Bump version: 0.38.0-beta.2 → 0.38.0-beta.3

2026-08-28 17:08:43 +00:00 · 2026-08-21 10:25:20 +00:00
181 changed files with 3179 additions and 23432 deletions
@@ -1,5 +1,5 @@
 [tool.bumpversion]
-current_version = "0.38.0-beta.11"
+current_version = "0.38.0-beta.3"
 parse = """(?x)
    (?P<major>0|[1-9]\\d*)\\.
    (?P<minor>0|[1-9]\\d*)\\.
@@ -17,18 +17,6 @@ updates:
    # newer minimum versions.
    versioning-strategy: lockfile-only
    groups:
-      # The arrow-rs and datafusion crates are released in lockstep and have to
-      # move together, so keep them in one PR instead of one per sub-crate.
-      # Listed first: a dependency joins the first group it matches.
-      arrow-datafusion:
-        patterns:
-          - arrow
-          - arrow-*
-          - parquet
-          - parquet-*
-          - datafusion
-          - datafusion-*
-          - object_store
      rust-minor-patch:
        update-types:
          - minor
@@ -129,12 +129,6 @@ jobs:
      # link.exe is single-threaded and the long pole on Windows builds. Use
      # rustc's bundled lld-link instead.
      CARGO_TARGET_X86_64_PC_WINDOWS_MSVC_LINKER: rust-lld
-      # Fat LTO of the cdylib is single-threaded and the peak-memory step of the
-      # build. ThinLTO parallelizes it across the runner's cores, at some cost
-      # to runtime performance on our least performance-sensitive platform.
-      # Matches what the nodejs Windows builds already do in npm-publish.yml.
-      CARGO_PROFILE_RELEASE_LTO: thin
-      CARGO_PROFILE_RELEASE_CODEGEN_UNITS: 16
    steps:
      - uses: actions/checkout@v6
        with:
@@ -3455,8 +3455,8 @@ checksum = "42703706b716c37f96a77aea830392ad231f44c9e9a67872fa5548707e11b11c"

 [[package]]
 name = "fsst"
-version = "12.0.0-beta.2"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+version = "11.0.0-beta.16"
+source = "git+https://github.com/lance-format/lance.git?tag=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "arrow-array",
 "rand 0.9.5",
@@ -4815,8 +4815,8 @@ checksum = "e037a2e1d8d5fdbd49b16a4ea09d5d6401c1f29eca5ff29d03d3824dba16256a"

 [[package]]
 name = "lance"
-version = "12.0.0-beta.2"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+version = "11.0.0-beta.16"
+source = "git+https://github.com/lance-format/lance.git?tag=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "arc-swap",
 "arrow",
@@ -4888,8 +4888,8 @@ dependencies = [

 [[package]]
 name = "lance-arrow"
-version = "12.0.0-beta.2"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+version = "11.0.0-beta.16"
+source = "git+https://github.com/lance-format/lance.git?tag=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "arrow-array",
 "arrow-buffer",
@@ -4911,7 +4911,7 @@ 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=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "arrow-array",
 "arrow-buffer",
@@ -4925,7 +4925,7 @@ 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=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "arrow-array",
 "arrow-schema",
@@ -4934,8 +4934,8 @@ dependencies = [

 [[package]]
 name = "lance-bitpacking"
-version = "12.0.0-beta.2"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+version = "11.0.0-beta.16"
+source = "git+https://github.com/lance-format/lance.git?tag=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "arrayref",
 "crunchy",
@@ -4945,8 +4945,8 @@ dependencies = [

 [[package]]
 name = "lance-core"
-version = "12.0.0-beta.2"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+version = "11.0.0-beta.16"
+source = "git+https://github.com/lance-format/lance.git?tag=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "arrow-array",
 "arrow-buffer",
@@ -4983,8 +4983,8 @@ dependencies = [

 [[package]]
 name = "lance-datafusion"
-version = "12.0.0-beta.2"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+version = "11.0.0-beta.16"
+source = "git+https://github.com/lance-format/lance.git?tag=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "arrow",
 "arrow-array",
@@ -5013,8 +5013,8 @@ dependencies = [

 [[package]]
 name = "lance-datagen"
-version = "12.0.0-beta.2"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+version = "11.0.0-beta.16"
+source = "git+https://github.com/lance-format/lance.git?tag=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "arrow",
 "arrow-array",
@@ -5031,8 +5031,8 @@ dependencies = [

 [[package]]
 name = "lance-derive"
-version = "12.0.0-beta.2"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+version = "11.0.0-beta.16"
+source = "git+https://github.com/lance-format/lance.git?tag=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "proc-macro2",
 "quote",
@@ -5041,8 +5041,8 @@ dependencies = [

 [[package]]
 name = "lance-encoding"
-version = "12.0.0-beta.2"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+version = "11.0.0-beta.16"
+source = "git+https://github.com/lance-format/lance.git?tag=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "arrow-arith",
 "arrow-array",
@@ -5075,8 +5075,8 @@ dependencies = [

 [[package]]
 name = "lance-file"
-version = "12.0.0-beta.2"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+version = "11.0.0-beta.16"
+source = "git+https://github.com/lance-format/lance.git?tag=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "arrow-arith",
 "arrow-array",
@@ -5107,8 +5107,8 @@ dependencies = [

 [[package]]
 name = "lance-index"
-version = "12.0.0-beta.2"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+version = "11.0.0-beta.16"
+source = "git+https://github.com/lance-format/lance.git?tag=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "arc-swap",
 "arrow",
@@ -5172,8 +5172,8 @@ dependencies = [

 [[package]]
 name = "lance-index-core"
-version = "12.0.0-beta.2"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+version = "11.0.0-beta.16"
+source = "git+https://github.com/lance-format/lance.git?tag=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "arrow-array",
 "arrow-schema",
@@ -5195,8 +5195,8 @@ dependencies = [

 [[package]]
 name = "lance-io"
-version = "12.0.0-beta.2"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+version = "11.0.0-beta.16"
+source = "git+https://github.com/lance-format/lance.git?tag=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "arrow",
 "arrow-array",
@@ -5222,11 +5222,7 @@ dependencies = [
 "pin-project",
 "prost",
 "rand 0.9.5",
- "reqsign-core",
- "reqsign-file-read-tokio",
- "reqsign-google",
 "serde",
- "serde_json",
 "tempfile",
 "tokio",
 "tracing",
@@ -5236,8 +5232,8 @@ dependencies = [

 [[package]]
 name = "lance-linalg"
-version = "12.0.0-beta.2"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+version = "11.0.0-beta.16"
+source = "git+https://github.com/lance-format/lance.git?tag=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "arrow-array",
 "arrow-schema",
@@ -5251,8 +5247,8 @@ dependencies = [

 [[package]]
 name = "lance-namespace"
-version = "12.0.0-beta.2"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+version = "11.0.0-beta.16"
+source = "git+https://github.com/lance-format/lance.git?tag=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "arrow",
 "async-trait",
@@ -5264,8 +5260,8 @@ dependencies = [

 [[package]]
 name = "lance-namespace-impls"
-version = "12.0.0-beta.2"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+version = "11.0.0-beta.16"
+source = "git+https://github.com/lance-format/lance.git?tag=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "arrow",
 "arrow-ipc",
@@ -5318,8 +5314,8 @@ dependencies = [

 [[package]]
 name = "lance-select"
-version = "12.0.0-beta.2"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+version = "11.0.0-beta.16"
+source = "git+https://github.com/lance-format/lance.git?tag=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "arrow-array",
 "arrow-buffer",
@@ -5333,8 +5329,8 @@ dependencies = [

 [[package]]
 name = "lance-table"
-version = "12.0.0-beta.2"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+version = "11.0.0-beta.16"
+source = "git+https://github.com/lance-format/lance.git?tag=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "arrow",
 "arrow-array",
@@ -5374,8 +5370,8 @@ dependencies = [

 [[package]]
 name = "lance-testing"
-version = "12.0.0-beta.2"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+version = "11.0.0-beta.16"
+source = "git+https://github.com/lance-format/lance.git?tag=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "arrow-array",
 "arrow-schema",
@@ -5388,8 +5384,8 @@ dependencies = [

 [[package]]
 name = "lance-tokenizer"
-version = "12.0.0-beta.2"
-source = "git+https://github.com/lance-format/lance.git?tag=v12.0.0-beta.2#dafa4642658d996b3e31dde91e02f72db7860d7e"
+version = "11.0.0-beta.16"
+source = "git+https://github.com/lance-format/lance.git?tag=v11.0.0-beta.16#d514a61fa7e449dedcd03571eabd338b3c84b170"
 dependencies = [
 "frostem",
 "icu_segmenter",
@@ -5402,7 +5398,7 @@ dependencies = [

 [[package]]
 name = "lancedb"
-version = "0.38.0-beta.11"
+version = "0.38.0-beta.2"
 dependencies = [
 "ahash",
 "anyhow",
@@ -5490,7 +5486,7 @@ dependencies = [

 [[package]]
 name = "lancedb-nodejs"
-version = "0.38.0-beta.11"
+version = "0.38.0-beta.2"
 dependencies = [
 "arrow-array",
 "arrow-buffer",
@@ -5515,7 +5511,7 @@ dependencies = [

 [[package]]
 name = "lancedb-python"
-version = "0.38.0-beta.11"
+version = "0.38.0-beta.2"
 dependencies = [
 "arrow",
 "async-trait",
@@ -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-core = { "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.2", "tag" = "v12.0.0-beta.2", "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-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-index = { "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.2", "tag" = "v12.0.0-beta.2", "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-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-table = { "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.2", "tag" = "v12.0.0-beta.2", "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-encoding = { "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.2", "tag" = "v12.0.0-beta.2", "git" = "https://github.com/lance-format/lance.git" }
+lance = { "version" = "=11.0.0-beta.16", default-features = false, "tag" = "v11.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
+lance-core = { "version" = "=11.0.0-beta.16", "tag" = "v11.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
+lance-datagen = { "version" = "=11.0.0-beta.16", "tag" = "v11.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
+lance-file = { "version" = "=11.0.0-beta.16", "tag" = "v11.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
+lance-io = { "version" = "=11.0.0-beta.16", default-features = false, "tag" = "v11.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
+lance-index = { "version" = "=11.0.0-beta.16", "tag" = "v11.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
+lance-linalg = { "version" = "=11.0.0-beta.16", "tag" = "v11.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
+lance-namespace = { "version" = "=11.0.0-beta.16", "tag" = "v11.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
+lance-namespace-impls = { "version" = "=11.0.0-beta.16", default-features = false, "tag" = "v11.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
+lance-table = { "version" = "=11.0.0-beta.16", "tag" = "v11.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
+lance-testing = { "version" = "=11.0.0-beta.16", "tag" = "v11.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
+lance-datafusion = { "version" = "=11.0.0-beta.16", "tag" = "v11.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
+lance-encoding = { "version" = "=11.0.0-beta.16", "tag" = "v11.0.0-beta.16", "git" = "https://github.com/lance-format/lance.git" }
+lance-arrow = { "version" = "=11.0.0-beta.16", "tag" = "v11.0.0-beta.16", "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
@@ -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>0.38.0-beta.3</version>
 </dependency>
 ```

@@ -1,518 +0,0 @@
-[**@lancedb/lancedb**](../README.md) • **Docs**
-
-***
-
-[@lancedb/lancedb](../globals.md) / AutoQuery
-
-# Class: AutoQuery
-
-A builder for automatic string searches.
-
-Automatic search determines whether to use full-text or vector search from
-the table revision selected for each execution. This builder exposes the
-common operations supported by both query families.
-
-## Extends
-
- `StandardQueryBase`&lt;`NativeQuery` \| `NativeVectorQuery`&gt;
-
-## Properties
-
-### inner
-
-```ts
-protected inner: Query | VectorQuery | Promise<Query | VectorQuery>;
-```
-
-#### Inherited from
-
-`StandardQueryBase.inner`
-
-## Methods
-
-### analyzePlan()
-
-```ts
-analyzePlan(distributedMetrics?): Promise<string>
-```
-
-Executes the query and returns the physical query plan annotated with runtime metrics.
-
-This is useful for debugging and performance analysis, as it shows how the query was executed
-and includes metrics such as elapsed time, rows processed, and I/O statistics.
-
-#### Parameters
-
-* **distributedMetrics?**: [`AnalyzePlanDistributedMetrics`](../type-aliases/AnalyzePlanDistributedMetrics.md)
-    How distributed worker metrics are displayed for remote query plans.
-    Defaults to `"aggregate"`.
-
-#### Returns
-
-`Promise`&lt;`string`&gt;
-
-A query execution plan with runtime metrics for each step.
-
-#### Example
-
-```ts
-import * as lancedb from "@lancedb/lancedb"
-
-const db = await lancedb.connect("./.lancedb");
-const table = await db.createTable("my_table", [
-  { vector: [1.1, 0.9], id: "1" },
-]);
-
-const plan = await table.query().nearestTo([0.5, 0.2]).analyzePlan();
-
-Example output (with runtime metrics inlined):
-AnalyzeExec verbose=true, metrics=[]
- ProjectionExec: expr=[id@3 as id, vector@0 as vector, _distance@2 as _distance], metrics=[output_rows=1, elapsed_compute=3.292µs]
-  Take: columns="vector, _rowid, _distance, (id)", metrics=[output_rows=1, elapsed_compute=66.001µs, batches_processed=1, bytes_read=8, iops=1, requests=1]
-   CoalesceBatchesExec: target_batch_size=1024, metrics=[output_rows=1, elapsed_compute=3.333µs]
-    GlobalLimitExec: skip=0, fetch=10, metrics=[output_rows=1, elapsed_compute=167ns]
-     FilterExec: _distance@2 IS NOT NULL, metrics=[output_rows=1, elapsed_compute=8.542µs]
-      SortExec: TopK(fetch=10), expr=[_distance@2 ASC NULLS LAST], metrics=[output_rows=1, elapsed_compute=63.25µs, row_replacements=1]
-       KNNVectorDistance: metric=l2, metrics=[output_rows=1, elapsed_compute=114.333µs, output_batches=1]
-        LanceScan: uri=/path/to/data, projection=[vector], row_id=true, row_addr=false, ordered=false, metrics=[output_rows=1, elapsed_compute=103.626µs, bytes_read=549, iops=2, requests=2]
-```
-
-#### Inherited from
-
-`StandardQueryBase.analyzePlan`
-
-***
-
-### execute()
-
-```ts
-protected execute(options?): AsyncGenerator<RecordBatch<any>, void, unknown>
-```
-
-Execute the query and return the results as an
-
-#### Parameters
-
-* **options?**: `Partial`&lt;[`QueryExecutionOptions`](../interfaces/QueryExecutionOptions.md)&gt;
-
-#### Returns
-
-`AsyncGenerator`&lt;`RecordBatch`&lt;`any`&gt;, `void`, `unknown`&gt;
-
-#### See
-
- - AsyncIterator
-of
- - RecordBatch.
-
-By default, LanceDb will use many threads to calculate results and, when
-the result set is large, multiple batches will be processed at one time.
-This readahead is limited however and backpressure will be applied if this
-stream is consumed slowly (this constrains the maximum memory used by a
-single query)
-
-#### Inherited from
-
-`StandardQueryBase.execute`
-
-***
-
-### explainPlan()
-
-```ts
-explainPlan(verbose): Promise<string>
-```
-
-Generates an explanation of the query execution plan.
-
-#### Parameters
-
-* **verbose**: `boolean` = `false`
-    If true, provides a more detailed explanation. Defaults to false.
-
-#### Returns
-
-`Promise`&lt;`string`&gt;
-
-A Promise that resolves to a string containing the query execution plan explanation.
-
-#### Example
-
-```ts
-import * as lancedb from "@lancedb/lancedb"
-const db = await lancedb.connect("./.lancedb");
-const table = await db.createTable("my_table", [
-  { vector: [1.1, 0.9], id: "1" },
-]);
-const plan = await table.query().nearestTo([0.5, 0.2]).explainPlan();
-```
-
-#### Inherited from
-
-`StandardQueryBase.explainPlan`
-
-***
-
-### fastSearch()
-
-```ts
-fastSearch(): this
-```
-
-Skip searching un-indexed data. This can make search faster, but will miss
-any data that is not yet indexed.
-
-Use [Table#optimize](Table.md#optimize) to index all un-indexed data.
-
-#### Returns
-
-`this`
-
-#### Inherited from
-
-`StandardQueryBase.fastSearch`
-
-***
-
-### ~~filter()~~
-
-```ts
-filter(predicate): this
-```
-
-A filter statement to be applied to this query.
-
-#### Parameters
-
-* **predicate**: `string`
-
-#### Returns
-
-`this`
-
-#### See
-
-where
-
-#### Deprecated
-
-Use `where` instead
-
-#### Inherited from
-
-`StandardQueryBase.filter`
-
-***
-
-### fullTextSearch()
-
-```ts
-fullTextSearch(query, options?): this
-```
-
-#### Parameters
-
-* **query**: `string` \| [`FullTextQuery`](../interfaces/FullTextQuery.md)
-
-* **options?**: `Partial`&lt;[`FullTextSearchOptions`](../interfaces/FullTextSearchOptions.md)&gt;
-
-#### Returns
-
-`this`
-
-#### Inherited from
-
-`StandardQueryBase.fullTextSearch`
-
-***
-
-### limit()
-
-```ts
-limit(limit): this
-```
-
-Set the maximum number of results to return.
-
-By default, a plain search has no limit.  If this method is not
-called then every valid row from the table will be returned.
-
-#### Parameters
-
-* **limit**: `number`
-
-#### Returns
-
-`this`
-
-#### Inherited from
-
-`StandardQueryBase.limit`
-
-***
-
-### offset()
-
-```ts
-offset(offset): this
-```
-
-Set the number of rows to skip before returning results.
-
-This is useful for pagination.
-
-#### Parameters
-
-* **offset**: `number`
-
-#### Returns
-
-`this`
-
-#### Inherited from
-
-`StandardQueryBase.offset`
-
-***
-
-### orderBy()
-
-```ts
-orderBy(ordering): this
-```
-
-Sort the results by the specified column(s).
-
-#### Parameters
-
-* **ordering**: [`ColumnOrdering`](../interfaces/ColumnOrdering.md) \| [`ColumnOrdering`](../interfaces/ColumnOrdering.md)[]
-
-#### Returns
-
-`this`
-
-This query builder.
-
-#### Inherited from
-
-`StandardQueryBase.orderBy`
-
-***
-
-### outputSchema()
-
-```ts
-outputSchema(): Promise<Schema<any>>
-```
-
-Returns the schema of the output that will be returned by this query.
-
-This can be used to inspect the types and names of the columns that will be
-returned by the query before executing it.
-
-#### Returns
-
-`Promise`&lt;`Schema`&lt;`any`&gt;&gt;
-
-An Arrow Schema describing the output columns.
-
-#### Inherited from
-
-`StandardQueryBase.outputSchema`
-
-***
-
-### select()
-
-```ts
-select(columns): this
-```
-
-Return only the specified columns.
-
-By default a query will return all columns from the table.  However, this can have
-a very significant impact on latency.  LanceDb stores data in a columnar fashion.  This
-means we can finely tune our I/O to select exactly the columns we need.
-
-As a best practice you should always limit queries to the columns that you need.  If you
-pass in an array of column names then only those columns will be returned.
-
-You can also use this method to create new "dynamic" columns based on your existing columns.
-For example, you may not care about "a" or "b" but instead simply want "a + b".  This is often
-seen in the SELECT clause of an SQL query (e.g. `SELECT a+b FROM my_table`).
-
-To create dynamic columns you can pass in a Map<string, string>.  A column will be returned
-for each entry in the map.  The key provides the name of the column.  The value is
-an SQL string used to specify how the column is calculated.
-
-For example, an SQL query might state `SELECT a + b AS combined, c`.  The equivalent
-input to this method would be:
-
-#### Parameters
-
-* **columns**: `string` \| `string`[] \| `Record`&lt;`string`, `string`&gt; \| `Map`&lt;`string`, `string`&gt;
-
-#### Returns
-
-`this`
-
-#### Example
-
-```ts
-new Map([["combined", "a + b"], ["c", "c"]])
-
-Columns will always be returned in the order given, even if that order is different than
-the order used when adding the data.
-
-Note that you can pass in a `Record<string, string>` (e.g. an object literal). This method
-uses `Object.entries` which should preserve the insertion order of the object.  However,
-object insertion order is easy to get wrong and `Map` is more foolproof.
-```
-
-#### Inherited from
-
-`StandardQueryBase.select`
-
-***
-
-### toArray()
-
-```ts
-toArray(options?): Promise<any[]>
-```
-
-Collect the results as an array of objects.
-
-#### Parameters
-
-* **options?**: `Partial`&lt;[`QueryExecutionOptions`](../interfaces/QueryExecutionOptions.md)&gt;
-
-#### Returns
-
-`Promise`&lt;`any`[]&gt;
-
-#### Inherited from
-
-`StandardQueryBase.toArray`
-
-***
-
-### toArrow()
-
-```ts
-toArrow(options?): Promise<Table<any>>
-```
-
-Collect the results as an Arrow
-
-#### Parameters
-
-* **options?**: `Partial`&lt;[`QueryExecutionOptions`](../interfaces/QueryExecutionOptions.md)&gt;
-
-#### Returns
-
-`Promise`&lt;`Table`&lt;`any`&gt;&gt;
-
-#### See
-
-ArrowTable.
-
-#### Inherited from
-
-`StandardQueryBase.toArrow`
-
-***
-
-### useLsm()
-
-```ts
-useLsm(enable): this
-```
-
-Control MemWAL read routing for this query.
-
-By default (unset), when the table carries a MemWAL write spec (see
-[Table#setLsmWriteSpec](Table.md#setlsmwritespec)), reads are routed through the LSM scanner so
-they also return data written via the `mergeInsert` LSM path that has not yet
-been compacted into the base table (the active/frozen in-memory memtables and
-the flushed generations), deduplicated by primary key; a table without a spec
-reads the base table.
-
-#### Parameters
-
-* **enable**: `boolean`
-    `true` forces the LSM scanner and errors if the table has no
-    MemWAL write spec. `false` bypasses the MemWAL and reads the base table only,
-    even when a spec is present.
-    Note: the LSM scanner does not support every query shape (e.g. reranking,
-    hybrid search, `orderBy`). On a MemWAL table those shapes error unless
-    `useLsm(false)` is set, because a base-only read would silently exclude
-    un-compacted MemWAL data.
-
-#### Returns
-
-`this`
-
-#### Inherited from
-
-`StandardQueryBase.useLsm`
-
-***
-
-### where()
-
-```ts
-where(predicate): this
-```
-
-A filter statement to be applied to this query.
-
-The filter should be supplied as an SQL query string.  For example:
-
-#### Parameters
-
-* **predicate**: `string`
-
-#### Returns
-
-`this`
-
-#### Example
-
-```ts
-x > 10
-y > 0 AND y < 100
-x > 5 OR y = 'test'
-
-Filtering performance can often be improved by creating a scalar index
-on the filter column(s).
-
-Calling this multiple times combines the filters with a logical AND rather
-than replacing the previous filter.
-```
-
-#### Inherited from
-
-`StandardQueryBase.where`
-
-***
-
-### withRowId()
-
-```ts
-withRowId(): this
-```
-
-Whether to return the row id in the results.
-
-This column can be used to match results between different queries. For
-example, to match results from a full text search and a vector search in
-order to perform hybrid search.
-
-#### Returns
-
-`this`
-
-#### Inherited from
-
-`StandardQueryBase.withRowId`
@@ -37,31 +37,6 @@ latest and stays writable.

 ***

-### cherryPick()
-
-```ts
-cherryPick(fromBranch, dryRun): Promise<CherryPickResult>
-```
-
-Cherry-pick a branch onto main.
-
-Set `dryRun` to `true` to preview. A failed cherry-pick resolves
-with `status: "failed"` instead of throwing.
-
-#### Parameters
-
-* **fromBranch**: `string`
-    Branch to cherry-pick from.
-
-* **dryRun**: `boolean` = `false`
-    When true, only preview. Defaults to false.
-
-#### Returns
-
-`Promise`&lt;[`CherryPickResult`](../interfaces/CherryPickResult.md)&gt;
-
-***
-
 ### create()

 ```ts
@@ -137,3 +112,28 @@ List all branches, mapping name to branch metadata.
 #### Returns

 `Promise`&lt;`Record`&lt;`string`, [`BranchContents`](BranchContents.md)&gt;&gt;
+
+***
+
+### merge()
+
+```ts
+merge(fromBranch, dryRun): Promise<MergeBranchResult>
+```
+
+Merge a branch into main.
+
+Set `dryRun` to `true` to preview the merge. A rejected merge resolves
+with `status: "rejected"` instead of throwing.
+
+#### Parameters
+
+* **fromBranch**: `string`
+    Branch to merge from.
+
+* **dryRun**: `boolean` = `false`
+    When true, only preview the merge. Defaults to false.
+
+#### Returns
+
+`Promise`&lt;[`MergeBranchResult`](../interfaces/MergeBranchResult.md)&gt;
@@ -169,45 +169,6 @@ Creates a new empty Table

 ***

-### createMaterializedView()
-
-```ts
-abstract createMaterializedView(
-   name,
-   source,
-   options?): Promise<MaterializedView>
-```
-
-Define a materialized view named `name` over the table `source`.
-
-The view is created empty, with the query recorded in its schema
-metadata; `view.refresh()` computes the rows. The view is a normal
-table: it can be queried, indexed and searched, and it appears in
-`tableNames`. The source table must have stable row ids (create it with
-the `newTableEnableStableRowIds` storage option); they keep the view's
-provenance valid across source compactions and cannot be enabled after
-a table exists. Local databases only.
-
-#### Parameters
-
-* **name**: `string`
-
-* **source**: `string`
-
-* **options?**
-
-* **options.limit?**: `number`
-
-* **options.select?**: [`MaterializedViewSelect`](../type-aliases/MaterializedViewSelect.md)
-
-* **options.where?**: `string`
-
-#### Returns
-
-`Promise`&lt;[`MaterializedView`](MaterializedView.md)&gt;
-
-***
-
 ### createNamespace()

 ```ts
@@ -538,22 +499,6 @@ List server-side jobs across the database's tables.

 ***

-### listMaterializedViews()
-
-```ts
-abstract listMaterializedViews(): Promise<string[]>
-```
-
-The names of the materialized views in this database.
-
-Found by reading every table's schema, so this costs an open per table.
-
-#### Returns
-
-`Promise`&lt;`string`[]&gt;
-
-***
-
 ### listNamespaces()

 ```ts
@@ -584,90 +529,6 @@ Child namespace names and

 ***

-### listTables()
-
-#### listTables(options)
-
-```ts
-abstract listTables(options?): Promise<ListTablesResponse>
-```
-
-List a page of the tables in this database.
-
-To retrieve the tables after the page, pass the `pageToken` the response
-carries back in. A page can be shorter than `limit` without being the last
-one, so walk until a response carries no page token:
-
-```ts
-const names = [];
-let pageToken = undefined;
-do {
-  const page = await conn.listTables({ pageToken, limit: 100 });
-  names.push(...page.tables);
-  pageToken = page.pageToken;
-} while (pageToken);
-```
-
-##### Parameters
-
-* **options?**: `Partial`&lt;[`ListTablesOptions`](../interfaces/ListTablesOptions.md)&gt;
-    Pagination options
-    (`pageToken`, `limit`).
-
-##### Returns
-
-`Promise`&lt;[`ListTablesResponse`](../interfaces/ListTablesResponse.md)&gt;
-
-A page of table names and an
-  optional token for the tables after it.
-
-#### listTables(namespacePath, options)
-
-```ts
-abstract listTables(namespacePath?, options?): Promise<ListTablesResponse>
-```
-
-List a page of the tables in this database.
-
-##### Parameters
-
-* **namespacePath?**: `string`[]
-    The namespace path to list tables from
-    (defaults to root namespace)
-
-* **options?**: `Partial`&lt;[`ListTablesOptions`](../interfaces/ListTablesOptions.md)&gt;
-    Pagination options
-    (`pageToken`, `limit`).
-
-##### Returns
-
-`Promise`&lt;[`ListTablesResponse`](../interfaces/ListTablesResponse.md)&gt;
-
-A page of table names and an
-  optional token for the tables after it.
-
-***
-
-### openMaterializedView()
-
-```ts
-abstract openMaterializedView(name): Promise<MaterializedView>
-```
-
-Open the materialized view named `name`.
-
-Rejects a table that exists but is not a materialized view.
-
-#### Parameters
-
-* **name**: `string`
-
-#### Returns
-
-`Promise`&lt;[`MaterializedView`](MaterializedView.md)&gt;
-
-***
-
 ### openTable()

 ```ts
@@ -677,13 +538,18 @@ abstract openTable(
   options?): Promise<Table>
 ```

+Open a table in the database.
+
 #### Parameters

 * **name**: `string`
+    The name of the table

 * **namespacePath?**: `string`[]
+    The namespace path of the table (defaults to root namespace)

 * **options?**: `Partial`&lt;[`OpenTableOptions`](../interfaces/OpenTableOptions.md)&gt;
+    Additional options

 #### Returns

@@ -724,7 +590,7 @@ a "not supported" error.

 ***

-### ~~tableNames()~~
+### tableNames()

 #### tableNames(options)

@@ -746,10 +612,6 @@ Tables will be returned in lexicographical order.

