refactor(secrets): one binding list with a kind, and split the secret writes

`secret_env_bindings` took the general noun for one delivery mode. A binding is
the concept; how it arrives is a property of one. `secret_bindings` is a list of
`SecretBinding`, tagged by `kind`, so a later mode is a variant rather than a
sibling field -- and the rules that are per-Function, like how many Secrets it
may bind, stay answerable from one place.

The cost of one field is that an unknown kind is a decode error unless it is
caught. It is caught, the way `PythonRuntimeSpec` catches an unknown runtime:
`Unrecognized { kind }`, `#[non_exhaustive]`, and the payload dropped rather
than retained because the client does not proxy catalog values. A test pins it
-- a `file` binding from a newer server decodes, reports its kind, and
round-trips as its discriminator without failing the version around it.

A list has no key order to inherit, and the list is in the version hash, so
`bind_secrets` sorts it: a caller's argument order is not part of what a
Function is.

The Secret is named under `secret_ref`, not `secret`: the service scans Job
payloads for credential-shaped keys and refuses one called `secret` whatever it
holds. That guard is worth more blunt than argued with.

`create_secret` and `alter_secret` also stop sharing a request shape. They are
different operations to the service -- one refuses an existing name, the other
requires it -- and either may grow a field the other has no meaning for. What
they share is posting a body that must not be logged, which is a function.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01UfmeJ533rQDnPBkMtjerV6

.bumpversion.toml

@@ -1,5 +1,5 @@
 [tool.bumpversion]
-current_version = "0.38.0-beta.11"
+current_version = "0.39.0-beta.4"
 parse = """(?x)
     (?P<major>0|[1-9]\\d*)\\.
     (?P<minor>0|[1-9]\\d*)\\.

.github/dependabot.yml

@@ -44,3 +44,27 @@ updates:
       python-deps:
         patterns:
           - "*"
+
+  # The npm ecosystem covers pnpm lockfiles. There are two separate installs:
+  # the bindings themselves and the examples, which have their own lockfile.
+  # As with cargo and pip above, only bump the lockfile — the version ranges
+  # in package.json are our consumers' constraints, not ours.
+  - package-ecosystem: npm
+    directory: /nodejs
+    schedule:
+      interval: weekly
+    versioning-strategy: lockfile-only
+    groups:
+      nodejs-deps:
+        patterns:
+          - "*"
+
+  - package-ecosystem: npm
+    directory: /nodejs/examples
+    schedule:
+      interval: weekly
+    versioning-strategy: lockfile-only
+    groups:
+      nodejs-examples-deps:
+        patterns:
+          - "*"

.github/workflows/dev.yml

@@ -29,12 +29,14 @@ jobs:
     steps:
       - uses: actions/setup-node@v6
         with:
-          node-version: "18"
+          node-version: "24"
+      - uses: pnpm/action-setup@v6
+        with:
+          version: 11.1.1
       # These rules are disabled because Github will always ensure there
       # is a blank line between the title and the body and Github will
       # word wrap the description field to ensure a reasonable max line
       # length.
-      - run: npm install @commitlint/config-conventional
       - run: >
           echo 'module.exports = {
             "rules": {
@@ -43,7 +45,11 @@ jobs:
               "body-leading-blank": [0, "always"]
             }
           }' > .commitlintrc.js
-      - run: npx commitlint --extends @commitlint/config-conventional --verbose <<< $COMMIT_MSG
+      - run: >
+          pnpm dlx
+          --package @commitlint/cli@21.2.2
+          --package @commitlint/config-conventional@21.2.2
+          commitlint --extends @commitlint/config-conventional --verbose <<< $COMMIT_MSG
         env:
           COMMIT_MSG: >
             ${{ github.event.pull_request.title }}
@@ -54,7 +60,7 @@ jobs:
         with:
           script: |
             const message = `**ACTION NEEDED**
-              
+
               Lance follows the [Conventional Commits specification](https://www.conventionalcommits.org/en/v1.0.0/) for release automation.
 
               The PR title and description are used as the merge commit message.\

.github/workflows/docs-link-check.yml

@@ -56,7 +56,7 @@ jobs:
         uses: lycheeverse/lychee-action@e7477775783ea5526144ba13e8db5eec57747ce8 # v2.9.0
         with:
           # Restricted to http(s) on purpose. Much of docs/src is generated
-          # API reference (the js/ tree comes from `npm run docs` in nodejs)
+          # API reference (the js/ tree comes from `pnpm run docs` in nodejs)
           # and the hand-written pages use mkdocstrings cross-references and
           # nav-relative paths that only resolve in the site mkdocs builds,
           # not in this checkout, so relative links would be reported as

.github/workflows/docs.yml

@@ -55,9 +55,7 @@ jobs:
       - name: Set up node
         uses: actions/setup-node@v6
         with:
-          node-version: 20
+          node-version: 24
-          cache: 'npm'
-          cache-dependency-path: docs/package-lock.json
       - name: Install node dependencies
         working-directory: nodejs
         run: |

.github/workflows/nodejs.yml

@@ -47,9 +47,8 @@ jobs:
         version: 11.1.1
     - uses: actions/setup-node@v6
       with:
-        # pnpm 11 requires Node >= 22.13; use 24 since 22 hits EOL
+        # Build on a supported LTS; the matrix job below covers every
-        # in October. The library itself still supports Node >= 18
+        # Node version the library claims to support.
-        # (see test matrix below).
         node-version: 24
         cache: 'pnpm'
         cache-dependency-path: nodejs/pnpm-lock.yaml
@@ -84,7 +83,7 @@ jobs:
     timeout-minutes: 30
     strategy:
       matrix:
-        node-version: [ "18", "20" ]
+        node-version: [ "22", "24", "26" ]
     runs-on: "ubuntu-22.04"
     defaults:
       run:
@@ -101,9 +100,9 @@ jobs:
     - uses: actions/setup-node@v6
       name: Setup Node.js 24 for build
       with:
-        # pnpm 11 requires Node >= 22.13; use 24 since 22 hits EOL
+        # Build and install once on a fixed version so the generated docs
-        # in October. Build/install runs on Node 24; tests run on the
+        # are identical across matrix legs; the tests below then run on each
-        # matrix version below using direct jest invocation.
+        # supported Node version.
         node-version: 24
         cache: 'pnpm'
         cache-dependency-path: nodejs/pnpm-lock.yaml
@@ -152,9 +151,9 @@ jobs:
         S3_TEST: "1"
         # Newer @smithy/core uses dynamic ESM imports.
         NODE_OPTIONS: "--experimental-vm-modules"
-      # Invoke jest directly because pnpm 11 itself requires Node 22+
+      # Invoke the installed jest binary directly; the pnpm shim is set up
-      # while the matrix tests on older Node versions.
+      # against the build-phase Node, not the version selected above.
-      run: npx jest --verbose
+      run: node_modules/.bin/jest --verbose
     - name: Test examples
       working-directory: ./
       env:
@@ -164,7 +163,7 @@ jobs:
       run: |
         python ci/mock_openai.py &
         cd nodejs/examples
-        npx jest --testEnvironment jest-environment-node-single-context --verbose
+        node_modules/.bin/jest --testEnvironment jest-environment-node-single-context --verbose
   macos:
     timeout-minutes: 30
     # macos-15 ships a newer linker; the older macos-14 linker fails to insert
@@ -185,8 +184,7 @@ jobs:
         version: 11.1.1
     - uses: actions/setup-node@v6
       with:
-        # pnpm 11 requires Node >= 22.13; use 24 since 22 hits EOL
+        # pnpm 11 requires Node >= 22.13.
-        # in October.
         node-version: 24
         cache: 'pnpm'
         cache-dependency-path: nodejs/pnpm-lock.yaml

.github/workflows/npm-publish.yml

@@ -40,40 +40,31 @@ jobs:
           - target: aarch64-apple-darwin
             host: macos-latest
             features: fp16kernels
+            # Fat LTO was ~111 of this job's ~113 minutes.
+            lto: thin
+            codegen_units: 16
             pre_build: |-
               brew install protobuf
-              # Fat LTO (the workspace default in .cargo/config.toml) is
-              # single-threaded and is the peak-memory step of the build. On
-              # this runner it accounted for ~111 of the job's ~113 minutes,
-              # making it the critical path of the entire publish pipeline.
-              # ThinLTO parallelizes it across the runner's cores, for a few
-              # percent of runtime performance.
-              export CARGO_PROFILE_RELEASE_LTO=thin
-              export CARGO_PROFILE_RELEASE_CODEGEN_UNITS=16
           - target: x86_64-pc-windows-msvc
             host: windows-2025
             features: ","
+            # The lower peak also keeps this on the standard 4-core runner.
+            lto: thin
+            codegen_units: 16
             pre_build: |-
               choco install --no-progress protoc ninja nasm
               tail -n 1000 /c/ProgramData/chocolatey/logs/chocolatey.log
               # There is an issue where choco doesn't add nasm to the path
               export PATH="$PATH:/c/Program Files/NASM"
               nasm -v
-              # See the ThinLTO note on aarch64-apple-darwin above. Keeping
-              # peak memory down is also what lets this run on the standard
-              # 4-core runner: the 8-core larger runner was only needed to
-              # stop fat LTO from OOMing rustc-LLVM.
-              export CARGO_PROFILE_RELEASE_LTO=thin
-              export CARGO_PROFILE_RELEASE_CODEGEN_UNITS=16
           - target: aarch64-pc-windows-msvc
             host: windows-2025
             features: ","
+            lto: thin
+            codegen_units: 16
             pre_build: |-
                 choco install --no-progress protoc
                 rustup target add aarch64-pc-windows-msvc
-                # See the ThinLTO note on aarch64-apple-darwin above.
-                export CARGO_PROFILE_RELEASE_LTO=thin
-                export CARGO_PROFILE_RELEASE_CODEGEN_UNITS=16
           - target: x86_64-unknown-linux-gnu
             host: ubuntu-latest
             features: fp16kernels
@@ -103,6 +94,14 @@ jobs:
             # https://github.com/napi-rs/napi-rs/blob/main/debian-aarch64.Dockerfile
             docker: ghcr.io/napi-rs/napi-rs/nodejs-rust:lts-debian-aarch64
             features: "fp16kernels"
+            # Fat LTO OOM-killed rustc every nightly; even with lld it peaked
+            # at 31391 MiB of the runner's 32 GiB.
+            lto: thin
+            codegen_units: 16
+            # arm64 Linux links through GNU `ld` where x86_64 defaults to
+            # `rust-lld`, which is why only arm64 OOM'd. lld cut the largest
+            # linker process 7.0 -> 4.0 GiB (lancedb/sophon#7313).
+            linker: /tmp/aarch64-lld-clang
             pre_build: |-
               set -e &&
               apt-get update &&
@@ -112,9 +111,30 @@ jobs:
               # AT_HWCAP2 (added in Linux 3.17). Define it for aws-lc-sys.
               export CFLAGS="$CFLAGS -DAT_HWCAP2=26" &&
               rustup target add aarch64-unknown-linux-gnu
+              # Not `&&`-chained: in dash, errexit does not fire for a
+              # non-final command in an `&&` list, so failures were ignored.
+              #
+              # A wrapper rather than `-C link-arg` because the per-target
+              # rustflags variable does not reach every unit that links, while
+              # the linker variable does. `clang` because GCC silently ignores
+              # `-fuse-ld=lld` unless built with lld support. Two echoes
+              # because printf's newline escape gets rewritten to `;` between
+              # here and the container.
+              echo '#!/bin/sh' > /tmp/aarch64-lld-clang
+              echo 'exec clang --target=aarch64-unknown-linux-gnu --sysroot=/usr/aarch64-unknown-linux-gnu/aarch64-unknown-linux-gnu/sysroot --gcc-toolchain=/usr/aarch64-unknown-linux-gnu -fuse-ld=lld "$@"' >> /tmp/aarch64-lld-clang
+              chmod 0755 /tmp/aarch64-lld-clang
+              # Fail now, not at the cdylib link ~30 minutes later. Linking at
+              # all also proves lld resolved; clang errors out when it cannot.
+              echo 'int main(void){return 0;}' > /tmp/probe.c
+              /tmp/aarch64-lld-clang /tmp/probe.c -o /tmp/probe
+              readelf -h /tmp/probe | grep AArch64
           - target: aarch64-unknown-linux-musl
             host: ubuntu-2404-8x-x64
             features: ","
+            # Fat LTO took the whole runner down. lld cannot help: it died
+            # inside rustc's LLVM, before any linker was spawned.
+            lto: thin
+            codegen_units: 16
             pre_build: |-
               set -e &&
               sudo apt-get update &&
@@ -123,6 +143,19 @@ jobs:
               export EXTRA_ARGS="-x"
     name: build - ${{ matrix.settings.target }}
     runs-on: ${{ matrix.settings.host }}
+    # On the job, not exported from `pre_build`: `Swatinem/rust-cache` hashes
+    # `CARGO_*` into its cache key before any step runs, so a step-local export
+    # leaves the key unchanged while cargo still rebuilds cold. The ThinLTO
+    # legs had been doing that every run.
+    #
+    # Not `RUSTFLAGS`: setting it, even to "", discards every config-file
+    # rustflag, silently dropping .cargo/config.toml's `target-cpu` and
+    # `target-feature` from the published binaries.
+    env:
+      CARGO_PROFILE_RELEASE_LTO: ${{ matrix.settings.lto || 'fat' }}
+      CARGO_PROFILE_RELEASE_CODEGEN_UNITS: ${{ matrix.settings.codegen_units || '1' }}
+      # Empty elsewhere: a per-target variable is only read for that triple.
+      CARGO_TARGET_AARCH64_UNKNOWN_LINUX_GNU_LINKER: ${{ matrix.settings.linker }}
     defaults:
       run:
         working-directory: nodejs
@@ -135,8 +168,7 @@ jobs:
       - name: Setup node
         uses: actions/setup-node@v6
         with:
-          # pnpm 11 requires Node >= 22.13; use 24 since 22 hits EOL
+          # pnpm 11 requires Node >= 22.13.
-          # in October.
           node-version: 24
           cache: pnpm
           cache-dependency-path: nodejs/pnpm-lock.yaml
@@ -169,19 +201,15 @@ jobs:
           # creating ref). The nightly cadence also keeps entries inside
           # GitHub's 7-day eviction window, which a tag-only trigger would not.
           save-if: ${{ github.ref == 'refs/heads/main' }}
-      # Docker builds can use rust-cache too. `target/` already lives on the
+      # Docker builds can use rust-cache too: the workspace is bind-mounted, so
-      # host because the whole workspace is bind-mounted into the container, and
+      # `target/` lives on the host and rust-cache's prune keeps the entry
-      # rust-cache's prune and save run host-side, so they can manage it -- which
+      # small.
-      # is what keeps the entry to dependency artifacts rather than a multi-GB
-      # copy of everything.
       #
       # Two differences from the native builds. The container's CARGO_HOME is
-      # bind-mounted from `.cargo-cache` rather than the host's ~/.cargo, so that
+      # bind-mounted from `.cargo-cache` rather than ~/.cargo, so that is cached
-      # has to be cached explicitly. And the key is derived from the *host* rustc
+      # explicitly. And the key uses the *host* rustc version, not the compiler
-      # version, which is not the compiler that produced these artifacts; that is
+      # that built these artifacts -- safe, since cargo fingerprints the real
-      # safe because cargo fingerprints the real compiler and rebuilds on a
+      # one; a base-image bump just costs one cold build.
-      # mismatch, it just means a base-image toolchain bump costs one cold build
-      # instead of invalidating the key.
       - name: Cache cargo (docker builds)
         uses: Swatinem/rust-cache@v2
         if: ${{ matrix.settings.docker }}
@@ -210,14 +238,19 @@ jobs:
           # cache step above saves. Previously the registry mounts pointed at
           # `.cargo/...`, a path nothing cached, so the container re-downloaded
           # the whole crate registry on every run.
+          #
+          # `docker run` inherits nothing; `-e NAME` carries the job's `env:` in.
           options: "--user 0:0 -v ${{ github.workspace }}/.cargo-cache/git/db:/usr/local/cargo/git/db \
                     -v ${{ github.workspace }}/.cargo-cache/registry/cache:/usr/local/cargo/registry/cache \
                     -v ${{ github.workspace }}/.cargo-cache/registry/index:/usr/local/cargo/registry/index \
+                    -e CARGO_PROFILE_RELEASE_LTO \
+                    -e CARGO_PROFILE_RELEASE_CODEGEN_UNITS \
+                    -e CARGO_TARGET_AARCH64_UNKNOWN_LINUX_GNU_LINKER \
                     -v ${{ github.workspace }}:/build -w /build/nodejs"
           run: |
             set -e
             ${{ matrix.settings.pre_build }}
-            npx napi build --platform --release \
+            node_modules/.bin/napi build --platform --release \
               --features ${{ matrix.settings.features }} \
               --target ${{ matrix.settings.target }} \
               --dts ../lancedb/native.d.ts \
@@ -237,7 +270,7 @@ jobs:
       - name: Build
         run: |
           ${{ matrix.settings.pre_build }}
-          npx napi build --platform --release \
+          node_modules/.bin/napi build --platform --release \
               --features ${{ matrix.settings.features }} \
               --target ${{ matrix.settings.target }} \
               --dts ../lancedb/native.d.ts \
@@ -256,6 +289,18 @@ jobs:
         if: always()
         run: df -h
         shell: bash
+      - name: Report peak memory
+        if: always() && runner.os == 'Linux'
+        shell: bash
+        run: |
+          peak=$(find /sys/fs/cgroup -name memory.peak -readable \
+                 -exec cat {} + 2>/dev/null | sort -n | tail -1)
+          if [ -n "$peak" ]; then
+            echo "peak memory: $((peak / 1024 / 1024)) MiB"
+          else
+            echo "peak memory: unavailable (no readable cgroup v2 memory.peak)"
+          fi
+          free -g || true
       - name: Upload artifact
         uses: actions/upload-artifact@v7
         with:
@@ -293,7 +338,7 @@ jobs:
           - target: aarch64-unknown-linux-gnu
             host: ubuntu-2404-8x-arm64
         node:
-          - '20'
+          - '22'
     runs-on: ${{ matrix.settings.host }}
     defaults:
       run:
@@ -339,9 +384,9 @@ jobs:
       - name: Move built files
         run: cp dist/native.d.ts dist/native.js dist/*.node lancedb/
       - name: Test bindings
-        # Invoke jest directly because pnpm 11 itself requires Node 22+
+        # Invoke the installed jest binary directly; the pnpm shim is set up
-        # while the matrix tests on older Node versions.
+        # against the install-phase Node, not the version selected above.
-        run: npx jest --verbose
+        run: node_modules/.bin/jest --verbose
   publish:
     name: Publish
     runs-on: ubuntu-latest

.github/workflows/rust.yml

@@ -232,7 +232,10 @@ jobs:
           ALL_FEATURES=`cargo metadata --format-version=1 --no-deps \
             | jq -r '.packages[] | .features | keys | .[]' \
             | grep -v s3-test | sort | uniq | paste -s -d "," -`
-          cargo test --profile ci --features $ALL_FEATURES --locked
+          # Run doctests before test binaries fill the runner disk. Examples are
+          # already built by the Linux job, so avoid retaining them here.
+          cargo test --profile ci --features $ALL_FEATURES --locked --doc
+          cargo test --profile ci --features $ALL_FEATURES --locked --lib --tests
 
   windows:
     strategy:

.github/workflows/update_package_lock_run.yml

@@ -1,22 +0,0 @@
-name: Update package-lock.json
-
-on:
-  workflow_dispatch:
-
-permissions:
-  contents: read
-
-jobs:
-  publish:
-    runs-on: ubuntu-latest
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v6
-        with:
-          ref: main
-          persist-credentials: false
-          fetch-depth: 0
-          lfs: true
-      - uses: ./.github/workflows/update_package_lock
-        with:
-          github_token: ${{ secrets.LANCEDB_RELEASE_TOKEN }}

.github/workflows/update_package_lock_run_nodejs.yml

@@ -1,22 +0,0 @@
-name: Update NodeJs package-lock.json
-
-on:
-  workflow_dispatch:
-
-permissions:
-  contents: read
-
-jobs:
-  publish:
-    runs-on: ubuntu-latest
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v6
-        with:
-          ref: main
-          persist-credentials: false
-          fetch-depth: 0
-          lfs: true
-      - uses: ./.github/workflows/update_package_lock_nodejs
-        with:
-          github_token: ${{ secrets.LANCEDB_RELEASE_TOKEN }}

.pre-commit-config.yaml

@@ -20,7 +20,10 @@ repos:
     hooks:
       - id: local-biome-check
         name: biome check
-        entry: npx @biomejs/biome@1.8.3 check --config-path nodejs/biome.json nodejs/
+        # Use the biome from nodejs/package.json rather than a separately
+        # pinned one: the two drifted apart and disagreed on formatting, so
+        # this hook rejected code that `pnpm lint` accepted.
+        entry: nodejs/node_modules/.bin/biome check --config-path nodejs/biome.json nodejs/
         language: system
         types: [text]
         files: "nodejs/.*"

AGENTS.md

@@ -38,7 +38,7 @@ Before committing changes, run formatting for every language you touched. At min
 * Rust changes: run `cargo fmt --all`.
 * Python changes: run `ruff format .` and `ruff check .` from the repository root,
   and run targeted tests through `cd python && uv run ...`.
-* TypeScript changes: run the relevant `npm`/`pnpm` lint, format, build, and docs commands in `nodejs`.
+* TypeScript changes: run the relevant `pnpm` lint, format, build, and docs commands in `nodejs`.
 
 Before creating a PR, the exact value passed to `gh pr create --title` must follow
 Conventional Commits, such as `fix: support nested field paths in native index creation`
@@ -101,12 +101,12 @@ Python bindings changes:
 TypeScript bindings changes:
 
 1. Add napi-rs method binding on `Table` in `nodejs/src/table.rs`.
-2. Run `npm run build` to generate TypeScript definitions.
+2. Run `pnpm build` to generate TypeScript definitions.
 3. Add typescript method on abstract class `Table` in `nodejs/src/table.ts`.
 4. Add concrete method on `LocalTable` class in `nodejs/src/native_table.ts`.
     * Note: despite the name, this class is also used for remote tables.
 5. Add test in `nodejs/__test__/table.test.ts`.
-6. Run `npm run docs` to generate TypeScript documentation.
+6. Run `pnpm run docs` to generate TypeScript documentation.
 
 ## Python API reference
 

Cargo.lock

@@ -332,6 +332,34 @@ dependencies = [
  "num-traits",
 ]
 
+[[package]]
+name = "arrow-flight"
+version = "58.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b2dbe34824c639e43136af8f106992792ab456540d54b880bc320a3192502d2e"
+dependencies = [
+ "arrow-arith",
+ "arrow-array",
+ "arrow-buffer",
+ "arrow-cast",
+ "arrow-data",
+ "arrow-ipc",
+ "arrow-ord",
+ "arrow-row",
+ "arrow-schema",
+ "arrow-select",
+ "arrow-string",
+ "base64 0.22.1",
+ "bytes",
+ "futures",
+ "once_cell",
+ "paste",
+ "prost",
+ "prost-types",
+ "tonic",
+ "tonic-prost",
+]
+
 [[package]]
 name = "arrow-ipc"
 version = "58.4.0"
@@ -535,9 +563,9 @@ dependencies = [
 
 [[package]]
 name = "async-trait"
-version = "0.1.91"
+version = "0.1.92"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ae36dc4177970ef04fde5178d3e2429882def40e57a451f919c098f72baa6cec"
+checksum = "82f6aeea286b8eb4dd3431a1be1b59d290ace00f5bfd8e2a159bc2a05e2c1667"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -1129,7 +1157,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "edca88bc138befd0323b20752846e6587272d3b03b0343c8ea28a6f819e6e71f"
 dependencies = [
  "async-trait",
- "axum-core",
+ "axum-core 0.4.5",
  "bytes",
  "futures-util",
  "http 1.5.0",
@@ -1138,7 +1166,7 @@ dependencies = [
  "hyper 1.9.0",
  "hyper-util",
  "itoa",
- "matchit",
+ "matchit 0.7.3",
  "memchr",
  "mime",
  "percent-encoding",
@@ -1156,6 +1184,31 @@ dependencies = [
  "tracing",
 ]
 
+[[package]]
+name = "axum"
+version = "0.8.9"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "31b698c5f9a010f6573133b09e0de5408834d0c82f8d7475a89fc1867a71cd90"
+dependencies = [
+ "axum-core 0.5.6",
+ "bytes",
+ "futures-util",
+ "http 1.5.0",
+ "http-body 1.1.0",
+ "http-body-util",
+ "itoa",
+ "matchit 0.8.4",
+ "memchr",
+ "mime",
+ "percent-encoding",
+ "pin-project-lite",
+ "serde_core",
+ "sync_wrapper",
+ "tower",
+ "tower-layer",
+ "tower-service",
+]
+
 [[package]]
 name = "axum-core"
 version = "0.4.5"
@@ -1177,6 +1230,24 @@ dependencies = [
  "tracing",
 ]
 
+[[package]]
+name = "axum-core"
+version = "0.5.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "08c78f31d7b1291f7ee735c1c6780ccde7785daae9a9206026862dab7d8792d1"
+dependencies = [
+ "bytes",
+ "futures-core",
+ "http 1.5.0",
+ "http-body 1.1.0",
+ "http-body-util",
+ "mime",
+ "pin-project-lite",
+ "sync_wrapper",
+ "tower-layer",
+ "tower-service",
+]
+
 [[package]]
 name = "backoff"
 version = "0.4.0"
@@ -1443,9 +1514,9 @@ checksum = "175812e0be2bccb6abe50bb8d566126198344f707e304f45c648fd8f2cc0365e"
 
 [[package]]
 name = "bytemuck"
-version = "1.25.0"
+version = "1.25.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c8efb64bd706a16a1bdde310ae86b351e4d21550d98d056f22f8a7f7a2183fec"
+checksum = "95832e849adfb21180ccb6826a99da14e5d266ae5c2e668e1602cf234f153797"
 dependencies = [
  "bytemuck_derive",
 ]
@@ -1597,9 +1668,9 @@ checksum = "613afe47fcd5fac7ccf1db93babcb082c5994d996f20b8b159f2ad1658eb5724"
 
 [[package]]
 name = "chacha20"
-version = "0.10.0"
+version = "0.10.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "6f8d983286843e49675a4b7a2d174efe136dc93a18d69130dd18198a6c167601"
+checksum = "65c35e4b699c7e15ccbe7ee35c005e4fc0a278d22238a2857e6ce2dadeda1b06"
 dependencies = [
  "cfg-if 1.0.4",
  "cpufeatures 0.3.0",
@@ -3455,8 +3526,8 @@ checksum = "42703706b716c37f96a77aea830392ad231f44c9e9a67872fa5548707e11b11c"
 
 [[package]]
 name = "fsst"
-version = "12.0.0-beta.2"
+version = "12.0.0-beta.14"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "arrow-array",
  "rand 0.9.5",
@@ -4815,8 +4886,8 @@ checksum = "e037a2e1d8d5fdbd49b16a4ea09d5d6401c1f29eca5ff29d03d3824dba16256a"
 
 [[package]]
 name = "lance"
-version = "12.0.0-beta.2"
+version = "12.0.0-beta.14"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "arc-swap",
  "arrow",
@@ -4888,8 +4959,8 @@ dependencies = [
 
 [[package]]
 name = "lance-arrow"
-version = "12.0.0-beta.2"
+version = "12.0.0-beta.14"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "arrow-array",
  "arrow-buffer",
@@ -4911,7 +4982,7 @@ dependencies = [
 [[package]]
 name = "lance-arrow-scalar"
 version = "58.0.0"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "arrow-array",
  "arrow-buffer",
@@ -4925,7 +4996,7 @@ dependencies = [
 [[package]]
 name = "lance-arrow-stats"
 version = "58.0.0"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "arrow-array",
  "arrow-schema",
@@ -4934,8 +5005,8 @@ dependencies = [
 
 [[package]]
 name = "lance-bitpacking"
-version = "12.0.0-beta.2"
+version = "12.0.0-beta.14"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "arrayref",
  "crunchy",
@@ -4945,8 +5016,8 @@ dependencies = [
 
 [[package]]
 name = "lance-core"
-version = "12.0.0-beta.2"
+version = "12.0.0-beta.14"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "arrow-array",
  "arrow-buffer",
@@ -4983,8 +5054,8 @@ dependencies = [
 
 [[package]]
 name = "lance-datafusion"
-version = "12.0.0-beta.2"
+version = "12.0.0-beta.14"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "arrow",
  "arrow-array",
@@ -5000,6 +5071,7 @@ dependencies = [
  "datafusion-functions",
  "datafusion-physical-expr",
  "futures",
+ "half",
  "jsonb",
  "lance-arrow",
  "lance-core",
@@ -5013,8 +5085,8 @@ dependencies = [
 
 [[package]]
 name = "lance-datagen"
-version = "12.0.0-beta.2"
+version = "12.0.0-beta.14"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "arrow",
  "arrow-array",
@@ -5031,8 +5103,8 @@ dependencies = [
 
 [[package]]
 name = "lance-derive"
-version = "12.0.0-beta.2"
+version = "12.0.0-beta.14"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -5041,8 +5113,8 @@ dependencies = [
 
 [[package]]
 name = "lance-encoding"
-version = "12.0.0-beta.2"
+version = "12.0.0-beta.14"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "arrow-arith",
  "arrow-array",
@@ -5075,8 +5147,8 @@ dependencies = [
 
 [[package]]
 name = "lance-file"
-version = "12.0.0-beta.2"
+version = "12.0.0-beta.14"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "arrow-arith",
  "arrow-array",
@@ -5107,8 +5179,8 @@ dependencies = [
 
 [[package]]
 name = "lance-index"
-version = "12.0.0-beta.2"
+version = "12.0.0-beta.14"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "arc-swap",
  "arrow",
@@ -5172,8 +5244,8 @@ dependencies = [
 
 [[package]]
 name = "lance-index-core"
-version = "12.0.0-beta.2"
+version = "12.0.0-beta.14"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "arrow-array",
  "arrow-schema",
@@ -5195,8 +5267,8 @@ dependencies = [
 
 [[package]]
 name = "lance-io"
-version = "12.0.0-beta.2"
+version = "12.0.0-beta.14"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "arrow",
  "arrow-array",
@@ -5236,8 +5308,8 @@ dependencies = [
 
 [[package]]
 name = "lance-linalg"
-version = "12.0.0-beta.2"
+version = "12.0.0-beta.14"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "arrow-array",
  "arrow-schema",
@@ -5251,27 +5323,29 @@ dependencies = [
 
 [[package]]
 name = "lance-namespace"
-version = "12.0.0-beta.2"
+version = "12.0.0-beta.14"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "arrow",
  "async-trait",
  "bytes",
  "lance-core",
  "lance-namespace-reqwest-client",
+ "serde",
+ "serde_json",
  "snafu 0.9.0",
 ]
 
 [[package]]
 name = "lance-namespace-impls"
-version = "12.0.0-beta.2"
+version = "12.0.0-beta.14"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "arrow",
  "arrow-ipc",
  "arrow-schema",
  "async-trait",
- "axum",
+ "axum 0.7.9",
  "base64 0.22.1",
  "bytes",
  "chrono",
@@ -5304,9 +5378,9 @@ dependencies = [
 
 [[package]]
 name = "lance-namespace-reqwest-client"
-version = "0.11.0"
+version = "0.12.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0a030196da1c994b63a96a4f0bf5b0cfa459fe6dadc9e962320246ca328da22a"
+checksum = "d8d23e54b1634d5bbb434f8dd33dc3c05f6e58d876a9a27b3b4aef58ddbe11af"
 dependencies = [
  "reqwest 0.12.28",
  "serde",
@@ -5318,8 +5392,8 @@ dependencies = [
 
 [[package]]
 name = "lance-select"
-version = "12.0.0-beta.2"
+version = "12.0.0-beta.14"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "arrow-array",
  "arrow-buffer",
@@ -5333,8 +5407,8 @@ dependencies = [
 
 [[package]]
 name = "lance-table"
-version = "12.0.0-beta.2"
+version = "12.0.0-beta.14"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "arrow",
  "arrow-array",
@@ -5374,8 +5448,8 @@ dependencies = [
 
 [[package]]
 name = "lance-testing"
-version = "12.0.0-beta.2"
+version = "12.0.0-beta.14"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "arrow-array",
  "arrow-schema",
@@ -5388,8 +5462,8 @@ dependencies = [
 
 [[package]]
 name = "lance-tokenizer"
-version = "12.0.0-beta.2"
+version = "12.0.0-beta.14"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.14#a8a101774a1c9647065cc60137094feadbe55296"
 dependencies = [
  "frostem",
  "icu_segmenter",
@@ -5402,7 +5476,7 @@ dependencies = [
 
 [[package]]
 name = "lancedb"
-version = "0.38.0-beta.11"
+version = "0.39.0-beta.4"
 dependencies = [
  "ahash",
  "anyhow",
@@ -5411,6 +5485,7 @@ dependencies = [
  "arrow-buffer",
  "arrow-cast",
  "arrow-data",
+ "arrow-flight",
  "arrow-ipc",
  "arrow-ord",
  "arrow-schema",
@@ -5466,6 +5541,7 @@ dependencies = [
  "polars",
  "polars-arrow",
  "pprof 0.14.1",
+ "prost",
  "rand 0.9.5",
  "random_word",
  "regex",
@@ -5482,6 +5558,7 @@ dependencies = [
  "test-log",
  "tokenizers",
  "tokio",
+ "tonic",
  "url",
  "urlencoding",
  "uuid",
@@ -5490,7 +5567,7 @@ dependencies = [
 
 [[package]]
 name = "lancedb-nodejs"
-version = "0.38.0-beta.11"
+version = "0.39.0-beta.4"
 dependencies = [
  "arrow-array",
  "arrow-buffer",
@@ -5515,7 +5592,7 @@ dependencies = [
 
 [[package]]
 name = "lancedb-python"
-version = "0.38.0-beta.11"
+version = "0.39.0-beta.4"
 dependencies = [
  "arrow",
  "async-trait",
@@ -5539,6 +5616,7 @@ dependencies = [
  "serde_json",
  "snafu 0.8.9",
  "tokio",
+ "uuid",
 ]
 
 [[package]]
@@ -5748,9 +5826,9 @@ dependencies = [
 
 [[package]]
 name = "log"
-version = "0.4.33"
+version = "0.4.34"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0ceec5bc11778974d1bcb055b18002eba7f4b3518b6a0081b3af5f21666da9ad"
+checksum = "f9f8bd3e56ce4dfc153cf470fffbfa98c7620958b312ca5c3a4b8d5181fd13c6"
 
 [[package]]
 name = "loom"
@@ -5859,6 +5937,12 @@ version = "0.7.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "0e7465ac9959cc2b1404e8e2367b43684a6d13790fe23056cc8c6c5a6b7bcb94"
 
+[[package]]
+name = "matchit"
+version = "0.8.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "47e1ffaa40ddd1f3ed91f717a33c8c0ee23fff369e3aa8772b9605cc1d22f4c3"
+
 [[package]]
 name = "matrixmultiply"
 version = "0.3.10"
@@ -6001,9 +6085,9 @@ dependencies = [
 
 [[package]]
 name = "moka"
-version = "0.12.15"
+version = "0.12.16"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "957228ad12042ee839f93c8f257b62b4c0ab5eaae1d4fa60de53b27c9d7c5046"
+checksum = "4293f18e7567a1caf3c584855554377025c65e0aa445344d04171f5ad63d19b9"
 dependencies = [
  "async-lock",
  "crossbeam-channel",
@@ -6097,14 +6181,15 @@ dependencies = [
 
 [[package]]
 name = "napi"
-version = "3.11.0"
+version = "3.12.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "de33522036981030a75c231829566bc63414e08101a6f5ff4ac6cef19c8e0941"
+checksum = "58c5f4d5375213fdb7be2655e152386e82f026f9a5ba36a75556e11359aafe09"
 dependencies = [
  "bitflags 2.11.1",
  "chrono",
  "ctor 1.0.12",
  "futures",
+ "libc",
  "napi-build",
  "napi-sys",
  "nohash-hasher",
@@ -6116,15 +6201,15 @@ dependencies = [
 
 [[package]]
 name = "napi-build"
-version = "2.4.0"
+version = "2.4.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5282704fbe8d49b0cf8b08e3f33233416a528658f205c7e5ace63b582de0b11c"
+checksum = "60fdf9b392c50e7c4170fa633bd909490ed7835cea4c046776d1a4dd8d2ae0ab"
 
 [[package]]
 name = "napi-derive"
-version = "3.6.1"
+version = "3.6.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4d5c9c02556ea6dc99dffd36c1ce60141411657438501a125b675776d011ce92"
+checksum = "0fa55ea69990c90b888e9e77044410e304ce7f35de599dc6d0b5c1923d2e59af"
 dependencies = [
  "convert_case",
  "ctor 1.0.12",
@@ -6136,9 +6221,9 @@ dependencies = [
 
 [[package]]
 name = "napi-derive-backend"
-version = "6.1.1"
+version = "6.1.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d60b5d773ad46c698c8cc2cd9fde0b283d39cbb7f71c04bee633c7bdba4423bd"
+checksum = "df4056ac7c18e4438ccf0edaed4340ca0d269278c8ec19284f7b23cb039fd0ae"
 dependencies = [
  "convert_case",
  "proc-macro2",
@@ -7733,6 +7818,7 @@ dependencies = [
  "pyo3-build-config",
  "pyo3-ffi",
  "pyo3-macros",
+ "uuid",
 ]
 
 [[package]]
@@ -8601,9 +8687,9 @@ dependencies = [
 
 [[package]]
 name = "roaring"
-version = "0.11.4"
+version = "0.11.5"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "1dedc5658c6ecb3bdb5ef5f3295bb9253f42dcf3fd1402c03f6b1f7659c3c4a9"
+checksum = "18bd8a37d17a58532776dcdf6041ce64929adca78e8489d5cacbafe99229d3e1"
 dependencies = [
  "bytemuck",
  "byteorder",
@@ -9063,9 +9149,9 @@ dependencies = [
 
 [[package]]
 name = "serde_with"
-version = "3.21.0"
+version = "3.22.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "76a5c54c7310e7b8b9577c286d7e399ddd876c3e12b3ed917a8aabc4b96e9e8c"
+checksum = "ee78f1fbe43ac4a0e47aadb3dbd357b69eb0d3793e948624cd03dd2750ab1c0a"
 dependencies = [
  "base64 0.22.1",
  "bs58",
@@ -9073,6 +9159,7 @@ dependencies = [
  "hex",
  "indexmap 1.9.3",
  "indexmap 2.14.0",
+ "jiff",
  "schemars 0.9.0",
  "schemars 1.2.1",
  "serde_core",
@@ -9083,9 +9170,9 @@ dependencies = [
 
 [[package]]
 name = "serde_with_macros"
-version = "3.21.0"
+version = "3.22.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "84d57bc0c8b9a17920c178daa6bb924850d54a9c97ab45194bb8c17ad66bb660"
+checksum = "8705578779c2b6bd90d84d66eb2e206b708b1a4d7b9f17641b293545bf1c7e46"
 dependencies = [
  "darling 0.23.0",
  "proc-macro2",
@@ -10084,6 +10171,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "ac2a5518c70fa84342385732db33fb3f44bc4cc748936eb5833d2df34d6445ef"
 dependencies = [
  "async-trait",
+ "axum 0.8.9",
  "base64 0.22.1",
  "bytes",
  "h2 0.4.16",
@@ -10095,9 +10183,11 @@ dependencies = [
  "hyper-util",
  "percent-encoding",
  "pin-project",
+ "rustls-native-certs",
  "socket2 0.6.3",
  "sync_wrapper",
  "tokio",
+ "tokio-rustls 0.26.4",
  "tokio-stream",
  "tower",
  "tower-layer",
@@ -10452,9 +10542,9 @@ checksum = "06abde3611657adf66d383f00b093d7faecc7fa57071cce2578660c9f1010821"
 
