dependencies = [ "zerovec", ] -[[package]] -name = "icu_locid_transform" -version = "1.5.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "01d11ac35de8e40fdeda00d9e1e9d92525f3f9d887cdd7aa81d727596788b54e" -dependencies = [ - "displaydoc", - "icu_locid", - "icu_locid_transform_data", - "icu_provider", - "tinystr", - "zerovec", -] - -[[package]] -name = "icu_locid_transform_data" -version = "1.5.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "fdc8ff3388f852bede6b579ad4e978ab004f139284d7b28715f773507b946f6e" - [[package]] name = "icu_normalizer" -version = "1.5.0" +version = "2.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "19ce3e0da2ec68599d193c93d088142efd7f9c5d6fc9b803774855747dc6a84f" +checksum = "436880e8e18df4d7bbc06d58432329d6458cc84531f7ac5f024e93deadb37979" dependencies = [ "displaydoc", "icu_collections", @@ -1776,67 +1757,54 @@ dependencies = [ "icu_properties", "icu_provider", "smallvec", - "utf16_iter", - "utf8_iter", - "write16", "zerovec", ] [[package]] name = "icu_normalizer_data" -version = "1.5.0" +version = "2.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "f8cafbf7aa791e9b22bec55a167906f9e1215fd475cd22adfcf660e03e989516" +checksum = "00210d6893afc98edb752b664b8890f0ef174c8adbb8d0be9710fa66fbbf72d3" [[package]] name = "icu_properties" -version = "1.5.1" +version = "2.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "93d6020766cfc6302c15dbbc9c8778c37e62c14427cb7f6e601d849e092aeef5" +checksum = "2549ca8c7241c82f59c80ba2a6f415d931c5b58d24fb8412caa1a1f02c49139a" dependencies = [ "displaydoc", "icu_collections", - "icu_locid_transform", + "icu_locale_core", "icu_properties_data", "icu_provider", - "tinystr", + "potential_utf", + "zerotrie", "zerovec", ] [[package]] name = "icu_properties_data" -version = "1.5.0" +version = "2.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "67a8effbc3dd3e4ba1afa8ad918d5684b8868b3b26500753effea8d2eed19569" +checksum = "8197e866e47b68f8f7d95249e172903bec06004b18b2937f1095d40a0c57de04" [[package]] name = "icu_provider" -version = "1.5.0" +version = "2.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "6ed421c8a8ef78d3e2dbc98a973be2f3770cb42b606e3ab18d6237c4dfde68d9" +checksum = "03c80da27b5f4187909049ee2d72f276f0d9f99a42c306bd0131ecfe04d8e5af" dependencies = [ "displaydoc", - "icu_locid", - "icu_provider_macros", + "icu_locale_core", "stable_deref_trait", "tinystr", "writeable", "yoke", "zerofrom", + "zerotrie", "zerovec", ] -[[package]] -name = "icu_provider_macros" -version = "1.5.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "1ec89e9337638ecdc08744df490b221a7399bf8d164eb52a665454e60e075ad6" -dependencies = [ - "proc-macro2", - "quote", - "syn 2.0.100", -] - [[package]] name = "idna" version = "1.0.3" @@ -1850,9 +1818,9 @@ dependencies = [ [[package]] name = "idna_adapter" -version = "1.2.0" +version = "1.2.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "daca1df1c957320b2cf139ac61e7bd64fed304c5040df000a745aa1de3b4ef71" +checksum = "3acae9609540aa318d1bc588455225fb2085b9ed0c4f6bd0d9d5bcd86f1a0344" dependencies = [ "icu_normalizer", "icu_properties", @@ -1865,14 +1833,14 @@ source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "cea70ddb795996207ad57735b50c5982d8844f38ba9ee5f1aedcfb708a2aa11e" dependencies = [ "equivalent", - "hashbrown 0.15.1", + "hashbrown 0.15.3", ] [[package]] name = "indoc" -version = "2.0.5" +version = "2.0.6" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "b248f5224d1d606005e02c97f5aa4e88eeb230488bcc03bc9ca4d7991399f2b5" +checksum = "f4c7245a08504955605670dbf141fceab975f15ca21570696aebe9d2e71576bd" [[package]] name = "integer-encoding" @@ -1880,15 +1848,6 @@ version = "3.0.4" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "8bb03732005da905c88227371639bf1ad885cc712789c011c31c5fb3ab3ccf02" -[[package]] -name = "itertools" -version = "0.13.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "413ee7dfc52ee1a4949ceeb7dbc8a33f2d6c088194d9f922fb8318faf1f01186" -dependencies = [ - "either", -] - [[package]] name = "itertools" version = "0.14.0" @@ -1900,25 +1859,27 @@ dependencies = [ [[package]] name = "itoa" -version = "1.0.11" +version = "1.0.15" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "49f1f14873335454500d59611f1cf4a4b0f786f9ac11f4312a78e4cf2566695b" +checksum = "4a5f13b858c8d314ee3e8f639011f7ccefe71f97f96e50151fb991f267928e2c" [[package]] name = "jobserver" -version = "0.1.32" +version = "0.1.33" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "48d1dbcbbeb6a7fec7e059840aa538bd62aaccf972c7346c4d9d2059312853d0" +checksum = "38f262f097c174adebe41eb73d66ae9c06b2844fb0da69969647bbddd9b0538a" dependencies = [ + "getrandom 0.3.3", "libc", ] [[package]] name = "js-sys" -version = "0.3.72" +version = "0.3.77" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "6a88f1bda2bd75b0452a14784937d796722fdebfe50df998aeb3f0b7603019a9" +checksum = "1cfaf33c695fc6e08064efbc1f72ec937429614f25eef83af942d0e227c3a28f" dependencies = [ + "once_cell", "wasm-bindgen", ] @@ -1930,9 +1891,9 @@ checksum = "bbd2bcb4c963f2ddae06a2efc7e9f3591312473c50c6685e1f298068316e66fe" [[package]] name = "lexical-core" -version = "1.0.2" +version = "1.0.5" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "0431c65b318a590c1de6b8fd6e72798c92291d27762d94c9e6c37ed7a73d8458" +checksum = "b765c31809609075565a70b4b71402281283aeda7ecaf4818ac14a7b2ade8958" dependencies = [ "lexical-parse-float", "lexical-parse-integer", @@ -1943,9 +1904,9 @@ dependencies = [ [[package]] name = "lexical-parse-float" -version = "1.0.2" +version = "1.0.5" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "eb17a4bdb9b418051aa59d41d65b1c9be5affab314a872e5ad7f06231fb3b4e0" +checksum = "de6f9cb01fb0b08060209a057c048fcbab8717b4c1ecd2eac66ebfe39a65b0f2" dependencies = [ "lexical-parse-integer", "lexical-util", @@ -1954,9 +1915,9 @@ dependencies = [ [[package]] name = "lexical-parse-integer" -version = "1.0.2" +version = "1.0.5" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "5df98f4a4ab53bf8b175b363a34c7af608fe31f93cc1fb1bf07130622ca4ef61" +checksum = "72207aae22fc0a121ba7b6d479e42cbfea549af1479c3f3a4f12c70dd66df12e" dependencies = [ "lexical-util", "static_assertions", @@ -1964,18 +1925,18 @@ dependencies = [ [[package]] name = "lexical-util" -version = "1.0.3" +version = "1.0.6" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "85314db53332e5c192b6bca611fb10c114a80d1b831ddac0af1e9be1b9232ca0" +checksum = "5a82e24bf537fd24c177ffbbdc6ebcc8d54732c35b50a3f28cc3f4e4c949a0b3" dependencies = [ "static_assertions", ] [[package]] name = "lexical-write-float" -version = "1.0.2" +version = "1.0.5" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "6e7c3ad4e37db81c1cbe7cf34610340adc09c322871972f74877a712abc6c809" +checksum = "c5afc668a27f460fb45a81a757b6bf2f43c2d7e30cb5a2dcd3abf294c78d62bd" dependencies = [ "lexical-util", "lexical-write-integer", @@ -1984,9 +1945,9 @@ dependencies = [ [[package]] name = "lexical-write-integer" -version = "1.0.2" +version = "1.0.5" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "eb89e9f6958b83258afa3deed90b5de9ef68eef090ad5086c791cd2345610162" +checksum = "629ddff1a914a836fb245616a7888b62903aae58fa771e1d83943035efa0f978" dependencies = [ "lexical-util", "static_assertions", @@ -2010,9 +1971,9 @@ dependencies = [ [[package]] name = "libm" -version = "0.2.11" +version = "0.2.15" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "8355be11b20d696c8f18f6cc018c4e372165b1fa8126cef092399c9951984ffa" +checksum = "f9fbbcab51052fe104eb5e5d351cf728d30a5be1fe14d9be8a3b097481fb97de" [[package]] name = "libz-rs-sys" @@ -2025,15 +1986,15 @@ dependencies = [ [[package]] name = "linux-raw-sys" -version = "0.4.14" +version = "0.9.4" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "78b3ae25bc7c8c38cec158d1f2757ee79e9b3740fbc7ccf0e59e4b08d793fa89" +checksum = "cd945864f07fe9f5371a27ad7b52a172b4b499999f1d97574c9fa68373937e12" [[package]] name = "litemap" -version = "0.7.3" +version = "0.8.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "643cb0b8d4fcc284004d5fd0d67ccf61dfffadb7f75e1e71bc420f4688a3a704" +checksum = "241eaef5fd12c88705a01fc1066c48c4b36e0dd4377dcdc7ec3942cea7a69956" [[package]] name = "lock_api" @@ -2047,9 +2008,9 @@ dependencies = [ [[package]] name = "log" -version = "0.4.22" +version = "0.4.27" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "a7a70ba024b9dc04c27ea2f0c0548feb474ec5c54bba33a7f72f873a39d07b24" +checksum = "13dc2df351e3202783a1fe0d44375f7295ffb4049267b0f3018346dc122a1d94" [[package]] name = "lz4_flex" @@ -2181,18 +2142,18 @@ dependencies = [ [[package]] name = "object" -version = "0.36.5" +version = "0.36.7" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "aedf0a2d09c573ed1d8d85b30c119153926a2b36dce0ab28322c09a117a4683e" +checksum = "62948e14d923ea95ea2c7c86c71013138b66525b86bdc08d2dcc262bdb497b87" dependencies = [ "memchr", ] [[package]] name = "object_store" -version = "0.12.0" +version = "0.12.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e9ce831b09395f933addbc56d894d889e4b226eba304d4e7adbab591e26daf1e" +checksum = "d94ac16b433c0ccf75326388c893d2835ab7457ea35ab8ba5d745c053ef5fa16" dependencies = [ "async-trait", "bytes", @@ -2200,7 +2161,7 @@ dependencies = [ "futures", "http", "humantime", - "itertools 0.14.0", + "itertools", "parking_lot", "percent-encoding", "thiserror", @@ -2208,13 +2169,15 @@ dependencies = [ "tracing", "url", "walkdir", + "wasm-bindgen-futures", + "web-time", ] [[package]] name = "once_cell" -version = "1.20.2" +version = "1.21.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "1261fe7e33c73b354eab43b1273a57c8f967d0391e80353e51f764ac02cf6775" +checksum = "42f5e15c9953c5e4ccceeb2e7382a716482c34515315f7b03532b8b4e8393d2d" [[package]] name = "ordered-float" @@ -2250,9 +2213,9 @@ dependencies = [ [[package]] name = "parquet" -version = "55.0.0" +version = "55.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "cd31a8290ac5b19f09ad77ee7a1e6a541f1be7674ad410547d5f1eef6eef4a9c" +checksum = "be7b2d778f6b841d37083ebdf32e33a524acde1266b5884a8ca29bf00dfa1231" dependencies = [ "ahash", "arrow-array", @@ -2269,7 +2232,7 @@ dependencies = [ "flate2", "futures", "half", - "hashbrown 0.15.1", + "hashbrown 0.15.3", "lz4_flex", "num", "num-bigint", @@ -2317,18 +2280,18 @@ dependencies = [ [[package]] name = "phf" -version = "0.11.2" +version = "0.11.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "ade2d8b8f33c7333b51bcf0428d37e217e9f32192ae4772156f65063b8ce03dc" +checksum = "1fd6780a80ae0c52cc120a26a1a42c1ae51b247a253e4e06113d23d2c2edd078" dependencies = [ "phf_shared", ] [[package]] name = "phf_codegen" -version = "0.11.2" +version = "0.11.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e8d39688d359e6b34654d328e262234662d16cc0f60ec8dcbe5e718709342a5a" +checksum = "aef8048c789fa5e851558d709946d6d79a8ff88c0440c587967f8e94bfb1216a" dependencies = [ "phf_generator", "phf_shared", @@ -2336,9 +2299,9 @@ dependencies = [ [[package]] name = "phf_generator" -version = "0.11.2" +version = "0.11.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "48e4cc64c2ad9ebe670cb8fd69dd50ae301650392e81c05f9bfcb2d5bdbc24b0" +checksum = "3c80231409c20246a13fddb31776fb942c38553c51e871f8cbd687a4cfb5843d" dependencies = [ "phf_shared", "rand", @@ -2346,18 +2309,18 @@ dependencies = [ [[package]] name = "phf_shared" -version = "0.11.2" +version = "0.11.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "90fcb95eef784c2ac79119d1dd819e162b5da872ce6f3c3abe1e8ca1c082f72b" +checksum = "67eabc2ef2a60eb7faa00097bd1ffdb5bd28e62bf39990626a582201b7a754e5" dependencies = [ "siphasher", ] [[package]] name = "pin-project-lite" -version = "0.2.15" +version = "0.2.16" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "915a1e146535de9163f3987b8944ed8cf49a18bb0056bcebcdcece385cece4ff" +checksum = "3b3cff922bd51709b605d9ead9aa71031d81447142d828eb4a6eba76fe619f9b" [[package]] name = "pin-utils" @@ -2367,21 +2330,30 @@ checksum = "8b870d8c151b6f2fb93e84a13146138f05d02ed11c7e7c54f8826aaaf7c9f184" [[package]] name = "pkg-config" -version = "0.3.31" +version = "0.3.32" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "953ec861398dccce10c670dfeaf3ec4911ca479e9c02154b3a215178c5f566f2" +checksum = "7edddbd0b52d732b21ad9a5fab5c704c14cd949e5e9a1ec5929a24fded1b904c" [[package]] name = "portable-atomic" -version = "1.9.0" +version = "1.11.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "350e9b48cbc6b0e028b0473b114454c6316e57336ee184ceab6e53f72c178b3e" + +[[package]] +name = "potential_utf" +version = "0.1.2" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "cc9c68a3f6da06753e9335d63e27f6b9754dd1920d941135b7ea8224f141adb2" +checksum = "e5a7c30837279ca13e7c867e9e40053bc68740f988cb07f7ca6df43cc734b585" +dependencies = [ + "zerovec", +] [[package]] name = "ppv-lite86" -version = "0.2.20" +version = "0.2.21" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "77957b295656769bb8ad2b6a6b09d897d94f05c41b069aede1fcdaa675eaea04" +checksum = "85eae3c4ed2f50dcfe72643da4befc30deadb458a9b590d720cde2f2b1e97da9" dependencies = [ "zerocopy", ] @@ -2397,9 +2369,9 @@ dependencies = [ [[package]] name = "prost" -version = "0.13.3" +version = "0.13.5" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "7b0487d90e047de87f984913713b85c601c05609aad5b0df4b4573fbf69aa13f" +checksum = "2796faa41db3ec313a31f7624d9286acf277b52de526150b7e69f3debf891ee5" dependencies = [ "bytes", "prost-derive", @@ -2407,31 +2379,31 @@ dependencies = [ [[package]] name = "prost-derive" -version = "0.13.3" +version = "0.13.5" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e9552f850d5f0964a4e4d0bf306459ac29323ddfbae05e35a7c0d35cb0803cc5" +checksum = "8a56d757972c98b346a9b766e3f02746cde6dd1cd1d1d563472929fdd74bec4d" dependencies = [ "anyhow", - "itertools 0.13.0", + "itertools", "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] name = "psm" -version = "0.1.24" +version = "0.1.26" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "200b9ff220857e53e184257720a14553b2f4aa02577d2ed9842d45d4b9654810" +checksum = "6e944464ec8536cd1beb0bbfd96987eb5e3b72f2ecdafdc5c769a37f1fa2ae1f" dependencies = [ "cc", ] [[package]] name = "pyo3" -version = "0.23.4" +version = "0.23.5" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "57fe09249128b3173d092de9523eaa75136bf7ba85e0d69eca241c7939c933cc" +checksum = "7778bffd85cf38175ac1f545509665d0b9b92a198ca7941f131f85f7a4f9a872" dependencies = [ "cfg-if", "indoc", @@ -2447,9 +2419,9 @@ dependencies = [ [[package]] name = "pyo3-build-config" -version = "0.23.4" +version = "0.23.5" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "1cd3927b5a78757a0d71aa9dff669f903b1eb64b54142a9bd9f757f8fde65fd7" +checksum = "94f6cbe86ef3bf18998d9df6e0f3fc1050a8c5efa409bf712e661a4366e010fb" dependencies = [ "once_cell", "target-lexicon", @@ -2457,9 +2429,9 @@ dependencies = [ [[package]] name = "pyo3-ffi" -version = "0.23.4" +version = "0.23.5" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "dab6bb2102bd8f991e7749f130a70d05dd557613e39ed2deeee8e9ca0c4d548d" +checksum = "e9f1b4c431c0bb1c8fb0a338709859eed0d030ff6daa34368d3b152a63dfdd8d" dependencies = [ "libc", "pyo3-build-config", @@ -2467,27 +2439,27 @@ dependencies = [ [[package]] name = "pyo3-macros" -version = "0.23.4" +version = "0.23.5" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "91871864b353fd5ffcb3f91f2f703a22a9797c91b9ab497b1acac7b07ae509c7" +checksum = "fbc2201328f63c4710f68abdf653c89d8dbc2858b88c5d88b0ff38a75288a9da" dependencies = [ "proc-macro2", "pyo3-macros-backend", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] name = "pyo3-macros-backend" -version = "0.23.4" +version = "0.23.5" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "43abc3b80bc20f3facd86cd3c60beed58c3e2aa26213f3cda368de39c60a27e4" +checksum = "fca6726ad0f3da9c9de093d6f116a93c1a38e417ed73bf138472cf4064f72028" dependencies = [ "heck", "proc-macro2", "pyo3-build-config", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -2532,7 +2504,7 @@ version = "0.6.4" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "ec0be4795e2f6a28069bec0b5ff3e2ac9bafc99e6a9a7dc3547996c5c816922c" dependencies = [ - "getrandom 0.2.15", + "getrandom 0.2.16", ] [[package]] @@ -2552,14 +2524,14 @@ source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "76009fbe0614077fc1a2ce255e3a1881a2e3a3527097d5dc6d8212c585e7e38b" dependencies = [ "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] name = "redox_syscall" -version = "0.5.7" +version = "0.5.12" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "9b6dfecf2c74bce2466cabf93f6664d6998a69eb21e39f4207930065b27b771f" +checksum = "928fca9cf2aa042393a8325b9ead81d2f0df4cb12e1e24cef072922ccd99c5af" dependencies = [ "bitflags", ] @@ -2578,9 +2550,9 @@ dependencies = [ [[package]] name = "regex-automata" -version = "0.4.8" +version = "0.4.9" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "368758f23274712b504848e9d5a6f010445cc8b87a7cdb4d7cbee666c1288da3" +checksum = "809e8dc61f6de73b46c85f4c96486310fe304c434cfa43669d7b40f711150908" dependencies = [ "aho-corasick", "memchr", @@ -2619,28 +2591,28 @@ dependencies = [ [[package]] name = "rustix" -version = "0.38.40" +version = "1.0.7" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "99e4ea3e1cdc4b559b8e5650f9c8e5998e3e5c1343b4eaf034565f32318d63c0" +checksum = "c71e83d6afe7ff64890ec6b71d6a69bb8a610ab78ce364b3352876bb4c801266" dependencies = [ "bitflags", "errno", "libc", "linux-raw-sys", - "windows-sys 0.52.0", + "windows-sys", ] [[package]] name = "rustversion" -version = "1.0.18" +version = "1.0.20" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "0e819f2bc632f285be6d7cd36e25940d45b2391dd6d9b939e79de557f7014248" +checksum = "eded382c5f5f786b989652c49544c4877d9f015cc22e145a5ea8ea66c2921cd2" [[package]] name = "ryu" -version = "1.0.18" +version = "1.0.20" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "f3cb5ba0dc43242ce17de99c180e96db90b235b8a9fdc9543c96d2209116bd9f" +checksum = "28d3b2b1366ec20994f1fd18c3c594f05c5dd4bc44d8bb0c1c632c8d6829481f" [[package]] name = "same-file" @@ -2665,35 +2637,35 @@ checksum = "56e6fa9c48d24d85fb3de5ad847117517440f6beceb7798af16b4a87d616b8d0" [[package]] name = "seq-macro" -version = "0.3.5" +version = "0.3.6" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "a3f0bf26fd526d2a95683cd0f87bf103b8539e2ca1ef48ce002d67aad59aa0b4" +checksum = "1bc711410fbe7399f390ca1c3b60ad0f53f80e95c5eb935e52268a0e2cd49acc" [[package]] name = "serde" -version = "1.0.214" +version = "1.0.219" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "f55c3193aca71c12ad7890f1785d2b73e1b9f63a0bbc353c08ef26fe03fc56b5" +checksum = "5f0e2c6ed6606019b4e29e69dbaba95b11854410e5347d525002456dbbb786b6" dependencies = [ "serde_derive", ] [[package]] name = "serde_derive" -version = "1.0.214" +version = "1.0.219" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "de523f781f095e28fa605cdce0f8307e451cc0fd14e2eb4cd2e98a355b147766" +checksum = "5b0276cf7f2c73365f7157c8123c21cd9a50fbbd844757af28ca1f5925fc2a00" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] name = "serde_json" -version = "1.0.132" +version = "1.0.140" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "d726bfaff4b320266d395898905d0eba0345aae23b54aee3a737e260fd46db03" +checksum = "20068b6e96dc6c9bd23e01df8827e6c7e1f2fddd43c21810382803c136b99373" dependencies = [ "itoa", "memchr", @@ -2703,9 +2675,9 @@ dependencies = [ [[package]] name = "sha2" -version = "0.10.8" +version = "0.10.9" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "793db75ad2bcafc3ffa7c68b215fee268f537982cd901d132f89c6343f3a3dc8" +checksum = "a7507d819769d01a365ab707794a4084392c824f54a7a6a7862f8c3d0892b283" dependencies = [ "cfg-if", "cpufeatures", @@ -2726,9 +2698,9 @@ checksum = "e3a9fe34e3e7a50316060351f37187a3f546bce95496156754b601a5fa71b76e" [[package]] name = "siphasher" -version = "0.3.11" +version = "1.0.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "38b58827f4464d87d377d175e90bf58eb00fd8716ff0a62f80356b5e61555d0d" +checksum = "56199f7ddabf13fe5074ce809e7d3f42b42ae711800501b5b16ea82ad029c39d" [[package]] name = "slab" @@ -2741,9 +2713,9 @@ dependencies = [ [[package]] name = "smallvec" -version = "1.13.2" +version = "1.15.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "3c5e1a9a646d36c3599cd173a41282daf47c44583ad367b8e6837255952e5c67" +checksum = "8917285742e9f3e1683f0a9c4e6b57960b7314d0b08d30d1ecd426713ee2eee9" [[package]] name = "snap" @@ -2770,7 +2742,7 @@ checksum = "da5fc6819faabb412da764b99d3b713bb55083c11e7e0c00144d386cd6a1939c" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -2781,15 +2753,15 @@ checksum = "a8f112729512f8e442d81f95a8a7ddf2b7c6b8a1a6f509a95864142b30cab2d3" [[package]] name = "stacker" -version = "0.1.17" +version = "0.1.21" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "799c883d55abdb5e98af1a7b3f23b9b6de8ecada0ecac058672d7635eb48ca7b" +checksum = "cddb07e32ddb770749da91081d8d0ac3a16f1a569a18b20348cd371f5dead06b" dependencies = [ "cc", "cfg-if", "libc", "psm", - "windows-sys 0.59.0", + "windows-sys", ] [[package]] @@ -2798,25 +2770,6 @@ version = "1.1.0" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "a2eb9349b6444b326872e140eb1cf5e7c522154d69e7a0ffb0fb81c06b37543f" -[[package]] -name = "strum" -version = "0.26.3" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "8fec0f0aef304996cf250b31b5a10dee7980c85da9d759361292b8bca5a18f06" - -[[package]] -name = "strum_macros" -version = "0.26.4" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "4c6bee85a5a24955dc440386795aa378cd9cf82acd5f764469152d2270e581be" -dependencies = [ - "heck", - "proc-macro2", - "quote", - "rustversion", - "syn 2.0.100", -] - [[package]] name = "subtle" version = "2.6.1" @@ -2836,9 +2789,9 @@ dependencies = [ [[package]] name = "syn" -version = "2.0.100" +version = "2.0.101" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "b09a44accad81e1ba1cd74a32461ba89dee89095ba17b32f5d03683b1b1fc2a0" +checksum = "8ce2b7fc941b3a24138a0a7cf8e858bfc6a992e7978a068a5c760deb0ed43caf" dependencies = [ "proc-macro2", "quote", @@ -2847,13 +2800,13 @@ dependencies = [ [[package]] name = "synstructure" -version = "0.13.1" +version = "0.13.2" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "c8af7666ab7b6390ab78131fb5b0fce11d6b7a6951602017c35fa82800708971" +checksum = "728a70f3dbaf5bab7f0c4b1ac8d7ae5ea60a4b5549c8a5914361c99147a709d2" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -2864,15 +2817,15 @@ checksum = "61c41af27dd6d1e27b1b16b489db798443478cef1f06a660c96db617ba5de3b1" [[package]] name = "tempfile" -version = "3.14.0" +version = "3.20.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "28cce251fcbc87fac86a866eeb0d6c2d536fc16d06f184bb61aeae11aa4cee0c" +checksum = "e8a64e3985349f2441a1a9ef0b853f869006c3855f2cda6862a94d26ebb9d6a1" dependencies = [ - "cfg-if", "fastrand", + "getrandom 0.3.3", "once_cell", "rustix", - "windows-sys 0.59.0", + "windows-sys", ] [[package]] @@ -2892,7 +2845,7 @@ checksum = "7f7cf42b4507d8ea322120659672cf1b9dbb93f8f2d4ecfd6e51350ff5b17a1d" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -2917,9 +2870,9 @@ dependencies = [ [[package]] name = "tinystr" -version = "0.7.6" +version = "0.8.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "9117f5d4db391c1cf6927e7bea3db74b9a1c1add8f7eda9ffd5364f40f57b82f" +checksum = "5d4f6d1145dcb577acf783d4e601bc1d76a13337bb54e6233add580b07344c8b" dependencies = [ "displaydoc", "zerovec", @@ -2927,9 +2880,9 @@ dependencies = [ [[package]] name = "tokio" -version = "1.44.2" +version = "1.45.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e6b88822cbe49de4185e3a4cbf8321dd487cf5fe0c5c65695fef6346371e9c48" +checksum = "2513ca694ef9ede0fb23fe71a4ee4107cb102b9dc1930f6d0fd77aae068ae165" dependencies = [ "backtrace", "bytes", @@ -2945,7 +2898,7 @@ checksum = "6e06d43f1345a3bcd39f6a56dbb7dcab2ba47e68e8ac134855e7e2bdbaf8cab8" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -2963,9 +2916,9 @@ dependencies = [ [[package]] name = "tracing" -version = "0.1.40" +version = "0.1.41" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "c3523ab5a71916ccf420eebdf5521fcef02141234bbc0b8a49f2fdc4544364ef" +checksum = "784e0ac535deb450455cbfa28a6f0df145ea1bb7ae51b821cf5e7927fdcfbdd0" dependencies = [ "pin-project-lite", "tracing-attributes", @@ -2974,20 +2927,20 @@ dependencies = [ [[package]] name = "tracing-attributes" -version = "0.1.27" +version = "0.1.28" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "34704c8d6ebcbc939824180af020566b01a7c01f80641264eba0999f6c2b6be7" +checksum = "395ae124c09f9e6918a2310af6038fba074bcf474ac352496d5910dd59a2226d" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] name = "tracing-core" -version = "0.1.32" +version = "0.1.33" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "c06d3da6113f116aaee68e4d601191614c9053067f9ab7f6edbcb161237daa54" +checksum = "e672c95779cf947c5311f83787af4fa8fffd12fb27e4993211a84bdfd9610f9c" dependencies = [ "once_cell", ] @@ -3031,15 +2984,15 @@ checksum = "6af6ae20167a9ece4bcb41af5b80f8a1f1df981f6391189ce00fd257af04126a" [[package]] name = "typenum" -version = "1.17.0" +version = "1.18.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "42ff0bf0c66b8238c6f3b578df37d0b7848e55df8577b3f74f92a69acceeb825" +checksum = "1dccffe3ce07af9386bfd29e80c0ab1a8205a2fc34e4bcd40364df902cfa8f3f" [[package]] name = "unicode-ident" -version = "1.0.13" +version = "1.0.18" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e91b56cd4cadaeb79bbf1a5645f6b4f8dc5bde8834ad5894a8db35fda9efa1fe" +checksum = "5a5f39404a5da50712a4c1eecf25e90dd62b613502b7e925fd4e4d19b5c96512" [[package]] name = "unicode-segmentation" @@ -3049,15 +3002,15 @@ checksum = "f6ccf251212114b54433ec949fd6a7841275f9ada20dddd2f29e9ceea4501493" [[package]] name = "unicode-width" -version = "0.1.14" +version = "0.2.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "7dd6e30e90baa6f72411720665d41d89b9a3d039dc45b8faea1ddd07f617f6af" +checksum = "1fc81956842c57dac11422a97c3b8195a1ff727f06e85c84ed2e8aa277c9a0fd" [[package]] name = "unindent" -version = "0.2.3" +version = "0.2.4" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "c7de7d73e1754487cb58364ee906a499937a0dfabd86bcb980fa99ec8c8fa2ce" +checksum = "7264e107f553ccae879d21fbea1d6724ac785e8c3bfc762137959b5802826ef3" [[package]] name = "url" @@ -3070,12 +3023,6 @@ dependencies = [ "percent-encoding", ] -[[package]] -name = "utf16_iter" -version = "1.0.5" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "c8232dd3cdaed5356e0f716d285e4b40b932ac434100fe9b7e0e8e935b9e6246" - [[package]] name = "utf8_iter" version = "1.0.4" @@ -3088,7 +3035,7 @@ version = "1.16.0" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "458f7a779bf54acc9f347480ac654f68407d3aab21269a6e3c9f922acd9e2da9" dependencies = [ - "getrandom 0.3.2", + "getrandom 0.3.3", "js-sys", "wasm-bindgen", ] @@ -3126,35 +3073,48 @@ dependencies = [ [[package]] name = "wasm-bindgen" -version = "0.2.95" +version = "0.2.100" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "128d1e363af62632b8eb57219c8fd7877144af57558fb2ef0368d0087bddeb2e" +checksum = "1edc8929d7499fc4e8f0be2262a241556cfc54a0bea223790e71446f2aab1ef5" dependencies = [ "cfg-if", "once_cell", + "rustversion", "wasm-bindgen-macro", ] [[package]] name = "wasm-bindgen-backend" -version = "0.2.95" +version = "0.2.100" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "cb6dd4d3ca0ddffd1dd1c9c04f94b868c37ff5fac97c30b97cff2d74fce3a358" +checksum = "2f0a0651a5c2bc21487bde11ee802ccaf4c51935d0d3d42a6101f98161700bc6" dependencies = [ "bumpalo", "log", - "once_cell", "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", "wasm-bindgen-shared", ] +[[package]] +name = "wasm-bindgen-futures" +version = "0.4.50" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "555d470ec0bc3bb57890405e5d4322cc9ea83cebb085523ced7be4144dac1e61" +dependencies = [ + "cfg-if", + "js-sys", + "once_cell", + "wasm-bindgen", + "web-sys", +] + [[package]] name = "wasm-bindgen-macro" -version = "0.2.95" +version = "0.2.100" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e79384be7f8f5a9dd5d7167216f022090cf1f9ec128e6e6a482a2cb5c5422c56" +checksum = "7fe63fc6d09ed3792bd0897b314f53de8e16568c2b3f7982f468c0bf9bd0b407" dependencies = [ "quote", "wasm-bindgen-macro-support", @@ -3162,22 +3122,35 @@ dependencies = [ [[package]] name = "wasm-bindgen-macro-support" -version = "0.2.95" +version = "0.2.100" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "26c6ab57572f7a24a4985830b120de1594465e5d500f24afe89e16b4e833ef68" +checksum = "8ae87ea40c9f689fc23f209965b6fb8a99ad69aeeb0231408be24920604395de" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", "wasm-bindgen-backend", "wasm-bindgen-shared", ] [[package]] name = "wasm-bindgen-shared" -version = "0.2.95" +version = "0.2.100" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "1a05d73b933a847d6cccdda8f838a22ff101ad9bf93e33684f39c1f5f0eece3d" +dependencies = [ + "unicode-ident", +] + +[[package]] +name = "web-sys" +version = "0.3.77" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "65fc09f10666a9f147042251e0dda9c18f166ff7de300607007e96bdebc1068d" +checksum = "33b6dd2ef9186f1f2072e409e99cd22a975331a6b3591b12c764e0e55c60d5d2" +dependencies = [ + "js-sys", + "wasm-bindgen", +] [[package]] name = "web-time" @@ -3211,7 +3184,7 @@ version = "0.1.9" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "cf221c93e13a30d793f7645a0e7762c55d169dbb0a49671918a2319d289b10bb" dependencies = [ - "windows-sys 0.59.0", + "windows-sys", ] [[package]] @@ -3222,11 +3195,37 @@ checksum = "712e227841d057c1ee1cd2fb22fa7e5a5461ae8e48fa2ca79ec42cfc1931183f" [[package]] name = "windows-core" -version = "0.52.0" +version = "0.61.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "33ab640c8d7e35bf8ba19b884ba838ceb4fba93a4e8c65a9059d08afcfc683d9" +checksum = "46ec44dc15085cea82cf9c78f85a9114c463a369786585ad2882d1ff0b0acf40" dependencies = [ - "windows-targets", + "windows-implement", + "windows-interface", + "windows-link", + "windows-result", + "windows-strings", +] + +[[package]] +name = "windows-implement" +version = "0.60.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "a47fddd13af08290e67f4acabf4b459f647552718f683a7b415d290ac744a836" +dependencies = [ + "proc-macro2", + "quote", + "syn 2.0.101", +] + +[[package]] +name = "windows-interface" +version = "0.59.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "bd9211b69f8dcdfa817bfd14bf1c97c9188afa36f4750130fcdf3f400eca9fa8" +dependencies = [ + "proc-macro2", + "quote", + "syn 2.0.101", ] [[package]] @@ -3236,12 +3235,21 @@ source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "76840935b766e1b0a05c0066835fb9ec80071d4c09a16f6bd5f7e655e3c14c38" [[package]] -name = "windows-sys" -version = "0.52.0" +name = "windows-result" +version = "0.3.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "282be5f36a8ce781fad8c8ae18fa3f9beff57ec1b52cb3de0789201425d9a33d" +checksum = "4b895b5356fc36103d0f64dd1e94dfa7ac5633f1c9dd6e80fe9ec4adef69e09d" dependencies = [ - "windows-targets", + "windows-link", +] + +[[package]] +name = "windows-strings" +version = "0.4.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "2a7ab927b2637c19b3dbe0965e75d8f2d30bdd697a1516191cad2ec4df8fb28a" +dependencies = [ + "windows-link", ] [[package]] @@ -3326,17 +3334,11 @@ dependencies = [ "bitflags", ] -[[package]] -name = "write16" -version = "1.0.0" -source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "d1890f4022759daae28ed4fe62859b1236caebfc61ede2f63ed4e695f3f6d936" - [[package]] name = "writeable" -version = "0.5.5" +version = "0.6.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "1e9df38ee2d2c3c5948ea468a8406ff0db0b29ae1ffde1bcf20ef305bcc95c51" +checksum = "ea2f10b9bb0928dfb1b42b65e1f9e36f7f54dbdf08457afefb38afcdec4fa2bb" [[package]] name = "xz2" @@ -3349,9 +3351,9 @@ dependencies = [ [[package]] name = "yoke" -version = "0.7.4" +version = "0.8.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "6c5b1314b079b0930c31e3af543d8ee1757b1951ae1e1565ec704403a7240ca5" +checksum = "5f41bb01b8226ef4bfd589436a297c53d118f65921786300e427be8d487695cc" dependencies = [ "serde", "stable_deref_trait", @@ -3361,63 +3363,73 @@ dependencies = [ [[package]] name = "yoke-derive" -version = "0.7.4" +version = "0.8.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "28cc31741b18cb6f1d5ff12f5b7523e3d6eb0852bbbad19d73905511d9849b95" +checksum = "38da3c9736e16c5d3c8c597a9aaa5d1fa565d0532ae05e27c24aa62fb32c0ab6" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", "synstructure", ] [[package]] name = "zerocopy" -version = "0.7.35" +version = "0.8.25" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "1b9b4fd18abc82b8136838da5d50bae7bdea537c574d8dc1a34ed098d6c166f0" +checksum = "a1702d9583232ddb9174e01bb7c15a2ab8fb1bc6f227aa1233858c351a3ba0cb" dependencies = [ - "byteorder", "zerocopy-derive", ] [[package]] name = "zerocopy-derive" -version = "0.7.35" +version = "0.8.25" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "fa4f8080344d4671fb4e831a13ad1e68092748387dfc4f55e356242fae12ce3e" +checksum = "28a6e20d751156648aa063f3800b706ee209a32c0b4d9f24be3d980b01be55ef" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] name = "zerofrom" -version = "0.1.4" +version = "0.1.6" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "91ec111ce797d0e0784a1116d0ddcdbea84322cd79e5d5ad173daeba4f93ab55" +checksum = "50cc42e0333e05660c3587f3bf9d0478688e15d870fab3346451ce7f8c9fbea5" dependencies = [ "zerofrom-derive", ] [[package]] name = "zerofrom-derive" -version = "0.1.4" +version = "0.1.6" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "0ea7b4a3637ea8669cedf0f1fd5c286a17f3de97b8dd5a70a6c167a1730e63a5" +checksum = "d71e5d6e06ab090c67b5e44993ec16b72dcbaabc526db883a360057678b48502" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", "synstructure", ] +[[package]] +name = "zerotrie" +version = "0.2.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "36f0bbd478583f79edad978b407914f61b2972f5af6fa089686016be8f9af595" +dependencies = [ + "displaydoc", + "yoke", + "zerofrom", +] + [[package]] name = "zerovec" -version = "0.10.4" +version = "0.11.2" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "aa2b893d79df23bfb12d5461018d408ea19dfafe76c2c7ef6d4eba614f8ff079" +checksum = "4a05eb080e015ba39cc9e23bbe5e7fb04d5fb040350f99f34e338d5fdd294428" dependencies = [ "yoke", "zerofrom", @@ -3426,13 +3438,13 @@ dependencies = [ [[package]] name = "zerovec-derive" -version = "0.10.3" +version = "0.11.1" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "6eafa6dfb17584ea3e2bd6e76e0cc15ad7af12b09abdd1ca55961bed9b1063c6" +checksum = "5b96237efa0c878c64bd89c436f661be4e46b2f3eff1ebb976f7ef2321d2f58f" dependencies = [ "proc-macro2", "quote", - "syn 2.0.100", + "syn 2.0.101", ] [[package]] @@ -3443,27 +3455,27 @@ checksum = "868b928d7949e09af2f6086dfc1e01936064cc7a819253bce650d4e2a2d63ba8" [[package]] name = "zstd" -version = "0.13.2" +version = "0.13.3" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "fcf2b778a664581e31e389454a7072dab1647606d44f7feea22cd5abb9c9f3f9" +checksum = "e91ee311a569c327171651566e07972200e76fcfe2242a4fa446149a3881c08a" dependencies = [ "zstd-safe", ] [[package]] name = "zstd-safe" -version = "7.2.1" +version = "7.2.4" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "54a3ab4db68cea366acc5c897c7b4d4d1b8994a9cd6e6f841f8964566a419059" +checksum = "8f49c4d5f0abb602a93fb8736af2a4f4dd9512e36f7f570d66e65ff867ed3b9d" dependencies = [ "zstd-sys", ] [[package]] name = "zstd-sys" -version = "2.0.13+zstd.1.5.6" +version = "2.0.15+zstd.1.5.7" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "38ff0f21cfee8f97d94cef41359e0c89aa6113028ab0291aa8ca0038995a95aa" +checksum = "eb81183ddd97d0c74cedf1d50d85c8d08c1b8b68ee863bdee9e706eedba1a237" dependencies = [ "cc", "pkg-config", diff --git a/src/dataframe.rs b/src/dataframe.rs index cc226a0a1..ece8c4e0f 100644 --- a/src/dataframe.rs +++ b/src/dataframe.rs @@ -181,9 +181,7 @@ fn build_formatter_config_from_python(formatter: &Bound<'_, PyAny>) -> PyResult< }; // Return the validated config, converting String error to PyErr - config - .validate() - .map_err(|e| pyo3::exceptions::PyValueError::new_err(e))?; + config.validate().map_err(PyValueError::new_err)?; Ok(config) } From 24f0b1a1eee1c83ff8103ce49bb3ca6fad13b811 Mon Sep 17 00:00:00 2001 From: Crystal Zhou <45134936+crystalxyz@users.noreply.github.com> Date: Mon, 9 Jun 2025 15:08:46 -0400 Subject: [PATCH 002/220] Add a documentation build step in CI (#1139) --- .github/workflows/test.yaml | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/.github/workflows/test.yaml b/.github/workflows/test.yaml index 31c2febea..4ae081406 100644 --- a/.github/workflows/test.yaml +++ b/.github/workflows/test.yaml @@ -80,6 +80,13 @@ jobs: with: enable-cache: true + - name: Check documentation + if: ${{ matrix.python-version == '3.10' && matrix.toolchain == 'stable' }} + run: | + uv sync --dev --group docs --no-install-package datafusion + uv run --no-project maturin develop --uv + uv run --no-project docs/build.sh + - name: Run tests env: RUST_BACKTRACE: 1 From d6ef9bcb33857b4a350fbe749b93b27345bc817b Mon Sep 17 00:00:00 2001 From: kosiew Date: Sat, 14 Jun 2025 00:05:54 +0800 Subject: [PATCH 003/220] Add DataFrame API Documentation for DataFusion Python (#1132) * feat: add API reference documentation for DataFrame and index * feat: add tests for validating RST syntax, code blocks, and internal links in DataFrame API documentation * refactor: remove test script for DataFrame API documentation in RST format * fix: correct formatting inconsistencies in dataframe.rst * fix: correct header formatting in functions.rst * fix: adjust formatting for code block in dataframe.rst * fix: skip documentation for duplicate modules in autoapi configuration * fix: add cross reference to io pages --- docs/source/api/dataframe.rst | 387 ++++++++++++++++++ docs/source/api/index.rst | 27 ++ docs/source/conf.py | 3 + docs/source/index.rst | 2 + .../common-operations/functions.rst | 2 +- docs/source/user-guide/dataframe.rst | 3 +- 6 files changed, 422 insertions(+), 2 deletions(-) create mode 100644 docs/source/api/dataframe.rst create mode 100644 docs/source/api/index.rst diff --git a/docs/source/api/dataframe.rst b/docs/source/api/dataframe.rst new file mode 100644 index 000000000..a9e9e47c8 --- /dev/null +++ b/docs/source/api/dataframe.rst @@ -0,0 +1,387 @@ +.. Licensed to the Apache Software Foundation (ASF) under one +.. or more contributor license agreements. See the NOTICE file +.. distributed with this work for additional information +.. regarding copyright ownership. The ASF licenses this file +.. to you under the Apache License, Version 2.0 (the +.. "License"); you may not use this file except in compliance +.. with the License. You may obtain a copy of the License at + +.. http://www.apache.org/licenses/LICENSE-2.0 + +.. Unless required by applicable law or agreed to in writing, +.. software distributed under the License is distributed on an +.. "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +.. KIND, either express or implied. See the License for the +.. specific language governing permissions and limitations +.. under the License. + +================= +DataFrame API +================= + +Overview +-------- + +The ``DataFrame`` class is the core abstraction in DataFusion that represents tabular data and operations +on that data. DataFrames provide a flexible API for transforming data through various operations such as +filtering, projection, aggregation, joining, and more. + +A DataFrame represents a logical plan that is lazily evaluated. The actual execution occurs only when +terminal operations like ``collect()``, ``show()``, or ``to_pandas()`` are called. + +Creating DataFrames +------------------- + +DataFrames can be created in several ways: + +* From SQL queries via a ``SessionContext``: + + .. code-block:: python + + from datafusion import SessionContext + + ctx = SessionContext() + df = ctx.sql("SELECT * FROM your_table") + +* From registered tables: + + .. code-block:: python + + df = ctx.table("your_table") + +* From various data sources: + + .. code-block:: python + + # From CSV files (see :ref:`io_csv` for detailed options) + df = ctx.read_csv("path/to/data.csv") + + # From Parquet files (see :ref:`io_parquet` for detailed options) + df = ctx.read_parquet("path/to/data.parquet") + + # From JSON files (see :ref:`io_json` for detailed options) + df = ctx.read_json("path/to/data.json") + + # From Avro files (see :ref:`io_avro` for detailed options) + df = ctx.read_avro("path/to/data.avro") + + # From Pandas DataFrame + import pandas as pd + pandas_df = pd.DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]}) + df = ctx.from_pandas(pandas_df) + + # From Arrow data + import pyarrow as pa + batch = pa.RecordBatch.from_arrays( + [pa.array([1, 2, 3]), pa.array([4, 5, 6])], + names=["a", "b"] + ) + df = ctx.from_arrow(batch) + + For detailed information about reading from different data sources, see the :doc:`I/O Guide <../user-guide/io/index>`. + For custom data sources, see :ref:`io_custom_table_provider`. + +Common DataFrame Operations +--------------------------- + +DataFusion's DataFrame API offers a wide range of operations: + +.. code-block:: python + + from datafusion import column, literal + + # Select specific columns + df = df.select("col1", "col2") + + # Select with expressions + df = df.select(column("a") + column("b"), column("a") - column("b")) + + # Filter rows + df = df.filter(column("age") > literal(25)) + + # Add computed columns + df = df.with_column("full_name", column("first_name") + literal(" ") + column("last_name")) + + # Multiple column additions + df = df.with_columns( + (column("a") + column("b")).alias("sum"), + (column("a") * column("b")).alias("product") + ) + + # Sort data + df = df.sort(column("age").sort(ascending=False)) + + # Join DataFrames + df = df1.join(df2, on="user_id", how="inner") + + # Aggregate data + from datafusion import functions as f + df = df.aggregate( + [], # Group by columns (empty for global aggregation) + [f.sum(column("amount")).alias("total_amount")] + ) + + # Limit rows + df = df.limit(100) + + # Drop columns + df = df.drop("temporary_column") + +Terminal Operations +------------------- + +To materialize the results of your DataFrame operations: + +.. code-block:: python + + # Collect all data as PyArrow RecordBatches + result_batches = df.collect() + + # Convert to various formats + pandas_df = df.to_pandas() # Pandas DataFrame + polars_df = df.to_polars() # Polars DataFrame + arrow_table = df.to_arrow_table() # PyArrow Table + py_dict = df.to_pydict() # Python dictionary + py_list = df.to_pylist() # Python list of dictionaries + + # Display results + df.show() # Print tabular format to console + + # Count rows + count = df.count() + +HTML Rendering in Jupyter +------------------------- + +When working in Jupyter notebooks or other environments that support rich HTML display, +DataFusion DataFrames automatically render as nicely formatted HTML tables. This functionality +is provided by the ``_repr_html_`` method, which is automatically called by Jupyter. + +Basic HTML Rendering +~~~~~~~~~~~~~~~~~~~~ + +In a Jupyter environment, simply displaying a DataFrame object will trigger HTML rendering: + +.. code-block:: python + + # Will display as HTML table in Jupyter + df + + # Explicit display also uses HTML rendering + display(df) + +HTML Rendering Customization +---------------------------- + +DataFusion provides extensive customization options for HTML table rendering through the +``datafusion.html_formatter`` module. + +Configuring the HTML Formatter +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +You can customize how DataFrames are rendered by configuring the formatter: + +.. code-block:: python + + from datafusion.html_formatter import configure_formatter + + configure_formatter( + max_cell_length=30, # Maximum length of cell content before truncation + max_width=800, # Maximum width of table in pixels + max_height=400, # Maximum height of table in pixels + max_memory_bytes=2 * 1024 * 1024,# Maximum memory used for rendering (2MB) + min_rows_display=10, # Minimum rows to display + repr_rows=20, # Number of rows to display in representation + enable_cell_expansion=True, # Allow cells to be expandable on click + custom_css=None, # Custom CSS to apply + show_truncation_message=True, # Show message when data is truncated + style_provider=None, # Custom style provider class + use_shared_styles=True # Share styles across tables to reduce duplication + ) + +Custom Style Providers +~~~~~~~~~~~~~~~~~~~~~~ + +For advanced styling needs, you can create a custom style provider class: + +.. code-block:: python + + from datafusion.html_formatter import configure_formatter + + class CustomStyleProvider: + def get_cell_style(self) -> str: + return "background-color: #f5f5f5; color: #333; padding: 8px; border: 1px solid #ddd;" + + def get_header_style(self) -> str: + return "background-color: #4285f4; color: white; font-weight: bold; padding: 10px;" + + # Apply custom styling + configure_formatter(style_provider=CustomStyleProvider()) + +Custom Type Formatters +~~~~~~~~~~~~~~~~~~~~~~ + +You can register custom formatters for specific data types: + +.. code-block:: python + + from datafusion.html_formatter import get_formatter + + formatter = get_formatter() + + # Format integers with color based on value + def format_int(value): + return f' 100 else "blue"}">{value}' + + formatter.register_formatter(int, format_int) + + # Format date values + def format_date(value): + return f'{value.isoformat()}' + + formatter.register_formatter(datetime.date, format_date) + +Custom Cell Builders +~~~~~~~~~~~~~~~~~~~~ + +For complete control over cell rendering: + +.. code-block:: python + + formatter = get_formatter() + + def custom_cell_builder(value, row, col, table_id): + try: + num_value = float(value) + if num_value > 0: # Positive values get green + return f'{value}' + if num_value < 0: # Negative values get red + return f'{value}' + except (ValueError, TypeError): + pass + + # Default styling for non-numeric or zero values + return f'{value}' + + formatter.set_custom_cell_builder(custom_cell_builder) + +Custom Header Builders +~~~~~~~~~~~~~~~~~~~~~~ + +Similarly, you can customize the rendering of table headers: + +.. code-block:: python + + def custom_header_builder(field): + tooltip = f"Type: {field.type}" + return f'{field.name}' + + formatter.set_custom_header_builder(custom_header_builder) + +Managing Formatter State +-----------------------~ + +The HTML formatter maintains global state that can be managed: + +.. code-block:: python + + from datafusion.html_formatter import reset_formatter, reset_styles_loaded_state, get_formatter + + # Reset the formatter to default settings + reset_formatter() + + # Reset only the styles loaded state (useful when styles were loaded but need