Bump version: 0.24.0 → 0.24.1-beta.0

this also upgrades:
- datafusion 47.0 -> 48.0
- half 2.5.0 -> 2.6.0

to be consistent with lance

---------

Signed-off-by: BubbleCal <bubble-cal@outlook.com>

.bumpversion.toml

@@ -1,5 +1,5 @@
 [tool.bumpversion]
-current_version = "0.20.1-beta.1"
+current_version = "0.21.0"
 parse = """(?x)
     (?P<major>0|[1-9]\\d*)\\.
     (?P<minor>0|[1-9]\\d*)\\.

.github/workflows/npm-publish.yml

@@ -550,6 +550,9 @@ jobs:
           bash ci/update_lockfiles.sh
       - name: Push new commit
         uses: ad-m/github-push-action@master
+        with:
+          github_token: ${{ secrets.LANCEDB_RELEASE_TOKEN }}
+          branch: main
       - name: Notify Slack Action
         uses: ravsamhq/notify-slack-action@2.3.0
         if: ${{ always() }}

Cargo.lock

@@ -357,6 +357,8 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "73a47aa0c771b5381de2b7f16998d351a6f4eb839f1e13d48353e17e873d969b"
 dependencies = [
  "bitflags 2.9.1",
+ "serde",
+ "serde_json",
 ]
 
 [[package]]
@@ -1816,9 +1818,9 @@ dependencies = [
 
 [[package]]
 name = "datafusion"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ffe060b978f74ab446be722adb8a274e052e005bf6dfd171caadc3abaad10080"
+checksum = "8a11e19a7ccc5bb979c95c1dceef663eab39c9061b3bbf8d1937faf0f03bf41f"
 dependencies = [
  "arrow",
  "arrow-ipc",
@@ -1841,7 +1843,6 @@ dependencies = [
  "datafusion-functions-nested",
  "datafusion-functions-table",
  "datafusion-functions-window",
- "datafusion-macros",
  "datafusion-optimizer",
  "datafusion-physical-expr",
  "datafusion-physical-expr-common",
@@ -1852,9 +1853,9 @@ dependencies = [
  "futures",
  "itertools 0.14.0",
  "log",
- "object_store 0.12.2",
+ "object_store",
  "parking_lot",
- "rand 0.8.5",
+ "rand 0.9.1",
  "regex",
  "sqlparser 0.55.0",
  "tempfile",
@@ -1865,9 +1866,9 @@ dependencies = [
 
 [[package]]
 name = "datafusion-catalog"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "61fe34f401bd03724a1f96d12108144f8cd495a3cdda2bf5e091822fb80b7e66"
+checksum = "94985e67cab97b1099db2a7af11f31a45008b282aba921c1e1d35327c212ec18"
 dependencies = [
  "arrow",
  "async-trait",
@@ -1884,16 +1885,16 @@ dependencies = [
  "futures",
  "itertools 0.14.0",
  "log",
- "object_store 0.12.2",
+ "object_store",
  "parking_lot",
  "tokio",
 ]
 
 [[package]]
 name = "datafusion-catalog-listing"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a4411b8e3bce5e0fc7521e44f201def2e2d5d1b5f176fb56e8cdc9942c890f00"
+checksum = "e002df133bdb7b0b9b429d89a69aa77b35caeadee4498b2ce1c7c23a99516988"
 dependencies = [
  "arrow",
  "async-trait",
@@ -1908,15 +1909,15 @@ dependencies = [
  "datafusion-session",
  "futures",
  "log",
- "object_store 0.12.2",
+ "object_store",
  "tokio",
 ]
 
 [[package]]
 name = "datafusion-common"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0734015d81c8375eb5d4869b7f7ecccc2ee8d6cb81948ef737cd0e7b743bd69c"
+checksum = "e13242fc58fd753787b0a538e5ae77d356cb9d0656fa85a591a33c5f106267f6"
 dependencies = [
  "ahash",
  "arrow",
@@ -1927,7 +1928,7 @@ dependencies = [
  "indexmap 2.9.0",
  "libc",
  "log",
- "object_store 0.12.2",
+ "object_store",
  "paste",
  "sqlparser 0.55.0",
  "tokio",
@@ -1936,9 +1937,9 @@ dependencies = [
 
 [[package]]
 name = "datafusion-common-runtime"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5167bb1d2ccbb87c6bc36c295274d7a0519b14afcfdaf401d53cbcaa4ef4968b"
+checksum = "d2239f964e95c3a5d6b4a8cde07e646de8995c1396a7fd62c6e784f5341db499"
 dependencies = [
  "futures",
  "log",
@@ -1947,9 +1948,9 @@ dependencies = [
 
 [[package]]
 name = "datafusion-datasource"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "04e602dcdf2f50c2abf297cc2203c73531e6f48b29516af7695d338cf2a778b1"
+checksum = "2cf792579bc8bf07d1b2f68c2d5382f8a63679cce8fbebfd4ba95742b6e08864"
 dependencies = [
  "arrow",
  "async-trait",
@@ -1967,17 +1968,17 @@ dependencies = [
  "glob",
  "itertools 0.14.0",
  "log",
- "object_store 0.12.2",
+ "object_store",
- "rand 0.8.5",
+ "rand 0.9.1",
  "tokio",
  "url",
 ]
 
 [[package]]
 name = "datafusion-datasource-csv"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e3bb2253952dc32296ed5b84077cb2e0257fea4be6373e1c376426e17ead4ef6"
+checksum = "cfc114f9a1415174f3e8d2719c371fc72092ef2195a7955404cfe6b2ba29a706"
 dependencies = [
  "arrow",
  "async-trait",
@@ -1993,16 +1994,16 @@ dependencies = [
  "datafusion-physical-plan",
  "datafusion-session",
  "futures",
- "object_store 0.12.2",
+ "object_store",
  "regex",
  "tokio",
 ]
 
 [[package]]
 name = "datafusion-datasource-json"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5b8c7f47a5d2fe03bfa521ec9bafdb8a5c82de8377f60967c3663f00c8790352"
+checksum = "d88dd5e215c420a52362b9988ecd4cefd71081b730663d4f7d886f706111fc75"
 dependencies = [
  "arrow",
  "async-trait",
@@ -2018,22 +2019,22 @@ dependencies = [
  "datafusion-physical-plan",
  "datafusion-session",
  "futures",
- "object_store 0.12.2",
+ "object_store",
  "serde_json",
  "tokio",
 ]
 
 [[package]]
 name = "datafusion-doc"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a91f8c2c5788ef32f48ff56c68e5b545527b744822a284373ac79bba1ba47292"
+checksum = "e0e7b648387b0c1937b83cb328533c06c923799e73a9e3750b762667f32662c0"
 
 [[package]]
 name = "datafusion-execution"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "06f004d100f49a3658c9da6fb0c3a9b760062d96cd4ad82ccc3b7b69a9fb2f84"
+checksum = "9609d83d52ff8315283c6dad3b97566e877d8f366fab4c3297742f33dcd636c7"
 dependencies = [
  "arrow",
  "dashmap",
@@ -2041,18 +2042,18 @@ dependencies = [
  "datafusion-expr",
  "futures",
  "log",
- "object_store 0.12.2",
+ "object_store",
  "parking_lot",
- "rand 0.8.5",
+ "rand 0.9.1",
  "tempfile",
  "url",
 ]
 
 [[package]]
 name = "datafusion-expr"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7a4e4ce3802609be38eeb607ee72f6fe86c3091460de9dbfae9e18db423b3964"
+checksum = "e75230cd67f650ef0399eb00f54d4a073698f2c0262948298e5299fc7324da63"
 dependencies = [
  "arrow",
  "chrono",
@@ -2070,9 +2071,9 @@ dependencies = [
 
 [[package]]
 name = "datafusion-expr-common"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "422ac9cf3b22bbbae8cdf8ceb33039107fde1b5492693168f13bd566b1bcc839"
+checksum = "70fafb3a045ed6c49cfca0cd090f62cf871ca6326cc3355cb0aaf1260fa760b6"
 dependencies = [
  "arrow",
  "datafusion-common",
@@ -2083,9 +2084,9 @@ dependencies = [
 
 [[package]]
 name = "datafusion-functions"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2ddf0a0a2db5d2918349c978d42d80926c6aa2459cd8a3c533a84ec4bb63479e"
+checksum = "cdf9a9cf655265861a20453b1e58357147eab59bdc90ce7f2f68f1f35104d3bb"
 dependencies = [
  "arrow",
  "arrow-buffer",
@@ -2103,7 +2104,7 @@ dependencies = [
  "itertools 0.14.0",
  "log",
  "md-5",
- "rand 0.8.5",
+ "rand 0.9.1",
  "regex",
  "sha2",
  "unicode-segmentation",
@@ -2112,9 +2113,9 @@ dependencies = [
 
 [[package]]
 name = "datafusion-functions-aggregate"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "408a05dafdc70d05a38a29005b8b15e21b0238734dab1e98483fcb58038c5aba"
+checksum = "7f07e49733d847be0a05235e17b884d326a2fd402c97a89fe8bcf0bfba310005"
 dependencies = [
  "ahash",
  "arrow",
@@ -2133,9 +2134,9 @@ dependencies = [
 
 [[package]]
 name = "datafusion-functions-aggregate-common"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "756d21da2dd6c9bef97af1504970ff56cbf35d03fbd4ffd62827f02f4d2279d4"
+checksum = "4512607e10d72b0b0a1dc08f42cb5bd5284cb8348b7fea49dc83409493e32b1b"
 dependencies = [
  "ahash",
  "arrow",
@@ -2146,9 +2147,9 @@ dependencies = [
 