 `Promise`&lt;`string`[]&gt;

-##### Deprecated
-
-Use [Connection.listTables](Connection.md#listtables) instead.
-
 #### tableNames(namespacePath, options)

 ```ts
@@ -772,7 +634,3 @@ Tables will be returned in lexicographical order.
 ##### Returns

 `Promise`&lt;`string`[]&gt;
-
-##### Deprecated
-
-Use [Connection.listTables](Connection.md#listtables) instead.
@@ -1,101 +0,0 @@
-[**@lancedb/lancedb**](../README.md) • **Docs**
-
-***
-
-[@lancedb/lancedb](../globals.md) / MaterializedView
-
-# Class: MaterializedView
-
-A handle on a materialized view: its table plus its definition.
-
-Obtained from [Connection#createMaterializedView](Connection.md#creatematerializedview) or
-[Connection#openMaterializedView](Connection.md#openmaterializedview). The view is a normal table --
-queries, indexes and search all apply through [MaterializedView#table](MaterializedView.md#table)
-- whose contents are maintained by [MaterializedView#refresh](MaterializedView.md#refresh).
-
-## Constructors
-
-### new MaterializedView()
-
-```ts
-new MaterializedView(table): MaterializedView
-```
-
-#### Parameters
-
-* **table**: [`Table`](Table.md)
-
-#### Returns
-
-[`MaterializedView`](MaterializedView.md)
-
-## Accessors
-
-### name
-
-```ts
-get name(): string
-```
-
-#### Returns
-
-`string`
-
-## Methods
-
-### definition()
-
-```ts
-definition(): Promise<MaterializedViewDefinition>
-```
-
-The query that defines the view, read from its stored schema.
-
-#### Returns
-
-`Promise`&lt;[`MaterializedViewDefinition`](../interfaces/MaterializedViewDefinition.md)&gt;
-
-***
-
-### refresh()
-
-```ts
-refresh(options?): Promise<RefreshMaterializedViewResult>
-```
-
-Recompute the view from its source.
-
-The refresh is incremental when the source's changes can be reconciled
-into the view -- rows added, changed or removed since the last one --
-and otherwise rebuilds. `full` forces a rebuild; `sourceVersion`
-refreshes to that source version instead of the latest.
-
-Concurrent refreshes of one view do not duplicate its rows. Two that
-plan the same source rows conflict on commit, and the loser throws
-rather than writing them a second time.
-
-#### Parameters
-
-* **options?**
-
-* **options.full?**: `boolean`
-
-* **options.sourceVersion?**: `number`
-
-#### Returns
-
-`Promise`&lt;[`RefreshMaterializedViewResult`](../interfaces/RefreshMaterializedViewResult.md)&gt;
-
-***
-
-### table()
-
-```ts
-table(): Table
-```
-
-The view, as the table it is.
-
-#### Returns
-
-[`Table`](Table.md)
@@ -942,7 +942,7 @@ Get the schema of the table.
 abstract search(
   query,
   queryType?,
-   ftsColumns?): Query | VectorQuery | AutoQuery
+   ftsColumns?): Query | VectorQuery
 ```

 Create a search query to find the nearest neighbors
@@ -964,7 +964,7 @@ of the given query

 #### Returns

-[`Query`](Query.md) \| [`VectorQuery`](VectorQuery.md) \| [`AutoQuery`](AutoQuery.md)
+[`Query`](Query.md) \| [`VectorQuery`](VectorQuery.md)

 ***

@@ -1292,18 +1292,6 @@ abstract updateFieldMetadata(updates): Promise<UpdateFieldMetadataResult>

 Update per-field (column) metadata.

-The following keys are treated specially, by convention, and should be
-used when appropriate:
-
- `lancedb:description`: for a human-readable description of a field.
- `lancedb:tag:<name>`: for a user-defined key-value tag, where the suffix
-  names the tag category; e.g. `lancedb:tag:model: "clip"`.
- `lancedb:logical-column`: for a column grouping; e.g. `feature_v1` and
-  `feature_v2` might be in the same logical column.
- `lancedb:status`: for status options (`production`, `candidate`,
-  `deprecated`, `archived`) to designate the current life cycle state of
-  this column.
-
 #### Parameters

 * **updates**: [`FieldMetadataUpdate`](../interfaces/FieldMetadataUpdate.md)[]
@@ -18,7 +18,6 @@

 ## Classes

- [AutoQuery](classes/AutoQuery.md)
 - [BooleanQuery](classes/BooleanQuery.md)
 - [BoostQuery](classes/BoostQuery.md)
 - [BranchContents](classes/BranchContents.md)
@@ -29,7 +28,6 @@
 - [Job](classes/Job.md)
 - [MakeArrowTableOptions](classes/MakeArrowTableOptions.md)
 - [MatchQuery](classes/MatchQuery.md)
- [MaterializedView](classes/MaterializedView.md)
 - [MergeInsertBuilder](classes/MergeInsertBuilder.md)
 - [MultiMatchQuery](classes/MultiMatchQuery.md)
 - [NativeJsHeaderProvider](classes/NativeJsHeaderProvider.md)
@@ -61,9 +59,6 @@
 - [BranchIndexSummary](interfaces/BranchIndexSummary.md)
 - [BranchRowCountSummary](interfaces/BranchRowCountSummary.md)
 - [BucketStats](interfaces/BucketStats.md)
- [CherryPickError](interfaces/CherryPickError.md)
- [CherryPickPreview](interfaces/CherryPickPreview.md)
- [CherryPickResult](interfaces/CherryPickResult.md)
 - [ClientConfig](interfaces/ClientConfig.md)
 - [ColumnAlteration](interfaces/ColumnAlteration.md)
 - [ColumnOrdering](interfaces/ColumnOrdering.md)
@@ -101,12 +96,12 @@
 - [JobInfo](interfaces/JobInfo.md)
 - [ListNamespacesOptions](interfaces/ListNamespacesOptions.md)
 - [ListNamespacesResponse](interfaces/ListNamespacesResponse.md)
- [ListTablesOptions](interfaces/ListTablesOptions.md)
- [ListTablesResponse](interfaces/ListTablesResponse.md)
 - [LsmStats](interfaces/LsmStats.md)
 - [LsmWriteSpec](interfaces/LsmWriteSpec.md)
- [MaterializedViewDefinition](interfaces/MaterializedViewDefinition.md)
 - [MemtableStats](interfaces/MemtableStats.md)
+- [MergeBlocker](interfaces/MergeBlocker.md)
+- [MergeBranchResult](interfaces/MergeBranchResult.md)
+- [MergePreview](interfaces/MergePreview.md)
 - [MergeResult](interfaces/MergeResult.md)
 - [NativeOAuthConfig](interfaces/NativeOAuthConfig.md)
 - [OAuthConfig](interfaces/OAuthConfig.md)
@@ -115,7 +110,6 @@
 - [OptimizeStats](interfaces/OptimizeStats.md)
 - [QueryExecutionOptions](interfaces/QueryExecutionOptions.md)
 - [RefreshColumnResult](interfaces/RefreshColumnResult.md)
- [RefreshMaterializedViewResult](interfaces/RefreshMaterializedViewResult.md)
 - [RemovalStats](interfaces/RemovalStats.md)
 - [RenameTableOptions](interfaces/RenameTableOptions.md)
 - [RestNamespaceConfig](interfaces/RestNamespaceConfig.md)
@@ -148,7 +142,6 @@
 - [FieldLike](type-aliases/FieldLike.md)
 - [IntoSql](type-aliases/IntoSql.md)
 - [IntoVector](type-aliases/IntoVector.md)
- [MaterializedViewSelect](type-aliases/MaterializedViewSelect.md)
 - [MultiVector](type-aliases/MultiVector.md)
 - [RecordBatchLike](type-aliases/RecordBatchLike.md)
 - [SchemaLike](type-aliases/SchemaLike.md)
@@ -50,14 +50,6 @@ changedColumns: BranchColumnChange[];

 ***

-### errors
-
-```ts
-errors: CherryPickError[];
-```
-
-***
-
 ### fromBranch

 ```ts
@@ -74,6 +66,22 @@ mainVersion: number;

 ***

+### mergeBlockers
+
+```ts
+mergeBlockers: MergeBlocker[];
+```
+
+***
+
+### mergeable
+
+```ts
+mergeable: boolean;
+```
+
+***
+
 ### parentVersion

 ```ts
@@ -1,17 +0,0 @@
-[**@lancedb/lancedb**](../README.md) • **Docs**
-
-***
-
-[@lancedb/lancedb](../globals.md) / CherryPickPreview
-
-# Interface: CherryPickPreview
-
-Changes that would be, or were, promoted by a cherry-pick.
-
-## Properties
-
-### promotedColumns
-
-```ts
-promotedColumns: string[];
-```
@@ -17,8 +17,7 @@ metadata: Record<string, null | string>;
 ```

 Metadata key/value pairs. Merged into the field's existing metadata by
-default; a value of `null` deletes that key. See
-[Table.updateFieldMetadata](../classes/Table.md#updatefieldmetadata) for the conventional `lancedb:*` keys.
+default; a value of `null` deletes that key.

 ***

@@ -1,34 +0,0 @@
-[**@lancedb/lancedb**](../README.md) • **Docs**
-
-***
-
-[@lancedb/lancedb](../globals.md) / ListTablesOptions
-
-# Interface: ListTablesOptions
-
-## Properties
-
-### limit?
-
-```ts
-optional limit: number;
-```
-
-An upper bound on how many tables to return.
-
-A page may hold fewer than this and still not be the last one, so keep
-going while the response carries a page token rather than while pages are
-full.
-
-***
-
-### pageToken?
-
-```ts
-optional pageToken: string;
-```
-
-Token from a previous response, to resume listing where it left off.
-
-The token is opaque: it carries whatever the database needs to resume, and
-callers should not construct or interpret one.
@@ -1,23 +0,0 @@
-[**@lancedb/lancedb**](../README.md) • **Docs**
-
-***
-
-[@lancedb/lancedb](../globals.md) / ListTablesResponse
-
-# Interface: ListTablesResponse
-
-## Properties
-
-### pageToken?
-
-```ts
-optional pageToken: string;
-```
-
-***
-
-### tables
-
-```ts
-tables: string[];
-```
@@ -1,59 +0,0 @@
-[**@lancedb/lancedb**](../README.md) • **Docs**
-
-***
-
-[@lancedb/lancedb](../globals.md) / MaterializedViewDefinition
-
-# Interface: MaterializedViewDefinition
-
-The query that defines a materialized view.
-
-## Properties
-
-### filter?
-
-```ts
-optional filter: string;
-```
-
-SQL predicate selecting the source rows the view holds.
-
-***
-
-### inputs
-
-```ts
-inputs: string[];
-```
-
-Source columns the projections and filter read.
-
-***
-
-### limit?
-
-```ts
-optional limit: number;
-```
-
-Cap on the number of rows the view holds.
-
-***
-
-### projections
-
-```ts
-projections: [string, string][];
-```
-
-`[output column, SQL expression]` pairs, in view schema order.
-
-***
-
-### sourceTable
-
-```ts
-sourceTable: string;
-```
-
-Name of the source table, in the same database as the view.
@@ -2,11 +2,11 @@

 ***

-[@lancedb/lancedb](../globals.md) / CherryPickError
+[@lancedb/lancedb](../globals.md) / MergeBlocker

-# Interface: CherryPickError
+# Interface: MergeBlocker

-A reason why a cherry-pick cannot currently land.
+A reason why a branch cannot currently be merged.

 ## Properties

@@ -2,11 +2,11 @@

 ***

-[@lancedb/lancedb](../globals.md) / CherryPickResult
+[@lancedb/lancedb](../globals.md) / MergeBranchResult

-# Interface: CherryPickResult
+# Interface: MergeBranchResult

-Result of previewing or attempting a cherry-pick.
+Result of previewing or attempting a branch merge.

 ## Properties

@@ -29,7 +29,7 @@ optional mainVersionAfter: number;
 ### preview

 ```ts
-preview: CherryPickPreview;
+preview: MergePreview;
 ```

 ***
@@ -38,9 +38,9 @@ preview: CherryPickPreview;

 ```ts
 status:
-  | "failed"
  | "unknown"
+  | "rejected"
  | "ready"
  | "notImplemented"
-  | "cherryPicked";
+  | "merged";
 ```
@@ -0,0 +1,17 @@
+[**@lancedb/lancedb**](../README.md) • **Docs**
+
+***
+
+[@lancedb/lancedb](../globals.md) / MergePreview
+
+# Interface: MergePreview
+
+Changes that would be, or were, promoted by a branch merge.
+
+## Properties
+
+### promotedColumns
+
+```ts
+promotedColumns: string[];
+```
@@ -1,41 +0,0 @@
-[**@lancedb/lancedb**](../README.md) • **Docs**
-
-***
-
-[@lancedb/lancedb](../globals.md) / RefreshMaterializedViewResult
-
-# Interface: RefreshMaterializedViewResult
-
-## Properties
-
-### mode
-
-```ts
-mode: string;
-```
-
-How the view was brought up to date: "rebuild", "incremental" or "no_op".
-
-***
-
-### rowsWritten
-
-```ts
-rowsWritten: number;
-```
-
-***
-
-### sourceVersion
-
-```ts
-sourceVersion: number;
-```
-
-***
-
-### version
-
-```ts
-version: number;
-```
@@ -4,16 +4,11 @@

 [@lancedb/lancedb](../globals.md) / TableNamesOptions

-# Interface: ~~TableNamesOptions~~
-
-## Deprecated
-
-Use [ListTablesOptions](ListTablesOptions.md) with [Connection.listTables](../classes/Connection.md#listtables)
-instead.
+# Interface: TableNamesOptions

 ## Properties

-### ~~limit?~~
+### limit?

 ```ts
 optional limit: number;
@@ -23,7 +18,7 @@ An optional limit to the number of results to return.

 ***

-### ~~startAfter?~~
+### startAfter?

 ```ts
 optional startAfter: string;
@@ -25,12 +25,9 @@
 ### Type Aliases

 - [CreateReturnType](type-aliases/CreateReturnType.md)
- [EmbeddingMetadataEntry](type-aliases/EmbeddingMetadataEntry.md)
- [ResolvedEmbeddingFunctionConfig](type-aliases/ResolvedEmbeddingFunctionConfig.md)

 ### Functions

 - [LanceSchema](functions/LanceSchema.md)
 - [getRegistry](functions/getRegistry.md)
- [parseEmbeddingMetadata](functions/parseEmbeddingMetadata.md)
 - [register](functions/register.md)
@@ -10,12 +10,16 @@
 function getRegistry(): EmbeddingFunctionRegistry
 ```

-Get the global embedding function registry.
-
-LanceDB built-in providers are initialized when this public API is first
-used, so importing the root package does not change automatic search
-selection for tables without embedding metadata.
+Utility function to get the global instance of the registry

 ## Returns

 [`EmbeddingFunctionRegistry`](../classes/EmbeddingFunctionRegistry.md)
+
+`EmbeddingFunctionRegistry` The global instance of the registry
+
+## Example
+
+```ts
+const registry = getRegistry();
+const openai = registry.get("openai").create();
@@ -1,22 +0,0 @@
-[**@lancedb/lancedb**](../../../README.md) • **Docs**
-
-***
-
-[@lancedb/lancedb](../../../globals.md) / [embedding](../README.md) / parseEmbeddingMetadata
-
-# Function: parseEmbeddingMetadata()
-
-```ts
-function parseEmbeddingMetadata(json): EmbeddingMetadataEntry[]
-```
-
-The single parser for `embedding_functions` schema metadata: every reader
-goes through here, so the wire contract cannot fork between them.
-
-## Parameters
-
-* **json**: `string`
-
-## Returns
-
-[`EmbeddingMetadataEntry`](../type-aliases/EmbeddingMetadataEntry.md)[]
@@ -1,40 +0,0 @@
-[**@lancedb/lancedb**](../../../README.md) • **Docs**
-
-***
-
-[@lancedb/lancedb](../../../globals.md) / [embedding](../README.md) / EmbeddingMetadataEntry
-
-# Type Alias: EmbeddingMetadataEntry
-
-```ts
-type EmbeddingMetadataEntry: object;
-```
-
-One entry of the `embedding_functions` schema metadata, with the column
-keys normalized across the bindings' spellings.
-
-## Type declaration
-
-### model
-
-```ts
-model: EmbeddingFunction["TOptions"];
-```
-
-### name
-
-```ts
-name: string;
-```
-
-### sourceColumn
-
-```ts
-sourceColumn: string;
-```
-
-### vectorColumn
-
-```ts
-vectorColumn: string;
-```
@@ -1,22 +0,0 @@
-[**@lancedb/lancedb**](../../../README.md) • **Docs**
-
-***
-
-[@lancedb/lancedb](../../../globals.md) / [embedding](../README.md) / ResolvedEmbeddingFunctionConfig
-
-# Type Alias: ResolvedEmbeddingFunctionConfig
-
-```ts
-type ResolvedEmbeddingFunctionConfig: EmbeddingFunctionConfig & object;
-```
-
-An [EmbeddingFunctionConfig] read back from table metadata, where the
-vector column is always recorded.
-
-## Type declaration
-
-### vectorColumn
-
-```ts
-vectorColumn: string;
-```
@@ -1,14 +0,0 @@
-[**@lancedb/lancedb**](../README.md) • **Docs**
-
-***
-
-[@lancedb/lancedb](../globals.md) / MaterializedViewSelect
-
-# Type Alias: MaterializedViewSelect
-
-```ts
-type MaterializedViewSelect: (string | [string, string])[] | Record<string, string>;
-```
-
-The view's columns: column names, `[alias, SQL expression]` pairs, or a
-record of the same. A bare name projects itself.
@@ -102,12 +102,6 @@ listing a storage directory.

 ::: lancedb.job.AsyncJob

-## Materialized Views (Synchronous)
-
-::: lancedb.materialized_view.MaterializedView
-
-::: lancedb.materialized_view.MaterializedViewDefinition
-
 ## Expressions

 Type-safe expression builder for filters and projections. Use these instead
@@ -159,8 +153,6 @@ and combined with [BooleanQuery][lancedb.query.BooleanQuery].

 ::: lancedb.query.FullTextOperator

-::: lancedb.query.DocumentGranularity
-
 ::: lancedb.query.Occur

 ## Embeddings
@@ -223,14 +215,10 @@ tokens = list(
 Blob columns store large binary values out of line so they can be read lazily
 instead of being materialized with the rest of the row.

-`lancedb.BlobType` is `lance.blob.BlobType` when pylance is installed. Without
-pylance, LanceDB uses a matching `lance.blob.v2` extension type so blob columns
-still work. Queries return descriptors. Call
-[`fetch_blob_files`][lancedb.table.Table.fetch_blob_files] for lazy reads or
-[`fetch_blobs`][lancedb.table.Table.fetch_blobs] for eager bytes.
-
 ::: lancedb.blob

+::: lancedb.BlobType
+
 ::: lancedb._blob.BlobFile
    options:
      show_root_full_path: false
@@ -267,8 +255,6 @@ still work. Queries return descriptors. Call

 ::: lancedb.streaming.StreamingDataset

-::: lancedb.streaming.StreamingDataLoader
-
 ::: lancedb.permutation.permutation_builder

 ::: lancedb.permutation.PermutationBuilder
@@ -309,10 +295,6 @@ Table hold your actual data as a collection of records / rows.

 ::: lancedb.table.AsyncBranches

-## Materialized Views (Asynchronous)
-
-::: lancedb.materialized_view.AsyncMaterializedView
-
 ## Indices (Asynchronous)

 Indices can be created on a table to speed up queries. This section
@@ -8,7 +8,7 @@
    <parent>
      <groupId>com.lancedb</groupId>
      <artifactId>lancedb-parent</artifactId>
-      <version>0.38.0-beta.11</version>
+      <version>0.38.0-beta.3</version>
      <relativePath>../pom.xml</relativePath>
    </parent>

@@ -6,7 +6,7 @@

    <groupId>com.lancedb</groupId>
    <artifactId>lancedb-parent</artifactId>
-    <version>0.38.0-beta.11</version>
+    <version>0.38.0-beta.3</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>11.0.0-beta.16</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>
@@ -1,7 +1,7 @@
 [package]
 name = "lancedb-nodejs"
 edition.workspace = true
-version = "0.38.0-beta.11"
+version = "0.38.0-beta.3"
 publish = false
 license.workspace = true
 description.workspace = true
@@ -1,16 +1,11 @@
 // SPDX-License-Identifier: Apache-2.0
 // SPDX-FileCopyrightText: Copyright The LanceDB Authors
-import * as fs from "node:fs";
-import * as vm from "node:vm";
 import * as arrow15 from "apache-arrow-15";
 import * as arrow16 from "apache-arrow-16";
 import * as arrow17 from "apache-arrow-17";
 import * as arrow18 from "apache-arrow-18";

 import {
-  Field as CurrentField,
-  LargeBinary as CurrentLargeBinary,
-  Schema as CurrentSchema,
  Vector as CurrentVector,
  convertToTable,
  tableFromIPC as currentTableFromIPC,
@@ -41,59 +36,6 @@ function sampleRecords(): Array<Record<string, any>> {
    },
  ];
 }
-
-it("serializes an Arrow Table created in another JavaScript realm", async () => {
-  const context = vm.createContext({
-    TextDecoder,
-    TextEncoder,
-    console,
-    setTimeout,
-    clearTimeout,
-  });
-  vm.runInContext(
-    fs.readFileSync(
-      require.resolve("apache-arrow-15/Arrow.es2015.min"),
-      "utf8",
-    ),
-    context,
-  );
-  const foreignTable: unknown = vm.runInContext(
-    "Arrow.tableFromArrays({ id: new Int32Array([1, 2, 3]), text: ['foo', 'bar', 'baz'] })",
-    context,
-  );
-
-  const foreignMetadata = (
-    foreignTable as { schema: { metadata: Map<string, string> } }
-  ).schema.metadata;
-  expect(foreignMetadata).not.toBeInstanceOf(Map);
-
-  const buf = await fromDataToBuffer(
-    foreignTable as Parameters<typeof fromDataToBuffer>[0],
-  );
-  const actual = currentTableFromIPC(buf);
-
-  expect(actual.numRows).toBe(3);
-  expect(actual.getChild("id")?.toJSON()).toEqual([1, 2, 3]);
-  expect(actual.getChild("text")?.toJSON()).toEqual(["foo", "bar", "baz"]);
-});
-
-it("preserves field metadata from a provided schema", async function () {
-  const jsonMetadata = new Map([["ARROW:extension:name", "lance.json"]]);
-  const schema = new CurrentSchema([
-    new CurrentField("meta", new CurrentLargeBinary(), true, jsonMetadata),
-  ]);
-
-  const table = makeArrowTable(
-    [{ meta: Buffer.from(JSON.stringify({ source: "test" })) }],
-    { schema },
-  );
-
-  expect(table.schema.fields[0].metadata).toEqual(jsonMetadata);
-
-  const roundTripped = currentTableFromIPC(await fromTableToBuffer(table));
-  expect(roundTripped.schema.fields[0].metadata).toEqual(jsonMetadata);
-});
-
 describe.each([arrow15, arrow16, arrow17, arrow18])(
  "Arrow",
  (
@@ -231,36 +173,6 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
    }

    describe("The function makeArrowTable", function () {
-      it("accepts snake_case embedding metadata like camelCase", function () {
-        const spellings = [
-          // biome-ignore lint/style/useNamingConvention: the Python wire spelling
-          { source_column: "text", vector_column: "vector" },
-          { sourceColumn: "text", vectorColumn: "vector" },
-        ];
-        for (const columns of spellings) {
-          const schema = new Schema(
-            [
-              new Field("text", new Utf8(), false),
-              new Field(
-                "vector",
-                new FixedSizeList(3, new Field("item", new Float32(), true)),
-                false,
-              ),
-            ],
-            new Map([
-              [
-                "embedding_functions",
-                JSON.stringify([{ name: "mock", model: {}, ...columns }]),
-              ],
-            ]),
-          );
-          // The vector field is non-nullable and absent from the data; only a
-          // recognized embedding config makes that acceptable.
-          const table = makeArrowTable([{ text: "hello" }], { schema });
-          expect(table.numRows).toBe(1);
-        }
-      });
-
      it("will use data types from a provided schema instead of inference", async function () {
        const schema = new Schema([
          new Field("a", new Int32(), false),
@@ -573,137 +485,6 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
        );
      });

-      it("will allow matching inferred types across records", function () {
-        expect(() =>
-          makeArrowTable([{ value: 1 }, { value: 2 }]),
-        ).not.toThrow();
-      });
-
-      it("will reject mismatched inferred types across records", function () {
-        expect(() => makeArrowTable([{ value: 1 }, { value: "two" }])).toThrow(
-          "Failed to infer schema for data. Previously inferred type Float64 but found Utf8 for field value at row 1. Consider providing an explicit schema.",
-        );
-      });
-
-      it("will ignore generated dictionary IDs when comparing inferred types", function () {
-        const table = makeArrowTable([{ str: "a" }, { str: "b" }], {
-          dictionaryEncodeStrings: true,
-        });
-
-        expect(table.getChild("str")?.toJSON()).toEqual(["a", "b"]);
-      });
-
-      it("will preserve null values without treating them as type mismatches", function () {
-        for (const records of [
-          [{ vector: [1, 2, 3] }, { vector: null }],
-          [{ vector: null }, { vector: [1, 2, 3] }],
-        ]) {
-          const table = makeArrowTable(records);
-
-          expect(table.numRows).toBe(2);
-          expect(table.getChild("vector")?.nullCount).toBe(1);
-        }
-      });
-
-      it("will preserve empty variable-size lists", function () {
-        for (const records of [
-          [{ items: [1] }, { items: [] }],
-          [{ items: [] }, { items: [1] }],
-        ]) {
-          const table = makeArrowTable(records);
-          expect(
-            table
-              .getChild("items")
-              ?.toJSON()
-              .map((value) => value.toJSON()),
-          ).toEqual(records.map((record) => record.items));
-        }
-      });
-
-      it("will propagate deferred evidence through nested lists", function () {
-        for (const records of [
-          [{ items: [1] }, { items: [null] }],
-          [{ items: [null] }, { items: [1] }],
-          [{ items: [null, 1] }, { items: [2, null] }],
-        ]) {
-          const table = makeArrowTable(records);
-          expect(
-            table
-              .getChild("items")
-              ?.toJSON()
-              .map((value) => value.toJSON()),
-          ).toEqual(records.map((record) => record.items));
-        }
-
-        const nestedRecords = [{ items: [[1]] }, { items: [[null]] }];
-        const nestedTable = makeArrowTable(nestedRecords);
-        expect(
-          nestedTable
-            .getChild("items")
-            ?.toJSON()
-            .map((value) =>
-              value
-                .toJSON()
-                .map((nestedValue: { toJSON: () => unknown[] }) =>
-                  nestedValue.toJSON(),
-                ),
-            ),
-        ).toEqual(nestedRecords.map((record) => record.items));
-      });
-
-      it("will reject incompatible deferred evidence within a list", function () {
-        for (const items of [
-          [[], 1],
-          [1, []],
-          [[null], 1],
-          [1, [null]],
-        ]) {
-          expect(() => makeArrowTable([{ items }])).toThrow(
-            "Failed to infer data type for field items at row 0.",
-          );
-        }
-      });
-
-      it("will reject empty fixed-size lists", function () {
-        expect(() =>
-          makeArrowTable([{ vector: [1, 2, 3] }, { vector: [] }]),
-        ).toThrow(
-          "Failed to infer schema for data. Previously inferred type FixedSizeList[3]<Float32> but found List[0] for field vector at row 1.",
-        );
-      });
-
-      it("will reject inferred leaf and branch shape changes", function () {
-        expect(() =>
-          makeArrowTable([{ value: 1 }, { value: { nested: 2 } }]),
-        ).toThrow(
-          "Failed to infer schema for data. Previously inferred type Float64 but found Struct for field value at row 1.",
-        );
-        expect(() =>
-          makeArrowTable([{ value: { nested: 1 } }, { value: 2 }]),
-        ).toThrow(
-          "Failed to infer schema for data. Previously inferred type Struct but found Float64 for field value at row 1.",
-        );
-      });
-
-      it("will allow null values around inferred struct values", function () {
-        for (const { records, nullIndex } of [
-          {
-            records: [{ value: null }, { value: { nested: 2 } }],
-            nullIndex: 0,
-          },
-          {
-            records: [{ value: { nested: 1 } }, { value: null }],
-            nullIndex: 1,
-          },
-        ]) {
-          const table = makeArrowTable(records);
-          const values = table.getChild("value");
-
-          expect(values?.nullCount).toBe(1);
-          expect(values?.get(nullIndex)).toBeNull();
-        }
-      });
-
      it("will allow a schema to be provided", async function () {
        await checkTableCreation(
          async (records, _, schema) =>
@@ -4,13 +4,7 @@
 import { readdirSync } from "fs";
 import { Field, Float64, Schema } from "apache-arrow";
 import * as tmp from "tmp";
-import {
-  Connection,
-  ListTablesResponse,
-  Table,
-  connect,
-  connectNamespace,
-} from "../lancedb";
+import { Connection, Table, connect, connectNamespace } from "../lancedb";
 import { LocalTable } from "../lancedb/table";

 describe("when connecting", () => {
@@ -53,7 +47,6 @@ describe("given a connection", () => {
    await db.close();
    expect(db.isOpen()).toBe(false);
    await expect(db.tableNames()).rejects.toThrow("Connection is closed");
-    await expect(db.listTables()).rejects.toThrow("Connection is closed");
    await expect(db.renameTable("a", "b")).rejects.toThrow(
      "Connection is closed",
    );
@@ -136,66 +129,6 @@ describe("given a connection", () => {
    expect(tables).toEqual(["b", "c"]);
  });