reloading) + reset_styles_loaded_state() + + # Get the current formatter instance to make changes + formatter = get_formatter() + +Advanced Example: Dashboard-Style Formatting +------------------------------------------~~ + +This example shows how to create a dashboard-like styling for your DataFrames: + +.. code-block:: python + + from datafusion.html_formatter import configure_formatter, get_formatter + + # Define custom CSS + custom_css = """ + .datafusion-table { + font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif; + border-collapse: collapse; + width: 100%; + box-shadow: 0 2px 3px rgba(0,0,0,0.1); + } + .datafusion-table th { + position: sticky; + top: 0; + z-index: 10; + } + .datafusion-table tr:hover td { + background-color: #f1f7fa !important; + } + .datafusion-table .numeric-positive { + color: #0a7c00; + } + .datafusion-table .numeric-negative { + color: #d13438; + } + """ + + class DashboardStyleProvider: + def get_cell_style(self) -> str: + return "padding: 8px 12px; border-bottom: 1px solid #e0e0e0;" + + def get_header_style(self) -> str: + return ("background-color: #0078d4; color: white; font-weight: 600; " + "padding: 12px; text-align: left; border-bottom: 2px solid #005a9e;") + + # Apply configuration + configure_formatter( + max_height=500, + enable_cell_expansion=True, + custom_css=custom_css, + style_provider=DashboardStyleProvider(), + max_cell_length=50 + ) + + # Add custom formatters for numbers + formatter = get_formatter() + + def format_number(value): + try: + num = float(value) + cls = "numeric-positive" if num > 0 else "numeric-negative" if num < 0 else "" + return f'{value:,}' if cls else f'{value:,}' + except (ValueError, TypeError): + return str(value) + + formatter.register_formatter(int, format_number) + formatter.register_formatter(float, format_number) + +Best Practices +-------------- + +1. **Memory Management**: For large datasets, use ``max_memory_bytes`` to limit memory usage. + +2. **Responsive Design**: Set reasonable ``max_width`` and ``max_height`` values to ensure tables display well on different screens. + +3. **Style Optimization**: Use ``use_shared_styles=True`` to avoid duplicate style definitions when displaying multiple tables. + +4. **Reset When Needed**: Call ``reset_formatter()`` when you want to start fresh with default settings. + +5. **Cell Expansion**: Use ``enable_cell_expansion=True`` when cells might contain longer content that users may want to see in full. + +Additional Resources +-------------------- + +* :doc:`../user-guide/dataframe` - Complete guide to using DataFrames +* :doc:`../user-guide/io/index` - I/O Guide for reading data from various sources +* :doc:`../user-guide/data-sources` - Comprehensive data sources guide +* :ref:`io_csv` - CSV file reading +* :ref:`io_parquet` - Parquet file reading +* :ref:`io_json` - JSON file reading +* :ref:`io_avro` - Avro file reading +* :ref:`io_custom_table_provider` - Custom table providers +* `API Reference `_ - Full API reference diff --git a/docs/source/api/index.rst b/docs/source/api/index.rst new file mode 100644 index 000000000..7f58227ca --- /dev/null +++ b/docs/source/api/index.rst @@ -0,0 +1,27 @@ +.. Licensed to the Apache Software Foundation (ASF) under one +.. or more contributor license agreements. See the NOTICE file +.. distributed with this work for additional information +.. regarding copyright ownership. The ASF licenses this file +.. to you under the Apache License, Version 2.0 (the +.. "License"); you may not use this file except in compliance +.. with the License. You may obtain a copy of the License at + +.. http://www.apache.org/licenses/LICENSE-2.0 + +.. Unless required by applicable law or agreed to in writing, +.. software distributed under the License is distributed on an +.. "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +.. KIND, either express or implied. See the License for the +.. specific language governing permissions and limitations +.. under the License. + +============= +API Reference +============= + +This section provides detailed API documentation for the DataFusion Python library. + +.. toctree:: + :maxdepth: 2 + + dataframe diff --git a/docs/source/conf.py b/docs/source/conf.py index 0ca124fd1..28db17d35 100644 --- a/docs/source/conf.py +++ b/docs/source/conf.py @@ -80,6 +80,9 @@ def autoapi_skip_member_fn(app, what, name, obj, skip, options) -> bool: # noqa ("class", "datafusion.DataFrame"), ("class", "datafusion.SessionContext"), ("module", "datafusion.common"), + # Duplicate modules (skip module-level docs to avoid duplication) + ("module", "datafusion.col"), + ("module", "datafusion.udf"), # Deprecated ("class", "datafusion.substrait.serde"), ("class", "datafusion.substrait.plan"), diff --git a/docs/source/index.rst b/docs/source/index.rst index c18793822..ff1e47280 100644 --- a/docs/source/index.rst +++ b/docs/source/index.rst @@ -93,3 +93,5 @@ Example :hidden: :maxdepth: 1 :caption: API + + api/index diff --git a/docs/source/user-guide/common-operations/functions.rst b/docs/source/user-guide/common-operations/functions.rst index d458d3eb0..ccb47a4e7 100644 --- a/docs/source/user-guide/common-operations/functions.rst +++ b/docs/source/user-guide/common-operations/functions.rst @@ -132,7 +132,7 @@ The function :py:func:`~datafusion.functions.in_list` allows to check a column f Handling Missing Values -===================== +======================= DataFusion provides methods to handle missing values in DataFrames: diff --git a/docs/source/user-guide/dataframe.rst b/docs/source/user-guide/dataframe.rst index 11e3d7e72..23c65b5f6 100644 --- a/docs/source/user-guide/dataframe.rst +++ b/docs/source/user-guide/dataframe.rst @@ -122,7 +122,8 @@ Performance Optimization with Shared Styles The ``use_shared_styles`` parameter (enabled by default) optimizes performance when displaying multiple DataFrames in notebook environments: - .. code-block:: python +.. code-block:: python + from datafusion.html_formatter import StyleProvider, configure_formatter # Default: Use shared styles (recommended for notebooks) configure_formatter(use_shared_styles=True) From dc0d35a2cdb47f96a04c98d5cf1e7e79b0ba1467 Mon Sep 17 00:00:00 2001 From: kosiew Date: Mon, 16 Jun 2025 19:53:40 +0800 Subject: [PATCH 004/220] Add Interruptible Query Execution in Jupyter via KeyboardInterrupt Support (#1141) * fix: enhance error handling in async wait_for_future function * feat: implement async execution for execution plans in PySessionContext * fix: improve error message for execution failures in PySessionContext * fix: enhance error handling and improve execution plan retrieval in PyDataFrame * fix: ensure 'static lifetime for futures in wait_for_future function * fix: handle potential errors when caching DataFrame and retrieving execution plan * fix: flatten batches in PyDataFrame to ensure proper schema conversion * fix: correct error handling in batch processing for schema conversion * fix: flatten nested structure in PyDataFrame to ensure proper RecordBatch iteration * fix: improve error handling in PyDataFrame stream execution * fix: add utility to get Tokio Runtime with time enabled and update wait_for_future to use it * fix: store result of converting RecordBatches to PyArrow for debugging * fix: handle error from wait_for_future in PyDataFrame collect method * fix: propagate error from wait_for_future in collect_partitioned method * fix: enable IO in Tokio runtime with time support * main register_listing_table * Revert "main register_listing_table" This reverts commit 52a5efe2001455a3ad881968d468e5c7538e1ced. * fix: propagate error correctly from wait_for_future in PySessionContext methods * fix: simplify error handling in PySessionContext by unwrapping wait_for_future result * test: add interruption handling test for long-running queries in DataFusion * test: move test_collect_interrupted to test_dataframe.py * fix: add const for interval in wait_for_future utility * fix: use get_tokio_runtime instead of the custom get_runtime * Revert "fix: use get_tokio_runtime instead of the custom get_runtime" This reverts commit ca2d89289d0a702bbb38f34e88fb78ad61d20647. * fix: use get_tokio_runtime instead of the custom get_runtime * . * Revert "." This reverts commit b8ce3e446b74aac7a76f1cc8ce6501b453d4f13c. * fix: improve query interruption handling in test_collect_interrupted * fix: ensure proper handling of query interruption in test_collect_interrupted * fix: improve error handling in database table retrieval * refactor: add helper for async move * Revert "refactor: add helper for async move" This reverts commit faabf6dd90ac505934e7cb6dc3b69fddbe89e661. * move py_err_to_datafusion_err to errors.rs * add create_csv_read_options * fix * create_csv_read_options -> PyDataFusionResult * revert to before create_csv_read_options * refactor: simplify file compression type parsing in PySessionContext * fix: parse_compression_type once only * add create_ndjson_read_options * refactor comment for clarity in wait_for_future function * refactor wait_for_future to avoid spawn * remove unused py_err_to_datafusion_err function * add comment to clarify error handling in next method of PyRecordBatchStream * handle error from wait_for_future in PySubstraitSerializer * clarify comment on future pinning in wait_for_future function * refactor wait_for_future to use Duration for signal check interval * handle error from wait_for_future in count method of PyDataFrame * fix ruff errors * fix clippy errors * remove unused get_and_enter_tokio_runtime function and simplify wait_for_future * Refactor async handling in PySessionContext and PyDataFrame - Simplified async handling by removing unnecessary cloning of strings and context in various methods. - Streamlined the use of `wait_for_future` to directly handle futures without intermediate variables. - Improved error handling by directly propagating results from futures. - Enhanced readability by reducing boilerplate code in methods related to reading and writing data. - Updated the `wait_for_future` function to improve signal checking and future handling. * Organize imports in utils.rs for improved readability * map_err instead of panic * Fix error handling in async stream execution for PySessionContext and PyDataFrame --- python/tests/test_dataframe.py | 121 +++++++++++++++++++++++++++++++++ src/catalog.rs | 2 +- src/context.rs | 52 ++++++++------ src/dataframe.rs | 43 ++++++------ src/record_batch.rs | 2 +- src/substrait.rs | 11 +-- src/utils.rs | 46 +++++++++---- 7 files changed, 211 insertions(+), 66 deletions(-) diff --git a/python/tests/test_dataframe.py b/python/tests/test_dataframe.py index dd5f962b2..64220ce9c 100644 --- a/python/tests/test_dataframe.py +++ b/python/tests/test_dataframe.py @@ -14,9 +14,12 @@ # KIND, either express or implied. See the License for the # specific language governing permissions and limitations # under the License. +import ctypes import datetime import os import re +import threading +import time from typing import Any import pyarrow as pa @@ -2060,3 +2063,121 @@ def test_fill_null_all_null_column(ctx): # Check that all nulls were filled result = filled_df.collect()[0] assert result.column(1).to_pylist() == ["filled", "filled", "filled"] + + +def test_collect_interrupted(): + """Test that a long-running query can be interrupted with Ctrl-C. + + This test simulates a Ctrl-C keyboard interrupt by raising a KeyboardInterrupt + exception in the main thread during a long-running query execution. + """ + # Create a context and a DataFrame with a query that will run for a while + ctx = SessionContext() + + # Create a recursive computation that will run for some time + batches = [] + for i in range(10): + batch = pa.RecordBatch.from_arrays( + [ + pa.array(list(range(i * 1000, (i + 1) * 1000))), + pa.array([f"value_{j}" for j in range(i * 1000, (i + 1) * 1000)]), + ], + names=["a", "b"], + ) + batches.append(batch) + + # Register tables + ctx.register_record_batches("t1", [batches]) + ctx.register_record_batches("t2", [batches]) + + # Create a large join operation that will take time to process + df = ctx.sql(""" + WITH t1_expanded AS ( + SELECT + a, + b, + CAST(a AS DOUBLE) / 1.5 AS c, + CAST(a AS DOUBLE) * CAST(a AS DOUBLE) AS d + FROM t1 + CROSS JOIN (SELECT 1 AS dummy FROM t1 LIMIT 5) + ), + t2_expanded AS ( + SELECT + a, + b, + CAST(a AS DOUBLE) * 2.5 AS e, + CAST(a AS DOUBLE) * CAST(a AS DOUBLE) * CAST(a AS DOUBLE) AS f + FROM t2 + CROSS JOIN (SELECT 1 AS dummy FROM t2 LIMIT 5) + ) + SELECT + t1.a, t1.b, t1.c, t1.d, + t2.a AS a2, t2.b AS b2, t2.e, t2.f + FROM t1_expanded t1 + JOIN t2_expanded t2 ON t1.a % 100 = t2.a % 100 + WHERE t1.a > 100 AND t2.a > 100 + """) + + # Flag to track if the query was interrupted + interrupted = False + interrupt_error = None + main_thread = threading.main_thread() + + # Shared flag to indicate query execution has started + query_started = threading.Event() + max_wait_time = 5.0 # Maximum wait time in seconds + + # This function will be run in a separate thread and will raise + # KeyboardInterrupt in the main thread + def trigger_interrupt(): + """Poll for query start, then raise KeyboardInterrupt in the main thread""" + # Poll for query to start with small sleep intervals + start_time = time.time() + while not query_started.is_set(): + time.sleep(0.1) # Small sleep between checks + if time.time() - start_time > max_wait_time: + msg = f"Query did not start within {max_wait_time} seconds" + raise RuntimeError(msg) + + # Check if thread ID is available + thread_id = main_thread.ident + if thread_id is None: + msg = "Cannot get main thread ID" + raise RuntimeError(msg) + + # Use ctypes to raise exception in main thread + exception = ctypes.py_object(KeyboardInterrupt) + res = ctypes.pythonapi.PyThreadState_SetAsyncExc( + ctypes.c_long(thread_id), exception + ) + if res != 1: + # If res is 0, the thread ID was invalid + # If res > 1, we modified multiple threads + ctypes.pythonapi.PyThreadState_SetAsyncExc( + ctypes.c_long(thread_id), ctypes.py_object(0) + ) + msg = "Failed to raise KeyboardInterrupt in main thread" + raise RuntimeError(msg) + + # Start a thread to trigger the interrupt + interrupt_thread = threading.Thread(target=trigger_interrupt) + # we mark as daemon so the test process can exit even if this thread doesn't finish + interrupt_thread.daemon = True + interrupt_thread.start() + + # Execute the query and expect it to be interrupted + try: + # Signal that we're about to start the query + query_started.set() + df.collect() + except KeyboardInterrupt: + interrupted = True + except Exception as e: + interrupt_error = e + + # Assert that the query was interrupted properly + if not interrupted: + pytest.fail(f"Query was not interrupted; got error: {interrupt_error}") + + # Make sure the interrupt thread has finished + interrupt_thread.join(timeout=1.0) diff --git a/src/catalog.rs b/src/catalog.rs index 1e189a5aa..83f8d08cb 100644 --- a/src/catalog.rs +++ b/src/catalog.rs @@ -97,7 +97,7 @@ impl PyDatabase { } fn table(&self, name: &str, py: Python) -> PyDataFusionResult { - if let Some(table) = wait_for_future(py, self.database.table(name))? { + if let Some(table) = wait_for_future(py, self.database.table(name))?? { Ok(PyTable::new(table)) } else { Err(PyDataFusionError::Common(format!( diff --git a/src/context.rs b/src/context.rs index cc3d8e8e9..b0af566e4 100644 --- a/src/context.rs +++ b/src/context.rs @@ -34,7 +34,7 @@ use pyo3::prelude::*; use crate::catalog::{PyCatalog, PyTable}; use crate::dataframe::PyDataFrame; use crate::dataset::Dataset; -use crate::errors::{py_datafusion_err, PyDataFusionResult}; +use crate::errors::{py_datafusion_err, to_datafusion_err, PyDataFusionResult}; use crate::expr::sort_expr::PySortExpr; use crate::physical_plan::PyExecutionPlan; use crate::record_batch::PyRecordBatchStream; @@ -375,7 +375,7 @@ impl PySessionContext { None => { let state = self.ctx.state(); let schema = options.infer_schema(&state, &table_path); - wait_for_future(py, schema)? + wait_for_future(py, schema)?? } }; let config = ListingTableConfig::new(table_path) @@ -400,7 +400,7 @@ impl PySessionContext { /// Returns a PyDataFrame whose plan corresponds to the SQL statement. pub fn sql(&mut self, query: &str, py: Python) -> PyDataFusionResult { let result = self.ctx.sql(query); - let df = wait_for_future(py, result)?; + let df = wait_for_future(py, result)??; Ok(PyDataFrame::new(df)) } @@ -417,7 +417,7 @@ impl PySessionContext { SQLOptions::new() }; let result = self.ctx.sql_with_options(query, options); - let df = wait_for_future(py, result)?; + let df = wait_for_future(py, result)??; Ok(PyDataFrame::new(df)) } @@ -451,7 +451,7 @@ impl PySessionContext { self.ctx.register_table(&*table_name, Arc::new(table))?; - let table = wait_for_future(py, self._table(&table_name))?; + let table = wait_for_future(py, self._table(&table_name))??; let df = PyDataFrame::new(table); Ok(df) @@ -650,7 +650,7 @@ impl PySessionContext { .collect(); let result = self.ctx.register_parquet(name, path, options); - wait_for_future(py, result)?; + wait_for_future(py, result)??; Ok(()) } @@ -693,11 +693,11 @@ impl PySessionContext { if path.is_instance_of::() { let paths = path.extract::>()?; let result = self.register_csv_from_multiple_paths(name, paths, options); - wait_for_future(py, result)?; + wait_for_future(py, result)??; } else { let path = path.extract::()?; let result = self.ctx.register_csv(name, &path, options); - wait_for_future(py, result)?; + wait_for_future(py, result)??; } Ok(()) @@ -734,7 +734,7 @@ impl PySessionContext { options.schema = schema.as_ref().map(|x| &x.0); let result = self.ctx.register_json(name, path, options); - wait_for_future(py, result)?; + wait_for_future(py, result)??; Ok(()) } @@ -764,7 +764,7 @@ impl PySessionContext { options.schema = schema.as_ref().map(|x| &x.0); let result = self.ctx.register_avro(name, path, options); - wait_for_future(py, result)?; + wait_for_future(py, result)??; Ok(()) } @@ -825,9 +825,19 @@ impl PySessionContext { } pub fn table(&self, name: &str, py: Python) -> PyResult { - let x = wait_for_future(py, self.ctx.table(name)) + let res = wait_for_future(py, self.ctx.table(name)) .map_err(|e| PyKeyError::new_err(e.to_string()))?; - Ok(PyDataFrame::new(x)) + match res { + Ok(df) => Ok(PyDataFrame::new(df)), + Err(e) => { + if let datafusion::error::DataFusionError::Plan(msg) = &e { + if msg.contains("No table named") { + return Err(PyKeyError::new_err(msg.to_string())); + } + } + Err(py_datafusion_err(e)) + } + } } pub fn table_exist(&self, name: &str) -> PyDataFusionResult { @@ -865,10 +875,10 @@ impl PySessionContext { let df = if let Some(schema) = schema { options.schema = Some(&schema.0); let result = self.ctx.read_json(path, options); - wait_for_future(py, result)? + wait_for_future(py, result)?? } else { let result = self.ctx.read_json(path, options); - wait_for_future(py, result)? + wait_for_future(py, result)?? }; Ok(PyDataFrame::new(df)) } @@ -915,12 +925,12 @@ impl PySessionContext { let paths = path.extract::>()?; let paths = paths.iter().map(|p| p as &str).collect::>(); let result = self.ctx.read_csv(paths, options); - let df = PyDataFrame::new(wait_for_future(py, result)?); + let df = PyDataFrame::new(wait_for_future(py, result)??); Ok(df) } else { let path = path.extract::()?; let result = self.ctx.read_csv(path, options); - let df = PyDataFrame::new(wait_for_future(py, result)?); + let df = PyDataFrame::new(wait_for_future(py, result)??); Ok(df) } } @@ -958,7 +968,7 @@ impl PySessionContext { .collect(); let result = self.ctx.read_parquet(path, options); - let df = PyDataFrame::new(wait_for_future(py, result)?); + let df = PyDataFrame::new(wait_for_future(py, result)??); Ok(df) } @@ -978,10 +988,10 @@ impl PySessionContext { let df = if let Some(schema) = schema { options.schema = Some(&schema.0); let read_future = self.ctx.read_avro(path, options); - wait_for_future(py, read_future)? + wait_for_future(py, read_future)?? } else { let read_future = self.ctx.read_avro(path, options); - wait_for_future(py, read_future)? + wait_for_future(py, read_future)?? }; Ok(PyDataFrame::new(df)) } @@ -1021,8 +1031,8 @@ impl PySessionContext { let plan = plan.plan.clone(); let fut: JoinHandle> = rt.spawn(async move { plan.execute(part, Arc::new(ctx)) }); - let stream = wait_for_future(py, fut).map_err(py_datafusion_err)?; - Ok(PyRecordBatchStream::new(stream?)) + let stream = wait_for_future(py, async { fut.await.map_err(to_datafusion_err) })???; + Ok(PyRecordBatchStream::new(stream)) } } diff --git a/src/dataframe.rs b/src/dataframe.rs index ece8c4e0f..7711a0782 100644 --- a/src/dataframe.rs +++ b/src/dataframe.rs @@ -43,7 +43,7 @@ use pyo3::types::{PyCapsule, PyList, PyTuple, PyTupleMethods}; use tokio::task::JoinHandle; use crate::catalog::PyTable; -use crate::errors::{py_datafusion_err, PyDataFusionError}; +use crate::errors::{py_datafusion_err, to_datafusion_err, PyDataFusionError}; use crate::expr::sort_expr::to_sort_expressions; use crate::physical_plan::PyExecutionPlan; use crate::record_batch::PyRecordBatchStream; @@ -233,7 +233,7 @@ impl PyDataFrame { let (batches, has_more) = wait_for_future( py, collect_record_batches_to_display(self.df.as_ref().clone(), config), - )?; + )??; if batches.is_empty() { // This should not be reached, but do it for safety since we index into the vector below return Ok("No data to display".to_string()); @@ -256,7 +256,7 @@ impl PyDataFrame { let (batches, has_more) = wait_for_future( py, collect_record_batches_to_display(self.df.as_ref().clone(), config), - )?; + )??; if batches.is_empty() { // This should not be reached, but do it for safety since we index into the vector below return Ok("No data to display".to_string()); @@ -288,7 +288,7 @@ impl PyDataFrame { /// Calculate summary statistics for a DataFrame fn describe(&self, py: Python) -> PyDataFusionResult { let df = self.df.as_ref().clone(); - let stat_df = wait_for_future(py, df.describe())?; + let stat_df = wait_for_future(py, df.describe())??; Ok(Self::new(stat_df)) } @@ -391,7 +391,7 @@ impl PyDataFrame { /// Unless some order is specified in the plan, there is no /// guarantee of the order of the result. fn collect(&self, py: Python) -> PyResult> { - let batches = wait_for_future(py, self.df.as_ref().clone().collect()) + let batches = wait_for_future(py, self.df.as_ref().clone().collect())? .map_err(PyDataFusionError::from)?; // cannot use PyResult> return type due to // https://github.com/PyO3/pyo3/issues/1813 @@ -400,14 +400,14 @@ impl PyDataFrame { /// Cache DataFrame. fn cache(&self, py: Python) -> PyDataFusionResult { - let df = wait_for_future(py, self.df.as_ref().clone().cache())?; + let df = wait_for_future(py, self.df.as_ref().clone().cache())??; Ok(Self::new(df)) } /// Executes this DataFrame and collects all results into a vector of vector of RecordBatch /// maintaining the input partitioning. fn collect_partitioned(&self, py: Python) -> PyResult>> { - let batches = wait_for_future(py, self.df.as_ref().clone().collect_partitioned()) + let batches = wait_for_future(py, self.df.as_ref().clone().collect_partitioned())? .map_err(PyDataFusionError::from)?; batches @@ -511,7 +511,7 @@ impl PyDataFrame { /// Get the execution plan for this `DataFrame` fn execution_plan(&self, py: Python) -> PyDataFusionResult { - let plan = wait_for_future(py, self.df.as_ref().clone().create_physical_plan())?; + let plan = wait_for_future(py, self.df.as_ref().clone().create_physical_plan())??; Ok(plan.into()) } @@ -624,7 +624,7 @@ impl PyDataFrame { DataFrameWriteOptions::new(), Some(csv_options), ), - )?; + )??; Ok(()) } @@ -685,7 +685,7 @@ impl PyDataFrame { DataFrameWriteOptions::new(), Option::from(options), ), - )?; + )??; Ok(()) } @@ -697,7 +697,7 @@ impl PyDataFrame { .as_ref() .clone() .write_json(path, DataFrameWriteOptions::new(), None), - )?; + )??; Ok(()) } @@ -720,7 +720,7 @@ impl PyDataFrame { py: Python<'py>, requested_schema: Option>, ) -> PyDataFusionResult> { - let mut batches = wait_for_future(py, self.df.as_ref().clone().collect())?; + let mut batches = wait_for_future(py, self.df.as_ref().clone().collect())??; let mut schema: Schema = self.df.schema().to_owned().into(); if let Some(schema_capsule) = requested_schema { @@ -753,8 +753,8 @@ impl PyDataFrame { let df = self.df.as_ref().clone(); let fut: JoinHandle> = rt.spawn(async move { df.execute_stream().await }); - let stream = wait_for_future(py, fut).map_err(py_datafusion_err)?; - Ok(PyRecordBatchStream::new(stream?)) + let stream = wait_for_future(py, async { fut.await.map_err(to_datafusion_err) })???; + Ok(PyRecordBatchStream::new(stream)) } fn execute_stream_partitioned(&self, py: Python) -> PyResult> { @@ -763,14 +763,11 @@ impl PyDataFrame { let df = self.df.as_ref().clone(); let fut: JoinHandle>> = rt.spawn(async move { df.execute_stream_partitioned().await }); - let stream = wait_for_future(py, fut).map_err(py_datafusion_err)?; + let stream = wait_for_future(py, async { fut.await.map_err(to_datafusion_err) })? + .map_err(py_datafusion_err)? + .map_err(py_datafusion_err)?; - match stream { - Ok(batches) => Ok(batches.into_iter().map(PyRecordBatchStream::new).collect()), - _ => Err(PyValueError::new_err( - "Unable to execute stream partitioned", - )), - } + Ok(stream.into_iter().map(PyRecordBatchStream::new).collect()) } /// Convert to pandas dataframe with pyarrow @@ -815,7 +812,7 @@ impl PyDataFrame { // Executes this DataFrame to get the total number of rows. fn count(&self, py: Python) -> PyDataFusionResult { - Ok(wait_for_future(py, self.df.as_ref().clone().count())?) + Ok(wait_for_future(py, self.df.as_ref().clone().count())??) } /// Fill null values with a specified value for specific columns @@ -841,7 +838,7 @@ impl PyDataFrame { /// Print DataFrame fn print_dataframe(py: Python, df: DataFrame) -> PyDataFusionResult<()> { // Get string representation of record batches - let batches = wait_for_future(py, df.collect())?; + let batches = wait_for_future(py, df.collect())??; let batches_as_string = pretty::pretty_format_batches(&batches); let result = match batches_as_string { Ok(batch) => format!("DataFrame()\n{batch}"), diff --git a/src/record_batch.rs b/src/record_batch.rs index ec61c263f..a85f05423 100644 --- a/src/record_batch.rs +++ b/src/record_batch.rs @@ -63,7 +63,7 @@ impl PyRecordBatchStream { impl PyRecordBatchStream { fn next(&mut self, py: Python) -> PyResult { let stream = self.stream.clone(); - wait_for_future(py, next_stream(stream, true)) + wait_for_future(py, next_stream(stream, true))? } fn __next__(&mut self, py: Python) -> PyResult { diff --git a/src/substrait.rs b/src/substrait.rs index 1fefc0bbd..4da3738fb 100644 --- a/src/substrait.rs +++ b/src/substrait.rs @@ -72,7 +72,7 @@ impl PySubstraitSerializer { path: &str, py: Python, ) -> PyDataFusionResult<()> { - wait_for_future(py, serializer::serialize(sql, &ctx.ctx, path))?; + wait_for_future(py, serializer::serialize(sql, &ctx.ctx, path))??; Ok(()) } @@ -94,19 +94,20 @@ impl PySubstraitSerializer { ctx: PySessionContext, py: Python, ) -> PyDataFusionResult { - let proto_bytes: Vec = wait_for_future(py, serializer::serialize_bytes(sql, &ctx.ctx))?; + let proto_bytes: Vec = + wait_for_future(py, serializer::serialize_bytes(sql, &ctx.ctx))??; Ok(PyBytes::new(py, &proto_bytes).into()) } #[staticmethod] pub fn deserialize(path: &str, py: Python) -> PyDataFusionResult { - let plan = wait_for_future(py, serializer::deserialize(path))?; + let plan = wait_for_future(py, serializer::deserialize(path))??; Ok(PyPlan { plan: *plan }) } #[staticmethod] pub fn deserialize_bytes(proto_bytes: Vec, py: Python) -> PyDataFusionResult { - let plan = wait_for_future(py, serializer::deserialize_bytes(proto_bytes))?; + let plan = wait_for_future(py, serializer::deserialize_bytes(proto_bytes))??; Ok(PyPlan { plan: *plan }) } } @@ -143,7 +144,7 @@ impl PySubstraitConsumer { ) -> PyDataFusionResult { let session_state = ctx.ctx.state(); let result = consumer::from_substrait_plan(&session_state, &plan.plan); - let logical_plan = wait_for_future(py, result)?; + let logical_plan = wait_for_future(py, result)??; Ok(PyLogicalPlan::new(logical_plan)) } } diff --git a/src/utils.rs b/src/utils.rs index 0a24ab254..90d654385 100644 --- a/src/utils.rs +++ b/src/utils.rs @@ -15,19 +15,18 @@ // specific language governing permissions and limitations // under the License. -use crate::common::data_type::PyScalarValue; -use crate::errors::{PyDataFusionError, PyDataFusionResult}; -use crate::TokioRuntime; -use datafusion::common::ScalarValue; -use datafusion::execution::context::SessionContext; -use datafusion::logical_expr::Volatility; -use pyo3::exceptions::PyValueError; +use crate::{ + common::data_type::PyScalarValue, + errors::{PyDataFusionError, PyDataFusionResult}, + TokioRuntime, +}; +use datafusion::{ + common::ScalarValue, execution::context::SessionContext, logical_expr::Volatility, +}; use pyo3::prelude::*; -use pyo3::types::PyCapsule; -use std::future::Future; -use std::sync::OnceLock; -use tokio::runtime::Runtime; - +use pyo3::{exceptions::PyValueError, types::PyCapsule}; +use std::{future::Future, sync::OnceLock, time::Duration}; +use tokio::{runtime::Runtime, time::sleep}; /// Utility to get the Tokio Runtime from Python #[inline] pub(crate) fn get_tokio_runtime() -> &'static TokioRuntime { @@ -47,14 +46,31 @@ pub(crate) fn get_global_ctx() -> &'static SessionContext { CTX.get_or_init(SessionContext::new) } -/// Utility to collect rust futures with GIL released -pub fn wait_for_future(py: Python, f: F) -> F::Output +/// Utility to collect rust futures with GIL released and respond to +/// Python interrupts such as ``KeyboardInterrupt``. If a signal is +/// received while the future is running, the future is aborted and the +/// corresponding Python exception is raised. +pub fn wait_for_future(py: Python, fut: F) -> PyResult where F: Future + Send, F::Output: Send, { let runtime: &Runtime = &get_tokio_runtime().0; - py.allow_threads(|| runtime.block_on(f)) + const INTERVAL_CHECK_SIGNALS: Duration = Duration::from_millis(1_000); + + py.allow_threads(|| { + runtime.block_on(async { + tokio::pin!(fut); + loop { + tokio::select! { + res = &mut fut => break Ok(res), + _ = sleep(INTERVAL_CHECK_SIGNALS) => { + Python::with_gil(|py| py.check_signals())?; + } + } + } + }) + }) } pub(crate) fn parse_volatility(value: &str) -> PyDataFusionResult { From 13910780216893e7602d26632411c8ae06869485 Mon Sep 17 00:00:00 2001 From: Tim Saucer Date: Mon, 16 Jun 2025 17:18:55 -0400 Subject: [PATCH 005/220] feat: upgrade df48 dependency (#1143) * Upgrade to DF 48 * Update unit test * Resolve clippy warnings * Update wrapper test to look for __repr__ special function * Add __repr__ where missing * Error in return of __repr__ * Remove patch now that DF48 is released * Expose lit_with_metadata and add unit test --- Cargo.lock | 184 ++++++++++++++------------ Cargo.toml | 10 +- python/datafusion/__init__.py | 19 +++ python/datafusion/catalog.py | 12 ++ python/datafusion/context.py | 4 + python/datafusion/expr.py | 18 +++ python/datafusion/user_defined.py | 12 ++ python/tests/test_expr.py | 60 ++++++++- python/tests/test_wrapper_coverage.py | 7 +- src/context.rs | 47 +++++-- src/expr.rs | 34 +++-- src/expr/literal.rs | 16 ++- src/expr/window.rs | 29 ++-- src/functions.rs | 6 +- src/pyarrow_filter_expression.rs | 4 +- src/udwf.rs | 8 +- 16 files changed, 325 insertions(+), 145 deletions(-) diff --git a/Cargo.lock b/Cargo.lock index 39489ed94..112167cb4 100644 --- a/Cargo.lock +++ b/Cargo.lock @@ -359,6 +359,8 @@ source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "73a47aa0c771b5381de2b7f16998d351a6f4eb839f1e13d48353e17e873d969b" dependencies = [ "bitflags", + "serde", + "serde_json", ] [[package]] @@ -859,9 +861,9 @@ dependencies = [ [[package]] name = "datafusion" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "ffe060b978f74ab446be722adb8a274e052e005bf6dfd171caadc3abaad10080" +checksum = "cc6cb8c2c81eada072059983657d6c9caf3fddefc43b4a65551d243253254a96" dependencies = [ "arrow", "arrow-ipc", @@ -887,7 +889,6 @@ dependencies = [ "datafusion-functions-nested", "datafusion-functions-table", "datafusion-functions-window", - "datafusion-macros", "datafusion-optimizer", "datafusion-physical-expr", "datafusion-physical-expr-common", @@ -902,7 +903,7 @@ dependencies = [ "object_store", "parking_lot", "parquet", - "rand 0.8.5", + "rand 0.9.1", "regex", "sqlparser", "tempfile", @@ -915,9 +916,9 @@ dependencies = [ [[package]] name = "datafusion-catalog" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "61fe34f401bd03724a1f96d12108144f8cd495a3cdda2bf5e091822fb80b7e66" +checksum = "b7be8d1b627843af62e447396db08fe1372d882c0eb8d0ea655fd1fbc33120ee" dependencies = [ "arrow", "async-trait", @@ -941,9 +942,9 @@ dependencies = [ [[package]] name = "datafusion-catalog-listing" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "a4411b8e3bce5e0fc7521e44f201def2e2d5d1b5f176fb56e8cdc9942c890f00" +checksum = "38ab16c5ae43f65ee525fc493ceffbc41f40dee38b01f643dfcfc12959e92038" dependencies = [ "arrow", "async-trait", @@ -964,9 +965,9 @@ dependencies = [ [[package]] name = "datafusion-common" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "0734015d81c8375eb5d4869b7f7ecccc2ee8d6cb81948ef737cd0e7b743bd69c" +checksum = "d3d56b2ac9f476b93ca82e4ef5fb00769c8a3f248d12b4965af7e27635fa7e12" dependencies = [ "ahash", "apache-avro", @@ -989,9 +990,9 @@ dependencies = [ [[package]] name = "datafusion-common-runtime" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "5167bb1d2ccbb87c6bc36c295274d7a0519b14afcfdaf401d53cbcaa4ef4968b" +checksum = "16015071202d6133bc84d72756176467e3e46029f3ce9ad2cb788f9b1ff139b2" dependencies = [ "futures", "log", @@ -1000,9 +1001,9 @@ dependencies = [ [[package]] name = "datafusion-datasource" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "04e602dcdf2f50c2abf297cc2203c73531e6f48b29516af7695d338cf2a778b1" +checksum = "b77523c95c89d2a7eb99df14ed31390e04ab29b43ff793e562bdc1716b07e17b" dependencies = [ "arrow", "async-compression", @@ -1025,7 +1026,7 @@ dependencies = [ "log", "object_store", "parquet", - "rand 0.8.5", + "rand 0.9.1", "tempfile", "tokio", "tokio-util", @@ -1036,9 +1037,9 @@ dependencies = [ [[package]] name = "datafusion-datasource-avro" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "d4ea5111aab9d3f2a8bff570343cccb03ce4c203875ef5a566b7d6f1eb72559e" +checksum = "1371cb4ef13c2e3a15685d37a07398cf13e3b0a85e705024b769fc4c511f5fef" dependencies = [ "apache-avro", "arrow", @@ -1061,9 +1062,9 @@ dependencies = [ [[package]] name = "datafusion-datasource-csv" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e3bb2253952dc32296ed5b84077cb2e0257fea4be6373e1c376426e17ead4ef6" +checksum = "40d25c5e2c0ebe8434beeea997b8e88d55b3ccc0d19344293f2373f65bc524fc" dependencies = [ "arrow", "async-trait", @@ -1086,9 +1087,9 @@ dependencies = [ [[package]] name = "datafusion-datasource-json" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "5b8c7f47a5d2fe03bfa521ec9bafdb8a5c82de8377f60967c3663f00c8790352" +checksum = "3dc6959e1155741ab35369e1dc7673ba30fc45ed568fad34c01b7cb1daeb4d4c" dependencies = [ "arrow", "async-trait", @@ -1111,9 +1112,9 @@ dependencies = [ [[package]] name = "datafusion-datasource-parquet" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "27d15868ea39ed2dc266728b554f6304acd473de2142281ecfa1294bb7415923" +checksum = "b7a6afdfe358d70f4237f60eaef26ae5a1ce7cb2c469d02d5fc6c7fd5d84e58b" dependencies = [ "arrow", "async-trait", @@ -1136,21 +1137,21 @@ dependencies = [ "object_store", "parking_lot", "parquet", - "rand 0.8.5", + "rand 0.9.1", "tokio", ] [[package]] name = "datafusion-doc" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "a91f8c2c5788ef32f48ff56c68e5b545527b744822a284373ac79bba1ba47292" +checksum = "9bcd8a3e3e3d02ea642541be23d44376b5d5c37c2938cce39b3873cdf7186eea" [[package]] name = "datafusion-execution" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "06f004d100f49a3658c9da6fb0c3a9b760062d96cd4ad82ccc3b7b69a9fb2f84" +checksum = "670da1d45d045eee4c2319b8c7ea57b26cf48ab77b630aaa50b779e406da476a" dependencies = [ "arrow", "dashmap", @@ -1160,16 +1161,16 @@ dependencies = [ "log", "object_store", "parking_lot", - "rand 0.8.5", + "rand 0.9.1", "tempfile", "url", ] [[package]] name = "datafusion-expr" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "7a4e4ce3802609be38eeb607ee72f6fe86c3091460de9dbfae9e18db423b3964" +checksum = "b3a577f64bdb7e2cc4043cd97f8901d8c504711fde2dbcb0887645b00d7c660b" dependencies = [ "arrow", "chrono", @@ -1188,9 +1189,9 @@ dependencies = [ [[package]] name = "datafusion-expr-common" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "422ac9cf3b22bbbae8cdf8ceb33039107fde1b5492693168f13bd566b1bcc839" +checksum = "51b7916806ace3e9f41884f230f7f38ebf0e955dfbd88266da1826f29a0b9a6a" dependencies = [ "arrow", "datafusion-common", @@ -1201,9 +1202,9 @@ dependencies = [ [[package]] name = "datafusion-ffi" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "5cf3fe9ab492c56daeb7beed526690d33622d388b8870472e0b7b7f55490338c" +checksum = "980cca31de37f5dadf7ea18e4ffc2b6833611f45bed5ef9de0831d2abb50f1ef" dependencies = [ "abi_stable", "arrow", @@ -1211,7 +1212,9 @@ dependencies = [ "async-ffi", "async-trait", "datafusion", + "datafusion-functions-aggregate-common", "datafusion-proto", + "datafusion-proto-common", "futures", "log", "prost", @@ -1221,9 +1224,9 @@ dependencies = [ [[package]] name = "datafusion-functions" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "2ddf0a0a2db5d2918349c978d42d80926c6aa2459cd8a3c533a84ec4bb63479e" +checksum = "7fb31c9dc73d3e0c365063f91139dc273308f8a8e124adda9898db8085d68357" dependencies = [ "arrow", "arrow-buffer", @@ -1241,7 +1244,7 @@ dependencies = [ "itertools 0.14.0", "log", "md-5", - "rand 0.8.5", + "rand 0.9.1", "regex", "sha2", "unicode-segmentation", @@ -1250,9 +1253,9 @@ dependencies = [ [[package]] name = "datafusion-functions-aggregate" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "408a05dafdc70d05a38a29005b8b15e21b0238734dab1e98483fcb58038c5aba" +checksum = "ebb72c6940697eaaba9bd1f746a697a07819de952b817e3fb841fb75331ad5d4" dependencies = [ "ahash", "arrow", @@ -1271,9 +1274,9 @@ dependencies = [ [[package]] name = "datafusion-functions-aggregate-common" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "756d21da2dd6c9bef97af1504970ff56cbf35d03fbd4ffd62827f02f4d2279d4" +checksum = "d7fdc54656659e5ecd49bf341061f4156ab230052611f4f3609612a0da259696" dependencies = [ "ahash", "arrow", @@ -1284,9 +1287,9 @@ dependencies = [ [[package]] name = "datafusion-functions-nested" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "8d8d50f6334b378930d992d801a10ac5b3e93b846b39e4a05085742572844537" +checksum = "fad94598e3374938ca43bca6b675febe557e7a14eb627d617db427d70d65118b" dependencies = [ "arrow", "arrow-ord", @@ -1305,9 +1308,9 @@ dependencies = [ [[package]] name = "datafusion-functions-table" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "cc9a97220736c8fff1446e936be90d57216c06f28969f9ffd3b72ac93c958c8a" +checksum = "de2fc6c2946da5cab8364fb28b5cac3115f0f3a87960b235ed031c3f7e2e639b" dependencies = [ "arrow", "async-trait", @@ -1321,10 +1324,11 @@ dependencies = [ [[package]] name = "datafusion-functions-window" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "cefc2d77646e1aadd1d6a9c40088937aedec04e68c5f0465939912e1291f8193" +checksum = "3e5746548a8544870a119f556543adcd88fe0ba6b93723fe78ad0439e0fbb8b4" dependencies = [ + "arrow", "datafusion-common", "datafusion-doc", "datafusion-expr", @@ -1338,9 +1342,9 @@ dependencies = [ [[package]] name = "datafusion-functions-window-common" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "dd4aff082c42fa6da99ce0698c85addd5252928c908eb087ca3cfa64ff16b313" +checksum = "dcbe9404382cda257c434f22e13577bee7047031dfdb6216dd5e841b9465e6fe" dependencies = [ "datafusion-common", "datafusion-physical-expr-common", @@ -1348,9 +1352,9 @@ dependencies = [ [[package]] name = "datafusion-macros" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "df6f88d7ee27daf8b108ba910f9015176b36fbc72902b1ca5c2a5f1d1717e1a1" +checksum = "8dce50e3b637dab0d25d04d2fe79dfdca2b257eabd76790bffd22c7f90d700c8" dependencies = [ "datafusion-expr", "quote", @@ -1359,9 +1363,9 @@ dependencies = [ [[package]] name = "datafusion-optimizer" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "084d9f979c4b155346d3c34b18f4256e6904ded508e9554d90fed416415c3515" +checksum = "03cfaacf06445dc3bbc1e901242d2a44f2cae99a744f49f3fefddcee46240058" dependencies = [ "arrow", "chrono", @@ -1378,9 +1382,9 @@ dependencies = [ [[package]] name = "datafusion-physical-expr" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "64c536062b0076f4e30084065d805f389f9fe38af0ca75bcbac86bc5e9fbab65" +checksum = "1908034a89d7b2630898e06863583ae4c00a0dd310c1589ca284195ee3f7f8a6" dependencies = [ "ahash", "arrow", @@ -1395,14 +1399,14 @@ dependencies = [ "itertools 0.14.0", "log", "paste", - "petgraph", + "petgraph 0.8.2", ] [[package]] name = "datafusion-physical-expr-common" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "f8a92b53b3193fac1916a1c5b8e3f4347c526f6822e56b71faa5fb372327a863" +checksum = "47b7a12dd59ea07614b67dbb01d85254fbd93df45bcffa63495e11d3bdf847df" dependencies = [ "ahash", "arrow", @@ -1414,9 +1418,9 @@ dependencies = [ [[package]] name = "datafusion-physical-optimizer" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "6fa0a5ac94c7cf3da97bedabd69d6bbca12aef84b9b37e6e9e8c25286511b5e2" +checksum = "4371cc4ad33978cc2a8be93bd54a232d3f2857b50401a14631c0705f3f910aae" dependencies = [ "arrow", "datafusion-common", @@ -1433,9 +1437,9 @@ dependencies = [ [[package]] name = "datafusion-physical-plan" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "690c615db468c2e5fe5085b232d8b1c088299a6c63d87fd960a354a71f7acb55" +checksum = "dc47bc33025757a5c11f2cd094c5b6b5ed87f46fa33c023e6fdfa25fcbfade23" dependencies = [ "ahash", "arrow", @@ -1463,9 +1467,9 @@ dependencies = [ [[package]] name = "datafusion-proto" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "a4a1afb2bdb05de7ff65be6883ebfd4ec027bd9f1f21c46aa3afd01927160a83" +checksum = "d8f5d9acd7d96e3bf2a7bb04818373cab6e51de0356e3694b94905fee7b4e8b6" dependencies = [ "arrow", "chrono", @@ -1479,9 +1483,9 @@ dependencies = [ [[package]] name = "datafusion-proto-common" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "35b7a5876ebd6b564fb9a1fd2c3a2a9686b787071a256b47e4708f0916f9e46f" +checksum = "09ecb5ec152c4353b60f7a5635489834391f7a291d2b39a4820cd469e318b78e" dependencies = [ "arrow", "datafusion-common", @@ -1513,9 +1517,9 @@ dependencies = [ [[package]] name = "datafusion-session" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "ad229a134c7406c057ece00c8743c0c34b97f4e72f78b475fe17b66c5e14fa4f" +checksum = "d7485da32283985d6b45bd7d13a65169dcbe8c869e25d01b2cfbc425254b4b49" dependencies = [ "arrow", "async-trait", @@ -1537,9 +1541,9 @@ dependencies = [ [[package]] name = "datafusion-sql" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "64f6ab28b72b664c21a27b22a2ff815fd390ed224c26e89a93b5a8154a4e8607" +checksum = "a466b15632befddfeac68c125f0260f569ff315c6831538cbb40db754134e0df" dependencies = [ "arrow", "bigdecimal", @@ -1554,9 +1558,9 @@ dependencies = [ [[package]] name = "datafusion-substrait" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "061efc0937f0ce3abb37ed0d56cfa01dd0e654b90e408656d05e846c8b7599fe" +checksum = "f2f3973b1a4f6e9ee7fd99a22d58e1c06e6723a28dc911a60df575974c8339aa" dependencies = [ "async-recursion", "async-trait", @@ -2717,6 +2721,18 @@ dependencies = [ "indexmap", ] +[[package]] +name = "petgraph" +version = "0.8.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "54acf3a685220b533e437e264e4d932cfbdc4cc7ec0cd232ed73c08d03b8a7ca" +dependencies = [ + "fixedbitset", + "hashbrown 0.15.3", + "indexmap", + "serde", +] + [[package]] name = "phf" version = "0.11.3" @@ -2837,7 +2853,7 @@ dependencies = [ "log", "multimap", "once_cell", - "petgraph", + "petgraph 0.7.1", "prettyplease", "prost", "prost-types", @@ -3661,9 +3677,9 @@ dependencies = [ [[package]] name = "substrait" -version = "0.55.1" +version = "0.56.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "048fe52a3664881ccdfdc9bdb0f4e8805f3444ee64abf299d365c54f6a2ffabb" +checksum = "13de2e20128f2a018dab1cfa30be83ae069219a65968c6f89df66ad124de2397" dependencies = [ "heck", "pbjson", @@ -4016,9 +4032,9 @@ checksum = "1dccffe3ce07af9386bfd29e80c0ab1a8205a2fc34e4bcd40364df902cfa8f3f" [[package]] name = "typify" -version = "0.3.0" +version = "0.4.2" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e03ba3643450cfd95a1aca2e1938fef63c1c1994489337998aff4ad771f21ef8" +checksum = "6c6c647a34e851cf0260ccc14687f17cdcb8302ff1a8a687a24b97ca0f82406f" dependencies = [ "typify-impl", "typify-macro", @@ -4026,9 +4042,9 @@ dependencies = [ [[package]] name = "typify-impl" -version = "0.3.0" +version = "0.4.2" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "bce48219a2f3154aaa2c56cbf027728b24a3c8fe0a47ed6399781de2b3f3eeaf" +checksum = "741b7f1e2e1338c0bee5ad5a7d3a9bbd4e24c33765c08b7691810e68d879365d" dependencies = [ "heck", "log", @@ -4046,9 +4062,9 @@ dependencies = [ [[package]] name = "typify-macro" -version = "0.3.0" +version = "0.4.2" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "68b5780d745920ed73c5b7447496a9b5c42ed2681a9b70859377aec423ecf02b" +checksum = "7560adf816a1e8dad7c63d8845ef6e31e673e39eab310d225636779230cbedeb" dependencies = [ "proc-macro2", "quote", @@ -4116,9 +4132,9 @@ checksum = "b6c140620e7ffbb22c2dee59cafe6084a59b5ffc27a8859a5f0d494b5d52b6be" [[package]] name = "uuid" -version = "1.16.0" +version = "1.17.