 [[package]]
 name = "uuid"
-version = "1.24.0"
+version = "1.26.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "bf3923a6f5c4c6382e0b653c4117f48d631ea17f38ed86e2a828e6f7412f5239"
+checksum = "b5772d71c9be8a8a6ac2117d949c5b224c1b72241bb611d9a3012edcf8af7812"
 dependencies = [
  "getrandom 0.4.2",
  "js-sys",

Cargo.toml

@@ -13,20 +13,20 @@ categories = ["database-implementations"]
 rust-version = "1.91.0"
 
 [workspace.dependencies]
-lance = { "version" = "=12.0.0-beta.2", default-features = false, "tag" = "v12.0.0-beta.2", "git" = "https://github.com/lance-format/lance.git" }
+lance = { "version" = "=12.0.0-beta.14", default-features = false, "tag" = "v12.0.0-beta.14", "git" = "https://github.com/lance-format/lance.git" }
-lance-core = { "version" = "=12.0.0-beta.2", "tag" = "v12.0.0-beta.2", "git" = "https://github.com/lance-format/lance.git" }
+lance-core = { "version" = "=12.0.0-beta.14", "tag" = "v12.0.0-beta.14", "git" = "https://github.com/lance-format/lance.git" }
-lance-datagen = { "version" = "=12.0.0-beta.2", "tag" = "v12.0.0-beta.2", "git" = "https://github.com/lance-format/lance.git" }
+lance-datagen = { "version" = "=12.0.0-beta.14", "tag" = "v12.0.0-beta.14", "git" = "https://github.com/lance-format/lance.git" }
-lance-file = { "version" = "=12.0.0-beta.2", "tag" = "v12.0.0-beta.2", "git" = "https://github.com/lance-format/lance.git" }
+lance-file = { "version" = "=12.0.0-beta.14", "tag" = "v12.0.0-beta.14", "git" = "https://github.com/lance-format/lance.git" }
-lance-io = { "version" = "=12.0.0-beta.2", default-features = false, "tag" = "v12.0.0-beta.2", "git" = "https://github.com/lance-format/lance.git" }
+lance-io = { "version" = "=12.0.0-beta.14", default-features = false, "tag" = "v12.0.0-beta.14", "git" = "https://github.com/lance-format/lance.git" }
-lance-index = { "version" = "=12.0.0-beta.2", "tag" = "v12.0.0-beta.2", "git" = "https://github.com/lance-format/lance.git" }
+lance-index = { "version" = "=12.0.0-beta.14", "tag" = "v12.0.0-beta.14", "git" = "https://github.com/lance-format/lance.git" }
-lance-linalg = { "version" = "=12.0.0-beta.2", "tag" = "v12.0.0-beta.2", "git" = "https://github.com/lance-format/lance.git" }
+lance-linalg = { "version" = "=12.0.0-beta.14", "tag" = "v12.0.0-beta.14", "git" = "https://github.com/lance-format/lance.git" }
-lance-namespace = { "version" = "=12.0.0-beta.2", "tag" = "v12.0.0-beta.2", "git" = "https://github.com/lance-format/lance.git" }
+lance-namespace = { "version" = "=12.0.0-beta.14", "tag" = "v12.0.0-beta.14", "git" = "https://github.com/lance-format/lance.git" }
-lance-namespace-impls = { "version" = "=12.0.0-beta.2", default-features = false, "tag" = "v12.0.0-beta.2", "git" = "https://github.com/lance-format/lance.git" }
+lance-namespace-impls = { "version" = "=12.0.0-beta.14", default-features = false, "tag" = "v12.0.0-beta.14", "git" = "https://github.com/lance-format/lance.git" }
-lance-table = { "version" = "=12.0.0-beta.2", "tag" = "v12.0.0-beta.2", "git" = "https://github.com/lance-format/lance.git" }
+lance-table = { "version" = "=12.0.0-beta.14", "tag" = "v12.0.0-beta.14", "git" = "https://github.com/lance-format/lance.git" }
-lance-testing = { "version" = "=12.0.0-beta.2", "tag" = "v12.0.0-beta.2", "git" = "https://github.com/lance-format/lance.git" }
+lance-testing = { "version" = "=12.0.0-beta.14", "tag" = "v12.0.0-beta.14", "git" = "https://github.com/lance-format/lance.git" }
-lance-datafusion = { "version" = "=12.0.0-beta.2", "tag" = "v12.0.0-beta.2", "git" = "https://github.com/lance-format/lance.git" }
+lance-datafusion = { "version" = "=12.0.0-beta.14", "tag" = "v12.0.0-beta.14", "git" = "https://github.com/lance-format/lance.git" }
-lance-encoding = { "version" = "=12.0.0-beta.2", "tag" = "v12.0.0-beta.2", "git" = "https://github.com/lance-format/lance.git" }
+lance-encoding = { "version" = "=12.0.0-beta.14", "tag" = "v12.0.0-beta.14", "git" = "https://github.com/lance-format/lance.git" }
-lance-arrow = { "version" = "=12.0.0-beta.2", "tag" = "v12.0.0-beta.2", "git" = "https://github.com/lance-format/lance.git" }
+lance-arrow = { "version" = "=12.0.0-beta.14", "tag" = "v12.0.0-beta.14", "git" = "https://github.com/lance-format/lance.git" }
 lancedb = { path = "rust/lancedb", default-features = false }
 ahash = "0.8"
 # Note that this one does not include pyarrow
@@ -39,6 +39,7 @@ arrow-ord = "58.0.0"
 arrow-schema = "58.0.0"
 arrow-select = "58.0.0"
 arrow-cast = "58.0.0"
+arrow-flight = { version = "58.0.0", features = ["flight-sql-experimental"] }
 async-trait = "0"
 bytes = "1"
 datafusion = { version = "54.0.0", default-features = false }
@@ -71,7 +72,8 @@ serde = "1"
 serde_json = "1"
 tempfile = "3.5.0"
 tokio = { version = "1.23", features = ["rt-multi-thread", "sync"] }
-uuid = { version = "1.7.0", features = ["v4"] }
+tonic = { version = "0.14", features = ["tls-native-roots", "tls-ring"] }
+uuid = { version = "1.7.0", features = ["v4", "v7"] }
 chrono = { version = "0.4", default-features = false, features = ["clock"] }
 
 [profile.ci]

Makefile

@@ -5,5 +5,5 @@ licenses:
 	cd python && cargo about generate ../about.hbs -o RUST_THIRD_PARTY_LICENSES.html -c ../about.toml
 	cd python && uv sync --all-extras && uv tool run pip-licenses --python .venv/bin/python --format=markdown --with-urls --output-file=PYTHON_THIRD_PARTY_LICENSES.md
 	cd nodejs && cargo about generate ../about.hbs -o RUST_THIRD_PARTY_LICENSES.html -c ../about.toml
-	cd nodejs && npx license-checker --markdown --out NODEJS_THIRD_PARTY_LICENSES.md
+	cd nodejs && pnpm dlx license-checker@25 --markdown --out NODEJS_THIRD_PARTY_LICENSES.md
 	cd java && ./mvnw license:aggregate-add-third-party -q

ci/update_lockfiles.sh

@@ -12,16 +12,12 @@ done
 # This updates the lockfile without building
 cargo metadata --quiet > /dev/null
 
-pushd nodejs || exit 1
-npm install --package-lock-only --silent
-popd
-
 if git diff --quiet --exit-code; then
   echo "No lockfile changes to commit; skipping amend."
 elif $AMEND; then
-  git add Cargo.lock nodejs/package-lock.json
+  git add Cargo.lock
   git commit --amend --no-edit
 else
-  git add Cargo.lock nodejs/package-lock.json
+  git add Cargo.lock
   git commit -m "Update lockfiles"
 fi

deny.toml

@@ -131,18 +131,13 @@ allow = [
     "BSD-3-Clause",
     "ISC",
     "Unicode-3.0",
-    "Unicode-DFS-2016",
     "Zlib",
     "CC0-1.0",
     "MPL-2.0",
     "BSL-1.0",
-    "OpenSSL",
     # 0BSD ("BSD Zero Clause") is effectively public domain — no attribution
     # required. Pulled in by `mock_instant`.
     "0BSD",
-    # bzip2-1.0.6 is the permissive upstream bzip2 license (BSD-like). Pulled
-    # in by `libbz2-rs-sys`, the pure-Rust bzip2 implementation.
-    "bzip2-1.0.6",
     # CDLA-Permissive-2.0 is a permissive data license used by `webpki-roots`
     # for the Mozilla CA root bundle. Data-only, distribution-compatible.
     "CDLA-Permissive-2.0",
@@ -150,12 +145,7 @@ allow = [
 confidence-threshold = 0.8
 # Per-crate license exceptions: allow a license for a specific crate only,
 # rather than globally via the `allow` list above.
-exceptions = [
+exceptions = []
-    # CDDL-1.0 (copyleft) is pulled in only as a dev/profiling dependency via
-    # `inferno` -> `pprof` -> `lance-testing`; it is a test dependency that we
-    # do not distribute, so scope the allowance to `inferno` alone.
-    { allow = ["CDDL-1.0"], crate = "inferno" },
-]
 # Crates whose license cannot be determined from Cargo metadata but whose
 # license we've manually confirmed from upstream. Keep this list minimal.
 [[licenses.clarify]]

docs/README.md

@@ -47,22 +47,24 @@ pytest -vv python/tests/docs
 
 ### Checking typescript examples
 
-The `@lancedb/lancedb` package must be built before running the tests:
+The examples depend on `@lancedb/lancedb` at `file:../dist`, so the package must be
+built before running the tests. This uses pnpm; see the
+[Typescript contributing guide](../nodejs/CONTRIBUTING.md) for the toolchain setup.
 
 ```shell
 pushd nodejs
-npm ci
+pnpm install
-npm run build
+pnpm build
 popd
 ```
 
-Then you can run the examples by going to the `nodejs/examples` directory and
+Then you can run the examples by going to the `nodejs/examples` directory, which is a
-running the tests like a normal npm package:
+separate pnpm package with its own lockfile:
 
 ```shell
 pushd nodejs/examples
-npm ci
+pnpm install
-npm test
+pnpm test
 popd
 ```
 
@@ -84,6 +86,7 @@ The new files should be checked into the repository.
 
 ```shell
 pushd nodejs
-npm run docs
+# `pnpm docs` would invoke pnpm's built-in `docs` command, not the script.
+pnpm run docs
 popd
 ```

docs/openapi.yml

@@ -446,6 +446,15 @@ paths:
               properties:
                 column:
                   type: string
+                name:
+                  type: string
+                  description: Optional name for the created index.
+                replace:
+                  type: boolean
+                  default: true
+                  description: |
+                    Whether to replace an existing index with the same resolved
+                    name. Defaults to true.
                 metric_type:
                   type: string
                   nullable: false

docs/package-lock.json

@@ -1,135 +0,0 @@
-{
-  "name": "lancedb-docs-test",
-  "version": "1.0.0",
-  "lockfileVersion": 3,
-  "requires": true,
-  "packages": {
-    "": {
-      "name": "lancedb-docs-test",
-      "version": "1.0.0",
-      "license": "Apache 2",
-      "dependencies": {
-        "apache-arrow": "file:../node/node_modules/apache-arrow",
-        "vectordb": "file:../node"
-      },
-      "devDependencies": {
-        "@types/node": "^20.11.8",
-        "typescript": "^5.3.3"
-      }
-    },
-    "../node": {
-      "name": "vectordb",
-      "version": "0.21.2-beta.0",
-      "cpu": [
-        "x64",
-        "arm64"
-      ],
-      "license": "Apache-2.0",
-      "os": [
-        "darwin",
-        "linux",
-        "win32"
-      ],
-      "dependencies": {
-        "@neon-rs/load": "^0.0.74",
-        "axios": "^1.4.0"
-      },
-      "devDependencies": {
-        "@neon-rs/cli": "^0.0.160",
-        "@types/chai": "^4.3.4",
-        "@types/chai-as-promised": "^7.1.5",
-        "@types/mocha": "^10.0.1",
-        "@types/node": "^18.16.2",
-        "@types/sinon": "^10.0.15",
-        "@types/temp": "^0.9.1",
-        "@types/uuid": "^9.0.3",
-        "@typescript-eslint/eslint-plugin": "^5.59.1",
-        "apache-arrow-old": "npm:apache-arrow@13.0.0",
-        "cargo-cp-artifact": "^0.1",
-        "chai": "^4.3.7",
-        "chai-as-promised": "^7.1.1",
-        "eslint": "^8.39.0",
-        "eslint-config-standard-with-typescript": "^34.0.1",
-        "eslint-plugin-import": "^2.26.0",
-        "eslint-plugin-n": "^15.7.0",
-        "eslint-plugin-promise": "^6.1.1",
-        "mocha": "^10.2.0",
-        "openai": "^4.24.1",
-        "sinon": "^15.1.0",
-        "temp": "^0.9.4",
-        "ts-node": "^10.9.1",
-        "ts-node-dev": "^2.0.0",
-        "typedoc": "^0.24.7",
-        "typedoc-plugin-markdown": "^3.15.3",
-        "typescript": "^5.1.0",
-        "uuid": "^9.0.0"
-      },
-      "optionalDependencies": {
-        "@lancedb/vectordb-darwin-arm64": "0.21.2-beta.0",
-        "@lancedb/vectordb-darwin-x64": "0.21.2-beta.0",
-        "@lancedb/vectordb-linux-arm64-gnu": "0.21.2-beta.0",
-        "@lancedb/vectordb-linux-x64-gnu": "0.21.2-beta.0",
-        "@lancedb/vectordb-win32-x64-msvc": "0.21.2-beta.0"
-      },
-      "peerDependencies": {
-        "@apache-arrow/ts": "^14.0.2",
-        "apache-arrow": "^14.0.2"
-      }
-    },
-    "../node/node_modules/apache-arrow": {
-      "version": "14.0.2",
-      "license": "Apache-2.0",
-      "dependencies": {
-        "@types/command-line-args": "5.2.0",
-        "@types/command-line-usage": "5.0.2",
-        "@types/node": "20.3.0",
-        "@types/pad-left": "2.1.1",
-        "command-line-args": "5.2.1",
-        "command-line-usage": "7.0.1",
-        "flatbuffers": "23.5.26",
-        "json-bignum": "^0.0.3",
-        "pad-left": "^2.1.0",
-        "tslib": "^2.5.3"
-      },
-      "bin": {
-        "arrow2csv": "bin/arrow2csv.js"
-      }
-    },
-    "node_modules/@types/node": {
-      "version": "20.11.8",
-      "resolved": "https://registry.npmjs.org/@types/node/-/node-20.11.8.tgz",
-      "integrity": "sha512-i7omyekpPTNdv4Jb/Rgqg0RU8YqLcNsI12quKSDkRXNfx7Wxdm6HhK1awT3xTgEkgxPn3bvnSpiEAc7a7Lpyow==",
-      "dev": true,
-      "dependencies": {
-        "undici-types": "~5.26.4"
-      }
-    },
-    "node_modules/apache-arrow": {
-      "resolved": "../node/node_modules/apache-arrow",
-      "link": true
-    },
-    "node_modules/typescript": {
-      "version": "5.3.3",
-      "resolved": "https://registry.npmjs.org/typescript/-/typescript-5.3.3.tgz",
-      "integrity": "sha512-pXWcraxM0uxAS+tN0AG/BF2TyqmHO014Z070UsJ+pFvYuRSq8KH8DmWpnbXe0pEPDHXZV3FcAbJkijJ5oNEnWw==",
-      "dev": true,
-      "bin": {
-        "tsc": "bin/tsc",
-        "tsserver": "bin/tsserver"
-      },
-      "engines": {
-        "node": ">=14.17"
-      }
-    },
-    "node_modules/undici-types": {
-      "version": "5.26.5",
-      "resolved": "https://registry.npmjs.org/undici-types/-/undici-types-5.26.5.tgz",
-      "integrity": "sha512-JlCMO+ehdEIKqlFxk6IfVoAUVmgz7cU7zD/h9XZ0qzeosSHmUJVOzSQvvYSYWXkFXC+IfLKSIffhv0sVZup6pA==",
-      "dev": true
-    },
-    "node_modules/vectordb": {
-      "resolved": "../node",
-      "link": true
-    }
-  }
-}

docs/package.json

@@ -1,20 +0,0 @@
-{
-  "name": "lancedb-docs-test",
-  "version": "1.0.0",
-  "description": "auto-generated tests from doc",
-  "author": "dev@lancedb.com",
-  "license": "Apache 2",
-  "dependencies": {
-    "apache-arrow": "file:../node/node_modules/apache-arrow",
-    "vectordb": "file:../node"
-  },
-  "scripts": {
-    "build": "tsc -b && cd ../node && npm run build-release",
-    "example": "npm run build && node",
-    "test": "npm run build && ls dist/*.js | xargs -n 1 node"
-  },
-  "devDependencies": {
-    "@types/node": "^20.11.8",
-    "typescript": "^5.3.3"
-  }
-}

docs/src/java/java.md

@@ -14,7 +14,7 @@ Add the following dependency to your `pom.xml`:
 <dependency>
     <groupId>com.lancedb</groupId>
     <artifactId>lancedb-core</artifactId>
-    <version>0.38.0-beta.11</version>
+    <version>0.39.0-beta.4</version>
 </dependency>
 ```
 

docs/src/js/classes/Connection.md

@@ -448,26 +448,6 @@ on the returned job to know when cleanup has finished.
 
 ***
 
-### getJob()
-
-```ts
-abstract getJob(jobId): Promise<null | JobDescription>
-```
-
-Describe a single server-side job by id.
-
-Resolves to `null` when the server has no such job.
-
-#### Parameters
-
-* **jobId**: `string`
-
-#### Returns
-
-`Promise`&lt;`null` \| [`JobDescription`](../interfaces/JobDescription.md)&gt;
-
-***
-
 ### isOpen()
 
 ```ts
@@ -482,48 +462,6 @@ Return true if the connection has not been closed
 
 ***
 
-### job()
-
-```ts
-abstract job(jobId): Job
-```
-
-A [Job](Job.md) handle for a server-side job by id.
-
-The handle is constructed without a server round trip; an unknown id
-surfaces when the handle is used. Dropping the handle has no effect on
-the job itself.
-
-#### Parameters
-
-* **jobId**: `string`
-
-#### Returns
-
-[`Job`](Job.md)
-
-***
-
-### jobHistory()
-
-```ts
-abstract jobHistory(jobId?): Promise<Table<any>>
-```
-
-The lifecycle event history of a server-side job, as an Arrow table.
-
-Lists history across all jobs when `jobId` is omitted.
-
-#### Parameters
-
-* **jobId?**: `string`
-
-#### Returns
-
-`Promise`&lt;`Table`&lt;`any`&gt;&gt;
-
-***
-
 ### listJobs()
 
 ```ts
@@ -648,6 +586,30 @@ A page of table names and an
 
 ***
 
+### openJob()
+
+```ts
+abstract openJob(jobId): Promise<Job>
+```
+
+Open a server-side job by id, returning a handle with its record already
+populated. Rejects when the server has no such job, the way
+[Connection.openTable](Connection.md#opentable) does for a missing table.
+
+The returned [Job](Job.md) answers for its own state, specification,
+result, failure and event history, so there is no separate
+connection-level call for any of them.
+
+#### Parameters
+
+* **jobId**: `string`
+
+#### Returns
+
+`Promise`&lt;[`Job`](Job.md)&gt;
+
+***
+
 ### openMaterializedView()
 
 ```ts

docs/src/js/classes/Job.md

@@ -8,19 +8,46 @@
 
 A handle to an operation that may still be running.
 
-## Constructors
+The operation may already be complete when the handle is created.
 
-### new Job()
+The detail getters read what the handle last observed. Submitting an
+operation returns only a job id, so populating them eagerly would cost an
+extra round trip on every call:
+
+- [Job.refresh](Job.md#refresh) and [Job.status](Job.md#status) fetch the whole record.
+- [Job.wait](Job.md#wait) records the terminal state it establishes, but not the
+  rest of the record.
+- Everything is null until one of those runs.
+
+## Accessors
+
+### creationMs
 
 ```ts
-new Job(): Job
+get creationMs(): null | number
 ```
 
+When the job was created, in milliseconds since the epoch.
+
 #### Returns
 
-[`Job`](Job.md)
+`null` \| `number`
 
-## Accessors
+***
+
+### failure
+
+```ts
+get failure(): null | JobFailureInfo
+```
+
+Why the job failed, when it failed and the server reports a reason.
+
+#### Returns
+
+`null` \| [`JobFailureInfo`](../interfaces/JobFailureInfo.md)
+
+***
 
 ### id
 
@@ -28,8 +55,69 @@ new Job(): Job
 get id(): null | string
 ```
 
-Identifies the operation on the server that is running it. Operations
+Identifies the operation on the server that is running it.
-that run in this process have no server id. The value is opaque.
+
+Operations that run in this process have no server id. The value is
+opaque: parsing it or storing it to resume the job later is not supported.
+
+#### Returns
+
+`null` \| `string`
+
+***
+
+### jobType
+
+```ts
+get jobType(): null | string
+```
+
+The job's type, as the server names it. Null for an in-process job, which
+has no server-side record.
+
+#### Returns
+
+`null` \| `string`
+
+***
+
+### result
+
+```ts
+get result(): any
+```
+
+The job-type-specific terminal result. Null until the job succeeds, so a
+job that never terminates reports its progress through [Job.events](Job.md#events)
+instead.
+
+#### Returns
+
+`any`
+
+***
+
+### spec
+
+```ts
+get spec(): any
+```
+
+The job-type-specific specification it was submitted with.
+
+#### Returns
+
+`any`
+
+***
+
+### state
+
+```ts
+get state(): null | string
+```
+
+The last observed lifecycle state, without contacting the backend.
 
 #### Returns
 
@@ -51,18 +139,61 @@ Request cancellation. Cancelling a finished operation is a no-op.
 
 ***
 
+### events()
+
+```ts
+events(options?): Promise<Table<any>>
+```
+
+This job's recorded lifecycle events.
+
+Where the getters above report a terminal result only once the job reaches
+one, events are written as the job runs and outlive the workers that
+produced them. A distributed job records a `claim`/`claim_complete` pair
+per unit of work, each carrying `rows_processed`, so a job that never
+finishes still accounts for what it did.
+
+The server caps results at 1000 rows by default and 10,000 at most, and
+truncates without saying so, so pass `limit` for a job that emits an event
+per fragment. `filter` is a SQL-like expression over the `state`,
+`updated_by`, `emitted_from`, `emitted_by`, and `claim_entity` columns.
+
+#### Parameters
+
+* **options?**: [`JobEventsOptions`](../interfaces/JobEventsOptions.md)
+
+#### Returns
+
+`Promise`&lt;`Table`&lt;`any`&gt;&gt;
+
+***
+
+### refresh()
+
+```ts
+refresh(): Promise<void>
+```
+
+Ask the backend for this job's current state, and for a server-side job
+its full record, then cache it for the getters above.
+
+#### Returns
+
+`Promise`&lt;`void`&gt;
+
+***
+
 ### status()
 
 ```ts
 status(): Promise<string>
 ```
 
-The operation's current lifecycle state: "running", "finished",
+The operation's current lifecycle state: "running", "finished", "failed",
-"failed", or "cancelled".
+or "cancelled".
 
-A point snapshot; unlike [Job.wait](Job.md#wait) it does not block or reject
+A point snapshot; unlike [Job.wait](Job.md#wait) it does not block or reject on a
-on a terminal failure state. States a newer server reports that this
+terminal failure state. Also refreshes the getters above.
-client version does not know pass through as-is.
 
 #### Returns
 
@@ -70,6 +201,22 @@ client version does not know pass through as-is.
 
 ***
 
+### toString()
+
+```ts
+toString(): string
+```
+
+Every field the handle currently knows, one per line, with the JSON
+payloads indented -- a refresh job's spec and result are the point of
+printing it.
+
+#### Returns
+
+`string`
+
+***
+
 ### wait()
 
 ```ts

docs/src/js/classes/Table.md

@@ -676,9 +676,17 @@ List all the versions of the table
 abstract mergeInsert(on): MergeInsertBuilder
 ```
 
+Create a [MergeInsertBuilder](MergeInsertBuilder.md), which combines new data with the
+existing table in a single transaction — inserting, updating and deleting
+rows depending on how they match.
+
 #### Parameters
 
 * **on**: `string` \| `string`[]
+    The column, or columns, to match source rows against target
+    rows on. Typically a key or id column. Several columns match on the
+    composite key: a source row updates a target row only when it agrees on
+    every one of them.
 
 #### Returns
 

docs/src/js/globals.md

@@ -96,7 +96,7 @@
 - [IvfFlatOptions](interfaces/IvfFlatOptions.md)
 - [IvfPqOptions](interfaces/IvfPqOptions.md)
 - [IvfRqOptions](interfaces/IvfRqOptions.md)
-- [JobDescription](interfaces/JobDescription.md)
+- [JobEventsOptions](interfaces/JobEventsOptions.md)
 - [JobFailureInfo](interfaces/JobFailureInfo.md)
 - [JobInfo](interfaces/JobInfo.md)
 - [ListNamespacesOptions](interfaces/ListNamespacesOptions.md)

docs/src/js/interfaces/JobDescription.md

@@ -1,66 +0,0 @@
-[**@lancedb/lancedb**](../README.md) • **Docs**
-
-***
-
-[@lancedb/lancedb](../globals.md) / JobDescription
-
-# Interface: JobDescription
-
-A described job from `Connection.getJob`.
-
-## Properties
-
-### creationMs
-
-```ts
-creationMs: number;
-```
-
-When the job was created, in milliseconds since the epoch.
-
-***
-
-### failure?
-
-```ts
-optional failure: JobFailureInfo;
-```
-
-Why the job failed, when the job is failed and the server reports a
-reason.
-
-***
-
-### jobId
-
-```ts
-jobId: string;
-```
-
-***
-
-### jobType
-
-```ts
-jobType: string;
-```
-
-***
-
-### specJson?
-
-```ts
-optional specJson: string;
-```
-
-The job-type-specific specification as a JSON string, when present.
-
-***
-
-### state
-
-```ts
-state: string;
-```
-
-Lifecycle state: "running", "finished", "failed", or "cancelled".

docs/src/js/interfaces/JobEventsOptions.md

@@ -0,0 +1,29 @@
+[**@lancedb/lancedb**](../README.md) • **Docs**
+
+***
+
+[@lancedb/lancedb](../globals.md) / JobEventsOptions
+
+# Interface: JobEventsOptions
+
+Which of a job's events [Job.events](../classes/Job.md#events) returns.
+
+## Properties
+
+### filter?
+
+```ts
+optional filter: string;
+```
+
+SQL-like filter over the event columns.
+
+***
+
+### limit?
+
+```ts
+optional limit: number;
+```
+
+Maximum event rows to return, up to the server maximum of 10,000.

docs/src/js/interfaces/JobInfo.md

@@ -26,7 +26,7 @@ When the job was created, in milliseconds since the epoch.
 jobId: string;
 ```
 
-The job id -- what `Connection.getJob` and `Connection.cancelJob`
+The job id -- what `Connection.openJob` and `Connection.cancelJob`
 accept.
 
 ***

docs/src/js/interfaces/MaterializedViewDefinition.md

@@ -50,6 +50,16 @@ projections: [string, string][];
 