 [[package]]
 name = "datafusion-functions-nested"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8d8d50f6334b378930d992d801a10ac5b3e93b846b39e4a05085742572844537"
+checksum = "2ab331806e34f5545e5f03396e4d5068077395b1665795d8f88c14ec4f1e0b7a"
 dependencies = [
  "arrow",
  "arrow-ord",
@@ -2167,9 +2168,9 @@ dependencies = [
 
 [[package]]
 name = "datafusion-functions-table"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "cc9a97220736c8fff1446e936be90d57216c06f28969f9ffd3b72ac93c958c8a"
+checksum = "d4ac2c0be983a06950ef077e34e0174aa0cb9e346f3aeae459823158037ade37"
 dependencies = [
  "arrow",
  "async-trait",
@@ -2183,10 +2184,11 @@ dependencies = [
 
 [[package]]
 name = "datafusion-functions-window"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "cefc2d77646e1aadd1d6a9c40088937aedec04e68c5f0465939912e1291f8193"
+checksum = "36f3d92731de384c90906941d36dcadf6a86d4128409a9c5cd916662baed5f53"
 dependencies = [
+ "arrow",
  "datafusion-common",
  "datafusion-doc",
  "datafusion-expr",
@@ -2200,9 +2202,9 @@ dependencies = [
 
 [[package]]
 name = "datafusion-functions-window-common"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "dd4aff082c42fa6da99ce0698c85addd5252928c908eb087ca3cfa64ff16b313"
+checksum = "c679f8bf0971704ec8fd4249fcbb2eb49d6a12cc3e7a840ac047b4928d3541b5"
 dependencies = [
  "datafusion-common",
  "datafusion-physical-expr-common",
@@ -2210,9 +2212,9 @@ dependencies = [
 
 [[package]]
 name = "datafusion-macros"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "df6f88d7ee27daf8b108ba910f9015176b36fbc72902b1ca5c2a5f1d1717e1a1"
+checksum = "2821de7cb0362d12e75a5196b636a59ea3584ec1e1cc7dc6f5e34b9e8389d251"
 dependencies = [
  "datafusion-expr",
  "quote",
@@ -2221,9 +2223,9 @@ dependencies = [
 
 [[package]]
 name = "datafusion-optimizer"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "084d9f979c4b155346d3c34b18f4256e6904ded508e9554d90fed416415c3515"
+checksum = "1594c7a97219ede334f25347ad8d57056621e7f4f35a0693c8da876e10dd6a53"
 dependencies = [
  "arrow",
  "chrono",
@@ -2239,9 +2241,9 @@ dependencies = [
 
 [[package]]
 name = "datafusion-physical-expr"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "64c536062b0076f4e30084065d805f389f9fe38af0ca75bcbac86bc5e9fbab65"
+checksum = "dc6da0f2412088d23f6b01929dedd687b5aee63b19b674eb73d00c3eb3c883b7"
 dependencies = [
  "ahash",
  "arrow",
@@ -2256,14 +2258,14 @@ dependencies = [
  "itertools 0.14.0",
  "log",
  "paste",
- "petgraph",
+ "petgraph 0.8.2",
 ]
 
 [[package]]
 name = "datafusion-physical-expr-common"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "f8a92b53b3193fac1916a1c5b8e3f4347c526f6822e56b71faa5fb372327a863"
+checksum = "dcb0dbd9213078a593c3fe28783beaa625a4e6c6a6c797856ee2ba234311fb96"
 dependencies = [
  "ahash",
  "arrow",
@@ -2275,9 +2277,9 @@ dependencies = [
 
 [[package]]
 name = "datafusion-physical-optimizer"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "6fa0a5ac94c7cf3da97bedabd69d6bbca12aef84b9b37e6e9e8c25286511b5e2"
+checksum = "6d140854b2db3ef8ac611caad12bfb2e1e1de827077429322a6188f18fc0026a"
 dependencies = [
  "arrow",
  "datafusion-common",
@@ -2293,9 +2295,9 @@ dependencies = [
 
 [[package]]
 name = "datafusion-physical-plan"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "690c615db468c2e5fe5085b232d8b1c088299a6c63d87fd960a354a71f7acb55"
+checksum = "b46cbdf21a01206be76d467f325273b22c559c744a012ead5018dfe79597de08"
 dependencies = [
  "ahash",
  "arrow",
@@ -2323,9 +2325,9 @@ dependencies = [
 
 [[package]]
 name = "datafusion-session"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ad229a134c7406c057ece00c8743c0c34b97f4e72f78b475fe17b66c5e14fa4f"
+checksum = "3a72733766ddb5b41534910926e8da5836622316f6283307fd9fb7e19811a59c"
 dependencies = [
  "arrow",
  "async-trait",
@@ -2340,16 +2342,16 @@ dependencies = [
  "futures",
  "itertools 0.14.0",
  "log",
- "object_store 0.12.2",
+ "object_store",
  "parking_lot",
  "tokio",
 ]
 
 [[package]]
 name = "datafusion-sql"
-version = "47.0.0"
+version = "48.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "64f6ab28b72b664c21a27b22a2ff815fd390ed224c26e89a93b5a8154a4e8607"
+checksum = "c5162338cdec9cc7ea13a0e6015c361acad5ec1d88d83f7c86301f789473971f"
 dependencies = [
  "arrow",
  "bigdecimal",
@@ -2813,8 +2815,8 @@ checksum = "42703706b716c37f96a77aea830392ad231f44c9e9a67872fa5548707e11b11c"
 
 [[package]]
 name = "fsst"
-version = "0.30.0"
+version = "0.31.1"
-source = "git+https://github.com/lancedb/lance.git?tag=v0.30.0-beta.1#a499cfa06b7221b895bc13908cfc2ee7aadba46e"
+source = "git+https://github.com/lancedb/lance.git?tag=v0.31.1-beta.2#dff098a5aa66866197cfcd7ae7ca004aed02928f"
 dependencies = [
  "rand 0.8.5",
 ]
@@ -3280,9 +3282,9 @@ dependencies = [
 
 [[package]]
 name = "half"
-version = "2.5.0"
+version = "2.6.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7db2ff139bba50379da6aa0766b52fdcb62cb5b263009b09ed58ba604e14bbd1"
+checksum = "459196ed295495a68f7d7fe1d84f6c4b7ff0e21fe3017b2f283c6fac3ad803c9"
 dependencies = [
  "bytemuck",
  "cfg-if",
@@ -3906,8 +3908,8 @@ dependencies = [
 
 [[package]]
 name = "lance"
-version = "0.30.0"
+version = "0.31.1"
-source = "git+https://github.com/lancedb/lance.git?tag=v0.30.0-beta.1#a499cfa06b7221b895bc13908cfc2ee7aadba46e"
+source = "git+https://github.com/lancedb/lance.git?tag=v0.31.1-beta.2#dff098a5aa66866197cfcd7ae7ca004aed02928f"
 dependencies = [
  "arrow",
  "arrow-arith",
@@ -3947,10 +3949,9 @@ dependencies = [
  "lance-io",
  "lance-linalg",
  "lance-table",
- "lazy_static",
  "log",
  "moka",
- "object_store 0.11.2",
+ "object_store",
  "permutation",
  "pin-project",
  "prost",
@@ -3970,8 +3971,8 @@ dependencies = [
 
 [[package]]
 name = "lance-arrow"
-version = "0.30.0"
+version = "0.31.1"
-source = "git+https://github.com/lancedb/lance.git?tag=v0.30.0-beta.1#a499cfa06b7221b895bc13908cfc2ee7aadba46e"
+source = "git+https://github.com/lancedb/lance.git?tag=v0.31.1-beta.2#dff098a5aa66866197cfcd7ae7ca004aed02928f"
 dependencies = [
  "arrow-array",
  "arrow-buffer",
@@ -3988,8 +3989,8 @@ dependencies = [
 
 [[package]]
 name = "lance-core"
-version = "0.30.0"
+version = "0.31.1"
-source = "git+https://github.com/lancedb/lance.git?tag=v0.30.0-beta.1#a499cfa06b7221b895bc13908cfc2ee7aadba46e"
+source = "git+https://github.com/lancedb/lance.git?tag=v0.31.1-beta.2#dff098a5aa66866197cfcd7ae7ca004aed02928f"
 dependencies = [
  "arrow-array",
  "arrow-buffer",
@@ -4003,13 +4004,12 @@ dependencies = [
  "deepsize",
  "futures",
  "lance-arrow",
- "lazy_static",
  "libc",
  "log",
  "mock_instant",
  "moka",
  "num_cpus",
- "object_store 0.11.2",
+ "object_store",
  "pin-project",
  "prost",
  "rand 0.8.5",
@@ -4025,8 +4025,8 @@ dependencies = [
 
 [[package]]
 name = "lance-datafusion"
-version = "0.30.0"
+version = "0.31.1"
-source = "git+https://github.com/lancedb/lance.git?tag=v0.30.0-beta.1#a499cfa06b7221b895bc13908cfc2ee7aadba46e"
+source = "git+https://github.com/lancedb/lance.git?tag=v0.31.1-beta.2#dff098a5aa66866197cfcd7ae7ca004aed02928f"
 dependencies = [
  "arrow",
  "arrow-array",
@@ -4043,7 +4043,6 @@ dependencies = [
  "lance-arrow",
  "lance-core",
  "lance-datagen",
- "lazy_static",
  "log",
  "pin-project",
  "prost",
@@ -4055,8 +4054,8 @@ dependencies = [
 