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "458f7a779bf54acc9f347480ac654f68407d3aab21269a6e3c9f922acd9e2da9" +checksum = "3cf4199d1e5d15ddd86a694e4d0dffa9c323ce759fea589f00fef9d81cc1931d" dependencies = [ "getrandom 0.3.3", "js-sys", diff --git a/Cargo.toml b/Cargo.toml index 8107d76d3..4135e64e2 100644 --- a/Cargo.toml +++ b/Cargo.toml @@ -37,11 +37,11 @@ substrait = ["dep:datafusion-substrait"] tokio = { version = "1.45", features = ["macros", "rt", "rt-multi-thread", "sync"] } pyo3 = { version = "0.24", features = ["extension-module", "abi3", "abi3-py39"] } pyo3-async-runtimes = { version = "0.24", features = ["tokio-runtime"]} -arrow = { version = "55.0.0", features = ["pyarrow"] } -datafusion = { version = "47.0.0", features = ["avro", "unicode_expressions"] } -datafusion-substrait = { version = "47.0.0", optional = true } -datafusion-proto = { version = "47.0.0" } -datafusion-ffi = { version = "47.0.0" } +arrow = { version = "55.1.0", features = ["pyarrow"] } +datafusion = { version = "48.0.0", features = ["avro", "unicode_expressions"] } +datafusion-substrait = { version = "48.0.0", optional = true } +datafusion-proto = { version = "48.0.0" } +datafusion-ffi = { version = "48.0.0" } prost = "0.13.1" # keep in line with `datafusion-substrait` uuid = { version = "1.16", features = ["v4"] } mimalloc = { version = "0.1", optional = true, default-features = false, features = ["local_dynamic_tls"] } diff --git a/python/datafusion/__init__.py b/python/datafusion/__init__.py index c3468eb4a..4f7700251 100644 --- a/python/datafusion/__init__.py +++ b/python/datafusion/__init__.py @@ -21,6 +21,10 @@ See https://datafusion.apache.org/python for more information. """ +from __future__ import annotations + +from typing import Any + try: import importlib.metadata as importlib_metadata except ImportError: @@ -130,3 +134,18 @@ def str_lit(value): def lit(value) -> Expr: """Create a literal expression.""" return Expr.literal(value) + + +def literal_with_metadata(value: Any, metadata: dict[str, str]) -> Expr: + """Creates a new expression representing a scalar value with metadata. + + Args: + value: A valid PyArrow scalar value or easily castable to one. + metadata: Metadata to attach to the expression. + """ + return Expr.literal_with_metadata(value, metadata) + + +def lit_with_metadata(value: Any, metadata: dict[str, str]) -> Expr: + """Alias for literal_with_metadata.""" + return literal_with_metadata(value, metadata) diff --git a/python/datafusion/catalog.py b/python/datafusion/catalog.py index 6c3f188cc..67ab3ead2 100644 --- a/python/datafusion/catalog.py +++ b/python/datafusion/catalog.py @@ -34,6 +34,10 @@ def __init__(self, catalog: df_internal.Catalog) -> None: """This constructor is not typically called by the end user.""" self.catalog = catalog + def __repr__(self) -> str: + """Print a string representation of the catalog.""" + return self.catalog.__repr__() + def names(self) -> list[str]: """Returns the list of databases in this catalog.""" return self.catalog.names() @@ -50,6 +54,10 @@ def __init__(self, db: df_internal.Database) -> None: """This constructor is not typically called by the end user.""" self.db = db + def __repr__(self) -> str: + """Print a string representation of the database.""" + return self.db.__repr__() + def names(self) -> set[str]: """Returns the list of all tables in this database.""" return self.db.names() @@ -66,6 +74,10 @@ def __init__(self, table: df_internal.Table) -> None: """This constructor is not typically called by the end user.""" self.table = table + def __repr__(self) -> str: + """Print a string representation of the table.""" + return self.table.__repr__() + @property def schema(self) -> pa.Schema: """Returns the schema associated with this table.""" diff --git a/python/datafusion/context.py b/python/datafusion/context.py index 26c3d2e22..4ed465c99 100644 --- a/python/datafusion/context.py +++ b/python/datafusion/context.py @@ -496,6 +496,10 @@ def __init__( self.ctx = SessionContextInternal(config, runtime) + def __repr__(self) -> str: + """Print a string representation of the Session Context.""" + return self.ctx.__repr__() + @classmethod def global_ctx(cls) -> SessionContext: """Retrieve the global context as a `SessionContext` wrapper. diff --git a/python/datafusion/expr.py b/python/datafusion/expr.py index 9e58873d0..e785cab06 100644 --- a/python/datafusion/expr.py +++ b/python/datafusion/expr.py @@ -435,6 +435,20 @@ def literal(value: Any) -> Expr: value = pa.scalar(value) return Expr(expr_internal.RawExpr.literal(value)) + @staticmethod + def literal_with_metadata(value: Any, metadata: dict[str, str]) -> Expr: + """Creates a new expression representing a scalar value with metadata. + + Args: + value: A valid PyArrow scalar value or easily castable to one. + metadata: Metadata to attach to the expression. + """ + if isinstance(value, str): + value = pa.scalar(value, type=pa.string_view()) + value = value if isinstance(value, pa.Scalar) else pa.scalar(value) + + return Expr(expr_internal.RawExpr.literal_with_metadata(value, metadata)) + @staticmethod def string_literal(value: str) -> Expr: """Creates a new expression representing a UTF8 literal value. @@ -1172,6 +1186,10 @@ def __init__( end_bound = end_bound.cast(pa.uint64()) self.window_frame = expr_internal.WindowFrame(units, start_bound, end_bound) + def __repr__(self) -> str: + """Print a string representation of the window frame.""" + return self.window_frame.__repr__() + def get_frame_units(self) -> str: """Returns the window frame units for the bounds.""" return self.window_frame.get_frame_units() diff --git a/python/datafusion/user_defined.py b/python/datafusion/user_defined.py index 9ec3679a6..dd634c7fb 100644 --- a/python/datafusion/user_defined.py +++ b/python/datafusion/user_defined.py @@ -102,6 +102,10 @@ def __init__( name, func, input_types, return_type, str(volatility) ) + def __repr__(self) -> str: + """Print a string representation of the Scalar UDF.""" + return self._udf.__repr__() + def __call__(self, *args: Expr) -> Expr: """Execute the UDF. @@ -268,6 +272,10 @@ def __init__( str(volatility), ) + def __repr__(self) -> str: + """Print a string representation of the Aggregate UDF.""" + return self._udaf.__repr__() + def __call__(self, *args: Expr) -> Expr: """Execute the UDAF. @@ -604,6 +612,10 @@ def __init__( name, func, input_types, return_type, str(volatility) ) + def __repr__(self) -> str: + """Print a string representation of the Window UDF.""" + return self._udwf.__repr__() + def __call__(self, *args: Expr) -> Expr: """Execute the UDWF. diff --git a/python/tests/test_expr.py b/python/tests/test_expr.py index adca783b5..40a98dc4d 100644 --- a/python/tests/test_expr.py +++ b/python/tests/test_expr.py @@ -19,7 +19,14 @@ import pyarrow as pa import pytest -from datafusion import SessionContext, col, functions, lit +from datafusion import ( + SessionContext, + col, + functions, + lit, + lit_with_metadata, + literal_with_metadata, +) from datafusion.expr import ( Aggregate, AggregateFunction, @@ -103,7 +110,7 @@ def test_limit(test_ctx): plan = plan.to_variant() assert isinstance(plan, Limit) - assert "Skip: Some(Literal(Int64(5)))" in str(plan) + assert "Skip: Some(Literal(Int64(5), None))" in str(plan) def test_aggregate_query(test_ctx): @@ -824,3 +831,52 @@ def test_expr_functions(ctx, function, expected_result): assert len(result) == 1 assert result[0].column(0).equals(expected_result) + + +def test_literal_metadata(ctx): + result = ( + ctx.from_pydict({"a": [1]}) + .select( + lit(1).alias("no_metadata"), + lit_with_metadata(2, {"key1": "value1"}).alias("lit_with_metadata_fn"), + literal_with_metadata(3, {"key2": "value2"}).alias( + "literal_with_metadata_fn" + ), + ) + .collect() + ) + + expected_schema = pa.schema( + [ + pa.field("no_metadata", pa.int64(), nullable=False), + pa.field( + "lit_with_metadata_fn", + pa.int64(), + nullable=False, + metadata={"key1": "value1"}, + ), + pa.field( + "literal_with_metadata_fn", + pa.int64(), + nullable=False, + metadata={"key2": "value2"}, + ), + ] + ) + + expected = pa.RecordBatch.from_pydict( + { + "no_metadata": pa.array([1]), + "lit_with_metadata_fn": pa.array([2]), + "literal_with_metadata_fn": pa.array([3]), + }, + schema=expected_schema, + ) + + assert result[0] == expected + + # Testing result[0].schema == expected_schema does not check each key/value pair + # so we want to explicitly test these + for expected_field in expected_schema: + actual_field = result[0].schema.field(expected_field.name) + assert expected_field.metadata == actual_field.metadata diff --git a/python/tests/test_wrapper_coverage.py b/python/tests/test_wrapper_coverage.py index 926a65961..f484cb282 100644 --- a/python/tests/test_wrapper_coverage.py +++ b/python/tests/test_wrapper_coverage.py @@ -28,14 +28,14 @@ from enum import EnumMeta as EnumType -def missing_exports(internal_obj, wrapped_obj) -> None: +def missing_exports(internal_obj, wrapped_obj) -> None: # noqa: C901 """ Identify if any of the rust exposted structs or functions do not have wrappers. Special handling for: - Raw* classes: Internal implementation details that shouldn't be exposed - _global_ctx: Internal implementation detail - - __self__, __class__: Python special attributes + - __self__, __class__, __repr__: Python special attributes """ # Special case enums - EnumType overrides a some of the internal functions, # so check all of the values exist and move on @@ -45,6 +45,9 @@ def missing_exports(internal_obj, wrapped_obj) -> None: assert value in dir(wrapped_obj) return + if "__repr__" in internal_obj.__dict__ and "__repr__" not in wrapped_obj.__dict__: + pytest.fail(f"Missing __repr__: {internal_obj.__name__}") + for internal_attr_name in dir(internal_obj): wrapped_attr_name = internal_attr_name.removeprefix("Raw") assert wrapped_attr_name in dir(wrapped_obj) diff --git a/src/context.rs b/src/context.rs index b0af566e4..55c92a8fa 100644 --- a/src/context.rs +++ b/src/context.rs @@ -61,7 +61,7 @@ use datafusion::datasource::TableProvider; use datafusion::execution::context::{ DataFilePaths, SQLOptions, SessionConfig, SessionContext, TaskContext, }; -use datafusion::execution::disk_manager::DiskManagerConfig; +use datafusion::execution::disk_manager::DiskManagerMode; use datafusion::execution::memory_pool::{FairSpillPool, GreedyMemoryPool, UnboundedMemoryPool}; use datafusion::execution::options::ReadOptions; use datafusion::execution::runtime_env::RuntimeEnvBuilder; @@ -183,22 +183,49 @@ impl PyRuntimeEnvBuilder { } fn with_disk_manager_disabled(&self) -> Self { - let mut builder = self.builder.clone(); - builder = builder.with_disk_manager(DiskManagerConfig::Disabled); - Self { builder } + let mut runtime_builder = self.builder.clone(); + + let mut disk_mgr_builder = runtime_builder + .disk_manager_builder + .clone() + .unwrap_or_default(); + disk_mgr_builder.set_mode(DiskManagerMode::Disabled); + + runtime_builder = runtime_builder.with_disk_manager_builder(disk_mgr_builder); + Self { + builder: runtime_builder, + } } fn with_disk_manager_os(&self) -> Self { - let builder = self.builder.clone(); - let builder = builder.with_disk_manager(DiskManagerConfig::NewOs); - Self { builder } + let mut runtime_builder = self.builder.clone(); + + let mut disk_mgr_builder = runtime_builder + .disk_manager_builder + .clone() + .unwrap_or_default(); + disk_mgr_builder.set_mode(DiskManagerMode::OsTmpDirectory); + + runtime_builder = runtime_builder.with_disk_manager_builder(disk_mgr_builder); + Self { + builder: runtime_builder, + } } fn with_disk_manager_specified(&self, paths: Vec) -> Self { - let builder = self.builder.clone(); let paths = paths.iter().map(|s| s.into()).collect(); - let builder = builder.with_disk_manager(DiskManagerConfig::NewSpecified(paths)); - Self { builder } + let mut runtime_builder = self.builder.clone(); + + let mut disk_mgr_builder = runtime_builder + .disk_manager_builder + .clone() + .unwrap_or_default(); + disk_mgr_builder.set_mode(DiskManagerMode::Directories(paths)); + + runtime_builder = runtime_builder.with_disk_manager_builder(disk_mgr_builder); + Self { + builder: runtime_builder, + } } fn with_unbounded_memory_pool(&self) -> Self { diff --git a/src/expr.rs b/src/expr.rs index bc7dbeffd..6b1d01d65 100644 --- a/src/expr.rs +++ b/src/expr.rs @@ -15,10 +15,10 @@ // specific language governing permissions and limitations // under the License. -use datafusion::logical_expr::expr::{AggregateFunctionParams, WindowFunctionParams}; +use datafusion::logical_expr::expr::AggregateFunctionParams; use datafusion::logical_expr::utils::exprlist_to_fields; use datafusion::logical_expr::{ - ExprFuncBuilder, ExprFunctionExt, LogicalPlan, WindowFunctionDefinition, + lit_with_metadata, ExprFuncBuilder, ExprFunctionExt, LogicalPlan, WindowFunctionDefinition, }; use pyo3::IntoPyObjectExt; use pyo3::{basic::CompareOp, prelude::*}; @@ -150,7 +150,7 @@ impl PyExpr { Ok(PyScalarVariable::new(data_type, variables).into_bound_py_any(py)?) } Expr::Like(value) => Ok(PyLike::from(value.clone()).into_bound_py_any(py)?), - Expr::Literal(value) => Ok(PyLiteral::from(value.clone()).into_bound_py_any(py)?), + Expr::Literal(value, metadata) => Ok(PyLiteral::new_with_metadata(value.clone(), metadata.clone()).into_bound_py_any(py)?), Expr::BinaryExpr(expr) => Ok(PyBinaryExpr::from(expr.clone()).into_bound_py_any(py)?), Expr::Not(expr) => Ok(PyNot::new(*expr.clone()).into_bound_py_any(py)?), Expr::IsNotNull(expr) => Ok(PyIsNotNull::new(*expr.clone()).into_bound_py_any(py)?), @@ -282,6 +282,14 @@ impl PyExpr { lit(value.0).into() } + #[staticmethod] + pub fn literal_with_metadata( + value: PyScalarValue, + metadata: HashMap, + ) -> PyExpr { + lit_with_metadata(value.0, metadata).into() + } + #[staticmethod] pub fn column(value: &str) -> PyExpr { col(value).into() @@ -377,7 +385,7 @@ impl PyExpr { /// Extracts the Expr value into a PyObject that can be shared with Python pub fn python_value(&self, py: Python) -> PyResult { match &self.expr { - Expr::Literal(scalar_value) => scalar_to_pyarrow(scalar_value, py), + Expr::Literal(scalar_value, _) => scalar_to_pyarrow(scalar_value, py), _ => Err(py_type_err(format!( "Non Expr::Literal encountered in types: {:?}", &self.expr @@ -417,11 +425,13 @@ impl PyExpr { params: AggregateFunctionParams { args, .. }, .. }) - | Expr::ScalarFunction(ScalarFunction { args, .. }) - | Expr::WindowFunction(WindowFunction { - params: WindowFunctionParams { args, .. }, - .. - }) => Ok(args.iter().map(|arg| PyExpr::from(arg.clone())).collect()), + | Expr::ScalarFunction(ScalarFunction { args, .. }) => { + Ok(args.iter().map(|arg| PyExpr::from(arg.clone())).collect()) + } + Expr::WindowFunction(boxed_window_fn) => { + let args = &boxed_window_fn.params.args; + Ok(args.iter().map(|arg| PyExpr::from(arg.clone())).collect()) + } // Expr(s) that require more specific processing Expr::Case(Case { @@ -600,10 +610,10 @@ impl PyExpr { ) -> PyDataFusionResult { match &self.expr { Expr::AggregateFunction(agg_fn) => { - let window_fn = Expr::WindowFunction(WindowFunction::new( + let window_fn = Expr::WindowFunction(Box::new(WindowFunction::new( WindowFunctionDefinition::AggregateUDF(agg_fn.func.clone()), agg_fn.params.args.clone(), - )); + ))); add_builder_fns_to_window( window_fn, @@ -743,7 +753,7 @@ impl PyExpr { | Operator::QuestionPipe => Err(py_type_err(format!("Unsupported expr: ${op}"))), }, Expr::Cast(Cast { expr: _, data_type }) => DataTypeMap::map_from_arrow_type(data_type), - Expr::Literal(scalar_value) => DataTypeMap::map_from_scalar_value(scalar_value), + Expr::Literal(scalar_value, _) => DataTypeMap::map_from_scalar_value(scalar_value), _ => Err(py_type_err(format!( "Non Expr::Literal encountered in types: {:?}", expr diff --git a/src/expr/literal.rs b/src/expr/literal.rs index a660ac914..45303a104 100644 --- a/src/expr/literal.rs +++ b/src/expr/literal.rs @@ -18,11 +18,22 @@ use crate::errors::PyDataFusionError; use datafusion::common::ScalarValue; use pyo3::{prelude::*, IntoPyObjectExt}; +use std::collections::BTreeMap; #[pyclass(name = "Literal", module = "datafusion.expr", subclass)] #[derive(Clone)] pub struct PyLiteral { pub value: ScalarValue, + pub metadata: Option>, +} + +impl PyLiteral { + pub fn new_with_metadata( + value: ScalarValue, + metadata: Option>, + ) -> PyLiteral { + Self { value, metadata } + } } impl From for ScalarValue { @@ -33,7 +44,10 @@ impl From for ScalarValue { impl From for PyLiteral { fn from(value: ScalarValue) -> PyLiteral { - PyLiteral { value } + PyLiteral { + value, + metadata: None, + } } } diff --git a/src/expr/window.rs b/src/expr/window.rs index c5467bf94..052d9eeb4 100644 --- a/src/expr/window.rs +++ b/src/expr/window.rs @@ -16,7 +16,6 @@ // under the License. use datafusion::common::{DataFusionError, ScalarValue}; -use datafusion::logical_expr::expr::{WindowFunction, WindowFunctionParams}; use datafusion::logical_expr::{Expr, Window, WindowFrame, WindowFrameBound, WindowFrameUnits}; use pyo3::{prelude::*, IntoPyObjectExt}; use std::fmt::{self, Display, Formatter}; @@ -118,10 +117,9 @@ impl PyWindowExpr { /// Returns order by columns in a window function expression pub fn get_sort_exprs(&self, expr: PyExpr) -> PyResult> { match expr.expr.unalias() { - Expr::WindowFunction(WindowFunction { - params: WindowFunctionParams { order_by, .. }, - .. - }) => py_sort_expr_list(&order_by), + Expr::WindowFunction(boxed_window_fn) => { + py_sort_expr_list(&boxed_window_fn.params.order_by) + } other => Err(not_window_function_err(other)), } } @@ -129,10 +127,9 @@ impl PyWindowExpr { /// Return partition by columns in a window function expression pub fn get_partition_exprs(&self, expr: PyExpr) -> PyResult> { match expr.expr.unalias() { - Expr::WindowFunction(WindowFunction { - params: WindowFunctionParams { partition_by, .. }, - .. - }) => py_expr_list(&partition_by), + Expr::WindowFunction(boxed_window_fn) => { + py_expr_list(&boxed_window_fn.params.partition_by) + } other => Err(not_window_function_err(other)), } } @@ -140,10 +137,7 @@ impl PyWindowExpr { /// Return input args for window function pub fn get_args(&self, expr: PyExpr) -> PyResult> { match expr.expr.unalias() { - Expr::WindowFunction(WindowFunction { - params: WindowFunctionParams { args, .. }, - .. - }) => py_expr_list(&args), + Expr::WindowFunction(boxed_window_fn) => py_expr_list(&boxed_window_fn.params.args), other => Err(not_window_function_err(other)), } } @@ -151,7 +145,7 @@ impl PyWindowExpr { /// Return window function name pub fn window_func_name(&self, expr: PyExpr) -> PyResult { match expr.expr.unalias() { - Expr::WindowFunction(WindowFunction { fun, .. }) => Ok(fun.to_string()), + Expr::WindowFunction(boxed_window_fn) => Ok(boxed_window_fn.fun.to_string()), other => Err(not_window_function_err(other)), } } @@ -159,10 +153,9 @@ impl PyWindowExpr { /// Returns a Pywindow frame for a given window function expression pub fn get_frame(&self, expr: PyExpr) -> Option { match expr.expr.unalias() { - Expr::WindowFunction(WindowFunction { - params: WindowFunctionParams { window_frame, .. }, - .. - }) => Some(window_frame.into()), + Expr::WindowFunction(boxed_window_fn) => { + Some(boxed_window_fn.params.window_frame.into()) + } _ => None, } } diff --git a/src/functions.rs b/src/functions.rs index caa79b8ad..b2bafcb65 100644 --- a/src/functions.rs +++ b/src/functions.rs @@ -103,7 +103,7 @@ fn array_cat(exprs: Vec) -> PyExpr { #[pyo3(signature = (array, element, index=None))] fn array_position(array: PyExpr, element: PyExpr, index: Option) -> PyExpr { let index = ScalarValue::Int64(index); - let index = Expr::Literal(index); + let index = Expr::Literal(index, None); datafusion::functions_nested::expr_fn::array_position(array.into(), element.into(), index) .into() } @@ -334,7 +334,7 @@ fn window( .unwrap_or(WindowFrame::new(order_by.as_ref().map(|v| !v.is_empty()))); Ok(PyExpr { - expr: datafusion::logical_expr::Expr::WindowFunction(WindowFunction { + expr: datafusion::logical_expr::Expr::WindowFunction(Box::new(WindowFunction { fun, params: WindowFunctionParams { args: args.into_iter().map(|x| x.expr).collect::>(), @@ -351,7 +351,7 @@ fn window( window_frame, null_treatment: None, }, - }), + })), }) } diff --git a/src/pyarrow_filter_expression.rs b/src/pyarrow_filter_expression.rs index 4b4c86597..7fbb1dc2a 100644 --- a/src/pyarrow_filter_expression.rs +++ b/src/pyarrow_filter_expression.rs @@ -61,7 +61,7 @@ fn extract_scalar_list<'py>( .iter() .map(|expr| match expr { // TODO: should we also leverage `ScalarValue::to_pyarrow` here? - Expr::Literal(v) => match v { + Expr::Literal(v, _) => match v { // The unwraps here are for infallible conversions ScalarValue::Boolean(Some(b)) => Ok(b.into_bound_py_any(py)?), ScalarValue::Int8(Some(i)) => Ok(i.into_bound_py_any(py)?), @@ -106,7 +106,7 @@ impl TryFrom<&Expr> for PyArrowFilterExpression { let op_module = Python::import(py, "operator")?; let pc_expr: PyDataFusionResult> = match expr { Expr::Column(Column { name, .. }) => Ok(pc.getattr("field")?.call1((name,))?), - Expr::Literal(scalar) => Ok(scalar_to_pyarrow(scalar, py)?.into_bound(py)), + Expr::Literal(scalar, _) => Ok(scalar_to_pyarrow(scalar, py)?.into_bound(py)), Expr::BinaryExpr(BinaryExpr { left, op, right }) => { let operator = operator_to_py(op, &op_module)?; let left = PyArrowFilterExpression::try_from(left.as_ref())?.0; diff --git a/src/udwf.rs b/src/udwf.rs index defd9c522..a0c8cc59a 100644 --- a/src/udwf.rs +++ b/src/udwf.rs @@ -300,13 +300,9 @@ impl WindowUDFImpl for MultiColumnWindowUDF { &self.signature } - fn field(&self, field_args: WindowUDFFieldArgs) -> Result { + fn field(&self, field_args: WindowUDFFieldArgs) -> Result { // TODO: Should nullable always be `true`? - Ok(arrow::datatypes::Field::new( - field_args.name(), - self.return_type.clone(), - true, - )) + Ok(arrow::datatypes::Field::new(field_args.name(), self.return_type.clone(), true).into()) } // TODO: Enable passing partition_evaluator_args to python? From 9b6acec075f49d551a2b90608b0c7114de84d718 Mon Sep 17 00:00:00 2001 From: Michele Gregori Date: Thu, 19 Jun 2025 19:58:22 +0200 Subject: [PATCH 006/220] Support types other than String and Int for partition columns (#1154) * impl impl * fix test * format rust * support for old logic dasdas * also on io * fix formatting --------- Co-authored-by: michele gregori --- python/datafusion/context.py | 66 ++++++++++++++++++++++---- python/datafusion/io.py | 8 ++-- python/tests/test_sql.py | 26 ++++++----- src/context.rs | 89 +++++++++++++++++++++++------------- 4 files changed, 132 insertions(+), 57 deletions(-) diff --git a/python/datafusion/context.py b/python/datafusion/context.py index 4ed465c99..5b99b0d26 100644 --- a/python/datafusion/context.py +++ b/python/datafusion/context.py @@ -19,8 +19,11 @@ from __future__ import annotations +import warnings from typing import TYPE_CHECKING, Any, Protocol +import pyarrow as pa + try: from warnings import deprecated # Python 3.13+ except ImportError: @@ -42,7 +45,6 @@ import pandas as pd import polars as pl - import pyarrow as pa from datafusion.plan import ExecutionPlan, LogicalPlan @@ -539,7 +541,7 @@ def register_listing_table( self, name: str, path: str | pathlib.Path, - table_partition_cols: list[tuple[str, str]] | None = None, + table_partition_cols: list[tuple[str, str | pa.DataType]] | None = None, file_extension: str = ".parquet", schema: pa.Schema | None = None, file_sort_order: list[list[Expr | SortExpr]] | None = None, @@ -560,6 +562,7 @@ def register_listing_table( """ if table_partition_cols is None: table_partition_cols = [] + table_partition_cols = self._convert_table_partition_cols(table_partition_cols) file_sort_order_raw = ( [sort_list_to_raw_sort_list(f) for f in file_sort_order] if file_sort_order is not None @@ -778,7 +781,7 @@ def register_parquet( self, name: str, path: str | pathlib.Path, - table_partition_cols: list[tuple[str, str]] | None = None, + table_partition_cols: list[tuple[str, str | pa.DataType]] | None = None, parquet_pruning: bool = True, file_extension: str = ".parquet", skip_metadata: bool = True, @@ -806,6 +809,7 @@ def register_parquet( """ if table_partition_cols is None: table_partition_cols = [] + table_partition_cols = self._convert_table_partition_cols(table_partition_cols) self.ctx.register_parquet( name, str(path), @@ -869,7 +873,7 @@ def register_json( schema: pa.Schema | None = None, schema_infer_max_records: int = 1000, file_extension: str = ".json", - table_partition_cols: list[tuple[str, str]] | None = None, + table_partition_cols: list[tuple[str, str | pa.DataType]] | None = None, file_compression_type: str | None = None, ) -> None: """Register a JSON file as a table. @@ -890,6 +894,7 @@ def register_json( """ if table_partition_cols is None: table_partition_cols = [] + table_partition_cols = self._convert_table_partition_cols(table_partition_cols) self.ctx.register_json( name, str(path), @@ -906,7 +911,7 @@ def register_avro( path: str | pathlib.Path, schema: pa.Schema | None = None, file_extension: str = ".avro", - table_partition_cols: list[tuple[str, str]] | None = None, + table_partition_cols: list[tuple[str, str | pa.DataType]] | None = None, ) -> None: """Register an Avro file as a table. @@ -922,6 +927,7 @@ def register_avro( """ if table_partition_cols is None: table_partition_cols = [] + table_partition_cols = self._convert_table_partition_cols(table_partition_cols) self.ctx.register_avro( name, str(path), schema, file_extension, table_partition_cols ) @@ -981,7 +987,7 @@ def read_json( schema: pa.Schema | None = None, schema_infer_max_records: int = 1000, file_extension: str = ".json", - table_partition_cols: list[tuple[str, str]] | None = None, + table_partition_cols: list[tuple[str, str | pa.DataType]] | None = None, file_compression_type: str | None = None, ) -> DataFrame: """Read a line-delimited JSON data source. @@ -1001,6 +1007,7 @@ def read_json( """ if table_partition_cols is None: table_partition_cols = [] + table_partition_cols = self._convert_table_partition_cols(table_partition_cols) return DataFrame( self.ctx.read_json( str(path), @@ -1020,7 +1027,7 @@ def read_csv( delimiter: str = ",", schema_infer_max_records: int = 1000, file_extension: str = ".csv", - table_partition_cols: list[tuple[str, str]] | None = None, + table_partition_cols: list[tuple[str, str | pa.DataType]] | None = None, file_compression_type: str | None = None, ) -> DataFrame: """Read a CSV data source. @@ -1045,6 +1052,7 @@ def read_csv( """ if table_partition_cols is None: table_partition_cols = [] + table_partition_cols = self._convert_table_partition_cols(table_partition_cols) path = [str(p) for p in path] if isinstance(path, list) else str(path) @@ -1064,7 +1072,7 @@ def read_csv( def read_parquet( self, path: str | pathlib.Path, - table_partition_cols: list[tuple[str, str]] | None = None, + table_partition_cols: list[tuple[str, str | pa.DataType]] | None = None, parquet_pruning: bool = True, file_extension: str = ".parquet", skip_metadata: bool = True, @@ -1093,6 +1101,7 @@ def read_parquet( """ if table_partition_cols is None: table_partition_cols = [] + table_partition_cols = self._convert_table_partition_cols(table_partition_cols) file_sort_order = ( [sort_list_to_raw_sort_list(f) for f in file_sort_order] if file_sort_order is not None @@ -1114,7 +1123,7 @@ def read_avro( self, path: str | pathlib.Path, schema: pa.Schema | None = None, - file_partition_cols: list[tuple[str, str]] | None = None, + file_partition_cols: list[tuple[str, str | pa.DataType]] | None = None, file_extension: str = ".avro", ) -> DataFrame: """Create a :py:class:`DataFrame` for reading Avro data source. @@ -1130,6 +1139,7 @@ def read_avro( """ if file_partition_cols is None: file_partition_cols = [] + file_partition_cols = self._convert_table_partition_cols(file_partition_cols) return DataFrame( self.ctx.read_avro(str(path), schema, file_partition_cols, file_extension) ) @@ -1146,3 +1156,41 @@ def read_table(self, table: Table) -> DataFrame: def execute(self, plan: ExecutionPlan, partitions: int) -> RecordBatchStream: """Execute the ``plan`` and return the results.""" return RecordBatchStream(self.ctx.execute(plan._raw_plan, partitions)) + + @staticmethod + def _convert_table_partition_cols( + table_partition_cols: list[tuple[str, str | pa.DataType]], + ) -> list[tuple[str, pa.DataType]]: + warn = False + converted_table_partition_cols = [] + + for col, data_type in table_partition_cols: + if isinstance(data_type, str): + warn = True + if data_type == "string": + converted_data_type = pa.string() + elif data_type == "int": + converted_data_type = pa.int32() + else: + message = ( + f"Unsupported literal data type '{data_type}' for partition " + "column. Supported types are 'string' and 'int'" + ) + raise ValueError(message) + else: + converted_data_type = data_type + + converted_table_partition_cols.append((col, converted_data_type)) + + if warn: + message = ( + "using literals for table_partition_cols data types is deprecated," + "use pyarrow types instead" + ) + warnings.warn( + message, + category=DeprecationWarning, + stacklevel=2, + ) + + return converted_table_partition_cols diff --git a/python/datafusion/io.py b/python/datafusion/io.py index ef5ebf96f..551e20a6f 100644 --- a/python/datafusion/io.py +++ b/python/datafusion/io.py @@ -34,7 +34,7 @@ def read_parquet( path: str | pathlib.Path, - table_partition_cols: list[tuple[str, str]] | None = None, + table_partition_cols: list[tuple[str, str | pa.DataType]] | None = None, parquet_pruning: bool = True, file_extension: str = ".parquet", skip_metadata: bool = True, @@ -83,7 +83,7 @@ def read_json( schema: pa.Schema | None = None, schema_infer_max_records: int = 1000, file_extension: str = ".json", - table_partition_cols: list[tuple[str, str]] | None = None, + table_partition_cols: list[tuple[str, str | pa.DataType]] | None = None, file_compression_type: str | None = None, ) -> DataFrame: """Read a line-delimited JSON data source. @@ -124,7 +124,7 @@ def read_csv( delimiter: str = ",", schema_infer_max_records: int = 1000, file_extension: str = ".csv", - table_partition_cols: list[tuple[str, str]] | None = None, + table_partition_cols: list[tuple[str, str | pa.DataType]] | None = None, file_compression_type: str | None = None, ) -> DataFrame: """Read a CSV data source. @@ -171,7 +171,7 @@ def read_csv( def read_avro( path: str | pathlib.Path, schema: pa.Schema | None = None, - file_partition_cols: list[tuple[str, str]] | None = None, + file_partition_cols: list[tuple[str, str | pa.DataType]] | None = None, file_extension: str = ".avro", ) -> DataFrame: """Create a :py:class:`DataFrame` for reading Avro data source. diff --git a/python/tests/test_sql.py b/python/tests/test_sql.py index b6348e3a0..41cee4ef3 100644 --- a/python/tests/test_sql.py +++ b/python/tests/test_sql.py @@ -157,8 +157,10 @@ def test_register_parquet(ctx, tmp_path): assert result.to_pydict() == {"cnt": [100]} -@pytest.mark.parametrize("path_to_str", [True, False]) -def test_register_parquet_partitioned(ctx, tmp_path, path_to_str): +@pytest.mark.parametrize( + ("path_to_str", "legacy_data_type"), [(True, False), (False, False), (False, True)] +) +def test_register_parquet_partitioned(ctx, tmp_path, path_to_str, legacy_data_type): dir_root = tmp_path / "dataset_parquet_partitioned" dir_root.mkdir(exist_ok=False) (dir_root / "grp=a").mkdir(exist_ok=False) @@ -177,10 +179,12 @@ def test_register_parquet_partitioned(ctx, tmp_path, path_to_str): dir_root = str(dir_root) if path_to_str else dir_root + partition_data_type = "string" if legacy_data_type else pa.string() + ctx.register_parquet( "datapp", dir_root, - table_partition_cols=[("grp", "string")], + table_partition_cols=[("grp", partition_data_type)], parquet_pruning=True, file_extension=".parquet", ) @@ -488,9 +492,9 @@ def test_register_listing_table( ): dir_root = tmp_path / "dataset_parquet_partitioned" dir_root.mkdir(exist_ok=False) - (dir_root / "grp=a/date_id=20201005").mkdir(exist_ok=False, parents=True) - (dir_root / "grp=a/date_id=20211005").mkdir(exist_ok=False, parents=True) - (dir_root / "grp=b/date_id=20201005").mkdir(exist_ok=False, parents=True) + (dir_root / "grp=a/date=2020-10-05").mkdir(exist_ok=False, parents=True) + (dir_root / "grp=a/date=2021-10-05").mkdir(exist_ok=False, parents=True) + (dir_root / "grp=b/date=2020-10-05").mkdir(exist_ok=False, parents=True) table = pa.Table.from_arrays( [ @@ -501,13 +505,13 @@ def test_register_listing_table( names=["int", "str", "float"], ) pa.parquet.write_table( - table.slice(0, 3), dir_root / "grp=a/date_id=20201005/file.parquet" + table.slice(0, 3), dir_root / "grp=a/date=2020-10-05/file.parquet" ) pa.parquet.write_table( - table.slice(3, 2), dir_root / "grp=a/date_id=20211005/file.parquet" + table.slice(3, 2), dir_root / "grp=a/date=2021-10-05/file.parquet" ) pa.parquet.write_table( - table.slice(5, 10), dir_root / "grp=b/date_id=20201005/file.parquet" + table.slice(5, 10), dir_root / "grp=b/date=2020-10-05/file.parquet" ) dir_root = f"file://{dir_root}/" if path_to_str else dir_root @@ -515,7 +519,7 @@ def test_register_listing_table( ctx.register_listing_table( "my_table", dir_root, - table_partition_cols=[("grp", "string"), ("date_id", "int")], + table_partition_cols=[("grp", pa.string()), ("date", pa.date64())], file_extension=".parquet", schema=table.schema if pass_schema else None, file_sort_order=file_sort_order, @@ -531,7 +535,7 @@ def test_register_listing_table( assert dict(zip(rd["grp"], rd["count"])) == {"a": 5, "b": 2} result = ctx.sql( - "SELECT grp, COUNT(*) AS count FROM my_table WHERE date_id=20201005 GROUP BY grp" # noqa: E501 + "SELECT grp, COUNT(*) AS count FROM my_table WHERE date='2020-10-05' GROUP BY grp" # noqa: E501 ).collect() result = pa.Table.from_batches(result) diff --git a/src/context.rs b/src/context.rs index 55c92a8fa..6ce1f12bc 100644 --- a/src/context.rs +++ b/src/context.rs @@ -380,7 +380,7 @@ impl PySessionContext { &mut self, name: &str, path: &str, - table_partition_cols: Vec<(String, String)>, + table_partition_cols: Vec<(String, PyArrowType)>, file_extension: &str, schema: Option>, file_sort_order: Option>>, @@ -388,7 +388,12 @@ impl PySessionContext { ) -> PyDataFusionResult<()> { let options = ListingOptions::new(Arc::new(ParquetFormat::new())) .with_file_extension(file_extension) - .with_table_partition_cols(convert_table_partition_cols(table_partition_cols)?) + .with_table_partition_cols( + table_partition_cols + .into_iter() + .map(|(name, ty)| (name, ty.0)) + .collect::>(), + ) .with_file_sort_order( file_sort_order .unwrap_or_default() @@ -656,7 +661,7 @@ impl PySessionContext { &mut self, name: &str, path: &str, - table_partition_cols: Vec<(String, String)>, + table_partition_cols: Vec<(String, PyArrowType)>, parquet_pruning: bool, file_extension: &str, skip_metadata: bool, @@ -665,7 +670,12 @@ impl PySessionContext { py: Python, ) -> PyDataFusionResult<()> { let mut options = ParquetReadOptions::default() - .table_partition_cols(convert_table_partition_cols(table_partition_cols)?) + .table_partition_cols( + table_partition_cols + .into_iter() + .map(|(name, ty)| (name, ty.0)) + .collect::>(), + ) .parquet_pruning(parquet_pruning) .skip_metadata(skip_metadata); options.file_extension = file_extension; @@ -745,7 +755,7 @@ impl PySessionContext { schema: Option>, schema_infer_max_records: usize, file_extension: &str, - table_partition_cols: Vec<(String, String)>, + table_partition_cols: Vec<(String, PyArrowType)>, file_compression_type: Option, py: Python, ) -> PyDataFusionResult<()> { @@ -755,7 +765,12 @@ impl PySessionContext { let mut options = NdJsonReadOptions::default() .file_compression_type(parse_file_compression_type(file_compression_type)?) - .table_partition_cols(convert_table_partition_cols(table_partition_cols)?); + .table_partition_cols( + table_partition_cols + .into_iter() + .map(|(name, ty)| (name, ty.0)) + .collect::>(), + ); options.schema_infer_max_records = schema_infer_max_records; options.file_extension = file_extension; options.schema = schema.as_ref().map(|x| &x.0); @@ -778,15 +793,19 @@ impl PySessionContext { path: PathBuf, schema: Option>, file_extension: &str, - table_partition_cols: Vec<(String, String)>, + table_partition_cols: Vec<(String, PyArrowType)>, py: Python, ) -> PyDataFusionResult<()> { let path = path .to_str() .ok_or_else(|| PyValueError::new_err("Unable to convert path to a string"))?; - let mut options = AvroReadOptions::default() - .table_partition_cols(convert_table_partition_cols(table_partition_cols)?); + let mut options = AvroReadOptions::default().table_partition_cols( + table_partition_cols + .into_iter() + .map(|(name, ty)| (name, ty.0)) + .collect::>(), + ); options.file_extension = file_extension; options.schema = schema.as_ref().map(|x| &x.0); @@ -887,7 +906,7 @@ impl PySessionContext { schema: Option>, schema_infer_max_records: usize, file_extension: &str, - table_partition_cols: Vec<(String, String)>, + table_partition_cols: Vec<(String, PyArrowType)>, file_compression_type: Option, py: Python, ) -> PyDataFusionResult { @@ -895,7 +914,12 @@ impl PySessionContext { .to_str() .ok_or_else(|| PyValueError::new_err("Unable to convert path to a string"))?; let mut options = NdJsonReadOptions::default() - .table_partition_cols(convert_table_partition_cols(table_partition_cols)?) + .table_partition_cols( + table_partition_cols + .into_iter() + .map(|(name, ty)| (name, ty.0)) + .collect::>(), + ) .file_compression_type(parse_file_compression_type(file_compression_type)?); options.schema_infer_max_records = schema_infer_max_records; options.file_extension = file_extension; @@ -928,7 +952,7 @@ impl PySessionContext { delimiter: &str, schema_infer_max_records: usize, file_extension: &str, - table_partition_cols: Vec<(String, String)>, + table_partition_cols: Vec<(String, PyArrowType)>, file_compression_type: Option, py: Python, ) -> PyDataFusionResult { @@ -944,7 +968,12 @@ impl PySessionContext { .delimiter(delimiter[0]) .schema_infer_max_records(schema_infer_max_records) .file_extension(file_extension) - .table_partition_cols(convert_table_partition_cols(table_partition_cols)?) + .table_partition_cols( + table_partition_cols + .into_iter() + .map(|(name, ty)| (name, ty.0)) + .collect::>(), + ) .file_compression_type(parse_file_compression_type(file_compression_type)?); options.schema = schema.as_ref().map(|x| &x.0); @@ -974,7 +1003,7 @@ impl PySessionContext { pub fn read_parquet( &self, path: &str, - table_partition_cols: Vec<(String, String)>, + table_partition_cols: Vec<(String, PyArrowType)>, parquet_pruning: bool, file_extension: &str, skip_metadata: bool, @@ -983,7 +1012,12 @@ impl PySessionContext { py: Python, ) -> PyDataFusionResult { let mut options = ParquetReadOptions::default() - .table_partition_cols(convert_table_partition_cols(table_partition_cols)?) + .table_partition_cols( + table_partition_cols + .into_iter() + .map(|(name, ty)| (name, ty.0)) + .collect::>(), + ) .parquet_pruning(parquet_pruning) .skip_metadata(skip_metadata); options.file_extension = file_extension; @@ -1005,12 +1039,16 @@ impl PySessionContext { &self, path: &str, schema: Option>, - table_partition_cols: Vec<(String, String)>, + table_partition_cols: Vec<(String, PyArrowType)>, file_extension: &str, py: Python, ) -> PyDataFusionResult { - let mut options = AvroReadOptions::default() - .table_partition_cols(convert_table_partition_cols(table_partition_cols)?); + let mut options = AvroReadOptions::default().table_partition_cols( + table_partition_cols + .into_iter() + .map(|(name, ty)| (name, ty.0)) + .collect::>(), + ); options.file_extension = file_extension; let df = if let Some(schema) = schema { options.schema = Some(&schema.0); @@ -1109,21 +1147,6 @@ impl PySessionContext { } } -pub fn convert_table_partition_cols( - table_partition_cols: Vec<(String, String)>, -) -> PyDataFusionResult> { - table_partition_cols - .into_iter() - .map(|(name, ty)| match ty.as_str() { - "string" => Ok((name, DataType::Utf8)), - "int" => Ok((name, DataType::Int32)), - _ => Err(crate::errors::PyDataFusionError::Common(format!( - "Unsupported data type '{ty}' for partition column. Supported types are 'string' and 'int'" - ))), - }) - .collect::, _>>() -} - pub fn parse_file_compression_type( file_compression_type: Option, ) -> Result { From 98dc06b553b69576d50157542f761b44f456dcf2 Mon Sep 17 00:00:00 2001 From: Nuno Faria Date: Sat, 21 Jun 2025 15:20:21 +0100 Subject: [PATCH 007/220] feat: Support Parquet writer options (#1123) * feat: Support Parquet writer options * Create dedicated write_parquet_options function * Rename write_parquet_options to write_parquet_with_options * Merge remote-tracking branch 'origin/main' into write_parquet_options * Fix ruff errors --- python/datafusion/__init__.py | 4 +- python/datafusion/dataframe.py | 221 ++++++++++++++++++ python/tests/test_dataframe.py | 406 +++++++++++++++++++++++++++++++++ src/dataframe.rs | 126 +++++++++- src/lib.rs | 2 + 5 files changed, 757 insertions(+), 2 deletions(-) diff --git a/python/datafusion/__init__.py b/python/datafusion/__init__.py index 4f7700251..16d65f685 100644 --- a/python/datafusion/__init__.py +++ b/python/datafusion/__init__.py @@ -46,7 +46,7 @@ SessionContext, SQLOptions, ) -from .dataframe import DataFrame +from .dataframe import DataFrame, ParquetColumnOptions, ParquetWriterOptions from .expr import ( Expr, WindowFrame, @@ -80,6 +80,8 @@ "ExecutionPlan", "Expr", "LogicalPlan", + "ParquetColumnOptions", + "ParquetWriterOptions", "RecordBatch", "RecordBatchStream", "RuntimeEnvBuilder", diff --git a/python/datafusion/dataframe.py b/python/datafusion/dataframe.py index a1df7e080..769271c7e 100644 --- a/python/datafusion/dataframe.py +++ b/python/datafusion/dataframe.py @@ -38,6 +38,8 @@ from typing_extensions import deprecated # Python 3.12 from datafusion._internal import DataFrame as DataFrameInternal +from datafusion._internal import ParquetColumnOptions as ParquetColumnOptionsInternal +from datafusion._internal import ParquetWriterOptions as ParquetWriterOptionsInternal from datafusion.expr import Expr, SortExpr, sort_or_default from datafusion.plan import ExecutionPlan, LogicalPlan from datafusion.record_batch import RecordBatchStream @@ -114,6 +116,173 @@ def get_default_level(self) -> Optional[int]: return None +class ParquetWriterOptions: + """Advanced parquet writer options. + + Allows settings the writer options that apply to the entire file. Some options can + also be set on a column by column basis, with the field `column_specific_options` + (see `ParquetColumnOptions`). + + Attributes: + data_pagesize_limit: Sets best effort maximum size of data page in bytes. + write_batch_size: Sets write_batch_size in bytes. + writer_version: Sets parquet writer version. Valid values are `1.0` and + `2.0`. + skip_arrow_metadata: Skip encoding the embedded arrow metadata in the + KV_meta. + compression: Compression type to use. Default is "zstd(3)". + Available compression types are + - "uncompressed": No compression. + - "snappy": Snappy compression. + - "gzip(n)": Gzip compression with level n. + - "brotli(n)": Brotli compression with level n. + - "lz4": LZ4 compression. + - "lz4_raw": LZ4_RAW compression. + - "zstd(n)": Zstandard compression with level n. + dictionary_enabled: Sets if dictionary encoding is enabled. If None, uses + the default parquet writer setting. + dictionary_page_size_limit: Sets best effort maximum dictionary page size, + in bytes. + statistics_enabled: Sets if statistics are enabled for any column Valid + values are `none`, `chunk`, and `page`. If None, uses the default + parquet writer setting. + max_row_group_size: Target maximum number of rows in each row group + (defaults to 1M rows). Writing larger row groups requires more memory to + write, but can get better compression and be faster to read. + created_by: Sets "created by" property. + column_index_truncate_length: Sets column index truncate length. + statistics_truncate_length: Sets statistics truncate length. If None, uses + the default parquet writer setting. + data_page_row_count_limit: Sets best effort maximum number of rows in a data + page. + encoding: Sets default encoding for any column. Valid values are `plain`, + `plain_dictionary`, `rle`, `bit_packed`, `delta_binary_packed`, + `delta_length_byte_array`, `delta_byte_array`, `rle_dictionary`, and + `byte_stream_split`. If None, uses the default parquet writer setting. + bloom_filter_on_write: Write bloom filters for all columns when creating + parquet files. + bloom_filter_fpp: Sets bloom filter false positive probability. If None, + uses the default parquet writer setting + bloom_filter_ndv: Sets bloom filter number of distinct values. If None, uses + the default parquet writer setting. + allow_single_file_parallelism: Controls whether DataFusion will attempt to + speed up writing parquet files by serializing them in parallel. Each + column in each row group in each output file are serialized in parallel + leveraging a maximum possible core count of n_files * n_row_groups * + n_columns. + maximum_parallel_row_group_writers: By default parallel parquet writer is + tuned for minimum memory usage in a streaming execution plan. You may + see a performance benefit when writing large parquet files by increasing + `maximum_parallel_row_group_writers` and + `maximum_buffered_record_batches_per_stream` if your system has idle + cores and can tolerate additional memory usage. Boosting these values is + likely worthwhile when writing out already in-memory data, such as from + a cached data frame. + maximum_buffered_record_batches_per_stream: See + `maximum_parallel_row_group_writers`. + column_specific_options: Overrides options for specific columns. If a column + is not a part of this dictionary, it will use the parameters provided here. + """ + + def __init__( + self, + data_pagesize_limit: int = 1024 * 1024, + write_batch_size: int = 1024, + writer_version: str = "1.0", + skip_arrow_metadata: bool = False, + compression: Optional[str] = "zstd(3)", + dictionary_enabled: Optional[bool] = True, + dictionary_page_size_limit: int = 1024 * 1024, + statistics_enabled: Optional[str] = "page", + max_row_group_size: int = 1024 * 1024, + created_by: str = "datafusion-python", + column_index_truncate_length: Optional[int] = 64, + statistics_truncate_length: Optional[int] = None, + data_page_row_count_limit: int = 20_000, + encoding: Optional[str] = None, + bloom_filter_on_write: bool = False, + bloom_filter_fpp: Optional[float] = None, + bloom_filter_ndv: Optional[int] = None, + allow_single_file_parallelism: bool = True, + maximum_parallel_row_group_writers: int = 1, + maximum_buffered_record_batches_per_stream: int = 2, + column_specific_options: Optional[dict[str, ParquetColumnOptions]] = None, + ) -> None: + """Initialize the ParquetWriterOptions.""" + self.data_pagesize_limit = data_pagesize_limit + self.write_batch_size = write_batch_size + self.writer_version = writer_version + self.skip_arrow_metadata = skip_arrow_metadata + self.compression = compression + self.dictionary_enabled = dictionary_enabled + self.dictionary_page_size_limit = dictionary_page_size_limit + self.statistics_enabled = statistics_enabled + self.max_row_group_size = max_row_group_size + self.created_by = created_by + self.column_index_truncate_length = column_index_truncate_length + self.statistics_truncate_length = statistics_truncate_length + self.data_page_row_count_limit = data_page_row_count_limit + self.encoding = encoding + self.bloom_filter_on_write = bloom_filter_on_write + self.bloom_filter_fpp = bloom_filter_fpp + self.bloom_filter_ndv = bloom_filter_ndv + self.allow_single_file_parallelism = allow_single_file_parallelism + self.maximum_parallel_row_group_writers = maximum_parallel_row_group_writers + self.maximum_buffered_record_batches_per_stream = ( + maximum_buffered_record_batches_per_stream + ) + self.column_specific_options = column_specific_options + + +class ParquetColumnOptions: + """Parquet options for individual columns. + + Contains the available options that can be applied for an individual Parquet column, + replacing the global options in `ParquetWriterOptions`. + + Attributes: + encoding: Sets encoding for the column path. Valid values are: `plain`, + `plain_dictionary`, `rle`, `bit_packed`, `delta_binary_packed`, + `delta_length_byte_array`, `delta_byte_array`, `rle_dictionary`, and + `byte_stream_split`. These values are not case-sensitive. If `None`, uses + the default parquet options + dictionary_enabled: Sets if dictionary encoding is enabled for the column path. + If `None`, uses the default parquet options + compression: Sets default parquet compression codec for the column path. Valid + values are `uncompressed`, `snappy`, `gzip(level)`, `lzo`, `brotli(level)`, + `lz4`, `zstd(level)`, and `lz4_raw`. These values are not case-sensitive. If + `None`, uses the default parquet options. + statistics_enabled: Sets if statistics are enabled for the column Valid values + are: `none`, `chunk`, and `page` These values are not case sensitive. If + `None`, uses the default parquet options. + bloom_filter_enabled: Sets if bloom filter is enabled for the column path. If + `None`, uses the default parquet options. + bloom_filter_fpp: Sets bloom filter false positive probability for the column + path. If `None`, uses the default parquet options. + bloom_filter_ndv: Sets bloom filter number of distinct values. If `None`, uses + the default parquet options. + """ + + def __init__( + self, + encoding: Optional[str] = None, + dictionary_enabled: Optional[bool] = None, + compression: Optional[str] = None, + statistics_enabled: Optional[str] = None, + bloom_filter_enabled: Optional[bool] = None, + bloom_filter_fpp: Optional[float] = None, + bloom_filter_ndv: Optional[int] = None, + ) -> None: + """Initialize the ParquetColumnOptions.""" + self.encoding = encoding + self.dictionary_enabled = dictionary_enabled + self.compression = compression + self.statistics_enabled = statistics_enabled + self.bloom_filter_enabled = bloom_filter_enabled + self.bloom_filter_fpp = bloom_filter_fpp + self.bloom_filter_ndv = bloom_filter_ndv + + class DataFrame: """Two dimensional table representation of data. @@ -737,6 +906,58 @@ def write_parquet( self.df.write_parquet(str(path), compression.value, compression_level) + def write_parquet_with_options( + self, path: str | pathlib.Path, options: ParquetWriterOptions + ) -> None: + """Execute the :py:class:`DataFrame` and write the results to a Parquet file. + + Allows advanced writer options to be set with `ParquetWriterOptions`. + + Args: + path: Path of the Parquet file to write. + options: Sets the writer parquet options (see `ParquetWriterOptions`). + """ + options_internal = ParquetWriterOptionsInternal( + options.data_pagesize_limit, + options.write_batch_size, + options.writer_version, + options.skip_arrow_metadata, + options.compression, + options.dictionary_enabled, + options.dictionary_page_size_limit, + options.statistics_enabled, + options.max_row_group_size, + options.created_by, + options.column_index_truncate_length, + options.statistics_truncate_length, + options.data_page_row_count_limit, + options.encoding, + options.bloom_filter_on_write, + options.bloom_filter_fpp, + options.bloom_filter_ndv, + options.allow_single_file_parallelism, + options.maximum_parallel_row_group_writers, + options.maximum_buffered_record_batches_per_stream, + ) + + column_specific_options_internal = {} + for column, opts in (options.column_specific_options or {}).items(): + column_specific_options_internal[column] = ParquetColumnOptionsInternal( + bloom_filter_enabled=opts.bloom_filter_enabled, + encoding=opts.encoding, + dictionary_enabled=opts.dictionary_enabled, + compression=opts.compression, + statistics_enabled=opts.statistics_enabled, + bloom_filter_fpp=opts.bloom_filter_fpp, + bloom_filter_ndv=opts.bloom_filter_ndv, + ) + + self.df.write_parquet_with_options( + str(path), + options_internal, + column_specific_options_internal, + ) + def write_json(self, path: str | pathlib.Path) -> None: """Execute the :py:class:`DataFrame` and write the results to a JSON file. diff --git a/python/tests/test_dataframe.py b/python/tests/test_dataframe.py index 64220ce9c..3c9b97f23 100644 --- a/python/tests/test_dataframe.py +++ b/python/tests/test_dataframe.py @@ -27,6 +27,8 @@ import pytest from datafusion import ( DataFrame, + ParquetColumnOptions, + ParquetWriterOptions, SessionContext, WindowFrame, column, @@ -66,6 +68,21 @@ def df(): return ctx.from_arrow(batch) +@pytest.fixture +def large_df(): + ctx = SessionContext() + + rows = 100000 + data = { + "a": list(range(rows)), + "b": [f"s-{i}" for i in range(rows)], + "c": [float(i + 0.1) for i in range(rows)], + } + batch = pa.record_batch(data) + + return ctx.from_arrow(batch) + + @pytest.fixture def struct_df(): ctx = SessionContext() @@ -1632,6 +1649,395 @@ def test_write_compressed_parquet_default_compression_level(df, tmp_path, compre df.write_parquet(str(path), compression=compression) +def test_write_parquet_with_options_default_compression(df, tmp_path): + """Test that the default compression is ZSTD.""" + df.write_parquet(tmp_path) + + for file in tmp_path.rglob("*.parquet"): + metadata = pq.ParquetFile(file).metadata.to_dict() + for row_group in metadata["row_groups"]: + for col in row_group["columns"]: + assert col["compression"].lower() == "zstd" + + +@pytest.mark.parametrize( + "compression", + ["gzip(6)", "brotli(7)", "zstd(15)", "snappy", "uncompressed"], +) +def test_write_parquet_with_options_compression(df, tmp_path, compression): + import re + + path = tmp_path + df.write_parquet_with_options( + str(path), ParquetWriterOptions(compression=compression) + ) + + # test that the actual compression scheme is the one written + for _root, _dirs, files in os.walk(path): + for file in files: + if file.endswith(".parquet"): + metadata = pq.ParquetFile(tmp_path / file).metadata.to_dict() + for row_group in metadata["row_groups"]: + for col in row_group["columns"]: + assert col["compression"].lower() == re.sub( + r"\(\d+\)", "", compression + ) + + result = pq.read_table(str(path)).to_pydict() + expected = df.to_pydict() + + assert result == expected + + +@pytest.mark.parametrize( + "compression", + ["gzip(12)", "brotli(15)", "zstd(23)"], +) +def test_write_parquet_with_options_wrong_compression_level(df, tmp_path, compression): + path = tmp_path + + with pytest.raises(Exception, match=r"valid compression range .