-  it("should respect limit and page token when listing tables", async () => {
-    const db = await connect(tmpDir.name);
-
-    await db.createTable("b", [{ id: 1 }]);
-    await db.createTable("a", [{ id: 1 }]);
-    await db.createTable("c", [{ id: 1 }]);
-
-    const all = await db.listTables();
-    expect(all.tables).toEqual(["a", "b", "c"]);
-    expect(all.pageToken).toBeUndefined();
-
-    const first = await db.listTables({ limit: 1 });
-    expect(first.tables).toEqual(["a"]);
-    expect(first.pageToken).toBeDefined();
-
-    const second = await db.listTables({
-      limit: 1,
-      pageToken: first.pageToken,
-    });
-    expect(second.tables).toEqual(["b"]);
-  });
-
-  it("should visit every table exactly once when walking pages", async () => {
-    const db = await connect(tmpDir.name);
-
-    const created = ["a", "b", "c", "d", "e"];
-    for (const name of created) {
-      await db.createTable(name, [{ id: 1 }]);
-    }
-
-    const seen: string[] = [];
-    let pageToken: string | undefined = undefined;
-    do {
-      const page: ListTablesResponse = await db.listTables({
-        limit: 2,
-        pageToken,
-      });
-      seen.push(...page.tables);
-      pageToken = page.pageToken;
-    } while (pageToken);
-
-    expect(seen).toEqual(created);
-  });
-
-  it("should list tables in a namespace", async () => {
-    const db = await connect(tmpDir.name, {
-      // biome-ignore lint/style/useNamingConvention: opaque backend property key, must match Rust
-      namespaceClientProperties: { manifest_enabled: "true" },
-    });
-    await db.createNamespace(["child"]);
-    await db.createTable("nested", [{ id: 1 }], ["child"]);
-
-    await expect(db.listTables(["child"])).resolves.toEqual(
-      expect.objectContaining({ tables: ["nested"] }),
-    );
-    await expect(db.listTables()).resolves.toEqual(
-      expect.objectContaining({ tables: [] }),
-    );
-  });
-
  it("should create tables in v2 mode", async () => {
    const db = await connect(tmpDir.name);
    const data = [...Array(10000).keys()].map((i) => ({ id: i }));
@@ -187,58 +187,6 @@ describe("embedding functions", () => {
    const vector0 = JSON.parse(JSON.stringify(arr[0].vector));
    expect(vector0).toEqual([1, 2, 3]);
  });
-  it("should append multiple Python embeddings with the same alias", async () => {
-    @register("python-mock")
-    // biome-ignore lint/correctness/noUnusedVariables: the decorator registers this class
-    class MockEmbeddingFunction extends EmbeddingFunction<string> {
-      ndims() {
-        return 3;
-      }
-      embeddingDataType(): Float {
-        return new Float32();
-      }
-      async computeQueryEmbeddings(_data: string) {
-        return [1, 2, 3];
-      }
-      async computeSourceEmbeddings(data: string[]) {
-        return data.map((value) =>
-          value === "hello world" ? [1, 2, 3] : [4, 5, 6],
-        );
-      }
-    }
-
-    const metadata = new Map([
-      [
-        "embedding_functions",
-        '[{"source_column":"text1","vector_column":"vector1","name":"python-mock","model":{}},{"source_column":"text2","vector_column":"vector2","name":"python-mock","model":{}}]',
-      ],
-    ]);
-    const schema = new Schema(
-      [
-        new Field("text1", new Utf8(), true),
-        new Field("text2", new Utf8(), true),
-        new Field(
-          "vector1",
-          new FixedSizeList(3, new Field("item", new Float32(), true)),
-          true,
-        ),
-        new Field(
-          "vector2",
-          new FixedSizeList(3, new Field("item", new Float32(), true)),
-          true,
-        ),
-      ],
-      metadata,
-    );
-
-    const db = await connect(tmpDir.name);
-    const table = await db.createEmptyTable("test", schema);
-    await table.add([{ text1: "hello world", text2: "goodbye world" }]);
-
-    const rows = await table.query().toArray();
-    expect(JSON.parse(JSON.stringify(rows[0].vector1))).toEqual([1, 2, 3]);
-    expect(JSON.parse(JSON.stringify(rows[0].vector2))).toEqual([4, 5, 6]);
-  });

  it("should append generated vectors to a non-nullable schema", async () => {
    @register("non_nullable_schema_test")
@@ -539,52 +487,4 @@ describe("embedding functions", () => {
      expect(stringSchema3).toEqual(stringExpectedSchema);
    },
  );
-  test("parses one function writing several vector columns", async () => {
-    class MockEmbeddingFunction extends EmbeddingFunction<string> {
-      ndims() {
-        return 3;
-      }
-      embeddingDataType(): Float {
-        return new Float32();
-      }
-      async computeQueryEmbeddings(_data: string) {
-        return [1, 2, 3];
-      }
-      async computeSourceEmbeddings(data: string[]) {
-        return Array.from({ length: data.length }).fill([
-          1, 2, 3,
-        ]) as number[][];
-      }
-    }
-    const registry = getRegistry();
-    registry.register("multi_output_mock")(MockEmbeddingFunction);
-
-    // A materialized view can project one source vector column under two
-    // names, so a table's configuration names the same function twice.
-    const parsed = await registry.parseFunctions(
-      new Map([
-        [
-          "embedding_functions",
-          JSON.stringify([
-            {
-              name: "multi_output_mock",
-              sourceColumn: "text",
-              vectorColumn: "vector_a",
-              model: {},
-            },
-            {
-              name: "multi_output_mock",
-              sourceColumn: "text",
-              vectorColumn: "vector_b",
-              model: {},
-            },
-          ]),
-        ],
-      ]),
-    );
-
-    expect(
-      [...parsed.values()].map(({ vectorColumn }) => vectorColumn).sort(),
-    ).toEqual(["vector_a", "vector_b"]);
-  });
 });
@@ -1,95 +0,0 @@
-// SPDX-License-Identifier: Apache-2.0
-// SPDX-FileCopyrightText: Copyright The LanceDB Authors
-
-import { execFileSync } from "node:child_process";
-import { resolve } from "node:path";
-
-import type { OpenAIEmbeddingFunction } from "../lancedb/embedding/openai";
-import type { EmbeddingFunctionRegistry } from "../lancedb/embedding/registry";
-
-type EmbeddingModule = typeof import("../lancedb/embedding");
-type OpenAIModule = typeof import("../lancedb/embedding/openai");
-type RegistryModule = typeof import("../lancedb/embedding/registry");
-
-describe("embedding function registry", () => {
-  const registries: EmbeddingFunctionRegistry[] = [];
-
-  afterEach(() => {
-    for (const registry of registries) {
-      registry.reset();
-    }
-    registries.length = 0;
-  });
-
-  it("defers built-in providers until the public registry API is used", () => {
-    jest.isolateModules(() => {
-      const embedding = require("../lancedb/embedding") as EmbeddingModule;
-      const { getRegistry: getInternalRegistry } =
-        require("../lancedb/embedding/registry") as RegistryModule;
-      const registry = getInternalRegistry();
-      registries.push(registry);
-
-      expect(registry.length()).toBe(0);
-      expect(embedding.getRegistry()).toBe(registry);
-      expect(registry.get("openai")).toBeDefined();
-      expect(registry.get("huggingface")).toBeDefined();
-    });
-  });
-
-  it("preserves automatic FTS search in a fresh process", () => {
-    execFileSync(
-      process.execPath,
-      [resolve(__dirname, "fixtures", "auto_fts_search.cjs")],
-      { stdio: "pipe" },
-    );
-  });
-
-  it("shares registrations across duplicated provider module graphs", () => {
-    let registeringRegistry: EmbeddingFunctionRegistry | undefined;
-    let latestOpenAIConstructor: typeof OpenAIEmbeddingFunction | undefined;
-
-    jest.isolateModules(() => {
-      require("../lancedb/embedding/openai");
-      const { getRegistry } =
-        require("../lancedb/embedding/registry") as RegistryModule;
-      registeringRegistry = getRegistry();
-      registries.push(registeringRegistry);
-      expect(registeringRegistry.get("openai")).toBeDefined();
-    });
-
-    expect(() => {
-      jest.isolateModules(() => {
-        const { OpenAIEmbeddingFunction } =
-          require("../lancedb/embedding/openai") as OpenAIModule;
-        latestOpenAIConstructor = OpenAIEmbeddingFunction;
-        const { getRegistry } =
-          require("../lancedb/embedding/registry") as RegistryModule;
-        registries.push(getRegistry());
-      });
-    }).not.toThrow();
-
-    const previousApiKey = process.env.OPENAI_API_KEY;
-    process.env.OPENAI_API_KEY = "test";
-    try {
-      const latestOpenAI = registeringRegistry!
-        .get<OpenAIEmbeddingFunction>("openai")!
-        .create();
-      expect(latestOpenAI).toBeInstanceOf(latestOpenAIConstructor!);
-    } finally {
-      if (previousApiKey === undefined) {
-        delete process.env.OPENAI_API_KEY;
-      } else {
-        process.env.OPENAI_API_KEY = previousApiKey;
-      }
-    }
-
-    jest.isolateModules(() => {
-      const { getRegistry } =
-        require("../lancedb/embedding") as EmbeddingModule;
-      const publicRegistry = getRegistry();
-      registries.push(publicRegistry);
-      expect(publicRegistry).toBe(registeringRegistry);
-      expect(publicRegistry.get("openai")).toBeDefined();
-    });
-  });
-});
@@ -1,33 +0,0 @@
-// SPDX-License-Identifier: Apache-2.0
-// SPDX-FileCopyrightText: Copyright The LanceDB Authors
-
-const assert = require("node:assert/strict");
-const tmp = require("tmp");
-const { connect, embedding, Index } = require("../../dist");
-const { getRegistry } = require("../../dist/embedding/registry");
-
-async function main() {
-  assert.equal(typeof embedding.getRegistry, "function");
-  assert.equal(getRegistry().length(), 0);
-  assert.equal(embedding.getRegistry(), getRegistry());
-  assert.equal(getRegistry().length(), 2);
-
-  const dir = tmp.dirSync({ unsafeCleanup: true });
-  let db;
-  try {
-    db = await connect(dir.name);
-    const table = await db.createTable("docs", [{ text: "hello world" }]);
-    await table.createIndex("text", { config: Index.fts() });
-
-    const rows = await table.search("hello").toArray();
-    assert.equal(rows[0].text, "hello world");
-  } finally {
-    db?.close();
-    dir.removeCallback();
-  }
-}
-
-main().catch((error) => {
-  console.error(error);
-  process.exitCode = 1;
-});
@@ -1,147 +0,0 @@
-// SPDX-License-Identifier: Apache-2.0
-// SPDX-FileCopyrightText: Copyright The LanceDB Authors
-
-import * as tmp from "tmp";
-
-import { Connection, connect } from "../lancedb";
-import {
-  DEFINITION_META_KEY,
-  definitionFromMetadata,
-} from "../lancedb/materialized_view";
-
-describe("materialized views", () => {
-  let tmpDir: tmp.DirResult;
-  let db: Connection;
-
-  beforeEach(async () => {
-    tmpDir = tmp.dirSync({ unsafeCleanup: true });
-    db = await connect(tmpDir.name);
-    await db.createTable(
-      "people",
-      [
-        { name: "ada", age: 36 },
-        { name: "kid", age: 7 },
-        { name: "grace", age: 85 },
-      ],
-      { storageOptions: { newTableEnableStableRowIds: "true" } },
-    );
-  });
-  afterEach(() => tmpDir.removeCallback());
-
-  it("rejects a stored limit a number cannot carry", () => {
-    const big = new Map([
-      [
-        DEFINITION_META_KEY,
-        '{"kind":"select","source_table":"people","limit":9007199254740993}',
-      ],
-    ]);
-    expect(() => definitionFromMetadata(big, "v")).toThrow(
-      /too large to represent exactly/,
-    );
-
-    const safe = new Map([
-      [
-        DEFINITION_META_KEY,
-        '{"kind":"select","source_table":"people","limit":42}',
-      ],
-    ]);
-    expect(definitionFromMetadata(safe, "v").limit).toBe(42);
-  });
-
-  it("creates, refreshes and queries a view", async () => {
-    const view = await db.createMaterializedView("adults", "people", {
-      select: ["name", ["shout", "upper(name)"]],
-      where: "age >= 18",
-    });
-    expect(view.name).toBe("adults");
-    expect(await view.table().countRows()).toBe(0);
-
-    const result = await view.refresh();
-    expect(result.mode).toBe("rebuild");
-    expect(Number(result.rowsWritten)).toBe(2);
-
-    const rows = await view.table().query().toArray();
-    expect(rows.map((r) => r.shout).sort()).toEqual(["ADA", "GRACE"]);
-  });
-
-  it("round-trips the definition", async () => {
-    await db.createMaterializedView("adults", "people", {
-      where: "age >= 18",
-    });
-    const view = await db.openMaterializedView("adults");
-    const definition = await view.definition();
-    expect(definition.sourceTable).toBe("people");
-    expect(definition.filter).toBe("age >= 18");
-    expect(definition.projections).toEqual([
-      ["name", "`name`"],
-      ["age", "`age`"],
-    ]);
-    expect(definition.inputs).toEqual(["age", "name"]);
-  });
-
-  it("refreshes incrementally after an append", async () => {
-    const view = await db.createMaterializedView("copy", "people");
-    await view.refresh();
-
-    const people = await db.openTable("people");
-    await people.add([{ name: "alan", age: 41 }]);
-    const result = await view.refresh();
-    expect(result.mode).toBe("incremental");
-    expect(Number(result.rowsWritten)).toBe(1);
-    expect(await view.table().countRows()).toBe(4);
-
-    expect((await view.refresh()).mode).toBe("no_op");
-  });
-
-  it("lists views and rejects non-views", async () => {
-    await db.createMaterializedView("adults", "people", {
-      where: "age >= 18",
-    });
-    expect(await db.listMaterializedViews()).toEqual(["adults"]);
-    await expect(db.openMaterializedView("people")).rejects.toThrow(
-      "not a materialized view",
-    );
-  });
-
-  it("rejects an invalid expression at create time", async () => {
-    await expect(
-      db.createMaterializedView("bad", "people", {
-        select: [["x", "missing + 1"]],
-      }),
-    ).rejects.toThrow("missing");
-  });
-
-  it("rejects invalid numeric options before creating anything", async () => {
-    for (const limit of [-5, 1.5, Infinity, NaN]) {
-      await expect(
-        db.createMaterializedView("bad", "people", { limit }),
-      ).rejects.toThrow("non-negative integer");
-    }
-    expect(await db.listMaterializedViews()).toEqual([]);
-
-    const view = await db.createMaterializedView("copy", "people");
-    for (const sourceVersion of [-1, 1.5, Infinity, NaN]) {
-      await expect(view.refresh({ sourceVersion })).rejects.toThrow(
-        "non-negative integer",
-      );
-    }
-  });
-
-  it("quotes bare select names", async () => {
-    await db.createTable("odd_names", [{ "order item": "widget" }], {
-      storageOptions: { newTableEnableStableRowIds: "true" },
-    });
-    const view = await db.createMaterializedView("quoted", "odd_names", {
-      select: ["order item"],
-    });
-    const result = await view.refresh();
-    expect(Number(result.rowsWritten)).toBe(1);
-  });
-
-  it("requires stable row ids on the source", async () => {
-    await db.createTable("plain", [{ x: 1 }]);
-    await expect(db.createMaterializedView("v", "plain")).rejects.toThrow(
-      "stable row ids",
-    );
-  });
-});
@@ -106,77 +106,6 @@ describe.each([arrow15, arrow16, arrow17, arrow18])("Registry", (arrow) => {
      'Embedding function with alias "mock-embedding" already exists',
    );
  });
-  test("parseFunctions keeps entries sharing a function name", async () => {
-    class MockEmbeddingFunction extends EmbeddingFunction<string> {
-      ndims() {
-        return 3;
-      }
-      embeddingDataType() {
-        return new arrow.Float32() as apiArrow.Float;
-      }
-      async computeSourceEmbeddings(data: string[]) {
-        return data.map(() => [1, 2, 3]);
-      }
-    }
-    register("mock-embedding")(MockEmbeddingFunction);
-    const parsed = await getRegistry().parseFunctions(
-      new Map([
-        [
-          "embedding_functions",
-          JSON.stringify([
-            {
-              name: "mock-embedding",
-              sourceColumn: "text",
-              vectorColumn: "vector_a",
-              model: {},
-            },
-            {
-              name: "mock-embedding",
-              sourceColumn: "text",
-              vectorColumn: "vector_b",
-              model: {},
-            },
-          ]),
-        ],
-      ]),
-    );
-    expect([...parsed.values()].map((f) => f.vectorColumn)).toEqual([
-      "vector_a",
-      "vector_b",
-    ]);
-
-    // The Python bindings write snake_case keys.
-    const snake = await getRegistry().parseFunctions(
-      new Map([
-        [
-          "embedding_functions",
-          JSON.stringify([
-            {
-              name: "mock-embedding",
-              // biome-ignore lint/style/useNamingConvention: the Python wire spelling
-              source_column: "text",
-              // biome-ignore lint/style/useNamingConvention: the Python wire spelling
-              vector_column: "vector_a",
-              model: {},
-            },
-            {
-              name: "mock-embedding",
-              // biome-ignore lint/style/useNamingConvention: the Python wire spelling
-              source_column: "text",
-              // biome-ignore lint/style/useNamingConvention: the Python wire spelling
-              vector_column: "vector_b",
-              model: {},
-            },
-          ]),
-        ],
-      ]),
-    );
-    expect([...snake.keys()]).toEqual(["vector_a", "vector_b"]);
-    expect([...snake.values()].map((f) => f.sourceColumn)).toEqual([
-      "text",
-      "text",
-    ]);
-  });
  test("schema should contain correct metadata", async () => {
    class MockEmbeddingFunction extends EmbeddingFunction<string> {
      constructor(args: FunctionOptions = {}) {
@@ -75,25 +75,6 @@ async function withMockDatabase(
 }

 describe("remote connection", () => {
-  it("refuses materialized views before issuing any request", async () => {
-    const paths: string[] = [];
-    await withMockDatabase(
-      (req, res) => {
-        paths.push(req.url ?? "");
-        res.writeHead(404).end();
-      },
-      async (db) => {
-        await expect(db.openMaterializedView("secret_table")).rejects.toThrow(
-          /only on local databases/,
-        );
-        await expect(db.listMaterializedViews()).rejects.toThrow(
-          /only on local databases/,
-        );
-        expect(paths).toEqual([]);
-      },
-    );
-  });
-
  it("should accept partial connection options", async () => {
    await connect("db://test", {
      apiKey: "fake",
@@ -330,7 +311,7 @@ describe("remote connection", () => {
    expect(createIndexBody?.["custom_stop_words"]).toEqual(["the"]);
  });

-  it("diffs and cherry-picks remote branches", async () => {
+  it("diffs and merges remote branches", async () => {
    const sampleDiff = {
      fromBranch: "exp",
      parentVersion: 1,
@@ -352,9 +333,10 @@ describe("remote connection", () => {
      changedColumns: [],
      addedIndexes: [],
      removedIndexes: [],
-      errors: [],
+      mergeable: true,
+      mergeBlockers: [],
    };
-    const cherryPickBodies: Record<string, unknown>[] = [];
+    const mergeBodies: Record<string, unknown>[] = [];

    await withMockDatabase(
      (req, res) => {
@@ -384,16 +366,17 @@ describe("remote connection", () => {
              .end(JSON.stringify(sampleDiff));
            return;
          }
-          if (path.endsWith("/branches/cherry_pick/")) {
-            cherryPickBodies.push(body);
+          if (path.endsWith("/branches/merge/")) {
+            mergeBodies.push(body);
            const dryRun = body["dry_run"] === true;
            const response = {
-              status: dryRun ? "ready" : "failed",
+              status: dryRun ? "ready" : "rejected",
              diff: dryRun
                ? sampleDiff
                : {
                    ...sampleDiff,
-                    errors: [
+                    mergeable: false,
+                    mergeBlockers: [
                      { code: "baseMoved", message: "main has advanced" },
                    ],
                  },
@@ -415,19 +398,19 @@ describe("remote connection", () => {

        await expect(branches.diff("exp")).resolves.toEqual(sampleDiff);

-        const failed = await branches.cherryPick("exp");
-        expect(failed.status).toBe("failed");
-        expect(failed.diff.errors).toEqual([
+        const rejected = await branches.merge("exp");
+        expect(rejected.status).toBe("rejected");
+        expect(rejected.diff.mergeBlockers).toEqual([
          { code: "baseMoved", message: "main has advanced" },
        ]);

-        const preview = await branches.cherryPick("exp", true);
+        const preview = await branches.merge("exp", true);
        expect(preview.status).toBe("ready");
        expect(preview.preview.promotedColumns).toEqual(["tag"]);
      },
    );

-    expect(cherryPickBodies).toEqual([
+    expect(mergeBodies).toEqual([
      // biome-ignore lint/style/useNamingConvention: snake_case mandated by the server wire format
      { from_branch: "exp", dry_run: false },
      // biome-ignore lint/style/useNamingConvention: snake_case mandated by the server wire format
@@ -11,13 +11,10 @@ import * as arrow17 from "apache-arrow-17";
 import * as arrow18 from "apache-arrow-18";

 import {
-  AutoQuery,
  Connection,
  MatchQuery,
  PhraseQuery,
-  Query,
  Table,
-  VectorQuery,
  connect,
  tokenize,
 } from "../lancedb";
@@ -685,56 +682,6 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
  },
 );

-// https://github.com/lancedb/lancedb/issues/1963
-it("should query documents with LangChain PDF metadata", async () => {
-  const tmpDir = tmp.dirSync({ unsafeCleanup: true });
-  try {
-    const db = await connect(tmpDir.name);
-    const documents = [
-      {
-        text: "first page",
-        vector: [1, 0],
-        source: "first.pdf",
-        loc: { pageNumber: 1, lines: { from: 1, to: 12 } },
-        pdf: {
-          version: "1.10.100",
-          info: {
-            format: "PDF 1.7",
-            producer: "pdf.js",
-            creator: "Writer",
-          },
-          totalPages: 2,
-        },
-      },
-      {
-        text: "second page",
-        vector: [0, 1],
-        source: "second.pdf",
-        loc: { pageNumber: 2, lines: { from: 13, to: 24 } },
-        pdf: {
-          version: "1.10.100",
-          info: {
-            format: "PDF 1.7",
-            producer: "pdf.js",
-            creator: "Writer",
-          },
-          totalPages: 2,
-        },
-      },
-    ];
-    const documentsTable = await db.createTable("documents", documents);
-
-    const results = await documentsTable.query().toArray();
-
-    expect(results).toHaveLength(2);
-    expect(results[0].source).toBe("first.pdf");
-    expect(results[0].pdf.info.producer).toBe("pdf.js");
-    expect(results[1].loc.pageNumber).toBe(2);
-  } finally {
-    tmpDir.removeCallback();
-  }
-});
-
 describe("merge insert", () => {
  let tmpDir: tmp.DirResult;
  let table: Table;
@@ -1830,194 +1777,6 @@ describe("Read consistency interval", () => {
  });
 });

-describe("automatic search schema consistency", () => {
-  let tmpDir: tmp.DirResult;
-
-  class SchemaRefreshEmbedding extends EmbeddingFunction<string> {
-    ndims() {
-      return 2;
-    }
-
-    embeddingDataType() {
-      return new Float32();
-    }
-
-    async computeSourceEmbeddings(data: string[]) {
-      return data.map((value) => [value.length, 1]);
-    }
-
-    async computeQueryEmbeddings(value: string) {
-      return [value.length, 1];
-    }
-  }
-
-  function embeddingSchema() {
-    const func = new SchemaRefreshEmbedding();
-    return LanceSchema({
-      text: func.sourceField(new Utf8()),
-      vector: func.vectorField(),
-    });
-  }
-
-  beforeEach(() => {
-    getRegistry().reset();
-    register("schema-refresh")(SchemaRefreshEmbedding);
-    tmpDir = tmp.dirSync({ unsafeCleanup: true });
-  });
-
-  afterEach(() => {
-    getRegistry().reset();
-    tmpDir.removeCallback();
-  });
-
-  it("uses the schema refreshed from another connection", async () => {
-    const first = await connect(tmpDir.name, { readConsistencyInterval: 0 });
-    const second = await connect(tmpDir.name, { readConsistencyInterval: 0 });
-
-    try {
-      const stale = await first.createTable("docs", [{ text: "before" }], {
-        schema: embeddingSchema(),
-      });
-      const replacement = await second.createTable(
-        "docs",
-        [{ text: "after hello" }],
-        { mode: "overwrite" },
-      );
-      await replacement.createIndex("text", { config: Index.fts() });
-
-      const search = stale.search("hello");
-      expect(search).toBeInstanceOf(AutoQuery);
-      expect(search).not.toBeInstanceOf(Query);
-      expect(search).not.toBeInstanceOf(VectorQuery);
-      expect("nprobes" in search).toBe(false);
-
-      const rows = await search.toArray();
-      expect(rows[0].text).toBe("after hello");
-      expect((await stale.schema()).metadata.has("embedding_functions")).toBe(
-        false,
-      );
-    } finally {
-      first.close();
-      second.close();
-    }
-  });
-
-  it("tracks embedding metadata across checkout and restore", async () => {
-    const first = await connect(tmpDir.name, { readConsistencyInterval: 0 });
-    const second = await connect(tmpDir.name, { readConsistencyInterval: 0 });
-
-    try {
-      await first.createTable("docs", [{ text: "before" }], {
-        schema: embeddingSchema(),
-      });
-      const table = await second.createTable(
-        "docs",
-        [{ text: "after hello" }],
-        { mode: "overwrite" },
-      );
-      await table.createIndex("text", { config: Index.fts() });
-
-      await table.checkout(1);
-      expect((await table.search("before").toArray())[0].text).toBe("before");
-
-      await table.checkoutLatest();
-      expect((await table.search("hello").toArray())[0].text).toBe(
-        "after hello",
-      );
-
-      await table.checkout(1);
-      await table.restore();
-      expect((await table.search("before").toArray())[0].text).toBe("before");
-    } finally {
-      first.close();
-      second.close();
-    }
-  });
-
-  it("pins automatic search while computing an embedding", async () => {
-    let markStarted!: () => void;
-    let releaseEmbedding!: () => void;
-    const started = new Promise<void>((resolve) => {
-      markStarted = resolve;
-    });
-    const released = new Promise<void>((resolve) => {
-      releaseEmbedding = resolve;
-    });
-
-    class BlockingEmbedding extends SchemaRefreshEmbedding {
-      async computeQueryEmbeddings(value: string) {
-        markStarted();
-        await released;
-        return [value.length, 1];
-      }
-    }
-
-    register("schema-refresh-blocking")(BlockingEmbedding);
-    const func = new BlockingEmbedding();
-    const schema = LanceSchema({
-      text: func.sourceField(new Utf8()),
-      vector: func.vectorField(),
-    });
-    const first = await connect(tmpDir.name, { readConsistencyInterval: 0 });
-    const second = await connect(tmpDir.name, { readConsistencyInterval: 0 });
-
-    try {
-      const table = await first.createTable(
-        "docs",
-        [{ text: "hello before" }],
-        { schema },
-      );
-      const pending = table.search("hello").toArray();
-      await started;
-
-      const replacement = await second.createTable(
-        "docs",
-        [{ text: "hello after" }],
-        { mode: "overwrite" },
-      );
-      await replacement.createIndex("text", { config: Index.fts() });
-      releaseEmbedding();
-
-      expect((await pending)[0].text).toBe("hello before");
-    } finally {
-      releaseEmbedding();
-      first.close();
-      second.close();
-    }
-  });
-
-  it("refreshes a reused automatic search for every execution", async () => {
-    const first = await connect(tmpDir.name, { readConsistencyInterval: 0 });
-    const second = await connect(tmpDir.name, { readConsistencyInterval: 0 });
-
-    try {
-      const table = await first.createTable("docs", [
-        { text: "hello before", marker: "before" },
-      ]);
-      await table.createIndex("text", { config: Index.fts() });
-      const search = table.search("hello").select(["text"]);
-
-      const before = (await search.toArray())[0];
-      expect(before.text).toBe("hello before");
-      expect(before.marker).toBeUndefined();
-
-      const replacement = await second.createTable(
-        "docs",
-        [{ text: "hello after", marker: "after" }],
-        { mode: "overwrite" },
-      );
-      await replacement.createIndex("text", { config: Index.fts() });
-
-      const after = (await search.toArray())[0];
-      expect(after.text).toBe("hello after");
-      expect(after.marker).toBeUndefined();
-    } finally {
-      first.close();
-      second.close();
-    }
-  });
-});
-
 describe("schema evolution", function () {
  let tmpDir: tmp.DirResult;
  beforeEach(() => {
@@ -2585,24 +2344,7 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
      );
    });

-    test("full text search if only an unrelated embedding function is registered", async () => {
-      register("unused")(
-        class extends EmbeddingFunction<string> {
-          ndims() {
-            return 3;
-          }
-          embeddingDataType() {
-            return new Float32();
-          }
-          async computeQueryEmbeddings(_data: string) {
-            return [1, 2, 3];
-          }
-          async computeSourceEmbeddings(data: string[]) {
-            return data.map(() => [1, 2, 3]);
-          }
-        },
-      );
-
+    test("full text search if no embedding function provided", async () => {
      const db = await connect(tmpDir.name);
      const data = [
        { text: "hello world", vector: [0.1, 0.2, 0.3] },
@@ -2624,306 +2366,6 @@ describe.each([arrow15, arrow16, arrow17, arrow18])(
      expect(results2[0].text).toBe(data[1].text);
    });