 ***
 
+### sourceNamespace
+
+```ts
+sourceNamespace: string[];
+```
+
+Namespace holding the source table; empty is the root namespace.
+
+***
+
 ### sourceTable
 
 ```ts

docs/src/python/python.md

@@ -28,6 +28,59 @@ is also an [asynchronous API client](#connections-asynchronous).
 
 ::: lancedb.Session
 
+## Remote SQL
+
+Submit SQL against a remote LanceDB database through the connection.
+The connected database and `default_namespace_path=["public"]` are used for
+unqualified tables. Fully qualified references can still query other databases
+and namespaces available to the same deployment. `execute_query` returns a
+reader as soon as its initial result stream is available. `execute_query_async`
+returns a query handle immediately; use it to inspect progress, open a reader,
+or cancel the query. The SQL client is initialized by the first query and
+retained for the lifetime of the remote connection. Query ids are random,
+connection-scoped references rather than encoded SQL or durable resume tokens:
+
+```python
+import lancedb
+
+db = lancedb.connect(
+    "db://analytics",
+    api_key="ldb_...",
+    host_override="https://api.example.com",
+    sql_host_override="grpc+tls://sql.example.com:10026",
+)
+reader = db.execute_query(
+    """
+    SELECT events.id, accounts.name
+    FROM analytics.public.events AS events
+    JOIN users.public.accounts AS accounts ON events.user_id = accounts.id
+    """,
+    default_namespace_path=["public"],
+)
+for batch in reader:
+    print(batch.num_rows)
+
+query = db.execute_query_async("SELECT * FROM events")
+print(query.id)
+print(query.describe().status)
+for batch in query.reader():
+    print(batch.num_rows)
+
+# The async connection exposes the same lifecycle without blocking:
+# async_db = await lancedb.connect_async(
+#     "db://analytics",
+#     api_key="ldb_...",
+#     host_override="https://api.example.com",
+#     sql_host_override="grpc+tls://sql.example.com:10026",
+# )
+# reader = await async_db.execute_query("SELECT * FROM events")
+# query = await async_db.execute_query_async("SELECT * FROM events")
+# description = await async_db.describe_query(query.id)
+# async for batch in await query.reader():
+#     print(batch.num_rows)
+# await query.cancel()
+```
+
 ## Namespaces (Synchronous)
 
 A namespace-backed connection resolves tables through a
@@ -72,6 +125,10 @@ listing a storage directory.
 
 ::: lancedb.functions.UdfDefinition
 
+::: lancedb.secrets.EnvVarSecret
+
+::: lancedb.secrets.SecretInfo
+
 ::: lancedb.functions.FunctionRegistrationRequest
 
 ::: lancedb.functions.FunctionArtifactRequest
@@ -94,6 +151,8 @@ listing a storage directory.
 
 ::: lancedb.functions.OutputMapping
 
+::: lancedb.functions.AssignmentMapping
+
 ::: lancedb.functions.FunctionBinding
 
 ::: lancedb.functions.RefreshColumnResult
@@ -102,6 +161,18 @@ listing a storage directory.
 
 ::: lancedb.job.AsyncJob
 
+::: lancedb.job.JobInfo
+
+::: lancedb.job.JobDescription
+
+::: lancedb.job.JobFailureInfo
+
+::: lancedb.sql.Query
+
+::: lancedb.sql.AsyncQuery
+
+::: lancedb.sql.QueryDescription
+
 ## Materialized Views (Synchronous)
 
 ::: lancedb.materialized_view.MaterializedView
@@ -249,6 +320,12 @@ still work. Queries return descriptors. Call
 
 ::: lancedb.exceptions.MissingColumnError
 
+::: lancedb.exceptions.JobNotFoundError
+
+::: lancedb.exceptions.JobFailedError
+
+::: lancedb.exceptions.JobCancelledError
+
 ## Integrations
 
 ## Pydantic

docs/tsconfig.json

@@ -1,17 +0,0 @@
-{
-  "include": [
-    "src/*.ts",
-  ],
-  "compilerOptions": {
-    "target": "es2022",
-    "module": "nodenext",
-    "declaration": true,
-    "outDir": "./dist",
-    "strict": true,
-    "allowJs": true,
-    "resolveJsonModule": true,
-  },
-  "exclude": [
-    "./dist/*",
-  ]
-}

java/lancedb-core/pom.xml

@@ -8,7 +8,7 @@
     <parent>
       <groupId>com.lancedb</groupId>
       <artifactId>lancedb-parent</artifactId>
-      <version>0.38.0-beta.11</version>
+      <version>0.39.0-beta.4</version>
       <relativePath>../pom.xml</relativePath>
     </parent>
 

java/pom.xml

@@ -6,7 +6,7 @@
 
     <groupId>com.lancedb</groupId>
     <artifactId>lancedb-parent</artifactId>
-    <version>0.38.0-beta.11</version>
+    <version>0.39.0-beta.4</version>
     <packaging>pom</packaging>
     <name>${project.artifactId}</name>
     <description>LanceDB Java SDK Parent POM</description>
@@ -28,7 +28,7 @@
     <properties>
         <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
         <arrow.version>15.0.0</arrow.version>
-        <lance-core.version>12.0.0-beta.2</lance-core.version>
+        <lance-core.version>12.0.0-beta.14</lance-core.version>
         <spotless.skip>false</spotless.skip>
         <spotless.version>2.30.0</spotless.version>
         <spotless.java.googlejavaformat.version>1.7</spotless.java.googlejavaformat.version>

nodejs/Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "lancedb-nodejs"
 edition.workspace = true
-version = "0.38.0-beta.11"
+version = "0.39.0-beta.4"
 publish = false
 license.workspace = true
 description.workspace = true

nodejs/__test__/materialized_view.test.ts

@@ -48,6 +48,28 @@ describe("materialized views", () => {
     expect(definitionFromMetadata(safe, "v").limit).toBe(42);
   });
 
+  it("reads the namespaced select kind and refuses unknown kinds", () => {
+    // "namespaced_select" is the namespaced form of "select": same shape, a
+    // separate kind so readers that predate it refuse instead of resolving
+    // the source at the root.
+    const namespaced = new Map([
+      [
+        DEFINITION_META_KEY,
+        '{"kind":"namespaced_select","source_table":"people","source_namespace":["ns"]}',
+      ],
+    ]);
+    const definition = definitionFromMetadata(namespaced, "v");
+    expect(definition.sourceTable).toBe("people");
+    expect(definition.sourceNamespace).toEqual(["ns"]);
+
+    const unknown = new Map([
+      [DEFINITION_META_KEY, '{"kind":"select_v3","source_table":"people"}'],
+    ]);
+    expect(() => definitionFromMetadata(unknown, "v")).toThrow(
+      /cannot refresh/,
+    );
+  });
+
   it("creates, refreshes and queries a view", async () => {
     const view = await db.createMaterializedView("adults", "people", {
       select: ["name", ["shout", "upper(name)"]],

nodejs/__test__/package.test.ts

@@ -5,8 +5,8 @@ import packageJson = require("../package.json");
 
 describe("package metadata", () => {
   it("requires Node.js type declarations compatible with the runtime", () => {
-    expect(packageJson.engines.node).toBe(">= 18");
+    expect(packageJson.engines.node).toBe(">= 22");
-    expect(packageJson.peerDependencies["@types/node"]).toBe(">=18");
+    expect(packageJson.peerDependencies["@types/node"]).toBe(">=22");
     expect(packageJson.peerDependenciesMeta["@types/node"]).toEqual({
       optional: true,
     });

nodejs/__test__/remote.test.ts

@@ -3,6 +3,7 @@
 
 import * as http from "http";
 import { RequestListener } from "http";
+import packageJson = require("../package.json");
 import {
   ClientConfig,
   Connection,
@@ -70,7 +71,13 @@ async function withMockDatabase(
   try {
     await callback(db);
   } finally {
-    server.close();
+    // `close()` alone leaves the port bound until keep-alive sockets drain, so
+    // a single failing test would cascade into EADDRINUSE for every test after
+    // it. Destroy the connections and wait for the port to actually be free.
+    await new Promise<void>((resolve) => {
+      server.closeAllConnections();
+      server.close(() => resolve());
+    });
   }
 }
 
@@ -131,7 +138,7 @@ describe("remote connection", () => {
       (req, res) => {
         expect(req.headers["x-api-key"]).toEqual("fake");
         expect(req.headers["user-agent"]).toEqual(
-          `LanceDB-Node-Client/${process.env.npm_package_version}`,
+          `LanceDB-Node-Client/${packageJson.version}`,
         );
 
         const body = JSON.stringify({ tables: [] });
@@ -932,6 +939,7 @@ describe("remote connection jobs surface", () => {
     const { tableFromArrays, tableToIPC } = await import("apache-arrow");
     const eventsTable = tableFromArrays({ state: ["created", "succeeded"] });
     const eventsBody = Buffer.from(tableToIPC(eventsTable, "stream"));
+    const queryEventsPayloads: Record<string, unknown>[] = [];
 
     await withMockDatabase(
       (req, res) => {
@@ -960,6 +968,16 @@ describe("remote connection jobs surface", () => {
                 );
             }
           } else if (req.url === "/v1/jobs/describe") {
+            if (payload["job_id"] === "job-2") {
+              res
+                .writeHead(200, { "Content-Type": "application/json" })
+                .end(
+                  '{"job_id": "job-2", "job_type": "refresh_column", ' +
+                    '"job_state": "DONE", "creation_ms": 2000, ' +
+                    '"result": {"rows_assigned": 1000000}}',
+                );
+              return;
+            }
             if (payload["job_id"] !== "job-1") {
               res.writeHead(404).end("no such job");
               return;
@@ -981,6 +999,7 @@ describe("remote connection jobs surface", () => {
               .writeHead(200, { "Content-Type": "application/json" })
               .end('{"job_id": "job-1"}');
           } else if (req.url === "/v1/jobs/query_events") {
+            queryEventsPayloads.push(payload);
             res
               .writeHead(200, {
                 "Content-Type": "application/vnd.apache.arrow.stream",
@@ -997,22 +1016,65 @@ describe("remote connection jobs surface", () => {
         expect(jobs[0].state).toEqual("running");
         expect(jobs[1].state).toEqual("finished");
 
-        const description = await db.getJob("job-1");
-        expect(description?.state).toEqual("failed");
-        expect(JSON.parse(description?.specJson ?? "")).toEqual({
-          column: "vec",
-        });
-        expect(description?.failure?.message).toEqual("worker died");
-        expect(await db.getJob("missing")).toBeNull();
-
         expect(await db.cancelJob("job-1")).toBe(true);
         expect(await db.cancelJob("missing")).toBe(false);
 
-        const history = await db.jobHistory("job-1");
+        // Opening a job hands back a populated handle; a missing one rejects.
-        expect(history.numRows).toEqual(2);
+        await expect(db.openJob("missing")).rejects.toThrow("not found");
+        const finished = await db.openJob("job-2");
+        expect(finished.state).toEqual("finished");
+        expect(finished.result).toEqual({
+          // biome-ignore lint/style/useNamingConvention: snake_case mandated by the server wire format
+          rows_assigned: 1000000,
+        });
 
-        const job = db.job("job-1");
+        const job = await db.openJob("job-1");
         expect(job.id).toEqual("job-1");
+
+        // openJob already populated the handle; refresh() re-reads it.
+        expect(job.state).toEqual("failed");
+        await job.refresh();
+        expect(job.state).toEqual("failed");
+        expect(job.jobType).toEqual("create_index");
+        expect(job.creationMs).toEqual(1000);
+        expect(job.spec).toEqual({ column: "vec" });
+        expect(job.result).toBeNull();
+        expect(job.failure?.message).toEqual("worker died");
+
+        // The handle reaches its own events, supplying its job id.
+        const jobEvents = await job.events({
+          limit: 500,
+          filter: "state = 'claim_complete'",
+        });
+        expect(jobEvents.numRows).toEqual(2);
+        expect(queryEventsPayloads.pop()).toEqual({
+          // biome-ignore lint/style/useNamingConvention: snake_case mandated by the server wire format
+          job_id: "job-1",
+          limit: 500,
+          filter: "state = 'claim_complete'",
+        });
+
+        // Printing lays every known field out on its own line, with the JSON
+        // payloads indented rather than crammed onto one line.
+        expect(`${job}`).toEqual(
+          [
+            "Job(",
+            '    id="job-1",',
+            '    state="failed",',
+            '    jobType="create_index",',
+            "    creationMs=1000,",
+            "    spec={",
+            '        "column": "vec"',
+            "    },",
+            "    failure={",
+            '        "phase": "execute",',
+            '        "message": "worker died",',
+            '        "retryable": true',
+            "    },",
+            ")",
+          ].join("\n"),
+        );
+
         expect(await job.status()).toEqual("failed");
         await expect(job.wait()).rejects.toThrow("worker died");
       },

nodejs/__test__/table.test.ts

@@ -737,11 +737,12 @@ it("should query documents with LangChain PDF metadata", async () => {
 
 describe("merge insert", () => {
   let tmpDir: tmp.DirResult;
+  let conn: Connection;
   let table: Table;
 
   beforeEach(async () => {
     tmpDir = tmp.dirSync({ unsafeCleanup: true });
-    const conn = await connect(tmpDir.name);
+    conn = await connect(tmpDir.name);
 
     table = await conn.createTable("some_table", [
       { a: 1, b: "a" },
@@ -779,6 +780,38 @@ describe("merge insert", () => {
 
     expect(result.map((row) => ({ ...row }))).toEqual(expected);
   });
+  test("upsert on a composite key", async () => {
+    const composite = await conn.createTable("composite", [
+      { shard: "a", id: 1, val: "x" },
+      { shard: "a", id: 2, val: "y" },
+      { shard: "b", id: 1, val: "z" },
+    ]);
+
+    // ("a", 1) matches an existing row and updates it. ("b", 2) agrees with an
+    // existing row on each key column separately but on neither pair, so it is
+    // an insert.
+    const mergeInsertRes = await composite
+      .mergeInsert(["shard", "id"])
+      .whenMatchedUpdateAll()
+      .whenNotMatchedInsertAll()
+      .execute([
+        { shard: "a", id: 1, val: "X" },
+        { shard: "b", id: 2, val: "W" },
+      ]);
+    expect(mergeInsertRes.numUpdatedRows).toBe(1);
+    expect(mergeInsertRes.numInsertedRows).toBe(1);
+
+    const result = (await composite.toArrow())
+      .toArray()
+      .sort((a, b) => a.shard.localeCompare(b.shard) || a.id - b.id);
+
+    expect(result.map((row) => ({ ...row }))).toEqual([
+      { shard: "a", id: 1, val: "X" },
+      { shard: "a", id: 2, val: "y" },
+      { shard: "b", id: 1, val: "z" },
+      { shard: "b", id: 2, val: "W" },
+    ]);
+  });
   test("conditional update", async () => {
     const newData = [
       { a: 2, b: "x" },

nodejs/examples/package.json

@@ -8,7 +8,8 @@
     "//1": "--experimental-vm-modules is needed to run jest with sentence-transformers",
     "//2": "--testEnvironment is needed to run jest with sentence-transformers",
     "//3": "See: https://github.com/huggingface/transformers.js/issues/57",
-    "test": "node --experimental-vm-modules node_modules/.bin/jest --testEnvironment jest-environment-node-single-context --verbose",
+    "//4": "jest is invoked by its JS entry, not node_modules/.bin/jest: under pnpm that path is a shell shim, which `node` cannot execute",
+    "test": "node --experimental-vm-modules node_modules/jest/bin/jest.js --testEnvironment jest-environment-node-single-context --verbose",
     "lint": "biome check *.ts && biome format *.ts",
     "lint-ci": "biome ci .",
     "lint-fix": "biome check --write *.ts && pnpm format",

nodejs/lancedb/connection.ts

@@ -1,7 +1,6 @@
 // SPDX-License-Identifier: Apache-2.0
 // SPDX-FileCopyrightText: Copyright The LanceDB Authors
 
-import { tableFromIPC } from "apache-arrow";
 import {
   Data,
   SchemaLike,
@@ -16,6 +15,7 @@ import {
   makeEmptyTable,
 } from "./arrow";
 import { EmbeddingFunctionConfig, getRegistry } from "./embedding/registry";
+import { Job } from "./job";
 import {
   MaterializedView,
   MaterializedViewSelect,
@@ -27,8 +27,6 @@ import type {
   CreateNamespaceResponse,
   DescribeNamespaceResponse,
   DropNamespaceResponse,
-  Job,
-  JobDescription,
   JobInfo,
   ListNamespacesResponse,
   ListTablesResponse,
@@ -557,24 +555,19 @@ export abstract class Connection {
   ): Promise<void>;
 
   /**
-   * A {@link Job} handle for a server-side job by id.
+   * Open a server-side job by id, returning a handle with its record already
+   * populated. Rejects when the server has no such job, the way
+   * {@link Connection.openTable} does for a missing table.
    *
-   * The handle is constructed without a server round trip; an unknown id
+   * The returned {@link Job} answers for its own state, specification,
-   * surfaces when the handle is used. Dropping the handle has no effect on
+   * result, failure and event history, so there is no separate
-   * the job itself.
+   * connection-level call for any of them.
    */
-  abstract job(jobId: string): Job;
+  abstract openJob(jobId: string): Promise<Job>;
 
   /** List server-side jobs across the database's tables. */
   abstract listJobs(): Promise<JobInfo[]>;
 
-  /**
-   * Describe a single server-side job by id.
-   *
-   * Resolves to `null` when the server has no such job.
-   */
-  abstract getJob(jobId: string): Promise<JobDescription | null>;
-
   /**
    * Request cancellation of a server-side job by id.
    *
@@ -582,13 +575,6 @@ export abstract class Connection {
    * such job exists. Cancelling an already-terminal job is a no-op success.
    */
   abstract cancelJob(jobId: string): Promise<boolean>;
-
-  /**
-   * The lifecycle event history of a server-side job, as an Arrow table.
-   *
-   * Lists history across all jobs when `jobId` is omitted.
-   */
-  abstract jobHistory(jobId?: string): Promise<ArrowTable>;
 }
 
 /** @hideconstructor */
@@ -869,7 +855,7 @@ export class LocalConnection extends Connection {
   }
 
   async dropTableAsync(name: string, namespacePath?: string[]): Promise<Job> {
-    return this.inner.dropTableAsync(name, namespacePath ?? []);
+    return new Job(await this.inner.dropTableAsync(name, namespacePath ?? []));
   }
 
   async dropAllTables(namespacePath?: string[]): Promise<void> {
@@ -928,29 +914,17 @@ export class LocalConnection extends Connection {
     );
   }
 
-  job(jobId: string): Job {
+  async openJob(jobId: string): Promise<Job> {
-    return this.inner.job(jobId);
+    return new Job(await this.inner.openJob(jobId));
   }
 
   async listJobs(): Promise<JobInfo[]> {
     return this.inner.listJobs();
   }
 
-  async getJob(jobId: string): Promise<JobDescription | null> {
-    return this.inner.getJob(jobId);
-  }
-
   async cancelJob(jobId: string): Promise<boolean> {
     return this.inner.cancelJob(jobId);
   }
-
-  async jobHistory(jobId?: string): Promise<ArrowTable> {
-    const buf = await this.inner.jobHistory(jobId);
-    if (buf.length === 0) {
-      return new ArrowTable();
-    }
-    return tableFromIPC(buf);
-  }
 }
 
 /**

nodejs/lancedb/index.ts

@@ -94,13 +94,9 @@ export {
   RenameTableOptions,
 } from "./connection";
 
-export {
+export { JobFailureInfo, JobInfo, Session } from "./native.js";
-  Job,
+
-  JobDescription,
+export { Job, JobEventsOptions } from "./job";
-  JobFailureInfo,
-  JobInfo,
-  Session,
-} from "./native.js";
 
 export {
   AutoQuery,

nodejs/lancedb/job.ts

@@ -0,0 +1,188 @@
+// SPDX-License-Identifier: Apache-2.0
+// SPDX-FileCopyrightText: Copyright The LanceDB Authors
+
+import { Table as ArrowTable, tableFromIPC } from "apache-arrow";
+import { JobFailureInfo, Job as NativeJob } from "./native";
+
+/** Which of a job's events {@link Job.events} returns. */
+export interface JobEventsOptions {
+  /** Maximum event rows to return, up to the server maximum of 10,000. */
+  limit?: number;
+  /** SQL-like filter over the event columns. */
+  filter?: string;
+}
+
+/**
+ * A handle to an operation that may still be running.
+ *
+ * The operation may already be complete when the handle is created.
+ *
+ * The detail getters read what the handle last observed. Submitting an
+ * operation returns only a job id, so populating them eagerly would cost an
+ * extra round trip on every call:
+ *
+ * - {@link Job.refresh} and {@link Job.status} fetch the whole record.
+ * - {@link Job.wait} records the terminal state it establishes, but not the
+ *   rest of the record.
+ * - Everything is null until one of those runs.
+ *
+ * @hideconstructor
+ */
+export class Job {
+  private readonly inner: NativeJob;
+
+  constructor(inner: NativeJob) {
+    this.inner = inner;
+  }
+
+  /**
+   * Identifies the operation on the server that is running it.
+   *
+   * Operations that run in this process have no server id. The value is
+   * opaque: parsing it or storing it to resume the job later is not supported.
+   */
+  get id(): string | null {
+    return this.inner.id ?? null;
+  }
+
+  /** The last observed lifecycle state, without contacting the backend. */
+  get state(): string | null {
+    return this.inner.state ?? null;
+  }
+
+  /**
+   * The job's type, as the server names it. Null for an in-process job, which
+   * has no server-side record.
+   */
+  get jobType(): string | null {
+    return this.inner.jobType ?? null;
+  }
+
+  /** When the job was created, in milliseconds since the epoch. */
+  get creationMs(): number | null {
+    return this.inner.creationMs ?? null;
+  }
+
+  /** The job-type-specific specification it was submitted with. */
+  // biome-ignore lint/suspicious/noExplicitAny: shape varies by job type
+  get spec(): any | null {
+    return parseJson(this.inner.specJson);
+  }
+
+  /**
+   * The job-type-specific terminal result. Null until the job succeeds, so a
+   * job that never terminates reports its progress through {@link Job.events}
+   * instead.
+   */
+  // biome-ignore lint/suspicious/noExplicitAny: shape varies by job type
+  get result(): any | null {
+    return parseJson(this.inner.resultJson);
+  }
+
+  /** Why the job failed, when it failed and the server reports a reason. */
+  get failure(): JobFailureInfo | null {
+    return this.inner.failure ?? null;
+  }
+
+  /**
+   * The operation's current lifecycle state: "running", "finished", "failed",
+   * or "cancelled".
+   *
+   * A point snapshot; unlike {@link Job.wait} it does not block or reject on a
+   * terminal failure state. Also refreshes the getters above.
+   */
+  async status(): Promise<string> {
+    return this.inner.status();
+  }
+
+  /** Wait until the operation reaches a terminal state. */
+  async wait(): Promise<void> {
+    return this.inner.wait();
+  }
+
+  /** Request cancellation. Cancelling a finished operation is a no-op. */
+  async cancel(): Promise<void> {
+    return this.inner.cancel();
+  }
+
+  /**
+   * Ask the backend for this job's current state, and for a server-side job
+   * its full record, then cache it for the getters above.
+   */
+  async refresh(): Promise<void> {
+    return this.inner.refresh();
+  }
+
+  /**
+   * This job's recorded lifecycle events.
+   *
+   * Where the getters above report a terminal result only once the job reaches
+   * one, events are written as the job runs and outlive the workers that
+   * produced them. A distributed job records a `claim`/`claim_complete` pair
+   * per unit of work, each carrying `rows_processed`, so a job that never
+   * finishes still accounts for what it did.
+   *
+   * The server caps results at 1000 rows by default and 10,000 at most, and
+   * truncates without saying so, so pass `limit` for a job that emits an event
+   * per fragment. `filter` is a SQL-like expression over the `state`,
+   * `updated_by`, `emitted_from`, `emitted_by`, and `claim_entity` columns.
+   */
+  async events(options?: JobEventsOptions): Promise<ArrowTable> {
+    const buf = await this.inner.events(options?.limit, options?.filter);
+    if (buf.length === 0) {
+      return new ArrowTable();
+    }
+    return tableFromIPC(buf);
+  }
+
+  /**
+   * Every field the handle currently knows, one per line, with the JSON
+   * payloads indented -- a refresh job's spec and result are the point of
+   * printing it.
+   */
+  toString(): string {
+    if (this.state === null) {
+      const known = this.id === null ? "" : `id=${JSON.stringify(this.id)}, `;
+      return `Job(${known}not refreshed)`;
+    }
+    const fields: string[] = [];
+    if (this.id !== null) {
+      fields.push(`id=${JSON.stringify(this.id)}`);
+    }
+    fields.push(`state=${JSON.stringify(this.state)}`);
+    if (this.jobType !== null) {
+      fields.push(`jobType=${JSON.stringify(this.jobType)}`);
+    }
+    if (this.creationMs !== null) {
+      fields.push(`creationMs=${this.creationMs}`);
+    }
+    for (const [name, value] of [
+      ["spec", this.spec],
+      ["result", this.result],
+    ] as const) {
+      if (value !== null) {
+        fields.push(`${name}=${indentJson(value)}`);
+      }
+    }
+    if (this.failure !== null) {
+      fields.push(`failure=${indentJson(this.failure)}`);
+    }
+    return `Job(${fields.map((field) => `\n${REPR_INDENT}${field},`).join("")}\n)`;
+  }
+
+  [Symbol.for("nodejs.util.inspect.custom")](): string {
+    return this.toString();
+  }
+}
+
+const REPR_INDENT = "    ";
+
+// biome-ignore lint/suspicious/noExplicitAny: shape varies by job type
+function indentJson(value: any): string {
+  return JSON.stringify(value, null, 4).replace(/\n/g, `\n${REPR_INDENT}`);
+}
+
+// biome-ignore lint/suspicious/noExplicitAny: shape varies by job type
+function parseJson(raw: string | null | undefined): any | null {
+  return raw === null || raw === undefined ? null : JSON.parse(raw);
+}

nodejs/lancedb/materialized_view.ts

@@ -19,6 +19,8 @@ export interface MaterializedViewDefinition {
   limit?: number;
   /** Source columns the projections and filter read. */
   inputs: string[];
+  /** Namespace holding the source table; empty is the root namespace. */
+  sourceNamespace: string[];
 }
 
 /**
@@ -78,7 +80,8 @@ export function definitionFromMetadata(
   }
   // biome-ignore lint/suspicious/noExplicitAny: raw JSON
   const value: any = JSON.parse(raw);
-  if (value.kind !== "select") {
+  // "namespaced_select" keeps older readers from resolving the source at root.
+  if (value.kind !== "select" && value.kind !== "namespaced_select") {
     throw new Error(
       `materialized view '${name}' is defined by '${value.kind}', which this ` +
         "version of lancedb cannot refresh",
@@ -103,6 +106,7 @@ export function definitionFromMetadata(
     filter: value.filter ?? undefined,
     limit,
     inputs: value.inputs ?? [],
+    sourceNamespace: value.source_namespace ?? [],
   };
 }
 

nodejs/lancedb/table.ts

@@ -19,6 +19,7 @@ import {
 
 import { EmbeddingFunctionConfig, getRegistry } from "./embedding/registry";
 import { IndexOptions } from "./indices";
+import { Job } from "./job";
 import { MergeInsertBuilder } from "./merge";
 import {
   AddColumnsResult,
@@ -30,7 +31,6 @@ import {
   DropColumnsResult,
   IndexConfig,
   IndexStatistics,
-  Job,
   LsmStats,
   Branches as NativeBranches,
   OptimizeStats,
@@ -919,6 +919,16 @@ export abstract class Table {
   /** Return the table as an arrow table */
   abstract toArrow(): Promise<ArrowTable>;
 
+  /**
+   * Create a {@link MergeInsertBuilder}, which combines new data with the
+   * existing table in a single transaction — inserting, updating and deleting
+   * rows depending on how they match.
+   *
+   * @param on - The column, or columns, to match source rows against target
+   * rows on. Typically a key or id column. Several columns match on the
+   * composite key: a source row updates a target row only when it agrees on
+   * every one of them.
+   */
   abstract mergeInsert(on: string | string[]): MergeInsertBuilder;
 
   /** List all the stats of a specified index
@@ -1114,13 +1124,15 @@ export class LocalTable extends Table {
   ): Promise<Job> {
     // biome-ignore lint/suspicious/noExplicitAny: skip
     const nativeIndex = (options?.config as any)?.inner;
-    return await this.inner.createIndexAsync(
+    return new Job(
-      nativeIndex,
+      await this.inner.createIndexAsync(
-      column,
+        nativeIndex,
-      options?.replace,
+        column,
-      options?.waitTimeoutSeconds,
+        options?.replace,
-      options?.name,
+        options?.waitTimeoutSeconds,
-      options?.train,
+        options?.name,
+        options?.train,
+      ),
     );
   }
 
@@ -1303,7 +1315,7 @@ export class LocalTable extends Table {
   }
 
   async refreshColumnAsync(column: string): Promise<Job> {
-    return await this.inner.refreshColumnAsync(column);
+    return new Job(await this.inner.refreshColumnAsync(column));
   }
 
   async refreshMaterializedView(

nodejs/npm/darwin-arm64/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-darwin-arm64",
-	"version": "0.38.0-beta.11",
+	"version": "0.39.0-beta.4",
 	"os": ["darwin"],
 	"cpu": ["arm64"],
 	"main": "lancedb.darwin-arm64.node",

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

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-arm64-gnu",
-	"version": "0.38.0-beta.11",
+	"version": "0.39.0-beta.4",
 	"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": "0.38.0-beta.11",
+	"version": "0.39.0-beta.4",
 	"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": "0.38.0-beta.11",
+	"version": "0.39.0-beta.4",
 	"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": "0.38.0-beta.11",
+	"version": "0.39.0-beta.4",
 	"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": "0.38.0-beta.11",
+  "version": "0.39.0-beta.4",
   "os": [
     "win32"
   ],

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

@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-win32-x64-msvc",
-	"version": "0.38.0-beta.11",
+	"version": "0.39.0-beta.4",
 	"os": ["win32"],
 	"cpu": ["x64"],
 	"main": "lancedb.win32-x64-msvc.node",

nodejs/package.json

@@ -11,7 +11,7 @@
     "ann"
   ],
   "private": false,
-  "version": "0.38.0-beta.11",
+  "version": "0.39.0-beta.4",
   "main": "dist/index.js",
   "exports": {
     ".": "./dist/index.js",
@@ -44,7 +44,7 @@
     "@biomejs/biome": "^1.7.3",
     "@jest/globals": "^29.7.0",
     "@napi-rs/cli": "3.7.0",
-    "@opentelemetry/sdk-metrics": "^1.30.0",
+    "@opentelemetry/sdk-metrics": "^2.10.0",
     "@types/axios": "^0.14.0",
     "@types/jest": "^29.1.2",
     "@types/node": "22.7.4",
@@ -56,7 +56,7 @@
     "eslint": "^8.57.0",
     "jest": "^29.7.0",
     "shx": "^0.3.4",
-    "tmp": "^0.2.3",
+    "tmp": "^0.2.7",
     "ts-jest": "^29.1.2",
     "typedoc": "0.26.4",
     "typedoc-plugin-markdown": "4.2.1",
@@ -67,7 +67,7 @@
     "timeout": "3m"
   },
   "engines": {
-    "node": ">= 18"
+    "node": ">= 22"
   },
   "packageManager": "pnpm@11.1.1",
   "cpu": ["x64", "arm64"],
@@ -101,7 +101,7 @@
     "openai": "4.29.2"
   },
   "peerDependencies": {
-    "@types/node": ">=18",
+    "@types/node": ">=22",
     "apache-arrow": ">=15.0.0 <=18.1.0"
   },
   "peerDependenciesMeta": {

nodejs/pnpm-workspace.yaml

@@ -16,3 +16,41 @@ allowBuilds:
   onnxruntime-node: true
   protobufjs: true
   sharp: true
+
+minimumReleaseAgeExclude:
+  - protobufjs@7.5.8
+  - tmp@0.2.6
+  - form-data@4.0.6
+  - tar@7.5.16
+  - markdown-it@14.1.2
+  - linkify-it@5.0.1
+  - js-yaml@3.15.0
+  - js-yaml@4.1.2
+  - protobufjs@7.6.1
+  - protobufjs@7.6.3
+  - '@babel/core@7.29.1'
+  - axios@1.18.0
+  - brace-expansion@2.1.2
+  - brace-expansion@1.1.16
+  - js-yaml@4.3.0
+  - tar@7.5.18
+  - tar@7.5.19
+  - tar@7.5.17
+  - protobufjs@7.6.5
+  - linkify-it@5.0.2
+  - sharp@0.35.0
+  - brace-expansion@1.1.17
+  - brace-expansion@2.1.3
+  - brace-expansion@2.1.4
+  - brace-expansion@1.1.18
+  - js-yaml@3.15.1
+  - js-yaml@4.3.1
+  - tar@7.5.21
+  - '@opentelemetry/core@2.8.0'
+
+# @huggingface/transformers pins sharp ^0.33.5 and no released version has moved
+# past ^0.34.5, all of which inherit the libvips CVEs in GHSA-f88m-g3jw-g9cj.
+# Force the patched line. sharp is only reached by transformers' image pipeline,
+# which LanceDB's text embedding function never uses.
+overrides:
+  sharp: ^0.35.4

nodejs/src/connection.rs

@@ -442,13 +442,15 @@ impl Connection {
         self.get_inner()?.drop_all_tables(&ns).await.default_error()
     }
 
-    /// A `Job` handle for a server-side job by id.
+    /// Open a server-side job by id, returning a handle with its record
+    /// already populated. Rejects when the server has no such job.
     ///
-    /// The handle is constructed without a server round trip; an unknown id
+    /// The returned handle answers for its own state, specification, result,
-    /// surfaces when the handle is used.
+    /// failure and event history, so there is no separate connection-level
-    #[napi]
+    /// call for any of them.
-    pub fn job(&self, job_id: String) -> napi::Result<crate::job::Job> {
+    #[napi(catch_unwind)]
-        let job = self.get_inner()?.job(job_id).default_error()?;
+    pub async fn open_job(&self, job_id: String) -> napi::Result<crate::job::Job> {
+        let job = self.get_inner()?.open_job(&job_id).await.default_error()?;
         Ok(crate::job::Job::new(job))
     }
 
@@ -459,17 +461,6 @@ impl Connection {
         Ok(jobs.into_iter().map(Into::into).collect())
     }
 
-    /// Describe a single server-side job by id. `null` when the server has
-    /// no such job.
-    #[napi(catch_unwind)]
-    pub async fn get_job(
-        &self,
-        job_id: String,
-    ) -> napi::Result<Option<crate::job::JobDescription>> {
-        let description = self.get_inner()?.get_job(&job_id).await.default_error()?;
-        Ok(description.map(Into::into))
-    }
-
     /// Request cancellation of a server-side job by id. Returns true if the
     /// server accepted the cancellation, false if no such job exists.
     #[napi(catch_unwind)]
@@ -477,34 +468,6 @@ impl Connection {
         self.get_inner()?.cancel_job(&job_id).await.default_error()
     }
 
-    /// The lifecycle event history of a server-side job (all jobs when
-    /// `job_id` is null), as an Arrow IPC stream buffer. Empty when there is
-    /// no history.
-    #[napi(catch_unwind)]
-    pub async fn job_history(&self, job_id: Option<String>) -> napi::Result<Buffer> {
-        let batches = self
-            .get_inner()?
-            .job_history(job_id.as_deref())
-            .await
-            .default_error()?;
-        let Some(first) = batches.first() else {
-            return Ok(Buffer::from(Vec::<u8>::new()));
-        };
-        let mut out = Vec::new();
-        let mut writer = arrow_ipc::writer::StreamWriter::try_new(&mut out, &first.schema())
-            .map_err(|e| napi::Error::from_reason(e.to_string()))?;
-        for batch in &batches {
-            writer
-                .write(batch)
-                .map_err(|e| napi::Error::from_reason(e.to_string()))?;
-        }
-        writer
-            .finish()
-            .map_err(|e| napi::Error::from_reason(e.to_string()))?;
-        drop(writer);
-        Ok(Buffer::from(out))
-    }
-
     #[napi(catch_unwind)]
     /// Describe a namespace and return its properties.
     pub async fn describe_namespace(

nodejs/src/job.rs

@@ -3,6 +3,9 @@
 
 use std::sync::Arc;
 
+use arrow_array::RecordBatch;
+use lancedb::job::JobEventsRequest;
+use napi::bindgen_prelude::Buffer;
 use napi_derive::napi;
 
 use crate::error::NapiErrorExt;
@@ -55,12 +58,98 @@ impl Job {
     pub async fn cancel(&self) -> napi::Result<()> {
         self.inner.cancel().await.default_error()
     }
+
+    /// Ask the backend for this job's current state, and for a server-side job
+    /// its full record, then cache it for the getters below.
+    ///
+    /// They are all null until this runs, because submitting an operation
+    /// returns only a job id. {@link Job.status} fetches the whole record too;
+    /// {@link Job.wait} records only the terminal state it establishes.
+    #[napi(catch_unwind)]
+    pub async fn refresh(&self) -> napi::Result<()> {
+        self.inner.refresh().await.default_error()
+    }
+
+    /// The last observed lifecycle state, without contacting the backend.
+    #[napi(getter)]
+    pub fn state(&self) -> Option<String> {
+        self.inner.state()
+    }
+
+    /// The job's type, as the server names it. Null for an in-process job,
+    /// which has no server-side record.
+    #[napi(getter)]
+    pub fn job_type(&self) -> Option<String> {
+        self.inner.job_type()
+    }
+
+    /// When the job was created, in milliseconds since the epoch.
+    #[napi(getter)]
+    pub fn creation_ms(&self) -> Option<i64> {
+        self.inner.creation_ms()
+    }
+
+    /// The job-type-specific specification as a JSON string, when present.
+    #[napi(getter)]
+    pub fn spec_json(&self) -> Option<String> {
+        self.inner.spec().map(|spec| spec.to_string())
+    }
+
+    /// The job-type-specific terminal result as a JSON string. Null until the
+    /// job succeeds, so a job that never terminates reports its progress
+    /// through {@link Job.events} instead.
+    #[napi(getter)]
+    pub fn result_json(&self) -> Option<String> {
+        self.inner.result().map(|result| result.to_string())
+    }
+
+    /// Why the job failed, when it failed and the server reports a reason.
+    #[napi(getter)]
+    pub fn failure(&self) -> Option<JobFailureInfo> {
+        self.inner.failure().map(|failure| JobFailureInfo {
+            phase: failure.phase,
+            message: failure.message,
+            retryable: failure.retryable,
+        })
+    }
+
+    /// This job's recorded lifecycle events, as an Arrow IPC stream buffer.
+    /// The TypeScript wrapper turns it into an Arrow table.
+    #[napi(catch_unwind)]
+    pub async fn events(&self, limit: Option<u32>, filter: Option<String>) -> napi::Result<Buffer> {
+        let batches = self
+            .inner
+            .events(JobEventsRequest { limit, filter })
+            .await
+            .default_error()?;
+        batches_to_ipc_buffer(&batches)
+    }
+}
+
+/// Serialise Arrow batches as a single IPC stream for the TypeScript layer.
+fn batches_to_ipc_buffer(batches: &[RecordBatch]) -> napi::Result<Buffer> {
+    let Some(first) = batches.first() else {
+        return Ok(Buffer::from(Vec::<u8>::new()));
+    };
+    let mut out = Vec::new();
+    let mut writer = arrow_ipc::writer::StreamWriter::try_new(&mut out, &first.schema())
+        .map_err(|e| napi::Error::from_reason(e.to_string()))?;
+    for batch in batches {
+        writer
+            .write(batch)
+            .map_err(|e| napi::Error::from_reason(e.to_string()))?;
+    }
+    writer
+        .finish()
+        .map_err(|e| napi::Error::from_reason(e.to_string()))?;
+    drop(writer);
+    Ok(Buffer::from(out))
 }
 
 /// A row from `Connection.listJobs`: one server-side job.
 #[napi(object)]
 pub struct JobInfo {
-    /// The job id -- what `Connection.getJob` and `Connection.cancelJob`
+    /// The job id -- what `Connection.openJob` and `Connection.cancelJob`
     /// accept.
     pub job_id: String,
     /// The table the job runs against, without URI or namespace.
@@ -91,36 +180,3 @@ pub struct JobFailureInfo {
     pub message: Option<String>,
     pub retryable: Option<bool>,
 }
-
-/// A described job from `Connection.getJob`.
-#[napi(object)]
-pub struct JobDescription {
-    pub job_id: String,
-    pub job_type: String,
-    /// Lifecycle state: "running", "finished", "failed", or "cancelled".
-    pub state: String,
-    /// When the job was created, in milliseconds since the epoch.
-    pub creation_ms: i64,
-    /// The job-type-specific specification as a JSON string, when present.
-    pub spec_json: Option<String>,
-    /// Why the job failed, when the job is failed and the server reports a
-    /// reason.
-    pub failure: Option<JobFailureInfo>,
-}
-
-impl From<lancedb::database::JobDescription> for JobDescription {
-    fn from(description: lancedb::database::JobDescription) -> Self {
-        Self {
-            job_id: description.job_id,
-            job_type: description.job_type,
-            state: description.state,
-            creation_ms: description.creation_ms,
-            spec_json: (!description.spec.is_null()).then(|| description.spec.to_string()),
-            failure: description.failure.map(|failure| JobFailureInfo {
-                phase: failure.phase,
-                message: failure.message,
-                retryable: failure.retryable,
-            }),
-        }
-    }
-}

nodejs/src/query.rs

@@ -664,7 +664,11 @@ impl JsFullTextQuery {
 }
 
 fn parse_fts_query(query: Object) -> napi::Result<FullTextSearchQuery> {
-    if let Ok(Some(query)) = query.get::<&JsFullTextQuery>("query") {
+    // `&JsFullTextQuery` recovers a native class reference through napi's borrow-tracked
+    // path, which is only usable from generated `#[napi]` argument conversion. This is a
+    // manual lookup on a nested `Object` property instead, so use `ClassInstance`, which
+    // unwraps the class without requiring a borrow scope.
+    if let Ok(Some(query)) = query.get::<ClassInstance<JsFullTextQuery>>("query") {
         Ok(FullTextSearchQuery::new_query(query.inner.clone()))
     } else if let Ok(Some(query_text)) = query.get::<String>("query") {
         let mut query_text = query_text;

python/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "lancedb-python"
-version = "0.38.0-beta.11"
+version = "0.39.0-beta.4"
 publish = false
 edition.workspace = true
 description = "Python bindings for LanceDB"
@@ -28,7 +28,7 @@ env_logger.workspace = true
 log.workspace = true
 # Maturin enables extension-module mode for Python builds. Keeping it out of
 # Cargo features lets Rust unit tests link against libpython.
-pyo3 = { version = "0.28", features = ["abi3-py310", "chrono"] }
+pyo3 = { version = "0.28", features = ["abi3-py310", "chrono", "uuid"] }
 chrono.workspace = true
 pyo3-async-runtimes = { version = "0.28", features = [
     "attributes",
@@ -40,6 +40,7 @@ serde.workspace = true
 serde_json.workspace = true
 snafu.workspace = true
 tokio.workspace = true
+uuid.workspace = true
 libc = "0.2"
 
 [build-dependencies]

python/pyproject.toml

@@ -139,6 +139,7 @@ include = [
     "python/lancedb/exceptions.py",
     "python/lancedb/background_loop.py",
     "python/lancedb/schema.py",