*? exceeded."): + df.write_parquet_with_options( + str(path), ParquetWriterOptions(compression=compression) + ) + + +@pytest.mark.parametrize("compression", ["wrong", "wrong(12)"]) +def test_write_parquet_with_options_invalid_compression(df, tmp_path, compression): + path = tmp_path + + with pytest.raises(Exception, match="Unknown or unsupported parquet compression"): + df.write_parquet_with_options( + str(path), ParquetWriterOptions(compression=compression) + ) + + +@pytest.mark.parametrize( + ("writer_version", "format_version"), + [("1.0", "1.0"), ("2.0", "2.6"), (None, "1.0")], +) +def test_write_parquet_with_options_writer_version( + df, tmp_path, writer_version, format_version +): + """Test the Parquet writer version. Note that writer_version=2.0 results in + format_version=2.6""" + if writer_version is None: + df.write_parquet_with_options(tmp_path, ParquetWriterOptions()) + else: + df.write_parquet_with_options( + tmp_path, ParquetWriterOptions(writer_version=writer_version) + ) + + for file in tmp_path.rglob("*.parquet"): + parquet = pq.ParquetFile(file) + metadata = parquet.metadata.to_dict() + assert metadata["format_version"] == format_version + + +@pytest.mark.parametrize("writer_version", ["1.2.3", "custom-version", "0"]) +def test_write_parquet_with_options_wrong_writer_version(df, tmp_path, writer_version): + """Test that invalid writer versions in Parquet throw an exception.""" + with pytest.raises( + Exception, match="Unknown or unsupported parquet writer version" + ): + df.write_parquet_with_options( + tmp_path, ParquetWriterOptions(writer_version=writer_version) + ) + + +@pytest.mark.parametrize("dictionary_enabled", [True, False, None]) +def test_write_parquet_with_options_dictionary_enabled( + df, tmp_path, dictionary_enabled +): + """Test enabling/disabling the dictionaries in Parquet.""" + df.write_parquet_with_options( + tmp_path, ParquetWriterOptions(dictionary_enabled=dictionary_enabled) + ) + # by default, the dictionary is enabled, so None results in True + result = dictionary_enabled if dictionary_enabled is not None else True + + for file in tmp_path.rglob("*.parquet"): + parquet = pq.ParquetFile(file) + metadata = parquet.metadata.to_dict() + + for row_group in metadata["row_groups"]: + for col in row_group["columns"]: + assert col["has_dictionary_page"] == result + + +@pytest.mark.parametrize( + ("statistics_enabled", "has_statistics"), + [("page", True), ("chunk", True), ("none", False), (None, True)], +) +def test_write_parquet_with_options_statistics_enabled( + df, tmp_path, statistics_enabled, has_statistics +): + """Test configuring the statistics in Parquet. In pyarrow we can only check for + column-level statistics, so "page" and "chunk" are tested in the same way.""" + df.write_parquet_with_options( + tmp_path, ParquetWriterOptions(statistics_enabled=statistics_enabled) + ) + + for file in tmp_path.rglob("*.parquet"): + parquet = pq.ParquetFile(file) + metadata = parquet.metadata.to_dict() + + for row_group in metadata["row_groups"]: + for col in row_group["columns"]: + if has_statistics: + assert col["statistics"] is not None + else: + assert col["statistics"] is None + + +@pytest.mark.parametrize("max_row_group_size", [1000, 5000, 10000, 100000]) +def test_write_parquet_with_options_max_row_group_size( + large_df, tmp_path, max_row_group_size +): + """Test configuring the max number of rows per group in Parquet. These test cases + guarantee that the number of rows for each row group is max_row_group_size, given + the total number of rows is a multiple of max_row_group_size.""" + large_df.write_parquet_with_options( + tmp_path, ParquetWriterOptions(max_row_group_size=max_row_group_size) + ) + + for file in tmp_path.rglob("*.parquet"): + parquet = pq.ParquetFile(file) + metadata = parquet.metadata.to_dict() + for row_group in metadata["row_groups"]: + assert row_group["num_rows"] == max_row_group_size + + +@pytest.mark.parametrize("created_by", ["datafusion", "datafusion-python", "custom"]) +def test_write_parquet_with_options_created_by(df, tmp_path, created_by): + """Test configuring the created by metadata in Parquet.""" + df.write_parquet_with_options(tmp_path, ParquetWriterOptions(created_by=created_by)) + + for file in tmp_path.rglob("*.parquet"): + parquet = pq.ParquetFile(file) + metadata = parquet.metadata.to_dict() + assert metadata["created_by"] == created_by + + +@pytest.mark.parametrize("statistics_truncate_length", [5, 25, 50]) +def test_write_parquet_with_options_statistics_truncate_length( + df, tmp_path, statistics_truncate_length +): + """Test configuring the truncate limit in Parquet's row-group-level statistics.""" + ctx = SessionContext() + data = { + "a": [ + "a_the_quick_brown_fox_jumps_over_the_lazy_dog", + "m_the_quick_brown_fox_jumps_over_the_lazy_dog", + "z_the_quick_brown_fox_jumps_over_the_lazy_dog", + ], + "b": ["a_smaller", "m_smaller", "z_smaller"], + } + df = ctx.from_arrow(pa.record_batch(data)) + df.write_parquet_with_options( + tmp_path, + ParquetWriterOptions(statistics_truncate_length=statistics_truncate_length), + ) + + for file in tmp_path.rglob("*.parquet"): + parquet = pq.ParquetFile(file) + metadata = parquet.metadata.to_dict() + + for row_group in metadata["row_groups"]: + for col in row_group["columns"]: + statistics = col["statistics"] + assert len(statistics["min"]) <= statistics_truncate_length + assert len(statistics["max"]) <= statistics_truncate_length + + +def test_write_parquet_with_options_default_encoding(tmp_path): + """Test that, by default, Parquet files are written with dictionary encoding. + Note that dictionary encoding is not used for boolean values, so it is not tested + here.""" + ctx = SessionContext() + data = { + "a": [1, 2, 3], + "b": ["1", "2", "3"], + "c": [1.01, 2.02, 3.03], + } + df = ctx.from_arrow(pa.record_batch(data)) + df.write_parquet_with_options(tmp_path, ParquetWriterOptions()) + + for file in tmp_path.rglob("*.parquet"): + parquet = pq.ParquetFile(file) + metadata = parquet.metadata.to_dict() + + for row_group in metadata["row_groups"]: + for col in row_group["columns"]: + assert col["encodings"] == ("PLAIN", "RLE", "RLE_DICTIONARY") + + +@pytest.mark.parametrize( + ("encoding", "data_types", "result"), + [ + ("plain", ["int", "float", "str", "bool"], ("PLAIN", "RLE")), + ("rle", ["bool"], ("RLE",)), + ("delta_binary_packed", ["int"], ("RLE", "DELTA_BINARY_PACKED")), + ("delta_length_byte_array", ["str"], ("RLE", "DELTA_LENGTH_BYTE_ARRAY")), + ("delta_byte_array", ["str"], ("RLE", "DELTA_BYTE_ARRAY")), + ("byte_stream_split", ["int", "float"], ("RLE", "BYTE_STREAM_SPLIT")), + ], +) +def test_write_parquet_with_options_encoding(tmp_path, encoding, data_types, result): + """Test different encodings in Parquet in their respective support column types.""" + ctx = SessionContext() + + data = {} + for data_type in data_types: + if data_type == "int": + data["int"] = [1, 2, 3] + elif data_type == "float": + data["float"] = [1.01, 2.02, 3.03] + elif data_type == "str": + data["str"] = ["a", "b", "c"] + elif data_type == "bool": + data["bool"] = [True, False, True] + + df = ctx.from_arrow(pa.record_batch(data)) + df.write_parquet_with_options( + tmp_path, ParquetWriterOptions(encoding=encoding, dictionary_enabled=False) + ) + + for file in tmp_path.rglob("*.parquet"): + parquet = pq.ParquetFile(file) + metadata = parquet.metadata.to_dict() + + for row_group in metadata["row_groups"]: + for col in row_group["columns"]: + assert col["encodings"] == result + + +@pytest.mark.parametrize("encoding", ["bit_packed"]) +def test_write_parquet_with_options_unsupported_encoding(df, tmp_path, encoding): + """Test that unsupported Parquet encodings do not work.""" + # BaseException is used since this throws a Rust panic: https://github.com/PyO3/pyo3/issues/3519 + with pytest.raises(BaseException, match="Encoding .*? is not supported"): + df.write_parquet_with_options(tmp_path, ParquetWriterOptions(encoding=encoding)) + + +@pytest.mark.parametrize("encoding", ["non_existent", "unknown", "plain123"]) +def test_write_parquet_with_options_invalid_encoding(df, tmp_path, encoding): + """Test that invalid Parquet encodings do not work.""" + with pytest.raises(Exception, match="Unknown or unsupported parquet encoding"): + df.write_parquet_with_options(tmp_path, ParquetWriterOptions(encoding=encoding)) + + +@pytest.mark.parametrize("encoding", ["plain_dictionary", "rle_dictionary"]) +def test_write_parquet_with_options_dictionary_encoding_fallback( + df, tmp_path, encoding +): + """Test that the dictionary encoding cannot be used as fallback in Parquet.""" + # BaseException is used since this throws a Rust panic: https://github.com/PyO3/pyo3/issues/3519 + with pytest.raises( + BaseException, match="Dictionary encoding can not be used as fallback encoding" + ): + df.write_parquet_with_options(tmp_path, ParquetWriterOptions(encoding=encoding)) + + +def test_write_parquet_with_options_bloom_filter(df, tmp_path): + """Test Parquet files with and without (default) bloom filters. Since pyarrow does + not expose any information about bloom filters, the easiest way to confirm that they + are actually written is to compare the file size.""" + path_no_bloom_filter = tmp_path / "1" + path_bloom_filter = tmp_path / "2" + + df.write_parquet_with_options(path_no_bloom_filter, ParquetWriterOptions()) + df.write_parquet_with_options( + path_bloom_filter, ParquetWriterOptions(bloom_filter_on_write=True) + ) + + size_no_bloom_filter = 0 + for file in path_no_bloom_filter.rglob("*.parquet"): + size_no_bloom_filter += os.path.getsize(file) + + size_bloom_filter = 0 + for file in path_bloom_filter.rglob("*.parquet"): + size_bloom_filter += os.path.getsize(file) + + assert size_no_bloom_filter < size_bloom_filter + + +def test_write_parquet_with_options_column_options(df, tmp_path): + """Test writing Parquet files with different options for each column, which replace + the global configs (when provided).""" + data = { + "a": [1, 2, 3], + "b": ["a", "b", "c"], + "c": [False, True, False], + "d": [1.01, 2.02, 3.03], + "e": [4, 5, 6], + } + + column_specific_options = { + "a": ParquetColumnOptions(statistics_enabled="none"), + "b": ParquetColumnOptions(encoding="plain", dictionary_enabled=False), + "c": ParquetColumnOptions( + compression="snappy", encoding="rle", dictionary_enabled=False + ), + "d": ParquetColumnOptions( + compression="zstd(6)", + encoding="byte_stream_split", + dictionary_enabled=False, + statistics_enabled="none", + ), + # column "e" will use the global configs + } + + results = { + "a": { + "statistics": False, + "compression": "brotli", + "encodings": ("PLAIN", "RLE", "RLE_DICTIONARY"), + }, + "b": { + "statistics": True, + "compression": "brotli", + "encodings": ("PLAIN", "RLE"), + }, + "c": { + "statistics": True, + "compression": "snappy", + "encodings": ("RLE",), + }, + "d": { + "statistics": False, + "compression": "zstd", + "encodings": ("RLE", "BYTE_STREAM_SPLIT"), + }, + "e": { + "statistics": True, + "compression": "brotli", + "encodings": ("PLAIN", "RLE", "RLE_DICTIONARY"), + }, + } + + ctx = SessionContext() + df = ctx.from_arrow(pa.record_batch(data)) + df.write_parquet_with_options( + tmp_path, + ParquetWriterOptions( + compression="brotli(8)", column_specific_options=column_specific_options + ), + ) + + for file in tmp_path.rglob("*.parquet"): + parquet = pq.ParquetFile(file) + metadata = parquet.metadata.to_dict() + + for row_group in metadata["row_groups"]: + for col in row_group["columns"]: + column_name = col["path_in_schema"] + result = results[column_name] + assert (col["statistics"] is not None) == result["statistics"] + assert col["compression"].lower() == result["compression"].lower() + assert col["encodings"] == result["encodings"] + + def test_dataframe_export(df) -> None: # Guarantees that we have the canonical implementation # reading our dataframe export diff --git a/src/dataframe.rs b/src/dataframe.rs index 7711a0782..3d68db279 100644 --- a/src/dataframe.rs +++ b/src/dataframe.rs @@ -15,6 +15,7 @@ // specific language governing permissions and limitations // under the License. +use std::collections::HashMap; use std::ffi::CString; use std::sync::Arc; @@ -27,7 +28,7 @@ use datafusion::arrow::datatypes::Schema; use datafusion::arrow::pyarrow::{PyArrowType, ToPyArrow}; use datafusion::arrow::util::pretty; use datafusion::common::UnnestOptions; -use datafusion::config::{CsvOptions, TableParquetOptions}; +use datafusion::config::{CsvOptions, ParquetColumnOptions, ParquetOptions, TableParquetOptions}; use datafusion::dataframe::{DataFrame, DataFrameWriteOptions}; use datafusion::datasource::TableProvider; use datafusion::error::DataFusionError; @@ -185,6 +186,101 @@ fn build_formatter_config_from_python(formatter: &Bound<'_, PyAny>) -> PyResult< Ok(config) } +/// Python mapping of `ParquetOptions` (includes just the writer-related options). +#[pyclass(name = "ParquetWriterOptions", module = "datafusion", subclass)] +#[derive(Clone, Default)] +pub struct PyParquetWriterOptions { + options: ParquetOptions, +} + +#[pymethods] +impl PyParquetWriterOptions { + #[new] + #[allow(clippy::too_many_arguments)] + pub fn new( + data_pagesize_limit: usize, + write_batch_size: usize, + writer_version: String, + skip_arrow_metadata: bool, + compression: Option, + dictionary_enabled: Option, + dictionary_page_size_limit: usize, + statistics_enabled: Option, + max_row_group_size: usize, + created_by: String, + column_index_truncate_length: Option, + statistics_truncate_length: Option, + data_page_row_count_limit: usize, + encoding: Option, + bloom_filter_on_write: bool, + bloom_filter_fpp: Option, + bloom_filter_ndv: Option, + allow_single_file_parallelism: bool, + maximum_parallel_row_group_writers: usize, + maximum_buffered_record_batches_per_stream: usize, + ) -> Self { + Self { + options: ParquetOptions { + data_pagesize_limit, + write_batch_size, + writer_version, + skip_arrow_metadata, + compression, + dictionary_enabled, + dictionary_page_size_limit, + statistics_enabled, + max_row_group_size, + created_by, + column_index_truncate_length, + statistics_truncate_length, + data_page_row_count_limit, + encoding, + bloom_filter_on_write, + bloom_filter_fpp, + bloom_filter_ndv, + allow_single_file_parallelism, + maximum_parallel_row_group_writers, + maximum_buffered_record_batches_per_stream, + ..Default::default() + }, + } + } +} + +/// Python mapping of `ParquetColumnOptions`. +#[pyclass(name = "ParquetColumnOptions", module = "datafusion", subclass)] +#[derive(Clone, Default)] +pub struct PyParquetColumnOptions { + options: ParquetColumnOptions, +} + +#[pymethods] +impl PyParquetColumnOptions { + #[new] + pub fn new( + bloom_filter_enabled: Option, + encoding: Option, + dictionary_enabled: Option, + compression: Option, + statistics_enabled: Option, + bloom_filter_fpp: Option, + bloom_filter_ndv: Option, + ) -> Self { + Self { + options: ParquetColumnOptions { + bloom_filter_enabled, + encoding, + dictionary_enabled, + compression, + statistics_enabled, + bloom_filter_fpp, + bloom_filter_ndv, + ..Default::default() + }, + } + } +} + /// A PyDataFrame is a representation of a logical plan and an API to compose statements. /// Use it to build a plan and `.collect()` to execute the plan and collect the result. /// The actual execution of a plan runs natively on Rust and Arrow on a multi-threaded environment. @@ -689,6 +785,34 @@ impl PyDataFrame { Ok(()) } + /// Write a `DataFrame` to a Parquet file, using advanced options. + fn write_parquet_with_options( + &self, + path: &str, + options: PyParquetWriterOptions, + column_specific_options: HashMap, + py: Python, + ) -> PyDataFusionResult<()> { + let table_options = TableParquetOptions { + global: options.options, + column_specific_options: column_specific_options + .into_iter() + .map(|(k, v)| (k, v.options)) + .collect(), + ..Default::default() + }; + + wait_for_future( + py, + self.df.as_ref().clone().write_parquet( + path, + DataFrameWriteOptions::new(), + Option::from(table_options), + ), + )??; + Ok(()) + } + /// Executes a query and writes the results to a partitioned JSON file. fn write_json(&self, path: &str, py: Python) -> PyDataFusionResult<()> { wait_for_future( diff --git a/src/lib.rs b/src/lib.rs index 7dced1fbd..1293eee3c 100644 --- a/src/lib.rs +++ b/src/lib.rs @@ -86,6 +86,8 @@ fn _internal(py: Python, m: Bound<'_, PyModule>) -> PyResult<()> { m.add_class::()?; m.add_class::()?; m.add_class::()?; + m.add_class::()?; + m.add_class::()?; m.add_class::()?; m.add_class::()?; m.add_class::()?; From 1812a0d3e88976f51b40ccfd1a02fddb17ddb8bb Mon Sep 17 00:00:00 2001 From: Kyle Barron Date: Tue, 24 Jun 2025 10:21:53 -0700 Subject: [PATCH 008/220] Fix signature of `__arrow_c_stream__` (#1168) --- python/datafusion/dataframe.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/python/datafusion/dataframe.py b/python/datafusion/dataframe.py index 769271c7e..1fd63bdc6 100644 --- a/python/datafusion/dataframe.py +++ b/python/datafusion/dataframe.py @@ -1053,7 +1053,7 @@ def unnest_columns(self, *columns: str, preserve_nulls: bool = True) -> DataFram columns = list(columns) return DataFrame(self.df.unnest_columns(columns, preserve_nulls=preserve_nulls)) - def __arrow_c_stream__(self, requested_schema: pa.Schema) -> Any: + def __arrow_c_stream__(self, requested_schema: object | None = None) -> object: """Export an Arrow PyCapsule Stream. This will execute and collect the DataFrame. We will attempt to respect the From 0d3c37f9379fd12406022c7edc3fee056866a694 Mon Sep 17 00:00:00 2001 From: kosiew Date: Wed, 25 Jun 2025 07:33:03 +0800 Subject: [PATCH 009/220] Consolidate DataFrame Docs: Merge HTML Rendering Section as Subpage (#1161) * docs: unify dataframe documentation (#2) * docs: update links in basics and rendering documentation * docs: add API reference section to main index * docs: add license information to rendering documentation * Move API Reference under User Guide > Dataframe * Merge data from dataframe api reference page into main dataframe page --------- Co-authored-by: Tim Saucer --- docs/source/api/dataframe.rst | 387 ------------------ docs/source/api/index.rst | 27 -- docs/source/index.rst | 4 +- docs/source/user-guide/basics.rst | 2 +- docs/source/user-guide/dataframe/index.rst | 209 ++++++++++ .../rendering.rst} | 103 ++--- 6 files changed, 265 insertions(+), 467 deletions(-) delete mode 100644 docs/source/api/dataframe.rst delete mode 100644 docs/source/api/index.rst create mode 100644 docs/source/user-guide/dataframe/index.rst rename docs/source/user-guide/{dataframe.rst => dataframe/rendering.rst} (72%) diff --git a/docs/source/api/dataframe.rst b/docs/source/api/dataframe.rst deleted file mode 100644 index a9e9e47c8..000000000 --- a/docs/source/api/dataframe.rst +++ /dev/null @@ -1,387 +0,0 @@ -.. Licensed to the Apache Software Foundation (ASF) under one -.. or more contributor license agreements. See the NOTICE file -.. distributed with this work for additional information -.. regarding copyright ownership. The ASF licenses this file -.. to you under the Apache License, Version 2.0 (the -.. "License"); you may not use this file except in compliance -.. with the License. You may obtain a copy of the License at - -.. http://www.apache.org/licenses/LICENSE-2.0 - -.. Unless required by applicable law or agreed to in writing, -.. software distributed under the License is distributed on an -.. "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY -.. KIND, either express or implied. See the License for the -.. specific language governing permissions and limitations -.. under the License. - -================= -DataFrame API -================= - -Overview --------- - -The ``DataFrame`` class is the core abstraction in DataFusion that represents tabular data and operations -on that data. DataFrames provide a flexible API for transforming data through various operations such as -filtering, projection, aggregation, joining, and more. - -A DataFrame represents a logical plan that is lazily evaluated. The actual execution occurs only when -terminal operations like ``collect()``, ``show()``, or ``to_pandas()`` are called. - -Creating DataFrames -------------------- - -DataFrames can be created in several ways: - -* From SQL queries via a ``SessionContext``: - - .. code-block:: python - - from datafusion import SessionContext - - ctx = SessionContext() - df = ctx.sql("SELECT * FROM your_table") - -* From registered tables: - - .. code-block:: python - - df = ctx.table("your_table") - -* From various data sources: - - .. code-block:: python - - # From CSV files (see :ref:`io_csv` for detailed options) - df = ctx.read_csv("path/to/data.csv") - - # From Parquet files (see :ref:`io_parquet` for detailed options) - df = ctx.read_parquet("path/to/data.parquet") - - # From JSON files (see :ref:`io_json` for detailed options) - df = ctx.read_json("path/to/data.json") - - # From Avro files (see :ref:`io_avro` for detailed options) - df = ctx.read_avro("path/to/data.avro") - - # From Pandas DataFrame - import pandas as pd - pandas_df = pd.DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]}) - df = ctx.from_pandas(pandas_df) - - # From Arrow data - import pyarrow as pa - batch = pa.RecordBatch.from_arrays( - [pa.array([1, 2, 3]), pa.array([4, 5, 6])], - names=["a", "b"] - ) - df = ctx.from_arrow(batch) - - For detailed information about reading from different data sources, see the :doc:`I/O Guide <../user-guide/io/index>`. - For custom data sources, see :ref:`io_custom_table_provider`. - -Common DataFrame Operations ---------------------------- - -DataFusion's DataFrame API offers a wide range of operations: - -.. code-block:: python - - from datafusion import column, literal - - # Select specific columns - df = df.select("col1", "col2") - - # Select with expressions - df = df.select(column("a") + column("b"), column("a") - column("b")) - - # Filter rows - df = df.filter(column("age") > literal(25)) - - # Add computed columns - df = df.with_column("full_name", column("first_name") + literal(" ") + column("last_name")) - - # Multiple column additions - df = df.with_columns( - (column("a") + column("b")).alias("sum"), - (column("a") * column("b")).alias("product") - ) - - # Sort data - df = df.sort(column("age").sort(ascending=False)) - - # Join DataFrames - df = df1.join(df2, on="user_id", how="inner") - - # Aggregate data - from datafusion import functions as f - df = df.aggregate( - [], # Group by columns (empty for global aggregation) - [f.sum(column("amount")).alias("total_amount")] - ) - - # Limit rows - df = df.limit(100) - - # Drop columns - df = df.drop("temporary_column") - -Terminal Operations -------------------- - -To materialize the results of your DataFrame operations: - -.. code-block:: python - - # Collect all data as PyArrow RecordBatches - result_batches = df.collect() - - # Convert to various formats - pandas_df = df.to_pandas() # Pandas DataFrame - polars_df = df.to_polars() # Polars DataFrame - arrow_table = df.to_arrow_table() # PyArrow Table - py_dict = df.to_pydict() # Python dictionary - py_list = df.to_pylist() # Python list of dictionaries - - # Display results - df.show() # Print tabular format to console - - # Count rows - count = df.count() - -HTML Rendering in Jupyter -------------------------- - -When working in Jupyter notebooks or other environments that support rich HTML display, -DataFusion DataFrames automatically render as nicely formatted HTML tables. This functionality -is provided by the ``_repr_html_`` method, which is automatically called by Jupyter. - -Basic HTML Rendering -~~~~~~~~~~~~~~~~~~~~ - -In a Jupyter environment, simply displaying a DataFrame object will trigger HTML rendering: - -.. code-block:: python - - # Will display as HTML table in Jupyter - df - - # Explicit display also uses HTML rendering - display(df) - -HTML Rendering Customization ----------------------------- - -DataFusion provides extensive customization options for HTML table rendering through the -``datafusion.html_formatter`` module. - -Configuring the HTML Formatter -~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - -You can customize how DataFrames are rendered by configuring the formatter: - -.. code-block:: python - - from datafusion.html_formatter import configure_formatter - - configure_formatter( - max_cell_length=30, # Maximum length of cell content before truncation - max_width=800, # Maximum width of table in pixels - max_height=400, # Maximum height of table in pixels - max_memory_bytes=2 * 1024 * 1024,# Maximum memory used for rendering (2MB) - min_rows_display=10, # Minimum rows to display - repr_rows=20, # Number of rows to display in representation - enable_cell_expansion=True, # Allow cells to be expandable on click - custom_css=None, # Custom CSS to apply - show_truncation_message=True, # Show message when data is truncated - style_provider=None, # Custom style provider class - use_shared_styles=True # Share styles across tables to reduce duplication - ) - -Custom Style Providers -~~~~~~~~~~~~~~~~~~~~~~ - -For advanced styling needs, you can create a custom style provider class: - -.. code-block:: python - - from datafusion.html_formatter import configure_formatter - - class CustomStyleProvider: - def get_cell_style(self) -> str: - return "background-color: #f5f5f5; color: #333; padding: 8px; border: 1px solid #ddd;" - - def get_header_style(self) -> str: - return "background-color: #4285f4; color: white; font-weight: bold; padding: 10px;" - - # Apply custom styling - configure_formatter(style_provider=CustomStyleProvider()) - -Custom Type Formatters -~~~~~~~~~~~~~~~~~~~~~~ - -You can register custom formatters for specific data types: - -.. code-block:: python - - from datafusion.html_formatter import get_formatter - - formatter = get_formatter() - - # Format integers with color based on value - def format_int(value): - return f' 100 else "blue"}">{value}' - - formatter.register_formatter(int, format_int) - - # Format date values - def format_date(value): - return f'{value.isoformat()}' - - formatter.register_formatter(datetime.date, format_date) - -Custom Cell Builders -~~~~~~~~~~~~~~~~~~~~ - -For complete control over cell rendering: - -.. code-block:: python - - formatter = get_formatter() - - def custom_cell_builder(value, row, col, table_id): - try: - num_value = float(value) - if num_value > 0: # Positive values get green - return f'{value}' - if num_value < 0: # Negative values get red - return f'{value}' - except (ValueError, TypeError): - pass - - # Default styling for non-numeric or zero values - return f'{value}' - - formatter.set_custom_cell_builder(custom_cell_builder) - -Custom Header Builders -~~~~~~~~~~~~~~~~~~~~~~ - -Similarly, you can customize the rendering of table headers: - -.. code-block:: python - - def custom_header_builder(field): - tooltip = f"Type: {field.type}" - return f'{field.name}' - - formatter.set_custom_header_builder(custom_header_builder) - -Managing Formatter State ------------------------~ - -The HTML formatter maintains global state that can be managed: - -.. code-block:: python - - from datafusion.html_formatter import reset_formatter, reset_styles_loaded_state, get_formatter - - # Reset the formatter to default settings - reset_formatter() - - # Reset only the styles loaded state (useful when styles were loaded but need reloading) - reset_styles_loaded_state() - - # Get the current formatter instance to make changes - formatter = get_formatter() - -Advanced Example: Dashboard-Style Formatting -------------------------------------------~~ - -This example shows how to create a dashboard-like styling for your DataFrames: - -.. code-block:: python - - from datafusion.html_formatter import configure_formatter, get_formatter - - # Define custom CSS - custom_css = """ - .datafusion-table { - font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif; - border-collapse: collapse; - width: 100%; - box-shadow: 0 2px 3px rgba(0,0,0,0.1); - } - .datafusion-table th { - position: sticky; - top: 0; - z-index: 10; - } - .datafusion-table tr:hover td { - background-color: #f1f7fa !important; - } - .datafusion-table .numeric-positive { - color: #0a7c00; - } - .datafusion-table .numeric-negative { - color: #d13438; - } - """ - - class DashboardStyleProvider: - def get_cell_style(self) -> str: - return "padding: 8px 12px; border-bottom: 1px solid #e0e0e0;" - - def get_header_style(self) -> str: - return ("background-color: #0078d4; color: white; font-weight: 600; " - "padding: 12px; text-align: left; border-bottom: 2px solid #005a9e;") - - # Apply configuration - configure_formatter( - max_height=500, - enable_cell_expansion=True, - custom_css=custom_css, - style_provider=DashboardStyleProvider(), - max_cell_length=50 - ) - - # Add custom formatters for numbers - formatter = get_formatter() - - def format_number(value): - try: - num = float(value) - cls = "numeric-positive" if num > 0 else "numeric-negative" if num < 0 else "" - return f'{value:,}' if cls else f'{value:,}' - except (ValueError, TypeError): - return str(value) - - formatter.register_formatter(int, format_number) - formatter.register_formatter(float, format_number) - -Best Practices --------------- - -1. **Memory Management**: For large datasets, use ``max_memory_bytes`` to limit memory usage. - -2. **Responsive Design**: Set reasonable ``max_width`` and ``max_height`` values to ensure tables display well on different screens. - -3. **Style Optimization**: Use ``use_shared_styles=True`` to avoid duplicate style definitions when displaying multiple tables. - -4. **Reset When Needed**: Call ``reset_formatter()`` when you want to start fresh with default settings. - -5. **Cell Expansion**: Use ``enable_cell_expansion=True`` when cells might contain longer content that users may want to see in full. - -Additional Resources --------------------- - -* :doc:`../user-guide/dataframe` - Complete guide to using DataFrames -* :doc:`../user-guide/io/index` - I/O Guide for reading data from various sources -* :doc:`../user-guide/data-sources` - Comprehensive data sources guide -* :ref:`io_csv` - CSV file reading -* :ref:`io_parquet` - Parquet file reading -* :ref:`io_json` - JSON file reading -* :ref:`io_avro` - Avro file reading -* :ref:`io_custom_table_provider` - Custom table providers -* `API Reference `_ - Full API reference diff --git a/docs/source/api/index.rst b/docs/source/api/index.rst deleted file mode 100644 index 7f58227ca..000000000 --- a/docs/source/api/index.rst +++ /dev/null @@ -1,27 +0,0 @@ -.. Licensed to the Apache Software Foundation (ASF) under one -.. or more contributor license agreements. See the NOTICE file -.. distributed with this work for additional information -.. regarding copyright ownership. The ASF licenses this file -.. to you under the Apache License, Version 2.0 (the -.. "License"); you may not use this file except in compliance -.. with the License. You may obtain a copy of the License at - -.. http://www.apache.org/licenses/LICENSE-2.0 - -.. Unless required by applicable law or agreed to in writing, -.. software distributed under the License is distributed on an -.. "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY -.. KIND, either express or implied. See the License for the -.. specific language governing permissions and limitations -.. under the License. - -============= -API Reference -============= - -This section provides detailed API documentation for the DataFusion Python library. - -.. toctree:: - :maxdepth: 2 - - dataframe diff --git a/docs/source/index.rst b/docs/source/index.rst index ff1e47280..adec60f48 100644 --- a/docs/source/index.rst +++ b/docs/source/index.rst @@ -72,7 +72,7 @@ Example user-guide/introduction user-guide/basics user-guide/data-sources - user-guide/dataframe + user-guide/dataframe/index user-guide/common-operations/index user-guide/io/index user-guide/configuration @@ -93,5 +93,3 @@ Example :hidden: :maxdepth: 1 :caption: API - - api/index diff --git a/docs/source/user-guide/basics.rst b/docs/source/user-guide/basics.rst index 2975d9a6b..7c6820461 100644 --- a/docs/source/user-guide/basics.rst +++ b/docs/source/user-guide/basics.rst @@ -73,7 +73,7 @@ DataFrames are typically created by calling a method on :py:class:`~datafusion.c calling the transformation methods, such as :py:func:`~datafusion.dataframe.DataFrame.filter`, :py:func:`~datafusion.dataframe.DataFrame.select`, :py:func:`~datafusion.dataframe.DataFrame.aggregate`, and :py:func:`~datafusion.dataframe.DataFrame.limit` to build up a query definition. -For more details on working with DataFrames, including visualization options and conversion to other formats, see :doc:`dataframe`. +For more details on working with DataFrames, including visualization options and conversion to other formats, see :doc:`dataframe/index`. Expressions ----------- diff --git a/docs/source/user-guide/dataframe/index.rst b/docs/source/user-guide/dataframe/index.rst new file mode 100644 index 000000000..f69485af7 --- /dev/null +++ b/docs/source/user-guide/dataframe/index.rst @@ -0,0 +1,209 @@ +.. Licensed to the Apache Software Foundation (ASF) under one +.. or more contributor license agreements. See the NOTICE file +.. distributed with this work for additional information +.. regarding copyright ownership. The ASF licenses this file +.. to you under the Apache License, Version 2.0 (the +.. "License"); you may not use this file except in compliance +.. with the License. You may obtain a copy of the License at + +.. http://www.apache.org/licenses/LICENSE-2.0 + +.. Unless required by applicable law or agreed to in writing, +.. software distributed under the License is distributed on an +.. "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +.. KIND, either express or implied. See the License for the +.. specific language governing permissions and limitations +.. under the License. + +DataFrames +========== + +Overview +-------- + +The ``DataFrame`` class is the core abstraction in DataFusion that represents tabular data and operations +on that data. DataFrames provide a flexible API for transforming data through various operations such as +filtering, projection, aggregation, joining, and more. + +A DataFrame represents a logical plan that is lazily evaluated. The actual execution occurs only when +terminal operations like ``collect()``, ``show()``, or ``to_pandas()`` are called. + +Creating DataFrames +------------------- + +DataFrames can be created in several ways: + +* From SQL queries via a ``SessionContext``: + + .. code-block:: python + + from datafusion import SessionContext + + ctx = SessionContext() + df = ctx.sql("SELECT * FROM your_table") + +* From registered tables: + + .. code-block:: python + + df = ctx.table("your_table") + +* From various data sources: + + .. code-block:: python + + # From CSV files (see :ref:`io_csv` for detailed options) + df = ctx.read_csv("path/to/data.csv") + + # From Parquet files (see :ref:`io_parquet` for detailed options) + df = ctx.read_parquet("path/to/data.parquet") + + # From JSON files (see :ref:`io_json` for detailed options) + df = ctx.read_json("path/to/data.json") + + # From Avro files (see :ref:`io_avro` for detailed options) + df = ctx.read_avro("path/to/data.avro") + + # From Pandas DataFrame + import pandas as pd + pandas_df = pd.DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]}) + df = ctx.from_pandas(pandas_df) + + # From Arrow data + import pyarrow as pa + batch = pa.RecordBatch.from_arrays( + [pa.array([1, 2, 3]), pa.array([4, 5, 6])], + names=["a", "b"] + ) + df = ctx.from_arrow(batch) + +For detailed information about reading from different data sources, see the :doc:`I/O Guide <../io/index>`. +For custom data sources, see :ref:`io_custom_table_provider`. + +Common DataFrame Operations +--------------------------- + +DataFusion's DataFrame API offers a wide range of operations: + +.. code-block:: python + + from datafusion import column, literal + + # Select specific columns + df = df.select("col1", "col2") + + # Select with expressions + df = df.select(column("a") + column("b"), column("a") - column("b")) + + # Filter rows + df = df.filter(column("age") > literal(25)) + + # Add computed columns + df = df.with_column("full_name", column("first_name") + literal(" ") + column("last_name")) + + # Multiple column additions + df = df.with_columns( + (column("a") + column("b")).alias("sum"), + (column("a") * column("b")).alias("product") + ) + + # Sort data + df = df.sort(column("age").sort(ascending=False)) + + # Join DataFrames + df = df1.join(df2, on="user_id", how="inner") + + # Aggregate data + from datafusion import functions as f + df = df.aggregate( + [], # Group by columns (empty for global aggregation) + [f.sum(column("amount")).alias("total_amount")] + ) + + # Limit rows + df = df.limit(100) + + # Drop columns + df = df.drop("temporary_column") + +Terminal Operations +------------------- + +To materialize the results of your DataFrame operations: + +.. code-block:: python + + # Collect all data as PyArrow RecordBatches + result_batches = df.collect() + + # Convert to various formats + pandas_df = df.to_pandas() # Pandas DataFrame + polars_df = df.to_polars() # Polars DataFrame + arrow_table = df.to_arrow_table() # PyArrow Table + py_dict = df.to_pydict() # Python dictionary + py_list = df.to_pylist() # Python list of dictionaries + + # Display results + df.show() # Print tabular format to console + + # Count rows + count = df.count() + +HTML Rendering +-------------- + +When working in Jupyter notebooks or other environments that support HTML rendering, DataFrames will +automatically display as formatted HTML tables. For detailed information about customizing HTML +rendering, formatting options, and advanced styling, see :doc:`rendering`. + +Core Classes +------------ + +**DataFrame** + The main DataFrame class for building and executing queries. + + See: :py:class:`datafusion.DataFrame` + +**SessionContext** + The primary entry point for creating DataFrames from various data sources. + + Key methods for DataFrame creation: + + * :py:meth:`~datafusion.SessionContext.read_csv` - Read CSV files + * :py:meth:`~datafusion.SessionContext.read_parquet` - Read Parquet files + * :py:meth:`~datafusion.SessionContext.read_json` - Read JSON files + * :py:meth:`~datafusion.SessionContext.read_avro` - Read Avro files + * :py:meth:`~datafusion.SessionContext.table` - Access registered tables + * :py:meth:`~datafusion.SessionContext.sql` - Execute SQL queries + * :py:meth:`~datafusion.SessionContext.from_pandas` - Create from Pandas DataFrame + * :py:meth:`~datafusion.SessionContext.from_arrow` - Create from Arrow data + + See: :py:class:`datafusion.SessionContext` + +Expression Classes +------------------ + +**Expr** + Represents expressions that can be used in DataFrame operations. + + See: :py:class:`datafusion.Expr` + +**Functions for creating expressions:** + +* :py:func:`datafusion.column` - Reference a column by name +* :py:func:`datafusion.literal` - Create a literal value expression + +Built-in Functions +------------------ + +DataFusion provides many built-in functions for data manipulation: + +* :py:mod:`datafusion.functions` - Mathematical, string, date/time, and aggregation functions + +For a complete list of available functions, see the :py:mod:`datafusion.functions` module documentation. + + +.. toctree:: + :maxdepth: 1 + + rendering diff --git a/docs/source/user-guide/dataframe.rst b/docs/source/user-guide/dataframe/rendering.rst similarity index 72% rename from docs/source/user-guide/dataframe.rst rename to docs/source/user-guide/dataframe/rendering.rst index 23c65b5f6..4c37c7471 100644 --- a/docs/source/user-guide/dataframe.rst +++ b/docs/source/user-guide/dataframe/rendering.rst @@ -15,59 +15,37 @@ .. specific language governing permissions and limitations .. under the License. -DataFrames -========== +HTML Rendering in Jupyter +========================= -Overview --------- +When working in Jupyter notebooks or other