 [[package]]
 name = "lance-datagen"
-version = "0.30.0"
+version = "0.31.1"
-source = "git+https://github.com/lancedb/lance.git?tag=v0.30.0-beta.1#a499cfa06b7221b895bc13908cfc2ee7aadba46e"
+source = "git+https://github.com/lancedb/lance.git?tag=v0.31.1-beta.2#dff098a5aa66866197cfcd7ae7ca004aed02928f"
 dependencies = [
  "arrow",
  "arrow-array",
@@ -4067,12 +4066,13 @@ dependencies = [
  "hex",
  "rand 0.8.5",
  "rand_xoshiro",
+ "random_word 0.5.0",
 ]
 
 [[package]]
 name = "lance-encoding"
-version = "0.30.0"
+version = "0.31.1"
-source = "git+https://github.com/lancedb/lance.git?tag=v0.30.0-beta.1#a499cfa06b7221b895bc13908cfc2ee7aadba46e"
+source = "git+https://github.com/lancedb/lance.git?tag=v0.31.1-beta.2#dff098a5aa66866197cfcd7ae7ca004aed02928f"
 dependencies = [
  "arrayref",
  "arrow",
@@ -4093,7 +4093,6 @@ dependencies = [
  "itertools 0.13.0",
  "lance-arrow",
  "lance-core",
- "lazy_static",
  "log",
  "lz4",
  "num-traits",
@@ -4106,13 +4105,14 @@ dependencies = [
  "snafu",
  "tokio",
  "tracing",
+ "xxhash-rust",
  "zstd",
 ]
 
 [[package]]
 name = "lance-file"
-version = "0.30.0"
+version = "0.31.1"
-source = "git+https://github.com/lancedb/lance.git?tag=v0.30.0-beta.1#a499cfa06b7221b895bc13908cfc2ee7aadba46e"
+source = "git+https://github.com/lancedb/lance.git?tag=v0.31.1-beta.2#dff098a5aa66866197cfcd7ae7ca004aed02928f"
 dependencies = [
  "arrow-arith",
  "arrow-array",
@@ -4133,7 +4133,7 @@ dependencies = [
  "lance-io",
  "log",
  "num-traits",
- "object_store 0.11.2",
+ "object_store",
  "prost",
  "prost-build",
  "prost-types",
@@ -4146,8 +4146,8 @@ dependencies = [
 
 [[package]]
 name = "lance-index"
-version = "0.30.0"
+version = "0.31.1"
-source = "git+https://github.com/lancedb/lance.git?tag=v0.30.0-beta.1#a499cfa06b7221b895bc13908cfc2ee7aadba46e"
+source = "git+https://github.com/lancedb/lance.git?tag=v0.31.1-beta.2#dff098a5aa66866197cfcd7ae7ca004aed02928f"
 dependencies = [
  "arrow",
  "arrow-array",
@@ -4180,11 +4180,10 @@ dependencies = [
  "lance-io",
  "lance-linalg",
  "lance-table",
- "lazy_static",
  "log",
  "moka",
  "num-traits",
- "object_store 0.11.2",
+ "object_store",
  "prost",
  "prost-build",
  "rand 0.8.5",
@@ -4202,8 +4201,8 @@ dependencies = [
 
 [[package]]
 name = "lance-io"
-version = "0.30.0"
+version = "0.31.1"
-source = "git+https://github.com/lancedb/lance.git?tag=v0.30.0-beta.1#a499cfa06b7221b895bc13908cfc2ee7aadba46e"
+source = "git+https://github.com/lancedb/lance.git?tag=v0.31.1-beta.2#dff098a5aa66866197cfcd7ae7ca004aed02928f"
 dependencies = [
  "arrow",
  "arrow-arith",
@@ -4225,9 +4224,8 @@ dependencies = [
  "futures",
  "lance-arrow",
  "lance-core",
- "lazy_static",
  "log",
- "object_store 0.11.2",
+ "object_store",
  "path_abs",
  "pin-project",
  "prost",
@@ -4242,8 +4240,8 @@ dependencies = [
 
 [[package]]
 name = "lance-linalg"
-version = "0.30.0"
+version = "0.31.1"
-source = "git+https://github.com/lancedb/lance.git?tag=v0.30.0-beta.1#a499cfa06b7221b895bc13908cfc2ee7aadba46e"
+source = "git+https://github.com/lancedb/lance.git?tag=v0.31.1-beta.2#dff098a5aa66866197cfcd7ae7ca004aed02928f"
 dependencies = [
  "arrow-array",
  "arrow-ord",
@@ -4255,7 +4253,6 @@ dependencies = [
  "half",
  "lance-arrow",
  "lance-core",
- "lazy_static",
  "log",
  "num-traits",
  "rand 0.8.5",
@@ -4266,8 +4263,8 @@ dependencies = [
 
 [[package]]
 name = "lance-table"
-version = "0.30.0"
+version = "0.31.1"
-source = "git+https://github.com/lancedb/lance.git?tag=v0.30.0-beta.1#a499cfa06b7221b895bc13908cfc2ee7aadba46e"
+source = "git+https://github.com/lancedb/lance.git?tag=v0.31.1-beta.2#dff098a5aa66866197cfcd7ae7ca004aed02928f"
 dependencies = [
  "arrow",
  "arrow-array",
@@ -4286,9 +4283,8 @@ dependencies = [
  "lance-core",
  "lance-file",
  "lance-io",
- "lazy_static",
  "log",
- "object_store 0.11.2",
+ "object_store",
  "prost",
  "prost-build",
  "prost-types",
@@ -4306,8 +4302,8 @@ dependencies = [
 
 [[package]]
 name = "lance-testing"
-version = "0.30.0"
+version = "0.31.1"
-source = "git+https://github.com/lancedb/lance.git?tag=v0.30.0-beta.1#a499cfa06b7221b895bc13908cfc2ee7aadba46e"
+source = "git+https://github.com/lancedb/lance.git?tag=v0.31.1-beta.2#dff098a5aa66866197cfcd7ae7ca004aed02928f"
 dependencies = [
  "arrow-array",
  "arrow-schema",
@@ -4318,7 +4314,7 @@ dependencies = [
 
 [[package]]
 name = "lancedb"
-version = "0.20.1-beta.1"
+version = "0.21.0"
 dependencies = [
  "arrow",
  "arrow-array",
@@ -4365,12 +4361,12 @@ dependencies = [
  "log",
  "moka",
  "num-traits",
- "object_store 0.11.2",
+ "object_store",
  "pin-project",
  "polars",
  "polars-arrow",
  "rand 0.9.1",
- "random_word",
+ "random_word 0.4.3",
  "regex",
  "reqwest",
  "rstest",
@@ -4405,7 +4401,7 @@ dependencies = [
 
 [[package]]
 name = "lancedb-node"
-version = "0.20.1-beta.1"
+version = "0.21.0"
 dependencies = [
  "arrow-array",
  "arrow-ipc",
@@ -4422,7 +4418,7 @@ dependencies = [
  "lancedb",
  "lzma-sys",
  "neon",
- "object_store 0.11.2",
+ "object_store",
  "once_cell",
  "snafu",
  "tokio",
@@ -4430,7 +4426,7 @@ dependencies = [
 
 [[package]]
 name = "lancedb-nodejs"
-version = "0.20.1-beta.1"
+version = "0.21.0"
 dependencies = [
  "arrow-array",
  "arrow-ipc",
@@ -4450,7 +4446,7 @@ dependencies = [
 
 [[package]]
 name = "lancedb-python"
-version = "0.23.1-beta.1"
+version = "0.24.0"
 dependencies = [
  "arrow",
  "env_logger",
@@ -5159,38 +5155,6 @@ dependencies = [
  "memchr",
 ]
 
-[[package]]
-name = "object_store"
-version = "0.11.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "3cfccb68961a56facde1163f9319e0d15743352344e7808a11795fb99698dcaf"
-dependencies = [
- "async-trait",
- "base64 0.22.1",
- "bytes",
- "chrono",
- "futures",
- "httparse",
- "humantime",
- "hyper 1.6.0",
- "itertools 0.13.0",
- "md-5",
- "parking_lot",
- "percent-encoding",
- "quick-xml",
- "rand 0.8.5",
- "reqwest",
- "ring",
- "rustls-pemfile 2.2.0",
- "serde",
- "serde_json",
- "snafu",
- "tokio",
- "tracing",
- "url",
- "walkdir",
-]
-
 [[package]]
 name = "object_store"
 version = "0.12.2"
@@ -5198,14 +5162,28 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "7781f96d79ed0f961a7021424ab01840efbda64ae7a505aaea195efc91eaaec4"
 dependencies = [
  "async-trait",
+ "base64 0.22.1",
  "bytes",
  "chrono",
+ "form_urlencoded",
  "futures",
  "http 1.3.1",
+ "http-body-util",
+ "httparse",
  "humantime",
+ "hyper 1.6.0",
  "itertools 0.14.0",
+ "md-5",
  "parking_lot",
  "percent-encoding",
+ "quick-xml",
+ "rand 0.9.1",
+ "reqwest",
+ "ring",
+ "rustls-pemfile 2.2.0",
+ "serde",
+ "serde_json",
+ "serde_urlencoded",
  "thiserror 2.0.12",
  "tokio",
  "tracing",
@@ -5383,6 +5361,18 @@ dependencies = [
  "indexmap 2.9.0",
 ]
 
+[[package]]
+name = "petgraph"
+version = "0.8.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "54acf3a685220b533e437e264e4d932cfbdc4cc7ec0cd232ed73c08d03b8a7ca"
+dependencies = [
+ "fixedbitset",
+ "hashbrown 0.15.4",
+ "indexmap 2.9.0",
+ "serde",
+]
+
 [[package]]
 name = "phf"
 version = "0.11.3"
@@ -5915,7 +5905,7 @@ dependencies = [
  "log",
  "multimap",
  "once_cell",
- "petgraph",
+ "petgraph 0.7.1",
  "prettyplease",
  "prost",
  "prost-types",
@@ -6257,6 +6247,20 @@ dependencies = [
  "unicase",
 ]
 