+    "python/lancedb/sql.py",
     "python/lancedb/remote/__init__.py",
     "python/lancedb/remote/errors.py",
     "python/lancedb/embeddings/__init__.py",

python/python/lancedb/__init__.py

@@ -22,7 +22,11 @@ from .remote.db import RemoteDBConnection
 from .expr import Expr, col, lit, func
 from .schema import blob, vector
 from .job import AsyncJob, Job
+from .sql import AsyncQuery as AsyncSqlQuery
+from .sql import Query as SqlQuery
+from .sql import QueryDescription
 from .functions import (
+    AssignmentMapping as AssignmentMapping,
     FunctionArtifactRequest as FunctionArtifactRequest,
     FunctionApplication as FunctionApplication,
     FunctionBinding as FunctionBinding,
@@ -33,6 +37,8 @@ from .functions import (
     UdfDefinition as UdfDefinition,
     udf as udf,
 )
+from .secrets import EnvVarSecret as EnvVarSecret
+from .secrets import SecretInfo as SecretInfo
 from .materialized_view import (
     AsyncMaterializedView,
     MaterializedView,
@@ -101,6 +107,7 @@ def connect(
     api_key: Optional[str] = None,
     region: str = "us-east-1",
     host_override: Optional[str] = None,
+    sql_host_override: Optional[str] = None,
     read_consistency_interval: Optional[timedelta] = None,
     request_thread_pool: Optional[Union[int, ThreadPoolExecutor]] = None,
     client_config: Union[ClientConfig, Dict[str, Any], None] = None,
@@ -129,6 +136,9 @@ def connect(
         The region to use for LanceDB Cloud.
     host_override: str, optional
         The override url for LanceDB Cloud.
+    sql_host_override: str, optional
+        The remote SQL service endpoint override. The client connects lazily when SQL
+        is first executed and retains that connection.
     read_consistency_interval: timedelta, default None
         The interval at which to check for updates to the table from other
         processes. If None, then consistency is not checked. For performance
@@ -270,6 +280,7 @@ def connect(
             api_key,
             region,
             host_override,
+            sql_host_override=sql_host_override,
             # TODO: remove this (deprecation warning downstream)
             request_thread_pool=request_thread_pool,
             client_config=client_config,
@@ -412,6 +423,7 @@ def deserialize_conn(
             parsed["api_key"],
             parsed.get("region", "us-east-1"),
             host_override=parsed.get("host_override"),
+            sql_host_override=parsed.get("sql_host_override"),
             client_config=parsed.get("client_config"),
             storage_options=storage_options,
         )
@@ -425,6 +437,7 @@ async def connect_async(
     api_key: Optional[str] = None,
     region: str = "us-east-1",
     host_override: Optional[str] = None,
+    sql_host_override: Optional[str] = None,
     read_consistency_interval: Optional[timedelta] = None,
     client_config: Optional[Union[ClientConfig, Dict[str, Any]]] = None,
     storage_options: Optional[Dict[str, str]] = None,
@@ -447,6 +460,9 @@ async def connect_async(
         The region to use for LanceDB Cloud.
     host_override: str, optional
         The override url for LanceDB Cloud.
+    sql_host_override: str, optional
+        The remote SQL service endpoint override. The client connects lazily when SQL
+        is first executed and retains that connection.
     read_consistency_interval: timedelta, default None
         The interval at which to check for updates to the table from other
         processes. If None, then consistency is not checked. For performance
@@ -534,6 +550,7 @@ async def connect_async(
             api_key,
             region,
             host_override,
+            sql_host_override,
             read_consistency_interval_secs,
             client_config,
             storage_options,
@@ -556,6 +573,7 @@ __all__ = [
     "connect_namespace_async",
     "AsyncConnection",
     "AsyncJob",
+    "AsyncSqlQuery",
     "AsyncLanceNamespaceDBConnection",
     "AsyncTable",
     "FtsToken",
@@ -570,6 +588,8 @@ __all__ = [
     "vector",
     "DBConnection",
     "Job",
+    "QueryDescription",
+    "SqlQuery",
     "LanceDBConnection",
     "LanceNamespaceDBConnection",
     "LsmWriteSpec",

python/python/lancedb/_lancedb.pyi

@@ -1,6 +1,7 @@
 from datetime import date, datetime, timedelta
 from decimal import Decimal
 from typing import Dict, List, Optional, Tuple, Any, TypedDict, Union, Literal
+from uuid import UUID
 
 import pyarrow as pa
 
@@ -147,15 +148,25 @@ class Connection(object):
         start_after: Optional[str],
         limit: Optional[int],
     ) -> list[str]: ...  # Deprecated: Use list_tables instead
-    def job(self, job_id: str) -> Job: ...
+    async def open_job(self, job_id: str) -> Job: ...
     async def create_function_async(self, request_json: str) -> Job: ...
     async def get_function(self, name: str, version: str) -> str: ...
+    async def list_functions(self) -> List[str]: ...
+    async def drop_function(self, name: str, version: str) -> bool: ...
+    async def create_secret(self, name: str, value: str) -> None: ...
+    async def alter_secret(self, name: str, value: str) -> None: ...
+    async def list_secrets(self) -> List[str]: ...
+    async def drop_secret(self, name: str) -> None: ...
+    async def describe_secret(self, name: str) -> Dict[str, str]: ...
     async def list_jobs(self) -> List[JobInfo]: ...
-    async def get_job(self, job_id: str) -> Optional[JobDescription]: ...
     async def cancel_job(self, job_id: str) -> bool: ...
-    async def job_history(
+    async def execute_query_async(
-        self, job_id: Optional[str] = None
+        self,
-    ) -> List[pa.RecordBatch]: ...
+        query: str,
+        *,
+        default_namespace_path: Optional[List[str]] = None,
+    ) -> SqlQuery: ...
+    async def describe_query(self, query_id: UUID) -> QueryDescription: ...
     async def create_table(
         self,
         name: str,
@@ -234,9 +245,20 @@ class BlobFile:
 class Job:
     @property
     def id(self) -> Optional[str]: ...
+    @property
+    def _state(self) -> Optional[str]: ...
+    @property
+    def _description(self) -> Optional[JobDescription]: ...
     async def status(self) -> str: ...
     async def wait(self) -> Optional[str]: ...
     async def cancel(self) -> None: ...
+    async def refresh(self) -> None: ...
+    async def events(
+        self,
+        *,
+        limit: Optional[int] = None,
+        filter: Optional[str] = None,
+    ) -> pa.Table: ...
 
 class JobInfo:
     @property
@@ -268,10 +290,33 @@ class JobDescription:
     @property
     def creation_ms(self) -> int: ...
     @property
-    def spec_json(self) -> Optional[str]: ...
+    def _spec_json(self) -> Optional[str]: ...
+    @property
+    def _result_json(self) -> Optional[str]: ...
+    @property
+    def spec(self) -> Optional[Any]: ...
+    @property
+    def result(self) -> Optional[Any]: ...
     @property
     def failure(self) -> Optional[JobFailureInfo]: ...
 
+class SqlQuery:
+    @property
+    def id(self) -> UUID: ...
+    async def describe(self) -> QueryDescription: ...
+    async def reader(self) -> RecordBatchStream: ...
+    async def cancel(self) -> None: ...
+
+class QueryDescription:
+    @property
+    def id(self) -> UUID: ...
+    @property
+    def status(self) -> str: ...
+    @property
+    def progress(self) -> Optional[float]: ...
+    @property
+    def expires_at(self) -> Optional[datetime]: ...
+
 class Table:
     def name(self) -> str: ...
     def __repr__(self) -> str: ...
@@ -450,6 +495,7 @@ async def connect(
     api_key: Optional[str],
     region: Optional[str],
     host_override: Optional[str],
+    sql_host_override: Optional[str],
     read_consistency_interval: Optional[float],
     client_config: Optional[Union[ClientConfig, Dict[str, Any]]],
     storage_options: Optional[Dict[str, str]],
@@ -606,6 +652,7 @@ class FullTextQuery:
 class PyQueryRequest:
     limit: Optional[int]
     offset: Optional[int]
+    take_offsets: Optional[List[int]]
     filter: Optional[Union[str, bytes]]
     full_text_search: Optional[FullTextQuery]
     select: Optional[Union[str, List[str]]]

python/python/lancedb/db.py

@@ -16,10 +16,11 @@ from typing import (
     Iterable,
     List,
     Literal,
-    Mapping,
     Optional,
+    Sequence,
     Union,
 )
+from uuid import UUID
 
 if sys.version_info >= (3, 12):
     from typing import override
@@ -48,12 +49,16 @@ from . import __version__
 from ._lancedb import connect as lancedb_connect  # type: ignore
 from .functions import FunctionVersion, UdfDefinition
 from .job import AsyncJob, Job, _typed_job
+from .sql import AsyncQuery as AsyncSqlQuery
+from .sql import Query as SqlQuery
+from .sql import QueryDescription
 from .materialized_view import (
     AsyncMaterializedView,
     MaterializedView,
     SelectArg,
     normalize_select,
 )
+from .secrets import EnvVarSecret, SecretInfo, validate_secret_name
 from .table import (
     AsyncTable,
     LanceTable,
@@ -69,10 +74,11 @@ import deprecation
 
 if TYPE_CHECKING:
     import pyarrow as pa
+    from .arrow import AsyncRecordBatchReader
     from .pydantic import LanceModel
 
     from ._lancedb import Connection as LanceDbConnection
-    from ._lancedb import JobDescription, JobInfo
+    from ._lancedb import JobInfo
     from .common import DATA, URI
     from .embeddings import EmbeddingFunctionConfig
     from ._lancedb import Session
@@ -692,16 +698,34 @@ class DBConnection(EnforceOverrides):
         self,
         definition: UdfDefinition,
         *,
-        secrets: Optional[Mapping[str, str]] = None,
+        secrets: Optional[Sequence[EnvVarSecret]] = None,
     ) -> FunctionVersion:
         """Register a scalar Python UDF and wait for its immutable version.
 
-        ``secrets`` must contain exactly the names declared by
-        ``@udf(secrets=[...])``. Values are sent in the create request and
-        stored server-side in the private execution artifact; returned
-        Function and Job metadata contain only the declared names.
         This is the blocking counterpart of :meth:`create_function_async`.
         Local connections raise ``NotImplementedError``.
+
+        Parameters
+        ----------
+        definition : UdfDefinition
+            A callable decorated with [udf][lancedb.udf].
+        secrets : sequence of EnvVarSecret, optional
+            One [EnvVarSecret][lancedb.secrets.EnvVarSecret] per credential the
+            Function needs, each naming a Secret and the environment variable
+            its value arrives in. The Function's source is unchanged by this;
+            it reads the variable the way it already did.
+
+        Examples
+        --------
+        ```python
+        db.create_secret("openai-prod", os.environ["OPENAI_API_KEY"])
+        db.create_function(
+            analyze_caption,
+            secrets=[
+                EnvVarSecret(secret="openai-prod", env_variable="OPENAI_API_KEY")
+            ],
+        )
+        ```
         """
         return self.create_function_async(definition, secrets=secrets).wait()
 
@@ -709,14 +733,10 @@ class DBConnection(EnforceOverrides):
         self,
         definition: UdfDefinition,
         *,
-        secrets: Optional[Mapping[str, str]] = None,
+        secrets: Optional[Sequence[EnvVarSecret]] = None,
     ) -> Job[FunctionVersion]:
         """Register a scalar Python UDF through the remote Function catalog.
 
-        ``secrets`` must contain exactly the names declared by
-        ``@udf(secrets=[...])``. Values are sent in the create request and
-        stored server-side in the private execution artifact; returned
-        Function and Job metadata contain only the declared names.
         Submission returns a typed job. The immutable Function version becomes
         available only when :meth:`Job.wait` succeeds. Local connections raise
         ``NotImplementedError``.
@@ -731,26 +751,107 @@ class DBConnection(EnforceOverrides):
             "Function catalog operations are not supported for this connection type"
         )
 
-    def job(self, job_id: str) -> Job:
+    def list_functions(self) -> List[FunctionVersion]:
-        """A [Job][lancedb.job.Job] handle for a server-side job by id.
+        """List every published immutable Function version.
 
-        The handle is constructed without a server round trip; an unknown id
+        Results are ordered by Function name then version. Local connections
-        surfaces when the handle is used. Dropping the handle has no effect
+        raise ``NotImplementedError``.
-        on the job itself.
+
+        Examples
+        --------
+        List the identities available to use in Function-backed columns:
+
+        ```python
+        [(function.name, function.version) for function in db.list_functions()]
+        ```
         """
-        raise NotImplementedError("job is not supported for this connection type")
+        raise NotImplementedError(
+            "Function catalog operations are not supported for this connection type"
+        )
+
+    def drop_function(self, name: str, *, version: str) -> bool:
+        """Drop one exact immutable Function version from the remote catalog.
+
+        Returns True when the version changed to Dropped and False for an
+        idempotent replay. Local connections raise NotImplementedError.
+        """
+        raise NotImplementedError(
+            "Function catalog operations are not supported for this connection type"
+        )
+
+    def create_secret(self, name: str, value: str) -> None:
+        """Create a named Secret in this database.
+
+        Fails if the name is taken, so a create never silently becomes a
+        rotation. Nothing reads the value back: it is bound to a Function by
+        name and resolved by the service when that Function runs. Local
+        connections raise ``NotImplementedError``.
+        """
+        raise NotImplementedError(
+            "Secret operations are not supported for this connection type"
+        )
+
+    def alter_secret(self, name: str, value: str) -> None:
+        """Replace the credential behind an existing Secret.
+
+        Fails if it does not exist. Every Function bound to the Secret uses the
+        new value from its next job, and no new Function version is created --
+        which is how a rotation reaches columns pinned to a version registered
+        before it. Local connections raise ``NotImplementedError``.
+        """
+        raise NotImplementedError(
+            "Secret operations are not supported for this connection type"
+        )
+
+    def list_secrets(self) -> List[str]:
+        """The names of every Secret in this database.
+
+        Names only. No method returns a stored credential, by construction
+        rather than by policy. Local connections raise ``NotImplementedError``.
+        """
+        raise NotImplementedError(
+            "Secret operations are not supported for this connection type"
+        )
+
+    def drop_secret(self, name: str) -> None:
+        """Drop a Secret.
+
+        Functions bound to it fail at their next job, naming the Secret; that
+        is the revocation path. The name becomes free to reuse, and a new
+        Secret under it is picked up by everything still bound to that name.
+        Local connections raise ``NotImplementedError``.
+        """
+        raise NotImplementedError(
+            "Secret operations are not supported for this connection type"
+        )
+
+    def describe_secret(self, name: str) -> SecretInfo:
+        """What this database records about a Secret: name and timestamps.
+
+        Never the value -- there is no code path that could return one. Local
+        connections raise ``NotImplementedError``.
+        """
+        raise NotImplementedError(
+            "Secret operations are not supported for this connection type"
+        )
+
+    def open_job(self, job_id: str) -> Job:
+        """Open a server-side job by id, returning a handle with its record
+        already populated.
+
+        The returned [Job][lancedb.job.Job] answers for its own state,
+        specification, result, failure and event history, so there is no
+        separate connection-level call for any of them.
+
+        Raises `JobNotFoundError` when the server has no such job, the way
+        `open_table` does for a missing table.
+        """
+        raise NotImplementedError("open_job is not supported for this connection type")
 
     def list_jobs(self) -> List[JobInfo]:
         """List server-side jobs across the database's tables."""
         raise NotImplementedError("list_jobs is not supported for this connection type")
 
-    def get_job(self, job_id: str) -> Optional[JobDescription]:
-        """Describe a single server-side job by id.
-
-        Returns None when the server has no such job.
-        """
-        raise NotImplementedError("get_job is not supported for this connection type")
-
     def cancel_job(self, job_id: str) -> bool:
         """Request cancellation of a server-side job by id.
 
@@ -762,14 +863,38 @@ class DBConnection(EnforceOverrides):
             "cancel_job is not supported for this connection type"
         )
 
-    def job_history(self, job_id: Optional[str] = None) -> List[pa.RecordBatch]:
+    def execute_query(
-        """The lifecycle event history of a server-side job, as Arrow batches.
+        self,
+        query: str,
+        *,
+        default_namespace_path: Optional[List[str]] = None,
+    ) -> pa.RecordBatchReader:
+        """Execute SQL and return a blocking Arrow reader.
 
-        Lists history across all jobs when `job_id` is None.
+        This submits through :meth:`execute_query_async` and waits until the
+        initial result stream is readable. It does not wait for the full query
+        to finish.
         """
-        raise NotImplementedError(
+        return self.execute_query_async(
-            "job_history is not supported for this connection type"
+            query,
-        )
+            default_namespace_path=default_namespace_path,
+        ).reader()
+
+    def execute_query_async(
+        self,
+        query: str,
+        *,
+        default_namespace_path: Optional[List[str]] = None,
+    ) -> SqlQuery:
+        """Start executing SQL and return its query handle.
+
+        Local connections do not support SQL.
+        """
+        raise NotImplementedError("SQL is not supported for this connection type")
+
+    def describe_query(self, query_id: UUID) -> QueryDescription:
+        """Describe a submitted SQL query by its connection-scoped id."""
+        raise NotImplementedError("SQL is not supported for this connection type")
 
 
 class LanceDBConnection(DBConnection):
@@ -866,6 +991,7 @@ class LanceDBConnection(DBConnection):
                 None,
                 None,
                 None,
+                None,
                 read_consistency_interval_secs,
                 None,
                 storage_options,
@@ -1414,21 +1540,18 @@ class LanceDBConnection(DBConnection):
         )
 
     @override
-    def job(self, job_id: str) -> Job:
+    def open_job(self, job_id: str) -> Job:
-        """A [Job][lancedb.job.Job] handle for a server-side job by id.
+        """Open a server-side job by id. See
-
+        [DBConnection.open_job][lancedb.db.DBConnection.open_job].
-        The handle is constructed without a server round trip; an unknown id
-        surfaces when the handle is used. Dropping the handle has no effect
-        on the job itself.
         """
-        return Job(self._conn.job(job_id))
+        return Job(LOOP.run(self._conn.open_job(job_id)))
 
     @override
     def create_function_async(
         self,
         definition: UdfDefinition,
         *,
-        secrets: Optional[Mapping[str, str]] = None,
+        secrets: Optional[Sequence[EnvVarSecret]] = None,
     ) -> Job[FunctionVersion]:
         job = LOOP.run(self._conn.create_function_async(definition, secrets=secrets))
         return Job(job)
@@ -1437,19 +1560,39 @@ class LanceDBConnection(DBConnection):
     def get_function(self, name: str, *, version: str) -> FunctionVersion:
         return LOOP.run(self._conn.get_function(name, version=version))
 
+    @override
+    def list_functions(self) -> List[FunctionVersion]:
+        return LOOP.run(self._conn.list_functions())
+
+    @override
+    def drop_function(self, name: str, *, version: str) -> bool:
+        return LOOP.run(self._conn.drop_function(name, version=version))
+
+    @override
+    def create_secret(self, name: str, value: str) -> None:
+        LOOP.run(self._conn.create_secret(name, value))
+
+    @override
+    def alter_secret(self, name: str, value: str) -> None:
+        LOOP.run(self._conn.alter_secret(name, value))
+
+    @override
+    def list_secrets(self) -> List[str]:
+        return LOOP.run(self._conn.list_secrets())
+
+    @override
+    def drop_secret(self, name: str) -> None:
+        LOOP.run(self._conn.drop_secret(name))
+
+    @override
+    def describe_secret(self, name: str) -> SecretInfo:
+        return LOOP.run(self._conn.describe_secret(name))
+
     @override
     def list_jobs(self) -> List[JobInfo]:
         """List server-side jobs across the database's tables."""
         return LOOP.run(self._conn.list_jobs())
 
-    @override
-    def get_job(self, job_id: str) -> Optional[JobDescription]:
-        """Describe a single server-side job by id.
-
-        Returns None when the server has no such job.
-        """
-        return LOOP.run(self._conn.get_job(job_id))
-
     @override
     def cancel_job(self, job_id: str) -> bool:
         """Request cancellation of a server-side job by id.
@@ -1460,14 +1603,6 @@ class LanceDBConnection(DBConnection):
         """
         return LOOP.run(self._conn.cancel_job(job_id))
 
-    @override
-    def job_history(self, job_id: Optional[str] = None) -> List[pa.RecordBatch]:
-        """The lifecycle event history of a server-side job, as Arrow batches.
-
-        Lists history across all jobs when `job_id` is None.
-        """
-        return LOOP.run(self._conn.job_history(job_id))
-
     @override
     def namespace_client(self) -> LanceNamespace:
         """Get the equivalent namespace client for this connection.
@@ -2238,53 +2373,90 @@ class AsyncConnection(object):
             namespace_path = []
         await self._inner.drop_all_tables(namespace_path=namespace_path)
 
-    def job(self, job_id: str) -> AsyncJob:
+    async def open_job(self, job_id: str) -> AsyncJob:
-        """An [AsyncJob][lancedb.job.AsyncJob] handle for a server-side job
+        """Open a server-side job by id. See
-        by id.
+        [DBConnection.open_job][lancedb.db.DBConnection.open_job].
-
-        The handle is constructed without a server round trip; an unknown id
-        surfaces when the handle is used. Dropping the handle has no effect
-        on the job itself.
         """
-        return AsyncJob(self._inner.job(job_id))
+        return AsyncJob(await self._inner.open_job(job_id))
 
     async def create_function_async(
         self,
         definition: UdfDefinition,
         *,
-        secrets: Optional[Mapping[str, str]] = None,
+        secrets: Optional[Sequence[EnvVarSecret]] = None,
     ) -> AsyncJob[FunctionVersion]:
         """Register a scalar Python UDF through the remote Function catalog.
 
-        ``secrets`` must contain exactly the names declared by
-        ``@udf(secrets=[...])``. Values are sent in the create request and
-        stored server-side in the private execution artifact; returned
-        Function and Job metadata contain only the declared names.
         The returned typed job resolves to the immutable Function version.
-        Local connections raise ``NotImplementedError``.
+        ``secrets`` is a sequence of
+        [EnvVarSecret][lancedb.secrets.EnvVarSecret], each naming a Secret and
+        the environment variable its value arrives in. Local connections raise
+        ``NotImplementedError``.
         """
         if not isinstance(definition, UdfDefinition):
             raise TypeError("create_function_async requires a @udf definition")
-        inner = await self._inner.create_function_async(
+        request = definition.bind_secrets(secrets)
-            definition._submission_json(secrets)
+        inner = await self._inner.create_function_async(request.to_canonical_json())
-        )
         return _typed_job(inner, FunctionVersion.from_json)
 
     async def get_function(self, name: str, *, version: str) -> FunctionVersion:
         """Open one exact immutable Function version from the remote catalog."""
         return FunctionVersion.from_json(await self._inner.get_function(name, version))
 
+    async def list_functions(self) -> List[FunctionVersion]:
+        """List every published immutable Function version.
+
+        Results are ordered by Function name then version. Local connections
+        raise ``NotImplementedError``.
+        """
+        return [
+            FunctionVersion.from_json(value)
+            for value in await self._inner.list_functions()
+        ]
+
+    async def drop_function(self, name: str, *, version: str) -> bool:
+        """Drop one exact immutable Function version from the remote catalog."""
+        return await self._inner.drop_function(name, version)
+
+    async def create_secret(self, name: str, value: str) -> None:
+        """Create a named Secret in this database.
+
+        Fails if the name is taken, so a create never silently becomes a
+        rotation. Nothing reads the value back.
+        """
+        await self._inner.create_secret(validate_secret_name(name), value)
+
+    async def alter_secret(self, name: str, value: str) -> None:
+        """Replace the credential behind an existing Secret.
+
+        Fails if it does not exist. Bound Functions use the new value from
+        their next job, with no new Function version.
+        """
+        await self._inner.alter_secret(validate_secret_name(name), value)
+
+    async def list_secrets(self) -> List[str]:
+        """The names of every Secret in this database. Names only."""
+        return await self._inner.list_secrets()
+
+    async def drop_secret(self, name: str) -> None:
+        """Drop a Secret. Bound Functions fail at their next job."""
+        await self._inner.drop_secret(validate_secret_name(name))
+
+    async def describe_secret(self, name: str) -> SecretInfo:
+        """What this database records about a Secret. Never the value."""
+        name, created_at_millis, updated_at_millis = await self._inner.describe_secret(
+            validate_secret_name(name)
+        )
+        return SecretInfo(
+            name=name,
+            created_at_millis=created_at_millis,
+            updated_at_millis=updated_at_millis,
+        )
+
     async def list_jobs(self) -> List[JobInfo]:
         """List server-side jobs across the database's tables."""
         return await self._inner.list_jobs()
 
-    async def get_job(self, job_id: str) -> Optional[JobDescription]:
-        """Describe a single server-side job by id.
-
-        Returns None when the server has no such job.
-        """
-        return await self._inner.get_job(job_id)
-
     async def cancel_job(self, job_id: str) -> bool:
         """Request cancellation of a server-side job by id.
 
@@ -2294,12 +2466,46 @@ class AsyncConnection(object):
         """
         return await self._inner.cancel_job(job_id)
 
-    async def job_history(self, job_id: Optional[str] = None) -> List[pa.RecordBatch]:
+    async def execute_query(
-        """The lifecycle event history of a server-side job, as Arrow batches.
+        self,
+        query: str,
+        *,
+        default_namespace_path: Optional[List[str]] = None,
+    ) -> AsyncRecordBatchReader:
+        """Execute SQL and return an asynchronous Arrow reader.
 
-        Lists history across all jobs when `job_id` is None.
+        This submits through :meth:`execute_query_async` and waits until the
+        initial result stream is readable. It does not wait for the full query
+        to finish.
         """
-        return await self._inner.job_history(job_id)
+        submitted = await self.execute_query_async(
+            query,
+            default_namespace_path=default_namespace_path,
+        )
+        return await submitted.reader()
+
+    async def execute_query_async(
+        self,
+        query: str,
+        *,
+        default_namespace_path: Optional[List[str]] = None,
+    ) -> AsyncSqlQuery:
+        """Start executing SQL and return its query handle.
+
+        The database from ``connect_async`` is used for unqualified database
+        references. The namespace defaults to ``["public"]``. Local
+        connections raise ``NotImplementedError``.
+        """
+        return AsyncSqlQuery(
+            await self._inner.execute_query_async(
+                query,
+                default_namespace_path=default_namespace_path,
+            )
+        )
+
+    async def describe_query(self, query_id: UUID) -> QueryDescription:
+        """Describe a submitted SQL query by its connection-scoped id."""
+        return await self._inner.describe_query(query_id)
 
     async def namespace_client(self) -> LanceNamespace:
         """Get the equivalent namespace client for this connection.

python/python/lancedb/exceptions.py

@@ -35,3 +35,9 @@ class JobCancelledError(RuntimeError):
     """Exception raised when an asynchronous job was cancelled."""
 
     pass
+
+
+class JobNotFoundError(ValueError):
+    """Exception raised when opening a job the server does not have."""
+
+    pass

python/python/lancedb/functions.py

@@ -25,7 +25,7 @@ import re
 import sys
 import textwrap
 import types
-from collections.abc import Mapping
+from collections.abc import Mapping, Sequence
 from datetime import date, datetime
 from typing import (
     Annotated,
@@ -49,11 +49,26 @@ from pydantic import (
     model_validator,
 )
 
+from .schema import is_blob_v2_field as _is_blob_v2_field
+from .secrets import EnvVarSecret
+
 _Int32 = conint(strict=True, ge=-(2**31), le=2**31 - 1)
 _UInt32 = conint(strict=True, ge=0, le=2**32 - 1)
 _UInt64 = conint(strict=True, ge=0, le=2**64 - 1)
 
 
+def _validate_gpu_wire_marker(value: Any) -> bool:
+    if value is not True:
+        raise ValueError("runtime.gpu must be true")
+    return True
+
+
+def _normalize_gpu_marker(value: bool) -> Optional[bool]:
+    if not isinstance(value, bool):
+        raise ValueError("gpu must be a boolean")
+    return True if value else None
+
+
 class _FrozenDict(dict):
     def _immutable(self, *args, **kwargs):
         raise TypeError("remote canonical values are immutable")
@@ -212,6 +227,20 @@ class FunctionOutput(_OpenRemoteValue):
     fields: tuple[FunctionResultField, ...] = ()
 
 
+class SecretBinding(_RemoteValue):
+    """How a Secret reaches the Function that binds it.
+
+    One list rather than a field per delivery mode: a binding is the concept,
+    and how it arrives is a property of one. ``kind`` is open, so a binding a
+    newer service introduces decodes here instead of failing the whole
+    FunctionVersion.
+    """
+
+    kind: str
+    variable: Optional[str] = None
+    secret_ref: Optional[str] = None
+
+
 class FunctionSignature(_RemoteValue):
     inputs: tuple[FunctionParameter, ...]
     output: FunctionOutput
@@ -229,7 +258,7 @@ class PythonEnvironmentSpec(_RemoteValue):
 
 
 class PythonRuntimeSpec(_RemoteValue):
-    """Remote runtime definition with non-secret environment values.
+    """Remote runtime definition with environment values.
 
     V1 supports ``kind="python"``. Newer runtime kinds remain readable, while
     their unknown payload fields are intentionally not retained by the client.
@@ -239,6 +268,23 @@ class PythonRuntimeSpec(_RemoteValue):
     python_version: Optional[str] = None
     environment: Optional[PythonEnvironmentSpec] = None
     env: Optional[Mapping[str, str]] = None
+    gpu: Optional[bool] = None
+
+    @model_validator(mode="before")
+    @classmethod
+    def _discard_unknown_runtime_payload(cls, value):
+        if isinstance(value, Mapping):
+            kind = value.get("kind")
+            if isinstance(kind, str) and kind not in {"python", "python_v2"}:
+                return {"kind": kind}
+        return value
+
+    @field_validator("gpu", mode="before")
+    @classmethod
+    def _validate_gpu_marker(cls, value):
+        if value is None:
+            return None
+        return _validate_gpu_wire_marker(value)
 
     @model_validator(mode="after")
     def _validate_runtime_kind(self):
@@ -247,18 +293,28 @@ class PythonRuntimeSpec(_RemoteValue):
                 raise ValueError("python runtime requires python_version")
             if self.environment is None:
                 raise ValueError("python runtime requires environment")
+            if self.gpu is not None:
+                raise ValueError("python runtime with gpu requires kind='python_v2'")
+        elif self.kind == "python_v2":
+            if self.python_version is None:
+                raise ValueError("python_v2 runtime requires python_version")
+            if self.environment is None:
+                raise ValueError("python_v2 runtime requires environment")
+            if self.gpu is None:
+                raise ValueError("python_v2 runtime requires gpu")
         else:
             object.__setattr__(self, "python_version", None)
             object.__setattr__(self, "environment", None)
             object.__setattr__(self, "env", None)
+            object.__setattr__(self, "gpu", None)
         return self
 
 
 class FunctionVersion(_RemoteValue):
     """An exact immutable Function version returned by Enterprise.
 
-    Scheduling resources, priority, concurrency, and retry policy belong to
+    The GPU execution requirement is part of this identity. CPU and memory sizing,
-    the submitting Job and are not part of this identity.
+    priority, concurrency, and retry policy belong to the execution platform.
     """
 
     name: str
@@ -268,7 +324,7 @@ class FunctionVersion(_RemoteValue):
     runtime: PythonRuntimeSpec
     runtime_digest: str
     environment_digest: str
-    required_secrets: tuple[str, ...] = ()
+    secret_bindings: tuple[SecretBinding, ...] = ()
     created_at: str
 
     def __call__(self, **inputs: Any) -> FunctionApplication:
@@ -332,15 +388,16 @@ class FunctionVersion(_RemoteValue):
 class FunctionRegistrationRequest(_RemoteValue):
     """Stable remote registration envelope produced by :func:`udf`.
 
-    Only secret names are represented. Secret values are supplied separately
+    Credential values deliberately have no field here. The only secret-shaped
-    when the definition is submitted and are not part of this durable value.
+    thing a client sends is ``secret_bindings``: the name of a Secret the
+    database already holds, which the remote service resolves at execution.
     """
 
     name: str
     artifact: FunctionArtifactRequest
     signature: FunctionSignature
     runtime: PythonRuntimeSpec
-    required_secrets: tuple[str, ...] = ()
+    secret_bindings: tuple[SecretBinding, ...] = ()
 
 
 class FunctionVersionRef(_OpenRemoteValue):
@@ -435,11 +492,7 @@ class InputBinding(_RemoteValue):
 
 
 class OutputMapping(_RemoteValue):
-    """One stable result-field mapping.
+    """One stable result-field mapping."""
-
-    Assignment state is outside the Slice 1 client contract. During the NULL
-    transition Lance exposes no public cell-flag identifier to persist here.
-    """
 
     result_field: str
     output_name: str
@@ -449,6 +502,13 @@ class OutputMapping(_RemoteValue):
     nullable: bool
 
 
+class AssignmentMapping(_RemoteValue):
+    """Internal physical column preserving flattened struct validity."""
+
+    output_name: str
+    output_field_id: _Int32
+
+
 class FunctionBinding(_RemoteValue):
     """Immutable Function binding persisted by the Enterprise table service."""
 
@@ -456,6 +516,7 @@ class FunctionBinding(_RemoteValue):
     function: FunctionVersionRef
     inputs: tuple[InputBinding, ...]
     outputs: tuple[OutputMapping, ...]
+    assignment: Optional[AssignmentMapping] = None
     input_schema: Optional[Mapping[str, Any]] = None
     output_schema: Optional[Mapping[str, Any]] = None
 
@@ -485,27 +546,15 @@ class RefreshColumnResult(_RemoteValue):
 
 
 _FUNCTION_NAME = re.compile(r"^[A-Za-z_][A-Za-z0-9_.-]*$")
-_SECRET_NAME = re.compile(r"^[A-Za-z_][A-Za-z0-9_]*$")
+_DECLARED_SECRET = re.compile(r"^[A-Za-z_][A-Za-z0-9_]*$")
-# Keep this byte limit aligned with Sophon's MAX_FUNCTION_SECRET_VALUE_BYTES.
-_MAX_FUNCTION_SECRET_VALUE_BYTES = 64 * 1024
-_MAX_FUNCTION_SECRET_VALUES_BYTES = 512 * 1024
 
-
+_FUNCTION_BLOB_V2_TYPE = "blob_v2"
-def _validate_secret_value(name: str, value: Any) -> str:
+_ARROW_EXTENSION_NAME_KEY = "ARROW:extension:name"
-    """Validate one secret value before building the create request."""
+_BLOB_V2_EXTENSION_NAME = "lance.blob.v2"
-    if not isinstance(value, str):
+_NESTED_BLOB_COLLECTION_ERROR = (
-        raise TypeError(f"Function secret {name!r} value must be a string")
+    "unsupported Arrow type for Function signature: Blob v2 fields nested under "
-    if not value:
+    "collection types are not supported"
-        raise ValueError(f"Function secret {name!r} value must be non-empty")
+)
-    if "\0" in value:
-        raise ValueError(f"Function secret {name!r} value must not contain NUL")
-    value_bytes = len(value.encode("utf-8"))
-    if value_bytes > _MAX_FUNCTION_SECRET_VALUE_BYTES:
-        raise ValueError(
-            f"Function secret {name!r} value exceeds the "
-            f"{_MAX_FUNCTION_SECRET_VALUE_BYTES}-byte limit"
-        )
-    return value
 
 
 _GRAMMAR_PRIMITIVES = (
@@ -522,6 +571,7 @@ _GRAMMAR_PRIMITIVES = (
     (pa.float32(), "float32"),
     (pa.float64(), "float64"),
     (pa.string(), "utf8"),
+    (pa.large_string(), "large_utf8"),
     (pa.binary(), "binary"),
     (pa.date32(), "date32"),
     (pa.date64(), "date64"),
@@ -529,31 +579,258 @@ _GRAMMAR_PRIMITIVES = (
 
 
 def _canonical_arrow_type(data_type: pa.DataType) -> str:
-    """The server's V1 Function type grammar. Anything outside it is rejected
+    """The compact Function grammar, or canonical exact JSON for nested types."""
-    here rather than at registration."""
+    grammar = _grammar_arrow_type(data_type)
+    if grammar is not None:
+        return grammar
+    exact = _exact_arrow_type(data_type)
+    return json.dumps(exact, ensure_ascii=False, sort_keys=True, separators=(",", ":"))
+
+
+def _grammar_arrow_type(data_type: pa.DataType) -> Optional[str]:
     for candidate, name in _GRAMMAR_PRIMITIVES:
         if data_type == candidate:
             return name
     if pa.types.is_list(data_type) or pa.types.is_large_list(data_type):
+        item = _grammar_list_item(data_type)
+        if item is None:
+            return None
         prefix = "list" if pa.types.is_list(data_type) else "large_list"
-        return f"{prefix}<{_canonical_list_item(data_type)}>"
+        return f"{prefix}<{item}>"
     if pa.types.is_fixed_size_list(data_type) and data_type.list_size > 0:
-        return (
+        item = _grammar_list_item(data_type)
-            f"fixed_size_list<{_canonical_list_item(data_type)}, {data_type.list_size}>"
+        if item is not None:
-        )
+            return f"fixed_size_list<{item}, {data_type.list_size}>"
-    raise TypeError(f"unsupported Arrow type for Function signature: {data_type}")
+    return None
 
 
-def _canonical_list_item(data_type: pa.DataType) -> str:
+def _grammar_list_item(data_type: pa.DataType) -> Optional[str]:
     """The grammar names only the item type; it always means a non-nullable
-    child called `item`, so any other child metadata cannot be represented."""
+    child called `item`, so other child properties require exact JSON."""
     child = data_type.value_field
     if child.name != "item" or child.nullable or child.metadata:
+        return None
+    return _grammar_arrow_type(child.type)
+
+
+def _validate_exact_arrow_field(field: pa.Field) -> None:
+    if not field.name:
         raise TypeError(
-            "unsupported Arrow type for Function signature: list items must be a "
+            "unsupported Arrow type for Function signature: field names "
-            f"non-nullable field named 'item', got {child}"
+            "must not be empty"
         )
-    return _canonical_arrow_type(child.type)
+    if _is_blob_v2_field(field):
+        if not _has_supported_blob_v2_layout(field):
+            raise TypeError(
+                "unsupported Arrow type for Function signature: lance.blob.v2 "
+                f"requires a supported Blob storage layout, got {field}"
+            )
+        metadata = {
+            (key.decode() if isinstance(key, bytes) else key): (
+                value.decode() if isinstance(value, bytes) else value
+            )
+            for key, value in (field.metadata or {}).items()
+        }
+        if metadata and metadata != {
+            _ARROW_EXTENSION_NAME_KEY: _BLOB_V2_EXTENSION_NAME
+        }:
+            raise TypeError(
+                "unsupported Arrow type for Function signature: lance.blob.v2 "
+                "field metadata must contain only its canonical extension marker"
+            )
+    elif field.metadata:
+        raise TypeError(
+            "unsupported Arrow type for Function signature: field metadata "
+            f"is not supported, got {field}"
+        )
+
+
+def _has_supported_blob_v2_layout(field: pa.Field) -> bool:
+    data_type = field.type
+    if isinstance(data_type, pa.ExtensionType):
+        data_type = data_type.storage_type
+    if not pa.types.is_struct(data_type):
+        return False
+
+    fields = tuple(data_type)
+
+    def matches(spec, compare_nullable) -> bool:
+        return len(fields) == len(spec) and all(
+            actual.name == name
+            and actual.type == expected_type