environments that support rich HTML display, +DataFusion DataFrames automatically render as nicely formatted HTML tables. This functionality +is provided by the ``_repr_html_`` method, which is automatically called by Jupyter to provide +a richer visualization than plain text output. -DataFusion's DataFrame API provides a powerful interface for building and executing queries against data sources. -It offers a familiar API similar to pandas and other DataFrame libraries, but with the performance benefits of Rust -and Arrow. +Basic HTML Rendering +-------------------- -A DataFrame represents a logical plan that can be composed through operations like filtering, projection, and aggregation. -The actual execution happens when terminal operations like ``collect()`` or ``show()`` are called. - -Basic Usage ------------ +In a Jupyter environment, simply displaying a DataFrame object will trigger HTML rendering: .. code-block:: python - import datafusion - from datafusion import col, lit + # Will display as HTML table in Jupyter + df - # Create a context and register a data source - ctx = datafusion.SessionContext() - ctx.register_csv("my_table", "path/to/data.csv") - - # Create and manipulate a DataFrame - df = ctx.sql("SELECT * FROM my_table") - - # Or use the DataFrame API directly - df = (ctx.table("my_table") - .filter(col("age") > lit(25)) - .select([col("name"), col("age")])) - - # Execute and collect results - result = df.collect() - - # Display the first few rows - df.show() + # Explicit display also uses HTML rendering + display(df) -HTML Rendering --------------- - -When working in Jupyter notebooks or other environments that support HTML rendering, DataFrames will -automatically display as formatted HTML tables, making it easier to visualize your data. +Customizing HTML Rendering +--------------------------- -The ``_repr_html_`` method is called automatically by Jupyter to render a DataFrame. This method -controls how DataFrames appear in notebook environments, providing a richer visualization than -plain text output. +DataFusion provides extensive customization options for HTML table rendering through the +``datafusion.html_formatter`` module. -Customizing HTML Rendering --------------------------- +Configuring the HTML Formatter +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ -You can customize how DataFrames are rendered in HTML by configuring the formatter: +You can customize how DataFrames are rendered by configuring the formatter: .. code-block:: python @@ -91,7 +69,7 @@ You can customize how DataFrames are rendered in HTML by configuring the formatt The formatter settings affect all DataFrames displayed after configuration. Custom Style Providers ----------------------- +----------------------- For advanced styling needs, you can create a custom style provider: @@ -118,7 +96,8 @@ For advanced styling needs, you can create a custom style provider: configure_formatter(style_provider=MyStyleProvider()) Performance Optimization with Shared Styles -------------------------------------------- +-------------------------------------------- + The ``use_shared_styles`` parameter (enabled by default) optimizes performance when displaying multiple DataFrames in notebook environments: @@ -138,7 +117,7 @@ When ``use_shared_styles=True``: - Applies consistent styling across all DataFrames Creating a Custom Formatter ---------------------------- +---------------------------- For complete control over rendering, you can implement a custom formatter: @@ -184,7 +163,7 @@ Get the current formatter settings: print(formatter.theme) Contextual Formatting ---------------------- +---------------------- You can also use a context manager to temporarily change formatting settings: @@ -207,12 +186,38 @@ Memory and Display Controls You can control how much data is displayed and how much memory is used for rendering: - .. code-block:: python - +.. code-block:: python + configure_formatter( max_memory_bytes=4 * 1024 * 1024, # 4MB maximum memory for display min_rows_display=50, # Always show at least 50 rows repr_rows=20 # Show 20 rows in __repr__ output ) -These parameters help balance comprehensive data display against performance considerations. \ No newline at end of file +These parameters help balance comprehensive data display against performance considerations. + +Best Practices +-------------- + +1. **Global Configuration**: Use ``configure_formatter()`` at the beginning of your notebook to set up consistent formatting for all DataFrames. + +2. **Memory Management**: Set appropriate ``max_memory_bytes`` limits to prevent performance issues with large datasets. + +3. **Shared Styles**: Keep ``use_shared_styles=True`` (default) for better performance in notebooks with multiple DataFrames. + +4. **Reset When Needed**: Call ``reset_formatter()`` when you want to start fresh with default settings. + +5. **Cell Expansion**: Use ``enable_cell_expansion=True`` when cells might contain longer content that users may want to see in full. + +Additional Resources +-------------------- + +* :doc:`../dataframe/index` - Complete guide to using DataFrames +* :doc:`../io/index` - I/O Guide for reading data from various sources +* :doc:`../data-sources` - Comprehensive data sources guide +* :ref:`io_csv` - CSV file reading +* :ref:`io_parquet` - Parquet file reading +* :ref:`io_json` - JSON file reading +* :ref:`io_avro` - Avro file reading +* :ref:`io_custom_table_provider` - Custom table providers +* `API Reference `_ - Full API reference From ef62fa897e6e8334779f556aad4c0da89ebdc86f Mon Sep 17 00:00:00 2001 From: kosiew Date: Wed, 25 Jun 2025 19:27:01 +0800 Subject: [PATCH 010/220] Add compression_level support to ParquetWriterOptions and enhance write_parquet to accept full options object (#1169) * feat: Add Parquet writer option autodetection * Add compression_level to ParquetWriterOptions * fix ruff errors * feat: Add overloads for write_parquet method to support various compression options --- python/datafusion/__init__.py | 3 +-- python/datafusion/dataframe.py | 40 +++++++++++++++++++++++++++++++--- python/tests/test_dataframe.py | 16 ++++++++++++++ 3 files changed, 54 insertions(+), 5 deletions(-) diff --git a/python/datafusion/__init__.py b/python/datafusion/__init__.py index 16d65f685..69062fd30 100644 --- a/python/datafusion/__init__.py +++ b/python/datafusion/__init__.py @@ -30,13 +30,12 @@ except ImportError: import importlib_metadata -from datafusion.col import col, column - from . import functions, object_store, substrait, unparser # The following imports are okay to remain as opaque to the user. from ._internal import Config from .catalog import Catalog, Database, Table +from .col import col, column from .common import ( DFSchema, ) diff --git a/python/datafusion/dataframe.py b/python/datafusion/dataframe.py index 1fd63bdc6..49c0b2f8b 100644 --- a/python/datafusion/dataframe.py +++ b/python/datafusion/dataframe.py @@ -191,6 +191,7 @@ def __init__( writer_version: str = "1.0", skip_arrow_metadata: bool = False, compression: Optional[str] = "zstd(3)", + compression_level: Optional[int] = None, dictionary_enabled: Optional[bool] = True, dictionary_page_size_limit: int = 1024 * 1024, statistics_enabled: Optional[str] = "page", @@ -213,7 +214,10 @@ def __init__( self.write_batch_size = write_batch_size self.writer_version = writer_version self.skip_arrow_metadata = skip_arrow_metadata - self.compression = compression + if compression_level is not None: + self.compression = f"{compression}({compression_level})" + else: + self.compression = compression self.dictionary_enabled = dictionary_enabled self.dictionary_page_size_limit = dictionary_page_size_limit self.statistics_enabled = statistics_enabled @@ -870,10 +874,34 @@ def write_csv(self, path: str | pathlib.Path, with_header: bool = False) -> None """ self.df.write_csv(str(path), with_header) + @overload + def write_parquet( + self, + path: str | pathlib.Path, + compression: str, + compression_level: int | None = None, + ) -> None: ... + + @overload + def write_parquet( + self, + path: str | pathlib.Path, + compression: Compression = Compression.ZSTD, + compression_level: int | None = None, + ) -> None: ... + + @overload + def write_parquet( + self, + path: str | pathlib.Path, + compression: ParquetWriterOptions, + compression_level: None = None, + ) -> None: ... + def write_parquet( self, path: str | pathlib.Path, - compression: Union[str, Compression] = Compression.ZSTD, + compression: Union[str, Compression, ParquetWriterOptions] = Compression.ZSTD, compression_level: int | None = None, ) -> None: """Execute the :py:class:`DataFrame` and write the results to a Parquet file. @@ -894,7 +922,13 @@ def write_parquet( recommended range is 1 to 22, with the default being 4. Higher levels provide better compression but slower speed. """ - # Convert string to Compression enum if necessary + if isinstance(compression, ParquetWriterOptions): + if compression_level is not None: + msg = "compression_level should be None when using ParquetWriterOptions" + raise ValueError(msg) + self.write_parquet_with_options(path, compression) + return + if isinstance(compression, str): compression = Compression.from_str(compression) diff --git a/python/tests/test_dataframe.py b/python/tests/test_dataframe.py index 3c9b97f23..deaa30b3d 100644 --- a/python/tests/test_dataframe.py +++ b/python/tests/test_dataframe.py @@ -2038,6 +2038,22 @@ def test_write_parquet_with_options_column_options(df, tmp_path): assert col["encodings"] == result["encodings"] +def test_write_parquet_options(df, tmp_path): + options = ParquetWriterOptions(compression="gzip", compression_level=6) + df.write_parquet(str(tmp_path), options) + + result = pq.read_table(str(tmp_path)).to_pydict() + expected = df.to_pydict() + + assert result == expected + + +def test_write_parquet_options_error(df, tmp_path): + options = ParquetWriterOptions(compression="gzip", compression_level=6) + with pytest.raises(ValueError): + df.write_parquet(str(tmp_path), options, compression_level=1) + + def test_dataframe_export(df) -> None: # Guarantees that we have the canonical implementation # reading our dataframe export From 03921d9629c576afc43b63ad67172f70abcd73e8 Mon Sep 17 00:00:00 2001 From: Tim Saucer Date: Wed, 25 Jun 2025 08:20:49 -0400 Subject: [PATCH 011/220] feat: dataframe string formatter (#1170) * Add string formatter * Rename html_formatter to dataframe_formatter * Add deprecation warning * Small adjustments based on user feedback --- python/datafusion/__init__.py | 2 +- python/datafusion/dataframe.py | 15 +- python/datafusion/dataframe_formatter.py | 739 +++++++++++++++++++++++ python/datafusion/html_formatter.py | 700 +-------------------- python/tests/test_dataframe.py | 4 +- src/dataframe.rs | 109 ++-- 6 files changed, 829 insertions(+), 740 deletions(-) create mode 100644 python/datafusion/dataframe_formatter.py diff --git a/python/datafusion/__init__.py b/python/datafusion/__init__.py index 69062fd30..fd7f4fc06 100644 --- a/python/datafusion/__init__.py +++ b/python/datafusion/__init__.py @@ -46,11 +46,11 @@ SQLOptions, ) from .dataframe import DataFrame, ParquetColumnOptions, ParquetWriterOptions +from .dataframe_formatter import configure_formatter from .expr import ( Expr, WindowFrame, ) -from .html_formatter import configure_formatter from .io import read_avro, read_csv, read_json, read_parquet from .plan import ExecutionPlan, LogicalPlan from .record_batch import RecordBatch, RecordBatchStream diff --git a/python/datafusion/dataframe.py b/python/datafusion/dataframe.py index 49c0b2f8b..991e6875a 100644 --- a/python/datafusion/dataframe.py +++ b/python/datafusion/dataframe.py @@ -52,7 +52,6 @@ import polars as pl import pyarrow as pa - from datafusion._internal import DataFrame as DataFrameInternal from datafusion._internal import expr as expr_internal from enum import Enum @@ -327,6 +326,20 @@ def __repr__(self) -> str: def _repr_html_(self) -> str: return self.df._repr_html_() + @staticmethod + def default_str_repr( + batches: list[pa.RecordBatch], + schema: pa.Schema, + has_more: bool, + table_uuid: str | None = None, + ) -> str: + """Return the default string representation of a DataFrame. + + This method is used by the default formatter and implemented in Rust for + performance reasons. + """ + return DataFrameInternal.default_str_repr(batches, schema, has_more, table_uuid) + def describe(self) -> DataFrame: """Return the statistics for this DataFrame. diff --git a/python/datafusion/dataframe_formatter.py b/python/datafusion/dataframe_formatter.py new file mode 100644 index 000000000..27f00f9c3 --- /dev/null +++ b/python/datafusion/dataframe_formatter.py @@ -0,0 +1,739 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. +"""HTML formatting utilities for DataFusion DataFrames.""" + +from __future__ import annotations + +from typing import ( + Any, + Callable, + Optional, + Protocol, + runtime_checkable, +) + +from datafusion._internal import DataFrame as DataFrameInternal + + +def _validate_positive_int(value: Any, param_name: str) -> None: + """Validate that a parameter is a positive integer. + + Args: + value: The value to validate + param_name: Name of the parameter (used in error message) + + Raises: + ValueError: If the value is not a positive integer + """ + if not isinstance(value, int) or value <= 0: + msg = f"{param_name} must be a positive integer" + raise ValueError(msg) + + +def _validate_bool(value: Any, param_name: str) -> None: + """Validate that a parameter is a boolean. + + Args: + value: The value to validate + param_name: Name of the parameter (used in error message) + + Raises: + TypeError: If the value is not a boolean + """ + if not isinstance(value, bool): + msg = f"{param_name} must be a boolean" + raise TypeError(msg) + + +@runtime_checkable +class CellFormatter(Protocol): + """Protocol for cell value formatters.""" + + def __call__(self, value: Any) -> str: + """Format a cell value to string representation.""" + ... + + +@runtime_checkable +class StyleProvider(Protocol): + """Protocol for HTML style providers.""" + + def get_cell_style(self) -> str: + """Get the CSS style for table cells.""" + ... + + def get_header_style(self) -> str: + """Get the CSS style for header cells.""" + ... + + +class DefaultStyleProvider: + """Default implementation of StyleProvider.""" + + def get_cell_style(self) -> str: + """Get the CSS style for table cells. + + Returns: + CSS style string + """ + return ( + "border: 1px solid black; padding: 8px; text-align: left; " + "white-space: nowrap;" + ) + + def get_header_style(self) -> str: + """Get the CSS style for header cells. + + Returns: + CSS style string + """ + return ( + "border: 1px solid black; padding: 8px; text-align: left; " + "background-color: #f2f2f2; white-space: nowrap; min-width: fit-content; " + "max-width: fit-content;" + ) + + +class DataFrameHtmlFormatter: + """Configurable HTML formatter for DataFusion DataFrames. + + This class handles the HTML rendering of DataFrames for display in + Jupyter notebooks and other rich display contexts. + + This class supports extension through composition. Key extension points: + - Provide a custom StyleProvider for styling cells and headers + - Register custom formatters for specific types + - Provide custom cell builders for specialized cell rendering + + Args: + max_cell_length: Maximum characters to display in a cell before truncation + max_width: Maximum width of the HTML table in pixels + max_height: Maximum height of the HTML table in pixels + max_memory_bytes: Maximum memory in bytes for rendered data (default: 2MB) + min_rows_display: Minimum number of rows to display + repr_rows: Default number of rows to display in repr output + enable_cell_expansion: Whether to add expand/collapse buttons for long cell + values + custom_css: Additional CSS to include in the HTML output + show_truncation_message: Whether to display a message when data is truncated + style_provider: Custom provider for cell and header styles + use_shared_styles: Whether to load styles and scripts only once per notebook + session + """ + + # Class variable to track if styles have been loaded in the notebook + _styles_loaded = False + + def __init__( + self, + max_cell_length: int = 25, + max_width: int = 1000, + max_height: int = 300, + max_memory_bytes: int = 2 * 1024 * 1024, # 2 MB + min_rows_display: int = 20, + repr_rows: int = 10, + enable_cell_expansion: bool = True, + custom_css: Optional[str] = None, + show_truncation_message: bool = True, + style_provider: Optional[StyleProvider] = None, + use_shared_styles: bool = True, + ) -> None: + """Initialize the HTML formatter. + + Parameters + ---------- + max_cell_length : int, default 25 + Maximum length of cell content before truncation. + max_width : int, default 1000 + Maximum width of the displayed table in pixels. + max_height : int, default 300 + Maximum height of the displayed table in pixels. + max_memory_bytes : int, default 2097152 (2MB) + Maximum memory in bytes for rendered data. + min_rows_display : int, default 20 + Minimum number of rows to display. + repr_rows : int, default 10 + Default number of rows to display in repr output. + enable_cell_expansion : bool, default True + Whether to allow cells to expand when clicked. + custom_css : str, optional + Custom CSS to apply to the HTML table. + show_truncation_message : bool, default True + Whether to show a message indicating that content has been truncated. + style_provider : StyleProvider, optional + Provider of CSS styles for the HTML table. If None, DefaultStyleProvider + is used. + use_shared_styles : bool, default True + Whether to use shared styles across multiple tables. + + Raises: + ------ + ValueError + If max_cell_length, max_width, max_height, max_memory_bytes, + min_rows_display, or repr_rows is not a positive integer. + TypeError + If enable_cell_expansion, show_truncation_message, or use_shared_styles is + not a boolean, + or if custom_css is provided but is not a string, + or if style_provider is provided but does not implement the StyleProvider + protocol. + """ + # Validate numeric parameters + _validate_positive_int(max_cell_length, "max_cell_length") + _validate_positive_int(max_width, "max_width") + _validate_positive_int(max_height, "max_height") + _validate_positive_int(max_memory_bytes, "max_memory_bytes") + _validate_positive_int(min_rows_display, "min_rows_display") + _validate_positive_int(repr_rows, "repr_rows") + + # Validate boolean parameters + _validate_bool(enable_cell_expansion, "enable_cell_expansion") + _validate_bool(show_truncation_message, "show_truncation_message") + _validate_bool(use_shared_styles, "use_shared_styles") + + # Validate custom_css + if custom_css is not None and not isinstance(custom_css, str): + msg = "custom_css must be None or a string" + raise TypeError(msg) + + # Validate style_provider + if style_provider is not None and not isinstance(style_provider, StyleProvider): + msg = "style_provider must implement the StyleProvider protocol" + raise TypeError(msg) + + self.max_cell_length = max_cell_length + self.max_width = max_width + self.max_height = max_height + self.max_memory_bytes = max_memory_bytes + self.min_rows_display = min_rows_display + self.repr_rows = repr_rows + self.enable_cell_expansion = enable_cell_expansion + self.custom_css = custom_css + self.show_truncation_message = show_truncation_message + self.style_provider = style_provider or DefaultStyleProvider() + self.use_shared_styles = use_shared_styles + # Registry for custom type formatters + self._type_formatters: dict[type, CellFormatter] = {} + # Custom cell builders + self._custom_cell_builder: Optional[Callable[[Any, int, int, str], str]] = None + self._custom_header_builder: Optional[Callable[[Any], str]] = None + + def register_formatter(self, type_class: type, formatter: CellFormatter) -> None: + """Register a custom formatter for a specific data type. + + Args: + type_class: The type to register a formatter for + formatter: Function that takes a value of the given type and returns + a formatted string + """ + self._type_formatters[type_class] = formatter + + def set_custom_cell_builder( + self, builder: Callable[[Any, int, int, str], str] + ) -> None: + """Set a custom cell builder function. + + Args: + builder: Function that takes (value, row, col, table_id) and returns HTML + """ + self._custom_cell_builder = builder + + def set_custom_header_builder(self, builder: Callable[[Any], str]) -> None: + """Set a custom header builder function. + + Args: + builder: Function that takes a field and returns HTML + """ + self._custom_header_builder = builder + + @classmethod + def is_styles_loaded(cls) -> bool: + """Check if HTML styles have been loaded in the current session. + + This method is primarily intended for debugging UI rendering issues + related to style loading. + + Returns: + True if styles have been loaded, False otherwise + + Example: + >>> from datafusion.dataframe_formatter import DataFrameHtmlFormatter + >>> DataFrameHtmlFormatter.is_styles_loaded() + False + """ + return cls._styles_loaded + + def format_html( + self, + batches: list, + schema: Any, + has_more: bool = False, + table_uuid: str | None = None, + ) -> str: + """Format record batches as HTML. + + This method is used by DataFrame's _repr_html_ implementation and can be + called directly when custom HTML rendering is needed. + + Args: + batches: List of Arrow RecordBatch objects + schema: Arrow Schema object + has_more: Whether there are more batches not shown + table_uuid: Unique ID for the table, used for JavaScript interactions + + Returns: + HTML string representation of the data + + Raises: + TypeError: If schema is invalid and no batches are provided + """ + if not batches: + return "No data to display" + + # Validate schema + if schema is None or not hasattr(schema, "__iter__"): + msg = "Schema must be provided" + raise TypeError(msg) + + # Generate a unique ID if none provided + table_uuid = table_uuid or f"df-{id(batches)}" + + # Build HTML components + html = [] + + # Only include styles and scripts if: + # 1. Not using shared styles, OR + # 2. Using shared styles but they haven't been loaded yet + include_styles = ( + not self.use_shared_styles or not DataFrameHtmlFormatter._styles_loaded + ) + + if include_styles: + html.extend(self._build_html_header()) + # If we're using shared styles, mark them as loaded + if self.use_shared_styles: + DataFrameHtmlFormatter._styles_loaded = True + + html.extend(self._build_table_container_start()) + + # Add table header and body + html.extend(self._build_table_header(schema)) + html.extend(self._build_table_body(batches, table_uuid)) + + html.append("") + html.append("") + + # Add footer (JavaScript and messages) + if include_styles and self.enable_cell_expansion: + html.append(self._get_javascript()) + + # Always add truncation message if needed (independent of styles) + if has_more and self.show_truncation_message: + html.append("
Data truncated due to size.
") + + return "\n".join(html) + + def format_str( + self, + batches: list, + schema: Any, + has_more: bool = False, + table_uuid: str | None = None, + ) -> str: + """Format record batches as a string. + + This method is used by DataFrame's __repr__ implementation and can be + called directly when string rendering is needed. + + Args: + batches: List of Arrow RecordBatch objects + schema: Arrow Schema object + has_more: Whether there are more batches not shown + table_uuid: Unique ID for the table, used for JavaScript interactions + + Returns: + String representation of the data + + Raises: + TypeError: If schema is invalid and no batches are provided + """ + return DataFrameInternal.default_str_repr(batches, schema, has_more, table_uuid) + + def _build_html_header(self) -> list[str]: + """Build the HTML header with CSS styles.""" + html = [] + html.append("") + return html + + def _build_table_container_start(self) -> list[str]: + """Build the opening tags for the table container.""" + html = [] + html.append( + f'
' + ) + html.append('') + return html + + def _build_table_header(self, schema: Any) -> list[str]: + """Build the HTML table header with column names.""" + html = [] + html.append("") + html.append("") + for field in schema: + if self._custom_header_builder: + html.append(self._custom_header_builder(field)) + else: + html.append( + f"" + ) + html.append("") + html.append("") + return html + + def _build_table_body(self, batches: list, table_uuid: str) -> list[str]: + """Build the HTML table body with data rows.""" + html = [] + html.append("") + + row_count = 0 + for batch in batches: + for row_idx in range(batch.num_rows): + row_count += 1 + html.append("") + + for col_idx, column in enumerate(batch.columns): + # Get the raw value from the column + raw_value = self._get_cell_value(column, row_idx) + + # Always check for type formatters first to format the value + formatted_value = self._format_cell_value(raw_value) + + # Then apply either custom cell builder or standard cell formatting + if self._custom_cell_builder: + # Pass both the raw value and formatted value to let the + # builder decide + cell_html = self._custom_cell_builder( + raw_value, row_count, col_idx, table_uuid + ) + html.append(cell_html) + else: + # Standard cell formatting with formatted value + if ( + len(str(raw_value)) > self.max_cell_length + and self.enable_cell_expansion + ): + cell_html = self._build_expandable_cell( + formatted_value, row_count, col_idx, table_uuid + ) + else: + cell_html = self._build_regular_cell(formatted_value) + html.append(cell_html) + + html.append("") + + html.append("") + return html + + def _get_cell_value(self, column: Any, row_idx: int) -> Any: + """Extract a cell value from a column. + + Args: + column: Arrow array + row_idx: Row index + + Returns: + The raw cell value + """ + try: + value = column[row_idx] + + if hasattr(value, "as_py"): + return value.as_py() + except (AttributeError, TypeError): + pass + else: + return value + + def _format_cell_value(self, value: Any) -> str: + """Format a cell value for display. + + Uses registered type formatters if available. + + Args: + value: The cell value to format + + Returns: + Formatted cell value as string + """ + # Check for custom type formatters + for type_cls, formatter in self._type_formatters.items(): + if isinstance(value, type_cls): + return formatter(value) + + # If no formatter matched, return string representation + return str(value) + + def _build_expandable_cell( + self, formatted_value: str, row_count: int, col_idx: int, table_uuid: str + ) -> str: + """Build an expandable cell for long content.""" + short_value = str(formatted_value)[: self.max_cell_length] + return ( + f"" + ) + + def _build_regular_cell(self, formatted_value: str) -> str: + """Build a regular table cell.""" + return ( + f"" + ) + + def _build_html_footer(self, has_more: bool) -> list[str]: + """Build the HTML footer with JavaScript and messages.""" + html = [] + + # Add JavaScript for interactivity only if cell expansion is enabled + # and we're not using the shared styles approach + if self.enable_cell_expansion and not self.use_shared_styles: + html.append(self._get_javascript()) + + # Add truncation message if needed + if has_more and self.show_truncation_message: + html.append("
Data truncated due to size.
") + + return html + + def _get_default_css(self) -> str: + """Get default CSS styles for the HTML table.""" + return """ + .expandable-container { + display: inline-block; + max-width: 200px; + } + .expandable { + white-space: nowrap; + overflow: hidden; + text-overflow: ellipsis; + display: block; + } + .full-text { + display: none; + white-space: normal; + } + .expand-btn { + cursor: pointer; + color: blue; + text-decoration: underline; + border: none; + background: none; + font-size: inherit; + display: block; + margin-top: 5px; + } + """ + + def _get_javascript(self) -> str: + """Get JavaScript code for interactive elements.""" + return """ + + """ + + +class FormatterManager: + """Manager class for the global DataFrame HTML formatter instance.""" + + _default_formatter: DataFrameHtmlFormatter = DataFrameHtmlFormatter() + + @classmethod + def set_formatter(cls, formatter: DataFrameHtmlFormatter) -> None: + """Set the global DataFrame HTML formatter. + + Args: + formatter: The formatter instance to use globally + """ + cls._default_formatter = formatter + _refresh_formatter_reference() + + @classmethod + def get_formatter(cls) -> DataFrameHtmlFormatter: + """Get the current global DataFrame HTML formatter. + + Returns: + The global HTML formatter instance + """ + return cls._default_formatter + + +def get_formatter() -> DataFrameHtmlFormatter: + """Get the current global DataFrame HTML formatter. + + This function is used by the DataFrame._repr_html_ implementation to access + the shared formatter instance. It can also be used directly when custom + HTML rendering is needed. + + Returns: + The global HTML formatter instance + + Example: + >>> from datafusion.html_formatter import get_formatter + >>> formatter = get_formatter() + >>> formatter.max_cell_length = 50 # Increase cell length + """ + return FormatterManager.get_formatter() + + +def set_formatter(formatter: DataFrameHtmlFormatter) -> None: + """Set the global DataFrame HTML formatter. + + Args: + formatter: The formatter instance to use globally + + Example: + >>> from datafusion.html_formatter import get_formatter, set_formatter + >>> custom_formatter = DataFrameHtmlFormatter(max_cell_length=100) + >>> set_formatter(custom_formatter) + """ + FormatterManager.set_formatter(formatter) + + +def configure_formatter(**kwargs: Any) -> None: + """Configure the global DataFrame HTML formatter. + + This function creates a new formatter with the provided configuration + and sets it as the global formatter for all DataFrames. + + Args: + **kwargs: Formatter configuration parameters like max_cell_length, + max_width, max_height, enable_cell_expansion, etc. + + Raises: + ValueError: If any invalid parameters are provided + + Example: + >>> from datafusion.html_formatter import configure_formatter + >>> configure_formatter( + ... max_cell_length=50, + ... max_height=500, + ... enable_cell_expansion=True, + ... use_shared_styles=True + ... ) + """ + # Valid parameters accepted by DataFrameHtmlFormatter + valid_params = { + "max_cell_length", + "max_width", + "max_height", + "max_memory_bytes", + "min_rows_display", + "repr_rows", + "enable_cell_expansion", + "custom_css", + "show_truncation_message", + "style_provider", + "use_shared_styles", + } + + # Check for invalid parameters + invalid_params = set(kwargs) - valid_params + if invalid_params: + msg = ( + f"Invalid formatter parameters: {', '.join(invalid_params)}. " + f"Valid parameters are: {', '.join(valid_params)}" + ) + raise ValueError(msg) + + # Create and set formatter with validated parameters + set_formatter(DataFrameHtmlFormatter(**kwargs)) + + +def reset_formatter() -> None: + """Reset the global DataFrame HTML formatter to default settings. + + This function creates a new formatter with default configuration + and sets it as the global formatter for all DataFrames. + + Example: + >>> from datafusion.html_formatter import reset_formatter + >>> reset_formatter() # Reset formatter to default settings + """ + formatter = DataFrameHtmlFormatter() + # Reset the styles_loaded flag to ensure styles will be reloaded + DataFrameHtmlFormatter._styles_loaded = False + set_formatter(formatter) + + +def reset_styles_loaded_state() -> None: + """Reset the styles loaded state to force reloading of styles. + + This can be useful when switching between notebook sessions or + when styles need to be refreshed. + + Example: + >>> from datafusion.html_formatter import reset_styles_loaded_state + >>> reset_styles_loaded_state() # Force styles to reload in next render + """ + DataFrameHtmlFormatter._styles_loaded = False + + +def _refresh_formatter_reference() -> None: + """Refresh formatter reference in any modules using it. + + This helps ensure that changes to the formatter are reflected in existing + DataFrames that might be caching the formatter reference. + """ + # This is a no-op but signals modules to refresh their reference diff --git a/python/datafusion/html_formatter.py b/python/datafusion/html_formatter.py index 12a7e4553..65eb1f042 100644 --- a/python/datafusion/html_formatter.py +++ b/python/datafusion/html_formatter.py @@ -14,698 +14,16 @@ # KIND, either express or implied. See the License for the # specific language governing permissions and limitations # under the License. -"""HTML formatting utilities for DataFusion DataFrames.""" -from __future__ import annotations +"""Deprecated module for dataframe formatting.""" -from typing import ( - Any, - Callable, - Optional, - Protocol, - runtime_checkable, -) - - -def _validate_positive_int(value: Any, param_name: str) -> None: - """Validate that a parameter is a positive integer. - - Args: - value: The value to validate - param_name: Name of the parameter (used in error message) - - Raises: - ValueError: If the value is not a positive integer - """ - if not isinstance(value, int) or value <= 0: - msg = f"{param_name} must be a positive integer" - raise ValueError(msg) - - -def _validate_bool(value: Any, param_name: str) -> None: - """Validate that a parameter is a boolean. - - Args: - value: The value to validate - param_name: Name of the parameter (used in error message) - - Raises: - TypeError: If the value is not a boolean - """ - if not isinstance(value, bool): - msg = f"{param_name} must be a boolean" - raise TypeError(msg) - - -@runtime_checkable -class CellFormatter(Protocol): - """Protocol for cell value formatters.""" - - def __call__(self, value: Any) -> str: - """Format a cell value to string representation.""" - ... - - -@runtime_checkable -class StyleProvider(Protocol): - """Protocol for HTML style providers.""" - - def get_cell_style(self) -> str: - """Get the CSS style for table cells.""" - ... - - def get_header_style(self) -> str: - """Get the CSS style for header cells.""" - ... - - -class DefaultStyleProvider: - """Default implementation of StyleProvider.""" - - def get_cell_style(self) -> str: - """Get the CSS style for table cells. - - Returns: - CSS style string - """ - return ( - "border: 1px solid black; padding: 8px; text-align: left; " - "white-space: nowrap;" - ) - - def get_header_style(self) -> str: - """Get the CSS style for header cells. - - Returns: - CSS style string - """ - return ( - "border: 1px solid black; padding: 8px; text-align: left; " - "background-color: #f2f2f2; white-space: nowrap; min-width: fit-content; " - "max-width: fit-content;" - ) - - -class DataFrameHtmlFormatter: - """Configurable HTML formatter for DataFusion DataFrames. - - This class handles the HTML rendering of DataFrames for display in - Jupyter notebooks and other rich display contexts. - - This class supports extension through composition. Key extension points: - - Provide a custom StyleProvider for styling cells and headers - - Register custom formatters for specific types - - Provide custom cell builders for specialized cell rendering - - Args: - max_cell_length: Maximum characters to display in a cell before truncation - max_width: Maximum width of the HTML table in pixels - max_height: Maximum height of the HTML table in pixels - max_memory_bytes: Maximum memory in bytes for rendered data (default: 2MB) - min_rows_display: Minimum number of rows to display - repr_rows: Default number of rows to display in repr output - enable_cell_expansion: Whether to add expand/collapse buttons for long cell - values - custom_css: Additional CSS to include in the HTML output - show_truncation_message: Whether to display a message when data is truncated - style_provider: Custom provider for cell and header styles - use_shared_styles: Whether to load styles and scripts only once per notebook - session - """ - - # Class variable to track if styles have been loaded in the notebook - _styles_loaded = False - - def __init__( - self, - max_cell_length: int = 25, - max_width: int = 1000, - max_height: int = 300, - max_memory_bytes: int = 2 * 1024 * 1024, # 2 MB - min_rows_display: int = 20, - repr_rows: int = 10, - enable_cell_expansion: bool = True, - custom_css: Optional[str] = None, - show_truncation_message: bool = True, - style_provider: Optional[StyleProvider] = None, - use_shared_styles: bool = True, - ) -> None: - """Initialize the HTML formatter. - - Parameters - ---------- - max_cell_length : int, default 25 - Maximum length of cell content before truncation. - max_width : int, default 1000 - Maximum width of the displayed table in pixels. - max_height : int, default 300 - Maximum height of the displayed table in pixels. - max_memory_bytes : int, default 2097152 (2MB) - Maximum memory in bytes for rendered data. - min_rows_display : int, default 20 - Minimum number of rows to display. - repr_rows : int, default 10 - Default number of rows to display in repr output. - enable_cell_expansion : bool, default True - Whether to allow cells to expand when clicked. - custom_css : str, optional - Custom CSS to apply to the HTML table. - show_truncation_message : bool, default True - Whether to show a message indicating that content has been truncated. - style_provider : StyleProvider, optional - Provider of CSS styles for the HTML table. If None, DefaultStyleProvider - is used. - use_shared_styles : bool, default True - Whether to use shared styles across multiple tables. - - Raises: - ------ - ValueError - If max_cell_length, max_width, max_height, max_memory_bytes, - min_rows_display, or repr_rows is not a positive integer. - TypeError - If enable_cell_expansion, show_truncation_message, or use_shared_styles is - not a boolean, - or if custom_css is provided but is not a string, - or if style_provider is provided but does not implement the StyleProvider - protocol. - """ - # Validate numeric parameters - _validate_positive_int(max_cell_length, "max_cell_length") - _validate_positive_int(max_width, "max_width") - _validate_positive_int(max_height, "max_height") - _validate_positive_int(max_memory_bytes, "max_memory_bytes") - _validate_positive_int(min_rows_display, "min_rows_display") - _validate_positive_int(repr_rows, "repr_rows") - - # Validate boolean parameters - _validate_bool(enable_cell_expansion, "enable_cell_expansion") - _validate_bool(show_truncation_message, "show_truncation_message") - _validate_bool(use_shared_styles, "use_shared_styles") - - # Validate custom_css - if custom_css is not None and not isinstance(custom_css, str): - msg = "custom_css must be None or a string" - raise TypeError(msg) - - # Validate style_provider - if style_provider is not None and not isinstance(style_provider, StyleProvider): - msg = "style_provider must implement the StyleProvider protocol" - raise TypeError(msg) - - self.max_cell_length = max_cell_length - self.max_width = max_width - self.max_height = max_height - self.max_memory_bytes = max_memory_bytes - self.min_rows_display = min_rows_display - self.repr_rows = repr_rows - self.enable_cell_expansion = enable_cell_expansion - self.custom_css = custom_css - self.show_truncation_message = show_truncation_message - self.style_provider = style_provider or DefaultStyleProvider() - self.use_shared_styles = use_shared_styles - # Registry for custom type formatters - self._type_formatters: dict[type, CellFormatter] = {} - # Custom cell builders - self._custom_cell_builder: Optional[Callable[[Any, int, int, str], str]] = None - self._custom_header_builder: Optional[Callable[[Any], str]] = None - - def register_formatter(self, type_class: type, formatter: CellFormatter) -> None: - """Register a custom formatter for a specific data type. - - Args: - type_class: The type to register a formatter for - formatter: Function that takes a value of the given type and returns - a formatted string - """ - self._type_formatters[type_class] = formatter - - def set_custom_cell_builder( - self, builder: Callable[[Any, int, int, str], str] - ) -> None: - """Set a custom cell builder function. - - Args: - builder: Function that takes (value, row, col, table_id) and returns HTML - """ - self._custom_cell_builder = builder - - def set_custom_header_builder(self, builder: Callable[[Any], str]) -> None: - """Set a custom header builder function. - - Args: - builder: Function that takes a field and returns HTML - """ - self._custom_header_builder = builder - - @classmethod - def is_styles_loaded(cls) -> bool: - """Check if HTML styles have been loaded in the current session. - - This method is primarily intended for debugging UI rendering issues - related to style loading. - - Returns: - True if styles have been loaded, False otherwise - - Example: - >>> from datafusion.html_formatter import DataFrameHtmlFormatter - >>> DataFrameHtmlFormatter.is_styles_loaded() - False - """ - return cls._styles_loaded - - def format_html( - self, - batches: list, - schema: Any, - has_more: bool = False, - table_uuid: str | None = None, - ) -> str: - """Format record batches as HTML. - - This method is used by DataFrame's _repr_html_ implementation and can be - called directly when custom HTML rendering is needed. - - Args: - batches: List of Arrow RecordBatch objects - schema: Arrow Schema object - has_more: Whether there are more batches not shown - table_uuid: Unique ID for the table, used for JavaScript interactions - - Returns: - HTML string representation of the data - - Raises: - TypeError: If schema is invalid and no batches are provided - """ - if not batches: - return "No data to display" - - # Validate schema - if schema is None or not hasattr(schema, "__iter__"): - msg = "Schema must be provided" - raise TypeError(msg) - - # Generate a unique ID if none provided - table_uuid = table_uuid or f"df-{id(batches)}" - - # Build HTML components - html = [] - - # Only include styles and scripts if: - # 1. Not using shared styles, OR - # 2. Using shared styles but they haven't been loaded yet - include_styles = ( - not self.use_shared_styles or not DataFrameHtmlFormatter._styles_loaded - ) - - if include_styles: - html.extend(self._build_html_header()) - # If we're using shared styles, mark them as loaded - if self.use_shared_styles: - DataFrameHtmlFormatter._styles_loaded = True - - html.extend(self._build_table_container_start()) - - # Add table header and body - html.extend(self._build_table_header(schema)) - html.extend(self._build_table_body(batches, table_uuid)) - - html.append("
" + f"{field.name}
" + f"
" + "" + "" + f"{formatted_value}" + f"" + f"
" + f"
{formatted_value}
") - html.append("
") - - # Add footer (JavaScript and messages) - if include_styles and self.enable_cell_expansion: - html.append(self._get_javascript()) - - # Always add truncation message if needed (independent of styles) - if has_more and self.show_truncation_message: - html.append("
Data truncated due to size.
") - - return "\n".join(html) - - def _build_html_header(self) -> list[str]: - """Build the HTML header with CSS styles.""" - html = [] - html.append("") - return html +import warnings - def _build_table_container_start(self) -> list[str]: - """Build the opening tags for the table container.""" - html = [] - html.append( - f'
' - ) - html.append('') - return html +from datafusion.dataframe_formatter import * # noqa: F403 - def _build_table_header(self, schema: Any) -> list[str]: - """Build the HTML table header with column names.""" - html = [] - html.append("") - html.append("") - for field in schema: - if self._custom_header_builder: - html.append(self._custom_header_builder(field)) - else: - html.append( - f"" - ) - html.append("") - html.append("") - return html - - def _build_table_body(self, batches: list, table_uuid: str) -> list[str]: - """Build the HTML table body with data rows.""" - html = [] - html.append("") - - row_count = 0 - for batch in batches: - for row_idx in range(batch.num_rows): - row_count += 1 - html.append("") - - for col_idx, column in enumerate(batch.columns): - # Get the raw value from the column - raw_value = self._get_cell_value(column, row_idx) - - # Always check for type formatters first to format the value - formatted_value = self._format_cell_value(raw_value) - - # Then apply either custom cell builder or standard cell formatting - if self._custom_cell_builder: - # Pass both the raw value and formatted value to let the - # builder decide - cell_html = self._custom_cell_builder( - raw_value, row_count, col_idx, table_uuid - ) - html.append(cell_html) - else: - # Standard cell formatting with formatted value - if ( - len(str(raw_value)) > self.max_cell_length - and self.enable_cell_expansion - ): - cell_html = self._build_expandable_cell( - formatted_value, row_count, col_idx, table_uuid - ) - else: - cell_html = self._build_regular_cell(formatted_value) - html.append(cell_html) - - html.append("") - - html.append("") - return html - - def _get_cell_value(self, column: Any, row_idx: int) -> Any: - """Extract a cell value from a column. - - Args: - column: Arrow array - row_idx: Row index - - Returns: - The raw cell value - """ - try: - value = column[row_idx] - - if hasattr(value, "as_py"): - return value.as_py() - except (AttributeError, TypeError): - pass - else: - return value - - def _format_cell_value(self, value: Any) -> str: - """Format a cell value for display. - - Uses registered type formatters if available. - - Args: - value: The cell value to format - - Returns: - Formatted cell value as string - """ - # Check for custom type formatters - for type_cls, formatter in self._type_formatters.items(): - if isinstance(value, type_cls): - return formatter(value) - - # If no formatter matched, return string representation - return str(value) - - def _build_expandable_cell( - self, formatted_value: str, row_count: int, col_idx: int, table_uuid: str - ) -> str: - """Build an expandable cell for long content.""" - short_value = str(formatted_value)[: self.max_cell_length] - return ( - f"" - ) - - def _build_regular_cell(self, formatted_value: str) -> str: - """Build a regular table cell.""" - return ( - f"" - ) - - def _build_html_footer(self, has_more: bool) -> list[str]: - """Build the HTML footer with JavaScript and messages.""" - html = [] - - # Add JavaScript for interactivity only if cell expansion is enabled - # and we're not using the shared styles approach - if self.enable_cell_expansion and not self.use_shared_styles: - html.append(self._get_javascript()) - - # Add truncation message if needed - if has_more and self.show_truncation_message: - html.append("
Data truncated due to size.