+[[package]]
+name = "random_word"
+version = "0.5.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "bcd87d2e3f99cc11e6c7fc518f09e63e194f7243b4cf30c979b0c524d04fbd90"
+dependencies = [
+ "ahash",
+ "brotli",
+ "once_cell",
+ "paste",
+ "rand 0.8.5",
+ "unicase",
+]
+
 [[package]]
 name = "rangemap"
 version = "1.5.1"

Cargo.toml

@@ -21,14 +21,14 @@ categories = ["database-implementations"]
 rust-version = "1.78.0"
 
 [workspace.dependencies]
-lance = { "version" = "=0.30.0", "features" = ["dynamodb"], tag = "v0.30.0-beta.1", git="https://github.com/lancedb/lance.git" }
+lance = { "version" = "=0.31.1", tag="v0.31.1-beta.2", git="https://github.com/lancedb/lance.git", features = ["dynamodb"] }
-lance-io = { version = "=0.30.0", tag = "v0.30.0-beta.1", git="https://github.com/lancedb/lance.git" }
+lance-io = { "version" = "=0.31.1", tag="v0.31.1-beta.2", git="https://github.com/lancedb/lance.git" }
-lance-index = { version = "=0.30.0", tag = "v0.30.0-beta.1", git="https://github.com/lancedb/lance.git" }
+lance-index = { "version" = "=0.31.1", tag="v0.31.1-beta.2", git="https://github.com/lancedb/lance.git" }
-lance-linalg = { version = "=0.30.0", tag = "v0.30.0-beta.1", git="https://github.com/lancedb/lance.git" }
+lance-linalg = { "version" = "=0.31.1", tag="v0.31.1-beta.2", git="https://github.com/lancedb/lance.git" }
-lance-table = { version = "=0.30.0", tag = "v0.30.0-beta.1", git="https://github.com/lancedb/lance.git" }
+lance-table = { "version" = "=0.31.1", tag="v0.31.1-beta.2", git="https://github.com/lancedb/lance.git" }
-lance-testing = { version = "=0.30.0", tag = "v0.30.0-beta.1", git="https://github.com/lancedb/lance.git" }
+lance-testing = { "version" = "=0.31.1", tag="v0.31.1-beta.2", git="https://github.com/lancedb/lance.git" }
-lance-datafusion = { version = "=0.30.0", tag = "v0.30.0-beta.1", git="https://github.com/lancedb/lance.git" }
+lance-datafusion = { "version" = "=0.31.1", tag="v0.31.1-beta.2", git="https://github.com/lancedb/lance.git" }
-lance-encoding = { version = "=0.30.0", tag = "v0.30.0-beta.1", git="https://github.com/lancedb/lance.git" }
+lance-encoding = { "version" = "=0.31.1", tag="v0.31.1-beta.2", git="https://github.com/lancedb/lance.git" }
 # Note that this one does not include pyarrow
 arrow = { version = "55.1", optional = false }
 arrow-array = "55.1"
@@ -39,20 +39,20 @@ arrow-schema = "55.1"
 arrow-arith = "55.1"
 arrow-cast = "55.1"
 async-trait = "0"
-datafusion = { version = "47.0", default-features = false }
+datafusion = { version = "48.0", default-features = false }
-datafusion-catalog = "47.0"
+datafusion-catalog = "48.0"
-datafusion-common = { version = "47.0", default-features = false }
+datafusion-common = { version = "48.0", default-features = false }
-datafusion-execution = "47.0"
+datafusion-execution = "48.0"
-datafusion-expr = "47.0"
+datafusion-expr = "48.0"
-datafusion-physical-plan = "47.0"
+datafusion-physical-plan = "48.0"
 env_logger = "0.11"
-half = { "version" = "=2.5.0", default-features = false, features = [
+half = { "version" = "=2.6.0", default-features = false, features = [
     "num-traits",
 ] }
 futures = "0"
 log = "0.4"
 moka = { version = "0.12", features = ["future"] }
-object_store = "0.11.0"
+object_store = "0.12.0"
 pin-project = "1.0.7"
 snafu = "0.8"
 url = "2"

docs/src/notebooks/Multivector_on_LanceDB.ipynb

@@ -428,7 +428,7 @@
         "\n",
         "**Why?**  \n",
         "Embedding the UFO dataset and ingesting it into LanceDB takes **~2 hours on a T4 GPU**. To save time:  \n",
-        "- **Use the pre-prepared table with index created ** (provided below) to proceed directly to step7: search.  \n",
+        "- **Use the pre-prepared table with index created** (provided below) to proceed directly to **Step 7**: search.  \n",
         "- **Step 5a** contains the full ingestion code for reference (run it only if necessary).  \n",
         "- **Step 6** contains the details on creating the index on the multivector column"
       ]

java/core/pom.xml

@@ -8,7 +8,7 @@
     <parent>
         <groupId>com.lancedb</groupId>
         <artifactId>lancedb-parent</artifactId>
-        <version>0.20.1-beta.1</version>
+        <version>0.21.0-final.0</version>
         <relativePath>../pom.xml</relativePath>
     </parent>
 

java/pom.xml

@@ -6,7 +6,7 @@
 
     <groupId>com.lancedb</groupId>
     <artifactId>lancedb-parent</artifactId>
-    <version>0.20.1-beta.1</version>
+    <version>0.21.0-final.0</version>
     <packaging>pom</packaging>
 
     <name>LanceDB Parent</name>

node/package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "vectordb",
-  "version": "0.20.1-beta.1",
+  "version": "0.21.0",
   "lockfileVersion": 3,
   "requires": true,
   "packages": {
     "": {
       "name": "vectordb",
-      "version": "0.20.1-beta.1",
+      "version": "0.21.0",
       "cpu": [
         "x64",
         "arm64"
@@ -52,11 +52,11 @@
         "uuid": "^9.0.0"
       },
       "optionalDependencies": {
-        "@lancedb/vectordb-darwin-arm64": "0.20.1-beta.1",
+        "@lancedb/vectordb-darwin-arm64": "0.21.0",
-        "@lancedb/vectordb-darwin-x64": "0.20.1-beta.1",
+        "@lancedb/vectordb-darwin-x64": "0.21.0",
-        "@lancedb/vectordb-linux-arm64-gnu": "0.20.1-beta.1",
+        "@lancedb/vectordb-linux-arm64-gnu": "0.21.0",
-        "@lancedb/vectordb-linux-x64-gnu": "0.20.1-beta.1",
+        "@lancedb/vectordb-linux-x64-gnu": "0.21.0",
-        "@lancedb/vectordb-win32-x64-msvc": "0.20.1-beta.1"
+        "@lancedb/vectordb-win32-x64-msvc": "0.21.0"
       },
       "peerDependencies": {
         "@apache-arrow/ts": "^14.0.2",
@@ -327,9 +327,9 @@
       }
     },
     "node_modules/@lancedb/vectordb-darwin-arm64": {
-      "version": "0.20.1-beta.1",
+      "version": "0.21.0",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.20.1-beta.1.tgz",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-arm64/-/vectordb-darwin-arm64-0.21.0.tgz",
-      "integrity": "sha512-DPD8gwFQz5aENYYbTFS/l3YX/rqzS6Kj2B4IZERccVFULQsdR5YwtaAfFwTMp7NSnsjWKwJAknohiMZlJr4njQ==",
+      "integrity": "sha512-FTKbdYG36mvQ75tId+esyRfRjIBzryRhAp/6h51tiXy8gsq/TButuiPdqIXeonNModEjhu8wkzsGFwgjCcePow==",
       "cpu": [
         "arm64"
       ],
@@ -339,9 +339,9 @@
       ]
     },
     "node_modules/@lancedb/vectordb-darwin-x64": {
-      "version": "0.20.1-beta.1",
+      "version": "0.21.0",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.20.1-beta.1.tgz",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-darwin-x64/-/vectordb-darwin-x64-0.21.0.tgz",
-      "integrity": "sha512-lTPtlRSTC08UgQW5Bv8WYhdbogAgUJ+9ejg+UE+fwP9gEsgEKXL/SHBm+9gmAlTo7LbrxJjg0CtCde/mW68UTw==",
+      "integrity": "sha512-vGaFBr2sQZWE0mudg3LGTHiRE7p2Qce2ogiE2VAf1DLAJ4MrIhgVmEttf966ausIwNCgml+5AzUntw6zC0Oyuw==",
       "cpu": [
         "x64"
       ],
@@ -351,9 +351,9 @@
       ]
     },
     "node_modules/@lancedb/vectordb-linux-arm64-gnu": {
-      "version": "0.20.1-beta.1",
+      "version": "0.21.0",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.20.1-beta.1.tgz",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-arm64-gnu/-/vectordb-linux-arm64-gnu-0.21.0.tgz",
-      "integrity": "sha512-w/3O9FvwQiGegYsM21yZ0FezfOFVsW7HttYwwPzZMZaCpK3/i+LvZVSqwO4qXHHJBtHgKevonINyvVlg5487aQ==",
+      "integrity": "sha512-KlxqhnX4eBN6rDqrPgf/x/vLpnHK2UcIzNLpiOZzSAhooCmKmnNpfs/EXt+KRFloEQMy25AHpMpqkSPv1Q2oDA==",
       "cpu": [
         "arm64"
       ],
@@ -363,9 +363,9 @@
       ]
     },
     "node_modules/@lancedb/vectordb-linux-x64-gnu": {
-      "version": "0.20.1-beta.1",
+      "version": "0.21.0",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.20.1-beta.1.tgz",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-linux-x64-gnu/-/vectordb-linux-x64-gnu-0.21.0.tgz",
-      "integrity": "sha512-rq7Q6Lq9kJmBcgwplYQVJmRbyeP+xPVmXyyQfAO3IjekqeSsyjj1HoCZYqZIfBZyN5ELiSvIJB0731aKf9pr1A==",
+      "integrity": "sha512-t7dkFV6kga3rqXR1rH460GdpSVuY0tw7CIc0KqsIIkBcXzUPA1n0QDoazdwPQ1MXzG/+F5WWCTp3dYWx2vP0Lw==",
       "cpu": [
         "x64"
       ],
@@ -375,9 +375,9 @@
       ]
     },
     "node_modules/@lancedb/vectordb-win32-x64-msvc": {
-      "version": "0.20.1-beta.1",
+      "version": "0.21.0",
-      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.20.1-beta.1.tgz",
+      "resolved": "https://registry.npmjs.org/@lancedb/vectordb-win32-x64-msvc/-/vectordb-win32-x64-msvc-0.21.0.tgz",
-      "integrity": "sha512-kHra0SEXeMKdgqi5h0igsqHcBr73hKBhEVJBa8VTv1DUv6Jvazwl4B4ueqllcyD4k3vvOTb2XzZomm7dhQ9QnA==",
+      "integrity": "sha512-yovkW61RECBTsu0S527BX1uW0jCAZK9MAsJTknXmDjp78figx4/AyI5ajT63u/Uo4EKoheeNiiLdyU4v+A9YVw==",
       "cpu": [
         "x64"
       ],