+            and (not check_nullable or actual.nullable == nullable)
+            for actual, (name, expected_type, nullable), check_nullable in zip(
+                fields, spec, compare_nullable
+            )
+        )
+
+    logical_minimal = (
+        ("data", pa.large_binary(), True),
+        ("uri", pa.utf8(), True),
+    )
+    logical_full = logical_minimal + (
+        ("position", pa.uint64(), True),
+        ("size", pa.uint64(), True),
+    )
+    prepared = (
+        ("kind", pa.uint8(), True),
+        ("data", pa.large_binary(), True),
+        ("uri", pa.utf8(), True),
+        ("blob_id", pa.uint32(), True),
+        ("blob_size", pa.uint64(), True),
+        ("position", pa.uint64(), True),
+    )
+    descriptor = (
+        ("kind", pa.uint8(), False),
+        ("position", pa.uint64(), False),
+        ("size", pa.uint64(), False),
+        ("blob_id", pa.uint32(), False),
+        ("blob_uri", pa.utf8(), False),
+    )
+    return (
+        matches(logical_minimal, (True, True))
+        or matches(logical_full, (True, True, False, False))
+        or matches(prepared, (True,) * len(prepared))
+        or matches(descriptor, (False,) * len(descriptor))
+    )
+
+
+def _canonical_arrow_field(field: pa.Field) -> str:
+    _validate_exact_arrow_field(field)
+    if _is_blob_v2_field(field):
+        return _FUNCTION_BLOB_V2_TYPE
+    return _canonical_arrow_type(field.type)
+
+
+def _blob_storage_type(field: pa.Field) -> pa.DataType:
+    data_type = field.type
+    if isinstance(data_type, pa.ExtensionType):
+        return data_type.storage_type
+    return data_type
+
+
+def _exact_blob_storage_type(field: pa.Field) -> dict[str, Any]:
+    storage = _blob_storage_type(field)
+    if not pa.types.is_struct(storage):
+        raise TypeError(
+            "unsupported Arrow type for Function signature: lance.blob.v2 "
+            "requires struct storage"
+        )
+    return {
+        "type": "struct",
+        "fields": [
+            {
+                "name": child.name,
+                "nullable": child.nullable,
+                "type": (
+                    {"type": "large_binary"}
+                    if pa.types.is_large_binary(child.type)
+                    else _exact_arrow_type(child.type)
+                ),
+            }
+            for child in storage
+        ],
+    }
+
+
+def _data_type_has_blob_v2(data_type: pa.DataType) -> bool:
+    if pa.types.is_struct(data_type):
+        return any(
+            _is_blob_v2_field(field) or _data_type_has_blob_v2(field.type)
+            for field in data_type
+        )
+    if (
+        pa.types.is_list(data_type)
+        or pa.types.is_large_list(data_type)
+        or pa.types.is_fixed_size_list(data_type)
+    ):
+        field = data_type.value_field
+        return _is_blob_v2_field(field) or _data_type_has_blob_v2(field.type)
+    if pa.types.is_map(data_type):
+        return any(
+            _is_blob_v2_field(field) or _data_type_has_blob_v2(field.type)
+            for field in (data_type.key_field, data_type.item_field)
+        )
+    return False
+
+
+def _exact_arrow_field(
+    field: pa.Field, *, inside_collection: bool = False
+) -> dict[str, Any]:
+    _validate_exact_arrow_field(field)
+    if _is_blob_v2_field(field):
+        if inside_collection:
+            raise TypeError(_NESTED_BLOB_COLLECTION_ERROR)
+        return {
+            "name": field.name,
+            "nullable": field.nullable,
+            "type": _exact_blob_storage_type(field),
+            "metadata": {
+                _ARROW_EXTENSION_NAME_KEY: _BLOB_V2_EXTENSION_NAME,
+            },
+        }
+    value = {
+        "name": field.name,
+        "nullable": field.nullable,
+        "type": _exact_arrow_type(field.type, inside_collection=inside_collection),
+    }
+    return value
+
+
+def _exact_arrow_type(
+    data_type: pa.DataType, *, inside_collection: bool = False
+) -> dict[str, Any]:
+    for candidate, name in _GRAMMAR_PRIMITIVES:
+        if data_type == candidate:
+            return {"type": name}
+    if pa.types.is_struct(data_type):
+        fields = list(data_type)
+        names = [field.name for field in fields]
+        if not fields or len(set(names)) != len(names):
+            raise TypeError(
+                "unsupported Arrow type for Function signature: structs must have "
+                "non-empty, uniquely named fields"
+            )
+        return {
+            "type": "struct",
+            "fields": [
+                _exact_arrow_field(field, inside_collection=inside_collection)
+                for field in fields
+            ],
+        }
+    if (
+        pa.types.is_list(data_type)
+        or pa.types.is_large_list(data_type)
+        or pa.types.is_fixed_size_list(data_type)
+    ):
+        if pa.types.is_fixed_size_list(data_type):
+            if data_type.value_field.name != "item":
+                raise TypeError(
+                    "unsupported Arrow type for Function signature: fixed-size list "
+                    "items must be named 'item'"
+                )
+            if data_type.list_size <= 0:
+                raise TypeError(
+                    f"unsupported Arrow type for Function signature: {data_type}"
+                )
+        value: dict[str, Any] = {
+            "type": (
+                "list"
+                if pa.types.is_list(data_type)
+                else "large_list"
+                if pa.types.is_large_list(data_type)
+                else "fixed_size_list"
+            ),
+            "fields": [
+                _exact_arrow_field(data_type.value_field, inside_collection=True)
+            ],
+        }
+        if pa.types.is_fixed_size_list(data_type):
+            value["length"] = data_type.list_size
+        return value
+    if pa.types.is_map(data_type) and _data_type_has_blob_v2(data_type):
+        raise TypeError(_NESTED_BLOB_COLLECTION_ERROR)
+    raise TypeError(f"unsupported Arrow type for Function signature: {data_type}")
 
 
 def _list_of(item: pa.DataType) -> pa.DataType:
@@ -627,8 +904,15 @@ def _callable_parameters(function: Callable[..., Any]) -> tuple[inspect.Paramete
 
 def _function_output(output: pa.DataType | pa.Field | pa.Schema) -> FunctionOutput:
     if isinstance(output, pa.Schema):
+        if output.metadata:
+            raise TypeError("Function output schema metadata is not supported")
         fields = tuple(output)
-    elif isinstance(output, pa.Field) and pa.types.is_struct(output.type):
+    elif (
+        isinstance(output, pa.Field)
+        and not _is_blob_v2_field(output)
+        and pa.types.is_struct(output.type)
+    ):
+        _validate_exact_arrow_field(output)
         if output.nullable:
             raise ValueError("Function output must be non-nullable")
         fields = tuple(output.type)
@@ -644,18 +928,19 @@ def _function_output(output: pa.DataType | pa.Field | pa.Schema) -> FunctionOutp
             raise TypeError(
                 "output_schema must be a PyArrow DataType, Field, or Schema"
             )
+        _validate_exact_arrow_field(field)
         if field.nullable:
             raise ValueError("Function output must be non-nullable")
         return FunctionOutput(
             kind="scalar",
-            arrow_type=_canonical_arrow_type(field.type),
+            arrow_type=_canonical_arrow_field(field),
             nullable=False,
         )
 
     if not fields:
         raise ValueError("named-struct Function output must contain at least one field")
-    if any(field.nullable for field in fields):
+    for field in fields:
-        raise ValueError("Function output fields must be non-nullable")
+        _validate_exact_arrow_field(field)
     names = [field.name for field in fields]
     if len(set(names)) != len(names):
         raise ValueError("Function output field names must be unique")
@@ -664,8 +949,8 @@ def _function_output(output: pa.DataType | pa.Field | pa.Schema) -> FunctionOutp
         fields=tuple(
             FunctionResultField(
                 name=field.name,
-                arrow_type=_canonical_arrow_type(field.type),
+                arrow_type=_canonical_arrow_field(field),
-                nullable=False,
+                nullable=field.nullable,
             )
             for field in fields
         ),
@@ -684,6 +969,10 @@ def _infer_signature(
     if input_schema is not None:
         if not isinstance(input_schema, pa.Schema):
             raise TypeError("input_schema must be a PyArrow Schema")
+        if input_schema.metadata:
+            raise TypeError("Function input schema metadata is not supported")
+        for field in input_schema:
+            _validate_exact_arrow_field(field)
         expected = tuple(parameter.name for parameter in parameters)
         actual = tuple(input_schema.names)
         if actual != expected:
@@ -694,7 +983,7 @@ def _infer_signature(
         inputs = tuple(
             FunctionParameter(
                 name=field.name,
-                arrow_type=_canonical_arrow_type(field.type),
+                arrow_type=_canonical_arrow_field(field),
                 nullable=field.nullable,
             )
             for field in input_schema
@@ -717,7 +1006,9 @@ def _infer_signature(
         inputs.append(
             FunctionParameter(
                 name=parameter.name,
-                arrow_type=_canonical_arrow_type(data_type),
+                arrow_type=_canonical_arrow_field(
+                    pa.field(parameter.name, data_type, nullable=nullable)
+                ),
                 nullable=nullable,
             )
         )
@@ -936,8 +1227,8 @@ class UdfDefinition:
         output_schema: Optional[pa.DataType | pa.Field | pa.Schema],
         pip: tuple[str, ...],
         env: Mapping[str, str],
-        secrets: tuple[str, ...],
         python_version: Optional[str],
+        gpu: bool = False,
         conda: tuple[str, ...] = (),
         conda_channels: tuple[str, ...] = (),
     ):
@@ -963,26 +1254,17 @@ class UdfDefinition:
             for key, value in environment.items()
         ):
             raise TypeError("Function env keys and values must be strings")
-        required_secrets = tuple(sorted(set(secrets)))
-        invalid_secrets = [
-            secret for secret in required_secrets if not _SECRET_NAME.fullmatch(secret)
-        ]
-        if invalid_secrets:
-            raise ValueError(f"invalid Function secret names: {invalid_secrets!r}")
-        overlap = set(environment) & set(required_secrets)
-        if overlap:
-            raise ValueError(
-                f"Function env and secret names must be disjoint: {sorted(overlap)!r}"
-            )
         signature = _infer_signature(function, input_schema, output_schema)
         source = _package_source(function)
         digest = f"sha256:{hashlib.sha256(source).hexdigest()}"
+        gpu_marker = _normalize_gpu_marker(gpu)
         runtime = PythonRuntimeSpec(
-            kind="python",
+            kind="python_v2" if gpu_marker is not None else "python",
             python_version=python_version
             or f"{sys.version_info.major}.{sys.version_info.minor}",
             environment=environment_spec,
             env=environment,
+            gpu=gpu_marker,
         )
         self._function = function
         self._request = FunctionRegistrationRequest(
@@ -1002,64 +1284,74 @@ class UdfDefinition:
             ),
             signature=signature,
             runtime=runtime,
-            required_secrets=required_secrets,
         )
         functools.update_wrapper(self, function)
 
     @property
     def registration_request(self) -> FunctionRegistrationRequest:
-        """The immutable, value-free client model for a Function submission."""
+        """The immutable request sent by ``create_function_async``.
+
+        Carries no secret bindings. Binding is a registration-time decision,
+        so a Function bound to Secrets is registered through :meth:`bind_secrets`,
+        which is what ``create_function`` calls.
+        """
         return self._request
 
-    def _submission_json(self, secrets: Optional[Mapping[str, str]]) -> str:
+    def bind_secrets(
-        """Build one registration submission without retaining values on self."""
+        self, secrets: Optional[Sequence[EnvVarSecret]]
-        if secrets is None:
+    ) -> FunctionRegistrationRequest:
-            secret_values: Mapping[str, str] = {}
+        """The registration request for this definition bound to ``secrets``.
-        elif not isinstance(secrets, Mapping):
-            raise TypeError("Function secrets must be a mapping of names to strings")
-        else:
-            secret_values = secrets
 
-        if any(not isinstance(name, str) for name in secret_values):
+        Binding does not change the Function's source: each
-            raise TypeError("Function secret names must be strings")
+        [EnvVarSecret][lancedb.secrets.EnvVarSecret] names a Secret and the
-        expected = set(self._request.required_secrets)
+        environment variable its value should arrive in, and the Function reads
-        provided = set(secret_values)
+        that variable the way it already did. Whether the named Secrets exist is
-        if provided != expected:
+        the server's answer, not this one.
-            missing = sorted(expected - provided)
+        """
-            unexpected = sorted(provided - expected)
+        bindings = () if secrets is None else tuple(secrets)
-            details = []
+        wrong_type = [
-            if missing:
+            binding for binding in bindings if not isinstance(binding, EnvVarSecret)
-                details.append(f"missing: {missing!r}")
+        ]
-            if unexpected:
+        if wrong_type:
-                details.append(f"unexpected: {unexpected!r}")
+            kinds = sorted({type(binding).__name__ for binding in wrong_type})
+            raise TypeError(
+                f"Function secrets must be EnvVarSecret values, not {kinds!r}; a "
+                "credential value is never sent to this API"
+            )
+        variables = [binding.env_variable for binding in bindings]
+        duplicates = sorted({name for name in variables if variables.count(name) > 1})
+        if duplicates:
             raise ValueError(
-                "Function secret values must exactly match the declared secrets ("
+                "a Function binds each environment variable once; duplicated: "
-                + "; ".join(details)
+                f"{duplicates!r}"
-                + ")"
+            )
+        # `env` is ordinary configuration carried in the definition, so a name in
+        # both would have a value visible in the Function's record and a value
+        # that is not. Refuse rather than pick.
+        environment = self._request.runtime.env or {}
+        overlap = sorted(set(environment) & set(variables))
+        if overlap:
+            raise ValueError(
+                f"Function env and secret bindings must be disjoint: {overlap!r}"
+            )
+        if not bindings:
+            return self._request
+        # Sorted, because the list is carried in the FunctionVersion hash and a
+        # caller's argument order is not part of what a Function is.
+        resolved = tuple(
+            sorted(
+                (
+                    SecretBinding(
+                        kind="env",
+                        variable=binding.env_variable,
+                        secret_ref=binding.secret,
+                    )
+                    for binding in bindings
+                ),
+                key=lambda binding: (binding.kind, binding.variable or ""),
             )
-
-        canonical_values = {}
-        total_bytes = 0
-        for name in sorted(secret_values):
-            value = _validate_secret_value(name, secret_values[name])
-            total_bytes += len(value.encode("utf-8"))
-            if total_bytes > _MAX_FUNCTION_SECRET_VALUES_BYTES:
-                raise ValueError(
-                    "Function secret values exceed the "
-                    f"{_MAX_FUNCTION_SECRET_VALUES_BYTES}-byte request limit"
-                )
-            canonical_values[name] = value
-
-        submission = self._request._known_dict()
-        if canonical_values:
-            submission["secret_values"] = canonical_values
-        return json.dumps(
-            submission,
-            ensure_ascii=False,
-            allow_nan=False,
-            sort_keys=True,
-            separators=(",", ":"),
         )
+        return self._request._copy(update={"secret_bindings": resolved})
 
     def __call__(self, *args, **kwargs):
         return self._function(*args, **kwargs)
@@ -1078,8 +1370,8 @@ def udf(
     output_schema: Optional[pa.DataType | pa.Field | pa.Schema] = None,
     pip: tuple[str, ...] | list[str] = (),
     env: Optional[Mapping[str, str]] = None,
-    secrets: tuple[str, ...] | list[str] = (),
     python_version: Optional[str] = None,
+    gpu: bool = False,
     conda: tuple[str, ...] | list[str] = (),
     conda_channels: tuple[str, ...] | list[str] = (),
 ) -> Callable[[Callable[..., Any]], UdfDefinition]: ...
@@ -1093,8 +1385,8 @@ def udf(
     output_schema: Optional[pa.DataType | pa.Field | pa.Schema] = None,
     pip: tuple[str, ...] | list[str] = (),
     env: Optional[Mapping[str, str]] = None,
-    secrets: tuple[str, ...] | list[str] = (),
     python_version: Optional[str] = None,
+    gpu: bool = False,
     conda: tuple[str, ...] | list[str] = (),
     conda_channels: tuple[str, ...] | list[str] = (),
 ):
@@ -1102,8 +1394,9 @@ def udf(
 
     Input and output signatures are inferred from supported annotations. For
     Arrow types annotations cannot express precisely, pass ``input_schema``
-    and ``output_schema`` together. Nullable outputs are rejected because V1
+    and ``output_schema`` together. Scalar outputs must be non-nullable. Every
-    uses physical NULL to represent unassigned computed-column rows.
+    named-struct field may be nullable; Enterprise preserves the struct's
+    validity when the result is expanded into sibling columns.
 
     Parameters
     ----------
@@ -1115,8 +1408,8 @@ def udf(
         Explicit input fields in the exact order of the callable parameters.
         Must be provided together with ``output_schema``.
     output_schema : pyarrow.DataType, pyarrow.Field, or pyarrow.Schema, optional
-        Explicit scalar or named-struct output. Must be non-nullable and be
+        Explicit scalar or named-struct output. Scalar outputs must be
-        provided together with ``input_schema``.
+        non-nullable. Must be provided together with ``input_schema``.
     pip : sequence of str, optional
         Pip requirements for the remote environment.
     conda : sequence of str, optional
@@ -1124,12 +1417,15 @@ def udf(
     conda_channels : sequence of str, optional
         Conda channels in priority order; requires ``conda``.
     env : mapping of str to str, optional
-        Non-secret environment variables. Use ``secrets`` for credentials.
+        Environment variables included in the Function definition. Not for
-    secrets : sequence of str, optional
+        credentials -- these are ordinary configuration, stored with the
-        Names of secrets required by the callable. Supply their values separately
+        Function and visible wherever it is.
-        to ``create_function`` or ``create_function_async``.
     python_version : str, optional
         Remote Python major/minor version. Defaults to the client version.
+    gpu : bool, default False
+        Whether every remote execution requires a GPU. The execution platform
+        selects one compatible GPU for each worker. The requirement is part of
+        the immutable Function version.
 
     The packaged artifact is a snapshot: the function source plus exactly
     the module-level names it references (modules as imports, importable
@@ -1149,15 +1445,16 @@ def udf(
     Examples
     --------
     >>> from lancedb import udf
-    >>> @udf(pip=["numpy==2.2.0"], secrets=["MODEL_TOKEN"])
+    >>> @udf(pip=["numpy==2.2.0"])
     ... def score(value: float) -> float:
     ...     return value * 2
     >>> score(1.5)
     3.0
-    >>> db.create_function(  # doctest: +SKIP
+    >>> @udf(pip=["cupy-cuda12x"], gpu=True)
-    ...     score, secrets={"MODEL_TOKEN": "user-secret-value"}
+    ... def gpu_score(value: int) -> int:
-    ... )
+    ...     return value * 2
-
+    >>> gpu_score.registration_request.runtime.gpu
+    True
     """
 
     def decorate(target: Callable[..., Any]) -> UdfDefinition:
@@ -1168,8 +1465,8 @@ def udf(
             output_schema=output_schema,
             pip=tuple(pip),
             env={} if env is None else env,
-            secrets=tuple(secrets),
             python_version=python_version,
+            gpu=gpu,
             conda=tuple(conda),
             conda_channels=tuple(conda_channels),
         )
@@ -1180,6 +1477,7 @@ def udf(
 
 
 __all__ = [
+    "AssignmentMapping",
     "ApplicationInput",
     "FunctionApplication",
     "FunctionArtifact",

python/python/lancedb/job.py

@@ -4,15 +4,27 @@
 """Handles to operations a server may run asynchronously."""
 
 import asyncio
+import json
 from datetime import timedelta
 from typing import Any, Callable, Generic, Optional, TypeVar, cast
 
+import pyarrow as pa
+
 from lancedb.background_loop import LOOP
 
 from . import _lancedb
+from ._lancedb import JobDescription, JobFailureInfo, JobInfo
 
 T = TypeVar("T")
 
+__all__ = [
+    "AsyncJob",
+    "Job",
+    "JobDescription",
+    "JobFailureInfo",
+    "JobInfo",
+]
+
 
 class AsyncJob(Generic[T]):
     """A handle to an operation that may still be running.
@@ -78,6 +90,149 @@ class AsyncJob(Generic[T]):
             return
         await self._inner.cancel()
 
+    async def refresh(self) -> None:
+        """Ask the backend for this job's current state, and for a server-side
+        job its full record, then cache it for the properties below.
+
+        The properties are all `None` until this runs, because submitting an
+        operation returns only a job id. `status` fetches the whole record too;
+        `wait` records only the terminal state it establishes.
+        """
+        if self._inner is None:
+            return
+        await self._inner.refresh()
+
+    @property
+    def state(self) -> Optional[str]:
+        """The last observed lifecycle state, without contacting the backend.
+
+        `None` until the handle has talked to it. See :meth:`AsyncJob.refresh`.
+        """
+        if self._inner is None:
+            return "finished"
+        return self._inner._state
+
+    @property
+    def job_type(self) -> Optional[str]:
+        """The job's type, as the server names it.
+
+        `None` for an in-process job, which has no server-side record.
+        """
+        return self._field("job_type")
+
+    @property
+    def creation_ms(self) -> Optional[int]:
+        """When the job was created, in milliseconds since the epoch."""
+        return self._field("creation_ms")
+
+    @property
+    def spec(self) -> Optional[Any]:
+        """The job-type-specific specification it was submitted with."""
+        return self._field("spec")
+
+    @property
+    def result(self) -> Optional[Any]:
+        """The job-type-specific terminal result, as reported data rather than
+        the typed model :meth:`AsyncJob.wait` returns.
+
+        `None` until the job succeeds, so a job that never terminates reports
+        its progress through :meth:`AsyncJob.events` instead.
+        """
+        return self._field("result")
+
+    @property
+    def failure(self) -> Optional[JobFailureInfo]:
+        """Why the job failed, when it failed and the server reports a reason."""
+        return self._field("failure")
+
+    @property
+    def _spec_json(self) -> Optional[str]:
+        return self._field("_spec_json")
+
+    @property
+    def _result_json(self) -> Optional[str]:
+        return self._field("_result_json")
+
+    def _field(self, name: str) -> Optional[Any]:
+        description = self._inner._description if self._inner is not None else None
+        return getattr(description, name) if description is not None else None
+
+    async def events(
+        self,
+        *,
+        limit: Optional[int] = None,
+        filter: Optional[str] = None,
+    ) -> "pa.Table":
+        """This job's recorded lifecycle events.
+
+        Where the properties above report a terminal result only once the job
+        reaches one, events are written as the job runs and outlive the workers
+        that produced them. A distributed job records a `claim`/`claim_complete`
+        pair per unit of work, each carrying `rows_processed`, so a job that
+        never finishes still accounts for what it did.
+
+        Parameters
+        ----------
+        limit: int, optional
+            Maximum event rows to return. The server caps results at 1000 by
+            default and 10,000 at most, and truncates without saying so, so
+            pass this for a job that emits an event per fragment.
+        filter: str, optional
+            SQL-like expression over the `state`, `updated_by`, `emitted_from`,
+            `emitted_by`, and `claim_entity` columns, such as
+            ``state = 'claim_complete'``.
+        """
+        if self._inner is None:
+            raise NotImplementedError(
+                "job event history is only available for server-side jobs"
+            )
+        return await self._inner.events(limit=limit, filter=filter)
+
+    def __repr__(self) -> str:
+        return _job_repr("AsyncJob", self)
+
+
+_REPR_INDENT = " " * 4
+
+
+def _repr_payload(value: Any) -> str:
+    """Render a job payload as indented JSON, aligned under its field."""
+    try:
+        rendered = json.dumps(value, indent=4)
+    except TypeError:
+        return repr(value)
+    return rendered.replace("\n", "\n" + _REPR_INDENT)
+
+
+def _job_repr(kind: str, job: Any) -> str:
+    """Render every field the handle currently knows, omitting the rest.
+
+    One field per line, with the JSON payloads indented, because a refresh
+    job's spec and result are the point of printing it.
+    """
+    state = job.state
+    if state is None:
+        # Nothing has been fetched yet, so there is nothing to lay out.
+        known = f"id={job.id!r}, " if job.id is not None else ""
+        return f"{kind}({known}not refreshed)"
+
+    fields = []
+    if job.id is not None:
+        fields.append(f"id={job.id!r}")
+    fields.append(f"state={state!r}")
+    for name in ("job_type", "creation_ms"):
+        value = getattr(job, name)
+        if value is not None:
+            fields.append(f"{name}={value!r}")
+    for name in ("spec", "result"):
+        value = getattr(job, name)
+        if value is not None:
+            fields.append(f"{name}={_repr_payload(value)}")
+    if job.failure is not None:
+        fields.append(f"failure={job.failure!r}")
+    body = "".join(f"\n{_REPR_INDENT}{field}," for field in fields)
+    return f"{kind}({body}\n)"
+
 
 class Job(Generic[T]):
     """Synchronous counterpart of `AsyncJob` with the same result type."""
@@ -122,6 +277,75 @@ class Job(Generic[T]):
             return
         LOOP.run(self._inner.cancel())
 
+    def refresh(self) -> None:
+        """Ask the backend for this job's current state and record.
+
+        See :meth:`AsyncJob.refresh`.
+        """
+        if self._inner is None:
+            return
+        LOOP.run(self._inner.refresh())
+
+    @property
+    def state(self) -> Optional[str]:
+        """The last observed lifecycle state. See :attr:`AsyncJob.state`."""
+        return self._inner.state if self._inner is not None else "finished"
+
+    @property
+    def job_type(self) -> Optional[str]:
+        """The job's type. See :attr:`AsyncJob.job_type`."""
+        return self._field("job_type")
+
+    @property
+    def creation_ms(self) -> Optional[int]:
+        """When the job was created. See :attr:`AsyncJob.creation_ms`."""
+        return self._field("creation_ms")
+
+    @property
+    def spec(self) -> Optional[Any]:
+        """The job's specification. See :attr:`AsyncJob.spec`."""
+        return self._field("spec")
+
+    @property
+    def result(self) -> Optional[Any]:
+        """The job's terminal result. See :attr:`AsyncJob.result`."""
+        return self._field("result")
+
+    @property
+    def failure(self) -> Optional[JobFailureInfo]:
+        """Why the job failed. See :attr:`AsyncJob.failure`."""
+        return self._field("failure")
+
+    @property
+    def _spec_json(self) -> Optional[str]:
+        return self._field("_spec_json")
+
+    @property
+    def _result_json(self) -> Optional[str]:
+        return self._field("_result_json")
+
+    def _field(self, name: str) -> Optional[Any]:
+        return getattr(self._inner, name) if self._inner is not None else None
+
+    def events(
+        self,
+        *,
+        limit: Optional[int] = None,
+        filter: Optional[str] = None,
+    ) -> "pa.Table":
+        """This job's recorded lifecycle events.
+
+        See :meth:`AsyncJob.events`.
+        """
+        if self._inner is None:
+            raise NotImplementedError(
+                "job event history is only available for server-side jobs"
+            )
+        return LOOP.run(self._inner.events(limit=limit, filter=filter))
+
+    def __repr__(self) -> str:
+        return _job_repr("Job", self)
+
 
 def _typed_job(
     inner: "_lancedb.Job", result_decoder: Callable[[str], T]

python/python/lancedb/materialized_view.py

@@ -42,6 +42,8 @@ class MaterializedViewDefinition:
     """Cap on the number of rows the view holds."""
     inputs: List[str] = field(default_factory=list)
     """Source columns the projections and filter read."""
+    source_namespace: List[str] = field(default_factory=list)
+    """Namespace holding the source table; empty is the root namespace."""
 
 
 def _definition_from_schema(
@@ -53,7 +55,8 @@ def _definition_from_schema(
         raise ValueError(f"Table '{name}' is not a materialized view")
     value = json.loads(raw)
     kind = value.get("kind")
-    if kind != "select":
+    # "namespaced_select" keeps older readers from resolving the source at root.
+    if kind not in ("select", "namespaced_select"):
         raise NotImplementedError(
             f"materialized view '{name}' is defined by '{kind}', which this "
             "version of lancedb cannot refresh"
@@ -66,6 +69,7 @@ def _definition_from_schema(
         filter=value.get("filter"),
         limit=value.get("limit"),
         inputs=value.get("inputs", []),
+        source_namespace=value.get("source_namespace", []),
     )
 
 

python/python/lancedb/namespace.py

@@ -12,6 +12,7 @@ from __future__ import annotations
 
 import sys
 from typing import TYPE_CHECKING, Any, Dict, Iterable, List, Optional, Union
+from uuid import UUID
 
 if sys.version_info >= (3, 12):
     from typing import override
@@ -48,8 +49,11 @@ from lancedb._lancedb import (
     connect_namespace_client as _connect_namespace_client,
 )
 from lancedb.background_loop import LOOP
+from lancedb.arrow import AsyncRecordBatchReader
 from lancedb.db import AsyncConnection, DBConnection
 from lancedb.job import AsyncJob, Job
+from lancedb.sql import AsyncQuery as AsyncSqlQuery
+from lancedb.sql import QueryDescription
 from lance_namespace import (
     LanceNamespace,
     connect as namespace_connect,
@@ -1447,6 +1451,37 @@ class AsyncLanceNamespaceDBConnection:
             namespace_path=namespace_path, page_token=page_token, limit=limit
         )
 
+    async def execute_query(
+        self,
+        query: str,
+        *,
+        default_namespace_path: Optional[List[str]] = None,
+    ) -> AsyncRecordBatchReader:
+        """Execute SQL when supported by the underlying connection."""
+        return await self._inner.execute_query(
+            query,
+            default_namespace_path=default_namespace_path,
+        )
+
+    async def execute_query_async(
+        self,
+        query: str,
+        *,
+        default_namespace_path: Optional[List[str]] = None,
+    ) -> AsyncSqlQuery:
+        """Start executing SQL when supported by the underlying connection.
+
+        Namespace-backed local connections do not support SQL.
+        """
+        return await self._inner.execute_query_async(
+            query,
+            default_namespace_path=default_namespace_path,
+        )
+
+    async def describe_query(self, query_id: UUID) -> QueryDescription:
+        """Describe a submitted SQL query when supported."""
+        return await self._inner.describe_query(query_id)
+
     async def namespace_client(self) -> LanceNamespace:
         """Get the namespace client for this connection.
 

python/python/lancedb/query.py

@@ -109,6 +109,7 @@ def _query_is_plain_scan(query: Query) -> bool:
     return (
         query.vector is None
         and query.full_text_query is None
+        and query.take_offsets is None
         and not query.postfilter
         and not query.order_by
     )
@@ -804,6 +805,10 @@ class Query(pydantic.BaseModel):
     # offset to start fetching results from
     offset: Optional[int] = None
 
+    # Dataset offsets whose duplicate occurrences must be restored after lookup.
+    # This is populated when a take query is converted to this serializable form.
+    take_offsets: Optional[List[int]] = None
+
     # if true, will only search the indexed data
     fast_search: Optional[bool] = None
 
@@ -825,6 +830,7 @@ class Query(pydantic.BaseModel):
         query = cls()
         query.limit = req.limit
         query.offset = req.offset
+        query.take_offsets = req.take_offsets
         query.filter = req.filter
         query.full_text_query = req.full_text_search
         query.columns = req.select

python/python/lancedb/remote/db.py

@@ -7,8 +7,18 @@ import json
 import logging
 from concurrent.futures import ThreadPoolExecutor
 import sys
-from typing import TYPE_CHECKING, Any, Dict, Iterable, List, Mapping, Optional, Union
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Dict,
+    Iterable,
+    List,
+    Optional,
+    Sequence,
+    Union,
+)
 from urllib.parse import urlparse
+from uuid import UUID
 import warnings
 
 if sys.version_info >= (3, 12):
@@ -25,10 +35,13 @@ from ..common import DATA
 from ..db import DBConnection, LOOP
 from ..functions import FunctionVersion, UdfDefinition
 from ..job import AsyncJob, Job
+from ..sql import Query as SqlQuery
+from ..sql import QueryDescription
 from ..materialized_view import MaterializedView, SelectArg
+from ..secrets import EnvVarSecret, SecretInfo
 
 if TYPE_CHECKING:
-    from .._lancedb import JobDescription, JobInfo
+    from .._lancedb import JobInfo
 from ..embeddings import EmbeddingFunctionConfig
 from lance_namespace import (
     LanceNamespace,
@@ -116,6 +129,7 @@ class RemoteDBConnection(DBConnection):
         read_timeout: Optional[float] = None,
         storage_options: Optional[Dict[str, str]] = None,
         read_consistency_interval: Optional[timedelta] = None,
+        sql_host_override: Optional[str] = None,
     ):
         """Connect to a remote LanceDB database."""
         if isinstance(client_config, dict):
@@ -161,6 +175,7 @@ class RemoteDBConnection(DBConnection):
         self.api_key = api_key
         self.region = region
         self.host_override = host_override
+        self.sql_host_override = sql_host_override
         self.storage_options = storage_options
         self.db_name = parsed.netloc
 
@@ -175,6 +190,7 @@ class RemoteDBConnection(DBConnection):
                 api_key=api_key,
                 region=region,
                 host_override=host_override,
+                sql_host_override=sql_host_override,
                 client_config=client_config,
                 storage_options=storage_options,
                 read_consistency_interval=read_consistency_interval,
@@ -193,6 +209,7 @@ class RemoteDBConnection(DBConnection):
                 "api_key": self.api_key,
                 "region": self.region,
                 "host_override": self.host_override,
+                "sql_host_override": self.sql_host_override,
                 "client_config": _client_config_to_dict(self.client_config),
                 "storage_options": self.storage_options,
             }
@@ -732,43 +749,59 @@ class RemoteDBConnection(DBConnection):
         )
 
     @override
-    def job(self, job_id: str) -> Job:
+    def open_job(self, job_id: str) -> Job:
-        """A [Job][lancedb.job.Job] handle for a server-side job by id.
+        """Open a server-side job by id. See
-
+        [DBConnection.open_job][lancedb.db.DBConnection.open_job].
-        The handle is constructed without a server round trip; an unknown id
-        surfaces when the handle is used. Dropping the handle has no effect
-        on the job itself.
         """
-        return Job(self._conn.job(job_id))
+        return Job(LOOP.run(self._conn.open_job(job_id)))
 
     @override
     def create_function_async(
         self,
         definition: UdfDefinition,
         *,
-        secrets: Optional[Mapping[str, str]] = None,
+        secrets: Optional[Sequence[EnvVarSecret]] = None,
     ) -> Job[FunctionVersion]:
-        return Job(
+        job = LOOP.run(self._conn.create_function_async(definition, secrets=secrets))
-            LOOP.run(self._conn.create_function_async(definition, secrets=secrets))
+        return Job(job)
-        )
 
     @override
     def get_function(self, name: str, *, version: str) -> FunctionVersion:
         return LOOP.run(self._conn.get_function(name, version=version))
 
+    @override
+    def list_functions(self) -> List[FunctionVersion]:
+        return LOOP.run(self._conn.list_functions())
+
+    @override
+    def drop_function(self, name: str, *, version: str) -> bool:
+        return LOOP.run(self._conn.drop_function(name, version=version))
+
+    @override
+    def create_secret(self, name: str, value: str) -> None:
+        LOOP.run(self._conn.create_secret(name, value))
+
+    @override
+    def alter_secret(self, name: str, value: str) -> None:
+        LOOP.run(self._conn.alter_secret(name, value))
+
+    @override
+    def describe_secret(self, name: str) -> SecretInfo:
+        return LOOP.run(self._conn.describe_secret(name))
+
+    @override
+    def list_secrets(self) -> List[str]:
+        return LOOP.run(self._conn.list_secrets())
+
+    @override
+    def drop_secret(self, name: str) -> None:
+        LOOP.run(self._conn.drop_secret(name))
+
     @override
     def list_jobs(self) -> List["JobInfo"]:
         """List server-side jobs across the database's tables."""
         return LOOP.run(self._conn.list_jobs())
 
-    @override
-    def get_job(self, job_id: str) -> Optional["JobDescription"]:
-        """Describe a single server-side job by id.
-
-        Returns None when the server has no such job.
-        """
-        return LOOP.run(self._conn.get_job(job_id))
-
     @override
     def cancel_job(self, job_id: str) -> bool:
         """Request cancellation of a server-side job by id.
@@ -780,12 +813,35 @@ class RemoteDBConnection(DBConnection):
         return LOOP.run(self._conn.cancel_job(job_id))
 
     @override
-    def job_history(self, job_id: Optional[str] = None) -> List[pa.RecordBatch]:
+    def execute_query_async(
-        """The lifecycle event history of a server-side job, as Arrow batches.
+        self,
+        query: str,
+        *,
+        default_namespace_path: Optional[List[str]] = None,
+    ) -> SqlQuery:
+        """Start executing SQL through this remote connection.
 
-        Lists history across all jobs when `job_id` is None.
+        Unqualified tables use this connection's database and the
+        ``["public"]`` namespace by default. Fully qualified table names may
+        reference other databases available to the same deployment.
         """
-        return LOOP.run(self._conn.job_history(job_id))
+        return SqlQuery(
+            LOOP.run(
+                self._conn.execute_query_async(
+                    query,
+                    default_namespace_path=default_namespace_path,
+                )
+            )
+        )
+
+    @override
+    def describe_query(self, query_id: UUID) -> QueryDescription:
+        """Describe a submitted SQL query by its connection-scoped id."""
+        return LOOP.run(
+            self._conn.describe_query(
+                query_id,
+            )
+        )
 
     @override
     def namespace_client(self) -> LanceNamespace:

python/python/lancedb/remote/header.py

@@ -177,4 +177,7 @@ class OAuthProvider(HeaderProvider):
         if not self._current_token:
             raise RuntimeError("Failed to obtain OAuth token")
 
-        return {"Authorization": f"Bearer {self._current_token}"}
+        return {
+            "Authorization": f"Bearer {self._current_token}",
+            "x-lancedb-credential-type": "oidc",
+        }

python/python/lancedb/remote/table.py

@@ -67,7 +67,15 @@ from ..query import (
     LanceTakeQueryBuilder,
     LanceVectorQueryBuilder,
 )
-from ..table import AsyncTable, BlobMode, Branches, IndexStatistics, Query, Table, Tags
+from ..table import (
+    AsyncTable,
+    BlobMode,
+    Branches,
+    IndexStatistics,
+    Query,
+    Table,
+    Tags,
+)
 from ..types import BaseTokenizerType
 
 
@@ -540,6 +548,7 @@ class RemoteTable(Table):
         LOOP.run(
             self._table.create_index(
                 column,
+                replace=replace,
                 config=config,
                 wait_timeout=wait_timeout,
                 name=name,

python/python/lancedb/secrets.py

@@ -0,0 +1,174 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright The LanceDB Authors
+
+"""Named Secrets, and the bindings that deliver them to Functions.
+
+A Secret is a database-scoped named credential. Nothing in this module holds a
+value: :class:`EnvVarSecret` names one and says which environment variable it