-    test("auto search stays consistent with the active revision", async () => {
-      let initCalls = 0;
-      let queryCalls = 0;
-      let markStarted!: () => void;
-      const started = new Promise<void>((resolve) => {
-        markStarted = resolve;
-      });
-      let releaseEmbedding!: () => void;
-      const embeddingReleased = new Promise<void>((resolve) => {
-        releaseEmbedding = resolve;
-      });
-
-      @register("refresh-test")
-      class TestEmbedding extends EmbeddingFunction<string> {
-        async init() {
-          initCalls += 1;
-        }
-        ndims() {
-          return 1;
-        }
-        embeddingDataType() {
-          return new arrow.Float32();
-        }
-        async computeQueryEmbeddings(value: string) {
-          queryCalls += 1;
-          if (value === "blocked") {
-            markStarted();
-            await embeddingReleased;
-          }
-          return value === "greetings" ? [0.1] : [0.2];
-        }
-        async computeSourceEmbeddings(values: string[]) {
-          return values.map((value) =>
-            value === "hello world" ? [0.1] : [0.2],
-          );
-        }
-      }
-
-      const writer = await connect(tmpDir.name);
-      await writer.createTable("test", [{ text: "plain", vector: [0.0] }]);
-      const reader = await connect(tmpDir.name, {
-        readConsistencyInterval: 0,
-      });
-      const tracked = await reader.openTable("test");
-      type SnapshotCountingNative = {
-        querySnapshot: () => Promise<unknown>;
-      };
-      const native = (tracked as unknown as { inner: SnapshotCountingNative })
-        .inner;
-      const querySnapshot = native.querySnapshot.bind(native);
-      let snapshotCalls = 0;
-      native.querySnapshot = async () => {
-        snapshotCalls += 1;
-        return await querySnapshot();
-      };
-      const autoQuery = tracked.search("greetings").select(["text"]).limit(1);
-
-      const func = new TestEmbedding();
-      const schema = LanceSchema({
-        text: func.sourceField(new arrow.Utf8()),
-        vector: func.vectorField(),
-      });
-      const data = [{ text: "hello world" }, { text: "goodbye world" }];
-      await writer.createTable("test", data, { mode: "overwrite", schema });
-      const baselineInitCalls = initCalls;
-
-      expect(
-        (await tracked.schema()).metadata.get("embedding_functions"),
-      ).toBeDefined();
-      const results = await autoQuery.toArray();
-      expect(results[0].text).toBe(data[0].text);
-      expect(initCalls).toBe(baselineInitCalls + 1);
-      expect(queryCalls).toBe(1);
-      expect(snapshotCalls).toBe(1);
-
-      const repeatedResults = await autoQuery.toArray();
-      expect(repeatedResults[0].text).toBe(data[0].text);
-      expect(initCalls).toBe(baselineInitCalls + 1);
-      expect(queryCalls).toBe(1);
-      expect(snapshotCalls).toBe(2);
-
-      const pending = tracked
-        .search("blocked")
-        .select(["text"])
-        .limit(1)
-        .toArray();
-      await started;
-
-      const ftsData = [
-        { text: "greetings from full text", vector: [0.0] },
-        { text: "blocked from full text", vector: [0.0] },
-      ];
-      const ftsTable = await writer.createTable("test", ftsData, {
-        mode: "overwrite",
-      });
-      await ftsTable.createIndex("text", { config: Index.fts() });
-      releaseEmbedding();
-
-      const pendingResults = await pending;
-      expect(pendingResults[0].text).toBe(data[1].text);
-
-      expect(
-        (await tracked.schema()).metadata.get("embedding_functions"),
-      ).toBeUndefined();
-      const ftsResults = await autoQuery.toArray();
-      expect(ftsResults[0].text).toBe(ftsData[0].text);
-    });
-
-    test("auto search keeps newer preparation during a revision race", async () => {
-      let aCalls = 0;
-      let bCalls = 0;
-      let markAStarted!: () => void;
-      const aStarted = new Promise<void>((resolve) => {
-        markAStarted = resolve;
-      });
-      let releaseA!: () => void;
-      const aReleased = new Promise<void>((resolve) => {
-        releaseA = resolve;
-      });
-      let markBStarted!: () => void;
-      const bStarted = new Promise<void>((resolve) => {
-        markBStarted = resolve;
-      });
-      let releaseB!: () => void;
-      const bReleased = new Promise<void>((resolve) => {
-        releaseB = resolve;
-      });
-
-      @register("race-a")
-      class EmbeddingA extends EmbeddingFunction<string> {
-        ndims() {
-          return 1;
-        }
-        embeddingDataType() {
-          return new arrow.Float32();
-        }
-        async computeQueryEmbeddings() {
-          aCalls += 1;
-          markAStarted();
-          await aReleased;
-          return [0.1];
-        }
-        async computeSourceEmbeddings(values: string[]) {
-          return values.map(() => [0.1]);
-        }
-      }
-
-      @register("race-b")
-      class EmbeddingB extends EmbeddingFunction<string> {
-        ndims() {
-          return 1;
-        }
-        embeddingDataType() {
-          return new arrow.Float32();
-        }
-        async computeQueryEmbeddings() {
-          bCalls += 1;
-          markBStarted();
-          await bReleased;
-          return [0.2];
-        }
-        async computeSourceEmbeddings(values: string[]) {
-          return values.map(() => [0.2]);
-        }
-      }
-
-      const writer = await connect(tmpDir.name);
-      const embeddingA = new EmbeddingA();
-      const schemaA = LanceSchema({
-        text: embeddingA.sourceField(new arrow.Utf8()),
-        vector: embeddingA.vectorField(),
-      });
-      await writer.createTable("race", [{ text: "revision a" }], {
-        schema: schemaA,
-      });
-      const reader = await connect(tmpDir.name, {
-        readConsistencyInterval: 0,
-      });
-      const tracked = await reader.openTable("race");
-      const query = tracked.search("query");
-
-      const first = query.toArray();
-      await aStarted;
-
-      const embeddingB = new EmbeddingB();
-      const schemaB = LanceSchema({
-        text: embeddingB.sourceField(new arrow.Utf8()),
-        vector: embeddingB.vectorField(),
-      });
-      await writer.createTable("race", [{ text: "revision b" }], {
-        mode: "overwrite",
-        schema: schemaB,
-      });
-      const second = query.toArray();
-      await bStarted;
-
-      releaseA();
-      releaseB();
-      await Promise.all([first, second]);
-      expect(aCalls).toBe(1);
-      expect(bCalls).toBe(1);
-    });
-
-    test("stale FTS routing keeps newer vector preparation", async () => {
-      let vectorCalls = 0;
-      let markVectorStarted!: () => void;
-      const vectorStarted = new Promise<void>((resolve) => {
-        markVectorStarted = resolve;
-      });
-      let releaseVector!: () => void;
-      const vectorReleased = new Promise<void>((resolve) => {
-        releaseVector = resolve;
-      });
-
-      @register("stale-fts-race")
-      class RaceEmbedding extends EmbeddingFunction<string> {
-        ndims() {
-          return 1;
-        }
-        embeddingDataType() {
-          return new arrow.Float32();
-        }
-        async computeQueryEmbeddings() {
-          vectorCalls += 1;
-          markVectorStarted();
-          await vectorReleased;
-          return [0.1];
-        }
-        async computeSourceEmbeddings(values: string[]) {
-          return values.map(() => [0.1]);
-        }
-      }
-
-      const writer = await connect(tmpDir.name);
-      const ftsTable = await writer.createTable("stale_fts", [
-        { text: "hello", vector: [0.0] },
-      ]);
-      await ftsTable.createIndex("text", { config: Index.fts() });
-
-      const reader = await connect(tmpDir.name, {
-        readConsistencyInterval: 0,
-      });
-      const tracked = await reader.openTable("stale_fts");
-      type Snapshot = {
-        schema: () => Promise<Buffer>;
-      };
-      type NativeWithSnapshot = {
-        querySnapshot: () => Promise<Snapshot>;
-      };
-      const native = (tracked as unknown as { inner: NativeWithSnapshot })
-        .inner;
-      const querySnapshot = native.querySnapshot.bind(native);
-      let snapshotCalls = 0;
-      let markStaleSchemaStarted!: () => void;
-      const staleSchemaStarted = new Promise<void>((resolve) => {
-        markStaleSchemaStarted = resolve;
-      });
-      let releaseStaleSchema!: () => void;
-      const staleSchemaReleased = new Promise<void>((resolve) => {
-        releaseStaleSchema = resolve;
-      });
-      native.querySnapshot = async () => {
-        const snapshot = await querySnapshot();
-        snapshotCalls += 1;
-        if (snapshotCalls === 1) {
-          const schema = snapshot.schema.bind(snapshot);
-          snapshot.schema = async () => {
-            markStaleSchemaStarted();
-            await staleSchemaReleased;
-            return await schema();
-          };
-        }
-        return snapshot;
-      };
-
-      const query = tracked.search("hello");
-      const staleFtsExecution = query.toArray();
-      await staleSchemaStarted;
-
-      const embedding = new RaceEmbedding();
-      const vectorSchema = LanceSchema({
-        text: embedding.sourceField(new arrow.Utf8()),
-        vector: embedding.vectorField(),
-      });
-      await writer.createTable("stale_fts", [{ text: "hello" }], {
-        mode: "overwrite",
-        schema: vectorSchema,
-      });
-
-      const vectorExecution = query.toArray();
-      await vectorStarted;
-      releaseStaleSchema();
-      await staleFtsExecution;
-      releaseVector();
-      await vectorExecution;
-
-      await query.toArray();
-      expect(vectorCalls).toBe(1);
-    });
-
    test("tokenizes FTS queries by column or index name", async () => {
      const db = await connect(tmpDir.name);
      const data = [
@@ -3474,30 +2916,6 @@ describe("column name options", () => {
    expect(results[1].query_index).toBe(1);
  });

-  test("observes promised additional vectors while the query is pending", async () => {
-    const initialVector = new Promise<number[]>(() => undefined);
-    const query = table.query().nearestTo(initialVector);
-    const unhandled: unknown[] = [];
-    const onUnhandled = (reason: unknown) => unhandled.push(reason);
-    process.on("unhandledRejection", onUnhandled);
-
-    try {
-      query.addQueryVector(Promise.reject(new Error("extra vector failed")));
-      await new Promise<void>((resolve) => setImmediate(resolve));
-      expect(unhandled).toEqual([]);
-
-      const rejectedQuery = table
-        .query()
-        .nearestTo([0.1, 0.2])
-        .addQueryVector(Promise.reject(new Error("consumed vector failed")));
-      await expect(rejectedQuery.toArray()).rejects.toThrow(
-        "consumed vector failed",
-      );
-    } finally {
-      process.off("unhandledRejection", onUnhandled);
-    }
-  });
-
  test("index and search multivectors", async () => {
    const db = await connect(tmpDir.name);
    const data = [];
@@ -3535,7 +2953,7 @@ describe("column name options", () => {
      .limit(10)
      .toArray();
    expect(results2.length).toBe(10);
-  }, 30_000);
+  });
 });

 describe("when creating an empty table", () => {
@@ -3561,27 +2979,6 @@ describe("when creating an empty table", () => {
    expect((actualSchema.fields[1].type as Float64).precision).toBe(2);
  });

-  it("can add and query JSON data", async () => {
-    const schema = new Schema([
-      new Field("id", new Int32(), true),
-      new Field(
-        "meta",
-        new Utf8(),
-        true,
-        new Map([["ARROW:extension:name", "arrow.json"]]),
-      ),
-    ]);
-    const table = await con.createEmptyTable("json", schema);
-    const meta = JSON.stringify({ x: 1 });
-
-    await table.add([{ id: 1, meta }]);
-
-    const rows = await table.query().toArray();
-    expect(rows).toHaveLength(1);
-    expect(rows[0].id).toBe(1);
-    expect(rows[0].meta).toBe(meta);
-  });
-
  it("can create an empty table from schema that specifies field types by name", async () => {
    const schemaLike = {
      fields: [
@@ -170,7 +170,7 @@ test("basic table examples", async () => {
    // --8<-- [end:create_index]

    // --8<-- [start:delete_rows]
-    await tbl.delete("item = 'fizz'");
+    await tbl.delete('item = "fizz"');
    // --8<-- [end:delete_rows]

    // --8<-- [start:drop_table]
@@ -5,6 +5,7 @@ import {
  Data as ArrowData,
  Table as ArrowTable,
  Binary,
+  Bool,
  BufferType,
  DataType,
  DateUnit,
@@ -17,7 +18,12 @@ import {
  FixedSizeList,
  Float,
  Float32,
+  Float64,
  Int,
+  Int8,
+  Int16,
+  Int32,
+  Int64,
  LargeBinary,
  List,
  Null,
@@ -30,29 +36,33 @@ import {
  Struct,
  Timestamp,
  Type,
+  Uint8,
+  Uint16,
+  Uint32,
  Utf8,
  Vector,
  makeVector as arrowMakeVector,
-  util as arrowUtil,
  vectorFromArray as badVectorFromArray,
  makeBuilder,
  makeData,
 } from "apache-arrow";
 import { Buffers } from "apache-arrow/data";
-import { typedArrayToArrowType } from "./arrow_type";
 import { type EmbeddingFunction } from "./embedding/embedding_function";
-import {
-  EmbeddingFunctionConfig,
-  getRegistry,
-  parseEmbeddingMetadata,
-} from "./embedding/registry";
+import { EmbeddingFunctionConfig, getRegistry } from "./embedding/registry";
 import {
  sanitizeField,
  sanitizeSchema,
  sanitizeTable,
  sanitizeType,
 } from "./sanitize";
-import { inferSchema } from "./schema";
+
+/**
+ * Check if a field name indicates a vector column.
+ */
+function nameSuggestsVectorColumn(fieldName: string): boolean {
+  const nameLower = fieldName.toLowerCase();
+  return nameLower.includes("vector") || nameLower.includes("embedding");
+}

 export * from "apache-arrow";
 export type SchemaLike =
@@ -72,7 +82,8 @@ export type FieldLike =
    };

 export type DataLike =
-  | import("apache-arrow").Data
+  // biome-ignore lint/suspicious/noExplicitAny: <explanation>
+  | import("apache-arrow").Data<Struct<any>>
  | {
      // biome-ignore lint/suspicious/noExplicitAny: <explanation>
      type: any;
@@ -81,7 +92,6 @@ export type DataLike =
      stride: number;
      nullable: boolean;
      children: DataLike[];
-      dictionary?: { data: readonly DataLike[] };
      get nullCount(): number;
      // biome-ignore lint/suspicious/noExplicitAny: <explanation>
      values: Buffers<any>[BufferType.DATA];
@@ -445,6 +455,110 @@ export function makeArrowTable(
  return new ArrowTable(inferredSchema, finalColumns);
 }

+function inferSchema(
+  data: Array<Record<string, unknown>>,
+  schema: Schema | undefined,
+  opts: MakeArrowTableOptions,
+): Schema {
+  // We will collect all fields we see in the data.
+  const pathTree = new PathTree<DataType>();
+
+  for (const [rowI, row] of data.entries()) {
+    for (const [path, value] of rowPathsAndValues(row)) {
+      if (!pathTree.has(path)) {
+        // First time seeing this field.
+        if (schema !== undefined) {
+          const field = getFieldForPath(schema, path);
+          if (field === undefined) {
+            throw new Error(
+              `Found field not in schema: ${path.join(".")} at row ${rowI}`,
+            );
+          } else {
+            pathTree.set(path, field.type);
+          }
+        } else {
+          const inferredType = inferType(value, path, opts);
+          if (inferredType === undefined) {
+            throw new Error(`Failed to infer data type for field ${path.join(
+              ".",
+            )} at row ${rowI}. \
+                             Consider providing an explicit schema.`);
+          }
+          pathTree.set(path, inferredType);
+        }
+      } else if (schema === undefined) {
+        const currentType = pathTree.get(path);
+        const newType = inferType(value, path, opts);
+        if (currentType !== newType) {
+          new Error(`Failed to infer schema for data. Previously inferred type \
+                     ${currentType} but found ${newType} at row ${rowI}. Consider \
+                     providing an explicit schema.`);
+        }
+      }
+    }
+  }
+
+  if (schema === undefined) {
+    function fieldsFromPathTree(pathTree: PathTree<DataType>): Field[] {
+      const fields = [];
+      for (const [name, value] of pathTree.map.entries()) {
+        if (value instanceof PathTree) {
+          const children = fieldsFromPathTree(value);
+          fields.push(new Field(name, new Struct(children), true));
+        } else {
+          fields.push(new Field(name, value, true));
+        }
+      }
+      return fields;
+    }
+    const fields = fieldsFromPathTree(pathTree);
+    return new Schema(fields);
+  } else {
+    function takeMatchingFields(
+      fields: Field[],
+      pathTree: PathTree<DataType>,
+    ): Field[] {
+      const outFields = [];
+      for (const field of fields) {
+        if (pathTree.map.has(field.name)) {
+          const value = pathTree.get([field.name]);
+          if (value instanceof PathTree) {
+            const struct = field.type as Struct;
+            const children = takeMatchingFields(struct.children, value);
+            outFields.push(
+              new Field(field.name, new Struct(children), field.nullable),
+            );
+          } else {
+            outFields.push(
+              new Field(field.name, value as DataType, field.nullable),
+            );
+          }
+        }
+      }
+      return outFields;
+    }
+    const fields = takeMatchingFields(schema.fields, pathTree);
+    return new Schema(fields);
+  }
+}
+
+function* rowPathsAndValues(
+  row: Record<string, unknown>,
+  basePath: string[] = [],
+): Generator<[string[], unknown]> {
+  for (const [key, value] of Object.entries(row)) {
+    if (isObject(value)) {
+      yield* rowPathsAndValues(value, [...basePath, key]);
+    } else {
+      // Skip undefined values - they should be treated the same as missing fields
+      // for embedding function purposes
+      if (value !== undefined) {
+        yield [[...basePath, key], value];
+      }
+    }
+  }
+}
+
 function isObject(value: unknown): value is Record<string, unknown> {
  return (
    typeof value === "object" &&
@@ -459,19 +573,146 @@ function isObject(value: unknown): value is Record<string, unknown> {
  );
 }

-function valueAtPath(datum: Record<string, unknown>, path: string[]): unknown {
-  let current: unknown = datum;
+function getFieldForPath(schema: Schema, path: string[]): Field | undefined {
+  let current: Field | Schema = schema;
  for (const key of path) {
-    if (current == null) {
-      return null;
-    }
-    if (isObject(current) && (Object.hasOwn(current, key) || key in current)) {
-      current = current[key];
+    if (current instanceof Schema) {
+      const field: Field | undefined = current.fields.find(
+        (f) => f.name === key,
+      );
+      if (field === undefined) {
+        return undefined;
+      }
+      current = field;
+    } else if (current instanceof Field && DataType.isStruct(current.type)) {
+      const struct: Struct = current.type;
+      const field = struct.children.find((f) => f.name === key);
+      if (field === undefined) {
+        return undefined;
+      }
+      current = field;
    } else {
      return undefined;
    }
  }
-  return current;
+  if (current instanceof Field) {
+    return current;
+  } else {
+    return undefined;
+  }
+}
+
+/**
+ * Try to infer which Arrow type to use for a given value.
+ *
+ * May return undefined if the type cannot be inferred.
+ */
+function inferType(
+  value: unknown,
+  path: string[],
+  opts: MakeArrowTableOptions,
+): DataType | undefined {
+  if (typeof value === "bigint") {
+    return new Int64();
+  } else if (typeof value === "number") {
+    // Even if it's an integer, it's safer to assume Float64. Users can
+    // always provide an explicit schema or use BigInt if they mean integer.
+    return new Float64();
+  } else if (typeof value === "string") {
+    if (opts.dictionaryEncodeStrings) {
+      return new Dictionary(new Utf8(), new Int32());
+    } else {
+      return new Utf8();
+    }
+  } else if (typeof value === "boolean") {
+    return new Bool();
+  } else if (value instanceof Buffer) {
+    return new Binary();
+  } else if (ArrayBuffer.isView(value) && !(value instanceof DataView)) {
+    const info = typedArrayToArrowType(value);
+    if (info !== undefined) {
+      const child = new Field("item", info.elementType, true);
+      return new FixedSizeList(info.length, child);
+    }
+    return undefined;
+  } else if (Array.isArray(value)) {
+    if (value.length === 0) {
+      return undefined; // Without any values we can't infer the type
+    }
+    if (path.length === 1 && Object.hasOwn(opts.vectorColumns, path[0])) {
+      const floatType = sanitizeType(opts.vectorColumns[path[0]].type);
+      return new FixedSizeList(
+        value.length,
+        new Field("item", floatType, true),
+      );
+    }
+    const valueType = inferType(value[0], path, opts);
+    if (valueType === undefined) {
+      return undefined;
+    }
+    // Try to automatically detect embedding columns.
+    if (nameSuggestsVectorColumn(path[path.length - 1])) {
+      // Check if value is a Uint8Array for integer vector type determination
+      if (value instanceof Uint8Array) {
+        // For integer vectors, we default to Uint8 (matching Python implementation)
+        const child = new Field("item", new Uint8(), true);
+        return new FixedSizeList(value.length, child);
+      } else {
+        // For float vectors, we default to Float32
+        const child = new Field("item", new Float32(), true);
+        return new FixedSizeList(value.length, child);
+      }
+    } else {
+      const child = new Field("item", valueType, true);
+      return new List(child);
+    }
+  } else {
+    // TODO: timestamp
+    return undefined;
+  }
+}
+
+class PathTree<V> {
+  map: Map<string, V | PathTree<V>>;
+
+  constructor(entries?: [string[], V][]) {
+    this.map = new Map();
+    if (entries !== undefined) {
+      for (const [path, value] of entries) {
+        this.set(path, value);
+      }
+    }
+  }
+  has(path: string[]): boolean {
+    let ref: PathTree<V> = this;
+    for (const part of path) {
+      if (!(ref instanceof PathTree) || !ref.map.has(part)) {
+        return false;
+      }
+      ref = ref.map.get(part) as PathTree<V>;
+    }
+    return true;
+  }
+  get(path: string[]): V | undefined {
+    let ref: PathTree<V> = this;
+    for (const part of path) {
+      if (!(ref instanceof PathTree) || !ref.map.has(part)) {
+        return undefined;
+      }
+      ref = ref.map.get(part) as PathTree<V>;
+    }
+    return ref as V;
+  }
+  set(path: string[], value: V): void {
+    let ref: PathTree<V> = this;
+    for (const part of path.slice(0, path.length - 1)) {
+      if (!ref.map.has(part)) {
+        ref.map.set(part, new PathTree<V>());
+      }
+      ref = ref.map.get(part) as PathTree<V>;
+    }
+    ref.map.set(path[path.length - 1], value);
+  }
 }

 function transposeData(
@@ -479,26 +720,37 @@ function transposeData(
  field: Field,
  path: string[] = [],
 ): Vector {
-  const valuesPath = [...path, field.name];
-  const values = data.map((datum) => valueAtPath(datum, valuesPath));
  if (field.type instanceof Struct) {
    const childFields = field.type.children;
+    const fullPath = [...path, field.name];
    const childVectors = childFields.map((child) => {
-      return transposeData(data, child, valuesPath);
+      return transposeData(data, child, fullPath);
    });
-    const nullCount = values.filter((value) => value === null).length;
    const structData = makeData({
      type: field.type,
-      length: values.length,
-      nullCount,
-      nullBitmap:
-        nullCount > 0
-          ? arrowUtil.packBools(values.map((value) => value !== null))
-          : undefined,
      children: childVectors as unknown as ArrowData<DataType>[],
    });
    return arrowMakeVector(structData);
  } else {
+    const valuesPath = [...path, field.name];
+    const values = data.map((datum) => {
+      let current: unknown = datum;
+      for (const key of valuesPath) {
+        if (current == null) {
+          return null;
+        }
+
+        if (
+          isObject(current) &&
+          (Object.hasOwn(current, key) || key in current)
+        ) {
+          current = current[key];
+        } else {
+          return null;
+        }
+      }
+      return current;
+    });
    return makeVector(values, field.type, undefined, field.nullable);
  }
 }
@@ -541,6 +793,32 @@ function makeListVector(lists: unknown[][]): Vector<unknown> {
  return listBuilder.finish().toVector();
 }

+/**
+ * Map a JS TypedArray instance to the corresponding Arrow element DataType
+ * and its length. Returns undefined if the value is not a recognized TypedArray.
+ */
+function typedArrayToArrowType(
+  value: ArrayBufferView,
+): { elementType: DataType; length: number } | undefined {
+  if (value instanceof Float32Array)
+    return { elementType: new Float32(), length: value.length };
+  if (value instanceof Float64Array)
+    return { elementType: new Float64(), length: value.length };
+  if (value instanceof Uint8Array)
+    return { elementType: new Uint8(), length: value.length };
+  if (value instanceof Uint16Array)
+    return { elementType: new Uint16(), length: value.length };
+  if (value instanceof Uint32Array)
+    return { elementType: new Uint32(), length: value.length };
+  if (value instanceof Int8Array)
+    return { elementType: new Int8(), length: value.length };
+  if (value instanceof Int16Array)
+    return { elementType: new Int16(), length: value.length };
+  if (value instanceof Int32Array)
+    return { elementType: new Int32(), length: value.length };
+  return undefined;
+}
+
 /** Helper function to convert an Array of JS values to an Arrow Vector */
 function makeVector(
  values: unknown[],
@@ -655,7 +933,7 @@ async function applyEmbeddingsFromMetadata(

  for (const functionEntry of functions.values()) {
    const sourceColumn = columns[functionEntry.sourceColumn];
-    const destColumn = functionEntry.vectorColumn;
+    const destColumn = functionEntry.vectorColumn ?? "vector";
    if (sourceColumn === undefined) {
      throw new Error(
        `Cannot apply embedding function because the source column '${functionEntry.sourceColumn}' was not present in the data`,
@@ -1107,10 +1385,11 @@ function validateSchemaEmbeddings(

        // Check schema metadata for embedding functions
        if (schema.metadata.has("embedding_functions")) {
-          const entries = parseEmbeddingMetadata(
+          const embeddings = JSON.parse(
            schema.metadata.get("embedding_functions")!,
          );
-          if (entries.some((f) => f.vectorColumn === field.name)) {
+          // biome-ignore lint/suspicious/noExplicitAny: we don't know the type of `f`
+          if (embeddings.find((f: any) => f["vectorColumn"] === field.name)) {
            hasEmbeddingFunction = true;
          }
        }
@@ -1180,12 +1459,8 @@ export function ensureNestedFieldsExist(
          completeRow[field.name] = row[field.name];
        }
      } else {
-        // Keep a missing struct valid while filling each of its children with
-        // null. This is distinct from an explicitly null struct value.
-        completeRow[field.name] =
-          field.type.constructor.name === "Struct"
-            ? ensureStructFieldsExist({}, field.type as Struct)
-            : null;
+        // Field is missing from the data - set to null
+        completeRow[field.name] = null;
      }
    }

@@ -1220,12 +1495,8 @@ function ensureStructFieldsExist(
        completeStruct[childField.name] = data[childField.name];
      }
    } else {
-      // Keep a missing struct valid while filling each of its children with
-      // null. This is distinct from an explicitly null struct value.
-      completeStruct[childField.name] =
-        childField.type.constructor.name === "Struct"
-          ? ensureStructFieldsExist({}, childField.type as Struct)
-          : null;
+      // Field is missing - set to null
+      completeStruct[childField.name] = null;
    }
  }

@@ -1,40 +0,0 @@
-// SPDX-License-Identifier: Apache-2.0
-// SPDX-FileCopyrightText: Copyright The LanceDB Authors
-
-import {
-  type DataType,
-  Float32,
-  Float64,
-  Int8,
-  Int16,
-  Int32,
-  Uint8,
-  Uint16,
-  Uint32,
-} from "apache-arrow";
-
-/**
- * Map a JS TypedArray instance to the corresponding Arrow element type and
- * length. Returns undefined when the view is not a supported TypedArray.
- */
-export function typedArrayToArrowType(
-  value: ArrayBufferView,
-): { elementType: DataType; length: number } | undefined {
-  if (value instanceof Float32Array)
-    return { elementType: new Float32(), length: value.length };
-  if (value instanceof Float64Array)
-    return { elementType: new Float64(), length: value.length };
-  if (value instanceof Uint8Array)
-    return { elementType: new Uint8(), length: value.length };
-  if (value instanceof Uint16Array)
-    return { elementType: new Uint16(), length: value.length };
-  if (value instanceof Uint32Array)
-    return { elementType: new Uint32(), length: value.length };
-  if (value instanceof Int8Array)
-    return { elementType: new Int8(), length: value.length };
-  if (value instanceof Int16Array)
-    return { elementType: new Int16(), length: value.length };
-  if (value instanceof Int32Array)
-    return { elementType: new Int32(), length: value.length };
-  return undefined;
-}
@@ -16,12 +16,6 @@ import {
  makeEmptyTable,
 } from "./arrow";
 import { EmbeddingFunctionConfig, getRegistry } from "./embedding/registry";
-import {
-  MaterializedView,
-  MaterializedViewSelect,
-  normalizeSelect,
-  validateNonNegativeInteger,
-} from "./materialized_view";
 import { Connection as LanceDbConnection } from "./native";
 import type {
  CreateNamespaceResponse,
@@ -31,14 +25,12 @@ import type {
  JobDescription,
  JobInfo,
  ListNamespacesResponse,
-  ListTablesResponse,
 } from "./native";
 export type {
  CreateNamespaceResponse,
  DescribeNamespaceResponse,
  DropNamespaceResponse,
  ListNamespacesResponse,
-  ListTablesResponse,
 };
 import { sanitizeTable } from "./sanitize";
 import { LocalTable, Table } from "./table";
@@ -136,10 +128,6 @@ export interface OpenTableOptions {
  indexCacheSize?: number;
 }