node/package.json

@@ -1,6 +1,6 @@
 {
   "name": "vectordb",
-  "version": "0.20.1-beta.1",
+  "version": "0.21.0",
   "description": " Serverless, low-latency vector database for AI applications",
   "private": false,
   "main": "dist/index.js",
@@ -89,10 +89,10 @@
     }
   },
   "optionalDependencies": {
-    "@lancedb/vectordb-darwin-x64": "0.20.1-beta.1",
+    "@lancedb/vectordb-darwin-x64": "0.21.0",
-    "@lancedb/vectordb-darwin-arm64": "0.20.1-beta.1",
+    "@lancedb/vectordb-darwin-arm64": "0.21.0",
-    "@lancedb/vectordb-linux-x64-gnu": "0.20.1-beta.1",
+    "@lancedb/vectordb-linux-x64-gnu": "0.21.0",
-    "@lancedb/vectordb-linux-arm64-gnu": "0.20.1-beta.1",
+    "@lancedb/vectordb-linux-arm64-gnu": "0.21.0",
-    "@lancedb/vectordb-win32-x64-msvc": "0.20.1-beta.1"
+    "@lancedb/vectordb-win32-x64-msvc": "0.21.0"
   }
 }

nodejs/Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "lancedb-nodejs"
 edition.workspace = true
-version = "0.20.1-beta.1"
+version = "0.21.0"
 license.workspace = true
 description.workspace = true
 repository.workspace = true

nodejs/__test__/table.test.ts

@@ -368,9 +368,9 @@ describe("merge insert", () => {
       { a: 4, b: "z" },
     ];
 
-    expect(
+    const result = (await table.toArrow()).toArray().sort((a, b) => a.a - b.a);
-      JSON.parse(JSON.stringify((await table.toArrow()).toArray())),
+
-    ).toEqual(expected);
+    expect(result.map((row) => ({ ...row }))).toEqual(expected);
   });
   test("conditional update", async () => {
     const newData = [
@@ -1650,13 +1650,25 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
       expect(resultSet.has("fob")).toBe(true);
       expect(resultSet.has("fo")).toBe(true);
       expect(resultSet.has("food")).toBe(true);
+
+      const prefixResults = await table
+        .search(
+          new MatchQuery("foo", "text", { fuzziness: 3, prefixLength: 3 }),
+        )
+        .toArray();
+      expect(prefixResults.length).toBe(2);
+      const resultSet2 = new Set(prefixResults.map((r) => r.text));
+      expect(resultSet2.has("foo")).toBe(true);
+      expect(resultSet2.has("food")).toBe(true);
     });
 
     test("full text search boolean query", async () => {
       const db = await connect(tmpDir.name);
       const data = [
-        { text: "hello world", vector: [0.1, 0.2, 0.3] },
+        { text: "The cat and dog are playing" },
-        { text: "goodbye world", vector: [0.4, 0.5, 0.6] },
+        { text: "The cat is sleeping" },
+        { text: "The dog is barking" },
+        { text: "The dog chases the cat" },
       ];
       const table = await db.createTable("test", data);
       await table.createIndex("text", {
@@ -1666,22 +1678,32 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
       const shouldResults = await table
         .search(
           new BooleanQuery([
-            [Occur.Should, new MatchQuery("hello", "text")],
+            [Occur.Should, new MatchQuery("cat", "text")],
-            [Occur.Should, new MatchQuery("goodbye", "text")],
+            [Occur.Should, new MatchQuery("dog", "text")],
           ]),
         )
         .toArray();
-      expect(shouldResults.length).toBe(2);
+      expect(shouldResults.length).toBe(4);
 
       const mustResults = await table
         .search(
           new BooleanQuery([
-            [Occur.Must, new MatchQuery("hello", "text")],
+            [Occur.Must, new MatchQuery("cat", "text")],
-            [Occur.Must, new MatchQuery("world", "text")],
+            [Occur.Must, new MatchQuery("dog", "text")],
           ]),
         )
         .toArray();
-      expect(mustResults.length).toBe(1);
+      expect(mustResults.length).toBe(2);
+
+      const mustNotResults = await table
+        .search(
+          new BooleanQuery([
+            [Occur.Must, new MatchQuery("cat", "text")],
+            [Occur.MustNot, new MatchQuery("dog", "text")],
+          ]),
+        )
+        .toArray();
+      expect(mustNotResults.length).toBe(1);
     });
 
     test.each([

nodejs/lancedb/query.ts

@@ -812,10 +812,12 @@ export enum Operator {
  *
  * - `Must`: The term must be present in the document.
  * - `Should`: The term should contribute to the document score, but is not required.
+ * - `MustNot`: The term must not be present in the document.
  */
 export enum Occur {
-  Must = "MUST",
   Should = "SHOULD",
+  Must = "MUST",
+  MustNot = "MUST_NOT",
 }
 
 /**
@@ -856,6 +858,7 @@ export class MatchQuery implements FullTextQuery {
    *   - `fuzziness`: The fuzziness level for the query (default is 0).
    *   - `maxExpansions`: The maximum number of terms to consider for fuzzy matching (default is 50).
    *   - `operator`: The logical operator to use for combining terms in the query (default is "OR").
+   *   - `prefixLength`: The number of beginning characters being unchanged for fuzzy matching.
    */
   constructor(
     query: string,
@@ -865,6 +868,7 @@ export class MatchQuery implements FullTextQuery {
       fuzziness?: number;
       maxExpansions?: number;
       operator?: Operator;
+      prefixLength?: number;
     },
   ) {
     let fuzziness = options?.fuzziness;
@@ -878,6 +882,7 @@ export class MatchQuery implements FullTextQuery {
       fuzziness,
       options?.maxExpansions ?? 50,
       options?.operator ?? Operator.Or,
+      options?.prefixLength ?? 0,
     );
   }
 

nodejs/npm/darwin-arm64/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-darwin-arm64",
-	"version": "0.20.1-beta.1",
+	"version": "0.21.0",
 	"os": ["darwin"],
 	"cpu": ["arm64"],
 	"main": "lancedb.darwin-arm64.node",

nodejs/npm/darwin-x64/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-darwin-x64",
-	"version": "0.20.1-beta.1",
+	"version": "0.21.0",
 	"os": ["darwin"],
 	"cpu": ["x64"],
 	"main": "lancedb.darwin-x64.node",

nodejs/npm/linux-arm64-gnu/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-arm64-gnu",
-	"version": "0.20.1-beta.1",
+	"version": "0.21.0",
 	"os": ["linux"],
 	"cpu": ["arm64"],
 	"main": "lancedb.linux-arm64-gnu.node",

nodejs/npm/linux-arm64-musl/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-arm64-musl",
-	"version": "0.20.1-beta.1",
+	"version": "0.21.0",
 	"os": ["linux"],
 	"cpu": ["arm64"],
 	"main": "lancedb.linux-arm64-musl.node",

nodejs/npm/linux-x64-gnu/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-x64-gnu",
-	"version": "0.20.1-beta.1",
+	"version": "0.21.0",
 	"os": ["linux"],
 	"cpu": ["x64"],
 	"main": "lancedb.linux-x64-gnu.node",

nodejs/npm/linux-x64-musl/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-x64-musl",
-	"version": "0.20.1-beta.1",
+	"version": "0.21.0",
 	"os": ["linux"],
 	"cpu": ["x64"],
 	"main": "lancedb.linux-x64-musl.node",

nodejs/npm/win32-arm64-msvc/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@lancedb/lancedb-win32-arm64-msvc",
-  "version": "0.20.1-beta.1",
+  "version": "0.21.0",
   "os": [
     "win32"
   ],

nodejs/npm/win32-x64-msvc/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-win32-x64-msvc",
-	"version": "0.20.1-beta.1",
+	"version": "0.21.0",
 	"os": ["win32"],
 	"cpu": ["x64"],
 	"main": "lancedb.win32-x64-msvc.node",