") - - return html - - def _get_default_css(self) -> str: - """Get default CSS styles for the HTML table.""" - return """ - .expandable-container { - display: inline-block; - max-width: 200px; - } - .expandable { - white-space: nowrap; - overflow: hidden; - text-overflow: ellipsis; - display: block; - } - .full-text { - display: none; - white-space: normal; - } - .expand-btn { - cursor: pointer; - color: blue; - text-decoration: underline; - border: none; - background: none; - font-size: inherit; - display: block; - margin-top: 5px; - } - """ - - def _get_javascript(self) -> str: - """Get JavaScript code for interactive elements.""" - return """ - - """ - - -class FormatterManager: - """Manager class for the global DataFrame HTML formatter instance.""" - - _default_formatter: DataFrameHtmlFormatter = DataFrameHtmlFormatter() - - @classmethod - def set_formatter(cls, formatter: DataFrameHtmlFormatter) -> None: - """Set the global DataFrame HTML formatter. - - Args: - formatter: The formatter instance to use globally - """ - cls._default_formatter = formatter - _refresh_formatter_reference() - - @classmethod - def get_formatter(cls) -> DataFrameHtmlFormatter: - """Get the current global DataFrame HTML formatter. - - Returns: - The global HTML formatter instance - """ - return cls._default_formatter - - -def get_formatter() -> DataFrameHtmlFormatter: - """Get the current global DataFrame HTML formatter. - - This function is used by the DataFrame._repr_html_ implementation to access - the shared formatter instance. It can also be used directly when custom - HTML rendering is needed. - - Returns: - The global HTML formatter instance - - Example: - >>> from datafusion.html_formatter import get_formatter - >>> formatter = get_formatter() - >>> formatter.max_cell_length = 50 # Increase cell length - """ - return FormatterManager.get_formatter() - - -def set_formatter(formatter: DataFrameHtmlFormatter) -> None: - """Set the global DataFrame HTML formatter. - - Args: - formatter: The formatter instance to use globally - - Example: - >>> from datafusion.html_formatter import get_formatter, set_formatter - >>> custom_formatter = DataFrameHtmlFormatter(max_cell_length=100) - >>> set_formatter(custom_formatter) - """ - FormatterManager.set_formatter(formatter) - - -def configure_formatter(**kwargs: Any) -> None: - """Configure the global DataFrame HTML formatter. - - This function creates a new formatter with the provided configuration - and sets it as the global formatter for all DataFrames. - - Args: - **kwargs: Formatter configuration parameters like max_cell_length, - max_width, max_height, enable_cell_expansion, etc. - - Raises: - ValueError: If any invalid parameters are provided - - Example: - >>> from datafusion.html_formatter import configure_formatter - >>> configure_formatter( - ... max_cell_length=50, - ... max_height=500, - ... enable_cell_expansion=True, - ... use_shared_styles=True - ... ) - """ - # Valid parameters accepted by DataFrameHtmlFormatter - valid_params = { - "max_cell_length", - "max_width", - "max_height", - "max_memory_bytes", - "min_rows_display", - "repr_rows", - "enable_cell_expansion", - "custom_css", - "show_truncation_message", - "style_provider", - "use_shared_styles", - } - - # Check for invalid parameters - invalid_params = set(kwargs) - valid_params - if invalid_params: - msg = ( - f"Invalid formatter parameters: {', '.join(invalid_params)}. " - f"Valid parameters are: {', '.join(valid_params)}" - ) - raise ValueError(msg) - - # Create and set formatter with validated parameters - set_formatter(DataFrameHtmlFormatter(**kwargs)) - - -def reset_formatter() -> None: - """Reset the global DataFrame HTML formatter to default settings. - - This function creates a new formatter with default configuration - and sets it as the global formatter for all DataFrames. - - Example: - >>> from datafusion.html_formatter import reset_formatter - >>> reset_formatter() # Reset formatter to default settings - """ - formatter = DataFrameHtmlFormatter() - # Reset the styles_loaded flag to ensure styles will be reloaded - DataFrameHtmlFormatter._styles_loaded = False - set_formatter(formatter) - - -def reset_styles_loaded_state() -> None: - """Reset the styles loaded state to force reloading of styles. - - This can be useful when switching between notebook sessions or - when styles need to be refreshed. - - Example: - >>> from datafusion.html_formatter import reset_styles_loaded_state - >>> reset_styles_loaded_state() # Force styles to reload in next render - """ - DataFrameHtmlFormatter._styles_loaded = False - - -def _refresh_formatter_reference() -> None: - """Refresh formatter reference in any modules using it. - - This helps ensure that changes to the formatter are reflected in existing - DataFrames that might be caching the formatter reference. - """ - # This is a no-op but signals modules to refresh their reference +warnings.warn( + "The module 'html_formatter' is deprecated and will be removed in the next release." + "Please use 'dataframe_formatter' instead.", + DeprecationWarning, + stacklevel=3, +) diff --git a/python/tests/test_dataframe.py b/python/tests/test_dataframe.py index deaa30b3d..c9ae38d8e 100644 --- a/python/tests/test_dataframe.py +++ b/python/tests/test_dataframe.py @@ -37,14 +37,14 @@ from datafusion import ( functions as f, ) -from datafusion.expr import Window -from datafusion.html_formatter import ( +from datafusion.dataframe_formatter import ( DataFrameHtmlFormatter, configure_formatter, get_formatter, reset_formatter, reset_styles_loaded_state, ) +from datafusion.expr import Window from pyarrow.csv import write_csv MB = 1024 * 1024 diff --git a/src/dataframe.rs b/src/dataframe.rs index 3d68db279..c2ad4771e 100644 --- a/src/dataframe.rs +++ b/src/dataframe.rs @@ -24,6 +24,7 @@ use arrow::compute::can_cast_types; use arrow::error::ArrowError; use arrow::ffi::FFI_ArrowSchema; use arrow::ffi_stream::FFI_ArrowArrayStream; +use arrow::pyarrow::FromPyArrow; use datafusion::arrow::datatypes::Schema; use datafusion::arrow::pyarrow::{PyArrowType, ToPyArrow}; use datafusion::arrow::util::pretty; @@ -150,9 +151,9 @@ fn get_python_formatter_with_config(py: Python) -> PyResult { Ok(PythonFormatter { formatter, config }) } -/// Get the Python formatter from the datafusion.html_formatter module +/// Get the Python formatter from the datafusion.dataframe_formatter module fn import_python_formatter(py: Python) -> PyResult> { - let formatter_module = py.import("datafusion.html_formatter")?; + let formatter_module = py.import("datafusion.dataframe_formatter")?; let get_formatter = formatter_module.getattr("get_formatter")?; get_formatter.call0() } @@ -295,6 +296,46 @@ impl PyDataFrame { pub fn new(df: DataFrame) -> Self { Self { df: Arc::new(df) } } + + fn prepare_repr_string(&self, py: Python, as_html: bool) -> PyDataFusionResult { + // Get the Python formatter and config + let PythonFormatter { formatter, config } = get_python_formatter_with_config(py)?; + let (batches, has_more) = wait_for_future( + py, + collect_record_batches_to_display(self.df.as_ref().clone(), config), + )??; + if batches.is_empty() { + // This should not be reached, but do it for safety since we index into the vector below + return Ok("No data to display".to_string()); + } + + let table_uuid = uuid::Uuid::new_v4().to_string(); + + // Convert record batches to PyObject list + let py_batches = batches + .into_iter() + .map(|rb| rb.to_pyarrow(py)) + .collect::>>()?; + + let py_schema = self.schema().into_pyobject(py)?; + + let kwargs = pyo3::types::PyDict::new(py); + let py_batches_list = PyList::new(py, py_batches.as_slice())?; + kwargs.set_item("batches", py_batches_list)?; + kwargs.set_item("schema", py_schema)?; + kwargs.set_item("has_more", has_more)?; + kwargs.set_item("table_uuid", table_uuid)?; + + let method_name = match as_html { + true => "format_html", + false => "format_str", + }; + + let html_result = formatter.call_method(method_name, (), Some(&kwargs))?; + let html_str: String = html_result.extract()?; + + Ok(html_str) + } } #[pymethods] @@ -321,18 +362,27 @@ impl PyDataFrame { } fn __repr__(&self, py: Python) -> PyDataFusionResult { - // Get the Python formatter config - let PythonFormatter { - formatter: _, - config, - } = get_python_formatter_with_config(py)?; - let (batches, has_more) = wait_for_future( - py, - collect_record_batches_to_display(self.df.as_ref().clone(), config), - )??; + self.prepare_repr_string(py, false) + } + + #[staticmethod] + #[expect(unused_variables)] + fn default_str_repr<'py>( + batches: Vec>, + schema: &Bound<'py, PyAny>, + has_more: bool, + table_uuid: &str, + ) -> PyResult { + let batches = batches + .into_iter() + .map(|batch| RecordBatch::from_pyarrow_bound(&batch)) + .collect::>>()? + .into_iter() + .filter(|batch| batch.num_rows() > 0) + .collect::>(); + if batches.is_empty() { - // This should not be reached, but do it for safety since we index into the vector below - return Ok("No data to display".to_string()); + return Ok("No data to display".to_owned()); } let batches_as_displ = @@ -347,38 +397,7 @@ impl PyDataFrame { } fn _repr_html_(&self, py: Python) -> PyDataFusionResult { - // Get the Python formatter and config - let PythonFormatter { formatter, config } = get_python_formatter_with_config(py)?; - let (batches, has_more) = wait_for_future( - py, - collect_record_batches_to_display(self.df.as_ref().clone(), config), - )??; - if batches.is_empty() { - // This should not be reached, but do it for safety since we index into the vector below - return Ok("No data to display".to_string()); - } - - let table_uuid = uuid::Uuid::new_v4().to_string(); - - // Convert record batches to PyObject list - let py_batches = batches - .into_iter() - .map(|rb| rb.to_pyarrow(py)) - .collect::>>()?; - - let py_schema = self.schema().into_pyobject(py)?; - - let kwargs = pyo3::types::PyDict::new(py); - let py_batches_list = PyList::new(py, py_batches.as_slice())?; - kwargs.set_item("batches", py_batches_list)?; - kwargs.set_item("schema", py_schema)?; - kwargs.set_item("has_more", has_more)?; - kwargs.set_item("table_uuid", table_uuid)?; - - let html_result = formatter.call_method("format_html", (), Some(&kwargs))?; - let html_str: String = html_result.extract()?; - - Ok(html_str) + self.prepare_repr_string(py, true) } /// Calculate summary statistics for a DataFrame From 954563429384078a9e85c56ad553c7e3be7ac52a Mon Sep 17 00:00:00 2001 From: Tim Saucer Date: Wed, 25 Jun 2025 11:29:35 -0400 Subject: [PATCH 012/220] feat: collect once during display() in jupyter notebooks (#1167) * Only collect one time during display() in jupyter notebooks * Check for juypter notebook environment specifically * Remove approach of checking environment which could not differentiate between jupyter console and notebook * Instead of trying to detect notebook vs console, collect one time when we have any kind if ipython environment. --- src/dataframe.rs | 36 ++++++++++++++++++++++++++---------- src/utils.rs | 11 +++++++++++ 2 files changed, 37 insertions(+), 10 deletions(-) diff --git a/src/dataframe.rs b/src/dataframe.rs index c2ad4771e..ab4749e35 100644 --- a/src/dataframe.rs +++ b/src/dataframe.rs @@ -51,7 +51,7 @@ use crate::physical_plan::PyExecutionPlan; use crate::record_batch::PyRecordBatchStream; use crate::sql::logical::PyLogicalPlan; use crate::utils::{ - get_tokio_runtime, py_obj_to_scalar_value, validate_pycapsule, wait_for_future, + get_tokio_runtime, is_ipython_env, py_obj_to_scalar_value, validate_pycapsule, wait_for_future, }; use crate::{ errors::PyDataFusionResult, @@ -289,21 +289,33 @@ impl PyParquetColumnOptions { #[derive(Clone)] pub struct PyDataFrame { df: Arc, + + // In IPython environment cache batches between __repr__ and _repr_html_ calls. + batches: Option<(Vec, bool)>, } impl PyDataFrame { /// creates a new PyDataFrame pub fn new(df: DataFrame) -> Self { - Self { df: Arc::new(df) } + Self { + df: Arc::new(df), + batches: None, + } } - fn prepare_repr_string(&self, py: Python, as_html: bool) -> PyDataFusionResult { + fn prepare_repr_string(&mut self, py: Python, as_html: bool) -> PyDataFusionResult { // Get the Python formatter and config let PythonFormatter { formatter, config } = get_python_formatter_with_config(py)?; - let (batches, has_more) = wait_for_future( - py, - collect_record_batches_to_display(self.df.as_ref().clone(), config), - )??; + + let should_cache = *is_ipython_env(py) && self.batches.is_none(); + let (batches, has_more) = match self.batches.take() { + Some(b) => b, + None => wait_for_future( + py, + collect_record_batches_to_display(self.df.as_ref().clone(), config), + )??, + }; + if batches.is_empty() { // This should not be reached, but do it for safety since we index into the vector below return Ok("No data to display".to_string()); @@ -313,7 +325,7 @@ impl PyDataFrame { // Convert record batches to PyObject list let py_batches = batches - .into_iter() + .iter() .map(|rb| rb.to_pyarrow(py)) .collect::>>()?; @@ -334,6 +346,10 @@ impl PyDataFrame { let html_result = formatter.call_method(method_name, (), Some(&kwargs))?; let html_str: String = html_result.extract()?; + if should_cache { + self.batches = Some((batches, has_more)); + } + Ok(html_str) } } @@ -361,7 +377,7 @@ impl PyDataFrame { } } - fn __repr__(&self, py: Python) -> PyDataFusionResult { + fn __repr__(&mut self, py: Python) -> PyDataFusionResult { self.prepare_repr_string(py, false) } @@ -396,7 +412,7 @@ impl PyDataFrame { Ok(format!("DataFrame()\n{batches_as_displ}{additional_str}")) } - fn _repr_html_(&self, py: Python) -> PyDataFusionResult { + fn _repr_html_(&mut self, py: Python) -> PyDataFusionResult { self.prepare_repr_string(py, true) } diff --git a/src/utils.rs b/src/utils.rs index 90d654385..f4e121fd5 100644 --- a/src/utils.rs +++ b/src/utils.rs @@ -39,6 +39,17 @@ pub(crate) fn get_tokio_runtime() -> &'static TokioRuntime { RUNTIME.get_or_init(|| TokioRuntime(tokio::runtime::Runtime::new().unwrap())) } +#[inline] +pub(crate) fn is_ipython_env(py: Python) -> &'static bool { + static IS_IPYTHON_ENV: OnceLock = OnceLock::new(); + IS_IPYTHON_ENV.get_or_init(|| { + py.import("IPython") + .and_then(|ipython| ipython.call_method0("get_ipython")) + .map(|ipython| !ipython.is_none()) + .unwrap_or(false) + }) +} + /// Utility to get the Global Datafussion CTX #[inline] pub(crate) fn get_global_ctx() -> &'static SessionContext { From 9362f53150e5423581757ed56883b3ca2c95b8a2 Mon Sep 17 00:00:00 2001 From: Tim Saucer Date: Wed, 2 Jul 2025 08:08:53 -0400 Subject: [PATCH 013/220] feat: python based catalog and schema provider (#1156) * Exposing FFI to python * Exposing FFI to python * Workin progress on python catalog * Flushing out schema and catalog providers * Adding implementation of python based catalog and schema providers * Small updates after rebase * Add default in memory options for adding schema and catalogs * Add support for creating in memory catalog and schema * Update from database to schema in unit tests * xfailed label no longer applies to these unit tests * Defining abstract methods for catalog and schema providers * Working through issues between custom catalog and build in schema * Check types on schema provider to return * Add docstring * Add documentation about how to use catalog and schema providers * Re-add module to all after rebase * Minor bugfix * Clippy updates from the new rust version --------- Co-authored-by: renato2099 --- Cargo.lock | 19 + Cargo.toml | 2 + docs/source/user-guide/data-sources.rst | 56 ++ examples/datafusion-ffi-example/Cargo.lock | 1 + examples/datafusion-ffi-example/Cargo.toml | 1 + .../python/tests/_test_catalog_provider.py | 60 +++ .../src/catalog_provider.rs | 179 +++++++ examples/datafusion-ffi-example/src/lib.rs | 3 + python/datafusion/__init__.py | 1 + python/datafusion/catalog.py | 195 ++++++- python/datafusion/context.py | 24 +- python/datafusion/dataframe.py | 9 +- python/tests/test_catalog.py | 173 ++++++- python/tests/test_context.py | 40 +- python/tests/test_sql.py | 30 +- python/tests/test_substrait.py | 4 +- src/catalog.rs | 490 ++++++++++++++++-- src/common/data_type.rs | 120 ++--- src/context.rs | 61 ++- src/expr.rs | 15 +- src/expr/aggregate.rs | 2 +- src/expr/aggregate_expr.rs | 2 +- src/expr/alias.rs | 2 +- src/expr/analyze.rs | 2 +- src/expr/between.rs | 2 +- src/expr/column.rs | 2 +- src/expr/copy_to.rs | 4 +- src/expr/create_catalog.rs | 2 +- src/expr/create_catalog_schema.rs | 2 +- src/expr/create_external_table.rs | 2 +- src/expr/create_function.rs | 2 +- src/expr/create_index.rs | 2 +- src/expr/create_memory_table.rs | 2 +- src/expr/create_view.rs | 2 +- src/expr/describe_table.rs | 2 +- src/expr/distinct.rs | 5 +- src/expr/drop_catalog_schema.rs | 2 +- src/expr/drop_function.rs | 2 +- src/expr/drop_table.rs | 2 +- src/expr/drop_view.rs | 2 +- src/expr/empty_relation.rs | 2 +- src/expr/filter.rs | 2 +- src/expr/join.rs | 2 +- src/expr/like.rs | 6 +- src/expr/limit.rs | 2 +- src/expr/projection.rs | 2 +- src/expr/recursive_query.rs | 2 +- src/expr/repartition.rs | 2 +- src/expr/sort.rs | 2 +- src/expr/sort_expr.rs | 2 +- src/expr/subquery.rs | 2 +- src/expr/subquery_alias.rs | 2 +- src/expr/table_scan.rs | 2 +- src/expr/union.rs | 2 +- src/expr/unnest.rs | 2 +- src/expr/unnest_expr.rs | 2 +- src/expr/window.rs | 11 +- src/functions.rs | 2 +- src/lib.rs | 10 +- src/physical_plan.rs | 3 +- src/sql/logical.rs | 3 +- src/utils.rs | 5 +- 62 files changed, 1340 insertions(+), 258 deletions(-) create mode 100644 examples/datafusion-ffi-example/python/tests/_test_catalog_provider.py create mode 100644 examples/datafusion-ffi-example/src/catalog_provider.rs diff --git a/Cargo.lock b/Cargo.lock index 112167cb4..a3e9336cf 100644 --- a/Cargo.lock +++ b/Cargo.lock @@ -165,6 +165,12 @@ dependencies = [ "zstd", ] +[[package]] +name = "arc-swap" +version = "1.7.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "69f7f8c3906b62b754cd5326047894316021dcfe5a194c8ea52bdd94934a3457" + [[package]] name = "arrayref" version = "0.3.9" @@ -1503,6 +1509,7 @@ dependencies = [ "datafusion-proto", "datafusion-substrait", "futures", + "log", "mimalloc", "object_store", "prost", @@ -1510,6 +1517,7 @@ dependencies = [ "pyo3", "pyo3-async-runtimes", "pyo3-build-config", + "pyo3-log", "tokio", "url", "uuid", @@ -2953,6 +2961,17 @@ dependencies = [ "pyo3-build-config", ] +[[package]] +name = "pyo3-log" +version = "0.12.4" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "45192e5e4a4d2505587e27806c7b710c231c40c56f3bfc19535d0bb25df52264" +dependencies = [ + "arc-swap", + "log", + "pyo3", +] + [[package]] name = "pyo3-macros" version = "0.24.2" diff --git a/Cargo.toml b/Cargo.toml index 4135e64e2..1f7895a50 100644 --- a/Cargo.toml +++ b/Cargo.toml @@ -37,6 +37,7 @@ substrait = ["dep:datafusion-substrait"] tokio = { version = "1.45", features = ["macros", "rt", "rt-multi-thread", "sync"] } pyo3 = { version = "0.24", features = ["extension-module", "abi3", "abi3-py39"] } pyo3-async-runtimes = { version = "0.24", features = ["tokio-runtime"]} +pyo3-log = "0.12.4" arrow = { version = "55.1.0", features = ["pyarrow"] } datafusion = { version = "48.0.0", features = ["avro", "unicode_expressions"] } datafusion-substrait = { version = "48.0.0", optional = true } @@ -49,6 +50,7 @@ async-trait = "0.1.88" futures = "0.3" object_store = { version = "0.12.1", features = ["aws", "gcp", "azure", "http"] } url = "2" +log = "0.4.27" [build-dependencies] prost-types = "0.13.1" # keep in line with `datafusion-substrait` diff --git a/docs/source/user-guide/data-sources.rst b/docs/source/user-guide/data-sources.rst index ba5967c97..9c95d58e0 100644 --- a/docs/source/user-guide/data-sources.rst +++ b/docs/source/user-guide/data-sources.rst @@ -185,3 +185,59 @@ the interface as describe in the :ref:`Custom Table Provider `_ is provided in the DataFusion repository. + +Catalog +======= + +A common technique for organizing tables is using a three level hierarchical approach. DataFusion +supports this form of organizing using the :py:class:`~datafusion.catalog.Catalog`, +:py:class:`~datafusion.catalog.Schema`, and :py:class:`~datafusion.catalog.Table`. By default, +a :py:class:`~datafusion.context.SessionContext` comes with a single Catalog and a single Schema +with the names ``datafusion`` and ``default``, respectively. + +The default implementation uses an in-memory approach to the catalog and schema. We have support +for adding additional in-memory catalogs and schemas. This can be done like in the following +example: + +.. code-block:: python + + from datafusion.catalog import Catalog, Schema + + my_catalog = Catalog.memory_catalog() + my_schema = Schema.memory_schema() + + my_catalog.register_schema("my_schema_name", my_schema) + + ctx.register_catalog("my_catalog_name", my_catalog) + +You could then register tables in ``my_schema`` and access them either through the DataFrame +API or via sql commands such as ``"SELECT * from my_catalog_name.my_schema_name.my_table"``. + +User Defined Catalog and Schema +------------------------------- + +If the in-memory catalogs are insufficient for your uses, there are two approaches you can take +to implementing a custom catalog and/or schema. In the below discussion, we describe how to +implement these for a Catalog, but the approach to implementing for a Schema is nearly +identical. + +DataFusion supports Catalogs written in either Rust or Python. If you write a Catalog in Rust, +you will need to export it as a Python library via PyO3. There is a complete example of a +catalog implemented this way in the +`examples folder `_ +of our repository. Writing catalog providers in Rust provides typically can lead to significant +performance improvements over the Python based approach. + +To implement a Catalog in Python, you will need to inherit from the abstract base class +:py:class:`~datafusion.catalog.CatalogProvider`. There are examples in the +`unit tests `_ of +implementing a basic Catalog in Python where we simply keep a dictionary of the +registered Schemas. + +One important note for developers is that when we have a Catalog defined in Python, we have +two different ways of accessing this Catalog. First, we register the catalog with a Rust +wrapper. This allows for any rust based code to call the Python functions as necessary. +Second, if the user access the Catalog via the Python API, we identify this and return back +the original Python object that implements the Catalog. This is an important distinction +for developers because we do *not* return a Python wrapper around the Rust wrapper of the +original Python object. diff --git a/examples/datafusion-ffi-example/Cargo.lock b/examples/datafusion-ffi-example/Cargo.lock index 075ebd5a1..e5a1ca8d1 100644 --- a/examples/datafusion-ffi-example/Cargo.lock +++ b/examples/datafusion-ffi-example/Cargo.lock @@ -1448,6 +1448,7 @@ dependencies = [ "arrow", "arrow-array", "arrow-schema", + "async-trait", "datafusion", "datafusion-ffi", "pyo3", diff --git a/examples/datafusion-ffi-example/Cargo.toml b/examples/datafusion-ffi-example/Cargo.toml index 0e17567b9..319163554 100644 --- a/examples/datafusion-ffi-example/Cargo.toml +++ b/examples/datafusion-ffi-example/Cargo.toml @@ -27,6 +27,7 @@ pyo3 = { version = "0.23", features = ["extension-module", "abi3", "abi3-py39"] arrow = { version = "55.0.0" } arrow-array = { version = "55.0.0" } arrow-schema = { version = "55.0.0" } +async-trait = "0.1.88" [build-dependencies] pyo3-build-config = "0.23" diff --git a/examples/datafusion-ffi-example/python/tests/_test_catalog_provider.py b/examples/datafusion-ffi-example/python/tests/_test_catalog_provider.py new file mode 100644 index 000000000..72aadf64c --- /dev/null +++ b/examples/datafusion-ffi-example/python/tests/_test_catalog_provider.py @@ -0,0 +1,60 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. + +from __future__ import annotations + +import pyarrow as pa +from datafusion import SessionContext +from datafusion_ffi_example import MyCatalogProvider + + +def test_catalog_provider(): + ctx = SessionContext() + + my_catalog_name = "my_catalog" + expected_schema_name = "my_schema" + expected_table_name = "my_table" + expected_table_columns = ["units", "price"] + + catalog_provider = MyCatalogProvider() + ctx.register_catalog_provider(my_catalog_name, catalog_provider) + my_catalog = ctx.catalog(my_catalog_name) + + my_catalog_schemas = my_catalog.names() + assert expected_schema_name in my_catalog_schemas + my_database = my_catalog.database(expected_schema_name) + assert expected_table_name in my_database.names() + my_table = my_database.table(expected_table_name) + assert expected_table_columns == my_table.schema.names + + result = ctx.table( + f"{my_catalog_name}.{expected_schema_name}.{expected_table_name}" + ).collect() + assert len(result) == 2 + + col0_result = [r.column(0) for r in result] + col1_result = [r.column(1) for r in result] + expected_col0 = [ + pa.array([10, 20, 30], type=pa.int32()), + pa.array([5, 7], type=pa.int32()), + ] + expected_col1 = [ + pa.array([1, 2, 5], type=pa.float64()), + pa.array([1.5, 2.5], type=pa.float64()), + ] + assert col0_result == expected_col0 + assert col1_result == expected_col1 diff --git a/examples/datafusion-ffi-example/src/catalog_provider.rs b/examples/datafusion-ffi-example/src/catalog_provider.rs new file mode 100644 index 000000000..54e61cf3e --- /dev/null +++ b/examples/datafusion-ffi-example/src/catalog_provider.rs @@ -0,0 +1,179 @@ +// Licensed to the Apache Software Foundation (ASF) under one +// or more contributor license agreements. See the NOTICE file +// distributed with this work for additional information +// regarding copyright ownership. The ASF licenses this file +// to you under the Apache License, Version 2.0 (the +// "License"); you may not use this file except in compliance +// with the License. You may obtain a copy of the License at +// +// http://www.apache.org/licenses/LICENSE-2.0 +// +// Unless required by applicable law or agreed to in writing, +// software distributed under the License is distributed on an +// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +// KIND, either express or implied. See the License for the +// specific language governing permissions and limitations +// under the License. + +use pyo3::{pyclass, pymethods, Bound, PyResult, Python}; +use std::{any::Any, fmt::Debug, sync::Arc}; + +use arrow::datatypes::Schema; +use async_trait::async_trait; +use datafusion::{ + catalog::{ + CatalogProvider, MemoryCatalogProvider, MemorySchemaProvider, SchemaProvider, TableProvider, + }, + common::exec_err, + datasource::MemTable, + error::{DataFusionError, Result}, +}; +use datafusion_ffi::catalog_provider::FFI_CatalogProvider; +use pyo3::types::PyCapsule; + +pub fn my_table() -> Arc { + use arrow::datatypes::{DataType, Field}; + use datafusion::common::record_batch; + + let schema = Arc::new(Schema::new(vec![ + Field::new("units", DataType::Int32, true), + Field::new("price", DataType::Float64, true), + ])); + + let partitions = vec![ + record_batch!( + ("units", Int32, vec![10, 20, 30]), + ("price", Float64, vec![1.0, 2.0, 5.0]) + ) + .unwrap(), + record_batch!( + ("units", Int32, vec![5, 7]), + ("price", Float64, vec![1.5, 2.5]) + ) + .unwrap(), + ]; + + Arc::new(MemTable::try_new(schema, vec![partitions]).unwrap()) +} + +#[derive(Debug)] +pub struct FixedSchemaProvider { + inner: MemorySchemaProvider, +} + +impl Default for FixedSchemaProvider { + fn default() -> Self { + let inner = MemorySchemaProvider::new(); + + let table = my_table(); + + let _ = inner.register_table("my_table".to_string(), table).unwrap(); + + Self { inner } + } +} + +#[async_trait] +impl SchemaProvider for FixedSchemaProvider { + fn as_any(&self) -> &dyn Any { + self + } + + fn table_names(&self) -> Vec { + self.inner.table_names() + } + + async fn table(&self, name: &str) -> Result>, DataFusionError> { + self.inner.table(name).await + } + + fn register_table( + &self, + name: String, + table: Arc, + ) -> Result>> { + self.inner.register_table(name, table) + } + + fn deregister_table(&self, name: &str) -> Result>> { + self.inner.deregister_table(name) + } + + fn table_exist(&self, name: &str) -> bool { + self.inner.table_exist(name) + } +} + +/// This catalog provider is intended only for unit tests. It prepopulates with one +/// schema and only allows for schemas named after four types of fruit. +#[pyclass( + name = "MyCatalogProvider", + module = "datafusion_ffi_example", + subclass +)] +#[derive(Debug)] +pub(crate) struct MyCatalogProvider { + inner: MemoryCatalogProvider, +} + +impl Default for MyCatalogProvider { + fn default() -> Self { + let inner = MemoryCatalogProvider::new(); + + let schema_name: &str = "my_schema"; + let _ = inner.register_schema(schema_name, Arc::new(FixedSchemaProvider::default())); + + Self { inner } + } +} + +impl CatalogProvider for MyCatalogProvider { + fn as_any(&self) -> &dyn Any { + self + } + + fn schema_names(&self) -> Vec { + self.inner.schema_names() + } + + fn schema(&self, name: &str) -> Option> { + self.inner.schema(name) + } + + fn register_schema( + &self, + name: &str, + schema: Arc, + ) -> Result>> { + self.inner.register_schema(name, schema) + } + + fn deregister_schema( + &self, + name: &str, + cascade: bool, + ) -> Result>> { + self.inner.deregister_schema(name, cascade) + } +} + +#[pymethods] +impl MyCatalogProvider { + #[new] + pub fn new() -> Self { + Self { + inner: Default::default(), + } + } + + pub fn __datafusion_catalog_provider__<'py>( + &self, + py: Python<'py>, + ) -> PyResult> { + let name = cr"datafusion_catalog_provider".into(); + let catalog_provider = + FFI_CatalogProvider::new(Arc::new(MyCatalogProvider::default()), None); + + PyCapsule::new(py, catalog_provider, Some(name)) + } +} diff --git a/examples/datafusion-ffi-example/src/lib.rs b/examples/datafusion-ffi-example/src/lib.rs index ae08c3b65..3a4cf2247 100644 --- a/examples/datafusion-ffi-example/src/lib.rs +++ b/examples/datafusion-ffi-example/src/lib.rs @@ -15,10 +15,12 @@ // specific language governing permissions and limitations // under the License. +use crate::catalog_provider::MyCatalogProvider; use crate::table_function::MyTableFunction; use crate::table_provider::MyTableProvider; use pyo3::prelude::*; +pub(crate) mod catalog_provider; pub(crate) mod table_function; pub(crate) mod table_provider; @@ -26,5 +28,6 @@ pub(crate) mod table_provider; fn datafusion_ffi_example(m: &Bound<'_, PyModule>) -> PyResult<()> { m.add_class::()?; m.add_class::()?; + m.add_class::()?; Ok(()) } diff --git a/python/datafusion/__init__.py b/python/datafusion/__init__.py index fd7f4fc06..e9d2dba75 100644 --- a/python/datafusion/__init__.py +++ b/python/datafusion/__init__.py @@ -92,6 +92,7 @@ "TableFunction", "WindowFrame", "WindowUDF", + "catalog", "col", "column", "common", diff --git a/python/datafusion/catalog.py b/python/datafusion/catalog.py index 67ab3ead2..536b3a790 100644 --- a/python/datafusion/catalog.py +++ b/python/datafusion/catalog.py @@ -19,18 +19,33 @@ from __future__ import annotations -from typing import TYPE_CHECKING +from abc import ABC, abstractmethod +from typing import TYPE_CHECKING, Protocol import datafusion._internal as df_internal if TYPE_CHECKING: import pyarrow as pa +try: + from warnings import deprecated # Python 3.13+ +except ImportError: + from typing_extensions import deprecated # Python 3.12 + + +__all__ = [ + "Catalog", + "CatalogProvider", + "Schema", + "SchemaProvider", + "Table", +] + class Catalog: """DataFusion data catalog.""" - def __init__(self, catalog: df_internal.Catalog) -> None: + def __init__(self, catalog: df_internal.catalog.RawCatalog) -> None: """This constructor is not typically called by the end user.""" self.catalog = catalog @@ -38,39 +53,95 @@ def __repr__(self) -> str: """Print a string representation of the catalog.""" return self.catalog.__repr__() - def names(self) -> list[str]: - """Returns the list of databases in this catalog.""" - return self.catalog.names() + def names(self) -> set[str]: + """This is an alias for `schema_names`.""" + return self.schema_names() + + def schema_names(self) -> set[str]: + """Returns the list of schemas in this catalog.""" + return self.catalog.schema_names() + + @staticmethod + def memory_catalog() -> Catalog: + """Create an in-memory catalog provider.""" + catalog = df_internal.catalog.RawCatalog.memory_catalog() + return Catalog(catalog) - def database(self, name: str = "public") -> Database: + def schema(self, name: str = "public") -> Schema: """Returns the database with the given ``name`` from this catalog.""" - return Database(self.catalog.database(name)) + schema = self.catalog.schema(name) + + return ( + Schema(schema) + if isinstance(schema, df_internal.catalog.RawSchema) + else schema + ) + + @deprecated("Use `schema` instead.") + def database(self, name: str = "public") -> Schema: + """Returns the database with the given ``name`` from this catalog.""" + return self.schema(name) + + def register_schema(self, name, schema) -> Schema | None: + """Register a schema with this catalog.""" + if isinstance(schema, Schema): + return self.catalog.register_schema(name, schema._raw_schema) + return self.catalog.register_schema(name, schema) + + def deregister_schema(self, name: str, cascade: bool = True) -> Schema | None: + """Deregister a schema from this catalog.""" + return self.catalog.deregister_schema(name, cascade) -class Database: - """DataFusion Database.""" +class Schema: + """DataFusion Schema.""" - def __init__(self, db: df_internal.Database) -> None: + def __init__(self, schema: df_internal.catalog.RawSchema) -> None: """This constructor is not typically called by the end user.""" - self.db = db + self._raw_schema = schema def __repr__(self) -> str: - """Print a string representation of the database.""" - return self.db.__repr__() + """Print a string representation of the schema.""" + return self._raw_schema.__repr__() + + @staticmethod + def memory_schema() -> Schema: + """Create an in-memory schema provider.""" + schema = df_internal.catalog.RawSchema.memory_schema() + return Schema(schema) def names(self) -> set[str]: - """Returns the list of all tables in this database.""" - return self.db.names() + """This is an alias for `table_names`.""" + return self.table_names() + + def table_names(self) -> set[str]: + """Returns the list of all tables in this schema.""" + return self._raw_schema.table_names def table(self, name: str) -> Table: - """Return the table with the given ``name`` from this database.""" - return Table(self.db.table(name)) + """Return the table with the given ``name`` from this schema.""" + return Table(self._raw_schema.table(name)) + + def register_table(self, name, table) -> None: + """Register a table provider in this schema.""" + if isinstance(table, Table): + return self._raw_schema.register_table(name, table.table) + return self._raw_schema.register_table(name, table) + + def deregister_table(self, name: str) -> None: + """Deregister a table provider from this schema.""" + return self._raw_schema.deregister_table(name) + + +@deprecated("Use `Schema` instead.") +class Database(Schema): + """See `Schema`.""" class Table: """DataFusion table.""" - def __init__(self, table: df_internal.Table) -> None: + def __init__(self, table: df_internal.catalog.RawTable) -> None: """This constructor is not typically called by the end user.""" self.table = table @@ -78,6 +149,11 @@ def __repr__(self) -> str: """Print a string representation of the table.""" return self.table.__repr__() + @staticmethod + def from_dataset(dataset: pa.dataset.Dataset) -> Table: + """Turn a pyarrow Dataset into a Table.""" + return Table(df_internal.catalog.RawTable.from_dataset(dataset)) + @property def schema(self) -> pa.Schema: """Returns the schema associated with this table.""" @@ -87,3 +163,86 @@ def schema(self) -> pa.Schema: def kind(self) -> str: """Returns the kind of table.""" return self.table.kind + + +class CatalogProvider(ABC): + """Abstract class for defining a Python based Catalog Provider.""" + + @abstractmethod + def schema_names(self) -> set[str]: + """Set of the names of all schemas in this catalog.""" + ... + + @abstractmethod + def schema(self, name: str) -> Schema | None: + """Retrieve a specific schema from this catalog.""" + ... + + def register_schema( # noqa: B027 + self, name: str, schema: SchemaProviderExportable | SchemaProvider | Schema + ) -> None: + """Add a schema to this catalog. + + This method is optional. If your catalog provides a fixed list of schemas, you + do not need to implement this method. + """ + + def deregister_schema(self, name: str, cascade: bool) -> None: # noqa: B027 + """Remove a schema from this catalog. + + This method is optional. If your catalog provides a fixed list of schemas, you + do not need to implement this method. + + Args: + name: The name of the schema to remove. + cascade: If true, deregister the tables within the schema. + """ + + +class SchemaProvider(ABC): + """Abstract class for defining a Python based Schema Provider.""" + + def owner_name(self) -> str | None: + """Returns the owner of the schema. + + This is an optional method. The default return is None. + """ + return None + + @abstractmethod + def table_names(self) -> set[str]: + """Set of the names of all tables in this schema.""" + ... + + @abstractmethod + def table(self, name: str) -> Table | None: + """Retrieve a specific table from this schema.""" + ... + + def register_table(self, name: str, table: Table) -> None: # noqa: B027 + """Add a table from this schema. + + This method is optional. If your schema provides a fixed list of tables, you do + not need to implement this method. + """ + + def deregister_table(self, name, cascade: bool) -> None: # noqa: B027 + """Remove a table from this schema. + + This method is optional. If your schema provides a fixed list of tables, you do + not need to implement this method. + """ + + @abstractmethod + def table_exist(self, name: str) -> bool: + """Returns true if the table exists in this schema.""" + ... + + +class SchemaProviderExportable(Protocol): + """Type hint for object that has __datafusion_schema_provider__ PyCapsule. + + https://docs.rs/datafusion/latest/datafusion/catalog/trait.SchemaProvider.html + """ + + def __datafusion_schema_provider__(self) -> object: ... diff --git a/python/datafusion/context.py b/python/datafusion/context.py index 5b99b0d26..bce51d644 100644 --- a/python/datafusion/context.py +++ b/python/datafusion/context.py @@ -29,7 +29,7 @@ except ImportError: from typing_extensions import deprecated # Python 3.12 -from datafusion.catalog import Catalog, Table +from datafusion.catalog import Catalog, CatalogProvider, Table from datafusion.dataframe import DataFrame from datafusion.expr import Expr, SortExpr, sort_list_to_raw_sort_list from datafusion.record_batch import RecordBatchStream @@ -80,6 +80,15 @@ class TableProviderExportable(Protocol): def __datafusion_table_provider__(self) -> object: ... # noqa: D105 +class CatalogProviderExportable(Protocol): + """Type hint for object that has __datafusion_catalog_provider__ PyCapsule. + + https://docs.rs/datafusion/latest/datafusion/catalog/trait.CatalogProvider.html + """ + + def __datafusion_catalog_provider__(self) -> object: ... # noqa: D105 + + class SessionConfig: """Session configuration options.""" @@ -749,6 +758,19 @@ def deregister_table(self, name: str) -> None: """Remove a table from the session.""" self.ctx.deregister_table(name) + def catalog_names(self) -> set[str]: + """Returns the list of catalogs in this context.""" + return self.ctx.catalog_names() + + def register_catalog_provider( + self, name: str, provider: CatalogProviderExportable | CatalogProvider | Catalog + ) -> None: + """Register a catalog provider.""" + if isinstance(provider, Catalog): + self.ctx.register_catalog_provider(name, provider.catalog) + else: + self.ctx.register_catalog_provider(name, provider) + def register_table_provider( self, name: str, provider: TableProviderExportable ) -> None: diff --git a/python/datafusion/dataframe.py b/python/datafusion/dataframe.py index 991e6875a..61cb09438 100644 --- a/python/datafusion/dataframe.py +++ b/python/datafusion/dataframe.py @@ -760,19 +760,16 @@ def join_on( exprs = [expr.expr for expr in on_exprs] return DataFrame(self.df.join_on(right.df, exprs, how)) - def explain(self, verbose: bool = False, analyze: bool = False) -> DataFrame: - """Return a DataFrame with the explanation of its plan so far. + def explain(self, verbose: bool = False, analyze: bool = False) -> None: + """Print an explanation of the DataFrame's plan so far. If ``analyze`` is specified, runs the plan and reports metrics. Args: verbose: If ``True``, more details will be included. analyze: If ``Tru`e``, the plan will run and metrics reported. - - Returns: - DataFrame with the explanation of its plan. """ - return DataFrame(self.df.explain(verbose, analyze)) + self.df.explain(verbose, analyze) def logical_plan(self) -> LogicalPlan: """Return the unoptimized ``LogicalPlan``. diff --git a/python/tests/test_catalog.py b/python/tests/test_catalog.py index 23b328458..1f9ecbfc3 100644 --- a/python/tests/test_catalog.py +++ b/python/tests/test_catalog.py @@ -14,9 +14,13 @@ # KIND, either express or implied. See the License for the # specific language governing permissions and limitations # under the License. +from __future__ import annotations +import datafusion as dfn import pyarrow as pa +import pyarrow.dataset as ds import pytest +from datafusion import SessionContext, Table # Note we take in `database` as a variable even though we don't use @@ -27,9 +31,9 @@ def test_basic(ctx, database): ctx.catalog("non-existent") default = ctx.catalog() - assert default.names() == ["public"] + assert default.names() == {"public"} - for db in [default.database("public"), default.database()]: + for db in [default.schema("public"), default.schema()]: assert db.names() == {"csv1", "csv", "csv2"} table = db.table("csv") @@ -41,3 +45,168 @@ def test_basic(ctx, database): pa.field("float", pa.float64(), nullable=True), ] ) + + +def create_dataset() -> Table: + batch = pa.RecordBatch.from_arrays( + [pa.array([1, 2, 3]), pa.array([4, 5, 6])], + names=["a", "b"], + ) + dataset = ds.dataset([batch]) + return Table.from_dataset(dataset) + + +class CustomSchemaProvider(dfn.catalog.SchemaProvider): + def __init__(self): + self.tables = {"table1": create_dataset()} + + def table_names(self) -> set[str]: + return set(self.tables.keys()) + + def register_table(self, name: str, table: Table): + self.tables[name] = table + + def deregister_table(self, name, cascade: bool = True): + del self.tables[name] + + def table(self, name: str) -> Table | None: + return self.tables[name] + + def table_exist(self, name: str) -> bool: + return name in self.tables + + +class CustomCatalogProvider(dfn.catalog.CatalogProvider): + def __init__(self): + self.schemas = {"my_schema": CustomSchemaProvider()} + + def schema_names(self) -> set[str]: + return set(self.schemas.keys()) + + def schema(self, name: str): + return self.schemas[name] + + def register_schema(self, name: str, schema: dfn.catalog.Schema): + self.schemas[name] = schema + + def deregister_schema(self, name, cascade: bool): + del self.schemas[name] + + +def test_python_catalog_provider(ctx: SessionContext): + ctx.register_catalog_provider("my_catalog", CustomCatalogProvider()) + + # Check the default catalog provider + assert ctx.catalog("datafusion").names() == {"public"} + + my_catalog = ctx.catalog("my_catalog") + assert my_catalog.names() == {"my_schema"} + + my_catalog.register_schema("second_schema", CustomSchemaProvider()) + assert my_catalog.schema_names() == {"my_schema", "second_schema"} + + my_catalog.deregister_schema("my_schema") + assert my_catalog.schema_names() == {"second_schema"} + + +def test_in_memory_providers(ctx: SessionContext): + catalog = dfn.catalog.Catalog.memory_catalog() + ctx.register_catalog_provider("in_mem_catalog", catalog) + + assert ctx.catalog_names() == {"datafusion", "in_mem_catalog"} + + schema = dfn.catalog.Schema.memory_schema() + catalog.register_schema("in_mem_schema", schema) + + schema.register_table("my_table", create_dataset()) + + batches = ctx.sql("select * from in_mem_catalog.in_mem_schema.my_table").collect() + + assert len(batches) == 1 + assert batches[0].column(0) == pa.array([1, 2, 3]) + assert batches[0].column(1) == pa.array([4, 5, 6]) + + +def test_python_schema_provider(ctx: SessionContext): + catalog = ctx.catalog() + + catalog.deregister_schema("public") + + catalog.register_schema("test_schema1", CustomSchemaProvider()) + assert catalog.names() == {"test_schema1"} + + catalog.register_schema("test_schema2", CustomSchemaProvider()) + catalog.deregister_schema("test_schema1") + assert catalog.names() == {"test_schema2"} + + +def test_python_table_provider(ctx: SessionContext): + catalog = ctx.catalog() + + catalog.register_schema("custom_schema", CustomSchemaProvider()) + schema = catalog.schema("custom_schema") + + assert schema.table_names() == {"table1"} + + schema.deregister_table("table1") + schema.register_table("table2", create_dataset()) + assert schema.table_names() == {"table2"} + + # Use the default schema instead of our custom schema + + schema = catalog.schema() + + schema.register_table("table3", create_dataset()) + assert schema.table_names() == {"table3"} + + schema.deregister_table("table3") + schema.register_table("table4", create_dataset()) + assert schema.table_names() == {"table4"} + + +def test_in_end_to_end_python_providers(ctx: SessionContext): + """Test registering all python providers and running a query against them.""" + + all_catalog_names = [ + "datafusion", + "custom_catalog", + "in_mem_catalog", + ] + + all_schema_names = [ + "custom_schema", + "in_mem_schema", + ] + + ctx.register_catalog_provider(all_catalog_names[1], CustomCatalogProvider()) + ctx.register_catalog_provider( + all_catalog_names[2], dfn.catalog.Catalog.memory_catalog() + ) + + for catalog_name in all_catalog_names: + catalog = ctx.catalog(catalog_name) + + # Clean out previous schemas if they exist so we can start clean + for schema_name in catalog.schema_names(): + catalog.deregister_schema(schema_name, cascade=False) + + catalog.register_schema(all_schema_names[0], CustomSchemaProvider()) + catalog.register_schema(all_schema_names[1], dfn.catalog.Schema.memory_schema()) + + for schema_name in all_schema_names: + schema = catalog.schema(schema_name) + + for table_name in schema.table_names(): + schema.deregister_table(table_name) + + schema.register_table("test_table", create_dataset()) + + for catalog_name in all_catalog_names: + for schema_name in all_schema_names: + table_full_name = f"{catalog_name}.