+should arrive in, and the value is resolved by the remote service when a
+Function bound to it runs. No API returns a stored credential, by construction
+rather than by policy -- there is no code path that could.
+"""
+
+from __future__ import annotations
+
+import re
+
+# The same characters LanceDB already admits in a namespace or table name, and
+# no positional rule on top of them: a segment may begin with `_`, `-` or `.`
+# today, so anything narrower would put Secrets out of reach inside namespaces
+# that already exist. Matches the service, which admits the same set.
+_SECRET_NAME = re.compile(r"^[A-Za-z0-9_.-]{1,255}$")
+_ENV_VARIABLE = re.compile(r"^[A-Za-z_][A-Za-z0-9_]*$")
+
+
+def validate_secret_name(name: str) -> str:
+    """Check a Secret name locally and return it unchanged."""
+    if not isinstance(name, str):
+        raise TypeError(f"Secret name must be a string, not {type(name).__name__}")
+    if not _SECRET_NAME.fullmatch(name):
+        raise ValueError(f"invalid Secret name: {name!r}")
+    return name
+
+
+def validate_env_variable(name: str) -> str:
+    """Check an environment variable name locally and return it unchanged."""
+    if not isinstance(name, str):
+        raise TypeError(
+            f"environment variable name must be a string, not {type(name).__name__}"
+        )
+    if not _ENV_VARIABLE.fullmatch(name):
+        raise ValueError(f"invalid environment variable name: {name!r}")
+    return name
+
+
+class EnvVarSecret:
+    """A Secret bound to the environment variable a Function's library reads.
+
+    Pass these in the ``secrets`` sequence of
+    [DBConnection.create_function][lancedb.db.DBConnection.create_function]. The
+    Function's source is unchanged by binding: it reads ``OPENAI_API_KEY`` the
+    way it always did, and the binding is what puts a value there.
+
+    This is a local value. Constructing it contacts no server, so it always
+    succeeds and says nothing about whether the Secret exists; that is checked
+    at registration, where a mistyped Secret name surfaces as a clear "does not
+    exist" naming both the Secret and the variable bound to it. A mistyped
+    *variable* name cannot be caught anywhere -- nothing knows which variables a
+    Function reads -- so it surfaces on the first rows instead.
+
+    The type exists so a credential cannot be passed by accident. A bare string
+    in the same position is a plausible-looking mistake with the opposite
+    meaning, and it reads identically in a diff.
+
+    Parameters
+    ----------
+    secret : str
+        The Secret's database-scoped name.
+    env_variable : str
+        The environment variable the Function reads it from.
+
+    Examples
+    --------
+    >>> from lancedb import EnvVarSecret
+    >>> binding = EnvVarSecret(secret="openai-prod", env_variable="OPENAI_API_KEY")
+    >>> binding.secret, binding.env_variable
+    ('openai-prod', 'OPENAI_API_KEY')
+    """
+
+    __slots__ = ("_secret", "_env_variable")
+
+    def __init__(self, secret: str, env_variable: str):
+        self._secret = validate_secret_name(secret)
+        self._env_variable = validate_env_variable(env_variable)
+
+    @property
+    def secret(self) -> str:
+        """The Secret's database-scoped name."""
+        return self._secret
+
+    @property
+    def env_variable(self) -> str:
+        """The environment variable the value is delivered in."""
+        return self._env_variable
+
+    def __repr__(self) -> str:
+        return (
+            f"EnvVarSecret(secret={self._secret!r}, "
+            f"env_variable={self._env_variable!r})"
+        )
+
+    def __eq__(self, other: object) -> bool:
+        return (
+            isinstance(other, EnvVarSecret)
+            and other._secret == self._secret
+            and other._env_variable == self._env_variable
+        )
+
+    def __hash__(self) -> int:
+        return hash((EnvVarSecret, self._secret, self._env_variable))
+
+
+class SecretInfo:
+    """What a database records about a Secret. Never its value.
+
+    Returned by
+    [DBConnection.describe_secret][lancedb.db.DBConnection.describe_secret].
+    """
+
+    __slots__ = ("_name", "_created_at_millis", "_updated_at_millis")
+
+    def __init__(self, name: str, created_at_millis: int, updated_at_millis: int):
+        self._name = name
+        self._created_at_millis = created_at_millis
+        self._updated_at_millis = updated_at_millis
+
+    @property
+    def name(self) -> str:
+        """The Secret's database-scoped name."""
+        return self._name
+
+    @property
+    def created_at_millis(self) -> int:
+        """When the Secret was created, in milliseconds since the Unix epoch."""
+        return self._created_at_millis
+
+    @property
+    def updated_at_millis(self) -> int:
+        """When the Secret's value was last rotated, in epoch milliseconds.
+
+        The only observable that a rotation landed: no API returns a credential,
+        so a caller confirms ``alter_secret`` took effect by watching this move.
+        """
+        return self._updated_at_millis
+
+    @classmethod
+    def from_json(cls, value: dict) -> "SecretInfo":
+        return cls(
+            name=value["name"],
+            created_at_millis=value["created_at_millis"],
+            updated_at_millis=value["updated_at_millis"],
+        )
+
+    def __repr__(self) -> str:
+        return (
+            f"SecretInfo(name={self._name!r}, "
+            f"created_at_millis={self._created_at_millis!r}, "
+            f"updated_at_millis={self._updated_at_millis!r})"
+        )
+
+    def __eq__(self, other: object) -> bool:
+        return (
+            isinstance(other, SecretInfo)
+            and other._name == self._name
+            and other._created_at_millis == self._created_at_millis
+            and other._updated_at_millis == self._updated_at_millis
+        )
+
+
+__all__ = [
+    "EnvVarSecret",
+    "SecretInfo",
+    "validate_env_variable",
+    "validate_secret_name",
+]

python/python/lancedb/sql.py

@@ -0,0 +1,88 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright The LanceDB Authors
+
+"""Handles to SQL queries running on a remote database."""
+
+from uuid import UUID
+
+import pyarrow as pa
+
+from lancedb.background_loop import LOOP
+
+from . import _lancedb
+from .arrow import AsyncRecordBatchReader
+
+QueryDescription = _lancedb.QueryDescription
+
+
+class AsyncQuery:
+    """A handle to a submitted SQL query on an asynchronous connection."""
+
+    def __init__(self, inner: "_lancedb.SqlQuery"):
+        self._inner = inner
+
+    @property
+    def id(self) -> UUID:
+        """The stable identifier scoped to the connection that submitted it."""
+        return self._inner.id
+
+    async def describe(self) -> QueryDescription:
+        """Get a point-in-time description of the query."""
+        return await self._inner.describe()
+
+    async def reader(self) -> AsyncRecordBatchReader:
+        """Wait for the initial result stream and return its Arrow reader.
+
+        Results are single-consumer. Calling this method more than once on the
+        same query raises an error. Later batches are streamed as they become
+        available without waiting for the full query to finish.
+        """
+        return AsyncRecordBatchReader(await self._inner.reader())
+
+    async def cancel(self) -> None:
+        """Request cancellation of the query."""
+        await self._inner.cancel()
+
+
+class Query:
+    """Synchronous counterpart of :class:`AsyncQuery`."""
+
+    def __init__(self, inner: AsyncQuery):
+        self._inner = inner
+
+    @property
+    def id(self) -> UUID:
+        """The stable identifier scoped to the connection that submitted it."""
+        return self._inner.id
+
+    def describe(self) -> QueryDescription:
+        """Get a point-in-time description of the query."""
+        return LOOP.run(self._inner.describe())
+
+    def reader(self) -> pa.RecordBatchReader:
+        """Wait for the initial result stream and return a blocking reader.
+
+        Results are single-consumer. Calling this method more than once on the
+        same query raises an error. Later batches block only until they become
+        available, without waiting for the full query to finish.
+        """
+        reader = LOOP.run(self._inner.reader())
+
+        def next_batch():
+            try:
+                return LOOP.run(reader.__anext__())
+            except StopAsyncIteration:
+                return None
+
+        def batches():
+            while (batch := next_batch()) is not None:
+                yield batch
+
+        return pa.RecordBatchReader.from_batches(reader.schema, batches())
+
+    def cancel(self) -> None:
+        """Request cancellation of the query."""
+        LOOP.run(self._inner.cancel())
+
+
+__all__ = ["AsyncQuery", "Query", "QueryDescription"]

python/python/lancedb/table.py

@@ -1547,7 +1547,9 @@ class Table(ABC):
         on: Union[str, Iterable[str]]
             A column (or columns) to join on.  This is how records from the
             source table and target table are matched.  Typically this is some
-            kind of key or id column.
+            kind of key or id column.  Passing several columns matches on the
+            composite key: a source row updates a target row only when it
+            agrees on every one of them.
 
         Examples
         --------
@@ -1678,9 +1680,9 @@ class Table(ABC):
         Offsets are mostly useful for sampling as the set of all valid offsets is easily
         known in advance to be [0, len(table)).
 
-        No guarantees are made regarding the order in which results are returned.  If
+        No guarantees are made regarding the order in which results are returned.
-        you desire an output order that matches the order of the given offsets, you will
+        Repeated offsets produce repeated rows, which makes this method suitable for
-        need to add the row offset column to the output and align it yourself.
+        sampling with replacement.
 
         Parameters
         ----------
@@ -1791,6 +1793,9 @@ class Table(ABC):
         The result has the same length and order as ``row_ids``. Null blobs
         produce null slots; valid empty blobs produce ``b""``.
 
+        ``_rowid`` values stay valid after compaction when the table has stable
+        row ids.
+
         Convenience for small payloads. For large values use
         :meth:`fetch_blob_files`.
         """
@@ -1808,6 +1813,9 @@ class Table(ABC):
         The result has the same length and order as ``requests``; null blobs
         produce null slots and empty ranges on non-null blobs produce ``b""``.
 
+        ``_rowid`` values stay valid after compaction when the table has stable
+        row ids.
+
         Row IDs can be obtained from a query with ``with_row_id(True)``. This
         API is currently supported only by local tables.
         """
@@ -1823,6 +1831,9 @@ class Table(ABC):
         ``_rowid`` or a ``_lance_row_id`` field on the blob descriptor. Null
         rows are ``None``. Remote tables require LanceDB Cloud server 0.5.0 or
         newer.
+
+        ``_rowid`` values stay valid after compaction when the table has stable
+        row ids.
         """
 
     @abstractmethod
@@ -2165,9 +2176,11 @@ class Table(ABC):
             Function columns are supported only on LanceDB Cloud and
             Enterprise.
         computed: Dict[str, str], optional
-            A map of column name to a SQL expression defining the column. The
+            A mapping from output column names to SQL expressions derives each
-            column's type and inputs are derived from the expression, so no
+            output field from its expression. A direct projection of a Blob v2
-            data type is supplied.
+            field inherits Blob v2 semantics; other expressions derive their
+            ordinary Arrow type. Mapping order is declaration and dependency
+            order.
 
             Unlike ``transforms``, the expression is stored rather than
             evaluated now: the column is committed with no values, and rows get
@@ -4088,6 +4101,7 @@ class LanceTable(Table):
             )
             and not self._route_pushdown_to_rust
             and self.current_branch() is None
+            and query.take_offsets is None
         ):
             from lancedb.namespace import _execute_server_side_query
 
@@ -5698,7 +5712,9 @@ class AsyncTable:
         on: Union[str, Iterable[str]]
             A column (or columns) to join on.  This is how records from the
             source table and target table are matched.  Typically this is some
-            kind of key or id column.
+            kind of key or id column.  Passing several columns matches on the
+            composite key: a source row updates a target row only when it
+            agrees on every one of them.
 
         Examples
         --------
@@ -5981,7 +5997,23 @@ class AsyncTable:
 
     def _sync_query_to_async(
         self, query: Query
-    ) -> AsyncHybridQuery | AsyncFTSQuery | AsyncVectorQuery | AsyncQuery:
+    ) -> (
+        AsyncHybridQuery
+        | AsyncFTSQuery
+        | AsyncVectorQuery
+        | AsyncQuery
+        | AsyncTakeQuery
+    ):
+        if query.take_offsets is not None:
+            take_query = self.take_offsets(query.take_offsets)
+            if query.columns:
+                take_query = take_query.select(query.columns)
+            if query.use_lsm is not None:
+                take_query = take_query.use_lsm(query.use_lsm)
+            if query.with_row_id:
+                take_query = take_query.with_row_id()
+            return take_query
+
         async_query = self.query()
         if query.limit is not None:
             async_query = async_query.limit(query.limit)
@@ -6046,6 +6078,7 @@ class AsyncTable:
                 self._namespace_client, self._pushdown_operations
             )
             and not self._route_pushdown_to_rust
+            and query.take_offsets is None
         ):
             from lancedb.namespace import _execute_server_side_query
 
@@ -6268,8 +6301,11 @@ class AsyncTable:
             Function columns are supported only on LanceDB Cloud and
             Enterprise.
         computed: Dict[str, str], optional
-            A map of column name to a SQL expression defining the column. The
+            A mapping from output column names to SQL expressions derives each
-            column's type and inputs are derived from the expression.
+            output field from its expression. A direct projection of a Blob v2
+            field inherits Blob v2 semantics; other expressions derive their
+            ordinary Arrow type. Mapping order is declaration and dependency
+            order.
 
             Unlike ``transforms``, the expression is stored rather than
             evaluated now: the column is committed with no values, and rows get
@@ -6540,6 +6576,9 @@ class AsyncTable:
         Offsets are mostly useful for sampling as the set of all valid offsets is easily
         known in advance to be [0, len(table)).
 
+        No guarantees are made regarding the order in which results are returned.
+        Repeated offsets produce repeated rows.
+
         Parameters
         ----------
         offsets: list[int]

python/python/tests/test_blob.py

@@ -66,6 +66,25 @@ def _row_ids_by_id(table):
     return dict(zip(hits["id"].to_pylist(), hits["_rowid"].to_pylist()))
 
 
+def _assert_missing_blob_row_ids(exc_info):
+    message = str(exc_info.value)
+    assert "row ids" in message
+    assert "rowaddr" not in message
+    assert "fragment" not in message
+
+
+def _assert_fetch_apis_reject_missing_row_ids(table, row_ids):
+    with pytest.raises(ValueError) as exc_info:
+        table.fetch_blobs("image", row_ids)
+    _assert_missing_blob_row_ids(exc_info)
+    with pytest.raises(ValueError) as exc_info:
+        table.fetch_blob_files("image", row_ids)
+    _assert_missing_blob_row_ids(exc_info)
+    with pytest.raises(ValueError) as exc_info:
+        table.fetch_blob_ranges("image", [(row_id, 0, 1) for row_id in row_ids])
+    _assert_missing_blob_row_ids(exc_info)
+
+
 def test_blob_factory_declares_v2_field():
     field = lancedb.blob("image")
     assert isinstance(field.type, pa.ExtensionType)
@@ -691,6 +710,25 @@ def test_fetch_blobs_accepts_query_result():
     assert {blobs[i].as_py() for i in range(len(blobs))} == {b"gamma"}
 
 
+def test_fetch_blobs_after_compact_with_stable_row_ids(tmp_path):
+    db = lancedb.connect(
+        tmp_path, storage_options={"new_table_enable_stable_row_ids": "true"}
+    )
+    schema = pa.schema([pa.field("id", pa.int64()), lancedb.blob("image")])
+    table = db.create_table("t", schema=schema)
+    table.add([{"id": 1, "image": b"frag-one"}])
+    table.add([{"id": 2, "image": b"frag-two"}])
+    by_id = _row_ids_by_id(table)
+    ids = [by_id[1], by_id[2]]
+
+    table.optimize()
+
+    blobs = table.fetch_blobs("image", ids)
+    assert blobs.to_pylist() == [b"frag-one", b"frag-two"]
+    ranges = table.fetch_blob_ranges("image", [(ids[0], 5, 3), (ids[1], 5, 3)])
+    assert ranges.to_pylist() == [b"one", b"two"]
+
+
 def test_fetch_blobs_preserves_null_and_empty_values():
     table = _blob_table(
         "nulls",
@@ -739,8 +777,25 @@ def test_fetch_blob_ranges_validates_requests():
     with pytest.raises(ValueError, match="offset \\+ length overflowed"):
         table.fetch_blob_ranges("image", [(row_id, 2**64 - 1, 1)])
 
-    with pytest.raises(ValueError, match="row IDs"):
+    with pytest.raises(ValueError) as exc_info:
         table.fetch_blob_ranges("image", [(2**64 - 1, 0, 1)])
+    _assert_missing_blob_row_ids(exc_info)
+
+
+def test_fetch_blob_apis_reject_missing_fragment_row_addr():
+    table = _blob_table("missing_frag", [{"id": 1, "image": b"x"}])
+    live = _row_ids_by_id(table)[1]
+    _assert_fetch_apis_reject_missing_row_ids(table, [1 << 32, live])
+
+
+def test_fetch_blob_apis_reject_deleted_row_ids():
+    table = _blob_table(
+        "deleted_rows",
+        [{"id": 1, "image": b"one"}, {"id": 2, "image": b"two"}],
+    )
+    by_id = _row_ids_by_id(table)
+    table.delete("id = 2")
+    _assert_fetch_apis_reject_missing_row_ids(table, [by_id[2], by_id[1]])
 
 
 def test_fetch_blob_ranges_empty_requests_returns_empty_array():

python/python/tests/test_first_class_function_slice1.py

@@ -13,6 +13,7 @@ from lancedb.functions import (
     FunctionBinding,
     FunctionVersion,
     PythonRuntimeSpec,
+    SecretBinding,
     RefreshColumnResult,
 )
 from lancedb.table import AsyncTable
@@ -38,6 +39,7 @@ def job_result(name: str) -> dict:
 
 
 def assert_no_secret_values(value):
+    """No client value models a resolved credential, at any nesting depth."""
     if isinstance(value, dict):
         for key, child in value.items():
             assert key not in {
@@ -109,7 +111,9 @@ def test_function_version_identity_is_immutable_and_exact():
     version = FunctionVersion.from_json(json.dumps(value))
     assert version.name == "embed"
     assert version.version == "fv_01K3EXACT"
-    assert version.required_secrets == ("HF_TOKEN",)
+    assert list(version.secret_bindings) == [
+        SecretBinding(kind="env", variable="HF_TOKEN", secret_ref="hf-prod")
+    ]
 
     with pytest.raises((TypeError, ValueError)):
         version.version = "fv_changed"
@@ -292,15 +296,27 @@ def test_refresh_result_rejects_non_u64_values(field):
         RefreshColumnResult.from_json(json.dumps(value))
 
 
-def test_canonical_client_values_contain_secret_names_only():
+def test_canonical_client_values_carry_bindings_and_no_credentials():
+    """A binding names a Secret; the credential behind it has no client field."""
     version = FunctionVersion.from_json(
         json.dumps(job_result("remote_function_job.json"))
     )
     canonical = json.loads(version.to_canonical_json())
-    assert canonical["required_secrets"] == ["HF_TOKEN"]
+    assert canonical["secret_bindings"] == [
+        {"kind": "env", "variable": "HF_TOKEN", "secret_ref": "hf-prod"}
+    ]
     assert_no_secret_values(canonical)
 
 
+def test_a_version_without_bindings_keeps_the_original_wire_shape():
+    """Every Function registered before Secrets existed serializes unchanged."""
+    value = job_result("remote_function_job.json")
+    del value["secret_bindings"]
+    version = FunctionVersion.from_json(json.dumps(value))
+    assert list(version.secret_bindings) == []
+    assert "secret_bindings" not in json.loads(version.to_canonical_json())
+
+
 class _FunctionDeclarationInner:
     def __init__(self):
         self.calls = []

python/python/tests/test_header_provider.py

@@ -54,7 +54,10 @@ class TestOAuthProvider:
         provider = OAuthProvider(fetcher)
         headers = provider.get_headers()
 
-        assert headers == {"Authorization": "Bearer token123"}
+        assert headers == {
+            "Authorization": "Bearer token123",
+            "x-lancedb-credential-type": "oidc",
+        }
         assert provider._current_token == "token123"
         assert provider._token_expires_at is not None
 
@@ -73,14 +76,20 @@ class TestOAuthProvider:
 
         # First call
         headers1 = provider.get_headers()
-        assert headers1 == {"Authorization": "Bearer token1"}
+        assert headers1 == {
+            "Authorization": "Bearer token1",
+            "x-lancedb-credential-type": "oidc",
+        }
 
         # Wait for token to expire
         time.sleep(1.1)
 
         # Second call should refresh
         headers2 = provider.get_headers()
-        assert headers2 == {"Authorization": "Bearer token2"}
+        assert headers2 == {
+            "Authorization": "Bearer token2",
+            "x-lancedb-credential-type": "oidc",
+        }
         assert call_count == 2
 
     def test_no_expiry_info(self):
@@ -92,12 +101,18 @@ class TestOAuthProvider:
         provider = OAuthProvider(fetcher)
         headers = provider.get_headers()
 
-        assert headers == {"Authorization": "Bearer permanent_token"}
+        assert headers == {
+            "Authorization": "Bearer permanent_token",
+            "x-lancedb-credential-type": "oidc",
+        }
         assert provider._token_expires_at is None
 
         # Should not refresh on second call
         headers2 = provider.get_headers()
-        assert headers2 == {"Authorization": "Bearer permanent_token"}
+        assert headers2 == {
+            "Authorization": "Bearer permanent_token",
+            "x-lancedb-credential-type": "oidc",
+        }
 
     def test_missing_access_token(self):
         """Test error handling when access_token is missing."""
@@ -121,7 +136,10 @@ class TestOAuthProvider:
         provider = OAuthProvider(fetcher)
         headers = provider.get_headers()
 
-        assert headers == {"Authorization": "Bearer sync_token"}
+        assert headers == {
+            "Authorization": "Bearer sync_token",
+            "x-lancedb-credential-type": "oidc",
+        }
 
 
 class TestClientConfigIntegration:

python/python/tests/test_materialized_views.py

@@ -266,3 +266,38 @@ async def test_async_namespace_connection_materialized_views(tmp_path):
             handle._route_pushdown_to_rust == through_namespace._route_pushdown_to_rust
         )
         assert handle._namespace_path == through_namespace._namespace_path
+
+
+def test_namespaced_select_kind_is_read_and_unknown_kinds_are_refused():
+    import json
+
+    import pyarrow as pa
+
+    from lancedb.materialized_view import _definition_from_schema
+
+    def schema_with(definition: dict) -> pa.Schema:
+        return pa.schema([pa.field("id", pa.int32())]).with_metadata(
+            {b"mv.definition": json.dumps(definition).encode()}
+        )
+
+    # "namespaced_select" is the namespaced form of "select": same shape,
+    # a separate kind so readers that predate it refuse instead of
+    # resolving the source at the root.
+    definition = _definition_from_schema(
+        schema_with(
+            {
+                "kind": "namespaced_select",
+                "source_table": "people",
+                "source_namespace": ["ns"],
+                "projections": [{"output": "name", "expression": "name"}],
+            }
+        ),
+        "v",
+    )
+    assert definition.source_table == "people"
+    assert definition.source_namespace == ["ns"]
+
+    with pytest.raises(NotImplementedError, match="cannot refresh"):
+        _definition_from_schema(
+            schema_with({"kind": "select_v3", "source_table": "people"}), "v"
+        )

python/python/tests/test_query.py

@@ -1923,6 +1923,21 @@ def test_take_queries(tmp_path):
         17,
     ]
 
+    # Duplicate offsets are occurrences, not set members. Ordering is unspecified.
+    assert sorted(table.take_offsets([5, 2, 5, 17]).to_pandas()["idx"].to_list()) == [
+        2,
+        5,
+        5,
+        17,
+    ]
+
+    # Converting a take builder to its serializable query representation must
+    # retain occurrence metadata and execute with the same multiplicity.
+    query = table.take_offsets([5, 2, 5, 17]).select(["idx"]).to_query_object()
+    assert query.take_offsets == [5, 2, 5, 17]
+    converted = table._execute_query(query).read_all()
+    assert sorted(converted["idx"].to_pylist()) == [2, 5, 5, 17]
+
     # Take by row id
     assert list(
         sorted(table.take_row_ids([5, 2, 17]).to_pandas()["idx"].to_list())

python/python/tests/test_remote_db.py

@@ -479,24 +479,49 @@ def test_remote_permutation_is_picklable():
                 match = re.search(
                     r"_rowoffset\s+in\s+\((.*?)\)", body["filter"], re.IGNORECASE
                 )
-                offsets = [int(o.strip()) for o in match.group(1).split(",")]
+                offsets = list(
+                    dict.fromkeys(int(o.strip()) for o in match.group(1).split(","))
+                )
             else:
                 offsets = list(range(len(rows)))
-            table = pa.table({"a": [rows[offset] for offset in offsets]})
+            columns = body.get("columns") or ["a"]
+            table = pa.table(
+                {
+                    column: (
+                        [rows[offset] for offset in offsets]
+                        if column == "a"
+                        else offsets
+                    )
+                    for column in columns
+                }
+            )
 
             request.send_response(200)
             request.send_header("Content-Type", "application/vnd.apache.arrow.file")
             request.end_headers()
             with pa.ipc.new_file(request.wfile, schema=table.schema) as writer:
-                writer.write_table(table)
+                writer.write_table(table, max_chunksize=2)
         else:
             request.send_response(404)
             request.end_headers()
 
     with mock_lancedb_connection(handler) as db:
-        permutation = Permutation.identity(db.open_table("test"))
+        table = db.open_table("test")
+        assert table.take_offsets([0, 2, 0, 4]).to_list() == [
+            {"a": 0},
+            {"a": 0},
+            {"a": 2},
+            {"a": 4},
+        ]
+
+        permutation = Permutation.identity(table)
         restored = pickle.loads(pickle.dumps(permutation))
-        assert restored.__getitems__([0, 2, 4]) == [{"a": 0}, {"a": 2}, {"a": 4}]
+        assert restored.__getitems__([0, 2, 0, 4]) == [
+            {"a": 0},
+            {"a": 2},
+            {"a": 0},
+            {"a": 4},
+        ]
 
 
 def test_create_table_exist_ok():
@@ -795,11 +820,13 @@ def test_table_create_indices():
         scalar_req = received_requests[0]
         assert "name" in scalar_req
         assert scalar_req["name"] == "custom_scalar_idx"
+        assert scalar_req["replace"] is False
 
         # Check FTS index request has custom name
         fts_req = received_requests[1]
         assert "name" in fts_req
         assert fts_req["name"] == "custom_fts_idx"
+        assert fts_req["replace"] is False
         assert fts_req["block_size"] == 256
         assert fts_req["custom_stop_words"] == ["cloud"]
 
@@ -807,6 +834,7 @@ def test_table_create_indices():
         vector_req = received_requests[2]
         assert "name" in vector_req
         assert vector_req["name"] == "custom_vector_idx"
+        assert "replace" not in vector_req
 
         table.wait_for_index(["custom_scalar_idx"], timedelta(seconds=2))
         table.wait_for_index(
@@ -1079,6 +1107,9 @@ def test_remote_create_index_new_api():
             table.create_index("text", config=FTS(block_size=256))
             # IvfRq via new API
             table.create_index("vector", config=IvfRq(distance_type="l2"))
+            table.create_index(
+                "vector", config=IvfPq(distance_type="l2"), replace=False
+            )
 
         # Legacy index_type="IVF_RQ" routes to IvfRq config under the hood.
         with pytest.warns(DeprecationWarning, match="create_index"):
@@ -1088,15 +1119,17 @@ def test_remote_create_index_new_api():
                 num_partitions=8,
             )
 
-        assert len(received_requests) == 5
+        assert len(received_requests) == 6
         assert [req["column"] for req in received_requests] == [
             "vector",
             "category",
             "text",
             "vector",
             "vector",
+            "vector",
         ]
         assert received_requests[2]["block_size"] == 256
+        assert received_requests[4]["replace"] is False
 
 
 def test_table_wait_for_index_timeout():
@@ -2434,7 +2467,7 @@ def test_remote_blob_byte_apis_not_supported_on_old_server():
 
 
 def test_remote_connection_jobs_surface():
-    from lancedb.exceptions import JobFailedError
+    from lancedb.exceptions import JobFailedError, JobNotFoundError
 
     schema = pa.schema([("state", pa.string())])
     batch = pa.record_batch([pa.array(["created", "done"])], schema=schema)
@@ -2442,6 +2475,7 @@ def test_remote_connection_jobs_surface():
     with pa.ipc.new_stream(sink, schema) as writer:
         writer.write_batch(batch)
     events_body = sink.getvalue().to_pybytes()
+    query_events_payloads = []
 
     def handler(request):
         content_len = int(request.headers.get("Content-Length", 0))
@@ -2479,6 +2513,22 @@ def test_remote_connection_jobs_surface():
             request.end_headers()
             request.wfile.write(json.dumps(rsp).encode())
         elif request.path == "/v1/jobs/describe":
+            if payload["job_id"] == "job-2":
+                request.send_response(200)
+                request.send_header("Content-Type", "application/json")
+                request.end_headers()
+                request.wfile.write(
+                    json.dumps(
+                        dict(
+                            job_id="job-2",
+                            job_type="refresh_column",
+                            job_state="DONE",
+                            creation_ms=2000,
+                            result=dict(rows_assigned=1000000, rows_failed=0),
+                        )
+                    ).encode()
+                )
+                return
             if payload["job_id"] != "job-1":
                 request.send_response(404)
                 request.end_headers()
@@ -2510,7 +2560,7 @@ def test_remote_connection_jobs_surface():
             request.end_headers()
             request.wfile.write(b'{"job_id": "job-1"}')
         elif request.path == "/v1/jobs/query_events":
-            assert payload["job_id"] == "job-1"
+            query_events_payloads.append(payload)
             request.send_response(200)
             request.send_header("Content-Type", "application/vnd.apache.arrow.stream")
             request.end_headers()
@@ -2526,24 +2576,109 @@ def test_remote_connection_jobs_surface():
         assert jobs[0].table == "t1"
         assert jobs[1].state == "finished"
 
-        description = db.get_job("job-1")
-        assert description.job_type == "create_index"
-        assert description.state == "failed"
-        assert json.loads(description.spec_json) == {"column": "vec"}
-        assert description.failure.message == "worker died"
-        assert description.failure.retryable is True
-        assert db.get_job("missing") is None
-
         assert db.cancel_job("job-1") is True
         assert db.cancel_job("missing") is False
 
-        batches = db.job_history("job-1")
+        # Opening a job hands back a populated handle; a missing one fails.
-        assert len(batches) == 1
+        with pytest.raises(JobNotFoundError, match="missing"):
-        assert batches[0].num_rows == 2
+            db.open_job("missing")
-        assert batches[0].column("state").to_pylist() == ["created", "done"]
+        finished = db.open_job("job-2")
+        assert finished.state == "finished"
+        assert finished.result == {"rows_assigned": 1000000, "rows_failed": 0}
 
-        job = db.job("job-1")
+        job = db.open_job("job-1")
         assert job.id == "job-1"
+        # Opening already populated the handle.
+        assert job.state == "failed"
+        assert job.spec == {"column": "vec"}
+        assert job.failure.message == "worker died"
         assert job.status() == "failed"
         with pytest.raises(JobFailedError, match="worker died"):
             job.wait(timeout=timedelta(seconds=5))
+
+
+def test_remote_job_handle_reports_its_own_detail():
+    schema = pa.schema([("state", pa.string())])
+    batch = pa.record_batch([pa.array(["claim_complete"])], schema=schema)
+    sink = pa.BufferOutputStream()
+    with pa.ipc.new_stream(sink, schema) as writer:
+        writer.write_batch(batch)
+    events_body = sink.getvalue().to_pybytes()
+    event_payloads = []
+
+    def handler(request):
+        content_len = int(request.headers.get("Content-Length", 0))
+        body = request.rfile.read(content_len) if content_len > 0 else b""
+        payload = json.loads(body) if body else {}
+        if request.path == "/v1/jobs/describe":
+            request.send_response(200)
+            request.send_header("Content-Type", "application/json")
+            request.end_headers()
+            request.wfile.write(
+                json.dumps(
+                    dict(
+                        job_id="job-1",
+                        job_type="refresh_column",
+                        job_state="DONE",
+                        creation_ms=2000,
+                        spec=dict(column="vec"),
+                        result=dict(rows_assigned=1000000),
+                    )
+                ).encode()
+            )
+        elif request.path == "/v1/jobs/query_events":
+            event_payloads.append(payload)
+            request.send_response(200)
+            request.send_header("Content-Type", "application/vnd.apache.arrow.stream")
+            request.end_headers()
+            request.wfile.write(events_body)
+        else:
+            request.send_response(404)
+            request.end_headers()
+
+    with mock_lancedb_connection(handler) as db:
+        job = db.open_job("job-1")
+
+        # Opening populates the handle in the same round trip.
+        assert job.state == "finished"
+        job.refresh()
+        assert job.job_type == "refresh_column"
+        assert job.creation_ms == 2000
+        assert job.spec == {"column": "vec"}
+        assert job.result == {"rows_assigned": 1000000}
+        assert job.failure is None
+        # The JSON payloads stay reachable, but as internal APIs.
+        assert json.loads(job._spec_json) == {"column": "vec"}
+        assert json.loads(job._result_json) == {"rows_assigned": 1000000}
+
+        # print() shows everything the handle knows and nothing it does not.
+        # print() lays every known field out on its own line, with the JSON
+        # payloads indented rather than crammed onto one line.
+        assert repr(job) == "\n".join(
+            [
+                "Job(",
+                "    id='job-1',",
+                "    state='finished',",
+                "    job_type='refresh_column',",
+                "    creation_ms=2000,",
+                "    spec={",
+                '        "column": "vec"',
+                "    },",
+                "    result={",
+                '        "rows_assigned": 1000000',
+                "    },",
+                ")",
+            ]
+        )
+        # Nothing it does not know shows up.
+        assert "failure" not in repr(job)
+
+        events = job.events(filter="state = 'claim_complete'", limit=500)
+        assert isinstance(events, pa.Table)
+        assert events.column("state").to_pylist() == ["claim_complete"]
+        # The handle supplies job_id; the caller only narrows the query.
+        assert event_payloads[-1] == {
+            "job_id": "job-1",
+            "limit": 500,
+            "filter": "state = 'claim_complete'",
+        }

python/python/tests/test_sql.py

@@ -0,0 +1,162 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright The LanceDB Authors
+
+from uuid import UUID
+
+import pytest
+import pyarrow as pa
+
+import lancedb
+from lancedb import _lancedb
+from lancedb.arrow import AsyncRecordBatchReader
+from lancedb.db import AsyncConnection
+from lancedb.remote.db import RemoteDBConnection
+from lancedb.sql import AsyncQuery, Query
+
+NIL_QUERY_ID = UUID(int=0)
+
+
+class FakeNativeQuery:
+    id = UUID("0198f1b2-c3d4-7e5f-8123-456789abcdef")
+
+    async def reader(self):
+        return pa.table({"value": [1, 2]})
+
+
+class FakeNativeConnection:
+    async def execute_query_async(self, query, *, default_namespace_path=None):
+        return FakeNativeQuery()
+
+
+class FakeAsyncConnection:
+    async def execute_query_async(self, query, *, default_namespace_path=None):
+        return AsyncQuery(FakeNativeQuery())
+
+
+def remote_connection(sql_host_override=None):
+    return lancedb.connect(
+        "db://analytics",
+        api_key="test-key",
+        host_override="http://localhost:10024",
+        sql_host_override=sql_host_override,
+    )
+
+
+def test_sql_is_connection_scoped():
+    assert hasattr(lancedb, "sql")
+    assert not callable(lancedb.sql)
+    assert not hasattr(_lancedb, "sql")
+    assert not hasattr(remote_connection(), "sql")
+    assert hasattr(remote_connection(), "execute_query")
+    assert hasattr(remote_connection(), "execute_query_async")
+    assert hasattr(remote_connection(), "describe_query")
+
+
+def test_query_id_is_uuid():
+    query = AsyncQuery(FakeNativeQuery())
+    assert isinstance(query.id, UUID)
+    assert Query(query).id == query.id
+
+
+def test_connection_serializes_sql_host_override():
+    endpoint = "grpc+tls://sql.example.com:10026"
+    restored = lancedb.deserialize_conn(
+        remote_connection(sql_host_override=endpoint).serialize()
+    )
+    assert restored.sql_host_override == endpoint
+
+
+@pytest.mark.asyncio
+async def test_async_sql_reader_is_record_batch_stream():
+    reader = await AsyncQuery(FakeNativeQuery()).reader()
+    assert isinstance(reader, AsyncRecordBatchReader)
+    assert (await reader.read_all())[0].column(0).to_pylist() == [1, 2]
+
+
+def test_sync_sql_reader_is_record_batch_reader():
+    reader = Query(AsyncQuery(FakeNativeQuery())).reader()
+    assert isinstance(reader, pa.RecordBatchReader)
+    assert reader.read_all().column(0).to_pylist() == [1, 2]
+
+
+def test_execute_query_returns_blocking_reader():
+    connection = RemoteDBConnection.__new__(RemoteDBConnection)
+    connection._conn = FakeAsyncConnection()
+    reader = connection.execute_query("SELECT 1")
+    assert isinstance(reader, pa.RecordBatchReader)
+    assert reader.read_all().column(0).to_pylist() == [1, 2]
+
+
+@pytest.mark.asyncio
+async def test_async_execute_query_returns_async_reader():
+    connection = AsyncConnection(FakeNativeConnection())
+    reader = await connection.execute_query("SELECT 1")
+    assert isinstance(reader, AsyncRecordBatchReader)
+    assert (await reader.read_all())[0].column(0).to_pylist() == [1, 2]
+
+
+def test_local_connection_rejects_sql(tmp_path):
+    connection = lancedb.connect(tmp_path)
+    with pytest.raises(NotImplementedError, match="SQL"):
+        connection.execute_query("SELECT 1")
+    with pytest.raises(NotImplementedError, match="SQL"):
+        connection.execute_query_async("SELECT 1")
+    with pytest.raises(NotImplementedError, match="SQL"):
+        connection.describe_query(NIL_QUERY_ID)
+
+
+@pytest.mark.asyncio
+async def test_local_async_connection_rejects_sql(tmp_path):
+    connection = await lancedb.connect_async(tmp_path)
+    with pytest.raises(NotImplementedError, match="SQL"):
+        await connection.execute_query("SELECT 1")
+    with pytest.raises(NotImplementedError, match="SQL"):
+        await connection.execute_query_async("SELECT 1")
+    with pytest.raises(NotImplementedError, match="SQL"):
+        await connection.describe_query(NIL_QUERY_ID)
+
+
+@pytest.mark.asyncio
+async def test_async_namespace_connection_rejects_sql(tmp_path):
+    connection = lancedb.connect_namespace_async("dir", {"root": str(tmp_path)})
+    with pytest.raises(NotImplementedError, match="SQL"):
+        await connection.execute_query("SELECT 1")
+    with pytest.raises(NotImplementedError, match="SQL"):
+        await connection.execute_query_async("SELECT 1")
+    with pytest.raises(NotImplementedError, match="SQL"):
+        await connection.describe_query(NIL_QUERY_ID)
+
+
+def test_describe_query_requires_uuid():
+    with pytest.raises(TypeError, match="UUID"):
+        remote_connection().describe_query(str(NIL_QUERY_ID))
+
+
+@pytest.mark.parametrize(
+    "default_namespace_path",
+    ["public", ("public",), [1]],
+)
+def test_execute_query_async_requires_namespace_path_list(default_namespace_path):
+    with pytest.raises(ValueError, match="default_namespace_path"):
+        remote_connection().execute_query_async(
+            "SELECT 1", default_namespace_path=default_namespace_path
+        )
+
+
+def test_execute_query_async_rejects_invalid_endpoint():