-/**
- * @deprecated Use {@link ListTablesOptions} with {@link Connection.listTables}
- * instead.
- */
 export interface TableNamesOptions {
  /**
   * If present, only return names that come lexicographically after the
@@ -153,24 +141,6 @@ export interface TableNamesOptions {
  limit?: number;
 }

-export interface ListTablesOptions {
-  /**
-   * Token from a previous response, to resume listing where it left off.
-   *
-   * The token is opaque: it carries whatever the database needs to resume, and
-   * callers should not construct or interpret one.
-   */
-  pageToken?: string;
-  /**
-   * An upper bound on how many tables to return.
-   *
-   * A page may hold fewer than this and still not be the last one, so keep
-   * going while the response carries a page token rather than while pages are
-   * full.
-   */
-  limit?: number;
-}
-
 export interface ListNamespacesOptions {
  /** Token from a previous response for pagination. */
  pageToken?: string;
@@ -255,7 +225,6 @@ export abstract class Connection {
   * @param {Partial<TableNamesOptions>} options - options to control the
   * paging / start point (backwards compatibility)
   *
-   * @deprecated Use {@link Connection.listTables} instead.
   */
  abstract tableNames(options?: Partial<TableNamesOptions>): Promise<string[]>;
  /**
@@ -266,94 +235,18 @@ export abstract class Connection {
   * @param {Partial<TableNamesOptions>} options - options to control the
   * paging / start point
   *
-   * @deprecated Use {@link Connection.listTables} instead.
   */
  abstract tableNames(
    namespacePath?: string[],
    options?: Partial<TableNamesOptions>,
  ): Promise<string[]>;

-  /**
-   * List a page of the tables in this database.
-   *
-   * To retrieve the tables after the page, pass the `pageToken` the response
-   * carries back in. A page can be shorter than `limit` without being the last
-   * one, so walk until a response carries no page token:
-   *
-   * ```ts
-   * const names = [];
-   * let pageToken = undefined;
-   * do {
-   *   const page = await conn.listTables({ pageToken, limit: 100 });
-   *   names.push(...page.tables);
-   *   pageToken = page.pageToken;
-   * } while (pageToken);
-   * ```
-   *
-   * @param {Partial<ListTablesOptions>} options - Pagination options
-   *   (`pageToken`, `limit`).
-   * @returns {Promise<ListTablesResponse>} A page of table names and an
-   *   optional token for the tables after it.
-   */
-  abstract listTables(
-    options?: Partial<ListTablesOptions>,
-  ): Promise<ListTablesResponse>;
-  /**
-   * List a page of the tables in this database.
-   *
-   * @param {string[]} namespacePath - The namespace path to list tables from
-   *   (defaults to root namespace)
-   * @param {Partial<ListTablesOptions>} options - Pagination options
-   *   (`pageToken`, `limit`).
-   * @returns {Promise<ListTablesResponse>} A page of table names and an
-   *   optional token for the tables after it.
-   */
-  abstract listTables(
-    namespacePath?: string[],
-    options?: Partial<ListTablesOptions>,
-  ): Promise<ListTablesResponse>;
-
  /**
   * Open a table in the database.
   * @param {string} name - The name of the table
   * @param {string[]} namespacePath - The namespace path of the table (defaults to root namespace)
   * @param {Partial<OpenTableOptions>} options - Additional options
   */
-  /**
-   * Define a materialized view named `name` over the table `source`.
-   *
-   * The view is created empty, with the query recorded in its schema
-   * metadata; `view.refresh()` computes the rows. The view is a normal
-   * table: it can be queried, indexed and searched, and it appears in
-   * `tableNames`. The source table must have stable row ids (create it with
-   * the `newTableEnableStableRowIds` storage option); they keep the view's
-   * provenance valid across source compactions and cannot be enabled after
-   * a table exists. Local databases only.
-   */
-  abstract createMaterializedView(
-    name: string,
-    source: string,
-    options?: {
-      select?: MaterializedViewSelect;
-      where?: string;
-      limit?: number;
-    },
-  ): Promise<MaterializedView>;
-
-  /**
-   * Open the materialized view named `name`.
-   *
-   * Rejects a table that exists but is not a materialized view.
-   */
-  abstract openMaterializedView(name: string): Promise<MaterializedView>;
-
-  /**
-   * The names of the materialized views in this database.
-   *
-   * Found by reading every table's schema, so this costs an open per table.
-   */
-  abstract listMaterializedViews(): Promise<string[]>;
-
  abstract openTable(
    name: string,
    namespacePath?: string[],
@@ -638,54 +531,6 @@ export class LocalConnection extends Connection {
    );
  }

-  async createMaterializedView(
-    name: string,
-    source: string,
-    options?: {
-      select?: MaterializedViewSelect;
-      where?: string;
-      limit?: number;
-    },
-  ): Promise<MaterializedView> {
-    validateNonNegativeInteger(options?.limit, "limit");
-    const innerTable = await this.inner.createMaterializedView(
-      name,
-      source,
-      normalizeSelect(options?.select),
-      options?.where,
-      options?.limit,
-    );
-    return new MaterializedView(new LocalTable(innerTable));
-  }
-
-  async openMaterializedView(name: string): Promise<MaterializedView> {
-    const innerTable = await this.inner.openMaterializedView(name);
-    return new MaterializedView(new LocalTable(innerTable));
-  }
-
-  async listMaterializedViews(): Promise<string[]> {
-    return await this.inner.listMaterializedViews();
-  }
-
-  async listTables(
-    namespacePathOrOptions?: string[] | Partial<ListTablesOptions>,
-    options?: Partial<ListTablesOptions>,
-  ): Promise<ListTablesResponse> {
-    // Detect if first argument is namespacePath array or options object
-    const namespacePath = Array.isArray(namespacePathOrOptions)
-      ? namespacePathOrOptions
-      : undefined;
-    const listTablesOptions = Array.isArray(namespacePathOrOptions)
-      ? options
-      : namespacePathOrOptions;
-
-    return this.inner.listTables(
-      namespacePath ?? [],
-      listTablesOptions?.pageToken,
-      listTablesOptions?.limit,
-    );
-  }
-
  async openTable(
    name: string,
    namespacePath?: string[],
@@ -4,15 +4,7 @@
 import { Field, Schema } from "../arrow";
 import { sanitizeType } from "../sanitize";
 import { EmbeddingFunction } from "./embedding_function";
-import {
-  EmbeddingFunctionConfig,
-  EmbeddingFunctionRegistry,
-  getRegistry as getGlobalRegistry,
-  registerBuiltIn,
-} from "./registry";
-
-type OpenAIModule = typeof import("./openai");
-type TransformersModule = typeof import("./transformers");
+import { EmbeddingFunctionConfig, getRegistry } from "./registry";

 export {
  FieldOptions,
@@ -22,39 +14,7 @@ export {
  EmbeddingFunctionConstructor,
 } from "./embedding_function";

-export {
-  EmbeddingFunctionRegistry,
-  parseEmbeddingMetadata,
-  register,
-} from "./registry";
-export type {
-  CreateReturnType,
-  EmbeddingFunctionConfig,
-  EmbeddingFunctionCreate,
-  EmbeddingMetadataEntry,
-  ResolvedEmbeddingFunctionConfig,
-} from "./registry";
-
-function initializeBuiltInProviders() {
-  const { OpenAIEmbeddingFunction } = require("./openai") as OpenAIModule;
-  const { TransformersEmbeddingFunction } =
-    require("./transformers") as TransformersModule;
-
-  registerBuiltIn("openai", OpenAIEmbeddingFunction);
-  registerBuiltIn("huggingface", TransformersEmbeddingFunction);
-}
-
-/**
- * Get the global embedding function registry.
- *
- * LanceDB built-in providers are initialized when this public API is first
- * used, so importing the root package does not change automatic search
- * selection for tables without embedding metadata.
- */
-export function getRegistry(): EmbeddingFunctionRegistry {
-  initializeBuiltInProviders();
-  return getGlobalRegistry();
-}
+export * from "./registry";

 /**
 * Create a schema with embedding functions.
@@ -5,13 +5,14 @@ import type OpenAI from "openai";
 import type { EmbeddingCreateParams } from "openai/resources/index";
 import { Float, Float32 } from "../arrow";
 import { EmbeddingFunction } from "./embedding_function";
-import { registerBuiltIn } from "./registry";
+import { register } from "./registry";

 export type OpenAIOptions = {
  apiKey: string;
  model: EmbeddingCreateParams["model"];
 };

+@register("openai")
 export class OpenAIEmbeddingFunction extends EmbeddingFunction<
  string,
  Partial<OpenAIOptions>
@@ -99,5 +100,3 @@ export class OpenAIEmbeddingFunction extends EmbeddingFunction<
    return response.data[0].embedding;
  }
 }
-
-registerBuiltIn("openai", OpenAIEmbeddingFunction);
@@ -7,10 +7,6 @@ import {
 } from "./embedding_function";
 import "reflect-metadata";

-const builtInFunctionsKey = Symbol.for(
-  "@lancedb/lancedb::embedding-built-in-functions::v1",
-);
-
 export type CreateReturnType<T> = T extends { init: () => Promise<void> }
  ? Promise<T>
  : T;
@@ -63,15 +59,6 @@ export class EmbeddingFunctionRegistry {
    };
  }

-  /** @ignore */
-  setBuiltIn<
-    T extends EmbeddingFunctionConstructor = EmbeddingFunctionConstructor,
-  >(name: string, ctor: T): T {
-    this.#functions.set(name, ctor);
-    Reflect.defineMetadata("lancedb::embedding::name", name, ctor);
-    return ctor;
-  }
-
  get<T extends EmbeddingFunction<unknown>>(
    name: string,
  ): EmbeddingFunctionCreate<T> | undefined;
@@ -109,7 +96,6 @@ export class EmbeddingFunctionRegistry {
   */
  reset(this: EmbeddingFunctionRegistry) {
    this.#functions.clear();
-    getBuiltInFunctions(this).clear();
  }

  /**
@@ -118,29 +104,41 @@ export class EmbeddingFunctionRegistry {
  async parseFunctions(
    this: EmbeddingFunctionRegistry,
    metadata: Map<string, string>,
-  ): Promise<Map<string, ResolvedEmbeddingFunctionConfig>> {
+  ): Promise<Map<string, EmbeddingFunctionConfig>> {
    if (!metadata.has("embedding_functions")) {
      return new Map();
+    } else {
+      type FunctionConfig = {
+        name: string;
+        sourceColumn: string;
+        vectorColumn: string;
+        model: EmbeddingFunction["TOptions"];
+      };
+
+      const functions = <FunctionConfig[]>(
+        JSON.parse(metadata.get("embedding_functions")!)
+      );
+
+      const items: [string, EmbeddingFunctionConfig][] = await Promise.all(
+        functions.map(async (f) => {
+          const fn = this.get(f.name);
+          if (!fn) {
+            throw new Error(`Function "${f.name}" not found in registry`);
+          }
+          const func = await this.get(f.name)!.create(f.model);
+          return [
+            f.name,
+            {
+              sourceColumn: f.sourceColumn,
+              vectorColumn: f.vectorColumn,
+              function: func,
+            },
+          ];
+        }),
+      );
+
+      return new Map(items);
    }
-    const entries = parseEmbeddingMetadata(
-      metadata.get("embedding_functions")!,
-    );
-    const items = await Promise.all(
-      entries.map(async (f): Promise<ResolvedEmbeddingFunctionConfig> => {
-        const fn = this.get(f.name);
-        if (!fn) {
-          throw new Error(`Function "${f.name}" not found in registry`);
-        }
-        const func = await fn.create(f.model);
-        return {
-          sourceColumn: f.sourceColumn,
-          vectorColumn: f.vectorColumn,
-          function: func,
-        };
-      }),
-    );
-    // Keyed by output column: one function may serve several columns.
-    return new Map(items.map((config) => [config.vectorColumn, config]));
  }
  // biome-ignore lint/suspicious/noExplicitAny: <explanation>
  functionToMetadata(conf: EmbeddingFunctionConfig): Record<string, any> {
@@ -197,56 +195,12 @@ export class EmbeddingFunctionRegistry {
  }
 }

-function getBuiltInFunctions(registry: EmbeddingFunctionRegistry): Set<string> {
-  const registryWithBuiltIns = registry as EmbeddingFunctionRegistry & {
-    [key: symbol]: Set<string> | undefined;
-  };
-  let builtInFunctions = registryWithBuiltIns[builtInFunctionsKey];
-  if (builtInFunctions === undefined) {
-    builtInFunctions = new Set<string>();
-    registryWithBuiltIns[builtInFunctionsKey] = builtInFunctions;
-  }
-  return builtInFunctions;
-}
-
-// Server bundlers can load the side-effect embedding entry points and the public
-// embedding API from separate module graphs. Keep their registry shared.
-const registryKey = Symbol.for(
-  "@lancedb/lancedb::embedding-function-registry::v1",
-);
-const registryGlobal = globalThis as typeof globalThis & {
-  [key: symbol]: EmbeddingFunctionRegistry | undefined;
-};
-
-function getGlobalRegistry(): EmbeddingFunctionRegistry {
-  const existingRegistry = registryGlobal[registryKey];
-  if (existingRegistry !== undefined) {
-    return existingRegistry;
-  }
-  const registry = new EmbeddingFunctionRegistry();
-  registryGlobal[registryKey] = registry;
-  return registry;
-}
-
-const _REGISTRY = getGlobalRegistry();
+const _REGISTRY = new EmbeddingFunctionRegistry();

 export function register(name?: string) {
  return _REGISTRY.register(name);
 }

-/** @ignore */
-export function registerBuiltIn<
-  T extends EmbeddingFunctionConstructor = EmbeddingFunctionConstructor,
->(name: string, ctor: T): T {
-  const builtInFunctions = getBuiltInFunctions(_REGISTRY);
-  if (builtInFunctions.has(name)) {
-    return _REGISTRY.setBuiltIn(name, ctor);
-  }
-  _REGISTRY.register(name)(ctor);
-  builtInFunctions.add(name);
-  return ctor;
-}
-
 /**
 * Utility function to get the global instance of the registry
 * @returns `EmbeddingFunctionRegistry` The global instance of the registry
@@ -264,52 +218,3 @@ export interface EmbeddingFunctionConfig {
  vectorColumn?: string;
  function: EmbeddingFunction;
 }
-
-/** An [EmbeddingFunctionConfig] read back from table metadata, where the
- * vector column is always recorded. */
-export type ResolvedEmbeddingFunctionConfig = EmbeddingFunctionConfig & {
-  vectorColumn: string;
-};
-
-/** One entry of the `embedding_functions` schema metadata, with the column
- * keys normalized across the bindings' spellings. */
-export type EmbeddingMetadataEntry = {
-  name: string;
-  sourceColumn: string;
-  vectorColumn: string;
-  model: EmbeddingFunction["TOptions"];
-};
-
-/** The single parser for `embedding_functions` schema metadata: every reader
- * goes through here, so the wire contract cannot fork between them. */
-export function parseEmbeddingMetadata(json: string): EmbeddingMetadataEntry[] {
-  // The wire format, honestly: the Python bindings write snake_case keys.
-  type Raw = {
-    name: string;
-    sourceColumn?: string;
-    // biome-ignore lint/style/useNamingConvention: the Python wire spelling
-    source_column?: string;
-    vectorColumn?: string;
-    // biome-ignore lint/style/useNamingConvention: the Python wire spelling
-    vector_column?: string;
-    model: EmbeddingFunction["TOptions"];
-  };
-  const entries = <Raw[]>JSON.parse(json);
-  const seen = new Set<string>();
-  return entries.map((f) => {
-    const sourceColumn = f.sourceColumn ?? f.source_column;
-    const vectorColumn = f.vectorColumn ?? f.vector_column;
-    if (sourceColumn === undefined || vectorColumn === undefined) {
-      throw new Error(
-        `Embedding function "${f.name}" metadata names no source or vector column`,
-      );
-    }
-    if (seen.has(vectorColumn)) {
-      throw new Error(
-        `Multiple embedding configs claim vector column "${vectorColumn}"`,
-      );
-    }
-    seen.add(vectorColumn);
-    return { name: f.name, sourceColumn, vectorColumn, model: f.model };
-  });
-}
@@ -3,7 +3,7 @@

 import { Float, Float32 } from "../arrow";
 import { EmbeddingFunction } from "./embedding_function";
-import { registerBuiltIn } from "./registry";
+import { register } from "./registry";

 export type XenovaTransformerOptions = {
  /** The wasm compatible model to use */
@@ -31,6 +31,7 @@ export type XenovaTransformerOptions = {
  };
 };

+@register("huggingface")
 export class TransformersEmbeddingFunction extends EmbeddingFunction<
  string,
  Partial<XenovaTransformerOptions>
@@ -157,8 +158,6 @@ export class TransformersEmbeddingFunction extends EmbeddingFunction<
  }
 }

-registerBuiltIn("huggingface", TransformersEmbeddingFunction);
-
 const tensorDiv = (
  src: import("@huggingface/transformers").Tensor,
  divBy: number,
@@ -21,11 +21,6 @@ import type { BaseTokenizer } from "./indices";
 import type { FtsToken } from "./table";

 // Re-export native header provider for use with connectWithHeaderProvider
-export {
-  MaterializedView,
-  MaterializedViewDefinition,
-  MaterializedViewSelect,
-} from "./materialized_view";
 export { JsHeaderProvider as NativeJsHeaderProvider } from "./native.js";

 // OpenTelemetry metrics bridge. Only the high-level entry point is public; the
@@ -56,7 +51,6 @@ export {
  AddResult,
  AddColumnsResult,
  RefreshColumnResult,
-  RefreshMaterializedViewResult,
  AlterColumnsResult,
  UpdateFieldMetadataResult,
  DeleteResult,
@@ -81,13 +75,11 @@ export {
  Connection,
  CreateTableOptions,
  TableNamesOptions,
-  ListTablesOptions,
  OpenTableOptions,
  ListNamespacesOptions,
  CreateNamespaceOptions,
  DropNamespaceOptions,
  ListNamespacesResponse,
-  ListTablesResponse,
  CreateNamespaceResponse,
  DropNamespaceResponse,
  DescribeNamespaceResponse,
@@ -103,7 +95,6 @@ export {
 } from "./native.js";

 export {
-  AutoQuery,
  ExecutableQuery,
  Query,
  QueryBase,
@@ -144,10 +135,10 @@ export {
  BranchColumnChange,
  BranchIndexSummary,
  BranchRowCountSummary,
-  CherryPickError,
+  MergeBlocker,
  BranchDiff,
-  CherryPickPreview,
-  CherryPickResult,
+  MergePreview,
+  MergeBranchResult,
  AddDataOptions,
  UpdateOptions,
  OptimizeOptions,
@@ -1,161 +0,0 @@
-// SPDX-License-Identifier: Apache-2.0
-// SPDX-FileCopyrightText: Copyright The LanceDB Authors
-
-import { RefreshMaterializedViewResult } from "./native";
-import { Table } from "./table";
-
-/** Schema metadata key holding a materialized view's definition. */
-export const DEFINITION_META_KEY = "mv.definition";
-
-/** The query that defines a materialized view. */
-export interface MaterializedViewDefinition {
-  /** Name of the source table, in the same database as the view. */
-  sourceTable: string;
-  /** `[output column, SQL expression]` pairs, in view schema order. */
-  projections: [string, string][];
-  /** SQL predicate selecting the source rows the view holds. */
-  filter?: string;
-  /** Cap on the number of rows the view holds. */
-  limit?: number;
-  /** Source columns the projections and filter read. */
-  inputs: string[];
-}
-
-/**
- * The view's columns: column names, `[alias, SQL expression]` pairs, or a
- * record of the same. A bare name projects itself.
- */
-export type MaterializedViewSelect =
-  | (string | [string, string])[]
-  | Record<string, string>;
-
-/**
- * @internal Reject a numeric option N-API would otherwise silently coerce:
- * `Infinity` reaches Rust as 0, `1.5` as 1.
- */
-export function validateNonNegativeInteger(
-  value: number | undefined,
-  name: string,
-): void {
-  if (value !== undefined && !(Number.isSafeInteger(value) && value >= 0)) {
-    throw new Error(`${name} must be a non-negative integer`);
-  }
-}
-
-/** @internal Quote a column name as a Lance SQL identifier (backticks). */
-function quoteIdentifier(name: string): string {
-  return "`" + name.replace(/`/g, "``") + "`";
-}
-
-/**
- * @internal Normalize a select argument into `[alias, expression]` pairs.
- * A bare name projects itself and is quoted, so any valid column name works;
- * pair and record entries are kept verbatim because their right side is an
- * expression.
- */
-export function normalizeSelect(
-  select?: MaterializedViewSelect,
-): [string, string][] | undefined {
-  if (select === undefined) {
-    return undefined;
-  }
-  if (Array.isArray(select)) {
-    return select.map((item) =>
-      typeof item === "string" ? [item, quoteIdentifier(item)] : item,
-    );
-  }
-  return Object.entries(select);
-}
-
-/** @internal Parse a definition off a table's stored schema metadata. */
-export function definitionFromMetadata(
-  metadata: Map<string, string>,
-  name: string,
-): MaterializedViewDefinition {
-  const raw = metadata.get(DEFINITION_META_KEY);
-  if (raw === undefined) {
-    throw new Error(`Table '${name}' is not a materialized view`);
-  }
-  // biome-ignore lint/suspicious/noExplicitAny: raw JSON
-  const value: any = JSON.parse(raw);
-  if (value.kind !== "select") {
-    throw new Error(
-      `materialized view '${name}' is defined by '${value.kind}', which this ` +
-        "version of lancedb cannot refresh",
-    );
-  }
-  const limit = value.limit ?? undefined;
-  // JSON.parse rounds integers past 2^53; every exact u64 parses to a safe
-  // integer and every rounded one does not, so this rejects precisely the
-  // values a number cannot carry.
-  if (limit !== undefined && !Number.isSafeInteger(limit)) {
-    throw new Error(
-      `materialized view '${name}' has a stored limit too large to represent exactly`,
-    );
-  }
-  return {
-    sourceTable: value.source_table,
-    // biome-ignore lint/suspicious/noExplicitAny: raw JSON
-    projections: (value.projections ?? []).map((p: any) => [
-      p.output,
-      p.expression,
-    ]),
-    filter: value.filter ?? undefined,
-    limit,
-    inputs: value.inputs ?? [],
-  };
-}
-
-/**
- * A handle on a materialized view: its table plus its definition.
- *
- * Obtained from {@link Connection#createMaterializedView} or
- * {@link Connection#openMaterializedView}. The view is a normal table --
- * queries, indexes and search all apply through {@link MaterializedView#table}
- * -- whose contents are maintained by {@link MaterializedView#refresh}.
- */
-export class MaterializedView {
-  private readonly inner: Table;
-
-  constructor(table: Table) {
-    this.inner = table;
-  }
-
-  get name(): string {
-    return this.inner.name;
-  }
-
-  /** The view, as the table it is. */
-  table(): Table {
-    return this.inner;
-  }
-
-  /** The query that defines the view, read from its stored schema. */
-  async definition(): Promise<MaterializedViewDefinition> {
-    const schema = await this.inner.schema();
-    return definitionFromMetadata(schema.metadata, this.name);
-  }
-
-  /**
-   * Recompute the view from its source.
-   *
-   * The refresh is incremental when the source's changes can be reconciled
-   * into the view -- rows added, changed or removed since the last one --
-   * and otherwise rebuilds. `full` forces a rebuild; `sourceVersion`
-   * refreshes to that source version instead of the latest.
-   *
-   * Concurrent refreshes of one view do not duplicate its rows. Two that
-   * plan the same source rows conflict on commit, and the loser throws
-   * rather than writing them a second time.
-   */
-  async refresh(options?: {
-    full?: boolean;
-    sourceVersion?: number;
-  }): Promise<RefreshMaterializedViewResult> {
-    validateNonNegativeInteger(options?.sourceVersion, "sourceVersion");
-    return await this.inner.refreshMaterializedView(
-      options?.full,
-      options?.sourceVersion,
-    );
-  }
-}
@@ -100,29 +100,6 @@ export interface FullTextSearchOptions {
  columns?: string | string[];
 }

-function nearestToNative(
-  inner: NativeQuery,
-  vector: Awaited<IntoVector>,
-): NativeVectorQuery {
-  const raw = Array.isArray(vector) ? null : extractVectorBuffer(vector);
-  if (raw) {
-    return inner.nearestToRaw(raw.data, raw.dtype);
-  }
-  return inner.nearestTo(Float32Array.from(vector as number[]));
-}
-
-function addQueryVectorToNative(
-  inner: NativeVectorQuery,
-  vector: Awaited<IntoVector>,
-) {
-  const raw = Array.isArray(vector) ? null : extractVectorBuffer(vector);
-  if (raw) {
-    inner.addQueryVectorRaw(raw.data, raw.dtype);
-  } else {
-    inner.addQueryVector(Float32Array.from(vector as number[]));
-  }
-}
-
 /** Common methods supported by all query types
 *
 * @see {@link Query}
@@ -134,15 +111,13 @@ export class QueryBase<
  NativeQueryType extends NativeQuery | NativeVectorQuery | NativeTakeQuery,
 > implements AsyncIterable<RecordBatch>
 {
-  protected inner!: NativeQueryType | Promise<NativeQueryType>;
-
  /**
   * @hidden
   */
-  protected constructor(inner?: NativeQueryType | Promise<NativeQueryType>) {
-    if (inner !== undefined) {
-      this.inner = inner;
-    }
+  protected constructor(
+    protected inner: NativeQueryType | Promise<NativeQueryType>,
+  ) {
+    // intentionally empty
  }

  // call a function on the inner (either a promise or the actual object)
@@ -160,15 +135,6 @@ export class QueryBase<
    }
  }

-  /**
-   * Return the native query used by the next terminal operation.
-   *
-   * @hidden
-   */
-  protected async getInner(): Promise<NativeQueryType> {
-    return this.inner;
-  }
-
  /**
   * Return only the specified columns.
   *
@@ -241,11 +207,16 @@ export class QueryBase<
  /**
   * @hidden
   */
-  protected async nativeExecute(
+  protected nativeExecute(
    options?: Partial<QueryExecutionOptions>,
  ): Promise<NativeBatchIterator> {
-    const inner = await this.getInner();
-    return inner.execute(options?.maxBatchLength, options?.timeoutMs);
+    if (this.inner instanceof Promise) {
+      return this.inner.then((inner) =>
+        inner.execute(options?.maxBatchLength, options?.timeoutMs),
+      );
+    } else {
+      return this.inner.execute(options?.maxBatchLength, options?.timeoutMs);
+    }
  }

  /**
@@ -274,7 +245,12 @@ export class QueryBase<
  /** Collect the results as an Arrow @see {@link ArrowTable}. */
  async toArrow(options?: Partial<QueryExecutionOptions>): Promise<ArrowTable> {
    const batches = [];
-    const inner = await this.getInner();
+    let inner;
+    if (this.inner instanceof Promise) {
+      inner = await this.inner;
+    } else {
+      inner = this.inner;
+    }
    for await (const batch of new RecordBatchIterable(inner, options)) {
      batches.push(batch);
    }
@@ -303,8 +279,11 @@ export class QueryBase<
   * @returns A Promise that resolves to a string containing the query execution plan explanation.
   */
  async explainPlan(verbose = false): Promise<string> {
-    const inner = await this.getInner();
-    return inner.explainPlan(verbose);
+    if (this.inner instanceof Promise) {
+      return this.inner.then((inner) => inner.explainPlan(verbose));
+    } else {
+      return this.inner.explainPlan(verbose);
+    }
  }

  /**
@@ -342,8 +321,13 @@ export class QueryBase<
    distributedMetrics?: AnalyzePlanDistributedMetrics,
  ): Promise<string> {
    const distributedMetricsMode = distributedMetrics ?? "aggregate";
-    const inner = await this.getInner();
-    return inner.analyzePlan(distributedMetricsMode);
+    if (this.inner instanceof Promise) {
+      return this.inner.then((inner) =>
+        inner.analyzePlan(distributedMetricsMode),
+      );
+    } else {
+      return this.inner.analyzePlan(distributedMetricsMode);
+    }
  }

  /**
@@ -355,8 +339,12 @@ export class QueryBase<
   * @returns An Arrow Schema describing the output columns.
   */
  async outputSchema(): Promise<import("./arrow").Schema> {
-    const inner = await this.getInner();
-    const schemaBuffer = await inner.outputSchema();
+    let schemaBuffer: Buffer;
+    if (this.inner instanceof Promise) {
+      schemaBuffer = await this.inner.then((inner) => inner.outputSchema());
+    } else {
+      schemaBuffer = await this.inner.outputSchema();
+    }
    const schema = tableFromIPC(schemaBuffer).schema;
    return schema;
  }
@@ -368,7 +356,7 @@ export class StandardQueryBase<
  extends QueryBase<NativeQueryType>
  implements ExecutableQuery
 {
-  constructor(inner?: NativeQueryType | Promise<NativeQueryType>) {
+  constructor(inner: NativeQueryType | Promise<NativeQueryType>) {
    super(inner);
  }

@@ -522,13 +510,6 @@ export class VectorQuery extends StandardQueryBase<NativeVectorQuery> {
    super(inner);
  }