nodejs/package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "@lancedb/lancedb",
-  "version": "0.20.1-beta.1",
+  "version": "0.21.0",
   "lockfileVersion": 3,
   "requires": true,
   "packages": {
     "": {
       "name": "@lancedb/lancedb",
-      "version": "0.20.1-beta.1",
+      "version": "0.21.0",
       "cpu": [
         "x64",
         "arm64"

nodejs/package.json

@@ -11,7 +11,7 @@
     "ann"
   ],
   "private": false,
-  "version": "0.20.1-beta.1",
+  "version": "0.21.0",
   "main": "dist/index.js",
   "exports": {
     ".": "./dist/index.js",

nodejs/src/query.rs

@@ -335,6 +335,7 @@ impl JsFullTextQuery {
         fuzziness: Option<u32>,
         max_expansions: u32,
         operator: String,
+        prefix_length: u32,
     ) -> napi::Result<Self> {
         Ok(Self {
             inner: MatchQuery::new(query)
@@ -347,6 +348,7 @@ impl JsFullTextQuery {
                         napi::Error::from_reason(format!("Invalid operator: {}", e))
                     })?,
                 )
+                .with_prefix_length(prefix_length)
                 .into(),
         })
     }

python/.bumpversion.toml

@@ -1,5 +1,5 @@
 [tool.bumpversion]
-current_version = "0.23.1-beta.2"
+current_version = "0.24.1-beta.0"
 parse = """(?x)
     (?P<major>0|[1-9]\\d*)\\.
     (?P<minor>0|[1-9]\\d*)\\.

python/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "lancedb-python"
-version = "0.23.1-beta.2"
+version = "0.24.1-beta.0"
 edition.workspace = true
 description = "Python bindings for LanceDB"
 license.workspace = true

python/pyproject.toml

@@ -85,7 +85,7 @@ embeddings = [
     "boto3>=1.28.57",
     "awscli>=1.29.57",
     "botocore>=1.31.57",
-    "ollama",
+    "ollama>=0.3.0",
     "ibm-watsonx-ai>=1.1.2",
 ]
 azure = ["adlfs>=2024.2.0"]

python/python/lancedb/embeddings/ollama.py

@@ -2,14 +2,15 @@
 # SPDX-FileCopyrightText: Copyright The LanceDB Authors
 
 from functools import cached_property
-from typing import TYPE_CHECKING, List, Optional, Union
+from typing import TYPE_CHECKING, List, Optional, Sequence, Union
+
+import numpy as np
 
 from ..util import attempt_import_or_raise
 from .base import TextEmbeddingFunction
 from .registry import register
 
 if TYPE_CHECKING:
-    import numpy as np
     import ollama
 
 
@@ -28,23 +29,21 @@ class OllamaEmbeddings(TextEmbeddingFunction):
     keep_alive: Optional[Union[float, str]] = None
     ollama_client_kwargs: Optional[dict] = {}
 
-    def ndims(self):
+    def ndims(self) -> int:
         return len(self.generate_embeddings(["foo"])[0])
 
-    def _compute_embedding(self, text) -> Union["np.array", None]:
+    def _compute_embedding(self, text: Sequence[str]) -> Sequence[Sequence[float]]:
-        return (
+        response = self._ollama_client.embed(
-            self._ollama_client.embeddings(
+            model=self.name,
-                model=self.name,
+            input=text,
-                prompt=text,
+            options=self.options,
-                options=self.options,
+            keep_alive=self.keep_alive,
-                keep_alive=self.keep_alive,
-            )["embedding"]
-            or None
         )
+        return response.embeddings
 
     def generate_embeddings(
-        self, texts: Union[List[str], "np.ndarray"]
+        self, texts: Union[List[str], np.ndarray]
-    ) -> list[Union["np.array", None]]:
+    ) -> list[Union[np.array, None]]:
         """
         Get the embeddings for the given texts
 
@@ -54,8 +53,8 @@ class OllamaEmbeddings(TextEmbeddingFunction):
             The texts to embed
         """
         # TODO retry, rate limit, token limit
-        embeddings = [self._compute_embedding(text) for text in texts]
+        embeddings = self._compute_embedding(texts)
-        return embeddings
+        return list(embeddings)
 
     @cached_property
     def _ollama_client(self) -> "ollama.Client":

python/python/lancedb/query.py

@@ -101,8 +101,9 @@ class FullTextOperator(str, Enum):
 
 
 class Occur(str, Enum):
-    MUST = "MUST"
     SHOULD = "SHOULD"
+    MUST = "MUST"
+    MUST_NOT = "MUST_NOT"
 
 
 @pydantic.dataclasses.dataclass
@@ -181,6 +182,9 @@ class MatchQuery(FullTextQuery):
         Can be either `AND` or `OR`.
         If `AND`, all terms in the query must match.
         If `OR`, at least one term in the query must match.
+    prefix_length : int, optional
+        The number of beginning characters being unchanged for fuzzy matching.
+        This is useful to achieve prefix matching.
     """
 
     query: str
@@ -189,6 +193,7 @@ class MatchQuery(FullTextQuery):
     fuzziness: int = pydantic.Field(0, kw_only=True)
     max_expansions: int = pydantic.Field(50, kw_only=True)
     operator: FullTextOperator = pydantic.Field(FullTextOperator.OR, kw_only=True)
+    prefix_length: int = pydantic.Field(0, kw_only=True)
 
     def query_type(self) -> FullTextQueryType:
         return FullTextQueryType.MATCH
@@ -1446,10 +1451,13 @@ class LanceFtsQueryBuilder(LanceQueryBuilder):
 
         query = self._query
         if self._phrase_query:
-            raise NotImplementedError(
+            if isinstance(query, str):
-                "Phrase query is not yet supported in Lance FTS. "
+                if not query.startswith('"') or not query.endswith('"'):
-                "Use tantivy-based index instead for now."
+                    query = f'"{query}"'
-            )
+            elif isinstance(query, FullTextQuery) and not isinstance(
+                query, PhraseQuery
+            ):
+                raise TypeError("Please use PhraseQuery for phrase queries.")
         query = self.to_query_object()
         results = self._table._execute_query(query, timeout=timeout)
         results = results.read_all()
@@ -3034,15 +3042,21 @@ class AsyncHybridQuery(AsyncQueryBase, AsyncVectorQueryBase):
         >>> asyncio.run(doctest_example()) # doctest: +ELLIPSIS, +NORMALIZE_WHITESPACE
         Vector Search Plan:
         ProjectionExec: expr=[vector@0 as vector, text@3 as text, _distance@2 as _distance]
-            Take: columns="vector, _rowid, _distance, (text)"
+          Take: columns="vector, _rowid, _distance, (text)"
-                CoalesceBatchesExec: target_batch_size=1024
+            CoalesceBatchesExec: target_batch_size=1024
-                GlobalLimitExec: skip=0, fetch=10
+              GlobalLimitExec: skip=0, fetch=10
-                    FilterExec: _distance@2 IS NOT NULL
+                FilterExec: _distance@2 IS NOT NULL
-                    SortExec: TopK(fetch=10), expr=[_distance@2 ASC NULLS LAST], preserve_partitioning=[false]
+                  SortExec: TopK(fetch=10), expr=[_distance@2 ASC NULLS LAST], preserve_partitioning=[false]
-                        KNNVectorDistance: metric=l2
+                    KNNVectorDistance: metric=l2
-                        LanceScan: uri=..., projection=[vector], row_id=true, row_addr=false, ordered=false
+                      LanceScan: uri=..., projection=[vector], row_id=true, row_addr=false, ordered=false
+        <BLANKLINE>
         FTS Search Plan:
-        LanceScan: uri=..., projection=[vector, text], row_id=false, row_addr=false, ordered=true
+        ProjectionExec: expr=[vector@2 as vector, text@3 as text, _score@1 as _score]
+          Take: columns="_rowid, _score, (vector), (text)"
+            CoalesceBatchesExec: target_batch_size=1024
+              GlobalLimitExec: skip=0, fetch=10
+                MatchQuery: query=hello
+        <BLANKLINE>
 
         Parameters
         ----------

python/python/lancedb/table.py

@@ -827,7 +827,7 @@ class Table(ABC):
         ordering_field_names: Optional[Union[str, List[str]]] = None,
         replace: bool = False,
         writer_heap_size: Optional[int] = 1024 * 1024 * 1024,
-        use_tantivy: bool = True,
+        use_tantivy: bool = False,
         tokenizer_name: Optional[str] = None,
         with_position: bool = False,
         # tokenizer configs:
@@ -864,7 +864,7 @@ class Table(ABC):
             The tokenizer to use for the index. Can be "raw", "default" or the 2 letter
             language code followed by "_stem". So for english it would be "en_stem".
             For available languages see: https://docs.rs/tantivy/latest/tantivy/tokenizer/enum.Language.html
-        use_tantivy: bool, default True
+        use_tantivy: bool, default False
             If True, use the legacy full-text search implementation based on tantivy.
             If False, use the new full-text search implementation based on lance-index.
         with_position: bool, default False
@@ -1970,7 +1970,7 @@ class LanceTable(Table):
         ordering_field_names: Optional[Union[str, List[str]]] = None,
         replace: bool = False,
         writer_heap_size: Optional[int] = 1024 * 1024 * 1024,
-        use_tantivy: bool = True,
+        use_tantivy: bool = False,
         tokenizer_name: Optional[str] = None,
         with_position: bool = False,
         # tokenizer configs:

python/python/tests/docs/test_search.py

@@ -6,7 +6,7 @@ import lancedb
 
 # --8<-- [end:import-lancedb]
 # --8<-- [start:import-numpy]
-from lancedb.query import BoostQuery, MatchQuery
+from lancedb.query import BooleanQuery, BoostQuery, MatchQuery, Occur
 import numpy as np
 import pyarrow as pa
 