+    connection = remote_connection(sql_host_override="invalid://localhost")
+    with pytest.raises(ValueError, match="sql_host_override"):
+        connection.execute_query_async("SELECT 1")
+
+
+@pytest.mark.parametrize(
+    "default_namespace_path",
+    [[""], ["café"], ["pub\tlic"], ["events$raw"]],
+)
+def test_execute_query_async_rejects_invalid_namespace_components(
+    default_namespace_path,
+):
+    with pytest.raises(ValueError, match="default_namespace_path"):
+        remote_connection().execute_query_async(
+            "SELECT 1", default_namespace_path=default_namespace_path
+        )

python/python/tests/test_table.py

@@ -2682,6 +2682,43 @@ def test_merge_insert(mem_db: DBConnection):
         )
 
 
+def test_merge_insert_composite_key(mem_db: DBConnection):
+    table = mem_db.create_table(
+        "my_table",
+        data=pa.table(
+            {
+                "shard": ["a", "a", "b"],
+                "id": [1, 2, 1],
+                "val": ["x", "y", "z"],
+            }
+        ),
+    )
+
+    # ("a", 1) matches an existing row and updates it. ("b", 2) agrees with an
+    # existing row on each key column separately but on neither pair, so it is
+    # an insert.
+    new_data = pa.table({"shard": ["a", "b"], "id": [1, 2], "val": ["X", "W"]})
+    res = (
+        table.merge_insert(["shard", "id"])
+        .when_matched_update_all()
+        .when_not_matched_insert_all()
+        .execute(new_data)
+    )
+    assert res.num_updated_rows == 1
+    assert res.num_inserted_rows == 1
+
+    expected = pa.table(
+        {
+            "shard": ["a", "a", "b", "b"],
+            "id": [1, 2, 1, 2],
+            "val": ["X", "y", "z", "W"],
+        }
+    )
+    assert table.to_arrow().sort_by([("shard", "ascending"), ("id", "ascending")]) == (
+        expected
+    )
+
+
 def test_merge_insert_nullable_pandas_into_pydantic_schema(mem_db: DBConnection):
     # Regression test for https://github.com/lancedb/lancedb/issues/2366
     pd = pytest.importorskip("pandas")
@@ -4087,6 +4124,29 @@ def test_computed_column_rejects_transforms_and_computed_together(tmp_path):
         table.add_columns({"a": "x + 1"}, computed={"b": "x * 2"})
 
 
+def test_computed_column_blob_projection_inherits_semantics(tmp_path):
+    schema = pa.schema([pa.field("id", pa.int64()), lancedb.blob("image")])
+    db = lancedb.connect(tmp_path)
+    table = db.create_table("computed_column_blob", schema=schema)
+    table.add(
+        [
+            {"id": 1, "image": b"hello"},
+            {"id": 2, "image": b""},
+            {"id": 3, "image": None},
+        ]
+    )
+
+    table.add_columns(computed={"image_copy": "image", "second_copy": "image_copy"})
+    assert table.refresh_column("image_copy").rows_filled == 2
+    assert table.refresh_column("second_copy").rows_filled == 2
+    assert table.blob_columns() == ["image", "image_copy", "second_copy"]
+
+    hits = table.search().with_row_id(True).limit(10).to_arrow()
+    rows = sorted(zip(hits["id"].to_pylist(), hits["_rowid"].to_pylist()))
+    copied = table.fetch_blobs("second_copy", [row_id for _, row_id in rows])
+    assert copied.to_pylist() == [b"hello", b"", None]
+
+
 @pytest.mark.asyncio
 async def test_computed_column_async(tmp_path):
     db = await lancedb.connect_async(tmp_path)

python/src/connection.rs

@@ -13,11 +13,7 @@ use crate::{
     runtime::future_into_py,
     table::Table,
 };
-use arrow::{
+use arrow::{datatypes::Schema, ffi_stream::ArrowArrayStreamReader, pyarrow::FromPyArrow};
-    datatypes::Schema,
-    ffi_stream::ArrowArrayStreamReader,
-    pyarrow::{FromPyArrow, ToPyArrow},
-};
 use lancedb::{
     connection::Connection as LanceConnection,
     connection::NamespaceClientPushdownOperation,
@@ -28,7 +24,7 @@ use pyo3::{
     Bound, FromPyObject, Py, PyAny, PyRef, PyResult, Python,
     exceptions::{PyRuntimeError, PyValueError},
     pyclass, pyfunction, pymethods,
-    types::{PyDict, PyDictMethods, PyList, PyListMethods},
+    types::{PyAnyMethods, PyDict, PyDictMethods, PyList},
 };
 
 #[pyclass]
@@ -86,6 +82,24 @@ impl Connection {
     }
 }
 
+fn parse_default_namespace_path(path: Option<Bound<'_, PyAny>>) -> PyResult<Vec<String>> {
+    match path {
+        Some(path) => {
+            if !path.is_instance_of::<PyList>() {
+                return Err(PyValueError::new_err(
+                    "Connection.execute_query_async default_namespace_path must be a list",
+                ));
+            }
+            path.extract::<Vec<String>>().map_err(|_| {
+                PyValueError::new_err(
+                    "Connection.execute_query_async default_namespace_path components must be strings",
+                )
+            })
+        }
+        None => Ok(vec!["public".to_string()]),
+    }
+}
+
 #[pymethods]
 impl Connection {
     fn __repr__(&self) -> String {
@@ -108,6 +122,40 @@ impl Connection {
         self.get_inner().map(|inner| inner.uri().to_string())
     }
 
+    #[pyo3(signature = (query, *, default_namespace_path=None))]
+    pub fn execute_query_async<'a>(
+        self_: PyRef<'a, Self>,
+        query: String,
+        default_namespace_path: Option<Bound<'_, PyAny>>,
+    ) -> PyResult<Bound<'a, PyAny>> {
+        let inner = self_.get_inner()?.clone();
+        let default_namespace_path = parse_default_namespace_path(default_namespace_path)?;
+        future_into_py(self_.py(), async move {
+            let operation = inner
+                .execute_query_async(query)
+                .default_namespace_path(default_namespace_path);
+            operation
+                .execute()
+                .await
+                .map(crate::sql::Query::new)
+                .infer_error()
+        })
+    }
+
+    pub fn describe_query<'a>(
+        self_: PyRef<'a, Self>,
+        query_id: uuid::Uuid,
+    ) -> PyResult<Bound<'a, PyAny>> {
+        let inner = self_.get_inner()?.clone();
+        future_into_py(self_.py(), async move {
+            inner
+                .describe_query(query_id)
+                .await
+                .map(crate::sql::QueryDescription::from)
+                .infer_error()
+        })
+    }
+
     #[pyo3(signature = ())]
     pub fn get_read_consistency_interval(self_: PyRef<'_, Self>) -> PyResult<Bound<'_, PyAny>> {
         let inner = self_.get_inner()?.clone();
@@ -592,9 +640,12 @@ impl Connection {
         })
     }
 
-    pub fn job(&self, job_id: String) -> PyResult<crate::job::Job> {
+    pub fn open_job(self_: PyRef<'_, Self>, job_id: String) -> PyResult<Bound<'_, PyAny>> {
-        let inner = self.get_inner()?.clone();
+        let inner = self_.get_inner()?.clone();
-        Ok(crate::job::Job::new(inner.job(job_id).infer_error()?))
+        future_into_py(self_.py(), async move {
+            let job = inner.open_job(&job_id).await.infer_error()?;
+            Ok(crate::job::Job::new(job))
+        })
     }
 
     pub fn create_function_async(
@@ -629,6 +680,77 @@ impl Connection {
         })
     }
 
+    pub fn list_functions(self_: PyRef<'_, Self>) -> PyResult<Bound<'_, PyAny>> {
+        let inner = self_.get_inner()?.clone();
+        future_into_py(self_.py(), async move {
+            inner
+                .list_functions()
+                .await
+                .infer_error()?
+                .into_iter()
+                .map(|function| function.to_canonical_json().infer_error())
+                .collect::<PyResult<Vec<_>>>()
+        })
+    }
+
+    pub fn drop_function(
+        self_: PyRef<'_, Self>,
+        name: String,
+        version: String,
+    ) -> PyResult<Bound<'_, PyAny>> {
+        let inner = self_.get_inner()?.clone();
+        future_into_py(self_.py(), async move {
+            inner.drop_function(name, version).await.infer_error()
+        })
+    }
+
+    pub fn create_secret(
+        self_: PyRef<'_, Self>,
+        name: String,
+        value: String,
+    ) -> PyResult<Bound<'_, PyAny>> {
+        let inner = self_.get_inner()?.clone();
+        future_into_py(self_.py(), async move {
+            inner.create_secret(name, value).await.infer_error()
+        })
+    }
+
+    pub fn alter_secret(
+        self_: PyRef<'_, Self>,
+        name: String,
+        value: String,
+    ) -> PyResult<Bound<'_, PyAny>> {
+        let inner = self_.get_inner()?.clone();
+        future_into_py(self_.py(), async move {
+            inner.alter_secret(name, value).await.infer_error()
+        })
+    }
+
+    pub fn list_secrets(self_: PyRef<'_, Self>) -> PyResult<Bound<'_, PyAny>> {
+        let inner = self_.get_inner()?.clone();
+        future_into_py(self_.py(), async move {
+            inner.list_secrets().await.infer_error()
+        })
+    }
+
+    pub fn drop_secret(self_: PyRef<'_, Self>, name: String) -> PyResult<Bound<'_, PyAny>> {
+        let inner = self_.get_inner()?.clone();
+        future_into_py(self_.py(), async move {
+            inner.drop_secret(name).await.infer_error()
+        })
+    }
+
+    /// Name and timestamps as a plain tuple. `SecretInfo` carries no value, so
+    /// there is none to filter out here. Timestamps stay integers rather than
+    /// going through a string, so the caller can compare two without parsing.
+    pub fn describe_secret(self_: PyRef<'_, Self>, name: String) -> PyResult<Bound<'_, PyAny>> {
+        let inner = self_.get_inner()?.clone();
+        future_into_py(self_.py(), async move {
+            let info = inner.describe_secret(name).await.infer_error()?;
+            Ok((info.name, info.created_at_millis, info.updated_at_millis))
+        })
+    }
+
     pub fn list_jobs(self_: PyRef<'_, Self>) -> PyResult<Bound<'_, PyAny>> {
         let inner = self_.get_inner()?.clone();
         future_into_py(self_.py(), async move {
@@ -640,42 +762,16 @@ impl Connection {
         })
     }
 
-    pub fn get_job(self_: PyRef<'_, Self>, job_id: String) -> PyResult<Bound<'_, PyAny>> {
-        let inner = self_.get_inner()?.clone();
-        future_into_py(self_.py(), async move {
-            let description = inner.get_job(&job_id).await.infer_error()?;
-            Ok(description.map(crate::job::JobDescription::from))
-        })
-    }
-
     pub fn cancel_job(self_: PyRef<'_, Self>, job_id: String) -> PyResult<Bound<'_, PyAny>> {
         let inner = self_.get_inner()?.clone();
         future_into_py(self_.py(), async move {
             inner.cancel_job(&job_id).await.infer_error()
         })
     }
-
-    #[pyo3(signature = (job_id=None))]
-    pub fn job_history(
-        self_: PyRef<'_, Self>,
-        job_id: Option<String>,
-    ) -> PyResult<Bound<'_, PyAny>> {
-        let inner = self_.get_inner()?.clone();
-        future_into_py(self_.py(), async move {
-            let batches = inner.job_history(job_id.as_deref()).await.infer_error()?;
-            Python::attach(|py| {
-                let list = PyList::empty(py);
-                for batch in batches {
-                    list.append(batch.to_pyarrow(py)?)?;
-                }
-                Ok(list.unbind())
-            })
-        })
-    }
 }
 
 #[pyfunction]
-#[pyo3(signature = (uri, api_key=None, region=None, host_override=None, read_consistency_interval=None, client_config=None, storage_options=None, session=None, manifest_enabled=false, namespace_client_properties=None, oauth_config=None))]
+#[pyo3(signature = (uri, api_key=None, region=None, host_override=None, sql_host_override=None, read_consistency_interval=None, client_config=None, storage_options=None, session=None, manifest_enabled=false, namespace_client_properties=None, oauth_config=None))]
 #[allow(clippy::too_many_arguments)]
 pub fn connect(
     py: Python<'_>,
@@ -683,6 +779,7 @@ pub fn connect(
     api_key: Option<String>,
     region: Option<String>,
     host_override: Option<String>,
+    sql_host_override: Option<String>,
     read_consistency_interval: Option<f64>,
     client_config: Option<PyClientConfig>,
     storage_options: Option<HashMap<String, String>>,
@@ -702,6 +799,12 @@ pub fn connect(
         if let Some(host_override) = host_override {
             builder = builder.host_override(&host_override);
         }
+        #[cfg(feature = "remote")]
+        if let Some(sql_host_override) = sql_host_override {
+            builder = builder.sql_host_override(&sql_host_override);
+        }
+        #[cfg(not(feature = "remote"))]
+        let _ = sql_host_override;
         if let Some(read_consistency_interval) = read_consistency_interval {
             let read_consistency_interval = Duration::from_secs_f64(read_consistency_interval);
             builder = builder.read_consistency_interval(read_consistency_interval);

python/src/error.rs

@@ -114,6 +114,12 @@ impl<T> PythonErrorExt<T> for std::result::Result<T, LanceError> {
                         .getattr(intern!(py, "JobCancelledError"))?;
                     Err(PyErr::from_value(cls.call1((err.to_string(),))?))
                 }),
+                LanceError::JobNotFound { .. } => Python::attach(|py| {
+                    let cls = py
+                        .import(intern!(py, "lancedb.exceptions"))?
+                        .getattr(intern!(py, "JobNotFoundError"))?;
+                    Err(PyErr::from_value(cls.call1((err.to_string(),))?))
+                }),
                 _ => self.runtime_error(),
             },
         }

python/src/job.rs

@@ -4,11 +4,50 @@
 use std::sync::Arc;
 
 use crate::runtime::future_into_py;
-use pyo3::{Bound, PyAny, PyRef, PyResult, pyclass, pymethods};
+use arrow::{
+    datatypes::Schema,
+    pyarrow::{IntoPyArrow, Table as PyArrowTable},
+};
+use lancedb::job::JobEventsRequest;
+use pyo3::{
+    Bound, PyAny, PyRef, PyResult, Python,
+    exceptions::PyValueError,
+    pyclass, pymethods,
+    types::{PyAnyMethods, PyDict, PyDictMethods},
+};
 use serde::Serialize;
 
 use crate::error::PythonErrorExt;
 
+const REPR_INDENT: &str = "    ";
+
+/// Parse a stored JSON payload into Python data. The bindings carry these as
+/// strings because that is what crosses the boundary cheaply; the public
+/// Python surface is the parsed form.
+fn parse_json_payload<'py>(
+    py: Python<'py>,
+    raw: Option<&str>,
+) -> PyResult<Option<Bound<'py, PyAny>>> {
+    match raw {
+        None => Ok(None),
+        Some(raw) => Ok(Some(py.import("json")?.call_method1("loads", (raw,))?)),
+    }
+}
+
+/// A payload rendered as indented JSON, aligned under the field that holds it.
+fn pretty_json_payload(py: Python<'_>, raw: Option<&str>) -> PyResult<Option<String>> {
+    let Some(parsed) = parse_json_payload(py, raw)? else {
+        return Ok(None);
+    };
+    let kwargs = PyDict::new(py);
+    kwargs.set_item("indent", 4)?;
+    let rendered: String = py
+        .import("json")?
+        .call_method("dumps", (parsed,), Some(&kwargs))?
+        .extract()?;
+    Ok(Some(rendered.replace('\n', &format!("\n{REPR_INDENT}"))))
+}
+
 #[pyclass]
 pub struct Job {
     inner: Arc<lancedb::Job<std::result::Result<Option<String>, String>>>,
@@ -67,6 +106,48 @@ impl Job {
             Ok(())
         })
     }
+
+    pub fn refresh(self_: PyRef<'_, Self>) -> PyResult<Bound<'_, PyAny>> {
+        let inner = self_.inner.clone();
+        future_into_py(self_.py(), async move {
+            inner.refresh().await.infer_error()?;
+            Ok(())
+        })
+    }
+
+    /// The last observed lifecycle state, without contacting the backend.
+    #[getter]
+    pub fn _state(&self) -> Option<String> {
+        self.inner.state()
+    }
+
+    /// The last observed server-side record. `None` for an in-process job.
+    #[getter]
+    pub fn _description(&self) -> Option<JobDescription> {
+        self.inner.description().map(JobDescription::from)
+    }
+
+    #[pyo3(signature = (*, limit=None, filter=None))]
+    pub fn events(
+        self_: PyRef<'_, Self>,
+        limit: Option<u32>,
+        filter: Option<String>,
+    ) -> PyResult<Bound<'_, PyAny>> {
+        let inner = self_.inner.clone();
+        let request = JobEventsRequest { limit, filter };
+        future_into_py(self_.py(), async move {
+            let batches = inner.events(request).await.infer_error()?;
+            Python::attach(|py| {
+                let schema = batches
+                    .first()
+                    .map(|batch| batch.schema())
+                    .unwrap_or_else(|| Arc::new(Schema::empty()));
+                let table = PyArrowTable::try_new(batches, schema)
+                    .map_err(|err| PyValueError::new_err(err.to_string()))?;
+                table.into_pyarrow(py).map(|table| table.unbind())
+            })
+        })
+    }
 }
 
 /// A row from `Connection.list_jobs`: one server-side job.
@@ -121,7 +202,7 @@ impl JobFailureInfo {
     }
 }
 
-/// A described job from `Connection.get_job`.
+/// The server-side record behind a `Job` handle.
 #[pyclass(get_all, skip_from_py_object)]
 #[derive(Clone)]
 pub struct JobDescription {
@@ -129,17 +210,49 @@ pub struct JobDescription {
     job_type: String,
     state: String,
     creation_ms: i64,
-    spec_json: Option<String>,
+    /// Internal: the wire form behind the `spec` property.
+    _spec_json: Option<String>,
+    /// Internal: the wire form behind the `result` property.
+    _result_json: Option<String>,
     failure: Option<JobFailureInfo>,
 }
 
 #[pymethods]
 impl JobDescription {
-    fn __repr__(&self) -> String {
+    /// The job-type-specific specification it was submitted with.
-        format!(
+    #[getter]
-            "JobDescription(job_id={:?}, job_type={:?}, state={:?}, creation_ms={})",
+    fn spec<'py>(&self, py: Python<'py>) -> PyResult<Option<Bound<'py, PyAny>>> {
-            self.job_id, self.job_type, self.state, self.creation_ms
+        parse_json_payload(py, self._spec_json.as_deref())
-        )
+    }
+
+    /// The job-type-specific terminal result. `None` until the job succeeds.
+    #[getter]
+    fn result<'py>(&self, py: Python<'py>) -> PyResult<Option<Bound<'py, PyAny>>> {
+        parse_json_payload(py, self._result_json.as_deref())
+    }
+
+    fn __repr__(&self, py: Python<'_>) -> PyResult<String> {
+        let mut fields = vec![
+            format!("job_id={:?}", self.job_id),
+            format!("job_type={:?}", self.job_type),
+            format!("state={:?}", self.state),
+            format!("creation_ms={}", self.creation_ms),
+        ];
+        // Lay the payloads out as indented JSON, the same way the `Job` repr
+        // does, so the two agree on how the same data looks.
+        for (name, payload) in [("spec", &self._spec_json), ("result", &self._result_json)] {
+            if let Some(rendered) = pretty_json_payload(py, payload.as_deref())? {
+                fields.push(format!("{name}={rendered}"));
+            }
+        }
+        if let Some(failure) = &self.failure {
+            fields.push(format!("failure={}", failure.__repr__()));
+        }
+        let body = fields
+            .iter()
+            .map(|field| format!("\n{REPR_INDENT}{field},"))
+            .collect::<String>();
+        Ok(format!("JobDescription({body}\n)"))
     }
 }
 
@@ -150,7 +263,11 @@ impl From<lancedb::database::JobDescription> for JobDescription {
             job_type: description.job_type,
             state: description.state,
             creation_ms: description.creation_ms,
-            spec_json: (!description.spec.is_null()).then(|| description.spec.to_string()),
+            _spec_json: (!description.spec.is_null()).then(|| description.spec.to_string()),
+            _result_json: description
+                .result
+                .filter(|result| !result.is_null())
+                .map(|result| result.to_string()),
             failure: description.failure.map(|failure| JobFailureInfo {
                 phase: failure.phase,
                 message: failure.message,

python/src/lib.rs

@@ -34,6 +34,7 @@ pub mod permutation;
 pub mod query;
 pub mod runtime;
 pub mod session;
+pub mod sql;
 pub mod table;
 pub mod util;
 
@@ -50,6 +51,8 @@ pub fn _lancedb(_py: Python, m: &Bound<'_, PyModule>) -> PyResult<()> {
     m.add_class::<crate::job::JobInfo>()?;
     m.add_class::<crate::job::JobDescription>()?;
     m.add_class::<crate::job::JobFailureInfo>()?;
+    m.add_class::<crate::sql::Query>()?;
+    m.add_class::<crate::sql::QueryDescription>()?;
     m.add_class::<PyBlobFile>()?;
     m.add_class::<IndexConfig>()?;
     m.add_class::<Query>()?;

python/src/query.rs

@@ -323,6 +323,7 @@ impl<'py> IntoPyObject<'py> for PyQueryVectors {
 pub struct PyQueryRequest {
     pub limit: Option<usize>,
     pub offset: Option<usize>,
+    pub take_offsets: Option<Vec<u64>>,
     pub filter: Option<PyQueryFilter>,
     pub full_text_search: Option<PyLanceDB<FtsQuery>>,
     pub select: PySelect,
@@ -353,6 +354,7 @@ impl From<AnyQuery> for PyQueryRequest {
             AnyQuery::Query(query_request) => Self {
                 limit: query_request.limit,
                 offset: query_request.offset,
+                take_offsets: query_request.take_offsets,
                 filter: query_request.filter.map(PyQueryFilter),
                 full_text_search: query_request
                     .full_text_search
@@ -381,6 +383,7 @@ impl From<AnyQuery> for PyQueryRequest {
             AnyQuery::VectorQuery(vector_query) => Self {
                 limit: vector_query.base.limit,
                 offset: vector_query.base.offset,
+                take_offsets: vector_query.base.take_offsets,
                 filter: vector_query.base.filter.map(PyQueryFilter),
                 full_text_search: None,
                 select_source_columns: PySelect::source_columns(&vector_query.base.select),

python/src/sql.rs

@@ -0,0 +1,90 @@
+// SPDX-License-Identifier: Apache-2.0
+// SPDX-FileCopyrightText: Copyright The LanceDB Authors
+
+use std::sync::Arc;
+
+use chrono::{DateTime, Utc};
+use pyo3::{Bound, PyAny, PyRef, PyResult, pyclass, pymethods};
+use uuid::Uuid;
+
+use crate::arrow::RecordBatchStream;
+use crate::error::PythonErrorExt;
+use crate::runtime::future_into_py;
+
+#[pyclass(name = "SqlQuery")]
+pub struct Query {
+    inner: Arc<lancedb::sql::Query>,
+}
+
+impl Query {
+    pub(crate) fn new(inner: lancedb::sql::Query) -> Self {
+        Self {
+            inner: Arc::new(inner),
+        }
+    }
+}
+
+#[pymethods]
+impl Query {
+    #[getter]
+    pub fn id(&self) -> Uuid {
+        self.inner.id()
+    }
+
+    pub fn describe(self_: PyRef<'_, Self>) -> PyResult<Bound<'_, PyAny>> {
+        let inner = self_.inner.clone();
+        future_into_py(self_.py(), async move {
+            inner
+                .describe()
+                .await
+                .map(QueryDescription::from)
+                .infer_error()
+        })
+    }
+
+    pub fn reader(self_: PyRef<'_, Self>) -> PyResult<Bound<'_, PyAny>> {
+        let inner = self_.inner.clone();
+        future_into_py(self_.py(), async move {
+            let stream = inner.reader().await.infer_error()?;
+            Ok(RecordBatchStream::new(stream))
+        })
+    }
+
+    pub fn cancel(self_: PyRef<'_, Self>) -> PyResult<Bound<'_, PyAny>> {
+        let inner = self_.inner.clone();
+        future_into_py(self_.py(), async move {
+            inner.cancel().await.infer_error()?;
+            Ok(())
+        })
+    }
+}
+
+#[pyclass(get_all, skip_from_py_object)]
+#[derive(Clone)]
+pub struct QueryDescription {
+    id: Uuid,
+    status: String,
+    progress: Option<f64>,
+    expires_at: Option<DateTime<Utc>>,
+}
+
+#[pymethods]
+impl QueryDescription {
+    fn __repr__(&self) -> String {
+        format!(
+            "QueryDescription(id={:?}, status={:?}, progress={:?}, expires_at={:?})",
+            self.id, self.status, self.progress, self.expires_at
+        )
+    }
+}
+
+impl From<lancedb::sql::QueryDescription> for QueryDescription {
+    fn from(description: lancedb::sql::QueryDescription) -> Self {
+        Self {
+            id: description.id,
+            status: description.status.to_string(),
+            progress: description.progress,
+            expires_at: description.expires_at,
+        }
+    }
+}

python/uv.lock

@@ -10,6 +10,9 @@ resolution-markers = [
     "python_full_version < '3.11'",
 ]
 
+[options]
+prerelease-mode = "allow"
+
 [[package]]
 name = "accelerate"
 version = "1.14.0"
@@ -799,7 +802,7 @@ name = "cuda-bindings"
 version = "13.3.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "cuda-pathfinder", marker = "(python_full_version < '3.14' and sys_platform == 'emscripten') or (python_full_version < '3.14' and sys_platform == 'win32') or (sys_platform != 'emscripten' and sys_platform != 'win32')" },
+    { name = "cuda-pathfinder" },
 ]
 wheels = [
     { url = "https://files.pythonhosted.org/packages/a9/21/8464d133752951c154feafb3b65c297e7d80f301183d220bec4c830f1441/cuda_bindings-13.3.1-cp310-cp310-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:120fcc53d57903df529c3486962c56528cba5b7d6c57c99537320ed9922c8b86", size = 6073403, upload-time = "2026-05-29T23:11:36.22Z" },
@@ -834,37 +837,37 @@ wheels = [
 
 [package.optional-dependencies]
 cublas = [
-    { name = "nvidia-cublas", marker = "(python_full_version < '3.14' and sys_platform == 'win32') or sys_platform == 'linux'" },
+    { name = "nvidia-cublas" },
 ]
 cudart = [
-    { name = "nvidia-cuda-runtime", marker = "(python_full_version < '3.14' and sys_platform == 'win32') or sys_platform == 'linux'" },
+    { name = "nvidia-cuda-runtime" },
 ]
 cufft = [
-    { name = "nvidia-cufft", marker = "(python_full_version < '3.14' and sys_platform == 'win32') or sys_platform == 'linux'" },
+    { name = "nvidia-cufft" },
 ]
 cufile = [
-    { name = "nvidia-cufile", marker = "sys_platform == 'linux'" },
+    { name = "nvidia-cufile" },
 ]
 cupti = [
-    { name = "nvidia-cuda-cupti", marker = "(python_full_version < '3.14' and sys_platform == 'win32') or sys_platform == 'linux'" },
+    { name = "nvidia-cuda-cupti" },
 ]
 curand = [
-    { name = "nvidia-curand", marker = "(python_full_version < '3.14' and sys_platform == 'win32') or sys_platform == 'linux'" },
+    { name = "nvidia-curand" },
 ]
 cusolver = [
-    { name = "nvidia-cusolver", marker = "(python_full_version < '3.14' and sys_platform == 'win32') or sys_platform == 'linux'" },
+    { name = "nvidia-cusolver" },
 ]
 cusparse = [
-    { name = "nvidia-cusparse", marker = "(python_full_version < '3.14' and sys_platform == 'win32') or sys_platform == 'linux'" },
+    { name = "nvidia-cusparse" },
 ]
 nvjitlink = [
-    { name = "nvidia-nvjitlink", marker = "(python_full_version < '3.14' and sys_platform == 'win32') or sys_platform == 'linux'" },
+    { name = "nvidia-nvjitlink" },
 ]
 nvrtc = [
-    { name = "nvidia-cuda-nvrtc", marker = "(python_full_version < '3.14' and sys_platform == 'win32') or sys_platform == 'linux'" },
+    { name = "nvidia-cuda-nvrtc" },
 ]
 nvtx = [
-    { name = "nvidia-nvtx", marker = "(python_full_version < '3.14' and sys_platform == 'win32') or sys_platform == 'linux'" },
+    { name = "nvidia-nvtx" },
 ]
 
 [[package]]
@@ -1023,7 +1026,7 @@ name = "exceptiongroup"
 version = "1.3.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "typing-extensions", marker = "python_full_version < '3.11'" },
+    { name = "typing-extensions" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/50/79/66800aadf48771f6b62f7eb014e352e5d06856655206165d775e675a02c9/exceptiongroup-1.3.1.tar.gz", hash = "sha256:8b412432c6055b0b7d14c310000ae93352ed6754f70fa8f7c34141f91c4e3219", size = 30371, upload-time = "2025-11-21T23:01:54.787Z" }
 wheels = [
@@ -1440,16 +1443,16 @@ resolution-markers = [
     "python_full_version < '3.11'",
 ]
 dependencies = [
-    { name = "cachetools", marker = "python_full_version < '3.11'" },
+    { name = "cachetools" },
-    { name = "certifi", marker = "python_full_version < '3.11'" },
+    { name = "certifi" },
-    { name = "httpx", marker = "python_full_version < '3.11'" },
+    { name = "httpx" },
-    { name = "ibm-cos-sdk", marker = "python_full_version < '3.11'" },
+    { name = "ibm-cos-sdk" },
-    { name = "lomond", marker = "python_full_version < '3.11'" },
+    { name = "lomond" },
-    { name = "packaging", marker = "python_full_version < '3.11'" },
+    { name = "packaging" },
-    { name = "pandas", version = "2.2.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.11'" },
+    { name = "pandas", version = "2.2.3", source = { registry = "https://pypi.org/simple" } },
-    { name = "requests", marker = "python_full_version < '3.11'" },
+    { name = "requests" },
-    { name = "tabulate", marker = "python_full_version < '3.11'" },
+    { name = "tabulate" },
-    { name = "urllib3", marker = "python_full_version < '3.11'" },
+    { name = "urllib3" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/c7/56/2e3df38a1f13062095d7bde23c87a92f3898982993a15186b1bfecbd206f/ibm_watsonx_ai-1.3.42.tar.gz", hash = "sha256:ee5be59009004245d957ce97d1227355516df95a2640189749487614fef674ff", size = 688651, upload-time = "2025-10-01T13:35:41.527Z" }
 wheels = [
@@ -1468,17 +1471,17 @@ resolution-markers = [
     "python_full_version == '3.11.*'",
 ]
 dependencies = [
-    { name = "cachetools", marker = "python_full_version >= '3.11'" },
+    { name = "cachetools" },
-    { name = "certifi", marker = "python_full_version >= '3.11'" },
+    { name = "certifi" },
-    { name = "httpx", marker = "python_full_version >= '3.11'" },
+    { name = "httpx" },
-    { name = "ibm-cos-sdk", marker = "python_full_version >= '3.11'" },
+    { name = "ibm-cos-sdk" },
-    { name = "lomond", marker = "python_full_version >= '3.11'" },
+    { name = "lomond" },
-    { name = "packaging", marker = "python_full_version >= '3.11'" },
+    { name = "packaging" },
-    { name = "pandas", version = "2.3.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11' and python_full_version < '3.14'" },
+    { name = "pandas", version = "2.3.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.14'" },
     { name = "pandas", version = "3.0.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.14'" },
-    { name = "requests", marker = "python_full_version >= '3.11'" },
+    { name = "requests" },
-    { name = "tabulate", marker = "python_full_version >= '3.11'" },
+    { name = "tabulate" },
-    { name = "urllib3", marker = "python_full_version >= '3.11'" },
+    { name = "urllib3" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/29/a3/c756b534696ab2f3f29882fdb7ca7198b7a5c94e10c0a3a327853d6d6b79/ibm_watsonx_ai-1.5.14.tar.gz", hash = "sha256:a756488bd57e87c0fc51be42dcba871143cfe0ac1e805c497c5047e1e4f13e9d", size = 735804, upload-time = "2026-06-22T12:32:43.85Z" }
 wheels = [
@@ -1554,17 +1557,17 @@ resolution-markers = [
     "python_full_version < '3.11'",
 ]
 dependencies = [
-    { name = "colorama", marker = "python_full_version < '3.11' and sys_platform == 'win32'" },
+    { name = "colorama", marker = "sys_platform == 'win32'" },
-    { name = "decorator", marker = "python_full_version < '3.11'" },
+    { name = "decorator" },
-    { name = "exceptiongroup", marker = "python_full_version < '3.11'" },
+    { name = "exceptiongroup" },
-    { name = "jedi", marker = "python_full_version < '3.11'" },
+    { name = "jedi" },
-    { name = "matplotlib-inline", marker = "python_full_version < '3.11'" },
+    { name = "matplotlib-inline" },
-    { name = "pexpect", marker = "python_full_version < '3.11' and sys_platform != 'emscripten' and sys_platform != 'win32'" },
+    { name = "pexpect", marker = "sys_platform != 'emscripten' and sys_platform != 'win32'" },
-    { name = "prompt-toolkit", marker = "python_full_version < '3.11'" },
+    { name = "prompt-toolkit" },
-    { name = "pygments", marker = "python_full_version < '3.11'" },
+    { name = "pygments" },
-    { name = "stack-data", marker = "python_full_version < '3.11'" },
+    { name = "stack-data" },
-    { name = "traitlets", marker = "python_full_version < '3.11'" },
+    { name = "traitlets" },
-    { name = "typing-extensions", marker = "python_full_version < '3.11'" },
+    { name = "typing-extensions" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/40/18/f8598d287006885e7136451fdea0755af4ebcbfe342836f24deefaed1164/ipython-8.39.0.tar.gz", hash = "sha256:4110ae96012c379b8b6db898a07e186c40a2a1ef5d57a7fa83166047d9da7624", size = 5513971, upload-time = "2026-03-27T10:02:13.94Z" }
 wheels = [
@@ -1583,18 +1586,18 @@ resolution-markers = [
     "python_full_version == '3.11.*'",
 ]
 dependencies = [
-    { name = "colorama", marker = "python_full_version >= '3.11' and sys_platform == 'win32'" },
+    { name = "colorama", marker = "sys_platform == 'win32'" },
-    { name = "decorator", marker = "python_full_version >= '3.11'" },
+    { name = "decorator" },
-    { name = "ipython-pygments-lexers", marker = "python_full_version >= '3.11'" },
+    { name = "ipython-pygments-lexers" },
-    { name = "jedi", marker = "python_full_version >= '3.11'" },
+    { name = "jedi" },
-    { name = "matplotlib-inline", marker = "python_full_version >= '3.11'" },
+    { name = "matplotlib-inline" },
-    { name = "pexpect", marker = "python_full_version >= '3.11' and sys_platform != 'emscripten' and sys_platform != 'win32'" },
+    { name = "pexpect", marker = "sys_platform != 'emscripten' and sys_platform != 'win32'" },
-    { name = "prompt-toolkit", marker = "python_full_version >= '3.11'" },
+    { name = "prompt-toolkit" },
-    { name = "psutil", marker = "python_full_version >= '3.11' and sys_platform != 'cygwin' and sys_platform != 'emscripten'" },
+    { name = "psutil", marker = "sys_platform != 'cygwin' and sys_platform != 'emscripten'" },
-    { name = "pygments", marker = "python_full_version >= '3.11'" },
+    { name = "pygments" },
-    { name = "stack-data", marker = "python_full_version >= '3.11'" },
+    { name = "stack-data" },
-    { name = "traitlets", marker = "python_full_version >= '3.11'" },
+    { name = "traitlets" },
-    { name = "typing-extensions", marker = "python_full_version == '3.11.*'" },
+    { name = "typing-extensions", marker = "python_full_version < '3.12'" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/53/59/165d3b4d75cc34add3122c4417ecb229085140ac573103c223cd01dde96f/ipython-9.15.0.tar.gz", hash = "sha256:da2819ce2aa83135257df830660b1176d986c3d2876db24df01974fa955b2756", size = 4442580, upload-time = "2026-06-26T11:03:35.913Z" }
 wheels = [
@@ -1606,7 +1609,7 @@ name = "ipython-pygments-lexers"
 version = "1.1.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "pygments", marker = "python_full_version >= '3.11'" },
+    { name = "pygments" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/ef/4c/5dd1d8af08107f88c7f741ead7a40854b8ac24ddf9ae850afbcf698aa552/ipython_pygments_lexers-1.1.1.tar.gz", hash = "sha256:09c0138009e56b6854f9535736f4171d855c8c08a563a0dcd8022f78355c7e81", size = 8393, upload-time = "2025-01-17T11:24:34.505Z" }
 wheels = [
@@ -2858,7 +2861,7 @@ name = "nvidia-cudnn-cu13"
 version = "9.19.0.56"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "nvidia-cublas", marker = "(python_full_version < '3.14' and sys_platform == 'emscripten') or (python_full_version < '3.14' and sys_platform == 'win32') or (sys_platform != 'emscripten' and sys_platform != 'win32')" },
+    { name = "nvidia-cublas" },
 ]
 wheels = [
     { url = "https://files.pythonhosted.org/packages/f1/84/26025437c1e6b61a707442184fa0c03d083b661adf3a3eecfd6d21677740/nvidia_cudnn_cu13-9.19.0.56-py3-none-manylinux_2_27_aarch64.whl", hash = "sha256:6ed29ffaee1176c612daf442e4dd6cfeb6a0caa43ddcbeb59da94953030b1be4", size = 433781201, upload-time = "2026-02-03T20:40:53.805Z" },
@@ -2870,7 +2873,7 @@ name = "nvidia-cufft"
 version = "12.0.0.61"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "nvidia-nvjitlink", marker = "(python_full_version < '3.14' and sys_platform == 'emscripten') or (python_full_version < '3.14' and sys_platform == 'win32') or (sys_platform != 'emscripten' and sys_platform != 'win32')" },
+    { name = "nvidia-nvjitlink" },
 ]
 wheels = [
     { url = "https://files.pythonhosted.org/packages/8b/ae/f417a75c0259e85c1d2f83ca4e960289a5f814ed0cea74d18c353d3e989d/nvidia_cufft-12.0.0.61-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:2708c852ef8cd89d1d2068bdbece0aa188813a0c934db3779b9b1faa8442e5f5", size = 214053554, upload-time = "2025-09-04T08:31:38.196Z" },
@@ -2900,9 +2903,9 @@ name = "nvidia-cusolver"
 version = "12.0.4.66"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "nvidia-cublas", marker = "(python_full_version < '3.14' and sys_platform == 'emscripten') or (python_full_version < '3.14' and sys_platform == 'win32') or (sys_platform != 'emscripten' and sys_platform != 'win32')" },
+    { name = "nvidia-cublas" },
-    { name = "nvidia-cusparse", marker = "(python_full_version < '3.14' and sys_platform == 'emscripten') or (python_full_version < '3.14' and sys_platform == 'win32') or (sys_platform != 'emscripten' and sys_platform != 'win32')" },
+    { name = "nvidia-cusparse" },
-    { name = "nvidia-nvjitlink", marker = "(python_full_version < '3.14' and sys_platform == 'emscripten') or (python_full_version < '3.14' and sys_platform == 'win32') or (sys_platform != 'emscripten' and sys_platform != 'win32')" },
+    { name = "nvidia-nvjitlink" },
 ]
 wheels = [
     { url = "https://files.pythonhosted.org/packages/c8/c3/b30c9e935fc01e3da443ec0116ed1b2a009bb867f5324d3f2d7e533e776b/nvidia_cusolver-12.0.4.66-py3-none-manylinux_2_27_aarch64.whl", hash = "sha256:02c2457eaa9e39de20f880f4bd8820e6a1cfb9f9a34f820eb12a155aa5bc92d2", size = 223467760, upload-time = "2025-09-04T08:33:04.222Z" },
@@ -2914,7 +2917,7 @@ name = "nvidia-cusparse"
 version = "12.6.3.3"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "nvidia-nvjitlink", marker = "(python_full_version < '3.14' and sys_platform == 'emscripten') or (python_full_version < '3.14' and sys_platform == 'win32') or (sys_platform != 'emscripten' and sys_platform != 'win32')" },
+    { name = "nvidia-nvjitlink" },
 ]
 wheels = [
     { url = "https://files.pythonhosted.org/packages/f8/94/5c26f33738ae35276672f12615a64bd008ed5be6d1ebcb23579285d960a9/nvidia_cusparse-12.6.3.3-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:80bcc4662f23f1054ee334a15c72b8940402975e0eab63178fc7e670aa59472c", size = 162155568, upload-time = "2025-09-04T08:33:42.864Z" },
@@ -3091,10 +3094,10 @@ resolution-markers = [
     "python_full_version < '3.11'",
 ]
 dependencies = [
-    { name = "numpy", version = "2.2.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.11'" },
+    { name = "numpy", version = "2.2.6", source = { registry = "https://pypi.org/simple" } },
-    { name = "python-dateutil", marker = "python_full_version < '3.11'" },
+    { name = "python-dateutil" },
-    { name = "pytz", marker = "python_full_version < '3.11'" },
+    { name = "pytz" },
-    { name = "tzdata", marker = "python_full_version < '3.11'" },
+    { name = "tzdata" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/9c/d6/9f8431bacc2e19dca897724cd097b1bb224a6ad5433784a44b587c7c13af/pandas-2.2.3.tar.gz", hash = "sha256:4f18ba62b61d7e192368b84517265a99b4d7ee8912f8708660fb4a366cc82667", size = 4399213, upload-time = "2024-09-20T13:10:04.827Z" }
 wheels = [
@@ -3143,11 +3146,11 @@ resolution-markers = [
     "python_full_version == '3.11.*'",
 ]
 dependencies = [
-    { name = "numpy", version = "2.4.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.11.*'" },
+    { name = "numpy", version = "2.4.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.12' or python_full_version >= '3.14'" },
     { name = "numpy", version = "2.5.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.12' and python_full_version < '3.14'" },
-    { name = "python-dateutil", marker = "python_full_version >= '3.11' and python_full_version < '3.14'" },
+    { name = "python-dateutil" },
-    { name = "pytz", marker = "python_full_version >= '3.11' and python_full_version < '3.14'" },
+    { name = "pytz" },
-    { name = "tzdata", marker = "python_full_version >= '3.11' and python_full_version < '3.14'" },
+    { name = "tzdata" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/33/01/d40b85317f86cf08d853a4f495195c73815fdf205eef3993821720274518/pandas-2.3.3.tar.gz", hash = "sha256:e05e1af93b977f7eafa636d043f9f94c7ee3ac81af99c13508215942e64c993b", size = 4495223, upload-time = "2025-09-29T23:34:51.853Z" }
 wheels = [
@@ -3210,9 +3213,9 @@ resolution-markers = [
     "python_full_version >= '3.14' and sys_platform != 'emscripten' and sys_platform != 'win32'",
 ]
 dependencies = [
-    { name = "numpy", version = "2.5.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.14'" },
+    { name = "numpy", version = "2.5.1", source = { registry = "https://pypi.org/simple" } },
-    { name = "python-dateutil", marker = "python_full_version >= '3.14'" },
+    { name = "python-dateutil" },
-    { name = "tzdata", marker = "(python_full_version >= '3.14' and sys_platform == 'emscripten') or (python_full_version >= '3.14' and sys_platform == 'win32')" },
+    { name = "tzdata", marker = "sys_platform == 'emscripten' or sys_platform == 'win32'" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/f8/87/4341c6252d1c47b08768c3d25ac487362bf403f0313ddae4a2a26c9b1b4c/pandas-3.0.3.tar.gz", hash = "sha256:696a4a00a2a2a35d4e5deb3fc946641b96c944f02230e4f76137fe35d806c4fc", size = 4651414, upload-time = "2026-05-11T18:54:29.21Z" }
 wheels = [
@@ -3320,7 +3323,7 @@ name = "pexpect"
 version = "4.9.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "ptyprocess", marker = "(python_full_version < '3.14' and sys_platform == 'emscripten') or (python_full_version < '3.14' and sys_platform == 'win32') or (sys_platform != 'emscripten' and sys_platform != 'win32')" },
+    { name = "ptyprocess" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/42/92/cc564bf6381ff43ce1f4d06852fc19a2f11d180f23dc32d9588bee2f149d/pexpect-4.9.0.tar.gz", hash = "sha256:ee7d41123f3c9911050ea2c2dac107568dc43b2d3b0c7557a33212c398ead30f", size = 166450, upload-time = "2023-11-25T09:07:26.339Z" }
 wheels = [
@@ -3912,8 +3915,8 @@ crypto = [
 
 [[package]]
 name = "pylance"
-version = "7.0.0"