{schema_name}.test_table" + + batches = ctx.sql(f"select * from {table_full_name}").collect() + + assert len(batches) == 1 + assert batches[0].column(0) == pa.array([1, 2, 3]) + assert batches[0].column(1) == pa.array([4, 5, 6]) diff --git a/python/tests/test_context.py b/python/tests/test_context.py index 4a15ac9cf..6dbcc0d5e 100644 --- a/python/tests/test_context.py +++ b/python/tests/test_context.py @@ -57,7 +57,7 @@ def test_runtime_configs(tmp_path, path_to_str): ctx = SessionContext(config, runtime) assert ctx is not None - db = ctx.catalog("foo").database("bar") + db = ctx.catalog("foo").schema("bar") assert db is not None @@ -70,7 +70,7 @@ def test_temporary_files(tmp_path, path_to_str): ctx = SessionContext(config, runtime) assert ctx is not None - db = ctx.catalog("foo").database("bar") + db = ctx.catalog("foo").schema("bar") assert db is not None @@ -91,7 +91,7 @@ def test_create_context_with_all_valid_args(): ctx = SessionContext(config, runtime) # verify that at least some of the arguments worked - ctx.catalog("foo").database("bar") + ctx.catalog("foo").schema("bar") with pytest.raises(KeyError): ctx.catalog("datafusion") @@ -105,7 +105,7 @@ def test_register_record_batches(ctx): ctx.register_record_batches("t", [[batch]]) - assert ctx.catalog().database().names() == {"t"} + assert ctx.catalog().schema().names() == {"t"} result = ctx.sql("SELECT a+b, a-b FROM t").collect() @@ -121,7 +121,7 @@ def test_create_dataframe_registers_unique_table_name(ctx): ) df = ctx.create_dataframe([[batch]]) - tables = list(ctx.catalog().database().names()) + tables = list(ctx.catalog().schema().names()) assert df assert len(tables) == 1 @@ -141,7 +141,7 @@ def test_create_dataframe_registers_with_defined_table_name(ctx): ) df = ctx.create_dataframe([[batch]], name="tbl") - tables = list(ctx.catalog().database().names()) + tables = list(ctx.catalog().schema().names()) assert df assert len(tables) == 1 @@ -155,7 +155,7 @@ def test_from_arrow_table(ctx): # convert to DataFrame df = ctx.from_arrow(table) - tables = list(ctx.catalog().database().names()) + tables = list(ctx.catalog().schema().names()) assert df assert len(tables) == 1 @@ -200,7 +200,7 @@ def test_from_arrow_table_with_name(ctx): # convert to DataFrame with optional name df = ctx.from_arrow(table, name="tbl") - tables = list(ctx.catalog().database().names()) + tables = list(ctx.catalog().schema().names()) assert df assert tables[0] == "tbl" @@ -213,7 +213,7 @@ def test_from_arrow_table_empty(ctx): # convert to DataFrame df = ctx.from_arrow(table) - tables = list(ctx.catalog().database().names()) + tables = list(ctx.catalog().schema().names()) assert df assert len(tables) == 1 @@ -228,7 +228,7 @@ def test_from_arrow_table_empty_no_schema(ctx): # convert to DataFrame df = ctx.from_arrow(table) - tables = list(ctx.catalog().database().names()) + tables = list(ctx.catalog().schema().names()) assert df assert len(tables) == 1 @@ -246,7 +246,7 @@ def test_from_pylist(ctx): ] df = ctx.from_pylist(data) - tables = list(ctx.catalog().database().names()) + tables = list(ctx.catalog().schema().names()) assert df assert len(tables) == 1 @@ -260,7 +260,7 @@ def test_from_pydict(ctx): data = {"a": [1, 2, 3], "b": [4, 5, 6]} df = ctx.from_pydict(data) - tables = list(ctx.catalog().database().names()) + tables = list(ctx.catalog().schema().names()) assert df assert len(tables) == 1 @@ -276,7 +276,7 @@ def test_from_pandas(ctx): pandas_df = pd.DataFrame(data) df = ctx.from_pandas(pandas_df) - tables = list(ctx.catalog().database().names()) + tables = list(ctx.catalog().schema().names()) assert df assert len(tables) == 1 @@ -292,7 +292,7 @@ def test_from_polars(ctx): polars_df = pd.DataFrame(data) df = ctx.from_polars(polars_df) - tables = list(ctx.catalog().database().names()) + tables = list(ctx.catalog().schema().names()) assert df assert len(tables) == 1 @@ -303,7 +303,7 @@ def test_from_polars(ctx): def test_register_table(ctx, database): default = ctx.catalog() - public = default.database("public") + public = default.schema("public") assert public.names() == {"csv", "csv1", "csv2"} table = public.table("csv") @@ -313,7 +313,7 @@ def test_register_table(ctx, database): def test_read_table(ctx, database): default = ctx.catalog() - public = default.database("public") + public = default.schema("public") assert public.names() == {"csv", "csv1", "csv2"} table = public.table("csv") @@ -323,7 +323,7 @@ def test_read_table(ctx, database): def test_deregister_table(ctx, database): default = ctx.catalog() - public = default.database("public") + public = default.schema("public") assert public.names() == {"csv", "csv1", "csv2"} ctx.deregister_table("csv") @@ -339,7 +339,7 @@ def test_register_dataset(ctx): dataset = ds.dataset([batch]) ctx.register_dataset("t", dataset) - assert ctx.catalog().database().names() == {"t"} + assert ctx.catalog().schema().names() == {"t"} result = ctx.sql("SELECT a+b, a-b FROM t").collect() @@ -356,7 +356,7 @@ def test_dataset_filter(ctx, capfd): dataset = ds.dataset([batch]) ctx.register_dataset("t", dataset) - assert ctx.catalog().database().names() == {"t"} + assert ctx.catalog().schema().names() == {"t"} df = ctx.sql("SELECT a+b, a-b FROM t WHERE a BETWEEN 2 and 3 AND b > 5") # Make sure the filter was pushed down in Physical Plan @@ -455,7 +455,7 @@ def test_dataset_filter_nested_data(ctx): dataset = ds.dataset([batch]) ctx.register_dataset("t", dataset) - assert ctx.catalog().database().names() == {"t"} + assert ctx.catalog().schema().names() == {"t"} df = ctx.table("t") diff --git a/python/tests/test_sql.py b/python/tests/test_sql.py index 41cee4ef3..c383edc60 100644 --- a/python/tests/test_sql.py +++ b/python/tests/test_sql.py @@ -75,7 +75,7 @@ def test_register_csv(ctx, tmp_path): ) ctx.register_csv("csv3", path, schema=alternative_schema) - assert ctx.catalog().database().names() == { + assert ctx.catalog().schema().names() == { "csv", "csv1", "csv2", @@ -150,7 +150,7 @@ def test_register_parquet(ctx, tmp_path): path = helpers.write_parquet(tmp_path / "a.parquet", helpers.data()) ctx.register_parquet("t", path) ctx.register_parquet("t1", str(path)) - assert ctx.catalog().database().names() == {"t", "t1"} + assert ctx.catalog().schema().names() == {"t", "t1"} result = ctx.sql("SELECT COUNT(a) AS cnt FROM t").collect() result = pa.Table.from_batches(result) @@ -188,7 +188,7 @@ def test_register_parquet_partitioned(ctx, tmp_path, path_to_str, legacy_data_ty parquet_pruning=True, file_extension=".parquet", ) - assert ctx.catalog().database().names() == {"datapp"} + assert ctx.catalog().schema().names() == {"datapp"} result = ctx.sql("SELECT grp, COUNT(*) AS cnt FROM datapp GROUP BY grp").collect() result = pa.Table.from_batches(result) @@ -204,7 +204,7 @@ def test_register_dataset(ctx, tmp_path, path_to_str): dataset = ds.dataset(path, format="parquet") ctx.register_dataset("t", dataset) - assert ctx.catalog().database().names() == {"t"} + assert ctx.catalog().schema().names() == {"t"} result = ctx.sql("SELECT COUNT(a) AS cnt FROM t").collect() result = pa.Table.from_batches(result) @@ -251,7 +251,7 @@ def test_register_json(ctx, tmp_path): ) ctx.register_json("json3", path, schema=alternative_schema) - assert ctx.catalog().database().names() == { + assert ctx.catalog().schema().names() == { "json", "json1", "json2", @@ -308,7 +308,7 @@ def test_execute(ctx, tmp_path): path = helpers.write_parquet(tmp_path / "a.parquet", pa.array(data)) ctx.register_parquet("t", path) - assert ctx.catalog().database().names() == {"t"} + assert ctx.catalog().schema().names() == {"t"} # count result = ctx.sql("SELECT COUNT(a) AS cnt FROM t WHERE a IS NOT NULL").collect() @@ -451,18 +451,10 @@ def test_udf( id="datetime_ns", ), # Not writtable to parquet - pytest.param( - helpers.data_timedelta("s"), id="timedelta_s", marks=pytest.mark.xfail - ), - pytest.param( - helpers.data_timedelta("ms"), id="timedelta_ms", marks=pytest.mark.xfail - ), - pytest.param( - helpers.data_timedelta("us"), id="timedelta_us", marks=pytest.mark.xfail - ), - pytest.param( - helpers.data_timedelta("ns"), id="timedelta_ns", marks=pytest.mark.xfail - ), + pytest.param(helpers.data_timedelta("s"), id="timedelta_s"), + pytest.param(helpers.data_timedelta("ms"), id="timedelta_ms"), + pytest.param(helpers.data_timedelta("us"), id="timedelta_us"), + pytest.param(helpers.data_timedelta("ns"), id="timedelta_ns"), ], ) def test_simple_select(ctx, tmp_path, arr): @@ -524,7 +516,7 @@ def test_register_listing_table( schema=table.schema if pass_schema else None, file_sort_order=file_sort_order, ) - assert ctx.catalog().database().names() == {"my_table"} + assert ctx.catalog().schema().names() == {"my_table"} result = ctx.sql( "SELECT grp, COUNT(*) AS count FROM my_table GROUP BY grp" diff --git a/python/tests/test_substrait.py b/python/tests/test_substrait.py index f367a447d..43aa327d4 100644 --- a/python/tests/test_substrait.py +++ b/python/tests/test_substrait.py @@ -34,7 +34,7 @@ def test_substrait_serialization(ctx): ctx.register_record_batches("t", [[batch]]) - assert ctx.catalog().database().names() == {"t"} + assert ctx.catalog().schema().names() == {"t"} # For now just make sure the method calls blow up substrait_plan = ss.Serde.serialize_to_plan("SELECT * FROM t", ctx) @@ -59,7 +59,7 @@ def test_substrait_file_serialization(ctx, tmp_path, path_to_str): ctx.register_record_batches("t", [[batch]]) - assert ctx.catalog().database().names() == {"t"} + assert ctx.catalog().schema().names() == {"t"} path = tmp_path / "substrait_plan" path = str(path) if path_to_str else path diff --git a/src/catalog.rs b/src/catalog.rs index 83f8d08cb..17d4ec3b8 100644 --- a/src/catalog.rs +++ b/src/catalog.rs @@ -15,44 +15,54 @@ // specific language governing permissions and limitations // under the License. -use std::collections::HashSet; -use std::sync::Arc; - -use pyo3::exceptions::PyKeyError; -use pyo3::prelude::*; - -use crate::errors::{PyDataFusionError, PyDataFusionResult}; -use crate::utils::wait_for_future; +use crate::dataset::Dataset; +use crate::errors::{py_datafusion_err, to_datafusion_err, PyDataFusionError, PyDataFusionResult}; +use crate::utils::{validate_pycapsule, wait_for_future}; +use async_trait::async_trait; +use datafusion::catalog::{MemoryCatalogProvider, MemorySchemaProvider}; +use datafusion::common::DataFusionError; use datafusion::{ arrow::pyarrow::ToPyArrow, catalog::{CatalogProvider, SchemaProvider}, datasource::{TableProvider, TableType}, }; +use datafusion_ffi::schema_provider::{FFI_SchemaProvider, ForeignSchemaProvider}; +use datafusion_ffi::table_provider::{FFI_TableProvider, ForeignTableProvider}; +use pyo3::exceptions::PyKeyError; +use pyo3::prelude::*; +use pyo3::types::PyCapsule; +use pyo3::IntoPyObjectExt; +use std::any::Any; +use std::collections::HashSet; +use std::sync::Arc; -#[pyclass(name = "Catalog", module = "datafusion", subclass)] +#[pyclass(name = "RawCatalog", module = "datafusion.catalog", subclass)] +#[derive(Clone)] pub struct PyCatalog { pub catalog: Arc, } -#[pyclass(name = "Database", module = "datafusion", subclass)] -pub struct PyDatabase { - pub database: Arc, +#[pyclass(name = "RawSchema", module = "datafusion.catalog", subclass)] +#[derive(Clone)] +pub struct PySchema { + pub schema: Arc, } -#[pyclass(name = "Table", module = "datafusion", subclass)] +#[pyclass(name = "RawTable", module = "datafusion.catalog", subclass)] +#[derive(Clone)] pub struct PyTable { pub table: Arc, } -impl PyCatalog { - pub fn new(catalog: Arc) -> Self { +impl From> for PyCatalog { + fn from(catalog: Arc) -> Self { Self { catalog } } } -impl PyDatabase { - pub fn new(database: Arc) -> Self { - Self { database } +impl From> for PySchema { + fn from(schema: Arc) -> Self { + Self { schema } } } @@ -68,36 +78,109 @@ impl PyTable { #[pymethods] impl PyCatalog { - fn names(&self) -> Vec { - self.catalog.schema_names() + #[new] + fn new(catalog: PyObject) -> Self { + let catalog_provider = + Arc::new(RustWrappedPyCatalogProvider::new(catalog)) as Arc; + catalog_provider.into() + } + + #[staticmethod] + fn memory_catalog() -> Self { + let catalog_provider = + Arc::new(MemoryCatalogProvider::default()) as Arc; + catalog_provider.into() + } + + fn schema_names(&self) -> HashSet { + self.catalog.schema_names().into_iter().collect() } #[pyo3(signature = (name="public"))] - fn database(&self, name: &str) -> PyResult { - match self.catalog.schema(name) { - Some(database) => Ok(PyDatabase::new(database)), - None => Err(PyKeyError::new_err(format!( - "Database with name {name} doesn't exist." - ))), - } + fn schema(&self, name: &str) -> PyResult { + let schema = self + .catalog + .schema(name) + .ok_or(PyKeyError::new_err(format!( + "Schema with name {name} doesn't exist." + )))?; + + Python::with_gil(|py| { + match schema + .as_any() + .downcast_ref::() + { + Some(wrapped_schema) => Ok(wrapped_schema.schema_provider.clone_ref(py)), + None => PySchema::from(schema).into_py_any(py), + } + }) + } + + fn register_schema(&self, name: &str, schema_provider: Bound<'_, PyAny>) -> PyResult<()> { + let provider = if schema_provider.hasattr("__datafusion_schema_provider__")? { + let capsule = schema_provider + .getattr("__datafusion_schema_provider__")? + .call0()?; + let capsule = capsule.downcast::().map_err(py_datafusion_err)?; + validate_pycapsule(capsule, "datafusion_schema_provider")?; + + let provider = unsafe { capsule.reference::() }; + let provider: ForeignSchemaProvider = provider.into(); + Arc::new(provider) as Arc + } else { + match schema_provider.extract::() { + Ok(py_schema) => py_schema.schema, + Err(_) => Arc::new(RustWrappedPySchemaProvider::new(schema_provider.into())) + as Arc, + } + }; + + let _ = self + .catalog + .register_schema(name, provider) + .map_err(py_datafusion_err)?; + + Ok(()) + } + + fn deregister_schema(&self, name: &str, cascade: bool) -> PyResult<()> { + let _ = self + .catalog + .deregister_schema(name, cascade) + .map_err(py_datafusion_err)?; + + Ok(()) } fn __repr__(&self) -> PyResult { - Ok(format!( - "Catalog(schema_names=[{}])", - self.names().join(";") - )) + let mut names: Vec = self.schema_names().into_iter().collect(); + names.sort(); + Ok(format!("Catalog(schema_names=[{}])", names.join(", "))) } } #[pymethods] -impl PyDatabase { - fn names(&self) -> HashSet { - self.database.table_names().into_iter().collect() +impl PySchema { + #[new] + fn new(schema_provider: PyObject) -> Self { + let schema_provider = + Arc::new(RustWrappedPySchemaProvider::new(schema_provider)) as Arc; + schema_provider.into() + } + + #[staticmethod] + fn memory_schema() -> Self { + let schema_provider = Arc::new(MemorySchemaProvider::default()) as Arc; + schema_provider.into() + } + + #[getter] + fn table_names(&self) -> HashSet { + self.schema.table_names().into_iter().collect() } fn table(&self, name: &str, py: Python) -> PyDataFusionResult { - if let Some(table) = wait_for_future(py, self.database.table(name))?? { + if let Some(table) = wait_for_future(py, self.schema.table(name))?? { Ok(PyTable::new(table)) } else { Err(PyDataFusionError::Common(format!( @@ -107,14 +190,49 @@ impl PyDatabase { } fn __repr__(&self) -> PyResult { - Ok(format!( - "Database(table_names=[{}])", - Vec::from_iter(self.names()).join(";") - )) + let mut names: Vec = self.table_names().into_iter().collect(); + names.sort(); + Ok(format!("Schema(table_names=[{}])", names.join(";"))) } - // register_table - // deregister_table + fn register_table(&self, name: &str, table_provider: Bound<'_, PyAny>) -> PyResult<()> { + let provider = if table_provider.hasattr("__datafusion_table_provider__")? { + let capsule = table_provider + .getattr("__datafusion_table_provider__")? + .call0()?; + let capsule = capsule.downcast::().map_err(py_datafusion_err)?; + validate_pycapsule(capsule, "datafusion_table_provider")?; + + let provider = unsafe { capsule.reference::() }; + let provider: ForeignTableProvider = provider.into(); + Arc::new(provider) as Arc + } else { + match table_provider.extract::() { + Ok(py_table) => py_table.table, + Err(_) => { + let py = table_provider.py(); + let provider = Dataset::new(&table_provider, py)?; + Arc::new(provider) as Arc + } + } + }; + + let _ = self + .schema + .register_table(name.to_string(), provider) + .map_err(py_datafusion_err)?; + + Ok(()) + } + + fn deregister_table(&self, name: &str) -> PyResult<()> { + let _ = self + .schema + .deregister_table(name) + .map_err(py_datafusion_err)?; + + Ok(()) + } } #[pymethods] @@ -125,6 +243,14 @@ impl PyTable { self.table.schema().to_pyarrow(py) } + #[staticmethod] + fn from_dataset(py: Python<'_>, dataset: &Bound<'_, PyAny>) -> PyResult { + let ds = Arc::new(Dataset::new(dataset, py).map_err(py_datafusion_err)?) + as Arc; + + Ok(Self::new(ds)) + } + /// Get the type of this table for metadata/catalog purposes. #[getter] fn kind(&self) -> &str { @@ -145,3 +271,285 @@ impl PyTable { // fn has_exact_statistics // fn supports_filter_pushdown } + +#[derive(Debug)] +pub(crate) struct RustWrappedPySchemaProvider { + schema_provider: PyObject, + owner_name: Option, +} + +impl RustWrappedPySchemaProvider { + pub fn new(schema_provider: PyObject) -> Self { + let owner_name = Python::with_gil(|py| { + schema_provider + .bind(py) + .getattr("owner_name") + .ok() + .map(|name| name.to_string()) + }); + + Self { + schema_provider, + owner_name, + } + } + + fn table_inner(&self, name: &str) -> PyResult>> { + Python::with_gil(|py| { + let provider = self.schema_provider.bind(py); + let py_table_method = provider.getattr("table")?; + + let py_table = py_table_method.call((name,), None)?; + if py_table.is_none() { + return Ok(None); + } + + if py_table.hasattr("__datafusion_table_provider__")? { + let capsule = provider.getattr("__datafusion_table_provider__")?.call0()?; + let capsule = capsule.downcast::().map_err(py_datafusion_err)?; + validate_pycapsule(capsule, "datafusion_table_provider")?; + + let provider = unsafe { capsule.reference::() }; + let provider: ForeignTableProvider = provider.into(); + + Ok(Some(Arc::new(provider) as Arc)) + } else { + if let Ok(inner_table) = py_table.getattr("table") { + if let Ok(inner_table) = inner_table.extract::() { + return Ok(Some(inner_table.table)); + } + } + + match py_table.extract::() { + Ok(py_table) => Ok(Some(py_table.table)), + Err(_) => { + let ds = Dataset::new(&py_table, py).map_err(py_datafusion_err)?; + Ok(Some(Arc::new(ds) as Arc)) + } + } + } + }) + } +} + +#[async_trait] +impl SchemaProvider for RustWrappedPySchemaProvider { + fn owner_name(&self) -> Option<&str> { + self.owner_name.as_deref() + } + + fn as_any(&self) -> &dyn Any { + self + } + + fn table_names(&self) -> Vec { + Python::with_gil(|py| { + let provider = self.schema_provider.bind(py); + + provider + .getattr("table_names") + .and_then(|names| names.extract::>()) + .unwrap_or_else(|err| { + log::error!("Unable to get table_names: {err}"); + Vec::default() + }) + }) + } + + async fn table( + &self, + name: &str, + ) -> datafusion::common::Result>, DataFusionError> { + self.table_inner(name).map_err(to_datafusion_err) + } + + fn register_table( + &self, + name: String, + table: Arc, + ) -> datafusion::common::Result>> { + let py_table = PyTable::new(table); + Python::with_gil(|py| { + let provider = self.schema_provider.bind(py); + let _ = provider + .call_method1("register_table", (name, py_table)) + .map_err(to_datafusion_err)?; + // Since the definition of `register_table` says that an error + // will be returned if the table already exists, there is no + // case where we want to return a table provider as output. + Ok(None) + }) + } + + fn deregister_table( + &self, + name: &str, + ) -> datafusion::common::Result>> { + Python::with_gil(|py| { + let provider = self.schema_provider.bind(py); + let table = provider + .call_method1("deregister_table", (name,)) + .map_err(to_datafusion_err)?; + if table.is_none() { + return Ok(None); + } + + // If we can turn this table provider into a `Dataset`, return it. + // Otherwise, return None. + let dataset = match Dataset::new(&table, py) { + Ok(dataset) => Some(Arc::new(dataset) as Arc), + Err(_) => None, + }; + + Ok(dataset) + }) + } + + fn table_exist(&self, name: &str) -> bool { + Python::with_gil(|py| { + let provider = self.schema_provider.bind(py); + provider + .call_method1("table_exist", (name,)) + .and_then(|pyobj| pyobj.extract()) + .unwrap_or(false) + }) + } +} + +#[derive(Debug)] +pub(crate) struct RustWrappedPyCatalogProvider { + pub(crate) catalog_provider: PyObject, +} + +impl RustWrappedPyCatalogProvider { + pub fn new(catalog_provider: PyObject) -> Self { + Self { catalog_provider } + } + + fn schema_inner(&self, name: &str) -> PyResult>> { + Python::with_gil(|py| { + let provider = self.catalog_provider.bind(py); + + let py_schema = provider.call_method1("schema", (name,))?; + if py_schema.is_none() { + return Ok(None); + } + + if py_schema.hasattr("__datafusion_schema_provider__")? { + let capsule = provider + .getattr("__datafusion_schema_provider__")? + .call0()?; + let capsule = capsule.downcast::().map_err(py_datafusion_err)?; + validate_pycapsule(capsule, "datafusion_schema_provider")?; + + let provider = unsafe { capsule.reference::() }; + let provider: ForeignSchemaProvider = provider.into(); + + Ok(Some(Arc::new(provider) as Arc)) + } else { + if let Ok(inner_schema) = py_schema.getattr("schema") { + if let Ok(inner_schema) = inner_schema.extract::() { + return Ok(Some(inner_schema.schema)); + } + } + match py_schema.extract::() { + Ok(inner_schema) => Ok(Some(inner_schema.schema)), + Err(_) => { + let py_schema = RustWrappedPySchemaProvider::new(py_schema.into()); + + Ok(Some(Arc::new(py_schema) as Arc)) + } + } + } + }) + } +} + +#[async_trait] +impl CatalogProvider for RustWrappedPyCatalogProvider { + fn as_any(&self) -> &dyn Any { + self + } + + fn schema_names(&self) -> Vec { + Python::with_gil(|py| { + let provider = self.catalog_provider.bind(py); + provider + .getattr("schema_names") + .and_then(|names| names.extract::>()) + .unwrap_or_else(|err| { + log::error!("Unable to get schema_names: {err}"); + Vec::default() + }) + }) + } + + fn schema(&self, name: &str) -> Option> { + self.schema_inner(name).unwrap_or_else(|err| { + log::error!("CatalogProvider schema returned error: {err}"); + None + }) + } + + fn register_schema( + &self, + name: &str, + schema: Arc, + ) -> datafusion::common::Result>> { + // JRIGHT HERE + // let py_schema: PySchema = schema.into(); + Python::with_gil(|py| { + let py_schema = match schema + .as_any() + .downcast_ref::() + { + Some(wrapped_schema) => wrapped_schema.schema_provider.as_any(), + None => &PySchema::from(schema) + .into_py_any(py) + .map_err(to_datafusion_err)?, + }; + + let provider = self.catalog_provider.bind(py); + let schema = provider + .call_method1("register_schema", (name, py_schema)) + .map_err(to_datafusion_err)?; + if schema.is_none() { + return Ok(None); + } + + let schema = Arc::new(RustWrappedPySchemaProvider::new(schema.into())) + as Arc; + + Ok(Some(schema)) + }) + } + + fn deregister_schema( + &self, + name: &str, + cascade: bool, + ) -> datafusion::common::Result>> { + Python::with_gil(|py| { + let provider = self.catalog_provider.bind(py); + let schema = provider + .call_method1("deregister_schema", (name, cascade)) + .map_err(to_datafusion_err)?; + if schema.is_none() { + return Ok(None); + } + + let schema = Arc::new(RustWrappedPySchemaProvider::new(schema.into())) + as Arc; + + Ok(Some(schema)) + }) + } +} + +pub(crate) fn init_module(m: &Bound<'_, PyModule>) -> PyResult<()> { + m.add_class::()?; + m.add_class::()?; + m.add_class::()?; + + Ok(()) +} diff --git a/src/common/data_type.rs b/src/common/data_type.rs index f5f8a6b06..5cf9d6e9f 100644 --- a/src/common/data_type.rs +++ b/src/common/data_type.rs @@ -172,7 +172,7 @@ impl DataTypeMap { SqlType::DATE, )), DataType::Duration(_) => Err(py_datafusion_err(DataFusionError::NotImplemented( - format!("{:?}", arrow_type), + format!("{arrow_type:?}"), ))), DataType::Interval(interval_unit) => Ok(DataTypeMap::new( DataType::Interval(*interval_unit), @@ -189,7 +189,7 @@ impl DataTypeMap { SqlType::BINARY, )), DataType::FixedSizeBinary(_) => Err(py_datafusion_err( - DataFusionError::NotImplemented(format!("{:?}", arrow_type)), + DataFusionError::NotImplemented(format!("{arrow_type:?}")), )), DataType::LargeBinary => Ok(DataTypeMap::new( DataType::LargeBinary, @@ -207,23 +207,22 @@ impl DataTypeMap { SqlType::VARCHAR, )), DataType::List(_) => Err(py_datafusion_err(DataFusionError::NotImplemented(format!( - "{:?}", - arrow_type + "{arrow_type:?}" )))), DataType::FixedSizeList(_, _) => Err(py_datafusion_err( - DataFusionError::NotImplemented(format!("{:?}", arrow_type)), + DataFusionError::NotImplemented(format!("{arrow_type:?}")), )), DataType::LargeList(_) => Err(py_datafusion_err(DataFusionError::NotImplemented( - format!("{:?}", arrow_type), + format!("{arrow_type:?}"), ))), DataType::Struct(_) => Err(py_datafusion_err(DataFusionError::NotImplemented( - format!("{:?}", arrow_type), + format!("{arrow_type:?}"), ))), DataType::Union(_, _) => Err(py_datafusion_err(DataFusionError::NotImplemented( - format!("{:?}", arrow_type), + format!("{arrow_type:?}"), ))), DataType::Dictionary(_, _) => Err(py_datafusion_err(DataFusionError::NotImplemented( - format!("{:?}", arrow_type), + format!("{arrow_type:?}"), ))), DataType::Decimal128(precision, scale) => Ok(DataTypeMap::new( DataType::Decimal128(*precision, *scale), @@ -236,23 +235,22 @@ impl DataTypeMap { SqlType::DECIMAL, )), DataType::Map(_, _) => Err(py_datafusion_err(DataFusionError::NotImplemented( - format!("{:?}", arrow_type), + format!("{arrow_type:?}"), ))), DataType::RunEndEncoded(_, _) => Err(py_datafusion_err( - DataFusionError::NotImplemented(format!("{:?}", arrow_type)), + DataFusionError::NotImplemented(format!("{arrow_type:?}")), )), DataType::BinaryView => Err(py_datafusion_err(DataFusionError::NotImplemented( - format!("{:?}", arrow_type), + format!("{arrow_type:?}"), ))), DataType::Utf8View => Err(py_datafusion_err(DataFusionError::NotImplemented(format!( - "{:?}", - arrow_type + "{arrow_type:?}" )))), DataType::ListView(_) => Err(py_datafusion_err(DataFusionError::NotImplemented( - format!("{:?}", arrow_type), + format!("{arrow_type:?}"), ))), DataType::LargeListView(_) => Err(py_datafusion_err(DataFusionError::NotImplemented( - format!("{:?}", arrow_type), + format!("{arrow_type:?}"), ))), } } @@ -379,8 +377,7 @@ impl DataTypeMap { "double" => Ok(DataType::Float64), "byte_array" => Ok(DataType::Utf8), _ => Err(PyValueError::new_err(format!( - "Unable to determine Arrow Data Type from Parquet String type: {:?}", - parquet_str_type + "Unable to determine Arrow Data Type from Parquet String type: {parquet_str_type:?}" ))), }; DataTypeMap::map_from_arrow_type(&arrow_dtype?) @@ -404,12 +401,10 @@ impl DataTypeMap { pub fn py_map_from_sql_type(sql_type: &SqlType) -> PyResult { match sql_type { SqlType::ANY => Err(py_datafusion_err(DataFusionError::NotImplemented(format!( - "{:?}", - sql_type + "{sql_type:?}" )))), SqlType::ARRAY => Err(py_datafusion_err(DataFusionError::NotImplemented(format!( - "{:?}", - sql_type + "{sql_type:?}" )))), SqlType::BIGINT => Ok(DataTypeMap::new( DataType::Int64, @@ -432,11 +427,10 @@ impl DataTypeMap { SqlType::CHAR, )), SqlType::COLUMN_LIST => Err(py_datafusion_err(DataFusionError::NotImplemented( - format!("{:?}", sql_type), + format!("{sql_type:?}"), ))), SqlType::CURSOR => Err(py_datafusion_err(DataFusionError::NotImplemented(format!( - "{:?}", - sql_type + "{sql_type:?}" )))), SqlType::DATE => Ok(DataTypeMap::new( DataType::Date64, @@ -449,8 +443,7 @@ impl DataTypeMap { SqlType::DECIMAL, )), SqlType::DISTINCT => Err(py_datafusion_err(DataFusionError::NotImplemented(format!( - "{:?}", - sql_type + "{sql_type:?}" )))), SqlType::DOUBLE => Ok(DataTypeMap::new( DataType::Decimal256(1, 1), @@ -458,7 +451,7 @@ impl DataTypeMap { SqlType::DOUBLE, )), SqlType::DYNAMIC_STAR => Err(py_datafusion_err(DataFusionError::NotImplemented( - format!("{:?}", sql_type), + format!("{sql_type:?}"), ))), SqlType::FLOAT => Ok(DataTypeMap::new( DataType::Decimal128(1, 1), @@ -466,8 +459,7 @@ impl DataTypeMap { SqlType::FLOAT, )), SqlType::GEOMETRY => Err(py_datafusion_err(DataFusionError::NotImplemented(format!( - "{:?}", - sql_type + "{sql_type:?}" )))), SqlType::INTEGER => Ok(DataTypeMap::new( DataType::Int8, @@ -475,55 +467,52 @@ impl DataTypeMap { SqlType::INTEGER, )), SqlType::INTERVAL => Err(py_datafusion_err(DataFusionError::NotImplemented(format!( - "{:?}", - sql_type + "{sql_type:?}" )))), SqlType::INTERVAL_DAY => Err(py_datafusion_err(DataFusionError::NotImplemented( - format!("{:?}", sql_type), + format!("{sql_type:?}"), ))), SqlType::INTERVAL_DAY_HOUR => Err(py_datafusion_err(DataFusionError::NotImplemented( - format!("{:?}", sql_type), + format!("{sql_type:?}"), ))), SqlType::INTERVAL_DAY_MINUTE => Err(py_datafusion_err( - DataFusionError::NotImplemented(format!("{:?}", sql_type)), + DataFusionError::NotImplemented(format!("{sql_type:?}")), )), SqlType::INTERVAL_DAY_SECOND => Err(py_datafusion_err( - DataFusionError::NotImplemented(format!("{:?}", sql_type)), + DataFusionError::NotImplemented(format!("{sql_type:?}")), )), SqlType::INTERVAL_HOUR => Err(py_datafusion_err(DataFusionError::NotImplemented( - format!("{:?}", sql_type), + format!("{sql_type:?}"), ))), SqlType::INTERVAL_HOUR_MINUTE => Err(py_datafusion_err( - DataFusionError::NotImplemented(format!("{:?}", sql_type)), + DataFusionError::NotImplemented(format!("{sql_type:?}")), )), SqlType::INTERVAL_HOUR_SECOND => Err(py_datafusion_err( - DataFusionError::NotImplemented(format!("{:?}", sql_type)), + DataFusionError::NotImplemented(format!("{sql_type:?}")), )), SqlType::INTERVAL_MINUTE => Err(py_datafusion_err(DataFusionError::NotImplemented( - format!("{:?}", sql_type), + format!("{sql_type:?}"), ))), SqlType::INTERVAL_MINUTE_SECOND => Err(py_datafusion_err( - DataFusionError::NotImplemented(format!("{:?}", sql_type)), + DataFusionError::NotImplemented(format!("{sql_type:?}")), )), SqlType::INTERVAL_MONTH => Err(py_datafusion_err(DataFusionError::NotImplemented( - format!("{:?}", sql_type), + format!("{sql_type:?}"), ))), SqlType::INTERVAL_SECOND => Err(py_datafusion_err(DataFusionError::NotImplemented( - format!("{:?}", sql_type), + format!("{sql_type:?}"), ))), SqlType::INTERVAL_YEAR => Err(py_datafusion_err(DataFusionError::NotImplemented( - format!("{:?}", sql_type), + format!("{sql_type:?}"), ))), SqlType::INTERVAL_YEAR_MONTH => Err(py_datafusion_err( - DataFusionError::NotImplemented(format!("{:?}", sql_type)), + DataFusionError::NotImplemented(format!("{sql_type:?}")), )), SqlType::MAP => Err(py_datafusion_err(DataFusionError::NotImplemented(format!( - "{:?}", - sql_type + "{sql_type:?}" )))), SqlType::MULTISET => Err(py_datafusion_err(DataFusionError::NotImplemented(format!( - "{:?}", - sql_type + "{sql_type:?}" )))), SqlType::NULL => Ok(DataTypeMap::new( DataType::Null, @@ -531,20 +520,16 @@ impl DataTypeMap { SqlType::NULL, )), SqlType::OTHER => Err(py_datafusion_err(DataFusionError::NotImplemented(format!( - "{:?}", - sql_type + "{sql_type:?}" )))), SqlType::REAL => Err(py_datafusion_err(DataFusionError::NotImplemented(format!( - "{:?}", - sql_type + "{sql_type:?}" )))), SqlType::ROW => Err(py_datafusion_err(DataFusionError::NotImplemented(format!( - "{:?}", - sql_type + "{sql_type:?}" )))), SqlType::SARG => Err(py_datafusion_err(DataFusionError::NotImplemented(format!( - "{:?}", - sql_type + "{sql_type:?}" )))), SqlType::SMALLINT => Ok(DataTypeMap::new( DataType::Int16, @@ -552,25 +537,22 @@ impl DataTypeMap { SqlType::SMALLINT, )), SqlType::STRUCTURED => Err(py_datafusion_err(DataFusionError::NotImplemented( - format!("{:?}", sql_type), + format!("{sql_type:?}"), ))), SqlType::SYMBOL => Err(py_datafusion_err(DataFusionError::NotImplemented(format!( - "{:?}", - sql_type + "{sql_type:?}" )))), SqlType::TIME => Err(py_datafusion_err(DataFusionError::NotImplemented(format!( - "{:?}", - sql_type + "{sql_type:?}" )))), SqlType::TIME_WITH_LOCAL_TIME_ZONE => Err(py_datafusion_err( - DataFusionError::NotImplemented(format!("{:?}", sql_type)), + DataFusionError::NotImplemented(format!("{sql_type:?}")), )), SqlType::TIMESTAMP => Err(py_datafusion_err(DataFusionError::NotImplemented(format!( - "{:?}", - sql_type + "{sql_type:?}" )))), SqlType::TIMESTAMP_WITH_LOCAL_TIME_ZONE => Err(py_datafusion_err( - DataFusionError::NotImplemented(format!("{:?}", sql_type)), + DataFusionError::NotImplemented(format!("{sql_type:?}")), )), SqlType::TINYINT => Ok(DataTypeMap::new( DataType::Int8, @@ -578,8 +560,7 @@ impl DataTypeMap { SqlType::TINYINT, )), SqlType::UNKNOWN => Err(py_datafusion_err(DataFusionError::NotImplemented(format!( - "{:?}", - sql_type + "{sql_type:?}" )))), SqlType::VARBINARY => Ok(DataTypeMap::new( DataType::LargeBinary, @@ -682,8 +663,7 @@ impl PyDataType { "datetime64" => Ok(DataType::Date64), "object" => Ok(DataType::Utf8), _ => Err(PyValueError::new_err(format!( - "Unable to determine Arrow Data Type from Arrow String type: {:?}", - arrow_str_type + "Unable to determine Arrow Data Type from Arrow String type: {arrow_str_type:?}" ))), }; Ok(PyDataType { diff --git a/src/context.rs b/src/context.rs index 6ce1f12bc..36133a33d 100644 --- a/src/context.rs +++ b/src/context.rs @@ -31,7 +31,7 @@ use uuid::Uuid; use pyo3::exceptions::{PyKeyError, PyValueError}; use pyo3::prelude::*; -use crate::catalog::{PyCatalog, PyTable}; +use crate::catalog::{PyCatalog, PyTable, RustWrappedPyCatalogProvider}; use crate::dataframe::PyDataFrame; use crate::dataset::Dataset; use crate::errors::{py_datafusion_err, to_datafusion_err, PyDataFusionResult}; @@ -49,6 +49,7 @@ use crate::utils::{get_global_ctx, get_tokio_runtime, validate_pycapsule, wait_f use datafusion::arrow::datatypes::{DataType, Schema, SchemaRef}; use datafusion::arrow::pyarrow::PyArrowType; use datafusion::arrow::record_batch::RecordBatch; +use datafusion::catalog::CatalogProvider; use datafusion::common::TableReference; use datafusion::common::{exec_err, ScalarValue}; use datafusion::datasource::file_format::file_compression_type::FileCompressionType; @@ -69,8 +70,10 @@ use datafusion::physical_plan::SendableRecordBatchStream; use datafusion::prelude::{ AvroReadOptions, CsvReadOptions, DataFrame, NdJsonReadOptions, ParquetReadOptions, }; +use datafusion_ffi::catalog_provider::{FFI_CatalogProvider, ForeignCatalogProvider}; use datafusion_ffi::table_provider::{FFI_TableProvider, ForeignTableProvider}; use pyo3::types::{PyCapsule, PyDict, PyList, PyTuple, PyType}; +use pyo3::IntoPyObjectExt; use tokio::task::JoinHandle; /// Configuration options for a SessionContext @@ -365,7 +368,7 @@ impl PySessionContext { } else { &upstream_host }; - let url_string = format!("{}{}", scheme, derived_host); + let url_string = format!("{scheme}{derived_host}"); let url = Url::parse(&url_string).unwrap(); self.ctx.runtime_env().register_object_store(&url, store); Ok(()) @@ -614,6 +617,34 @@ impl PySessionContext { Ok(()) } + pub fn register_catalog_provider( + &mut self, + name: &str, + provider: Bound<'_, PyAny>, + ) -> PyDataFusionResult<()> { + let provider = if provider.hasattr("__datafusion_catalog_provider__")? { + let capsule = provider + .getattr("__datafusion_catalog_provider__")? + .call0()?; + let capsule = capsule.downcast::().map_err(py_datafusion_err)?; + validate_pycapsule(capsule, "datafusion_catalog_provider")?; + + let provider = unsafe { capsule.reference::() }; + let provider: ForeignCatalogProvider = provider.into(); + Arc::new(provider) as Arc + } else { + match provider.extract::() { + Ok(py_catalog) => py_catalog.catalog, + Err(_) => Arc::new(RustWrappedPyCatalogProvider::new(provider.into())) + as Arc, + } + }; + + let _ = self.ctx.register_catalog(name, provider); + + Ok(()) + } + /// Construct datafusion dataframe from Arrow Table pub fn register_table_provider( &mut self, @@ -845,14 +876,24 @@ impl PySessionContext { } #[pyo3(signature = (name="datafusion"))] - pub fn catalog(&self, name: &str) -> PyResult { - match self.ctx.catalog(name) { - Some(catalog) => Ok(PyCatalog::new(catalog)), - None => Err(PyKeyError::new_err(format!( - "Catalog with name {} doesn't exist.", - &name, - ))), - } + pub fn catalog(&self, name: &str) -> PyResult { + let catalog = self.ctx.catalog(name).ok_or(PyKeyError::new_err(format!( + "Catalog with name {name} doesn't exist." + )))?; + + Python::with_gil(|py| { + match catalog + .as_any() + .downcast_ref::() + { + Some(wrapped_schema) => Ok(wrapped_schema.catalog_provider.clone_ref(py)), + None => PyCatalog::from(catalog).into_py_any(py), + } + }) + } + + pub fn catalog_names(&self) -> HashSet { + self.ctx.catalog_names().into_iter().collect() } pub fn tables(&self) -> HashSet { diff --git a/src/expr.rs b/src/expr.rs index 6b1d01d65..f1e002367 100644 --- a/src/expr.rs +++ b/src/expr.rs @@ -171,12 +171,10 @@ impl PyExpr { Expr::Cast(value) => Ok(cast::PyCast::from(value.clone()).into_bound_py_any(py)?), Expr::TryCast(value) => Ok(cast::PyTryCast::from(value.clone()).into_bound_py_any(py)?), Expr::ScalarFunction(value) => Err(py_unsupported_variant_err(format!( - "Converting Expr::ScalarFunction to a Python object is not implemented: {:?}", - value + "Converting Expr::ScalarFunction to a Python object is not implemented: {value:?}" ))), Expr::WindowFunction(value) => Err(py_unsupported_variant_err(format!( - "Converting Expr::WindowFunction to a Python object is not implemented: {:?}", - value + "Converting Expr::WindowFunction to a Python object is not implemented: {value:?}" ))), Expr::InList(value) => Ok(in_list::PyInList::from(value.clone()).into_bound_py_any(py)?), Expr::Exists(value) => Ok(exists::PyExists::from(value.clone()).into_bound_py_any(py)?), @@ -188,8 +186,7 @@ impl PyExpr { } #[allow(deprecated)] Expr::Wildcard { qualifier, options } => Err(py_unsupported_variant_err(format!( - "Converting Expr::Wildcard to a Python object is not implemented : {:?} {:?}", - qualifier, options + "Converting Expr::Wildcard to a Python object is not implemented : {qualifier:?} {options:?}" ))), Expr::GroupingSet(value) => { Ok(grouping_set::PyGroupingSet::from(value.clone()).into_bound_py_any(py)?) @@ -198,8 +195,7 @@ impl PyExpr { Ok(placeholder::PyPlaceholder::from(value.clone()).into_bound_py_any(py)?) } Expr::OuterReferenceColumn(data_type, column) => Err(py_unsupported_variant_err(format!( - "Converting Expr::OuterReferenceColumn to a Python object is not implemented: {:?} - {:?}", - data_type, column + "Converting Expr::OuterReferenceColumn to a Python object is not implemented: {data_type:?} - {column:?}" ))), Expr::Unnest(value) => Ok(unnest_expr::PyUnnestExpr::from(value.clone()).into_bound_py_any(py)?), } @@ -755,8 +751,7 @@ impl PyExpr { Expr::Cast(Cast { expr: _, data_type }) => DataTypeMap::map_from_arrow_type(data_type), Expr::Literal(scalar_value, _) => DataTypeMap::map_from_scalar_value(scalar_value), _ => Err(py_type_err(format!( - "Non Expr::Literal encountered in types: {:?}", - expr + "Non Expr::Literal encountered in types: {expr:?}" ))), } } diff --git a/src/expr/aggregate.rs b/src/expr/aggregate.rs index a99d83d23..fd4393271 100644 --- a/src/expr/aggregate.rs +++ b/src/expr/aggregate.rs @@ -116,7 +116,7 @@ impl PyAggregate { } fn __repr__(&self) -> PyResult { - Ok(format!("Aggregate({})", self)) + Ok(format!("Aggregate({self})")) } } diff --git a/src/expr/aggregate_expr.rs b/src/expr/aggregate_expr.rs index c09f116e3..7c5d3d31f 100644 --- a/src/expr/aggregate_expr.rs +++ b/src/expr/aggregate_expr.rs @@ -75,6 +75,6 @@ impl PyAggregateFunction { /// Get a String representation of this column fn __repr__(&self) -> String { - format!("{}", self) + format!("{self}") } } diff --git a/src/expr/alias.rs b/src/expr/alias.rs index e8e03cfad..40746f200 100644 --- a/src/expr/alias.rs +++ b/src/expr/alias.rs @@ -64,6 +64,6 @@ impl PyAlias { /// Get a String representation of this column fn __repr__(&self) -> String { - format!("{}", self) + format!("{self}") } } diff --git a/src/expr/analyze.rs b/src/expr/analyze.rs index 62f93cd26..e8081e95b 100644 --- a/src/expr/analyze.rs +++ b/src/expr/analyze.rs @@ -69,7 +69,7 @@ impl PyAnalyze { } fn __repr__(&self) -> PyResult { - Ok(format!("Analyze({})", self)) + Ok(format!("Analyze({self})")) } } diff --git a/src/expr/between.rs b/src/expr/between.rs index a2cac1442..817f1baae 100644 --- a/src/expr/between.rs +++ b/src/expr/between.rs @@ -71,6 +71,6 @@ impl PyBetween { } fn __repr__(&self) -> String { - format!("{}", self) + format!("{self}") } } diff --git a/src/expr/column.rs b/src/expr/column.rs index 365dbc0d2..50f316f1c 100644 --- a/src/expr/column.rs +++ b/src/expr/column.rs @@ -45,7 +45,7 @@ impl PyColumn { /// Get the column relation fn relation(&self) -> Option { - self.col.relation.as_ref().map(|r| format!("{}", r)) + self.col.relation.as_ref().map(|r| format!("{r}")) } /// Get the fully-qualified column name diff --git a/src/expr/copy_to.rs b/src/expr/copy_to.rs index ebfcb8ebc..473dabfed 100644 --- a/src/expr/copy_to.rs +++ b/src/expr/copy_to.rs @@ -106,7 +106,7 @@ impl PyCopyTo { } fn __repr__(&self) -> PyResult { - Ok(format!("CopyTo({})", self)) + Ok(format!("CopyTo({self})")) } fn __name__(&self) -> PyResult { @@ -129,7 +129,7 @@ impl Display for PyFileType { #[pymethods] impl PyFileType { fn __repr__(&self) -> PyResult { - Ok(format!("FileType({})", self)) + Ok(format!("FileType({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/create_catalog.rs b/src/expr/create_catalog.rs index f4ea0f517..d2d2ee8f6 100644 --- a/src/expr/create_catalog.rs +++ b/src/expr/create_catalog.rs @@ -81,7 +81,7 @@ impl PyCreateCatalog { } fn __repr__(&self) -> PyResult { - Ok(format!("CreateCatalog({})", self)) + Ok(format!("CreateCatalog({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/create_catalog_schema.rs b/src/expr/create_catalog_schema.rs index 85f447e1e..e794962f5 100644 --- a/src/expr/create_catalog_schema.rs +++ b/src/expr/create_catalog_schema.rs @@ -81,7 +81,7 @@ impl PyCreateCatalogSchema { } fn __repr__(&self) -> PyResult { - Ok(format!("CreateCatalogSchema({})", self)) + Ok(format!("CreateCatalogSchema({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/create_external_table.rs b/src/expr/create_external_table.rs index 01ce7d0ca..3e35af006 100644 --- a/src/expr/create_external_table.rs +++ b/src/expr/create_external_table.rs @@ -164,7 +164,7 @@ impl PyCreateExternalTable { } fn __repr__(&self) -> PyResult { - Ok(format!("CreateExternalTable({})", self)) + Ok(format!("CreateExternalTable({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/create_function.rs b/src/expr/create_function.rs index 6f3c3f0ff..c02ceebb1 100644 --- a/src/expr/create_function.rs +++ b/src/expr/create_function.rs @@ -163,7 +163,7 @@ impl PyCreateFunction { } fn __repr__(&self) -> PyResult { - Ok(format!("CreateFunction({})", self)) + Ok(format!("CreateFunction({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/create_index.rs b/src/expr/create_index.rs index 13dadbc3f..0f4b5011a 100644 --- a/src/expr/create_index.rs +++ b/src/expr/create_index.rs @@ -110,7 +110,7 @@ impl PyCreateIndex { } fn __repr__(&self) -> PyResult { - Ok(format!("CreateIndex({})", self)) + Ok(format!("CreateIndex({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/create_memory_table.rs b/src/expr/create_memory_table.rs index 8872b2d47..37f4d3420 100644 --- a/src/expr/create_memory_table.rs +++ b/src/expr/create_memory_table.rs @@ -78,7 +78,7 @@ impl PyCreateMemoryTable { } fn __repr__(&self) -> PyResult { - Ok(format!("CreateMemoryTable({})", self)) + Ok(format!("CreateMemoryTable({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/create_view.rs b/src/expr/create_view.rs index 87bb76876..718e404d0 100644 --- a/src/expr/create_view.rs +++ b/src/expr/create_view.rs @@ -75,7 +75,7 @@ impl PyCreateView { } fn __repr__(&self) -> PyResult { - Ok(format!("CreateView({})", self)) + Ok(format!("CreateView({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/describe_table.rs b/src/expr/describe_table.rs index 5658a13f2..6c48f3c77 100644 --- a/src/expr/describe_table.rs +++ b/src/expr/describe_table.rs @@ -61,7 +61,7 @@ impl PyDescribeTable { } fn __repr__(&self) -> PyResult { - Ok(format!("DescribeTable({})", self)) + Ok(format!("DescribeTable({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/distinct.rs b/src/expr/distinct.rs index b62b776f8..889e7099d 100644 --- a/src/expr/distinct.rs +++ b/src/expr/distinct.rs @@ -48,8 +48,7 @@ impl Display for PyDistinct { Distinct::All(input) => write!( f, "Distinct ALL - \nInput: {:?}", - input, + \nInput: {input:?}", ), Distinct::On(distinct_on) => { write!( @@ -71,7 +70,7 @@ impl PyDistinct { } fn __repr__(&self) -> PyResult { - Ok(format!("Distinct({})", self)) + Ok(format!("Distinct({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/drop_catalog_schema.rs b/src/expr/drop_catalog_schema.rs index b7420a99c..b4a4c521c 100644 --- a/src/expr/drop_catalog_schema.rs +++ b/src/expr/drop_catalog_schema.rs @@ -101,7 +101,7 @@ impl PyDropCatalogSchema { } fn __repr__(&self) -> PyResult { - Ok(format!("DropCatalogSchema({})", self)) + Ok(format!("DropCatalogSchema({self})")) } } diff --git a/src/expr/drop_function.rs b/src/expr/drop_function.rs index 9fbd78fdc..fca9eb94b 100644 --- a/src/expr/drop_function.rs +++ b/src/expr/drop_function.rs @@ -76,7 +76,7 @@ impl PyDropFunction { } fn __repr__(&self) -> PyResult { - Ok(format!("DropFunction({})", self)) + Ok(format!("DropFunction({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/drop_table.rs b/src/expr/drop_table.rs index 96983c1cf..3f442539a 100644 --- a/src/expr/drop_table.rs +++ b/src/expr/drop_table.rs @@ -70,7 +70,7 @@ impl PyDropTable { } fn __repr__(&self) -> PyResult { - Ok(format!("DropTable({})", self)) + Ok(format!("DropTable({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/drop_view.rs b/src/expr/drop_view.rs index 1d1ab1e59..6196c8bb5 100644 --- a/src/expr/drop_view.rs +++ b/src/expr/drop_view.rs @@ -83,7 +83,7 @@ impl PyDropView { } fn __repr__(&self) -> PyResult { - Ok(format!("DropView({})", self)) + Ok(format!("DropView({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/empty_relation.rs b/src/expr/empty_relation.rs index a1534ac15..758213423 100644 --- a/src/expr/empty_relation.rs +++ b/src/expr/empty_relation.rs @@ -65,7 +65,7 @@ impl PyEmptyRelation { /// Get a String representation of this column fn __repr__(&self) -> String { - format!("{}", self) + format!("{self}") } fn __name__(&self) -> PyResult { diff --git a/src/expr/filter.rs b/src/expr/filter.rs index 9bdb667cd..4fcb600cd 100644 --- a/src/expr/filter.rs +++ b/src/expr/filter.rs @@ -72,7 +72,7 @@ impl PyFilter { } fn __repr__(&self) -> String { - format!("Filter({})", self) + format!("Filter({self})") } } diff --git a/src/expr/join.rs b/src/expr/join.rs index 76ec532e7..b8d1d9da7 100644 --- a/src/expr/join.rs +++ b/src/expr/join.rs @@ -177,7 +177,7 @@ impl PyJoin { } fn __repr__(&self) -> PyResult { - Ok(format!("Join({})", self)) + Ok(format!("Join({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/like.rs b/src/expr/like.rs index 2e1f060bd..f180f5d4c 100644 --- a/src/expr/like.rs +++ b/src/expr/like.rs @@ -75,7 +75,7 @@ impl PyLike { } fn __repr__(&self) -> String { - format!("Like({})", self) + format!("Like({self})") } } @@ -133,7 +133,7 @@ impl PyILike { } fn __repr__(&self) -> String { - format!("Like({})", self) + format!("Like({self})") } } @@ -191,6 +191,6 @@ impl PySimilarTo { } fn __repr__(&self) -> String { - format!("Like({})", self) + format!("Like({self})") } } diff --git a/src/expr/limit.rs b/src/expr/limit.rs index c2a33ff89..92552814e 100644 --- a/src/expr/limit.rs +++ b/src/expr/limit.rs @@ -81,7 +81,7 @@ impl PyLimit { } fn __repr__(&self) -> PyResult { - Ok(format!("Limit({})", self)) + Ok(format!("Limit({self})")) } } diff --git a/src/expr/projection.rs b/src/expr/projection.rs index dc7e5e3c1..b5a9ef34a 100644 --- a/src/expr/projection.rs +++ b/src/expr/projection.rs @@ -85,7 +85,7 @@ impl PyProjection { } fn __repr__(&self) -> PyResult { - Ok(format!("Projection({})", self)) + Ok(format!("Projection({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/recursive_query.rs b/src/expr/recursive_query.rs index 65181f7d3..2517b7417 100644 --- a/src/expr/recursive_query.rs +++ b/src/expr/recursive_query.rs @@ -89,7 +89,7 @@ impl PyRecursiveQuery { } fn __repr__(&self) -> PyResult { - Ok(format!("RecursiveQuery({})", self)) + Ok(format!("RecursiveQuery({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/repartition.rs b/src/expr/repartition.rs index 3e782d6af..48b5e7041 100644 --- a/src/expr/repartition.rs +++ b/src/expr/repartition.rs @@ -108,7 +108,7 @@ impl PyRepartition { } fn __repr__(&self) -> PyResult { - Ok(format!("Repartition({})", self)) + Ok(format!("Repartition({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/sort.rs b/src/expr/sort.rs index ed4947591..79a8aee50 100644 --- a/src/expr/sort.rs +++ b/src/expr/sort.rs @@ -87,7 +87,7 @@ impl PySort { } fn __repr__(&self) -> PyResult { - Ok(format!("Sort({})", self)) + Ok(format!("Sort({self})")) } } diff --git a/src/expr/sort_expr.rs b/src/expr/sort_expr.rs index 12f74e4d8..79e35d978 100644 --- a/src/expr/sort_expr.rs +++ b/src/expr/sort_expr.rs @@ -85,6 +85,6 @@ impl PySortExpr { } fn __repr__(&self) -> String { - format!("{}", self) + format!("{self}") } } diff --git a/src/expr/subquery.rs b/src/expr/subquery.rs index 5ebfe6927..77f56f9a9 100644 --- a/src/expr/subquery.rs +++ b/src/expr/subquery.rs @@ -62,7 +62,7 @@ impl PySubquery { } fn __repr__(&self) -> PyResult { - Ok(format!("Subquery({})", self)) + Ok(format!("Subquery({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/subquery_alias.rs b/src/expr/subquery_alias.rs index 267a4d485..3302e7f23 100644 --- a/src/expr/subquery_alias.rs +++ b/src/expr/subquery_alias.rs @@ -72,7 +72,7 @@ impl PySubqueryAlias { } fn __repr__(&self) -> PyResult { - Ok(format!("SubqueryAlias({})", self)) + Ok(format!("SubqueryAlias({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/table_scan.rs b/src/expr/table_scan.rs index 6a0d53f0f..329964687 100644 --- a/src/expr/table_scan.rs +++ b/src/expr/table_scan.rs @@ -136,7 +136,7 @@ impl PyTableScan { } fn __repr__(&self) -> PyResult { - Ok(format!("TableScan({})", self)) + Ok(format!("TableScan({self})")) } } diff --git a/src/expr/union.rs b/src/expr/union.rs index 5a08ccc13..e0b221398 100644 --- a/src/expr/union.rs +++ b/src/expr/union.rs @@ -66,7 +66,7 @@ impl PyUnion { } fn __repr__(&self) -> PyResult { - Ok(format!("Union({})", self)) + Ok(format!("Union({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/unnest.rs b/src/expr/unnest.rs index 8e70e0990..c8833347f 100644 --- a/src/expr/unnest.rs +++ b/src/expr/unnest.rs @@ -66,7 +66,7 @@ impl PyUnnest { } fn __repr__(&self) -> PyResult { - Ok(format!("Unnest({})", self)) + Ok(format!("Unnest({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/unnest_expr.rs b/src/expr/unnest_expr.rs index 2234d24b1..634186ed8 100644 --- a/src/expr/unnest_expr.rs +++ b/src/expr/unnest_expr.rs @@ -58,7 +58,7 @@ impl PyUnnestExpr { } fn __repr__(&self) -> PyResult { - Ok(format!("UnnestExpr({})", self)) + Ok(format!("UnnestExpr({self})")) } fn __name__(&self) -> PyResult { diff --git a/src/expr/window.rs b/src/expr/window.rs index 052d9eeb4..a408731c2 100644 --- a/src/expr/window.rs +++ b/src/expr/window.rs @@ -185,8 +185,7 @@ impl PyWindowFrame { "groups" => WindowFrameUnits::Groups, _ => { return Err(py_datafusion_err(DataFusionError::NotImplemented(format!( - "{:?}", - units, + "{units:?}", )))); } }; @@ -197,8 +196,7 @@ impl PyWindowFrame { WindowFrameUnits::Rows => WindowFrameBound::Preceding(ScalarValue::UInt64(None)), WindowFrameUnits::Groups => { return Err(py_datafusion_err(DataFusionError::NotImplemented(format!( - "{:?}", - units, + "{units:?}", )))); } }, @@ -210,8 +208,7 @@ impl PyWindowFrame { WindowFrameUnits::Range => WindowFrameBound::Following(ScalarValue::UInt64(None)), WindowFrameUnits::Groups => { return Err(py_datafusion_err(DataFusionError::NotImplemented(format!( - "{:?}", - units, + "{units:?}", )))); } }, @@ -236,7 +233,7 @@ impl PyWindowFrame { /// Get a String representation of this window frame fn __repr__(&self) -> String { - format!("{}", self) + format!("{self}") } } diff --git a/src/functions.rs b/src/functions.rs index b2bafcb65..b40500b8b 100644 --- a/src/functions.rs +++ b/src/functions.rs @@ -937,7 +937,7 @@ pub(crate) fn init_module(m: &Bound<'_, PyModule>) -> PyResult<()> { m.add_wrapped(wrap_pyfunction!(left))?; m.add_wrapped(wrap_pyfunction!(length))?; m.add_wrapped(wrap_pyfunction!(ln))?; - m.add_wrapped(wrap_pyfunction!(log))?; + m.add_wrapped(wrap_pyfunction!(self::log))?; m.add_wrapped(wrap_pyfunction!(log10))?; m.add_wrapped(wrap_pyfunction!(log2))?; m.add_wrapped(wrap_pyfunction!(lower))?; diff --git a/src/lib.rs b/src/lib.rs index 1293eee3c..29d3f41da 100644 --- a/src/lib.rs +++ b/src/lib.rs @@ -77,10 +77,10 @@ pub(crate) struct TokioRuntime(tokio::runtime::Runtime); /// datafusion directory. #[pymodule] fn _internal(py: Python, m: Bound<'_, PyModule>) -> PyResult<()> { + // Initialize logging + pyo3_log::init(); + // Register the python classes - m.add_class::()?; - m.add_class::()?; - m.add_class::()?; m.add_class::()?; m.add_class::()?; m.add_class::()?; @@ -98,6 +98,10 @@ fn _internal(py: Python, m: Bound<'_, PyModule>) -> PyResult<()> { m.add_class::()?; m.add_class::()?; + let catalog = PyModule::new(py, "catalog")?; + catalog::init_module(&catalog)?; + m.add_submodule(&catalog)?; + // Register `common` as a submodule. Matching `datafusion-common` https://docs.rs/datafusion-common/latest/datafusion_common/ let common = PyModule::new(py, "common")?; common::init_module(&common)?; diff --git a/src/physical_plan.rs b/src/physical_plan.rs index f0be45c6a..49db643e1 100644 --- a/src/physical_plan.rs +++ b/src/physical_plan.rs @@ -78,8 +78,7 @@ impl PyExecutionPlan { let proto_plan = datafusion_proto::protobuf::PhysicalPlanNode::decode(bytes).map_err(|e| { PyRuntimeError::new_err(format!( - "Unable to decode logical node from serialized bytes: {}", - e + "Unable to decode logical node from serialized bytes: {e}" )) })?; diff --git a/src/sql/logical.rs b/src/sql/logical.rs index 198d68bdc..97d320470 100644 --- a/src/sql/logical.rs +++ b/src/sql/logical.rs @@ -201,8 +201,7 @@ impl PyLogicalPlan { let proto_plan = datafusion_proto::protobuf::LogicalPlanNode::decode(bytes).map_err(|e| { PyRuntimeError::new_err(format!( - "Unable to decode logical node from serialized bytes: {}", - e + "Unable to decode logical node from serialized bytes: {e}" )) })?; diff --git a/src/utils.rs b/src/utils.rs index f4e121fd5..3b30de5de 100644 --- a/src/utils.rs +++ b/src/utils.rs @@ -109,8 +109,7 @@ pub(crate) fn validate_pycapsule(capsule: &Bound, name: &str) -> PyRe let capsule_name = capsule_name.unwrap().to_str()?; if capsule_name != name { return Err(PyValueError::new_err(format!