@@ -191,6 +191,15 @@ def test_fts_fuzzy_query():
         "food",  # 1 insertion
     }
 
+    results = table.search(
+        MatchQuery("foo", "text", fuzziness=1, prefix_length=3)
+    ).to_pandas()
+    assert len(results) == 2
+    assert set(results["text"].to_list()) == {
+        "foo",
+        "food",
+    }
+
 
 @pytest.mark.skipif(
     os.name == "nt", reason="Need to fix https://github.com/lancedb/lance/issues/3905"
@@ -240,6 +249,60 @@ def test_fts_boost_query():
     )
 
 
+@pytest.mark.skipif(
+    os.name == "nt", reason="Need to fix https://github.com/lancedb/lance/issues/3905"
+)
+def test_fts_boolean_query(tmp_path):
+    uri = tmp_path / "boolean-example"
+    db = lancedb.connect(uri)
+    table = db.create_table(
+        "my_table_fts_boolean",
+        data=[
+            {"text": "The cat and dog are playing"},
+            {"text": "The cat is sleeping"},
+            {"text": "The dog is barking"},
+            {"text": "The dog chases the cat"},
+        ],
+        mode="overwrite",
+    )
+    table.create_fts_index("text", use_tantivy=False, replace=True)
+
+    # SHOULD
+    results = table.search(
+        MatchQuery("cat", "text") | MatchQuery("dog", "text")
+    ).to_pandas()
+    assert len(results) == 4
+    assert set(results["text"].to_list()) == {
+        "The cat and dog are playing",
+        "The cat is sleeping",
+        "The dog is barking",
+        "The dog chases the cat",
+    }
+    # MUST
+    results = table.search(
+        MatchQuery("cat", "text") & MatchQuery("dog", "text")
+    ).to_pandas()
+    assert len(results) == 2
+    assert set(results["text"].to_list()) == {
+        "The cat and dog are playing",
+        "The dog chases the cat",
+    }
+
+    # MUST NOT
+    results = table.search(
+        BooleanQuery(
+            [
+                (Occur.MUST, MatchQuery("cat", "text")),
+                (Occur.MUST_NOT, MatchQuery("dog", "text")),
+            ]
+        )
+    ).to_pandas()
+    assert len(results) == 1
+    assert set(results["text"].to_list()) == {
+        "The cat is sleeping",
+    }
+
+
 @pytest.mark.skipif(
     os.name == "nt", reason="Need to fix https://github.com/lancedb/lance/issues/3905"
 )

python/python/tests/test_query.py

@@ -775,6 +775,82 @@ async def test_explain_plan_async(table_async: AsyncTable):
     assert "KNN" in plan
 
 
+@pytest.mark.asyncio
+async def test_explain_plan_fts(table_async: AsyncTable):
+    """Test explain plan for FTS queries"""
+    # Create FTS index
+    from lancedb.index import FTS
+
+    await table_async.create_index("text", config=FTS())
+
+    # Test pure FTS query
+    query = await table_async.search("dog", query_type="fts", fts_columns="text")
+    plan = await query.explain_plan()
+    # Should show FTS details (issue #2465 is now fixed)
+    assert "MatchQuery: query=dog" in plan
+    assert "GlobalLimitExec" in plan  # Default limit
+
+    # Test FTS query with limit
+    query_with_limit = await table_async.search(
+        "dog", query_type="fts", fts_columns="text"
+    )
+    plan_with_limit = await query_with_limit.limit(1).explain_plan()
+    assert "MatchQuery: query=dog" in plan_with_limit
+    assert "GlobalLimitExec: skip=0, fetch=1" in plan_with_limit
+
+    # Test FTS query with offset and limit
+    query_with_offset = await table_async.search(
+        "dog", query_type="fts", fts_columns="text"
+    )
+    plan_with_offset = await query_with_offset.offset(1).limit(1).explain_plan()
+    assert "MatchQuery: query=dog" in plan_with_offset
+    assert "GlobalLimitExec: skip=1, fetch=1" in plan_with_offset
+
+
+@pytest.mark.asyncio
+async def test_explain_plan_vector_with_limit_offset(table_async: AsyncTable):
+    """Test explain plan for vector queries with limit and offset"""
+    # Test vector query with limit
+    plan_with_limit = await (
+        table_async.query().nearest_to(pa.array([1, 2])).limit(1).explain_plan()
+    )
+    assert "KNN" in plan_with_limit
+    assert "GlobalLimitExec: skip=0, fetch=1" in plan_with_limit
+
+    # Test vector query with offset and limit
+    plan_with_offset = await (
+        table_async.query()
+        .nearest_to(pa.array([1, 2]))
+        .offset(1)
+        .limit(1)
+        .explain_plan()
+    )
+    assert "KNN" in plan_with_offset
+    assert "GlobalLimitExec: skip=1, fetch=1" in plan_with_offset
+
+
+@pytest.mark.asyncio
+async def test_explain_plan_with_filters(table_async: AsyncTable):
+    """Test explain plan for queries with filters"""
+    # Test vector query with filter
+    plan_with_filter = await (
+        table_async.query().nearest_to(pa.array([1, 2])).where("id = 1").explain_plan()
+    )
+    assert "KNN" in plan_with_filter
+    assert "FilterExec" in plan_with_filter
+
+    # Test FTS query with filter
+    from lancedb.index import FTS
+
+    await table_async.create_index("text", config=FTS())
+    query_fts_filter = await table_async.search(
+        "dog", query_type="fts", fts_columns="text"
+    )
+    plan_fts_filter = await query_fts_filter.where("id = 1").explain_plan()
+    assert "MatchQuery: query=dog" in plan_fts_filter
+    assert "FilterExec: id@" in plan_fts_filter  # Should show filter details
+
+
 @pytest.mark.asyncio
 async def test_query_camelcase_async(tmp_path):
     db = await lancedb.connect_async(tmp_path)

python/python/tests/test_s3.py

@@ -245,7 +245,7 @@ def test_s3_dynamodb_sync(s3_bucket: str, commit_table: str, monkeypatch):
         NotImplementedError,
         match="Full-text search is only supported on the local filesystem",
     ):
-        table.create_fts_index("x")
+        table.create_fts_index("x", use_tantivy=True)
 
     # make sure list tables still works
     assert db.table_names() == ["test_ddb_sync"]

python/src/query.rs

@@ -50,8 +50,9 @@ impl FromPyObject<'_> for PyLanceDB<FtsQuery> {
                 let fuzziness = ob.getattr("fuzziness")?.extract()?;
                 let max_expansions = ob.getattr("max_expansions")?.extract()?;
                 let operator = ob.getattr("operator")?.extract::<String>()?;
+                let prefix_length = ob.getattr("prefix_length")?.extract()?;
 
-                Ok(PyLanceDB(
+                Ok(Self(
                     MatchQuery::new(query)
                         .with_column(Some(column))
                         .with_boost(boost)
@@ -60,6 +61,7 @@ impl FromPyObject<'_> for PyLanceDB<FtsQuery> {
                         .with_operator(Operator::try_from(operator.as_str()).map_err(|e| {
                             PyValueError::new_err(format!("Invalid operator: {}", e))
                         })?)
+                        .with_prefix_length(prefix_length)
                         .into(),
                 ))
             }
@@ -68,7 +70,7 @@ impl FromPyObject<'_> for PyLanceDB<FtsQuery> {
                 let column = ob.getattr("column")?.extract()?;
                 let slop = ob.getattr("slop")?.extract()?;
 
-                Ok(PyLanceDB(
+                Ok(Self(
                     PhraseQuery::new(query)
                         .with_column(Some(column))
                         .with_slop(slop)
@@ -76,10 +78,10 @@ impl FromPyObject<'_> for PyLanceDB<FtsQuery> {
                 ))
             }
             "BoostQuery" => {
-                let positive: PyLanceDB<FtsQuery> = ob.getattr("positive")?.extract()?;
+                let positive: Self = ob.getattr("positive")?.extract()?;
-                let negative: PyLanceDB<FtsQuery> = ob.getattr("negative")?.extract()?;
+                let negative: Self = ob.getattr("negative")?.extract()?;
                 let negative_boost = ob.getattr("negative_boost")?.extract()?;
-                Ok(PyLanceDB(
+                Ok(Self(
                     BoostQuery::new(positive.0, negative.0, negative_boost).into(),
                 ))
             }
@@ -101,18 +103,17 @@ impl FromPyObject<'_> for PyLanceDB<FtsQuery> {
                 let op = Operator::try_from(operator.as_str())
                     .map_err(|e| PyValueError::new_err(format!("Invalid operator: {}", e)))?;
 