-  /**
-   * @hidden
-   */
-  protected doVectorCall(fn: (inner: NativeVectorQuery) => void) {
-    super.doCall(fn);
-  }
-
  /**
   * Set the number of partitions to search (probe)
   *
@@ -556,7 +537,7 @@ export class VectorQuery extends StandardQueryBase<NativeVectorQuery> {
   * the minimum and maximum to the same value.
   */
  nprobes(nprobes: number): VectorQuery {
-    this.doVectorCall((inner) => inner.nprobes(nprobes));
+    super.doCall((inner) => inner.nprobes(nprobes));

    return this;
  }
@@ -570,7 +551,7 @@ export class VectorQuery extends StandardQueryBase<NativeVectorQuery> {
   * but will also increase latency.
   */
  minimumNprobes(minimumNprobes: number): VectorQuery {
-    this.doVectorCall((inner) => inner.minimumNprobes(minimumNprobes));
+    super.doCall((inner) => inner.minimumNprobes(minimumNprobes));
    return this;
  }

@@ -584,7 +565,7 @@ export class VectorQuery extends StandardQueryBase<NativeVectorQuery> {
   * potential false negatives.
   */
  maximumNprobes(maximumNprobes: number): VectorQuery {
-    this.doVectorCall((inner) => inner.maximumNprobes(maximumNprobes));
+    super.doCall((inner) => inner.maximumNprobes(maximumNprobes));
    return this;
  }

@@ -597,7 +578,7 @@ export class VectorQuery extends StandardQueryBase<NativeVectorQuery> {
   * `undefined` means no lower or upper bound.
   */
  distanceRange(lowerBound?: number, upperBound?: number): VectorQuery {
-    this.doVectorCall((inner) => inner.distanceRange(lowerBound, upperBound));
+    super.doCall((inner) => inner.distanceRange(lowerBound, upperBound));
    return this;
  }

@@ -611,7 +592,7 @@ export class VectorQuery extends StandardQueryBase<NativeVectorQuery> {
   * also increase the latency of your query. The default value is 1.5*limit.
   */
  ef(ef: number): VectorQuery {
-    this.doVectorCall((inner) => inner.ef(ef));
+    super.doCall((inner) => inner.ef(ef));
    return this;
  }

@@ -625,7 +606,7 @@ export class VectorQuery extends StandardQueryBase<NativeVectorQuery> {
   * whose data type is a fixed-size-list of floats.
   */
  column(column: string): VectorQuery {
-    this.doVectorCall((inner) => inner.column(column));
+    super.doCall((inner) => inner.column(column));
    return this;
  }

@@ -646,7 +627,7 @@ export class VectorQuery extends StandardQueryBase<NativeVectorQuery> {
  distanceType(
    distanceType: Required<IvfPqOptions>["distanceType"],
  ): VectorQuery {
-    this.doVectorCall((inner) => inner.distanceType(distanceType));
+    super.doCall((inner) => inner.distanceType(distanceType));
    return this;
  }

@@ -680,7 +661,7 @@ export class VectorQuery extends StandardQueryBase<NativeVectorQuery> {
   * distance between the query vector and the actual uncompressed vector.
   */
  refineFactor(refineFactor: number): VectorQuery {
-    this.doVectorCall((inner) => inner.refineFactor(refineFactor));
+    super.doCall((inner) => inner.refineFactor(refineFactor));
    return this;
  }

@@ -705,7 +686,7 @@ export class VectorQuery extends StandardQueryBase<NativeVectorQuery> {
   * factor can often help restore some of the results lost by post filtering.
   */
  postfilter(): VectorQuery {
-    this.doVectorCall((inner) => inner.postfilter());
+    super.doCall((inner) => inner.postfilter());
    return this;
  }

@@ -719,7 +700,7 @@ export class VectorQuery extends StandardQueryBase<NativeVectorQuery> {
   * calculate your recall to select an appropriate value for nprobes.
   */
  bypassVectorIndex(): VectorQuery {
-    this.doVectorCall((inner) => inner.bypassVectorIndex());
+    super.doCall((inner) => inner.bypassVectorIndex());
    return this;
  }

@@ -727,39 +708,43 @@ export class VectorQuery extends StandardQueryBase<NativeVectorQuery> {
   * Add a query vector to the search
   *
   * This method can be called multiple times to add multiple query vectors
-   * to the search. A column called `query_index` will be added to indicate the index
-   * of the query vector that produced the result. Flat searches share one table scan
-   * across the query vectors, avoiding the scan and memory amplification of running
-   * multiple queries concurrently. Indexed searches may still perform per-vector
-   * index work.
+   * to the search. If multiple query vectors are added, then they will be searched
+   * in parallel, and the results will be concatenated. A column called `query_index`
+   * will be added to indicate the index of the query vector that produced the result.
+   *
+   * Performance wise, this is equivalent to running multiple queries concurrently.
   */
  addQueryVector(vector: IntoVector): VectorQuery {
    if (vector instanceof Promise) {
-      // Observe the promise as soon as it is accepted. The existing native
-      // query may still be pending, and delaying observation until it resolves
-      // can otherwise surface a fast rejection as unhandled.
-      const settledVector = vector.then(
-        (value) => ({ status: "fulfilled" as const, value }),
-        (reason) => ({ status: "rejected" as const, reason }),
-      );
      const res = (async () => {
-        const inner = await this.getInner();
-        const outcome = await settledVector;
-        if (outcome.status === "rejected") {
-          throw outcome.reason;
+        try {
+          const v = await vector;
+          // biome-ignore lint/suspicious/noExplicitAny: we need to get the `inner`, but js has no package scoping
+          const value: any = this.addQueryVector(v);
+          const inner = value.inner as
+            | NativeVectorQuery
+            | Promise<NativeVectorQuery>;
+          return inner;
+        } catch (e) {
+          return Promise.reject(e);
        }
-        addQueryVectorToNative(inner, outcome.value);
-        return inner;
      })();
      return new VectorQuery(res);
    } else {
-      this.doVectorCall((inner) => addQueryVectorToNative(inner, vector));
+      super.doCall((inner) => {
+        const raw = Array.isArray(vector) ? null : extractVectorBuffer(vector);
+        if (raw) {
+          inner.addQueryVectorRaw(raw.data, raw.dtype);
+        } else {
+          inner.addQueryVector(Float32Array.from(vector as number[]));
+        }
+      });
      return this;
    }
  }

  rerank(reranker: Reranker): VectorQuery {
-    this.doVectorCall((inner) =>
+    super.doCall((inner) =>
      inner.rerank(async (args) => {
        const vecResults = await fromBufferToRecordBatch(args.vecResults);
        const ftsResults = await fromBufferToRecordBatch(args.ftsResults);
@@ -778,71 +763,6 @@ export class VectorQuery extends StandardQueryBase<NativeVectorQuery> {
  }
 }

-/**
- * Create a string query whose vector/FTS routing is resolved against the active
- * table schema when the query executes.
- *
- * @hidden
- */
-export function createAutoQuery(
-  table: NativeTable,
-  query: string,
-  columns: string[] | null,
-  getVector: (metadata: string) => Promise<Awaited<IntoVector>>,
-): AutoQuery {
-  type RouteSnapshot = {
-    table: NativeTable;
-    embeddingMetadata: string | undefined;
-  };
-  type CachedPreparation = {
-    metadata: string;
-    vector: Promise<Awaited<IntoVector>>;
-  };
-
-  let cachedPreparation: CachedPreparation | undefined;
-
-  const snapshotRoute = async (): Promise<RouteSnapshot> => {
-    const snapshot = await table.querySnapshot();
-    const schema = tableFromIPC(await snapshot.schema()).schema;
-    return {
-      table: snapshot,
-      embeddingMetadata: schema.metadata.get("embedding_functions"),
-    };
-  };
-
-  const createInner = async (): Promise<NativeQuery | NativeVectorQuery> => {
-    const route = await snapshotRoute();
-    if (route.embeddingMetadata === undefined) {
-      const inner = route.table.query();
-      inner.fullTextSearch({ query, columns });
-      return inner;
-    }
-
-    const metadata = route.embeddingMetadata;
-    if (cachedPreparation?.metadata !== metadata) {
-      cachedPreparation = {
-        metadata,
-        vector: Promise.resolve().then(() => getVector(metadata)),
-      };
-    }
-
-    const preparation = cachedPreparation;
-    let vector: Awaited<IntoVector>;
-    try {
-      vector = await preparation.vector;
-    } catch (error) {
-      if (cachedPreparation === preparation) {
-        cachedPreparation = undefined;
-      }
-      throw error;
-    }
-
-    return nearestToNative(route.table.query(), vector);
-  };
-
-  return new AutoQuery(createInner);
-}
-
 /**
 * A query that returns a subset of the rows in the table.
 *
@@ -868,51 +788,6 @@ export class TakeQuery extends QueryBase<NativeTakeQuery> {
  }
 }

-/**
- * A builder for automatic string searches.
- *
- * Automatic search determines whether to use full-text or vector search from
- * the table revision selected for each execution. This builder exposes the
- * common operations supported by both query families.
- *
- * @hideconstructor
- */
-export class AutoQuery extends StandardQueryBase<
-  NativeQuery | NativeVectorQuery
-> {
-  private readonly calls: Array<
-    (inner: NativeQuery | NativeVectorQuery) => void
-  > = [];
-
-  /** @hidden */
-  constructor(
-    private readonly createInner: () => Promise<
-      NativeQuery | NativeVectorQuery
-    >,
-  ) {
-    super();
-  }
-
-  /** @hidden */
-  protected override doCall(
-    fn: (inner: NativeQuery | NativeVectorQuery) => void,
-  ) {
-    this.calls.push(fn);
-  }
-
-  /** @hidden */
-  protected override async getInner(): Promise<
-    NativeQuery | NativeVectorQuery
-  > {
-    const calls = [...this.calls];
-    const inner = await this.createInner();
-    for (const call of calls) {
-      call(inner);
-    }
-    return inner;
-  }
-}
-
 /** A builder for LanceDB queries.
 *
 * @see {@link Table#query}, {@link Table#search}
@@ -965,19 +840,45 @@ export class Query extends StandardQueryBase<NativeQuery> {
   * a default `limit` of 10 will be used.  @see {@link Query#limit}
   */
  nearestTo(vector: IntoVector): VectorQuery {
-    const inner = this.inner;
-    if (inner instanceof Promise) {
-      const nativeQuery = inner.then(async (resolvedInner) =>
-        nearestToNative(resolvedInner, await vector),
-      );
+    const callNearestTo = (
+      inner: NativeQuery,
+      resolved: Float32Array | Float64Array | Uint8Array | number[],
+    ): NativeVectorQuery => {
+      const raw = Array.isArray(resolved)
+        ? null
+        : extractVectorBuffer(resolved);
+      if (raw) {
+        return inner.nearestToRaw(raw.data, raw.dtype);
+      }
+      return inner.nearestTo(Float32Array.from(resolved as number[]));
+    };
+
+    if (this.inner instanceof Promise) {
+      const nativeQuery = this.inner.then(async (inner) => {
+        const resolved = vector instanceof Promise ? await vector : vector;
+        return callNearestTo(inner, resolved);
+      });
      return new VectorQuery(nativeQuery);
    }
    if (vector instanceof Promise) {
-      return new VectorQuery(
-        vector.then((resolvedVector) => nearestToNative(inner, resolvedVector)),
-      );
+      const res = (async () => {
+        try {
+          const v = await vector;
+          // biome-ignore lint/suspicious/noExplicitAny: we need to get the `inner`, but js has no package scoping
+          const value: any = this.nearestTo(v);
+          const inner = value.inner as
+            | NativeVectorQuery
+            | Promise<NativeVectorQuery>;
+          return inner;
+        } catch (e) {
+          return Promise.reject(e);
+        }
+      })();
+      return new VectorQuery(res);
+    } else {
+      const vectorQuery = callNearestTo(this.inner, vector);
+      return new VectorQuery(vectorQuery);
    }
-    return new VectorQuery(nearestToNative(inner, vector));
  }

  nearestToText(query: string | FullTextQuery, columns?: string[]): Query {
@@ -94,24 +94,17 @@ export function sanitizeMetadata(
  if (metadataLike === undefined || metadataLike === null) {
    return undefined;
  }
-
-  let entries: IterableIterator<[unknown, unknown]>;
-  try {
-    entries = Map.prototype.entries.call(metadataLike);
-  } catch {
+  if (!(metadataLike instanceof Map)) {
    throw Error("Expected metadata, if present, to be a Map<string, string>");
  }
-
-  const metadata = new Map<string, string>();
-  for (const [key, value] of entries) {
-    if (typeof key !== "string" || typeof value !== "string") {
+  for (const item of metadataLike) {
+    if (typeof item[0] !== "string" || typeof item[1] !== "string") {
      throw Error(
        "Expected metadata, if present, to be a Map<string, string> but it had non-string keys or values",
      );
    }
-    metadata.set(key, value);
  }
-  return metadata;
+  return metadataLike as Map<string, string>;
 }

 export function sanitizeInt(typeLike: object) {
@@ -1,567 +0,0 @@
-// SPDX-License-Identifier: Apache-2.0
-// SPDX-FileCopyrightText: Copyright The LanceDB Authors
-
-import {
-  Binary,
-  Bool,
-  DataType,
-  Dictionary,
-  Field,
-  FixedSizeList,
-  Float32,
-  Float64,
-  Int32,
-  Int64,
-  List,
-  Schema,
-  Struct,
-  Utf8,
-  util as arrowUtil,
-} from "apache-arrow";
-import { typedArrayToArrowType } from "./arrow_type";
-import { sanitizeType } from "./sanitize";
-
-type InferenceOptions = {
-  dictionaryEncodeStrings: boolean;
-  vectorColumns: Record<string, { type: unknown }>;
-};
-
-/**
- * Infer the Arrow schema represented by a set of records.
- *
- * This is the intentionally small interface to schema inference. The stateful
- * details of combining partial type evidence are encapsulated below so callers
- * only need to provide records, an optional schema, and inference options.
- */
-export function inferSchema(
-  data: Array<Record<string, unknown>>,
-  schema: Schema | undefined,
-  options: InferenceOptions,
-): Schema {
-  return new SchemaInferrer(schema, options).infer(data);
-}
-
-class SchemaInferrer {
-  private readonly fields = new FieldTree();
-
-  constructor(
-    private readonly providedSchema: Schema | undefined,
-    private readonly options: InferenceOptions,
-  ) {}
-
-  infer(data: Array<Record<string, unknown>>): Schema {
-    for (const [row, record] of data.entries()) {
-      for (const [path, value] of recordPathsAndValues(record)) {
-        this.observe(path, value, row);
-      }
-    }
-
-    return this.providedSchema === undefined
-      ? new Schema(fieldsFromTree(this.fields))
-      : new Schema(matchingFields(this.providedSchema.fields, this.fields));
-  }
-
-  private observe(path: string[], value: unknown, row: number): void {
-    const current = this.fields.get(path);
-    if (current === undefined) {
-      this.addField(path, value, row);
-    } else if (this.providedSchema === undefined) {
-      this.updateInferredField(path, value, row, current);
-    }
-  }
-
-  private addField(path: string[], value: unknown, row: number): void {
-    if (this.providedSchema !== undefined) {
-      this.addSchemaField(this.providedSchema, path, row);
-      return;
-    }
-
-    const evidence =
-      this.inferType(value, path) ?? DeferredTypeEvidence.from(value, row);
-    if (evidence === undefined) {
-      throw typeInferenceError(path, row);
-    }
-
-    const conflict = this.fields.set(
-      path,
-      evidence,
-      (existing) =>
-        existing instanceof DeferredTypeEvidence && existing.isOnlyNulls(),
-    );
-    if (conflict !== undefined) {
-      throw branchConflictError(conflict, row, "Struct");
-    }
-  }
-
-  private addSchemaField(schema: Schema, path: string[], row: number): void {
-    const field = fieldAtPath(schema, path);
-    if (field === undefined) {
-      throw new Error(
-        `Found field not in schema: ${path.join(".")} at row ${row}`,
-      );
-    }
-
-    const conflict = this.fields.set(path, field.type);
-    if (conflict !== undefined) {
-      throw branchConflictError(conflict, row, "Struct");
-    }
-  }
-
-  private updateInferredField(
-    path: string[],
-    value: unknown,
-    row: number,
-    current: FieldNode,
-  ): void {
-    const newType = this.inferType(value, path);
-    const deferred = DeferredTypeEvidence.from(value, row);
-
-    if (current instanceof FieldTree) {
-      if (deferred?.isOnlyNulls()) {
-        return;
-      }
-      throw schemaInferenceError(
-        path,
-        row,
-        "Struct",
-        describeEvidence(newType ?? deferred),
-      );
-    }
-
-    if (current instanceof DeferredTypeEvidence) {
-      this.resolveDeferredField(path, row, current, newType, deferred);
-      return;
-    }
-
-    if (newType !== undefined) {
-      if (!inferredTypesEqual(current, newType)) {
-        throw schemaInferenceError(
-          path,
-          row,
-          describeEvidence(current),
-          describeEvidence(newType),
-        );
-      }
-      return;
-    }
-
-    if (deferred === undefined || !deferred.matches(current)) {
-      throw schemaInferenceError(
-        path,
-        row,
-        describeEvidence(current),
-        describeEvidence(deferred),
-      );
-    }
-  }
-
-  private resolveDeferredField(
-    path: string[],
-    row: number,
-    current: DeferredTypeEvidence,
-    newType: DataType | undefined,
-    deferred: DeferredTypeEvidence | undefined,
-  ): void {
-    if (newType !== undefined) {
-      if (!current.matches(newType)) {
-        throw schemaInferenceError(
-          path,
-          row,
-          current.describe(),
-          describeEvidence(newType),
-        );
-      }
-      this.fields.set(path, newType);
-      return;
-    }
-
-    if (deferred !== undefined) {
-      this.fields.set(path, current.merge(deferred));
-      return;
-    }
-
-    throw schemaInferenceError(
-      path,
-      row,
-      current.describe(),
-      describeEvidence(newType),
-    );
-  }
-
-  private inferType(value: unknown, path: string[]): DataType | undefined {
-    if (typeof value === "bigint") {
-      return new Int64();
-    }
-    if (typeof value === "number") {
-      return new Float64();
-    }
-    if (typeof value === "string") {
-      return this.options.dictionaryEncodeStrings
-        ? new Dictionary(new Utf8(), new Int32())
-        : new Utf8();
-    }
-    if (typeof value === "boolean") {
-      return new Bool();
-    }
-    if (value instanceof Buffer) {
-      return new Binary();
-    }
-    if (ArrayBuffer.isView(value) && !(value instanceof DataView)) {
-      const typedArray = typedArrayToArrowType(value);
-      return typedArray === undefined
-        ? undefined
-        : new FixedSizeList(
-            typedArray.length,
-            new Field("item", typedArray.elementType, true),
-          );
-    }
-    if (!Array.isArray(value) || value.length === 0) {
-      return undefined;
-    }
-
-    const configuredVector =
-      path.length === 1 ? this.options.vectorColumns[path[0]] : undefined;
-    if (configuredVector !== undefined) {
-      return new FixedSizeList(
-        value.length,
-        new Field("item", sanitizeType(configuredVector.type), true),
-      );
-    }
-
-    const itemType = this.inferArrayItemType(value, path);
-    if (itemType === undefined) {
-      return undefined;
-    }
-
-    return nameSuggestsVectorColumn(path[path.length - 1])
-      ? new FixedSizeList(value.length, new Field("item", new Float32(), true))
-      : new List(new Field("item", itemType, true));
-  }
-
-  private inferArrayItemType(
-    values: unknown[],
-    path: string[],
-  ): DataType | undefined {
-    let itemType: DataType | undefined;
-    const deferredItems: unknown[] = [];
-
-    for (const value of values) {
-      const candidate = this.inferType(value, path);
-      if (candidate === undefined) {
-        if (!isDeferredValue(value)) {
-          return undefined;
-        }
-        deferredItems.push(value);
-      } else if (itemType === undefined) {
-        itemType = candidate;
-      } else if (!inferredTypesEqual(itemType, candidate)) {
-        return undefined;
-      }
-    }
-
-    if (itemType === undefined) {
-      return undefined;
-    }
-    return deferredItems.every((value) =>
-      deferredValueMatchesType(value, itemType),
-    )
-      ? itemType
-      : undefined;
-  }
-}
-
-/** Nulls and empty/all-null lists that do not determine a type by themselves. */
-class DeferredTypeEvidence {
-  private constructor(
-    private readonly values: Array<{ value: unknown; row: number }>,
-  ) {}
-
-  static from(value: unknown, row: number): DeferredTypeEvidence | undefined {
-    return isDeferredValue(value)
-      ? new DeferredTypeEvidence([{ value, row }])
-      : undefined;
-  }
-
-  isOnlyNulls(): boolean {
-    return this.values.every(({ value }) => value == null);
-  }
-
-  matches(type: DataType): boolean {
-    return this.values.every(({ value }) =>
-      deferredValueMatchesType(value, type),
-    );
-  }
-
-  merge(other: DeferredTypeEvidence): DeferredTypeEvidence {
-    return new DeferredTypeEvidence([...this.values, ...other.values]);
-  }
-
-  describe(): string {
-    const list = this.values.find(({ value }) => Array.isArray(value));
-    return list === undefined
-      ? "null"
-      : `List[${(list.value as unknown[]).length}]`;
-  }
-
-  firstRow(): number {
-    return this.values[0].row;
-  }
-}
-
-type FieldNode = DataType | DeferredTypeEvidence | FieldTree;
-type LeafNode = Exclude<FieldNode, FieldTree>;
-type FieldConflict = { path: string[]; value: FieldNode };
-
-/** Nested field state, kept separate from Arrow's eventual Struct types. */
-class FieldTree {
-  private readonly children = new Map<string, FieldNode>();
-
-  get(path: string[]): FieldNode | undefined {
-    let current: FieldNode = this;
-    for (const part of path) {
-      if (!(current instanceof FieldTree)) {
-        return undefined;
-      }
-      const child = current.children.get(part);
-      if (child === undefined) {
-        return undefined;
-      }
-      current = child;
-    }
-    return current;
-  }
-
-  set(
-    path: string[],
-    value: LeafNode,
-    canReplaceLeaf: (value: LeafNode) => boolean = () => false,
-  ): FieldConflict | undefined {
-    let branch: FieldTree = this;
-    for (const [index, part] of path.slice(0, -1).entries()) {
-      const child = branch.children.get(part);
-      if (child === undefined || (isLeaf(child) && canReplaceLeaf(child))) {
-        const nextBranch = new FieldTree();
-        branch.children.set(part, nextBranch);
-        branch = nextBranch;
-      } else if (child instanceof FieldTree) {
-        branch = child;
-      } else {
-        return { path: path.slice(0, index + 1), value: child };
-      }
-    }
-
-    const name = path[path.length - 1];
-    const current = branch.children.get(name);
-    if (current instanceof FieldTree) {
-      return { path, value: current };
-    }
-    branch.children.set(name, value);
-    return undefined;
-  }
-
-  entries(): IterableIterator<[string, FieldNode]> {
-    return this.children.entries();
-  }
-
-  has(name: string): boolean {
-    return this.children.has(name);
-  }
-}
-
-function isLeaf(value: FieldNode): value is LeafNode {
-  return !(value instanceof FieldTree);
-}
-
-function fieldsFromTree(tree: FieldTree, path: string[] = []): Field[] {
-  const fields: Field[] = [];
-  for (const [name, value] of tree.entries()) {
-    if (value instanceof FieldTree) {
-      fields.push(
-        new Field(
-          name,
-          new Struct(fieldsFromTree(value, [...path, name])),
-          true,
-        ),
-      );
-    } else if (value instanceof DeferredTypeEvidence) {
-      throw typeInferenceError([...path, name], value.firstRow());
-    } else {
-      fields.push(new Field(name, value, true));
-    }
-  }
-  return fields;
-}
-
-function matchingFields(fields: Field[], tree: FieldTree): Field[] {
-  const matches: Field[] = [];
-  for (const field of fields) {
-    if (!tree.has(field.name)) {
-      continue;
-    }
-    const value = tree.get([field.name]);
-    if (value instanceof FieldTree) {
-      const struct = field.type as Struct;
-      matches.push(
-        new Field(
-          field.name,
-          new Struct(matchingFields(struct.children, value)),
-          field.nullable,
-          field.metadata,
-        ),
-      );
-    } else {
-      matches.push(field);
-    }
-  }
-  return matches;
-}
-
-function* recordPathsAndValues(
-  record: Record<string, unknown>,
-  path: string[] = [],
-): Generator<[string[], unknown]> {
-  for (const [name, value] of Object.entries(record)) {
-    if (isRecord(value)) {
-      yield* recordPathsAndValues(value, [...path, name]);
-    } else if (value !== undefined) {
-      yield [[...path, name], value];
-    }
-  }
-}
-
-function isRecord(value: unknown): value is Record<string, unknown> {
-  return (
-    typeof value === "object" &&
-    value !== null &&
-    !Array.isArray(value) &&
-    !(value instanceof RegExp) &&
-    !(value instanceof Date) &&
-    !(value instanceof Set) &&
-    !(value instanceof Map) &&
-    !(value instanceof Buffer) &&
-    !ArrayBuffer.isView(value)
-  );
-}
-
-function fieldAtPath(schema: Schema, path: string[]): Field | undefined {
-  let fields = schema.fields;
-  let field: Field | undefined;
-  for (const [index, name] of path.entries()) {
-    field = fields.find((candidate) => candidate.name === name);
-    if (field === undefined || index === path.length - 1) {
-      return field;
-    }
-    if (!DataType.isStruct(field.type)) {
-      return undefined;
-    }
-    fields = field.type.children;
-  }
-  return field;
-}
-
-function isDeferredValue(value: unknown): boolean {
-  return (
-    value == null || (Array.isArray(value) && value.every(isDeferredValue))
-  );
-}
-
-function deferredValueMatchesType(value: unknown, type: DataType): boolean {
-  if (value == null) {
-    return true;
-  }
-  if (!Array.isArray(value)) {
-    return false;
-  }
-  if (DataType.isList(type)) {
-    return value.every((item) =>
-      deferredValueMatchesType(item, type.valueType),
-    );
-  }
-  if (DataType.isFixedSizeList(type)) {
-    return (
-      value.length === type.listSize &&
-      value.every((item) => deferredValueMatchesType(item, type.valueType))
-    );
-  }
-  return false;
-}
-
-function inferredTypesEqual(current: DataType, candidate: DataType): boolean {
-  if (DataType.isDictionary(current)) {
-    return (
-      DataType.isDictionary(candidate) &&
-      current.isOrdered === candidate.isOrdered &&
-      inferredTypesEqual(current.indices, candidate.indices) &&
-      inferredTypesEqual(current.dictionary, candidate.dictionary)
-    );
-  }
-  if (DataType.isList(current)) {
-    return (
-      DataType.isList(candidate) &&
-      current.valueField.name === candidate.valueField.name &&
-      current.valueField.nullable === candidate.valueField.nullable &&
-      inferredTypesEqual(current.valueType, candidate.valueType)
-    );
-  }
-  if (DataType.isFixedSizeList(current)) {
-    return (
-      DataType.isFixedSizeList(candidate) &&
-      current.listSize === candidate.listSize &&
-      current.valueField.name === candidate.valueField.name &&
-      current.valueField.nullable === candidate.valueField.nullable &&
-      inferredTypesEqual(current.valueType, candidate.valueType)
-    );
-  }
-  return arrowUtil.compareTypes(current, candidate);
-}
-
-function describeEvidence(
-  evidence: DataType | DeferredTypeEvidence | undefined,
-): string {
-  if (evidence === undefined) {
-    return "an unsupported value";
-  }
-  return evidence instanceof DeferredTypeEvidence
-    ? evidence.describe()
-    : evidence.toString();
-}
-
-function branchConflictError(
-  conflict: FieldConflict,
-  row: number,
-  candidate: string,
-): Error {
-  return schemaInferenceError(
-    conflict.path,
-    row,
-    conflict.value instanceof FieldTree
-      ? "Struct"
-      : describeEvidence(conflict.value),
-    candidate,
-  );
-}
-
-function schemaInferenceError(
-  path: string[],
-  row: number,
-  currentType: string,
-  newType: string,
-): Error {
-  return new Error(
-    `Failed to infer schema for data. Previously inferred type ${currentType} ` +
-      `but found ${newType} for field ${path.join(".")} at row ${row}. ` +
-      "Consider providing an explicit schema.",
-  );
-}
-
-function typeInferenceError(path: string[], row: number): Error {
-  return new Error(
-    `Failed to infer data type for field ${path.join(".")} at row ${row}. ` +
-      "Consider providing an explicit schema.",
-  );
-}
-
-function nameSuggestsVectorColumn(name: string): boolean {
-  const normalized = name.toLowerCase();
-  return normalized.includes("vector") || normalized.includes("embedding");
-}
@@ -35,7 +35,6 @@ import {
  Branches as NativeBranches,
  OptimizeStats,
  RefreshColumnResult,
-  RefreshMaterializedViewResult,
  TableStatistics,
  Tags,
  UpdateFieldMetadataResult,
@@ -43,12 +42,10 @@ import {
  Table as _NativeTable,
 } from "./native";
 import {
-  AutoQuery,
  FullTextQuery,
  Query,
  TakeQuery,
  VectorQuery,
-  createAutoQuery,
  instanceOfFullTextQuery,
 } from "./query";
 import { sanitizeType } from "./sanitize";
@@ -525,7 +522,7 @@ export abstract class Table {
    query: string | IntoVector | MultiVector | FullTextQuery,
    queryType?: string,
    ftsColumns?: string | string[],
-  ): VectorQuery | Query | AutoQuery;
+  ): VectorQuery | Query;
  /**
   * Search the table with a given query vector.
   *
@@ -605,18 +602,6 @@ export abstract class Table {
   */
  abstract refreshColumnAsync(column: string): Promise<Job>;