( - "Expected name '{}' in PyCapsule, instead got '{}'", - name, capsule_name + "Expected name '{name}' in PyCapsule, instead got '{capsule_name}'" ))); } @@ -127,7 +126,7 @@ pub(crate) fn py_obj_to_scalar_value(py: Python, obj: PyObject) -> PyResult Date: Wed, 2 Jul 2025 09:59:02 -0400 Subject: [PATCH 014/220] feat: add FFI support for user defined functions (#1145) * Intermediate work adding ffi scalar udf * Add scalar UDF and example * Add aggregate udf via ffi * Initial commit for window ffi integration * Remove unused import --- docs/source/contributor-guide/ffi.rst | 2 +- examples/datafusion-ffi-example/Cargo.lock | 217 ++++++++++-------- examples/datafusion-ffi-example/Cargo.toml | 8 +- .../python/tests/_test_aggregate_udf.py | 77 +++++++ .../python/tests/_test_scalar_udf.py | 70 ++++++ .../python/tests/_test_window_udf.py | 89 +++++++ .../src/aggregate_udf.rs | 81 +++++++ .../src/catalog_provider.rs | 1 - examples/datafusion-ffi-example/src/lib.rs | 9 + .../datafusion-ffi-example/src/scalar_udf.rs | 91 ++++++++ .../datafusion-ffi-example/src/window_udf.rs | 81 +++++++ python/datafusion/user_defined.py | 107 ++++++++- src/functions.rs | 2 +- src/udaf.rs | 31 ++- src/udf.rs | 25 +- src/udwf.rs | 27 ++- 16 files changed, 805 insertions(+), 113 deletions(-) create mode 100644 examples/datafusion-ffi-example/python/tests/_test_aggregate_udf.py create mode 100644 examples/datafusion-ffi-example/python/tests/_test_scalar_udf.py create mode 100644 examples/datafusion-ffi-example/python/tests/_test_window_udf.py create mode 100644 examples/datafusion-ffi-example/src/aggregate_udf.rs create mode 100644 examples/datafusion-ffi-example/src/scalar_udf.rs create mode 100644 examples/datafusion-ffi-example/src/window_udf.rs diff --git a/docs/source/contributor-guide/ffi.rst b/docs/source/contributor-guide/ffi.rst index c1f9806b3..a40af1234 100644 --- a/docs/source/contributor-guide/ffi.rst +++ b/docs/source/contributor-guide/ffi.rst @@ -176,7 +176,7 @@ By convention the ``datafusion-python`` library expects a Python object that has ``TableProvider`` PyCapsule to have this capsule accessible by calling a function named ``__datafusion_table_provider__``. You can see a complete working example of how to share a ``TableProvider`` from one python library to DataFusion Python in the -`repository examples folder `_. +`repository examples folder `_. This section has been written using ``TableProvider`` as an example. It is the first extension that has been written using this approach and the most thoroughly implemented. diff --git a/examples/datafusion-ffi-example/Cargo.lock b/examples/datafusion-ffi-example/Cargo.lock index e5a1ca8d1..1b4ca6bee 100644 --- a/examples/datafusion-ffi-example/Cargo.lock +++ b/examples/datafusion-ffi-example/Cargo.lock @@ -323,6 +323,8 @@ source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "73a47aa0c771b5381de2b7f16998d351a6f4eb839f1e13d48353e17e873d969b" dependencies = [ "bitflags", + "serde", + "serde_json", ] [[package]] @@ -748,9 +750,9 @@ dependencies = [ [[package]] name = "datafusion" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "ffe060b978f74ab446be722adb8a274e052e005bf6dfd171caadc3abaad10080" +checksum = "cc6cb8c2c81eada072059983657d6c9caf3fddefc43b4a65551d243253254a96" dependencies = [ "arrow", "arrow-ipc", @@ -775,7 +777,6 @@ dependencies = [ "datafusion-functions-nested", "datafusion-functions-table", "datafusion-functions-window", - "datafusion-macros", "datafusion-optimizer", "datafusion-physical-expr", "datafusion-physical-expr-common", @@ -790,7 +791,7 @@ dependencies = [ "object_store", "parking_lot", "parquet", - "rand", + "rand 0.9.1", "regex", "sqlparser", "tempfile", @@ -803,9 +804,9 @@ dependencies = [ [[package]] name = "datafusion-catalog" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "61fe34f401bd03724a1f96d12108144f8cd495a3cdda2bf5e091822fb80b7e66" +checksum = "b7be8d1b627843af62e447396db08fe1372d882c0eb8d0ea655fd1fbc33120ee" dependencies = [ "arrow", "async-trait", @@ -829,9 +830,9 @@ dependencies = [ [[package]] name = "datafusion-catalog-listing" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "a4411b8e3bce5e0fc7521e44f201def2e2d5d1b5f176fb56e8cdc9942c890f00" +checksum = "38ab16c5ae43f65ee525fc493ceffbc41f40dee38b01f643dfcfc12959e92038" dependencies = [ "arrow", "async-trait", @@ -852,9 +853,9 @@ dependencies = [ [[package]] name = "datafusion-common" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "0734015d81c8375eb5d4869b7f7ecccc2ee8d6cb81948ef737cd0e7b743bd69c" +checksum = "d3d56b2ac9f476b93ca82e4ef5fb00769c8a3f248d12b4965af7e27635fa7e12" dependencies = [ "ahash", "arrow", @@ -876,9 +877,9 @@ dependencies = [ [[package]] name = "datafusion-common-runtime" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "5167bb1d2ccbb87c6bc36c295274d7a0519b14afcfdaf401d53cbcaa4ef4968b" +checksum = "16015071202d6133bc84d72756176467e3e46029f3ce9ad2cb788f9b1ff139b2" dependencies = [ "futures", "log", @@ -887,9 +888,9 @@ dependencies = [ [[package]] name = "datafusion-datasource" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "04e602dcdf2f50c2abf297cc2203c73531e6f48b29516af7695d338cf2a778b1" +checksum = "b77523c95c89d2a7eb99df14ed31390e04ab29b43ff793e562bdc1716b07e17b" dependencies = [ "arrow", "async-compression", @@ -912,7 +913,7 @@ dependencies = [ "log", "object_store", "parquet", - "rand", + "rand 0.9.1", "tempfile", "tokio", "tokio-util", @@ -923,9 +924,9 @@ dependencies = [ [[package]] name = "datafusion-datasource-csv" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e3bb2253952dc32296ed5b84077cb2e0257fea4be6373e1c376426e17ead4ef6" +checksum = "40d25c5e2c0ebe8434beeea997b8e88d55b3ccc0d19344293f2373f65bc524fc" dependencies = [ "arrow", "async-trait", @@ -948,9 +949,9 @@ dependencies = [ [[package]] name = "datafusion-datasource-json" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "5b8c7f47a5d2fe03bfa521ec9bafdb8a5c82de8377f60967c3663f00c8790352" +checksum = "3dc6959e1155741ab35369e1dc7673ba30fc45ed568fad34c01b7cb1daeb4d4c" dependencies = [ "arrow", "async-trait", @@ -973,9 +974,9 @@ dependencies = [ [[package]] name = "datafusion-datasource-parquet" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "27d15868ea39ed2dc266728b554f6304acd473de2142281ecfa1294bb7415923" +checksum = "b7a6afdfe358d70f4237f60eaef26ae5a1ce7cb2c469d02d5fc6c7fd5d84e58b" dependencies = [ "arrow", "async-trait", @@ -998,21 +999,21 @@ dependencies = [ "object_store", "parking_lot", "parquet", - "rand", + "rand 0.9.1", "tokio", ] [[package]] name = "datafusion-doc" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "a91f8c2c5788ef32f48ff56c68e5b545527b744822a284373ac79bba1ba47292" +checksum = "9bcd8a3e3e3d02ea642541be23d44376b5d5c37c2938cce39b3873cdf7186eea" [[package]] name = "datafusion-execution" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "06f004d100f49a3658c9da6fb0c3a9b760062d96cd4ad82ccc3b7b69a9fb2f84" +checksum = "670da1d45d045eee4c2319b8c7ea57b26cf48ab77b630aaa50b779e406da476a" dependencies = [ "arrow", "dashmap", @@ -1022,16 +1023,16 @@ dependencies = [ "log", "object_store", "parking_lot", - "rand", + "rand 0.9.1", "tempfile", "url", ] [[package]] name = "datafusion-expr" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "7a4e4ce3802609be38eeb607ee72f6fe86c3091460de9dbfae9e18db423b3964" +checksum = "b3a577f64bdb7e2cc4043cd97f8901d8c504711fde2dbcb0887645b00d7c660b" dependencies = [ "arrow", "chrono", @@ -1050,9 +1051,9 @@ dependencies = [ [[package]] name = "datafusion-expr-common" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "422ac9cf3b22bbbae8cdf8ceb33039107fde1b5492693168f13bd566b1bcc839" +checksum = "51b7916806ace3e9f41884f230f7f38ebf0e955dfbd88266da1826f29a0b9a6a" dependencies = [ "arrow", "datafusion-common", @@ -1063,9 +1064,9 @@ dependencies = [ [[package]] name = "datafusion-ffi" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "5cf3fe9ab492c56daeb7beed526690d33622d388b8870472e0b7b7f55490338c" +checksum = "980cca31de37f5dadf7ea18e4ffc2b6833611f45bed5ef9de0831d2abb50f1ef" dependencies = [ "abi_stable", "arrow", @@ -1073,7 +1074,9 @@ dependencies = [ "async-ffi", "async-trait", "datafusion", + "datafusion-functions-aggregate-common", "datafusion-proto", + "datafusion-proto-common", "futures", "log", "prost", @@ -1081,11 +1084,25 @@ dependencies = [ "tokio", ] +[[package]] +name = "datafusion-ffi-example" +version = "0.2.0" +dependencies = [ + "arrow", + "arrow-array", + "arrow-schema", + "async-trait", + "datafusion", + "datafusion-ffi", + "pyo3", + "pyo3-build-config", +] + [[package]] name = "datafusion-functions" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "2ddf0a0a2db5d2918349c978d42d80926c6aa2459cd8a3c533a84ec4bb63479e" +checksum = "7fb31c9dc73d3e0c365063f91139dc273308f8a8e124adda9898db8085d68357" dependencies = [ "arrow", "arrow-buffer", @@ -1103,7 +1120,7 @@ dependencies = [ "itertools", "log", "md-5", - "rand", + "rand 0.9.1", "regex", "sha2", "unicode-segmentation", @@ -1112,9 +1129,9 @@ dependencies = [ [[package]] name = "datafusion-functions-aggregate" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "408a05dafdc70d05a38a29005b8b15e21b0238734dab1e98483fcb58038c5aba" +checksum = "ebb72c6940697eaaba9bd1f746a697a07819de952b817e3fb841fb75331ad5d4" dependencies = [ "ahash", "arrow", @@ -1133,9 +1150,9 @@ dependencies = [ [[package]] name = "datafusion-functions-aggregate-common" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "756d21da2dd6c9bef97af1504970ff56cbf35d03fbd4ffd62827f02f4d2279d4" +checksum = "d7fdc54656659e5ecd49bf341061f4156ab230052611f4f3609612a0da259696" dependencies = [ "ahash", "arrow", @@ -1146,9 +1163,9 @@ dependencies = [ [[package]] name = "datafusion-functions-nested" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "8d8d50f6334b378930d992d801a10ac5b3e93b846b39e4a05085742572844537" +checksum = "fad94598e3374938ca43bca6b675febe557e7a14eb627d617db427d70d65118b" dependencies = [ "arrow", "arrow-ord", @@ -1167,9 +1184,9 @@ dependencies = [ [[package]] name = "datafusion-functions-table" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "cc9a97220736c8fff1446e936be90d57216c06f28969f9ffd3b72ac93c958c8a" +checksum = "de2fc6c2946da5cab8364fb28b5cac3115f0f3a87960b235ed031c3f7e2e639b" dependencies = [ "arrow", "async-trait", @@ -1183,10 +1200,11 @@ dependencies = [ [[package]] name = "datafusion-functions-window" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "cefc2d77646e1aadd1d6a9c40088937aedec04e68c5f0465939912e1291f8193" +checksum = "3e5746548a8544870a119f556543adcd88fe0ba6b93723fe78ad0439e0fbb8b4" dependencies = [ + "arrow", "datafusion-common", "datafusion-doc", "datafusion-expr", @@ -1200,9 +1218,9 @@ dependencies = [ [[package]] name = "datafusion-functions-window-common" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "dd4aff082c42fa6da99ce0698c85addd5252928c908eb087ca3cfa64ff16b313" +checksum = "dcbe9404382cda257c434f22e13577bee7047031dfdb6216dd5e841b9465e6fe" dependencies = [ "datafusion-common", "datafusion-physical-expr-common", @@ -1210,9 +1228,9 @@ dependencies = [ [[package]] name = "datafusion-macros" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "df6f88d7ee27daf8b108ba910f9015176b36fbc72902b1ca5c2a5f1d1717e1a1" +checksum = "8dce50e3b637dab0d25d04d2fe79dfdca2b257eabd76790bffd22c7f90d700c8" dependencies = [ "datafusion-expr", "quote", @@ -1221,9 +1239,9 @@ dependencies = [ [[package]] name = "datafusion-optimizer" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "084d9f979c4b155346d3c34b18f4256e6904ded508e9554d90fed416415c3515" +checksum = "03cfaacf06445dc3bbc1e901242d2a44f2cae99a744f49f3fefddcee46240058" dependencies = [ "arrow", "chrono", @@ -1240,9 +1258,9 @@ dependencies = [ [[package]] name = "datafusion-physical-expr" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "64c536062b0076f4e30084065d805f389f9fe38af0ca75bcbac86bc5e9fbab65" +checksum = "1908034a89d7b2630898e06863583ae4c00a0dd310c1589ca284195ee3f7f8a6" dependencies = [ "ahash", "arrow", @@ -1262,9 +1280,9 @@ dependencies = [ [[package]] name = "datafusion-physical-expr-common" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "f8a92b53b3193fac1916a1c5b8e3f4347c526f6822e56b71faa5fb372327a863" +checksum = "47b7a12dd59ea07614b67dbb01d85254fbd93df45bcffa63495e11d3bdf847df" dependencies = [ "ahash", "arrow", @@ -1276,9 +1294,9 @@ dependencies = [ [[package]] name = "datafusion-physical-optimizer" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "6fa0a5ac94c7cf3da97bedabd69d6bbca12aef84b9b37e6e9e8c25286511b5e2" +checksum = "4371cc4ad33978cc2a8be93bd54a232d3f2857b50401a14631c0705f3f910aae" dependencies = [ "arrow", "datafusion-common", @@ -1295,9 +1313,9 @@ dependencies = [ [[package]] name = "datafusion-physical-plan" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "690c615db468c2e5fe5085b232d8b1c088299a6c63d87fd960a354a71f7acb55" +checksum = "dc47bc33025757a5c11f2cd094c5b6b5ed87f46fa33c023e6fdfa25fcbfade23" dependencies = [ "ahash", "arrow", @@ -1325,9 +1343,9 @@ dependencies = [ [[package]] name = "datafusion-proto" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "a4a1afb2bdb05de7ff65be6883ebfd4ec027bd9f1f21c46aa3afd01927160a83" +checksum = "d8f5d9acd7d96e3bf2a7bb04818373cab6e51de0356e3694b94905fee7b4e8b6" dependencies = [ "arrow", "chrono", @@ -1341,9 +1359,9 @@ dependencies = [ [[package]] name = "datafusion-proto-common" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "35b7a5876ebd6b564fb9a1fd2c3a2a9686b787071a256b47e4708f0916f9e46f" +checksum = "09ecb5ec152c4353b60f7a5635489834391f7a291d2b39a4820cd469e318b78e" dependencies = [ "arrow", "datafusion-common", @@ -1352,9 +1370,9 @@ dependencies = [ [[package]] name = "datafusion-session" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "ad229a134c7406c057ece00c8743c0c34b97f4e72f78b475fe17b66c5e14fa4f" +checksum = "d7485da32283985d6b45bd7d13a65169dcbe8c869e25d01b2cfbc425254b4b49" dependencies = [ "arrow", "async-trait", @@ -1376,9 +1394,9 @@ dependencies = [ [[package]] name = "datafusion-sql" -version = "47.0.0" +version = "48.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "64f6ab28b72b664c21a27b22a2ff815fd390ed224c26e89a93b5a8154a4e8607" +checksum = "a466b15632befddfeac68c125f0260f569ff315c6831538cbb40db754134e0df" dependencies = [ "arrow", "bigdecimal", @@ -1441,20 +1459,6 @@ version = "2.3.0" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "37909eebbb50d72f9059c3b6d82c0463f2ff062c9e95845c43a6c9c0355411be" -[[package]] -name = "ffi-table-provider" -version = "0.1.0" -dependencies = [ - "arrow", - "arrow-array", - "arrow-schema", - "async-trait", - "datafusion", - "datafusion-ffi", - "pyo3", - "pyo3-build-config", -] - [[package]] name = "fixedbitset" version = "0.5.7" @@ -1488,6 +1492,12 @@ version = "1.0.7" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "3f9eec918d3f24069decb9af1554cad7c880e2da24a9afd88aca000531ab82c1" +[[package]] +name = "foldhash" +version = "0.1.5" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "d9c4f5dac5e15c24eb999c26181a6ca40b39fe946cbe4c263c7209467bc83af2" + [[package]] name = "form_urlencoded" version = "1.2.1" @@ -1666,6 +1676,11 @@ name = "hashbrown" version = "0.15.3" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "84b26c544d002229e640969970a2e74021aadf6e2f96372b9c58eff97de08eb3" +dependencies = [ + "allocator-api2", + "equivalent", + "foldhash", +] [[package]] name = "heck" @@ -2271,12 +2286,14 @@ checksum = "e3148f5046208a5d56bcfc03053e3ca6334e51da8dfb19b6cdc8b306fae3283e" [[package]] name = "petgraph" -version = "0.7.1" +version = "0.8.2" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "3672b37090dbd86368a4145bc067582552b29c27377cad4e0a306c97f9bd7772" +checksum = "54acf3a685220b533e437e264e4d932cfbdc4cc7ec0cd232ed73c08d03b8a7ca" dependencies = [ "fixedbitset", + "hashbrown 0.15.3", "indexmap", + "serde", ] [[package]] @@ -2305,7 +2322,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "3c80231409c20246a13fddb31776fb942c38553c51e871f8cbd687a4cfb5843d" dependencies = [ "phf_shared", - "rand", + "rand 0.8.5", ] [[package]] @@ -2484,19 +2501,27 @@ version = "0.8.5" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "34af8d1a0e25924bc5b7c43c079c942339d8f0a8b57c39049bef581b46327404" dependencies = [ - "libc", + "rand_core 0.6.4", +] + +[[package]] +name = "rand" +version = "0.9.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "9fbfd9d094a40bf3ae768db9361049ace4c0e04a4fd6b359518bd7b73a73dd97" +dependencies = [ "rand_chacha", - "rand_core", + "rand_core 0.9.3", ] [[package]] name = "rand_chacha" -version = "0.3.1" +version = "0.9.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "e6c10a63a0fa32252be49d21e7709d4d4baf8d231c2dbce1eaa8141b9b127d88" +checksum = "d3022b5f1df60f26e1ffddd6c66e8aa15de382ae63b3a0c1bfc0e4d3e3f325cb" dependencies = [ "ppv-lite86", - "rand_core", + "rand_core 0.9.3", ] [[package]] @@ -2504,8 +2529,14 @@ name = "rand_core" version = "0.6.4" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "ec0be4795e2f6a28069bec0b5ff3e2ac9bafc99e6a9a7dc3547996c5c816922c" + +[[package]] +name = "rand_core" +version = "0.9.3" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "99d9a13982dcf210057a8a78572b2217b667c3beacbf3a0d8b454f6f82837d38" dependencies = [ - "getrandom 0.2.16", + "getrandom 0.3.3", ] [[package]] @@ -3032,9 +3063,9 @@ checksum = "b6c140620e7ffbb22c2dee59cafe6084a59b5ffc27a8859a5f0d494b5d52b6be" [[package]] name = "uuid" -version = "1.16.0" +version = "1.17.0" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "458f7a779bf54acc9f347480ac654f68407d3aab21269a6e3c9f922acd9e2da9" +checksum = "3cf4199d1e5d15ddd86a694e4d0dffa9c323ce759fea589f00fef9d81cc1931d" dependencies = [ "getrandom 0.3.3", "js-sys", diff --git a/examples/datafusion-ffi-example/Cargo.toml b/examples/datafusion-ffi-example/Cargo.toml index 319163554..b26ab48e3 100644 --- a/examples/datafusion-ffi-example/Cargo.toml +++ b/examples/datafusion-ffi-example/Cargo.toml @@ -16,13 +16,13 @@ # under the License. [package] -name = "ffi-table-provider" -version = "0.1.0" +name = "datafusion-ffi-example" +version = "0.2.0" edition = "2021" [dependencies] -datafusion = { version = "47.0.0" } -datafusion-ffi = { version = "47.0.0" } +datafusion = { version = "48.0.0" } +datafusion-ffi = { version = "48.0.0" } pyo3 = { version = "0.23", features = ["extension-module", "abi3", "abi3-py39"] } arrow = { version = "55.0.0" } arrow-array = { version = "55.0.0" } diff --git a/examples/datafusion-ffi-example/python/tests/_test_aggregate_udf.py b/examples/datafusion-ffi-example/python/tests/_test_aggregate_udf.py new file mode 100644 index 000000000..7ea6b295c --- /dev/null +++ b/examples/datafusion-ffi-example/python/tests/_test_aggregate_udf.py @@ -0,0 +1,77 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. + +from __future__ import annotations + +import pyarrow as pa +from datafusion import SessionContext, col, udaf +from datafusion_ffi_example import MySumUDF + + +def setup_context_with_table(): + ctx = SessionContext() + + # Pick numbers here so we get the same value in both groups + # since we cannot be certain of the output order of batches + batch = pa.RecordBatch.from_arrays( + [ + pa.array([1, 2, 3, None], type=pa.int64()), + pa.array([1, 1, 2, 2], type=pa.int64()), + ], + names=["a", "b"], + ) + ctx.register_record_batches("test_table", [[batch]]) + return ctx + + +def test_ffi_aggregate_register(): + ctx = setup_context_with_table() + my_udaf = udaf(MySumUDF()) + ctx.register_udaf(my_udaf) + + result = ctx.sql("select my_custom_sum(a) from test_table group by b").collect() + + assert len(result) == 2 + assert result[0].num_columns == 1 + + result = [r.column(0) for r in result] + expected = [ + pa.array([3], type=pa.int64()), + pa.array([3], type=pa.int64()), + ] + + assert result == expected + + +def test_ffi_aggregate_call_directly(): + ctx = setup_context_with_table() + my_udaf = udaf(MySumUDF()) + + result = ( + ctx.table("test_table").aggregate([col("b")], [my_udaf(col("a"))]).collect() + ) + + assert len(result) == 2 + assert result[0].num_columns == 2 + + result = [r.column(1) for r in result] + expected = [ + pa.array([3], type=pa.int64()), + pa.array([3], type=pa.int64()), + ] + + assert result == expected diff --git a/examples/datafusion-ffi-example/python/tests/_test_scalar_udf.py b/examples/datafusion-ffi-example/python/tests/_test_scalar_udf.py new file mode 100644 index 000000000..0c949c34a --- /dev/null +++ b/examples/datafusion-ffi-example/python/tests/_test_scalar_udf.py @@ -0,0 +1,70 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. + +from __future__ import annotations + +import pyarrow as pa +from datafusion import SessionContext, col, udf +from datafusion_ffi_example import IsNullUDF + + +def setup_context_with_table(): + ctx = SessionContext() + + batch = pa.RecordBatch.from_arrays( + [pa.array([1, 2, 3, None])], + names=["a"], + ) + ctx.register_record_batches("test_table", [[batch]]) + return ctx + + +def test_ffi_scalar_register(): + ctx = setup_context_with_table() + my_udf = udf(IsNullUDF()) + ctx.register_udf(my_udf) + + result = ctx.sql("select my_custom_is_null(a) from test_table").collect() + + assert len(result) == 1 + assert result[0].num_columns == 1 + print(result) + + result = [r.column(0) for r in result] + expected = [ + pa.array([False, False, False, True], type=pa.bool_()), + ] + + assert result == expected + + +def test_ffi_scalar_call_directly(): + ctx = setup_context_with_table() + my_udf = udf(IsNullUDF()) + + result = ctx.table("test_table").select(my_udf(col("a"))).collect() + + assert len(result) == 1 + assert result[0].num_columns == 1 + print(result) + + result = [r.column(0) for r in result] + expected = [ + pa.array([False, False, False, True], type=pa.bool_()), + ] + + assert result == expected diff --git a/examples/datafusion-ffi-example/python/tests/_test_window_udf.py b/examples/datafusion-ffi-example/python/tests/_test_window_udf.py new file mode 100644 index 000000000..7d96994b9 --- /dev/null +++ b/examples/datafusion-ffi-example/python/tests/_test_window_udf.py @@ -0,0 +1,89 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. + +from __future__ import annotations + +import pyarrow as pa +from datafusion import SessionContext, col, udwf +from datafusion_ffi_example import MyRankUDF + + +def setup_context_with_table(): + ctx = SessionContext() + + # Pick numbers here so we get the same value in both groups + # since we cannot be certain of the output order of batches + batch = pa.RecordBatch.from_arrays( + [ + pa.array([40, 10, 30, 20], type=pa.int64()), + ], + names=["a"], + ) + ctx.register_record_batches("test_table", [[batch]]) + return ctx + + +def test_ffi_window_register(): + ctx = setup_context_with_table() + my_udwf = udwf(MyRankUDF()) + ctx.register_udwf(my_udwf) + + result = ctx.sql( + "select a, my_custom_rank() over (order by a) from test_table" + ).collect() + assert len(result) == 1 + assert result[0].num_columns == 2 + + results = [ + (result[0][0][idx].as_py(), result[0][1][idx].as_py()) for idx in range(4) + ] + results.sort() + + expected = [ + (10, 1), + (20, 2), + (30, 3), + (40, 4), + ] + assert results == expected + + +def test_ffi_window_call_directly(): + ctx = setup_context_with_table() + my_udwf = udwf(MyRankUDF()) + + result = ( + ctx.table("test_table") + .select(col("a"), my_udwf().order_by(col("a")).build()) + .collect() + ) + + assert len(result) == 1 + assert result[0].num_columns == 2 + + results = [ + (result[0][0][idx].as_py(), result[0][1][idx].as_py()) for idx in range(4) + ] + results.sort() + + expected = [ + (10, 1), + (20, 2), + (30, 3), + (40, 4), + ] + assert results == expected diff --git a/examples/datafusion-ffi-example/src/aggregate_udf.rs b/examples/datafusion-ffi-example/src/aggregate_udf.rs new file mode 100644 index 000000000..9481fe9c6 --- /dev/null +++ b/examples/datafusion-ffi-example/src/aggregate_udf.rs @@ -0,0 +1,81 @@ +// Licensed to the Apache Software Foundation (ASF) under one +// or more contributor license agreements. See the NOTICE file +// distributed with this work for additional information +// regarding copyright ownership. The ASF licenses this file +// to you under the Apache License, Version 2.0 (the +// "License"); you may not use this file except in compliance +// with the License. You may obtain a copy of the License at +// +// http://www.apache.org/licenses/LICENSE-2.0 +// +// Unless required by applicable law or agreed to in writing, +// software distributed under the License is distributed on an +// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +// KIND, either express or implied. See the License for the +// specific language governing permissions and limitations +// under the License. + +use arrow_schema::DataType; +use datafusion::error::Result as DataFusionResult; +use datafusion::functions_aggregate::sum::Sum; +use datafusion::logical_expr::function::AccumulatorArgs; +use datafusion::logical_expr::{Accumulator, AggregateUDF, AggregateUDFImpl, Signature}; +use datafusion_ffi::udaf::FFI_AggregateUDF; +use pyo3::types::PyCapsule; +use pyo3::{pyclass, pymethods, Bound, PyResult, Python}; +use std::any::Any; +use std::sync::Arc; + +#[pyclass(name = "MySumUDF", module = "datafusion_ffi_example", subclass)] +#[derive(Debug, Clone)] +pub(crate) struct MySumUDF { + inner: Arc, +} + +#[pymethods] +impl MySumUDF { + #[new] + fn new() -> Self { + Self { + inner: Arc::new(Sum::new()), + } + } + + fn __datafusion_aggregate_udf__<'py>( + &self, + py: Python<'py>, + ) -> PyResult> { + let name = cr"datafusion_aggregate_udf".into(); + + let func = Arc::new(AggregateUDF::from(self.clone())); + let provider = FFI_AggregateUDF::from(func); + + PyCapsule::new(py, provider, Some(name)) + } +} + +impl AggregateUDFImpl for MySumUDF { + fn as_any(&self) -> &dyn Any { + self + } + + fn name(&self) -> &str { + "my_custom_sum" + } + + fn signature(&self) -> &Signature { + self.inner.signature() + } + + fn return_type(&self, arg_types: &[DataType]) -> DataFusionResult { + self.inner.return_type(arg_types) + } + + fn accumulator(&self, acc_args: AccumulatorArgs) -> DataFusionResult> { + self.inner.accumulator(acc_args) + } + + fn coerce_types(&self, arg_types: &[DataType]) -> DataFusionResult> { + self.inner.coerce_types(arg_types) + } +} diff --git a/examples/datafusion-ffi-example/src/catalog_provider.rs b/examples/datafusion-ffi-example/src/catalog_provider.rs index 54e61cf3e..cd2616916 100644 --- a/examples/datafusion-ffi-example/src/catalog_provider.rs +++ b/examples/datafusion-ffi-example/src/catalog_provider.rs @@ -24,7 +24,6 @@ use datafusion::{ catalog::{ CatalogProvider, MemoryCatalogProvider, MemorySchemaProvider, SchemaProvider, TableProvider, }, - common::exec_err, datasource::MemTable, error::{DataFusionError, Result}, }; diff --git a/examples/datafusion-ffi-example/src/lib.rs b/examples/datafusion-ffi-example/src/lib.rs index 3a4cf2247..79af276fd 100644 --- a/examples/datafusion-ffi-example/src/lib.rs +++ b/examples/datafusion-ffi-example/src/lib.rs @@ -16,18 +16,27 @@ // under the License. use crate::catalog_provider::MyCatalogProvider; +use crate::aggregate_udf::MySumUDF; +use crate::scalar_udf::IsNullUDF; use crate::table_function::MyTableFunction; use crate::table_provider::MyTableProvider; +use crate::window_udf::MyRankUDF; use pyo3::prelude::*; pub(crate) mod catalog_provider; +pub(crate) mod aggregate_udf; +pub(crate) mod scalar_udf; pub(crate) mod table_function; pub(crate) mod table_provider; +pub(crate) mod window_udf; #[pymodule] fn datafusion_ffi_example(m: &Bound<'_, PyModule>) -> PyResult<()> { m.add_class::()?; m.add_class::()?; m.add_class::()?; + m.add_class::()?; + m.add_class::()?; + m.add_class::()?; Ok(()) } diff --git a/examples/datafusion-ffi-example/src/scalar_udf.rs b/examples/datafusion-ffi-example/src/scalar_udf.rs new file mode 100644 index 000000000..727666638 --- /dev/null +++ b/examples/datafusion-ffi-example/src/scalar_udf.rs @@ -0,0 +1,91 @@ +// Licensed to the Apache Software Foundation (ASF) under one +// or more contributor license agreements. See the NOTICE file +// distributed with this work for additional information +// regarding copyright ownership. The ASF licenses this file +// to you under the Apache License, Version 2.0 (the +// "License"); you may not use this file except in compliance +// with the License. You may obtain a copy of the License at +// +// http://www.apache.org/licenses/LICENSE-2.0 +// +// Unless required by applicable law or agreed to in writing, +// software distributed under the License is distributed on an +// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +// KIND, either express or implied. See the License for the +// specific language governing permissions and limitations +// under the License. + +use arrow_array::{Array, BooleanArray}; +use arrow_schema::DataType; +use datafusion::common::ScalarValue; +use datafusion::error::Result as DataFusionResult; +use datafusion::logical_expr::{ + ColumnarValue, ScalarFunctionArgs, ScalarUDF, ScalarUDFImpl, Signature, TypeSignature, + Volatility, +}; +use datafusion_ffi::udf::FFI_ScalarUDF; +use pyo3::types::PyCapsule; +use pyo3::{pyclass, pymethods, Bound, PyResult, Python}; +use std::any::Any; +use std::sync::Arc; + +#[pyclass(name = "IsNullUDF", module = "datafusion_ffi_example", subclass)] +#[derive(Debug, Clone)] +pub(crate) struct IsNullUDF { + signature: Signature, +} + +#[pymethods] +impl IsNullUDF { + #[new] + fn new() -> Self { + Self { + signature: Signature::new(TypeSignature::Any(1), Volatility::Immutable), + } + } + + fn __datafusion_scalar_udf__<'py>(&self, py: Python<'py>) -> PyResult> { + let name = cr"datafusion_scalar_udf".into(); + + let func = Arc::new(ScalarUDF::from(self.clone())); + let provider = FFI_ScalarUDF::from(func); + + PyCapsule::new(py, provider, Some(name)) + } +} + +impl ScalarUDFImpl for IsNullUDF { + fn as_any(&self) -> &dyn Any { + self + } + + fn name(&self) -> &str { + "my_custom_is_null" + } + + fn signature(&self) -> &Signature { + &self.signature + } + + fn return_type(&self, _arg_types: &[DataType]) -> DataFusionResult { + Ok(DataType::Boolean) + } + + fn invoke_with_args(&self, args: ScalarFunctionArgs) -> DataFusionResult { + let input = &args.args[0]; + + Ok(match input { + ColumnarValue::Array(arr) => match arr.is_nullable() { + true => { + let nulls = arr.nulls().unwrap(); + let nulls = BooleanArray::from_iter(nulls.iter().map(|x| Some(!x))); + ColumnarValue::Array(Arc::new(nulls)) + } + false => ColumnarValue::Scalar(ScalarValue::Boolean(Some(false))), + }, + ColumnarValue::Scalar(sv) => { + ColumnarValue::Scalar(ScalarValue::Boolean(Some(sv == &ScalarValue::Null))) + } + }) + } +} diff --git a/examples/datafusion-ffi-example/src/window_udf.rs b/examples/datafusion-ffi-example/src/window_udf.rs new file mode 100644 index 000000000..e0d397956 --- /dev/null +++ b/examples/datafusion-ffi-example/src/window_udf.rs @@ -0,0 +1,81 @@ +// Licensed to the Apache Software Foundation (ASF) under one +// or more contributor license agreements. See the NOTICE file +// distributed with this work for additional information +// regarding copyright ownership. The ASF licenses this file +// to you under the Apache License, Version 2.0 (the +// "License"); you may not use this file except in compliance +// with the License. You may obtain a copy of the License at +// +// http://www.apache.org/licenses/LICENSE-2.0 +// +// Unless required by applicable law or agreed to in writing, +// software distributed under the License is distributed on an +// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +// KIND, either express or implied. See the License for the +// specific language governing permissions and limitations +// under the License. + +use arrow_schema::{DataType, FieldRef}; +use datafusion::error::Result as DataFusionResult; +use datafusion::functions_window::rank::rank_udwf; +use datafusion::logical_expr::function::{PartitionEvaluatorArgs, WindowUDFFieldArgs}; +use datafusion::logical_expr::{PartitionEvaluator, Signature, WindowUDF, WindowUDFImpl}; +use datafusion_ffi::udwf::FFI_WindowUDF; +use pyo3::types::PyCapsule; +use pyo3::{pyclass, pymethods, Bound, PyResult, Python}; +use std::any::Any; +use std::sync::Arc; + +#[pyclass(name = "MyRankUDF", module = "datafusion_ffi_example", subclass)] +#[derive(Debug, Clone)] +pub(crate) struct MyRankUDF { + inner: Arc, +} + +#[pymethods] +impl MyRankUDF { + #[new] + fn new() -> Self { + Self { inner: rank_udwf() } + } + + fn __datafusion_window_udf__<'py>(&self, py: Python<'py>) -> PyResult> { + let name = cr"datafusion_window_udf".into(); + + let func = Arc::new(WindowUDF::from(self.clone())); + let provider = FFI_WindowUDF::from(func); + + PyCapsule::new(py, provider, Some(name)) + } +} + +impl WindowUDFImpl for MyRankUDF { + fn as_any(&self) -> &dyn Any { + self + } + + fn name(&self) -> &str { + "my_custom_rank" + } + + fn signature(&self) -> &Signature { + self.inner.signature() + } + + fn partition_evaluator( + &self, + partition_evaluator_args: PartitionEvaluatorArgs, + ) -> DataFusionResult> { + self.inner + .inner() + .partition_evaluator(partition_evaluator_args) + } + + fn field(&self, field_args: WindowUDFFieldArgs) -> DataFusionResult { + self.inner.inner().field(field_args) + } + + fn coerce_types(&self, arg_types: &[DataType]) -> DataFusionResult> { + self.inner.coerce_types(arg_types) + } +} diff --git a/python/datafusion/user_defined.py b/python/datafusion/user_defined.py index dd634c7fb..bd686acbb 100644 --- a/python/datafusion/user_defined.py +++ b/python/datafusion/user_defined.py @@ -22,7 +22,7 @@ import functools from abc import ABCMeta, abstractmethod from enum import Enum -from typing import TYPE_CHECKING, Any, Callable, Optional, TypeVar, overload +from typing import TYPE_CHECKING, Any, Callable, Optional, Protocol, TypeVar, overload import pyarrow as pa @@ -77,6 +77,12 @@ def __str__(self) -> str: return self.name.lower() +class ScalarUDFExportable(Protocol): + """Type hint for object that has __datafusion_scalar_udf__ PyCapsule.""" + + def __datafusion_scalar_udf__(self) -> object: ... # noqa: D105 + + class ScalarUDF: """Class for performing scalar user-defined functions (UDF). @@ -96,6 +102,9 @@ def __init__( See helper method :py:func:`udf` for argument details. """ + if hasattr(func, "__datafusion_scalar_udf__"): + self._udf = df_internal.ScalarUDF.from_pycapsule(func) + return if isinstance(input_types, pa.DataType): input_types = [input_types] self._udf = df_internal.ScalarUDF( @@ -134,6 +143,10 @@ def udf( name: Optional[str] = None, ) -> ScalarUDF: ... + @overload + @staticmethod + def udf(func: ScalarUDFExportable) -> ScalarUDF: ... + @staticmethod def udf(*args: Any, **kwargs: Any): # noqa: D417 """Create a new User-Defined Function (UDF). @@ -147,7 +160,10 @@ def udf(*args: Any, **kwargs: Any): # noqa: D417 Args: func (Callable, optional): Only needed when calling as a function. - Skip this argument when using ``udf`` as a decorator. + Skip this argument when using `udf` as a decorator. If you have a Rust + backed ScalarUDF within a PyCapsule, you can pass this parameter + and ignore the rest. They will be determined directly from the + underlying function. See the online documentation for more information. input_types (list[pa.DataType]): The data types of the arguments to ``func``. This list must be of the same length as the number of arguments. @@ -215,12 +231,31 @@ def wrapper(*args: Any, **kwargs: Any): return decorator + if hasattr(args[0], "__datafusion_scalar_udf__"): + return ScalarUDF.from_pycapsule(args[0]) + if args and callable(args[0]): # Case 1: Used as a function, require the first parameter to be callable return _function(*args, **kwargs) # Case 2: Used as a decorator with parameters return _decorator(*args, **kwargs) + @staticmethod + def from_pycapsule(func: ScalarUDFExportable) -> ScalarUDF: + """Create a Scalar UDF from ScalarUDF PyCapsule object. + + This function will instantiate a Scalar UDF that uses a DataFusion + ScalarUDF that is exported via the FFI bindings. + """ + name = str(func.__class__) + return ScalarUDF( + name=name, + func=func, + input_types=None, + return_type=None, + volatility=None, + ) + class Accumulator(metaclass=ABCMeta): """Defines how an :py:class:`AggregateUDF` accumulates values.""" @@ -242,6 +277,12 @@ def evaluate(self) -> pa.Scalar: """Return the resultant value.""" +class AggregateUDFExportable(Protocol): + """Type hint for object that has __datafusion_aggregate_udf__ PyCapsule.""" + + def __datafusion_aggregate_udf__(self) -> object: ... # noqa: D105 + + class AggregateUDF: """Class for performing scalar user-defined functions (UDF). @@ -263,6 +304,9 @@ def __init__( See :py:func:`udaf` for a convenience function and argument descriptions. """ + if hasattr(accumulator, "__datafusion_aggregate_udf__"): + self._udaf = df_internal.AggregateUDF.from_pycapsule(accumulator) + return self._udaf = df_internal.AggregateUDF( name, accumulator, @@ -307,7 +351,7 @@ def udaf( ) -> AggregateUDF: ... @staticmethod - def udaf(*args: Any, **kwargs: Any): # noqa: D417 + def udaf(*args: Any, **kwargs: Any): # noqa: D417, C901 """Create a new User-Defined Aggregate Function (UDAF). This class allows you to define an aggregate function that can be used in @@ -364,6 +408,10 @@ def udf4() -> Summarize: Args: accum: The accumulator python function. Only needed when calling as a function. Skip this argument when using ``udaf`` as a decorator. + If you have a Rust backed AggregateUDF within a PyCapsule, you can + pass this parameter and ignore the rest. They will be determined + directly from the underlying function. See the online documentation + for more information. input_types: The data types of the arguments to ``accum``. return_type: The data type of the return value. state_type: The data types of the intermediate accumulation. @@ -422,12 +470,32 @@ def wrapper(*args: Any, **kwargs: Any) -> Expr: return decorator + if hasattr(args[0], "__datafusion_aggregate_udf__"): + return AggregateUDF.from_pycapsule(args[0]) + if args and callable(args[0]): # Case 1: Used as a function, require the first parameter to be callable return _function(*args, **kwargs) # Case 2: Used as a decorator with parameters return _decorator(*args, **kwargs) + @staticmethod + def from_pycapsule(func: AggregateUDFExportable) -> AggregateUDF: + """Create an Aggregate UDF from AggregateUDF PyCapsule object. + + This function will instantiate a Aggregate UDF that uses a DataFusion + AggregateUDF that is exported via the FFI bindings. + """ + name = str(func.__class__) + return AggregateUDF( + name=name, + accumulator=func, + input_types=None, + return_type=None, + state_type=None, + volatility=None, + ) + class WindowEvaluator: """Evaluator class for user-defined window functions (UDWF). @@ -588,6 +656,12 @@ def include_rank(self) -> bool: return False +class WindowUDFExportable(Protocol): + """Type hint for object that has __datafusion_window_udf__ PyCapsule.""" + + def __datafusion_window_udf__(self) -> object: ... # noqa: D105 + + class WindowUDF: """Class for performing window user-defined functions (UDF). @@ -608,6 +682,9 @@ def __init__( See :py:func:`udwf` for a convenience function and argument descriptions. """ + if hasattr(func, "__datafusion_window_udf__"): + self._udwf = df_internal.WindowUDF.from_pycapsule(func) + return self._udwf = df_internal.WindowUDF( name, func, input_types, return_type, str(volatility) ) @@ -683,7 +760,10 @@ def biased_numbers() -> BiasedNumbers: Args: func: Only needed when calling as a function. Skip this argument when - using ``udwf`` as a decorator. + using ``udwf`` as a decorator. If you have a Rust backed WindowUDF + within a PyCapsule, you can pass this parameter and ignore the rest. + They will be determined directly from the underlying function. See + the online documentation for more information. input_types: The data types of the arguments. return_type: The data type of the return value. volatility: See :py:class:`Volatility` for allowed values. @@ -692,6 +772,9 @@ def biased_numbers() -> BiasedNumbers: Returns: A user-defined window function that can be used in window function calls. """ + if hasattr(args[0], "__datafusion_window_udf__"): + return WindowUDF.from_pycapsule(args[0]) + if args and callable(args[0]): # Case 1: Used as a function, require the first parameter to be callable return WindowUDF._create_window_udf(*args, **kwargs) @@ -759,6 +842,22 @@ def wrapper(*args: Any, **kwargs: Any) -> Expr: return decorator + @staticmethod + def from_pycapsule(func: WindowUDFExportable) -> WindowUDF: + """Create a Window UDF from WindowUDF PyCapsule object. + + This function will instantiate a Window UDF that uses a DataFusion + WindowUDF that is exported via the FFI bindings. + """ + name = str(func.__class__) + return WindowUDF( + name=name, + func=func, + input_types=None, + return_type=None, + volatility=None, + ) + class TableFunction: """Class for performing user-defined table functions (UDTF). diff --git a/src/functions.rs b/src/functions.rs index b40500b8b..eeef48385 100644 --- a/src/functions.rs +++ b/src/functions.rs @@ -682,7 +682,7 @@ pub fn approx_percentile_cont_with_weight( add_builder_fns_to_aggregate(agg_fn, None, filter, None, None) } -// We handle first_value explicitly because the signature expects an order_by +// We handle last_value explicitly because the signature expects an order_by // https://github.com/apache/datafusion/issues/12376 #[pyfunction] #[pyo3(signature = (expr, distinct=None, filter=None, order_by=None, null_treatment=None))] diff --git a/src/udaf.rs b/src/udaf.rs index 34a9cd51d..78f4e2b0c 100644 --- a/src/udaf.rs +++ b/src/udaf.rs @@ -19,6 +19,10 @@ use std::sync::Arc; use pyo3::{prelude::*, types::PyTuple}; +use crate::common::data_type::PyScalarValue; +use crate::errors::{py_datafusion_err, to_datafusion_err, PyDataFusionResult}; +use crate::expr::PyExpr; +use crate::utils::{parse_volatility, validate_pycapsule}; use datafusion::arrow::array::{Array, ArrayRef}; use datafusion::arrow::datatypes::DataType; use datafusion::arrow::pyarrow::{PyArrowType, ToPyArrow}; @@ -27,11 +31,8 @@ use datafusion::error::{DataFusionError, Result}; use datafusion::logical_expr::{ create_udaf, Accumulator, AccumulatorFactoryFunction, AggregateUDF, }; - -use crate::common::data_type::PyScalarValue; -use crate::errors::to_datafusion_err; -use crate::expr::PyExpr; -use crate::utils::parse_volatility; +use datafusion_ffi::udaf::{FFI_AggregateUDF, ForeignAggregateUDF}; +use pyo3::types::PyCapsule; #[derive(Debug)] struct RustAccumulator { @@ -183,6 +184,26 @@ impl PyAggregateUDF { Ok(Self { function }) } + #[staticmethod] + pub fn from_pycapsule(func: Bound<'_, PyAny>) -> PyDataFusionResult { + if func.hasattr("__datafusion_aggregate_udf__")? { + let capsule = func.getattr("__datafusion_aggregate_udf__")?.call0()?; + let capsule = capsule.downcast::().map_err(py_datafusion_err)?; + validate_pycapsule(capsule, "datafusion_aggregate_udf")?; + + let udaf = unsafe { capsule.reference::() }; + let udaf: ForeignAggregateUDF = udaf.try_into()?; + + Ok(Self { + function: udaf.into(), + }) + } else { + Err(crate::errors::PyDataFusionError::Common( + "__datafusion_aggregate_udf__ does not exist on AggregateUDF object.".to_string(), + )) + } + } + /// creates a new PyExpr with the call of the udf #[pyo3(signature = (*args))] fn __call__(&self, args: Vec) -> PyResult { diff --git a/src/udf.rs b/src/udf.rs index 574c9d7b5..de1e3f18c 100644 --- a/src/udf.rs +++ b/src/udf.rs @@ -17,6 +17,8 @@ use std::sync::Arc; +use datafusion_ffi::udf::{FFI_ScalarUDF, ForeignScalarUDF}; +use pyo3::types::PyCapsule; use pyo3::{prelude::*, types::PyTuple}; use datafusion::arrow::array::{make_array, Array, ArrayData, ArrayRef}; @@ -29,8 +31,9 @@ use datafusion::logical_expr::ScalarUDF; use datafusion::logical_expr::{create_udf, ColumnarValue}; use crate::errors::to_datafusion_err; +use crate::errors::{py_datafusion_err, PyDataFusionResult}; use crate::expr::PyExpr; -use crate::utils::parse_volatility; +use crate::utils::{parse_volatility, validate_pycapsule}; /// Create a Rust callable function from a python function that expects pyarrow arrays fn pyarrow_function_to_rust( @@ -105,6 +108,26 @@ impl PyScalarUDF { Ok(Self { function }) } + #[staticmethod] + pub fn from_pycapsule(func: Bound<'_, PyAny>) -> PyDataFusionResult { + if func.hasattr("__datafusion_scalar_udf__")? { + let capsule = func.getattr("__datafusion_scalar_udf__")?.call0()?; + let capsule = capsule.downcast::().map_err(py_datafusion_err)?; + validate_pycapsule(capsule, "datafusion_scalar_udf")?; + + let udf = unsafe { capsule.reference::() }; + let udf: ForeignScalarUDF = udf.try_into()?; + + Ok(Self { + function: udf.into(), + }) + } else { + Err(crate::errors::PyDataFusionError::Common( + "__datafusion_scalar_udf__ does not exist on ScalarUDF object.".to_string(), + )) + } + } + /// creates a new PyExpr with the call of the udf #[pyo3(signature = (*args))] fn __call__(&self, args: Vec) -> PyResult { diff --git a/src/udwf.rs b/src/udwf.rs index a0c8cc59a..4fb98916b 100644 --- a/src/udwf.rs +++ b/src/udwf.rs @@ -27,16 +27,17 @@ use pyo3::exceptions::PyValueError; use pyo3::prelude::*; use crate::common::data_type::PyScalarValue; -use crate::errors::to_datafusion_err; +use crate::errors::{py_datafusion_err, to_datafusion_err, PyDataFusionResult}; use crate::expr::PyExpr; -use crate::utils::parse_volatility; +use crate::utils::{parse_volatility, validate_pycapsule}; use datafusion::arrow::datatypes::DataType; use datafusion::arrow::pyarrow::{FromPyArrow, PyArrowType, ToPyArrow}; use datafusion::error::{DataFusionError, Result}; use datafusion::logical_expr::{ PartitionEvaluator, PartitionEvaluatorFactory, Signature, Volatility, WindowUDF, WindowUDFImpl, }; -use pyo3::types::{PyList, PyTuple}; +use datafusion_ffi::udwf::{FFI_WindowUDF, ForeignWindowUDF}; +use pyo3::types::{PyCapsule, PyList, PyTuple}; #[derive(Debug)] struct RustPartitionEvaluator { @@ -245,6 +246,26 @@ impl PyWindowUDF { Ok(self.function.call(args).into()) } + #[staticmethod] + pub fn from_pycapsule(func: Bound<'_, PyAny>) -> PyDataFusionResult { + if func.hasattr("__datafusion_window_udf__")? { + let capsule = func.getattr("__datafusion_window_udf__")?.call0()?; + let capsule = capsule.downcast::().map_err(py_datafusion_err)?; + validate_pycapsule(capsule, "datafusion_window_udf")?; + + let udwf = unsafe { capsule.reference::() }; + let udwf: ForeignWindowUDF = udwf.try_into()?; + + Ok(Self { + function: udwf.into(), + }) + } else { + Err(crate::errors::PyDataFusionError::Common( + "__datafusion_window_udf__ does not exist on WindowUDF object.".to_string(), + )) + } + } + fn __repr__(&self) -> PyResult { Ok(format!("WindowUDF({})", self.function.name())) } From 2e1b71369eefc97c22b82be84bbabb414f748fb9 Mon Sep 17 00:00:00 2001 From: kosiew Date: Fri, 4 Jul 2025 20:36:05 +0800 Subject: [PATCH 015/220] refactor: style loading logic in DataFrameHtmlFormatter (#1177) --- python/datafusion/dataframe_formatter.py | 117 ++++++++--------------- python/tests/test_dataframe.py | 60 +++++------- 2 files changed, 63 insertions(+), 114 deletions(-) diff --git a/python/datafusion/dataframe_formatter.py b/python/datafusion/dataframe_formatter.py index 27f00f9c3..2323224b8 100644 --- a/python/datafusion/dataframe_formatter.py +++ b/python/datafusion/dataframe_formatter.py @@ -135,9 +135,6 @@ class DataFrameHtmlFormatter: session """ - # Class variable to track if styles have been loaded in the notebook - _styles_loaded = False - def __init__( self, max_cell_length: int = 25, @@ -260,23 +257,6 @@ def set_custom_header_builder(self, builder: Callable[[Any], str]) -> None: """ self._custom_header_builder = builder - @classmethod - def is_styles_loaded(cls) -> bool: - """Check if HTML styles have been loaded in the current session. - - This method is primarily intended for debugging UI rendering issues - related to style loading. - - Returns: - True if styles have been loaded, False otherwise - - Example: - >>> from datafusion.dataframe_formatter import DataFrameHtmlFormatter - >>> DataFrameHtmlFormatter.is_styles_loaded() - False - """ - return cls._styles_loaded - def format_html( self, batches: list, @@ -315,18 +295,7 @@ def format_html( # Build HTML components html = [] - # Only include styles and scripts if: - # 1. Not using shared styles, OR - # 2. Using shared styles but they haven't been loaded yet - include_styles = ( - not self.use_shared_styles or not DataFrameHtmlFormatter._styles_loaded - ) - - if include_styles: - html.extend(self._build_html_header()) - # If we're using shared styles, mark them as loaded - if self.use_shared_styles: - DataFrameHtmlFormatter._styles_loaded = True + html.extend(self._build_html_header()) html.extend(self._build_table_container_start()) @@ -338,7 +307,7 @@ def format_html( html.append("") # Add footer (JavaScript and messages) - if include_styles and self.enable_cell_expansion: + if self.enable_cell_expansion: html.append(self._get_javascript()) # Always add truncation message if needed (independent of styles) @@ -375,14 +344,20 @@ def format_str( def _build_html_header(self) -> list[str]: """Build the HTML header with CSS styles.""" - html = [] - html.append("") + html.append(f"") return html def _build_table_container_start(self) -> list[str]: @@ -570,28 +545,31 @@ def _get_default_css(self) -> str: def _get_javascript(self) -> str: """Get JavaScript code for interactive elements.""" return """ - - """ + +""" class FormatterManager: @@ -712,24 +690,9 @@ def reset_formatter() -> None: >>> reset_formatter() # Reset formatter to default settings """ formatter = DataFrameHtmlFormatter() - # Reset the styles_loaded flag to ensure styles will be reloaded - DataFrameHtmlFormatter._styles_loaded = False set_formatter(formatter) -def reset_styles_loaded_state() -> None: - """Reset the styles loaded state to force reloading of styles. - - This can be useful when switching between notebook sessions or - when styles need to be refreshed. - - Example: - >>> from datafusion.html_formatter import reset_styles_loaded_state - >>> reset_styles_loaded_state() # Force styles to reload in next render - """ - DataFrameHtmlFormatter._styles_loaded = False - - def _refresh_formatter_reference() -> None: """Refresh formatter reference in any modules using it. diff --git a/python/tests/test_dataframe.py b/python/tests/test_dataframe.py index c9ae38d8e..a3870ead8 100644 --- a/python/tests/test_dataframe.py +++ b/python/tests/test_dataframe.py @@ -42,7 +42,6 @@ configure_formatter, get_formatter, reset_formatter, - reset_styles_loaded_state, ) from datafusion.expr import Window from pyarrow.csv import write_csv @@ -2177,27 +2176,15 @@ def test_html_formatter_shared_styles(df, clean_formatter_state): # First, ensure we're using shared styles configure_formatter(use_shared_styles=True) - # Get HTML output for first table - should include styles html_first = df._repr_html_() - - # Verify styles are included in first render - assert "
" - f"{field.name}
" - f"
" - "" - "" - f"{formatted_value}" - f"" - f"
" - f"
{formatted_value}