+version = "9.0.0rc1"
-source = { registry = "https://pypi.org/simple" }
+source = { registry = "https://pypi.fury.io/lance-format/" }
 dependencies = [
     { name = "lance-namespace" },
     { name = "numpy", version = "2.2.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.11'" },
@@ -3922,12 +3925,12 @@ dependencies = [
     { name = "pyarrow" },
 ]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ac/ad/2f64921bf346e7075aef24a72595db44821724a3d89a9a92dd24e79632aa/pylance-7.0.0-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:98422021975be76e72b1572f41b8c9abb3bee5bdc9bfa5e9ce731110a65ed4d1", size = 62134146, upload-time = "2026-05-27T21:59:37.459Z" },
+    { url = "https://pypi.fury.io/lance-format/-/ver_vEHBE/pylance-9.0.0rc1-cp310-abi3-macosx_11_0_arm64.whl", hash = "sha256:f0b6b02a1808bb3072ee7fe4e36614cae6f86302513e73ec7f55b2234a963b24" },
-    { url = "https://files.pythonhosted.org/packages/73/1c/c5a01bee0160b55d9a98895cbd33091d038f0a0995b121ab72e629008d02/pylance-7.0.0-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4bec86ee5b6fbd8bfc493e653f0a1fba0303cfe5492b9b46fc25ab908edc7183", size = 65373684, upload-time = "2026-05-27T22:04:01.584Z" },
+    { url = "https://pypi.fury.io/lance-format/-/ver_1Jipm4/pylance-9.0.0rc1-cp310-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:30f0ebf0d88034301819eb964f9236ce555aaa58e7ab89c5975a3e2250bbb405" },
-    { url = "https://files.pythonhosted.org/packages/eb/da/1fe8b8f7dbfe734d76af76acc994fc360a0d0c79a4874ef69f5a72a58fe3/pylance-7.0.0-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:881491432c53184e52f8d1db8d5f872f39a03f36fb104bec77b33d379519d8b5", size = 69458555, upload-time = "2026-05-27T22:16:50.567Z" },
+    { url = "https://pypi.fury.io/lance-format/-/ver_IvKxo/pylance-9.0.0rc1-cp310-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:44609ea2615ea6e684b85478d1694af2026458f61cf7895ecc75e238bfd17aa8" },
-    { url = "https://files.pythonhosted.org/packages/76/f0/dd505cf3fd0226ab9d94759acd713125af1d3bfacfd80bbd52e3b9f89509/pylance-7.0.0-cp39-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:18453999e7fff4f76b16d6b7882c9df0628bd142ff95e2461bd7dd5ee3fe0af3", size = 65394430, upload-time = "2026-05-27T22:05:30.923Z" },
+    { url = "https://pypi.fury.io/lance-format/-/ver_2hidj1/pylance-9.0.0rc1-cp310-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:182167a8dba9eeabffbffd53bd5b8548613d4d459b7cd7b34a840dd00cbb806f" },
-    { url = "https://files.pythonhosted.org/packages/17/ba/2357b81034f28eb00790e258ed140289a6a887a7468ca9df6349fd186b27/pylance-7.0.0-cp39-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:04a58051d408c60fe76d41a220dcaf8fea8fb6d1aa0ca78a709b60bc3cc8d19a", size = 69473470, upload-time = "2026-05-27T22:17:18.935Z" },
+    { url = "https://pypi.fury.io/lance-format/-/ver_1dFx3r/pylance-9.0.0rc1-cp310-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:8a63b11e814b7eab758bcaf0d6f97eb05ea86203d9fb0af718c462c24c7d6c9c" },
-    { url = "https://files.pythonhosted.org/packages/1f/ec/5c00b6303a67d787f9475141832cbdc513d674ac3dcaeef8a7b169905e65/pylance-7.0.0-cp39-abi3-win_amd64.whl", hash = "sha256:467d4864af047eaab4e1370e2f1e88e2c6f507c079874421116cb41d78bc3629", size = 74792863, upload-time = "2026-05-27T22:19:23.875Z" },
+    { url = "https://pypi.fury.io/lance-format/-/ver_2a8dSh/pylance-9.0.0rc1-cp310-abi3-win_amd64.whl", hash = "sha256:2ff8b953ae2b0550490c1a7efd210aa91bc223d200ffac28849056cfd7436d97" },
 ]
 
 [[package]]
@@ -4683,10 +4686,10 @@ resolution-markers = [
     "python_full_version < '3.11'",
 ]
 dependencies = [
-    { name = "joblib", marker = "python_full_version < '3.11'" },
+    { name = "joblib" },
-    { name = "numpy", version = "2.2.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.11'" },
+    { name = "numpy", version = "2.2.6", source = { registry = "https://pypi.org/simple" } },
-    { name = "scipy", version = "1.15.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.11'" },
+    { name = "scipy", version = "1.15.3", source = { registry = "https://pypi.org/simple" } },
-    { name = "threadpoolctl", marker = "python_full_version < '3.11'" },
+    { name = "threadpoolctl" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/98/c2/a7855e41c9d285dfe86dc50b250978105dce513d6e459ea66a6aeb0e1e0c/scikit_learn-1.7.2.tar.gz", hash = "sha256:20e9e49ecd130598f1ca38a1d85090e1a600147b9c02fa6f15d69cb53d968fda", size = 7193136, upload-time = "2025-09-09T08:21:29.075Z" }
 wheels = [
@@ -4734,13 +4737,13 @@ resolution-markers = [
     "python_full_version == '3.11.*'",
 ]
 dependencies = [
-    { name = "joblib", marker = "python_full_version >= '3.11'" },
+    { name = "joblib" },
-    { name = "narwhals", marker = "python_full_version >= '3.11'" },
+    { name = "narwhals" },
-    { name = "numpy", version = "2.4.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.11.*'" },
+    { name = "numpy", version = "2.4.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.12'" },
     { name = "numpy", version = "2.5.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.12'" },
-    { name = "scipy", version = "1.17.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.11.*'" },
+    { name = "scipy", version = "1.17.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.12'" },
     { name = "scipy", version = "1.18.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.12'" },
-    { name = "threadpoolctl", marker = "python_full_version >= '3.11'" },
+    { name = "threadpoolctl" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/fa/6f/37092bdb25f712817231799fc5674d8e704066a8a70c1d2d40517e18b4ab/scikit_learn-1.9.0.tar.gz", hash = "sha256:8833266989d3a5110178a9fae30783675460724d0e1efb13b14901d2c660c557", size = 7750767, upload-time = "2026-06-02T11:54:32.706Z" }
 wheels = [
@@ -4784,7 +4787,7 @@ resolution-markers = [
     "python_full_version < '3.11'",
 ]
 dependencies = [
-    { name = "numpy", version = "2.2.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.11'" },
+    { name = "numpy", version = "2.2.6", source = { registry = "https://pypi.org/simple" } },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/0f/37/6964b830433e654ec7485e45a00fc9a27cf868d622838f6b6d9c5ec0d532/scipy-1.15.3.tar.gz", hash = "sha256:eae3cf522bc7df64b42cad3925c876e1b0b6c35c1337c93e12c0f366f55b0eaf", size = 59419214, upload-time = "2025-05-08T16:13:05.955Z" }
 wheels = [
@@ -4843,7 +4846,7 @@ resolution-markers = [
     "python_full_version == '3.11.*'",
 ]
 dependencies = [
-    { name = "numpy", version = "2.4.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.11.*'" },
+    { name = "numpy", version = "2.4.6", source = { registry = "https://pypi.org/simple" } },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/7a/97/5a3609c4f8d58b039179648e62dd220f89864f56f7357f5d4f45c29eb2cc/scipy-1.17.1.tar.gz", hash = "sha256:95d8e012d8cb8816c226aef832200b1d45109ed4464303e997c5b13122b297c0", size = 30573822, upload-time = "2026-02-23T00:26:24.851Z" }
 wheels = [
@@ -4920,7 +4923,7 @@ resolution-markers = [
     "python_full_version >= '3.12' and python_full_version < '3.14'",
 ]
 dependencies = [
-    { name = "numpy", version = "2.5.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.12'" },
+    { name = "numpy", version = "2.5.1", source = { registry = "https://pypi.org/simple" } },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/a7/25/c2700dfaf6442b4effaa91af24ebce5dc9d31bb4a69706313aae70d72cd0/scipy-1.18.0.tar.gz", hash = "sha256:67b2ad2ad54c72ca6d04975a9b2df8c3638c34ddd5b28738e94fc2b57929d378", size = 30774447, upload-time = "2026-06-19T15:01:43.456Z" }
 wheels = [

rust/lancedb/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "lancedb"
-version = "0.38.0-beta.11"
+version = "0.39.0-beta.4"
 edition.workspace = true
 description = "LanceDB: A serverless, low-latency vector database for AI applications"
 license.workspace = true
@@ -21,6 +21,8 @@ arrow-select = { workspace = true }
 arrow-ord = { workspace = true }
 arrow-cast = { workspace = true }
 arrow-ipc.workspace = true
+arrow-flight = { workspace = true, optional = true }
+prost = { version = "0.14", optional = true }
 chrono = { workspace = true }
 datafusion-catalog.workspace = true
 datafusion-common.workspace = true
@@ -77,6 +79,7 @@ reqwest = { version = "0.12.0", default-features = false, features = [
     "rustls-tls-native-roots",
     "stream",
 ], optional = true }
+tonic = { workspace = true, optional = true }
 http = { version = "1", optional = true } # Matching what is in reqwest
 urlencoding = { version = "2", optional = true }
 uuid = { workspace = true, features = ["v5"] }
@@ -145,8 +148,11 @@ huggingface = [
 ]
 dynamodb = ["lance/dynamodb", "aws"]
 remote = [
+    "dep:arrow-flight",
+    "dep:prost",
     "dep:reqwest",
     "dep:http",
+    "dep:tonic",
     "dep:urlencoding",
     "lance-namespace-impls/rest",
     "lance-namespace-impls/rest-adapter",

rust/lancedb/src/blob.rs

@@ -7,7 +7,9 @@
 //! raw `Binary` / `LargeBinary` into the blob struct layout. Queries return
 //! small descriptors, not bytes.
 //!
-//! Blob tables require Lance file format >= 2.2 and stable row ids at create.
+//! Blob tables require Lance file format >= 2.2. `_rowid` values stay valid
+//! after compaction when the table has stable row ids. Overwrite is a new
+//! create and does not keep the previous table's stable row id setting.
 
 use std::ops::Range;
 use std::sync::Arc;
@@ -324,6 +326,7 @@ pub(crate) fn blob_column_names(schema: &Schema) -> Vec<String> {
 }
 
 /// Bumps storage format to at least [`LanceFileVersion::V2_2`] for blob schemas.
+/// Leaves `enable_stable_row_ids` unchanged.
 pub(crate) fn ensure_blob_storage_version(schema: &Schema, params: &mut WriteParams) {
     if !has_blob_columns(schema) {
         return;
@@ -385,6 +388,30 @@ fn ensure_all_row_ids_resolved(column: &str, requested: usize, resolved: usize)
     }
 }
 
+/// Lance take reports a missing physical row address as NotSupported or InvalidInput.
+fn map_blob_take_error(column: &str, requested: usize, err: lance::Error) -> Error {
+    let missing_row_addr = match &err {
+        lance::Error::NotSupported { source, .. } => {
+            source.to_string().contains("must not target deleted rows")
+        }
+        lance::Error::InvalidInput { source, .. } => source
+            .to_string()
+            .contains("belongs to non-existent fragment"),
+        _ => false,
+    };
+
+    if missing_row_addr {
+        Error::InvalidInput {
+            message: format!(
+                "blob read for column '{column}' requested {requested} row ids but some \
+                 do not exist in the table; pass row ids collected from this table"
+            ),
+        }
+    } else {
+        err.into()
+    }
+}
+
 /// Materialize blob-local ranges (same length and order as `requests`, nulls preserved).
 pub(crate) async fn take_blob_ranges_aligned(
     dataset: &Arc<Dataset>,
@@ -405,7 +432,8 @@ pub(crate) async fn take_blob_ranges_aligned(
         .with_row_ids(lance_requests)
         .preserve_order(true)
         .execute()
-        .await?;
+        .await
+        .map_err(|err| map_blob_take_error(column, requests.len(), err))?;
     ensure_all_row_ids_resolved(column, requests.len(), payloads.len())?;
 
     let mut builder = LargeBinaryBuilder::new();
@@ -434,7 +462,8 @@ pub(crate) async fn take_blobs_aligned(
         .with_row_ids(row_ids.to_vec())
         .preserve_order(true)
         .execute()
-        .await?;
+        .await
+        .map_err(|err| map_blob_take_error(column, row_ids.len(), err))?;
     ensure_all_row_ids_resolved(column, row_ids.len(), payloads.len())?;
 
     let mut builder = LargeBinaryBuilder::new();
@@ -458,7 +487,10 @@ pub(crate) async fn take_blob_files_aligned(
         return Ok(Vec::new());
     }
 
-    let handles = dataset.take_blobs(row_ids, column).await?;
+    let handles = dataset
+        .take_blobs(row_ids, column)
+        .await
+        .map_err(|err| map_blob_take_error(column, row_ids.len(), err))?;
     ensure_all_row_ids_resolved(column, row_ids.len(), handles.len())?;
     Ok(handles
         .into_iter()
@@ -504,6 +536,21 @@ mod tests {
             params.data_storage_version.unwrap().resolve(),
             ConcreteFileVersion::V2_2
         );
+        assert!(!params.enable_stable_row_ids);
+    }
+
+    #[test]
+    fn storage_version_leaves_stable_row_ids_enabled() {
+        let mut params = WriteParams {
+            enable_stable_row_ids: true,
+            ..Default::default()
+        };
+        ensure_blob_storage_version(&blob_schema(), &mut params);
+        assert!(params.enable_stable_row_ids);
+        assert_eq!(
+            params.data_storage_version.unwrap().resolve(),
+            ConcreteFileVersion::V2_2
+        );
     }
 
     #[test]
@@ -576,5 +623,6 @@ mod tests {
         let mut params = WriteParams::default();
         ensure_blob_storage_version(&schema, &mut params);
         assert!(params.data_storage_version.is_none());
+        assert!(!params.enable_stable_row_ids);
     }
 }

rust/lancedb/src/connection.rs

@@ -23,16 +23,20 @@ use crate::connection::create_table::CreateTableBuilder;
 use crate::data::scannable::Scannable;
 use crate::database::listing::ListingDatabase;
 use crate::database::{
-    CloneTableRequest, Database, DatabaseOptions, JobDescription, JobInfo, OpenTableRequest,
+    CloneTableRequest, Database, DatabaseOptions, JobInfo, OpenTableRequest, ReadConsistency,
-    ReadConsistency, TableNamesRequest,
+    TableNamesRequest,
 };
 use crate::embeddings::{EmbeddingRegistry, MemoryRegistry};
 use crate::error::{Error, Result};
 #[cfg(feature = "remote")]
 use crate::remote::{
     client::ClientConfig,
-    db::{OPT_REMOTE_API_KEY, OPT_REMOTE_HOST_OVERRIDE, OPT_REMOTE_REGION},
+    db::{
+        OPT_REMOTE_API_KEY, OPT_REMOTE_HOST_OVERRIDE, OPT_REMOTE_REGION,
+        OPT_REMOTE_SQL_HOST_OVERRIDE,
+    },
 };
+use crate::secrets::SecretInfo;
 use lance::io::ObjectStoreParams;
 pub use lance_file::version::LanceFileVersion;
 #[cfg(feature = "remote")]
@@ -322,6 +326,43 @@ pub struct CloneTableBuilder {
     request: CloneTableRequest,
 }
 
+/// Builder for asynchronously executing a SQL statement on a remote database.
+pub struct ExecuteQueryAsyncBuilder {
+    parent: Arc<dyn Database>,
+    query: String,
+    default_namespace_path: Vec<String>,
+}
+
+impl ExecuteQueryAsyncBuilder {
+    fn new(parent: Arc<dyn Database>, query: String) -> Self {
+        Self {
+            parent,
+            query,
+            default_namespace_path: vec!["public".to_string()],
+        }
+    }
+
+    /// Set the namespace used for unqualified table names.
+    ///
+    /// An empty path is treated as `public`, which is the SQL name for the
+    /// root Lance namespace.
+    pub fn default_namespace_path<I, S>(mut self, path: I) -> Self
+    where
+        I: IntoIterator<Item = S>,
+        S: Into<String>,
+    {
+        self.default_namespace_path = path.into_iter().map(Into::into).collect();
+        self
+    }
+
+    /// Start the statement and return its asynchronous query handle.
+    pub async fn execute(self) -> Result<crate::sql::Query> {
+        self.parent
+            .execute_query_async(&self.query, &self.default_namespace_path)
+            .await
+    }
+}
+
 impl CloneTableBuilder {
     fn new(parent: Arc<dyn Database>, target_table_name: String, source_uri: String) -> Self {
         Self {
@@ -405,6 +446,51 @@ impl Connection {
         &self.internal
     }
 
+    /// Start executing SQL on a remote LanceDB database.
+    ///
+    /// The query can reference tables in other databases with SQL dot notation.
+    /// Use [`ExecuteQueryAsyncBuilder::default_namespace_path`] to avoid qualifying
+    /// tables in the default namespace. Local connections return
+    /// [`Error::NotSupported`].
+    ///
+    /// # Example
+    ///
+    /// ```no_run
+    /// # async fn query(db: &lancedb::Connection) -> lancedb::Result<()> {
+    /// use futures::TryStreamExt;
+    ///
+    /// let query = db
+    ///     .execute_query_async("SELECT * FROM events LIMIT 10")
+    ///     .default_namespace_path(["public"])
+    ///     .execute()
+    ///     .await?;
+    /// println!("query id: {}", query.id());
+    /// let mut batches = query.reader().await?;
+    /// while let Some(batch) = batches.try_next().await? {
+    ///     println!("received {} rows", batch.num_rows());
+    /// }
+    /// # Ok(())
+    /// # }
+    /// ```
+    pub fn execute_query_async(&self, query: impl Into<String>) -> ExecuteQueryAsyncBuilder {
+        ExecuteQueryAsyncBuilder::new(self.internal.clone(), query.into())
+    }
+
+    /// Describe a submitted SQL query by its connection-scoped id.
+    ///
+    /// This performs one bounded status poll using state retained by this
+    /// connection. Running state with a live query handle is not evicted;
+    /// abandoned state has bounded retention, and server expiration is
+    /// honored. Terminal state is retained briefly.
+    /// Query ids are not portable to another connection. Local connections
+    /// return [`Error::NotSupported`].
+    pub async fn describe_query(
+        &self,
+        query_id: uuid::Uuid,
+    ) -> Result<crate::sql::QueryDescription> {
+        self.internal.describe_query(query_id).await
+    }
+
     /// Get the names of all tables in the database
     ///
     /// The names will be returned in lexicographical order (ascending)
@@ -501,6 +587,7 @@ impl Connection {
     /// Registration is remote-only and always asynchronous. Waiting on the
     /// returned typed job yields the durable [`crate::function::FunctionVersion`].
     /// Local databases return [`Error::NotSupported`].
+    ///
     pub async fn create_function_async(
         &self,
         request: crate::function::FunctionRegistrationRequest,
@@ -523,6 +610,96 @@ impl Connection {
             .await
     }
 
+    /// List every published immutable Function version in the remote catalog.
+    ///
+    /// Results are ordered by Function name then version. The client walks all
+    /// server pages before returning. Local databases return
+    /// [`Error::NotSupported`].
+    ///
+    /// # Example
+    ///
+    /// ```no_run
+    /// # async fn list_functions(
+    /// #     connection: &lancedb::Connection,
+    /// # ) -> Result<(), Box<dyn std::error::Error>> {
+    /// for function in connection.list_functions().await? {
+    ///     println!("{} {}", function.name(), function.version());
+    /// }
+    /// # Ok(())
+    /// # }
+    /// ```
+    pub async fn list_functions(&self) -> Result<Vec<crate::function::FunctionVersion>> {
+        self.internal.list_functions().await
+    }
+
+    /// Drop one exact immutable Function version from the remote catalog.
+    ///
+    /// Returns `true` when the server appended a Dropped transition and
+    /// `false` for an idempotent replay. Local databases return
+    /// [`Error::NotSupported`].
+    pub async fn drop_function(
+        &self,
+        name: impl AsRef<str>,
+        version: impl AsRef<str>,
+    ) -> Result<bool> {
+        self.internal
+            .drop_function(name.as_ref(), version.as_ref())
+            .await
+    }
+
+    /// Create a named Secret in this database.
+    ///
+    /// Fails if the name is taken, so a create can never silently become a
+    /// rotation. There is no API that reads a stored credential back; the only
+    /// consumer is a Function that binds the Secret by name. Local databases
+    /// return [`Error::NotSupported`].
+    pub async fn create_secret(&self, name: impl AsRef<str>, value: impl AsRef<str>) -> Result<()> {
+        self.internal
+            .create_secret(name.as_ref(), value.as_ref())
+            .await
+    }
+
+    /// Replace the credential behind an existing Secret.
+    ///
+    /// Fails if it does not exist. Every Function bound to the Secret resolves
+    /// the new value from its next execution, and no new Function version is
+    /// minted -- which is what lets a rotation reach columns pinned to a
+    /// version registered before it. Local databases return
+    /// [`Error::NotSupported`].
+    pub async fn alter_secret(&self, name: impl AsRef<str>, value: impl AsRef<str>) -> Result<()> {
+        self.internal
+            .alter_secret(name.as_ref(), value.as_ref())
+            .await
+    }
+
+    /// The names of every Secret in this database.
+    ///
+    /// Names only. No path in this API returns a stored credential, by
+    /// construction rather than by policy. Local databases return
+    /// [`Error::NotSupported`].
+    pub async fn list_secrets(&self) -> Result<Vec<String>> {
+        self.internal.list_secrets().await
+    }
+
+    /// Drop a Secret.
+    ///
+    /// Functions bound to it fail at their next job, naming the Secret; that
+    /// is the revocation path. The name becomes free to reuse, and a new
+    /// Secret under it is picked up by everything still bound to that name.
+    /// Local databases return [`Error::NotSupported`].
+    pub async fn drop_secret(&self, name: impl AsRef<str>) -> Result<()> {
+        self.internal.drop_secret(name.as_ref()).await
+    }
+
+    /// What this database records about one Secret: its name and timestamps.
+    ///
+    /// Never the value. The type it returns has no field for one, so this is a
+    /// property of the API rather than of what the caller chooses to read.
+    /// Local databases return [`Error::NotSupported`].
+    pub async fn describe_secret(&self, name: impl AsRef<str>) -> Result<SecretInfo> {
+        self.internal.describe_secret(name.as_ref()).await
+    }
+
     /// Rename a table in the database.
     ///
     /// This is only supported in LanceDB Cloud.
@@ -548,14 +725,34 @@ impl Connection {
         self.internal.read_consistency().await
     }
 
-    /// A [`crate::job::Job`] handle for a server-side job by id, suitable for
+    /// Open a server-side job by id, returning a handle with its record
-    /// waiting on or cancelling the job.
+    /// already populated. Fails with [`crate::Error::JobNotFound`] when the
+    /// server has no such job, the way [`Connection::open_table`] does for a
+    /// missing table.
     ///
-    /// The handle is constructed without a server round trip; an unknown id
+    /// This is the one way in: the returned [`crate::job::Job`] answers for
-    /// surfaces when the handle is used. Only server-backed databases support
+    /// its own state, specification, result, failure and event history, so
-    /// job handles by id.
+    /// there is no separate connection-level call for any of them.
-    pub fn job(&self, job_id: impl AsRef<str>) -> Result<crate::job::Job> {
+    ///
-        self.internal.job(job_id.as_ref())
+    /// # Example
+    ///
+    /// ```no_run
+    /// # use lancedb::job::JobEventsRequest;
+    /// # async fn open_job(
+    /// #     connection: &lancedb::Connection,
+    /// #     job_id: &str,
+    /// # ) -> Result<(), Box<dyn std::error::Error>> {
+    /// let job = connection.open_job(job_id).await?;
+    /// println!("{:?} {:?}", job.state(), job.result());
+    /// let done = job
+    ///     .events(JobEventsRequest::default().filter("state = 'claim_complete'"))
+    ///     .await?;
+    /// println!("{} completions", done.iter().map(|b| b.num_rows()).sum::<usize>());
+    /// # Ok(())
+    /// # }
+    /// ```
+    pub async fn open_job(&self, job_id: impl AsRef<str>) -> Result<crate::job::Job> {
+        self.internal.open_job(job_id.as_ref()).await
     }
 
     /// List server-side jobs across the database's tables.
@@ -563,24 +760,12 @@ impl Connection {
         self.internal.list_jobs().await
     }
 
-    /// Describe a single server-side job by id. `None` when the server has no
-    /// such job.
-    pub async fn get_job(&self, job_id: impl AsRef<str>) -> Result<Option<JobDescription>> {
-        self.internal.get_job(job_id.as_ref()).await
-    }
-
     /// Request cancellation of a server-side job by id. Returns true if the
     /// server accepted the cancellation, false if no such job exists.
     pub async fn cancel_job(&self, job_id: impl AsRef<str>) -> Result<bool> {
         self.internal.cancel_job(job_id.as_ref()).await
     }
 
-    /// The lifecycle event history of a server-side job (all jobs when
-    /// `job_id` is `None`), as recorded Arrow batches.
-    pub async fn job_history(&self, job_id: Option<&str>) -> Result<Vec<RecordBatch>> {
-        self.internal.job_history(job_id).await
-    }
-
     /// Drop a table in the database.
     ///
     /// # Arguments
@@ -827,6 +1012,19 @@ impl ConnectBuilder {
         self
     }
 
+    /// Set the SQL service host override for a remote connection.
+    ///
+    /// The SQL client is initialized lazily when the connection first executes
+    /// SQL and is retained for the connection's lifetime.
+    #[cfg(feature = "remote")]
+    pub fn sql_host_override(mut self, sql_host_override: &str) -> Self {
+        self.request.options.insert(
+            OPT_REMOTE_SQL_HOST_OVERRIDE.to_string(),
+            sql_host_override.to_string(),
+        );
+        self
+    }
+
     /// Set the database specific options
     ///
     /// See [crate::database::listing::ListingDatabaseOptions] for the options available for
@@ -1016,6 +1214,7 @@ impl ConnectBuilder {
 
         let mut merged_options = self.request.options.clone();
         Self::apply_env_defaults(&ENV_VARS_TO_STORAGE_OPTS, &mut merged_options);
+        let sql_host_override = merged_options.get(OPT_REMOTE_SQL_HOST_OVERRIDE).cloned();
         let options = RemoteDatabaseOptions::parse_from_map(&merged_options)?;
 
         let region = options.region.ok_or_else(|| Error::InvalidInput {
@@ -1057,11 +1256,15 @@ impl ConnectBuilder {
         }
 
         let storage_options = StorageOptions(options.storage_options.clone());
+        let host_overrides = crate::remote::db::RemoteHostOverrides {
+            rest: options.host_override,
+            sql: sql_host_override,
+        };
         let internal = Arc::new(crate::remote::db::RemoteDatabase::try_new(
             &self.request.uri,
             &api_key,
             &region,
-            options.host_override,
+            host_overrides,
             client_config,
             storage_options.into(),
             self.request.read_consistency_interval,
@@ -1355,6 +1558,23 @@ mod tests {
         assert_eq!(tc.connection.uri(), tc.uri);
     }
 
+    #[tokio::test]
+    async fn test_local_connection_rejects_sql_queries() {
+        let directory = tempdir().unwrap();
+        let connection = connect(directory.path().to_str().unwrap())
+            .execute()
+            .await
+            .unwrap();
+        assert!(matches!(
+            connection.execute_query_async("SELECT 1").execute().await,
+            Err(Error::NotSupported { .. })
+        ));
+        assert!(matches!(
+            connection.describe_query(uuid::Uuid::nil()).await,
+            Err(Error::NotSupported { .. })
+        ));
+    }
+
     #[cfg(feature = "remote")]
     #[test]
     fn test_apply_env_defaults() {

rust/lancedb/src/database.rs

@@ -18,8 +18,6 @@ use std::collections::HashMap;
 use std::sync::Arc;
 use std::time::Duration;
 
-use arrow_array::RecordBatch;
-
 use lance::dataset::ReadParams;
 use lance_namespace::LanceNamespace;
 use lance_namespace::models::{
@@ -30,6 +28,7 @@ use lance_namespace::models::{
 
 use crate::data::scannable::Scannable;
 use crate::error::Result;
+use crate::secrets::SecretInfo;
 use crate::table::{BaseTable, WriteOptions};
 
 pub mod listing;
@@ -206,8 +205,8 @@ pub enum ReadConsistency {
 /// compaction, column refresh, ...).
 #[derive(Debug, Clone)]
 pub struct JobInfo {
-    /// The job id -- what [`Database::get_job`] and [`Database::cancel_job`]
+    /// The job id -- what [`Database::open_job`] and
-    /// accept.
+    /// [`Database::cancel_job`] accept.
     pub job_id: String,
     /// The table the job runs against, without URI or namespace.
     pub table: String,
@@ -218,8 +217,8 @@ pub struct JobInfo {
     pub created_at_millis: i64,
 }
 
-/// A described job from [`Database::get_job`]: lifecycle state plus the
+/// The server-side record behind a [`crate::job::Job`] handle: lifecycle
-/// job-type-specific specification.
+/// state plus the job-type-specific specification and result.
 #[derive(Debug, Clone)]
 pub struct JobDescription {
     pub job_id: String,
@@ -230,6 +229,10 @@ pub struct JobDescription {
     pub creation_ms: i64,
     /// The job-type-specific specification. Null when the server omits it.
     pub spec: serde_json::Value,
+    /// The job-type-specific terminal result, for job types that define one.
+    /// `None` until the job succeeds, so a job that never terminates reports
+    /// its progress through [`crate::job::Job::events`] instead.
+    pub result: Option<serde_json::Value>,
     /// Why the job failed, when the job is failed and the server reports a
     /// reason.
     pub failure: Option<crate::error::JobFailure>,
@@ -247,6 +250,12 @@ fn function_catalog_not_supported<T>() -> Result<T> {
     })
 }
 
+fn secret_catalog_not_supported<T>() -> Result<T> {
+    Err(crate::error::Error::NotSupported {
+        message: "Secret operations are not supported by this database".to_string(),
+    })
+}
+
 /// The `Database` trait defines the interface for database implementations.
 ///
 /// A database is responsible for managing tables and their metadata.
@@ -307,30 +316,73 @@ pub trait Database:
     ) -> Result<crate::function::FunctionVersion> {
         function_catalog_not_supported()
     }
-    /// A [`crate::job::Job`] handle for a server-side job by id, suitable for
+    /// List every published immutable Function version in the remote catalog.
-    /// waiting on or cancelling the job. The handle is constructed without a
+    async fn list_functions(&self) -> Result<Vec<crate::function::FunctionVersion>> {
-    /// server round trip; an unknown id surfaces when the handle is used.
+        function_catalog_not_supported()
-    fn job(&self, _job_id: &str) -> Result<crate::job::Job> {
+    }
-        job_op_not_supported("job")
+    /// Drop one exact immutable Function version from the remote catalog.
+    async fn drop_function(&self, _name: &str, _version: &str) -> Result<bool> {
+        function_catalog_not_supported()
+    }
+    /// Create a named Secret in this database. Fails if the name is taken, so
+    /// a create can never silently become a rotation.
+    async fn create_secret(&self, _name: &str, _value: &str) -> Result<()> {
+        secret_catalog_not_supported()
+    }
+    /// Replace the credential behind an existing Secret. Fails if it does not
+    /// exist. Every Function bound to it resolves the new value from its next
+    /// execution, with no new Function version.
+    async fn alter_secret(&self, _name: &str, _value: &str) -> Result<()> {
+        secret_catalog_not_supported()
+    }
+    /// The names of every Secret in this database.
+    ///
+    /// Names only. No API path returns a stored credential, by construction
+    /// rather than by policy.
+    async fn list_secrets(&self) -> Result<Vec<String>> {
+        secret_catalog_not_supported()
+    }
+    /// Drop a Secret. Functions bound to it fail at their next job, which is
+    /// the revocation path.
+    async fn drop_secret(&self, _name: &str) -> Result<()> {
+        secret_catalog_not_supported()
+    }
+    /// What the database records about one Secret: its name and timestamps,
+    /// never its value.
+    async fn describe_secret(&self, _name: &str) -> Result<SecretInfo> {
+        secret_catalog_not_supported()
+    }
+    /// Open a job by id, returning a handle with its record already
+    /// populated. Fails with [`crate::Error::JobNotFound`] when the server has
+    /// no such job.
+    async fn open_job(&self, _job_id: &str) -> Result<crate::job::Job> {
+        job_op_not_supported("open_job")
     }
     /// List server-side jobs across the database's tables.
     async fn list_jobs(&self) -> Result<Vec<JobInfo>> {
         job_op_not_supported("list_jobs")
     }
-    /// Describe a single job by id. `None` when the server has no such job.
-    async fn get_job(&self, _job_id: &str) -> Result<Option<JobDescription>> {
-        job_op_not_supported("get_job")
-    }
     /// Request cancellation of a job by id. Returns true if the server
     /// accepted the cancellation, false if no such job exists. Cancelling an
     /// already-terminal job is a no-op success.
     async fn cancel_job(&self, _job_id: &str) -> Result<bool> {
         job_op_not_supported("cancel_job")
     }
-    /// The lifecycle event history of a job (all jobs when `job_id` is
+    /// Start executing a SQL statement on a remote database.
-    /// `None`), as recorded Arrow batches.
+    async fn execute_query_async(
-    async fn job_history(&self, _job_id: Option<&str>) -> Result<Vec<RecordBatch>> {
+        &self,
-        job_op_not_supported("job_history")
+        _query: &str,
+        _default_namespace_path: &[String],
+    ) -> Result<crate::sql::Query> {
+        Err(crate::error::Error::NotSupported {
+            message: "SQL is not supported by this database".to_string(),
+        })
+    }
+    /// Describe a submitted SQL query by its connection-scoped id.
+    async fn describe_query(&self, _query_id: uuid::Uuid) -> Result<crate::sql::QueryDescription> {
+        Err(crate::error::Error::NotSupported {
+            message: "SQL is not supported by this database".to_string(),
+        })
     }
     /// Open a table in the database
     async fn open_table(&self, request: OpenTableRequest) -> Result<Arc<dyn BaseTable>>;

rust/lancedb/src/database/listing.rs

@@ -18,7 +18,7 @@ use lance_table::io::commit::commit_handler_from_url;
 use object_store::local::LocalFileSystem;
 use snafu::ResultExt;
 
-use crate::blob::{ensure_blob_storage_version, has_blob_columns};
+use crate::blob::ensure_blob_storage_version;
 use crate::connection::ConnectRequest;
 use crate::database::ReadConsistency;
 use crate::database::namespace::LanceNamespaceDatabase;
@@ -827,7 +827,6 @@ impl ListingDatabase {
         if let Some(enable_stable_row_ids) = overrides
             .enable_stable_row_ids
             .or(self.new_table_config.enable_stable_row_ids)
-            .or(has_blob_columns(&data_schema).then_some(true))
         {
             write_params.enable_stable_row_ids = enable_stable_row_ids;
         }

rust/lancedb/src/database/namespace.rs

@@ -23,7 +23,7 @@ use lance_namespace_impls::ConnectBuilder;
 use lance_table::io::commit::CommitHandler;
 use lance_table::io::commit::external_manifest::ExternalManifestCommitHandler;
 
-use crate::blob::{ensure_blob_storage_version, has_blob_columns};
+use crate::blob::ensure_blob_storage_version;
 use crate::connection::NamespaceClientPushdownOperation;
 use crate::database::ReadConsistency;
 use crate::database::listing::{NewTableConfig, take_request_creation_overrides};
@@ -217,7 +217,6 @@ impl LanceNamespaceDatabase {
         if let Some(enable_stable_row_ids) = overrides
             .enable_stable_row_ids
             .or(self.new_table_config.enable_stable_row_ids)
-            .or(has_blob_columns(data_schema.as_ref()).then_some(true))
         {
             params.enable_stable_row_ids = enable_stable_row_ids;
         }
@@ -539,9 +538,7 @@ impl Database for LanceNamespaceDatabase {
         self.namespace
             .drop_table(drop_request)
             .await
-            .map_err(|e| Error::Runtime {
+            .map_err(|e| map_namespace_lance_error(e, name))?;
-                message: format!("Failed to drop table: {}", e),
-            })?;
 
         Ok(())
     }
@@ -1495,6 +1492,15 @@ mod tests {
             .expect("Failed to list tables");
         assert!(!table_names_after.contains(&"drop_test".to_string()));
 
+        let error = conn
+            .drop_table("drop_test", &["test_ns".into()])
+            .await
+            .expect_err("dropping a missing table should fail");
+        assert!(
+            matches!(error, Error::TableNotFound { ref name, .. } if name == "drop_test"),
+            "expected TableNotFound, got: {error:?}"
+        );
+
         // Verify: Cannot open dropped table
         let open_result = conn.open_table("drop_test").execute().await;
         assert!(open_result.is_err());

rust/lancedb/src/dataloader/permutation/reader.rs

@@ -31,7 +31,7 @@ use lance::io::RecordBatchStream;
 use lance_arrow::RecordBatchExt;
 use lance_core::ROW_ID;
 use lance_core::error::LanceOptionExt;
-use std::collections::HashMap;
+use std::collections::{HashMap, HashSet};
 use std::sync::Arc;
 
 /// Reads a permutation of a source table based on row IDs stored in a separate table
@@ -234,7 +234,14 @@ impl PermutationReader {
             .expect_ok()?
             .values();
 
-        let in_list: Vec<Expr> = row_ids.iter().map(|id| lit(*id)).collect();
+        let mut unique_row_ids = HashSet::with_capacity(num_rows);
+        let in_list: Vec<Expr> = row_ids
+            .iter()
+            .copied()
+            .filter(|row_id| unique_row_ids.insert(*row_id))
+            .map(lit)
+            .collect();
+        let num_unique_row_ids = unique_row_ids.len();
 
         let base_query = QueryRequest {
             filter: Some(QueryFilter::Datafusion(col(ROW_ID).in_list(in_list, false))),
@@ -247,7 +254,7 @@ impl PermutationReader {
             .query(
                 &AnyQuery::Query(base_query),
                 QueryExecutionOptions {
-                    max_batch_length: num_rows as u32,
+                    max_batch_length: num_unique_row_ids as u32,
                     ..Default::default()
                 },
             )
@@ -262,9 +269,9 @@ impl PermutationReader {
             });
         }
 
-        if batches.iter().map(|b| b.num_rows()).sum::<usize>() != num_rows {
+        if batches.iter().map(|b| b.num_rows()).sum::<usize>() != num_unique_row_ids {
             return Err(Error::InvalidInput {
-                message: "Base table returned different number of rows than the number of row IDs"
+                message: "Base table returned a different number of rows than the number of unique row IDs"
                     .to_string(),
             });
         }
@@ -504,6 +511,7 @@ impl PermutationReader {
             let table = Table::from(self.base_table.clone());
             let batches = table
                 .take_offsets(offsets.to_vec())
+                .preserve_order()
                 .select(selection.clone())
                 .execute()
                 .await?
@@ -803,10 +811,10 @@ mod tests {
         .unwrap();
 
         // Take offsets in reverse order and verify returned rows match that order
-        let offsets = vec![5, 3, 1, 0];
+        let offsets = vec![5, 3, 5, 1, 0];
         let batch = reader.take_offsets(&offsets, Select::All).await.unwrap();
 
-        assert_eq!(batch.num_rows(), 4);
+        assert_eq!(batch.num_rows(), 5);
 
         let idx_values = batch
             .column(0)
@@ -820,6 +828,52 @@ mod tests {
         assert_eq!(idx_values, expected);
     }
 
+    #[tokio::test]
+    async fn test_take_offsets_preserves_repeated_rows_in_permutation() {
+        let base_table = lance_datagen::gen_batch()
+            .col("idx", lance_datagen::array::step::<Int32Type>())
+            .into_mem_table("tbl", RowCount::from(5), BatchCount::from(1))
+            .await;
+        let base_row_ids = collect_column::<UInt64Type>(&base_table, "_rowid").await;
+        let permutation_row_ids = vec![
+            base_row_ids[3],
+            base_row_ids[1],
+            base_row_ids[3],
+            base_row_ids[2],
+        ];
+        let permutation_batch = RecordBatch::try_new(
+            Arc::new(Schema::new(vec![
+                Field::new("row_id", DataType::UInt64, false),
+                Field::new(SPLIT_ID_COLUMN, DataType::UInt64, false),
+            ])),
+            vec![
+                Arc::new(UInt64Array::from(permutation_row_ids)),
+                Arc::new(UInt64Array::from(vec![0; 4])),
+            ],
+        )
+        .unwrap();
+        let permutation_table = virtual_table("row_ids", &permutation_batch).await;
+        let reader = PermutationReader::try_from_tables(
+            base_table.base_table().clone(),
+            permutation_table.base_table().clone(),
+            0,
+        )
+        .await
+        .unwrap();
+
+        let batch = reader
+            .take_offsets(&[0, 1, 2, 3], Select::All)
+            .await
+            .unwrap();
+        let idx_values = batch
+            .column(0)
+            .as_primitive::<Int32Type>()
+            .values()
+            .to_vec();
+
+        assert_eq!(idx_values, vec![3, 1, 3, 2]);
+    }
+
     #[tokio::test]
     async fn test_take_offsets_with_column_selection() {
         let (base_table, row_ids_table, row_ids) = setup_permutation_tables(10).await;
@@ -883,17 +937,17 @@ mod tests {
             .unwrap();
 
         // With no permutation table, take_offsets uses the base table directly
-        let offsets = vec![0, 2, 4, 6];
+        let offsets = vec![0, 2, 0, 4, 6];
         let batch = reader.take_offsets(&offsets, Select::All).await.unwrap();
 
-        assert_eq!(batch.num_rows(), 4);
+        assert_eq!(batch.num_rows(), 5);
 
         let idx_values = batch
             .column(0)
             .as_primitive::<Int32Type>()
             .values()
             .to_vec();
-        assert_eq!(idx_values, vec![0, 2, 4, 6]);
+        assert_eq!(idx_values, vec![0, 2, 0, 4, 6]);
     }
 
     #[tokio::test]

rust/lancedb/src/error.rs

@@ -102,6 +102,8 @@ pub enum Error {
     },
     #[snafu(display("Job{} was cancelled", job_id.as_ref().map(|id| format!(" {id}")).unwrap_or_default()))]
     JobCancelled { job_id: Option<String> },
+    #[snafu(display("Job '{job_id}' was not found"))]
+    JobNotFound { job_id: String },
 
     // 3rd party / external errors
     #[snafu(display("object_store error: {source}"))]