-  /**
-   * Recompute this table's contents from its materialized-view definition.
-   *
-   * Plumbing for {@link MaterializedView.refresh}, which is the way to call
-   * it: rejects tables that carry no view definition. Local tables only.
-   * @ignore
-   */
-  abstract refreshMaterializedView(
-    full?: boolean,
-    sourceVersion?: number,
-  ): Promise<RefreshMaterializedViewResult>;
-
  /**
   * Alter the name or nullability of columns.
   * @param {ColumnAlteration[]} columnAlterations One or more alterations to
@@ -630,18 +615,6 @@ export abstract class Table {

  /**
   * Update per-field (column) metadata.
-   *
-   * The following keys are treated specially, by convention, and should be
-   * used when appropriate:
-   *
-   * - `lancedb:description`: for a human-readable description of a field.
-   * - `lancedb:tag:<name>`: for a user-defined key-value tag, where the suffix
-   *   names the tag category; e.g. `lancedb:tag:model: "clip"`.
-   * - `lancedb:logical-column`: for a column grouping; e.g. `feature_v1` and
-   *   `feature_v2` might be in the same logical column.
-   * - `lancedb:status`: for status options (`production`, `candidate`,
-   *   `deprecated`, `archived`) to designate the current life cycle state of
-   *   this column.
   * @param {FieldMetadataUpdate[]} updates One or more per-field updates. Each
   * update's metadata is merged into the field's existing metadata by default;
   * a value of `null` deletes that key, and `replace: true` swaps the whole map.
@@ -989,11 +962,10 @@ export class LocalTable extends Table {
    return this.inner.display();
  }

-  private async getEmbeddingFunctions(
-    inner: _NativeTable = this.inner,
-  ): Promise<Map<string, EmbeddingFunctionConfig>> {
-    const schemaBuf = await inner.schema();
-    const schema = tableFromIPC(schemaBuf).schema;
+  private async getEmbeddingFunctions(): Promise<
+    Map<string, EmbeddingFunctionConfig>
+  > {
+    const schema = await this.schema();
    const registry = getRegistry();
    return registry.parseFunctions(schema.metadata);
  }
@@ -1175,7 +1147,7 @@ export class LocalTable extends Table {
    query: string | IntoVector | MultiVector | FullTextQuery,
    queryType: string = "auto",
    ftsColumns?: string | string[],
-  ): VectorQuery | Query | AutoQuery {
+  ): VectorQuery | Query {
    if (typeof query !== "string" && !instanceOfFullTextQuery(query)) {
      if (queryType === "fts") {
        throw new Error("Cannot perform full text search on a vector query");
@@ -1190,28 +1162,14 @@ export class LocalTable extends Table {
      });
    }

-    if (queryType === "auto") {
-      if (instanceOfFullTextQuery(query)) {
-        return this.query().fullTextSearch(query, {
-          columns: ftsColumns,
-        });
-      }
-
-      const columns =
-        typeof ftsColumns === "string" ? [ftsColumns] : (ftsColumns ?? null);
-      return createAutoQuery(this.inner, query, columns, async (metadata) => {
-        const functions = await getRegistry().parseFunctions(
-          new Map([["embedding_functions", metadata]]),
-        );
-        // TODO: Support multiple embedding functions
-        const embeddingFunc: EmbeddingFunctionConfig | undefined = functions
-          .values()
-          .next().value;
-        // The route only calls this callback when embedding metadata exists.
-        // parseFunctions either yields a provider or reports malformed metadata.
-        if (!embeddingFunc)
-          throw new Error("Invalid embedding function metadata");
-        return await embeddingFunc.function.computeQueryEmbeddings(query);
+    // The query type is auto or vector
+    // fall back to full text search if no embedding functions are defined and the query is a string
+    if (
+      queryType === "auto" &&
+      (getRegistry().length() === 0 || instanceOfFullTextQuery(query))
+    ) {
+      return this.query().fullTextSearch(query, {
+        columns: ftsColumns,
      });
    }

@@ -1306,13 +1264,6 @@ export class LocalTable extends Table {
    return await this.inner.refreshColumnAsync(column);
  }

-  async refreshMaterializedView(
-    full?: boolean,
-    sourceVersion?: number,
-  ): Promise<RefreshMaterializedViewResult> {
-    return await this.inner.refreshMaterializedView(full, sourceVersion);
-  }
-
  async alterColumns(
    columnAlterations: ColumnAlteration[],
  ): Promise<AlterColumnsResult> {
@@ -1567,8 +1518,7 @@ export interface FieldMetadataUpdate {
  path: string;
  /**
   * Metadata key/value pairs. Merged into the field's existing metadata by
-   * default; a value of `null` deletes that key. See
-   * {@link Table.updateFieldMetadata} for the conventional `lancedb:*` keys.
+   * default; a value of `null` deletes that key.
   */
  metadata: Record<string, string | null>;
  /** If true, replace the field's entire metadata map instead of merging. */
@@ -1607,8 +1557,8 @@ export interface BranchRowCountSummary {
  deltaAvailable: boolean;
 }

-/** A reason why a cherry-pick cannot currently land. */
-export interface CherryPickError {
+/** A reason why a branch cannot currently be merged. */
+export interface MergeBlocker {
  code: string;
  message: string;
 }
@@ -1628,19 +1578,20 @@ export interface BranchDiff {
  changedColumns: BranchColumnChange[];
  addedIndexes: BranchIndexSummary[];
  removedIndexes: BranchIndexSummary[];
-  errors: CherryPickError[];
+  mergeable: boolean;
+  mergeBlockers: MergeBlocker[];
 }

-/** Changes that would be, or were, promoted by a cherry-pick. */
-export interface CherryPickPreview {
+/** Changes that would be, or were, promoted by a branch merge. */
+export interface MergePreview {
  promotedColumns: string[];
 }

-/** Result of previewing or attempting a cherry-pick. */
-export interface CherryPickResult {
-  status: "ready" | "failed" | "notImplemented" | "cherryPicked" | "unknown";
+/** Result of previewing or attempting a branch merge. */
+export interface MergeBranchResult {
+  status: "ready" | "rejected" | "notImplemented" | "merged" | "unknown";
  diff: BranchDiff;
-  preview: CherryPickPreview;
+  preview: MergePreview;
  mainVersionAfter?: number;
 }

@@ -1703,21 +1654,21 @@ export class Branches {
  }

  /**
-   * Cherry-pick a branch onto main.
+   * Merge a branch into main.
   *
-   * Set `dryRun` to `true` to preview. A failed cherry-pick resolves
-   * with `status: "failed"` instead of throwing.
+   * Set `dryRun` to `true` to preview the merge. A rejected merge resolves
+   * with `status: "rejected"` instead of throwing.
   *
-   * @param fromBranch Branch to cherry-pick from.
-   * @param dryRun When true, only preview. Defaults to false.
+   * @param fromBranch Branch to merge from.
+   * @param dryRun When true, only preview the merge. Defaults to false.
   */
-  async cherryPick(
+  async merge(
    fromBranch: string,
    dryRun: boolean = false,
-  ): Promise<CherryPickResult> {
-    return (await this.#inner.cherryPick(
+  ): Promise<MergeBranchResult> {
+    return (await this.#inner.merge(
      fromBranch,
      dryRun,
-    )) as unknown as CherryPickResult;
+    )) as unknown as MergeBranchResult;
  }
 }
@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-darwin-arm64",
-	"version": "0.38.0-beta.11",
+	"version": "0.38.0-beta.3",
 	"os": ["darwin"],
 	"cpu": ["arm64"],
 	"main": "lancedb.darwin-arm64.node",
@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-arm64-gnu",
-	"version": "0.38.0-beta.11",
+	"version": "0.38.0-beta.3",
 	"os": ["linux"],
 	"cpu": ["arm64"],
 	"main": "lancedb.linux-arm64-gnu.node",
@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-arm64-musl",
-	"version": "0.38.0-beta.11",
+	"version": "0.38.0-beta.3",
 	"os": ["linux"],
 	"cpu": ["arm64"],
 	"main": "lancedb.linux-arm64-musl.node",
@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-x64-gnu",
-	"version": "0.38.0-beta.11",
+	"version": "0.38.0-beta.3",
 	"os": ["linux"],
 	"cpu": ["x64"],
 	"main": "lancedb.linux-x64-gnu.node",
@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-linux-x64-musl",
-	"version": "0.38.0-beta.11",
+	"version": "0.38.0-beta.3",
 	"os": ["linux"],
 	"cpu": ["x64"],
 	"main": "lancedb.linux-x64-musl.node",
@@ -1,6 +1,6 @@
 {
  "name": "@lancedb/lancedb-win32-arm64-msvc",
-  "version": "0.38.0-beta.11",
+  "version": "0.38.0-beta.3",
  "os": [
    "win32"
  ],
@@ -1,6 +1,6 @@
 {
 	"name": "@lancedb/lancedb-win32-x64-msvc",
-	"version": "0.38.0-beta.11",
+	"version": "0.38.0-beta.3",
 	"os": ["win32"],
 	"cpu": ["x64"],
 	"main": "lancedb.win32-x64-msvc.node",
@@ -1,12 +1,12 @@
 {
  "name": "@lancedb/lancedb",
-  "version": "0.38.0-beta.11",
+  "version": "0.38.0-beta.2",
  "lockfileVersion": 3,
  "requires": true,
  "packages": {
    "": {
      "name": "@lancedb/lancedb",
-      "version": "0.38.0-beta.11",
+      "version": "0.38.0-beta.2",
      "cpu": [
        "x64",
        "arm64"
@@ -11,7 +11,7 @@
    "ann"
  ],
  "private": false,
-  "version": "0.38.0-beta.11",
+  "version": "0.38.0-beta.3",
  "main": "dist/index.js",
  "exports": {
    ".": "./dist/index.js",
@@ -17,7 +17,6 @@ use lancedb::connection::{ConnectBuilder, Connection as LanceDBConnection, conne

 use lance_namespace::models::{
    CreateNamespaceRequest, DescribeNamespaceRequest, DropNamespaceRequest, ListNamespacesRequest,
-    ListTablesRequest,
 };
 use lancedb::ipc::{ipc_file_to_batches, ipc_file_to_schema};

@@ -37,12 +36,6 @@ pub struct ListNamespacesResponse {
    pub page_token: Option<String>,
 }

-#[napi(object)]
-pub struct ListTablesResponse {
-    pub tables: Vec<String>,
-    pub page_token: Option<String>,
-}
-
 #[napi(object)]
 pub struct CreateNamespaceResponse {
    pub properties: Option<HashMap<String, String>>,
@@ -213,33 +206,6 @@ impl Connection {
        op.execute().await.default_error()
    }

-    /// List a page of tables in the database.
-    #[napi(catch_unwind)]
-    pub async fn list_tables(
-        &self,
-        namespace_path: Option<Vec<String>>,
-        page_token: Option<String>,
-        limit: Option<u32>,
-    ) -> napi::Result<ListTablesResponse> {
-        let request = ListTablesRequest {
-            // The root namespace is an empty path, not an absent one: a namespace-backed
-            // database rejects a request that names no namespace.
-            id: Some(namespace_path.unwrap_or_default()),
-            page_token,
-            limit: limit.map(|limit| i32::try_from(limit).unwrap_or(i32::MAX)),
-            ..Default::default()
-        };
-        let response = self
-            .get_inner()?
-            .list_tables(request)
-            .await
-            .default_error()?;
-        Ok(ListTablesResponse {
-            tables: response.tables,
-            page_token: response.page_token,
-        })
-    }
-
    /// Create table from a Apache Arrow IPC (file) buffer.
    ///
    /// Parameters:
@@ -300,58 +266,6 @@ impl Connection {
        Ok(Table::new(tbl))
    }

-    #[napi(catch_unwind)]
-    pub async fn create_materialized_view(
-        &self,
-        name: String,
-        source: String,
-        projections: Option<Vec<Vec<String>>>,
-        filter: Option<String>,
-        limit: Option<i64>,
-    ) -> napi::Result<Table> {
-        let mut builder = self.get_inner()?.create_materialized_view(name, source);
-        if let Some(projections) = projections {
-            let mut pairs = Vec::with_capacity(projections.len());
-            for pair in projections {
-                let [output, expression]: [String; 2] = pair.try_into().map_err(|_| {
-                    napi::Error::from_reason("each projection must be an [output, expression] pair")
-                })?;
-                pairs.push((output, expression));
-            }
-            builder = builder.select(pairs);
-        }
-        if let Some(filter) = filter {
-            builder = builder.only_if(filter);
-        }
-        if let Some(limit) = limit {
-            let limit = u64::try_from(limit)
-                .map_err(|_| napi::Error::from_reason("limit must be a non-negative integer"))?;
-            builder = builder.limit(limit);
-        }
-        let view = builder.execute().await.default_error()?;
-        Ok(Table::new(view.table().clone()))
-    }
-
-    #[napi(catch_unwind)]
-    pub async fn open_materialized_view(&self, name: String) -> napi::Result<Table> {
-        let view = self
-            .get_inner()?
-            .open_materialized_view(&name)
-            .await
-            .default_error()?;
-        Ok(Table::new(view.table().clone()))
-    }
-
-    #[napi(catch_unwind)]
-    pub async fn list_materialized_views(&self) -> napi::Result<Vec<String>> {
-        let views = self
-            .get_inner()?
-            .list_materialized_views()
-            .await
-            .default_error()?;
-        Ok(views.into_iter().map(|v| v.name).collect())
-    }
-
    #[napi(catch_unwind)]
    pub async fn open_table(
        &self,
@@ -14,12 +14,9 @@ pub struct Job {
 }

 impl Job {
-    pub(crate) fn new<T>(inner: lancedb::Job<T>) -> Self
-    where
-        T: Clone + Send + Sync + 'static,
-    {
+    pub(crate) fn new(inner: lancedb::Job) -> Self {
        Self {
-            inner: Arc::new(inner.map(|_| ())),
+            inner: Arc::new(inner),
        }
    }
 }
@@ -1,10 +1,6 @@
 // SPDX-License-Identifier: Apache-2.0
 // SPDX-FileCopyrightText: Copyright The LanceDB Authors

-// The materialized-view refresh future deepens the type graph past the
-// default trait-recursion depth; same raise as the core crate applies.
-#![recursion_limit = "256"]
-
 use std::collections::HashMap;

 use env_logger::Env;
@@ -278,13 +278,6 @@ impl Table {
        Ok(Query::new(self.inner_ref()?.query()))
    }

-    /// Return a read-only table handle pinned to the current query revision.
-    #[napi(catch_unwind)]
-    pub async fn query_snapshot(&self) -> napi::Result<Self> {
-        let snapshot = self.inner_ref()?.query_snapshot().await.default_error()?;
-        Ok(Self::new(snapshot))
-    }
-
    #[napi(catch_unwind)]
    pub fn take_offsets(&self, offsets: Vec<i64>) -> napi::Result<TakeQuery> {
        Ok(TakeQuery::new(
@@ -388,26 +381,6 @@ impl Table {
        Ok(crate::job::Job::new(job))
    }

-    #[napi(catch_unwind)]
-    pub async fn refresh_materialized_view(
-        &self,
-        full: Option<bool>,
-        source_version: Option<i64>,
-    ) -> napi::Result<RefreshMaterializedViewResult> {
-        let view = lancedb::MaterializedView::from_table(self.inner_ref()?.clone())
-            .await
-            .default_error()?;
-        let mut builder = view.refresh().full(full.unwrap_or(false));
-        if let Some(version) = source_version {
-            let version = u64::try_from(version).map_err(|_| {
-                napi::Error::from_reason("sourceVersion must be a non-negative integer")
-            })?;
-            builder = builder.source_version(version);
-        }
-        let result = builder.execute().await.default_error()?;
-        Ok(result.into())
-    }
-
    #[napi(catch_unwind)]
    pub async fn add_columns_with_schema(
        &self,
@@ -561,12 +534,6 @@ impl Table {
            .default_error()
    }

-    #[napi(catch_unwind)]
-    pub async fn checkout_current(&self) -> napi::Result<Self> {
-        let table = self.inner_ref()?.checkout_current().await.default_error()?;
-        Ok(Self::new(table))
-    }
-
    #[napi(catch_unwind)]
    pub async fn checkout(&self, version: i64) -> napi::Result<()> {
        self.inner_ref()?
@@ -1420,31 +1387,6 @@ pub struct RefreshColumnResult {
    pub version: i64,
 }

-#[napi(object)]
-pub struct RefreshMaterializedViewResult {
-    /// How the view was brought up to date: "rebuild", "incremental" or "no_op".
-    pub mode: String,
-    pub rows_written: i64,
-    pub source_version: i64,
-    pub version: i64,
-}
-
-impl From<lancedb::RefreshMaterializedViewResult> for RefreshMaterializedViewResult {
-    fn from(value: lancedb::RefreshMaterializedViewResult) -> Self {
-        let mode = match value.mode {
-            lancedb::RefreshMode::Rebuild => "rebuild",
-            lancedb::RefreshMode::Incremental => "incremental",
-            lancedb::RefreshMode::NoOp => "no_op",
-        };
-        Self {
-            mode: mode.to_string(),
-            rows_written: value.rows_written as i64,
-            source_version: value.source_version as i64,
-            version: value.version as i64,
-        }
-    }
-}
-
 impl From<lancedb::table::RefreshColumnResult> for RefreshColumnResult {
    fn from(value: lancedb::table::RefreshColumnResult) -> Self {
        Self {
@@ -1663,18 +1605,18 @@ impl Branches {
    }