-                Ok(PyLanceDB(q.with_operator(op).into()))
+                Ok(Self(q.with_operator(op).into()))
             }
             "BooleanQuery" => {
-                let queries: Vec<(String, PyLanceDB<FtsQuery>)> =
+                let queries: Vec<(String, Self)> = ob.getattr("queries")?.extract()?;
-                    ob.getattr("queries")?.extract()?;
                 let mut sub_queries = Vec::with_capacity(queries.len());
                 for (occur, q) in queries {
                     let occur = Occur::try_from(occur.as_str())
                         .map_err(|e| PyValueError::new_err(e.to_string()))?;
                     sub_queries.push((occur, q.0));
                 }
-                Ok(PyLanceDB(BooleanQuery::new(sub_queries).into()))
+                Ok(Self(BooleanQuery::new(sub_queries).into()))
             }
             name => Err(PyValueError::new_err(format!(
                 "Unsupported FTS query type: {}",
@@ -139,7 +140,8 @@ impl<'py> IntoPyObject<'py> for PyLanceDB<FtsQuery> {
                 kwargs.set_item("boost", query.boost)?;
                 kwargs.set_item("fuzziness", query.fuzziness)?;
                 kwargs.set_item("max_expansions", query.max_expansions)?;
-                kwargs.set_item("operator", operator_to_str(query.operator))?;
+                kwargs.set_item::<_, &str>("operator", query.operator.into())?;
+                kwargs.set_item("prefix_length", query.prefix_length)?;
                 namespace
                     .getattr(intern!(py, "MatchQuery"))?
                     .call((query.terms, query.column.unwrap()), Some(&kwargs))
@@ -152,8 +154,8 @@ impl<'py> IntoPyObject<'py> for PyLanceDB<FtsQuery> {
                     .call((query.terms, query.column.unwrap()), Some(&kwargs))
             }
             FtsQuery::Boost(query) => {
-                let positive = PyLanceDB(query.positive.as_ref().clone()).into_pyobject(py)?;
+                let positive = Self(query.positive.as_ref().clone()).into_pyobject(py)?;
-                let negative = PyLanceDB(query.negative.as_ref().clone()).into_pyobject(py)?;
+                let negative = Self(query.negative.as_ref().clone()).into_pyobject(py)?;
                 let kwargs = PyDict::new(py);
                 kwargs.set_item("negative_boost", query.negative_boost)?;
                 namespace
@@ -169,19 +171,25 @@ impl<'py> IntoPyObject<'py> for PyLanceDB<FtsQuery> {
                     .unzip();
                 let kwargs = PyDict::new(py);
                 kwargs.set_item("boosts", boosts)?;
-                kwargs.set_item("operator", operator_to_str(first.operator))?;
+                kwargs.set_item::<_, &str>("operator", first.operator.into())?;
                 namespace
                     .getattr(intern!(py, "MultiMatchQuery"))?
                     .call((first.terms.clone(), columns), Some(&kwargs))
             }
             FtsQuery::Boolean(query) => {
-                let mut queries = Vec::with_capacity(query.must.len() + query.should.len());
+                let mut queries: Vec<(&str, Bound<'py, PyAny>)> = Vec::with_capacity(
-                for q in query.must {
+                    query.should.len() + query.must.len() + query.must_not.len(),
-                    queries.push((occur_to_str(Occur::Must), PyLanceDB(q).into_pyobject(py)?));
+                );
-                }
                 for q in query.should {
-                    queries.push((occur_to_str(Occur::Should), PyLanceDB(q).into_pyobject(py)?));
+                    queries.push((Occur::Should.into(), Self(q).into_pyobject(py)?));
                 }
+                for q in query.must {
+                    queries.push((Occur::Must.into(), Self(q).into_pyobject(py)?));
+                }
+                for q in query.must_not {
+                    queries.push((Occur::MustNot.into(), Self(q).into_pyobject(py)?));
+                }
+
                 namespace
                     .getattr(intern!(py, "BooleanQuery"))?
                     .call1((queries,))
@@ -190,21 +198,6 @@ impl<'py> IntoPyObject<'py> for PyLanceDB<FtsQuery> {
     }
 }
 
-fn operator_to_str(op: Operator) -> &'static str {
-    match op {
-        Operator::And => "AND",
-        Operator::Or => "OR",
-    }
-}
-
-fn occur_to_str(occur: Occur) -> &'static str {
-    match occur {
-        Occur::Must => "MUST",
-        Occur::Should => "SHOULD",
-        Occur::MustNot => "MUST NOT",
-    }
-}
-
 // Python representation of query vector(s)
 #[derive(Clone)]
 pub struct PyQueryVectors(Vec<Arc<dyn Array>>);
@@ -569,7 +562,10 @@ impl FTSQuery {
     }
 
     pub fn explain_plan(self_: PyRef<'_, Self>, verbose: bool) -> PyResult<Bound<'_, PyAny>> {
-        let inner = self_.inner.clone();
+        let inner = self_
+            .inner
+            .clone()
+            .full_text_search(self_.fts_query.clone());
         future_into_py(self_.py(), async move {
             inner
                 .explain_plan(verbose)
@@ -579,7 +575,10 @@ impl FTSQuery {
     }
 
     pub fn analyze_plan(self_: PyRef<'_, Self>) -> PyResult<Bound<'_, PyAny>> {
-        let inner = self_.inner.clone();
+        let inner = self_
+            .inner
+            .clone()
+            .full_text_search(self_.fts_query.clone());
         future_into_py(self_.py(), async move {
             inner
                 .analyze_plan()

rust/ffi/node/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "lancedb-node"
-version = "0.20.1-beta.1"
+version = "0.21.0"
 description = "Serverless, low-latency vector database for AI applications"
 license.workspace = true
 edition.workspace = true

rust/lancedb/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "lancedb"
-version = "0.20.1-beta.1"
+version = "0.21.0"
 edition.workspace = true
 description = "LanceDB: A serverless, low-latency vector database for AI applications"
 license.workspace = true

rust/lancedb/src/catalog/listing.rs

@@ -105,7 +105,7 @@ impl ListingCatalog {
     }
 
     async fn open_path(path: &str) -> Result<Self> {
-        let (object_store, base_path) = ObjectStore::from_uri(path).await.unwrap();
+        let (object_store, base_path) = ObjectStore::from_uri(path).await?;
         if object_store.is_local() {
             Self::try_create_dir(path).context(CreateDirSnafu { path })?;
         }

rust/lancedb/src/io/object_store.rs

@@ -107,7 +107,7 @@ impl ObjectStore for MirroringObjectStore {
         self.primary.delete(location).await
     }
 
-    fn list(&self, prefix: Option<&Path>) -> BoxStream<'_, Result<ObjectMeta>> {
+    fn list(&self, prefix: Option<&Path>) -> BoxStream<'static, Result<ObjectMeta>> {
         self.primary.list(prefix)
     }
 

rust/lancedb/src/io/object_store/io_tracking.rs

@@ -119,7 +119,7 @@ impl ObjectStore for IoTrackingStore {
         let result = self.target.get(location).await;
         if let Ok(result) = &result {
             let num_bytes = result.range.end - result.range.start;
-            self.record_read(num_bytes as u64);
+            self.record_read(num_bytes);
         }
         result
     }
@@ -128,12 +128,12 @@ impl ObjectStore for IoTrackingStore {
         let result = self.target.get_opts(location, options).await;
         if let Ok(result) = &result {
             let num_bytes = result.range.end - result.range.start;
-            self.record_read(num_bytes as u64);
+            self.record_read(num_bytes);
         }
         result
     }
 
-    async fn get_range(&self, location: &Path, range: std::ops::Range<usize>) -> OSResult<Bytes> {
+    async fn get_range(&self, location: &Path, range: std::ops::Range<u64>) -> OSResult<Bytes> {
         let result = self.target.get_range(location, range).await;
         if let Ok(result) = &result {
             self.record_read(result.len() as u64);
@@ -144,7 +144,7 @@ impl ObjectStore for IoTrackingStore {
     async fn get_ranges(
         &self,
         location: &Path,
-        ranges: &[std::ops::Range<usize>],
+        ranges: &[std::ops::Range<u64>],
     ) -> OSResult<Vec<Bytes>> {
         let result = self.target.get_ranges(location, ranges).await;
         if let Ok(result) = &result {
@@ -170,7 +170,7 @@ impl ObjectStore for IoTrackingStore {
         self.target.delete_stream(locations)
     }
 
-    fn list(&self, prefix: Option<&Path>) -> BoxStream<'_, OSResult<ObjectMeta>> {
+    fn list(&self, prefix: Option<&Path>) -> BoxStream<'static, OSResult<ObjectMeta>> {
         self.record_read(0);
         self.target.list(prefix)
     }
@@ -179,7 +179,7 @@ impl ObjectStore for IoTrackingStore {
         &self,
         prefix: Option<&Path>,
         offset: &Path,
-    ) -> BoxStream<'_, OSResult<ObjectMeta>> {
+    ) -> BoxStream<'static, OSResult<ObjectMeta>> {
         self.record_read(0);
         self.target.list_with_offset(prefix, offset)
     }

rust/lancedb/src/table/datafusion.rs

@@ -392,9 +392,18 @@ pub mod tests {
                 } else {
                     expected_line.trim()
                 };
-                assert_eq!(&actual_trimmed[..expected_trimmed.len()], expected_trimmed);
+                assert_eq!(
+                    &actual_trimmed[..expected_trimmed.len()],
+                    expected_trimmed,
+                    "\nactual:\n{physical_plan}\nexpected:\n{expected}"
+                );
             }
-            assert_eq!(lines_checked, expected.lines().count());
+            assert_eq!(
+                lines_checked,
+                expected.lines().count(),
+                "\nlines_checked:\n{lines_checked}\nexpected:\n{}",
+                expected.lines().count()
+            );
         }
     }
 
@@ -477,9 +486,9 @@ pub mod tests {
         TestFixture::check_plan(
             plan,
             "MetadataEraserExec
-             RepartitionExec:...
              CoalesceBatchesExec:...
              FilterExec: i@0 >= 5
+             RepartitionExec:...
              ProjectionExec:...
              LanceScan:...",
         )

rust/lancedb/src/table/dataset.rs

@@ -129,7 +129,9 @@ impl DatasetRef {
                 dataset: ref mut ds,
                 ..
             } => {
-                *ds = dataset;
+                if dataset.manifest().version > ds.manifest().version {
+                    *ds = dataset;
+                }
             }
             _ => unreachable!("Dataset should be in latest mode at this point"),
         }