    #[napi(ts_return_type = "Promise<Record<string, unknown>>")]
-    pub async fn cherry_pick(
+    pub async fn merge(
        &self,
        from_branch: String,
        dry_run: Option<bool>,
    ) -> napi::Result<serde_json::Value> {
        let result = self
            .inner
-            .cherry_pick(&from_branch, dry_run.unwrap_or(false))
+            .merge_branch(&from_branch, dry_run.unwrap_or(false))
            .await
            .default_error()?;
        serde_json::to_value(result).map_err(|err| {
-            napi::Error::from_reason(format!("failed to serialize cherry-pick result: {err}"))
+            napi::Error::from_reason(format!("failed to serialize branch merge result: {err}"))
        })
    }
 }
@@ -0,0 +1,21 @@
+{
+  "name": "lancedb",
+  "description": "Write, review, debug, and document LanceDB pipelines in Python and TypeScript that work across local LanceDB OSS tables and remote LanceDB Enterprise/Cloud tables, with idiomatic query/search patterns and performance defaults for ingestion, indexing, filtering, and diagnostics.",
+  "version": "0.1.0",
+  "author": {
+    "name": "LanceDB"
+  },
+  "homepage": "https://www.lancedb.com",
+  "keywords": [
+    "lancedb",
+    "vector-search",
+    "full-text-search",
+    "hybrid-search",
+    "python",
+    "typescript",
+    "pipelines",
+    "ingestion",
+    "indexing",
+    "performance"
+  ]
+}
@@ -0,0 +1,33 @@
+{
+  "name": "lancedb",
+  "version": "0.1.0",
+  "description": "Codex plugin for building LanceDB pipelines in Python and TypeScript.",
+  "author": {
+    "name": "LanceDB"
+  },
+  "keywords": [
+    "lancedb",
+    "vector-search",
+    "full-text-search",
+    "hybrid-search",
+    "python",
+    "typescript",
+    "pipelines"
+  ],
+  "skills": "./skills/",
+  "interface": {
+    "displayName": "LanceDB",
+    "shortDescription": "Build LanceDB pipelines in Python and TypeScript.",
+    "longDescription": "Write, review, debug, and document LanceDB pipelines in Python and TypeScript that work across local LanceDB OSS tables and remote LanceDB Enterprise/Cloud tables, with idiomatic query/search patterns and performance defaults for ingestion, indexing, filtering, and diagnostics.",
+    "developerName": "LanceDB",
+    "websiteURL": "https://www.lancedb.com",
+    "category": "Developer Tools",
+    "capabilities": [
+      "Developer Tools"
+    ],
+    "defaultPrompt": "Create a LanceDB table, embed sample text, and run a vector search.",
+    "composerIcon": "./assets/logo.png",
+    "logo": "./assets/logo.png",
+    "logoDark": "./assets/logo-dark.png"
+  }
+}
@@ -0,0 +1,102 @@
+---
+name: lancedb
+description: Use when writing, reviewing, debugging, or documenting LanceDB pipelines in Python or TypeScript, especially code that should work across local LanceDB OSS tables and remote LanceDB Enterprise/Cloud tables. Helps avoid non-portable full-table materialization, choose idiomatic query/search patterns, apply LanceDB performance defaults for ingestion, indexing, filtering, and diagnostics, and resolve connections to the remote server for Enterprise-only operations such as jobs.
+---
+
+# Building LanceDB Pipelines
+
+Use this skill to produce LanceDB pipelines that are portable between local and remote tables (for LanceDB Enterprise/Cloud) and idiomatic for the selected SDK.
+
+## LanceDB Table Modes
+
+LanceDB has two common execution modes:
+
+- **Local table**: embedded, open source, in-process LanceDB. The client opens data from a local path or object storage URI and executes queries in the application process.
+- **Remote table**: LanceDB Enterprise/Cloud table opened through a `db://...` URI. The data may be very large, commonly backed by object storage, and queried through a remote service.
+
+Do NOT assume local-only table helpers exist on remote tables. If the user asks for LanceDB Enterprise, Cloud, `db://...`, production remote access, or a remote table, focus on the remote table path: use `search()` / `query()`, keep reads bounded with `select()` and `limit()`, and avoid table-level full materialization APIs.
+
+## Workflow
+
+1. Identify the SDK: Python, TypeScript, or both.
+2. Identify the table mode: local/embedded OSS, remote Enterprise/Cloud, or portable across both. If the user says "LanceDB Enterprise", choose the remote table path. If the task involves jobs in any way (listing, inspecting, creating, or canceling jobs), it is always the remote path and requires a remote server connection — see "Connecting to the LanceDB remote server" below before doing anything else.
+3. Read the matching topic reference before writing or changing code:
+   - Column metadata authoring (both SDKs): `references/column_metadata.md`
+   - Branch operations (both SDKs): `references/branch_ops.md`
+   - Remote server connection resolution (jobs, raw REST): `references/remote_connect.md`
+   - Job operations REST API (list/describe/cancel/query_events): `references/remote_jobs.md`
+
+   There is no bundled per-language guide. For exact method names, signatures, and options, look them up in the canonical sources instead of relying on memory:
+   - Python: `docs/src/python/python.md` (the hand-maintained API reference) and the source under `python/python/lancedb/` when working inside the LanceDB repo; otherwise <https://lancedb.github.io/lancedb/python/python/>.
+   - TypeScript: the generated typedoc under `docs/src/js/` and the source under `nodejs/lancedb/` when working inside the LanceDB repo; otherwise <https://lancedb.github.io/lancedb/js/globals/>.
+4. Apply the SDK invariants in "Per-SDK Invariants" below. Read `column_metadata.md` when the task is documenting, tagging, classifying, or grouping table columns (field descriptions, `lancedb:tag:*` tags, logical column families). Read `branch_ops.md` when the task involves branch lifecycle (list/create/delete), writing to a non-main branch, or verifying a change stayed off main. Read `remote_connect.md` when the task involves jobs or direct REST access to an Enterprise deployment, and `remote_jobs.md` for the job REST methods themselves (list, describe, cancel, query_events).
+5. For Python schemas, favor Pydantic models and validate records before writing. Use PyArrow schemas when Arrow-native, streaming, or highly dynamic data makes them materially better suited.
+6. Prefer `search()` or `query()` builders with explicit `select()` and `limit()` for reads.
+7. Avoid table-level full materialization in remote or portable code. This is the main local-vs-remote read pitfall.
+8. After a successful embedded OSS ingestion, call `table.optimize()`. Do not call it for Enterprise/Cloud; remote maintenance is automatic.
+9. For remote Enterprise/Cloud writes, never drop-then-reuse or `mode="overwrite"` the same table name — see "Enterprise: never drop-then-reuse the same table name" below. This is the main local-vs-remote write pitfall.
+10. If reviewing an existing file or repo, run `scripts/check_materialization.py` on the relevant paths and inspect each finding before editing.
+11. Cross-check unfamiliar or non-trivial API claims against the source tree instead of relying on memory.
+
+## Core Portability Rule
+
+Do not write code that assumes a local table API will exist on a remote table. Remote tables can be very large, so whole-table materialization helpers are intentionally unavailable or unsafe.
+
+This does **not** mean result conversion is forbidden. Bounded query/search result collection is normal:
+
+- Python: `table.search(...).select([...]).limit(10).to_pandas()`
+- TypeScript: `await table.search(...).select([...]).limit(10).toArray()`
+
+The unsafe pattern is table-level or unbounded collection, plus local-only dataset escape hatches in remote code:
+
+- Python: `table.to_pandas()`, `table.to_arrow()`, `table.to_polars()`; `table.to_lance()` is local/OSS-only dataset access, not materialization
+- TypeScript: `await table.toArrow()`, `await table.query().toArray()` without `limit()`
+
+## Per-SDK Invariants
+
+Python:
+
+- Result collectors: default to `.to_list()` (plain dicts, no extra dependency) or `.to_arrow()` (PyArrow ships with LanceDB). Use `.to_pandas()` / `.to_polars()` only when the project already declares that dependency — do not assume pandas or polars is installed.
+- Plain scans differ by client: the sync client has no `.query()` method — use `table.search()` with no argument; the async client uses `await async_table.query()`.
+
+TypeScript:
+
+- Collect bounded results with `.toArray()` (objects) or `.toArrow()` (Arrow) after `select()` and `limit()`.
+- For large reads, stream batches instead of collecting: `for await (const batch of table.query().where(...).select(...).limit(...)) { ... }`.
+
+Both SDKs:
+
+- Ingest in bulk or in batches of thousands of rows; never write per-row in a loop — each write creates a version and fragment, slowing ingestion and later queries.
+- Build a vector index once brute-force search is too slow (rule of thumb: beyond roughly 100K vectors locally), and scalar indexes for filtered columns and merge/upsert keys. Use index defaults unless the task states recall/latency requirements.
+
+## Enterprise: never drop-then-reuse the same table name
+
+LanceDB Enterprise/Cloud splits a **control plane** (DDL: create/drop/rename) from a **data plane** (query nodes that serve reads). Query nodes cache the resolved dataset for a table name for up to `table_cache_ttl` — **default 300 seconds (5 minutes)**. After you drop or overwrite a table, the control plane updates immediately but the data plane keeps serving the *old* dataset until that cache entry expires. During the window the two planes disagree.
+
+The failure this causes: you `drop_table("t")` then immediately `create_table("t", ...)` (or `create_table("t", ..., mode="overwrite")`). The DDL returns success, but every query against `t` returns **`500 Internal Server Error`** (the query node resolves the stale/deleted dataset), and a fresh `describe` may still show the *old* schema/version. It looks like your write silently failed; it didn't — the name is cached.
+
+**`mode="overwrite"` has the same problem** — it is a drop+create of the same name under the hood.
+
+Rules for portable Enterprise ingestion:
+
+1. **Never reuse a table name you just dropped/overwrote within the cache TTL.** Do not use `mode="overwrite"` to replace an existing Enterprise table in place.
+2. To (re)load data, **write to a fresh table name** (e.g. `<table>_v2`, or a run-stamped suffix). A brand-new name has no cached data-plane entry, so writes and reads work immediately.
+3. Before creating, `list_tables()` and **fail loudly if the name already exists** rather than overwriting — prompt for a new name.
+4. To land on a specific final name that is currently occupied by an old table: drop the old table, **wait out the TTL (~5 min), then `rename_table(fresh_name, final_name)`**. Renaming onto a name whose old dataset is still cached hits the same race, so the wait is mandatory. `rename_table` is a supported control-plane op.
+5. When you hand a table name back to a human, tell them which step still needs the propagation wait (usually: "the old `t` was dropped; run the rename in ~5 minutes").
+
+This is Enterprise/Cloud-specific. Local/OSS tables have no separate data plane, so `mode="overwrite"` and immediate same-name reuse are fine there.
+
+## Connecting to the LanceDB remote server
+
+LanceDB Enterprise/Cloud deployments are served by a server implementing the lance-namespace OpenAPI spec (<https://github.com/lance-format/lance-namespace/blob/main/docs/src/spec.yaml>). Every remote (`db://...`) connection talks to such a server, and some operations exist only there. In particular, **all operations around jobs (listing, inspecting, creating, or canceling jobs) run server-side** — there is no local/OSS equivalent. Before any job work, or any direct REST call to an Enterprise deployment, read `references/remote_connect.md` to resolve the base URL, credentials, and database header and to validate the connection. Then use the four job REST methods documented in `references/remote_jobs.md` (list, describe, cancel, query_events).
+
+## Script
+
+Run the scanner when reviewing or modifying an existing codebase:
+
+```bash
+python skills/lancedb/scripts/check_materialization.py path/to/file_or_dir
+```
+
+The script reports likely unsafe full-table materialization in Python and TypeScript. Treat results as review prompts, not automatic proof of a bug.
@@ -0,0 +1,6 @@
+interface:
+  display_name: "LanceDB"
+  short_description: "Build LanceDB pipelines in Python and TypeScript"
+  default_prompt: "Use $lancedb to create a table, embed sample text, and run a vector search."
+  icon_small: "./assets/icon.png"
+  icon_large: "./assets/icon.png"
@@ -0,0 +1,182 @@
+# Branch Operations
+
+Manage branches on a LanceDB table: list what exists, create new ones, delete stale ones, and direct read/write operations at a specific branch without touching main. Use for branch lifecycle tasks, experimental/isolated table versions, targeting an operation at a non-main branch, or confirming a mutation did not affect main.
+
+Works on local/OSS and remote Enterprise/Cloud tables, except merging a branch into main, which is Enterprise-only.
+
+## The branch model (important)
+
+Branches are isolated, writable lines of history forked from another branch (or a specific version). Writes on a branch never affect `main`.
+
+There is **no global "switch branch" state** — you never repoint the whole table at a branch. Instead, **operations are scoped by which table handle you use**:
+
+- The handle you got from `open_table(name)` / `openTable(name)` targets `main`.
+- `branches.create(...)` and `branches.checkout(...)` return a **new table handle scoped to that branch**. Every read/write on that handle (add, update, `update_field_metadata`, `create_index`, search, …) lands on the branch.
+- The original main handle is unaffected — keep it around to verify isolation.
+
+`branches.list()` returns only non-main branches. Main always exists and is not listed.
+
+## Python
+
+`table.branches` is a property returning the branch manager; `table.current_branch()` tells you what a handle is scoped to (`None` = main).
+
+```python
+table = db.open_table("products")            # scoped to main
+
+# list — dict of name -> metadata (parent_branch, parent_version, ...); {} = only main
+table.branches.list()
+
+# create: forks from main by default and returns a handle scoped to the new branch
+exp = table.branches.create("experiment-reindex")
+exp = table.branches.create("exp2", from_ref="main", from_version=None)  # optional fork point
+
+# checkout an existing branch -> branch-scoped handle
+wip = table.branches.checkout("wip-branch")
+# with version= it pins to that version (read-only detached view); omit to track latest, writable
+
+# operate on the branch simply by using its handle
+wip.update_field_metadata(
+    {"path": "category", "metadata": {"lancedb:description": "Product category label."}}
+)
+wip.create_scalar_index("category")
+
+# delete: removes only the branch pointer; main and row data remain intact
+table.branches.delete("stale-2024")
+
+# alternatively, open a branch handle directly from the connection
+wip = db.open_table("products", branch="wip-branch")
+
+exp.current_branch()    # "experiment-reindex"
+table.current_branch()  # None (main)
+```
+
+Async: same shape — `table.branches` returns `AsyncBranches`; `await table.branches.create(...)` etc.
+
+## TypeScript
+
+`table.branches()` is an **async method** returning the `Branches` manager; `table.currentBranch()` returns the scoped branch or `null` for main.
+
+```typescript
+const table = await db.openTable("products"); // scoped to main
+const branches = await table.branches();
+
+// list — Record<string, BranchContents>; {} = only main
+await branches.list();
+
+// create: forks from main by default, returns a Table scoped to the new branch
+const exp = await branches.create("experiment-reindex");
+const exp2 = await branches.create("exp2", "main" /* fromRef */, undefined /* fromVersion */);
+
+// checkout an existing branch -> branch-scoped Table
+const wip = await branches.checkout("wip-branch");
+// with a version arg it pins (read-only detached view); omit to track latest, writable
+
+// operate on the branch simply by using its handle
+await wip.updateFieldMetadata([
+  { path: "category", metadata: { "lancedb:description": "Product category label." } },
+]);
+await wip.createIndex("category");
+
+// delete: removes only the branch pointer; main and row data remain intact
+await branches.delete("stale-2024");
+
+// alternatively, open a branch handle directly from the connection
+const wip2 = await db.openTable("products", { branch: "wip-branch" });
+
+exp.currentBranch();   // "experiment-reindex"
+table.currentBranch(); // null (main)
+```
+
+## Verifying isolation
+
+After writing to a branch, confirm the change did NOT land on main by reading through both handles:
+
+```python
+wip = table.branches.checkout("wip-branch")
+wip.update_field_metadata({"path": "category", "metadata": {"lancedb:description": "..."}})
+
+assert b"lancedb:description" in (wip.schema.field("category").metadata or {})
+assert b"lancedb:description" not in (table.schema.field("category").metadata or {})  # main untouched
+```
+
+Two handles on the same branch see each other's writes (e.g. `table.branches.create("exp")` and `db.open_table(name, branch="exp")`); main stays isolated.
+
+## Merging a branch into main (Enterprise only)
+
+Merge is available through the SDKs (`table.branches.merge(...)`) on **Enterprise tables only** — it is not supported on Cloud or local/OSS tables, which raise `NotSupported`.
+
+`merge` takes the branch to merge **from** and a `dry_run` flag. Both the SDK method and the underlying REST endpoint **actually merge by default** (`dry_run=False`); pass `dry_run=True` to only preview. A rejected merge is **not an exception** — it returns a result with `status="rejected"` rather than raising, so inspect the return value. Use `branches.diff(from_branch)` to inspect a branch's pending diff without attempting a merge.
+
+```python
+exp = "experiment-reindex"
+
+# preview only — returns status="ready" if it would merge cleanly
+preview = table.branches.merge(exp, dry_run=True)
+
+# actually merge (default)
+result = table.branches.merge(exp)
+if result["status"] == "merged":
+    print("landed at", result["mainVersionAfter"])
+elif result["status"] == "rejected":
+    print(result["diff"]["mergeBlockers"])   # why it was refused
+
+# inspect a branch's pending diff without merging
+diff = table.branches.diff(exp)
+```
+
+Async: `await table.branches.merge(exp)`, `await table.branches.diff(exp)`.
+
+```typescript
+const branches = await table.branches();
+const exp = "experiment-reindex";
+
+// preview only (second arg is dryRun)
+const preview = await branches.merge(exp, true);
+
+// actually merge (default)
+const result = await branches.merge(exp);
+if (result.status === "merged") {
+  console.log("landed at", result.mainVersionAfter);
+} else if (result.status === "rejected") {
+  console.log(result.diff.mergeBlockers);
+}
+
+const diff = await branches.diff(exp);
+```
+
+The result is the wire JSON, containing `status` (`ready` on a passing dry run, `merged` on success, `rejected` when refused — also `notImplemented`/`unknown`), the branch `diff` (including `mergeBlockers` explaining any rejection), a `preview` of the columns that would be promoted, and — after a real merge — `mainVersionAfter`.
+
+### Merge preconditions
+
+Merge only **promotes newly added columns** onto main; it does not replay arbitrary commits. Practically, a branch is mergeable only if it has **exactly one commit since it was created, and that commit added a column**. The merge is rejected (`status: "rejected"`, with `mergeBlockers` set) if:
+
+- the branch was forked from another branch rather than directly from main
+- main has advanced since the branch was forked
+- the branch's rows changed since the fork (row counts must match main exactly)
+- the branch removed columns or changed a column's type/nullability
+- the branch added no columns (index-only changes are not merged)
+
+### Adding a column in a single commit
+
+Because the branch must contain just one column-adding commit, add the column with its values in one operation rather than add-then-backfill:
+
+1. **SQL transformation** — `add_columns` with a SQL expression computed from existing columns, so the column lands populated in one commit.
+2. **Precompute the values** — compute the column's values externally, then add the fully-populated column in a single operation (e.g. via `merge_insert`/`add_columns` with the data ready).
+3. **Lance-format-level data evolution (pylance)** — use Lance's data evolution with backfill, documented at <https://lance.org/guide/data_evolution/#with-data-backfill>.
+
+## Quick reference
+
+| Goal | Python | TypeScript |
+|------|--------|------------|
+| List branches (non-main) | `table.branches.list()` | `await (await table.branches()).list()` |
+| Create branch (off main) | `table.branches.create(name)` → branch handle | `await branches.create(name)` → branch `Table` |
+| Create from a fork point | `table.branches.create(name, from_ref=..., from_version=...)` | `await branches.create(name, fromRef, fromVersion)` |
+| Get a branch handle | `table.branches.checkout(name)` or `db.open_table(t, branch=name)` | `await branches.checkout(name)` or `await db.openTable(t, { branch: name })` |
+| Pin to a branch version (read-only) | `table.branches.checkout(name, version=v)` | `await branches.checkout(name, v)` |
+| Delete branch | `table.branches.delete(name)` | `await branches.delete(name)` |
+| Which branch is this handle on? | `table.current_branch()` (`None` = main) | `table.currentBranch()` (`null` = main) |
+| Target main | use the original (non-branch) handle | use the original (non-branch) handle |
+| Merge branch into main (Enterprise only) | `table.branches.merge(from_branch, dry_run=False)` | `await branches.merge(fromBranch, dryRun)` |
+| Preview a branch's pending diff (Enterprise only) | `table.branches.diff(from_branch)` | `await branches.diff(fromBranch)` |
+
+Branch names must be non-empty; empty names raise a validation error.
@@ -0,0 +1,183 @@
+# Column Metadata Authoring
+
+Write column-level descriptions, tags, and logical groupings onto a LanceDB table's schema. Use this when the user wants to document, annotate, tag, or classify what their table columns ARE (embeddings vs labels vs eval metrics, model provenance, version families, etc.).
+
+Works on local/OSS and remote Enterprise/Cloud tables alike — read the schema through the table handle, write through `update_field_metadata` (Python) / `updateFieldMetadata` (TypeScript).
+
+## Metadata key conventions
+
+All metadata uses namespaced keys:
+
+| Key | Purpose | Example value |
+|-----|---------|---------------|
+| `lancedb:description` | Human-readable explanation of what the column contains | `"CLIP ViT-L/14 image embedding, L2-normalized (768-dim)"` |
+| `lancedb:tag:<name>` | Flexible key-value tag; the suffix names the tag category | `lancedb:tag:field_type: "embedding"`, `lancedb:tag:model: "clip"`, `lancedb:tag:project_id: "foo"` |
+| `lancedb:logical-column` | Logical group/family this column belongs to | `"clip_features"` |
+
+Tags are open-ended — use whatever key suffix and value make sense given the user's intent. The tag suffix should describe *what is being classified* (e.g., `field_type`, `model`, `project_id`) and the value describes *how*. Multiple tags on the same column are fine — each is a separate key. All values are strings.
+
+## Step 1: Read the schema and existing metadata
+
+Read existing metadata before writing, to avoid redundant updates.
+
+Python — `table.schema` (sync property; async: `await table.schema()`) returns a `pyarrow.Schema`. **Arrow field metadata is bytes-keyed in Python**:
+
+```python
+schema = table.schema
+for field in schema:
+    meta = field.metadata or {}  # dict[bytes, bytes], e.g. {b"lancedb:description": b"..."}
+    print(field.name, field.type, field.nullable, meta)
+```
+
+TypeScript — `await table.schema()` returns an Arrow `Schema`; field metadata is a `Map<string, string>`:
+
+```typescript
+const schema = await table.schema();
+for (const field of schema.fields) {
+  console.log(field.name, field.type, field.nullable, field.metadata); // Map
+  // field.metadata.get("lancedb:description")
+}
+```
+
+For struct/nested fields, recurse into the field's children and address them as dot-paths (e.g., `parent.child`).
+
+If the user hasn't specified which columns to update, work with all columns.
+
+## Step 2: Generate metadata
+
+Decide what to generate based on the user's request.
+
+### Descriptions (`lancedb:description`)
+
+Base descriptions on:
+- The column name and Arrow type (e.g., `FixedSizeList` of floats → likely an embedding)
+- User-supplied context (upstream pipeline, sample values, domain knowledge)
+- Name patterns: `_embedding`/`_vec`/`_embed` → vector; `_label`/`_class` → label; `_score`/`_eval`/`_metric` → evaluation metric
+
+Be specific and concise. Good: `"Sentence-BERT embedding of the query text (768-dim)."` Not: `"An embedding column."`
+
+### Tags (`lancedb:tag:<name>`)
+
+Choose tag key names that match what the user asked to annotate. Common patterns:
+
+- Semantic field type → `lancedb:tag:field_type: "embedding"` / `"text"` / `"image"` / `"label"` / `"eval"` / `"id"` / `"metadata"`
+- Model or source → `lancedb:tag:model: "clip"` / `"bert"` / `"vit"`
+- Project affiliation → `lancedb:tag:project_id: "<name>"`
+- Version → `lancedb:tag:version: "v3"` (and `lancedb:tag:latest: "true"` for the newest)
+
+Use Arrow type as a hint: `FixedSizeList` + float → embedding; `Utf8`/`LargeUtf8` → text; `Binary` → image or blob.
+
+### Logical groupings (`lancedb:logical-column`)
+
+Look for naming patterns across columns:
+- `clip_v1`, `clip_v2`, `clip_v3` → logical column `"clip"`, latest is `v3`
+- `text_embed_20240101`, `text_embed_20240601` → logical column `"text_embed"`, latest is the most recent date suffix
+
+Write `lancedb:logical-column` on all members of a group. Mark the newest with `lancedb:tag:latest: "true"` (in addition to its version tag).
+
+## Step 3: Write the metadata
+
+Each update names a field by dot-path and carries a metadata map. Semantics (identical in both SDKs):
+
+- **Merge by default** (`replace` omitted/false) — preserves existing metadata the user didn't ask to change
+- `replace: true` swaps the field's entire metadata map — only if the user explicitly asks to overwrite
+- A value of `None`/`null` deletes that specific key
+- Batch all field updates into a single call when possible
+- Returns the new table version
+
+Python (sync and async take one dict per field, as varargs):
+
+```python
+res = table.update_field_metadata(
+    {
+        "path": "clip_v3",
+        "metadata": {
+            "lancedb:description": "CLIP ViT-L/14 image embedding, L2-normalized (1024-dim).",
+            "lancedb:tag:field_type": "embedding",
+            "lancedb:tag:model": "clip",
+            "lancedb:tag:version": "v3",
+            "lancedb:tag:latest": "true",
+            "lancedb:logical-column": "clip",
+        },
+    },
+    {
+        "path": "clip_v2",
+        "metadata": {
+            "lancedb:description": "CLIP ViT-B/32 image embedding (768-dim), superseded by v3.",
+            "lancedb:tag:field_type": "embedding",
+            "lancedb:tag:model": "clip",
+            "lancedb:tag:version": "v2",
+            "lancedb:logical-column": "clip",
+        },
+    },
+)
+print(res.version)  # new table version
+
+# merge semantics: add a key, delete one via None, keep the rest
+table.update_field_metadata(
+    {"path": "clip_v2", "metadata": {"lancedb:tag:archived": "true", "lancedb:tag:latest": None}}
+)
+```
+
+(`replace_field_metadata` is deprecated — use `update_field_metadata`.)
+
+TypeScript (takes an array of `FieldMetadataUpdate`):
+
+```typescript
+const res = await table.updateFieldMetadata([
+  {
+    path: "clip_v3",
+    metadata: {
+      "lancedb:description": "CLIP ViT-L/14 image embedding, L2-normalized (1024-dim).",
+      "lancedb:tag:field_type": "embedding",
+      "lancedb:tag:model": "clip",
+      "lancedb:tag:version": "v3",
+      "lancedb:tag:latest": "true",
+      "lancedb:logical-column": "clip",
+    },
+  },
+  {
+    path: "clip_v2",
+    metadata: {
+      "lancedb:description": "CLIP ViT-B/32 image embedding (768-dim), superseded by v3.",
+      "lancedb:tag:field_type": "embedding",
+      "lancedb:tag:model": "clip",
+      "lancedb:tag:version": "v2",
+      "lancedb:logical-column": "clip",
+    },
+  },
+]);
+console.log(res.version); // new table version
+
+// merge semantics: add a key, delete one via null, keep the rest
+await table.updateFieldMetadata([
+  { path: "clip_v2", metadata: { "lancedb:tag:archived": "true", "lancedb:tag:latest": null } },
+]);
+```
+
+## Step 4: Confirm
+
+Report back:
+- Which columns were updated and what was written
+- The new table version number (from the result)
+- Any columns skipped (e.g., already had up-to-date metadata)
+
+## Quick examples
+
+**"Write descriptions for all columns in the `product_embeddings` table"**
+1. Read `table.schema` → all fields + existing metadata
+2. Generate a `lancedb:description` for each column based on name + type
+3. One `update_field_metadata` call with all descriptions
+4. Report
+
+**"Tag the columns in `model_outputs` with their field type and model"**
+1. Read the schema
+2. For each field, classify by name + Arrow type → set `lancedb:tag:field_type` and `lancedb:tag:model` where applicable
+3. Write in one batched call
+4. Report
+
+**"Group the feature columns in `training_features` into logical families and mark the latest version"**
+1. Read the schema
+2. Find version patterns → assign `lancedb:logical-column` and `lancedb:tag:version`; mark newest with `lancedb:tag:latest: "true"`
+3. Write in one batched call
+4. Show the grouping
@@ -0,0 +1,45 @@
+# Connecting to a LanceDB remote server
+
+LanceDB Enterprise/Cloud deployments are served by a server implementing the
+lance-namespace OpenAPI spec
+(<https://github.com/lance-format/lance-namespace/blob/main/docs/src/spec.yaml>).
+Every remote (`db://...`) connection talks to such a server, and some operations
+exist only there. In particular, all operations around jobs (listing, inspecting,
+creating, or canceling jobs) run server-side — there is no local/OSS equivalent, so
+resolve a server connection before attempting any job work. The job REST methods
+themselves are documented in `references/remote_jobs.md`.
+
+Every request needs two things:
+
+1. **Base URL** — the server endpoint
+2. **Credentials** — an API key (`x-api-key` header over REST), and usually a database name (`x-lancedb-database` header)
+
+## Resolution steps
+
+1. If the user already gave a URL and API key (or said which environment they're working against), use that.
+2. Otherwise, look for credentials already available in the environment:
+   - Env vars like `LANCEDB_URI` / `LANCEDB_HOST` / `LANCEDB_API_KEY`
+   - A server endpoint already running or port-forwarded locally (the REST default port is 2333, i.e. `http://localhost:2333`)
+3. If you didn't find both pieces, ask the user directly: **"What's your LanceDB endpoint's URL, and what's your API key?"** Also ask which database to use if it isn't obvious. Don't guess or probe further — the user knows their deployment.
+
+## Validating the connection
+
+Make a cheap authenticated request and check the status before starting real work:
+
+```bash
+curl -s -w "\n%{http_code}" "{base_url}/v1/table/?limit=1" \
+  -H "x-api-key: <key>" \
+  -H "x-lancedb-database: <database>"
+```
+
+- `200` — connection, key, and database header all good
+- `401` — API key missing or wrong
+- `400` mentioning a database header — this deployment expects `x-lancedb-database`
+
+## Non-REST equivalents
+
+The same credentials work through the SDKs and CLI:
+
+- Python SDK: `lancedb.connect("db://<database>", api_key="<key>", host_override="<base_url>")`
+- TypeScript SDK: `await lancedb.connect("db://<database>", { apiKey: "<key>", hostOverride: "<base_url>" })`
+- `lancedb` CLI: a `[profiles.<name>]` entry in `~/.lancedb/config.toml` with `http_server_url`, `api_key`, `database`
@@ -0,0 +1,151 @@
+# Job operations over the LanceDB remote server REST API
+
+Jobs are server-side background operations on LanceDB Enterprise/Cloud — index builds,
+column backfills, materialized view refreshes, and similar async work. Endpoints that
+trigger async work (e.g. the column backfill or materialized view refresh endpoints)
+return a `job_id`; these four methods are how you track and manage those jobs.
+
+Resolve the connection first — see `references/remote_connect.md`. All four methods
+are **POST** requests under `{base_url}/v1/jobs/` with JSON bodies, and take the usual
+`x-api-key` / `x-lancedb-database` headers. If every job call returns `501`, job APIs
+are disabled on that deployment (the server has no job registry configured) — report
+that rather than retrying.
+
+## 1. List jobs — `POST /v1/jobs/list`
+
+The body is optional; an empty body lists everything. All fields are filters:
+
+```json
+{
+  "limit": 100,
+  "table_name": "my_table",
+  "job_type": "...",
+  "job_subtype": "...",
+  "state": "...",
+  "page_token": "..."
+}
+```
+
+```bash
+curl -s -X POST "{base_url}/v1/jobs/list" \
+  -H "x-api-key: <key>" -H "x-lancedb-database: <database>" \
+  -H "content-type: application/json" \
+  -d '{"table_name": "my_table"}'
+```
+
+Response:
+
+```json
+{
+  "jobs": [
+    {
+      "job_id": "...",
+      "table": "my_table",
+      "job_type": "...",
+      "job_subtype": "...",
+      "state": "done",
+      "created_at_millis": 1720000000000
+    }
+  ],
+  "page_token": "..."
+}
+```
+
+A `page_token` in the response means there are more results — pass it back in the next
+request to continue. Note list rows use a lowercase `state` string, while describe uses
+an uppercase `job_state`.
+
+## 2. Describe a job — `POST /v1/jobs/describe`
+
+Body: `{"job_id": "<id>"}`. Returns full detail for one job:
+
+```json
+{
+  "job_id": "...",
+  "job_type": "...",
+  "job_subtype": "...",
+  "job_state": "IN_PROGRESS",
+  "creation_ms": 1720000000000,
+  "spec": {},
+  "status": {}
+}
+```
+
+`job_state` is one of `IN_PROGRESS`, `CANCELLED`, `FAILED`, `DONE`. `spec` and `status`
+are job-type-specific JSON objects (the job's input specification and its current
+progress/status). Returns `404` for an unknown job id.
+
+## 3. Cancel a job — `POST /v1/jobs/cancel`
+
+Body: `{"job_id": "<id>"}`; response echoes `{"job_id": "<id>"}`. Cancellation is a
+service-level operation requiring the same administrative authorization as the
+`/admin` routes — a database-scoped API key that can list and describe jobs may still
+get a permission error here. Other errors: `404` unknown job, `409` state conflict
+(e.g. already in a terminal state), `429` too much write contention (safe to retry).
+
+## 4. Query job event history — `POST /v1/jobs/query_events`
+
+Returns the event history (state transitions, progress updates) for one or more jobs.
+Body: `{"job_id": "<id>"}` for one job, or `{"job_ids": ["<id>", ...]}` for a batch.
+Optional fields: `limit` (max event rows), `limit_per_job` (per job in a batch query),
+and `filter` — a SQL-like expression over the columns `state`, `updated_by`,
+`owner_component`, and `claim_entity`. (`full_text_search` is reserved and currently
+rejected as not implemented.)
+
+The response is **not JSON** — it is an Arrow IPC stream
+(`content-type: application/vnd.apache.arrow.stream`). Decode it, e.g. in Python:
+
+```python
+import pyarrow.ipc
+import requests
+
+resp = requests.post(
+    f"{base_url}/v1/jobs/query_events",
+    headers={"x-api-key": key, "x-lancedb-database": database},
+    json={"job_id": job_id},
+)
+resp.raise_for_status()
+events = pyarrow.ipc.open_stream(resp.content).read_all()
+```
+
+## Feature engineering (Geneva) jobs
+
+Feature engineering jobs — UDF column backfills and materialized view refreshes run
+through Geneva — are tracked **separately** from the `/v1/jobs` registry above. Their
+records live in a `geneva_jobs` table inside the database itself (in the `__system`
+namespace), and you access them through a Python `geneva` connection rather than the
+REST endpoints above:
+
+```python
+import geneva
+from geneva.jobs import JobStateManager
+
+# Same credentials as lancedb.connect / the REST API
+conn = geneva.connect("db://<database>", api_key="<key>", host_override="<base_url>")
+jsm = JobStateManager(conn)
+
+# List jobs. NOTE: status defaults to "RUNNING"; pass status=None for all jobs.
+# Statuses: PENDING | RUNNING | DONE | FAILED | CANCELLED
+jobs = jsm.list_jobs(table_name="my_table", status=None)
+
+# Fetch one job by id (returns a list of JobRecord)
+records = jsm.get("<job_id>")
+```
+
+Each `JobRecord` has `table_name`, `column_name`, `job_id`, `job_type`, `status`,
+`launched_at`, `completed_at`, `config`, `launched_by`, `manifest_id`, `cluster_name`,
+`metrics` (progress counters), `events` (human-readable history), and `updated_at`.
+For filters `list_jobs` doesn't support (e.g. time ranges), query the underlying table
+directly: `jsm.get_table(True).search().where("launched_at >= TIMESTAMP '...'")` —
+pass `True` to check out the latest version, since other processes update job state.
+
+Stale-status caveat: nothing reaps dead Geneva jobs, so a job can sit in
+`RUNNING`/`PENDING` forever if its worker died. Treat a job as effectively `FAILED`
+when it has been running longer than ~36 hours, or its `updated_at` is more than ~2
+hours old (this matches the heuristic the Geneva console UI applies on read).
+
+## Workflow tips
+
+- To wait for async work (a backfill, an index build), poll `describe` until
+  `job_state` leaves `IN_PROGRESS`; on `FAILED`, pull `status` and `query_events` for
+  the failure detail.
@@ -0,0 +1,137 @@
+#!/usr/bin/env python3
+"""Scan Python and TypeScript for likely unsafe LanceDB materialization."""
+
+from __future__ import annotations
+
+import argparse
+import re
+import sys
+from dataclasses import dataclass
+from pathlib import Path
+
+
+PY_FULL_TABLE = re.compile(r"\b\w+\.(to_pandas|to_arrow|to_polars)\s*\(")
+TS_TABLE_TO_ARROW = re.compile(r"\b\w+\.toArrow\s*\(")
+TS_QUERY_COLLECTOR = re.compile(r"\.query\s*\(\s*\)[\s\S]*?\.to(Array|Arrow)\s*\(")
+
+
+@dataclass
+class Finding:
+    path: Path
+    line: int
+    message: str
+    text: str
+
+
+def iter_files(paths: list[Path]) -> list[Path]:
+    files: list[Path] = []
+    for path in paths:
+        if path.is_dir():
+            files.extend(
+                p
+                for p in path.rglob("*")
+                if p.suffix in {".py", ".ts", ".tsx"} and "node_modules" not in p.parts
+            )
+        elif path.suffix in {".py", ".ts", ".tsx"}:
+            files.append(path)
+    return sorted(set(files))
+
+
+def line_number(text: str, offset: int) -> int:
+    return text.count("\n", 0, offset) + 1
+
+
+def scan_python(path: Path, text: str) -> list[Finding]:
+    findings: list[Finding] = []
+    for match in PY_FULL_TABLE.finditer(text):
+        line_start = text.rfind("\n", 0, match.start()) + 1
+        line_end = text.find("\n", match.start())
+        if line_end == -1:
+            line_end = len(text)
+        line = text[line_start:line_end].strip()
+        if ".search(" in line or ".query(" in line:
+            continue
+        findings.append(
+            Finding(
+                path,
+                line_number(text, match.start()),
+                f"Review Python `{match.group(1)}()` call; table-level materialization is not portable to remote tables.",
+                line,
+            )
+        )
+    return findings
+
+
+def statement_around(text: str, start: int, end: int) -> str:
+    before = max(text.rfind(";", 0, start), text.rfind("\n\n", 0, start))
+    after_candidates = [
+        pos for pos in (text.find(";", end), text.find("\n\n", end)) if pos != -1
+    ]
+    after = min(after_candidates) if after_candidates else len(text)
+    return text[before + 1 : after].strip()
+
+
+def scan_typescript(path: Path, text: str) -> list[Finding]:
+    findings: list[Finding] = []
+    for match in TS_TABLE_TO_ARROW.finditer(text):
+        stmt = statement_around(text, match.start(), match.end())
+        if ".query(" in stmt or ".search(" in stmt:
+            continue
+        findings.append(
+            Finding(
+                path,
+                line_number(text, match.start()),
+                "Review TypeScript `table.toArrow()`-style call; table-level materialization is not portable for large/remote tables.",
+                stmt.splitlines()[0].strip(),
+            )
+        )
+
+    for match in TS_QUERY_COLLECTOR.finditer(text):
+        stmt = statement_around(text, match.start(), match.end())
+        if ".limit(" in stmt:
+            continue
+        findings.append(
+            Finding(
+                path,
+                line_number(text, match.start()),
+                "Review unbounded TypeScript query collection; add `limit()` or stream batches.",
+                stmt.splitlines()[0].strip(),
+            )
+        )
+    return findings
+
+
+def scan_file(path: Path) -> list[Finding]:
+    text = path.read_text(encoding="utf-8", errors="replace")
+    if path.suffix == ".py":
+        return scan_python(path, text)
+    if path.suffix in {".ts", ".tsx"}:
+        return scan_typescript(path, text)
+    return []
+
+
+def main() -> int:
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument("paths", nargs="+", type=Path)
+    parser.add_argument(
+        "--no-fail", action="store_true", help="Always exit 0 after reporting findings."
+    )
+    args = parser.parse_args()
+
+    findings: list[Finding] = []
+    for path in iter_files(args.paths):
+        findings.extend(scan_file(path))
+
+    for finding in findings:
+        print(f"{finding.path}:{finding.line}: {finding.message}")
+        print(f"  {finding.text}")
+
+    if findings:
+        print(
+            f"\n{len(findings)} finding(s). Review manually; bounded query result conversion may be OK."
+        )
+    return 0 if args.no_fail or not findings else 1
+
+
+if __name__ == "__main__":
+    sys.exit(main())
@@ -1,6 +1,6 @@
 [package]
 name = "lancedb-python"
-version = "0.38.0-beta.11"
+version = "0.38.0-beta.3"
 publish = false
 edition.workspace = true
 description = "Python bindings for LanceDB"
@@ -26,9 +26,7 @@ lance-namespace-impls.workspace = true
 lance-io.workspace = true
 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 = ["extension-module", "abi3-py310", "chrono"] }
 chrono.workspace = true
 pyo3-async-runtimes = { version = "0.28", features = [
    "attributes",
@@ -43,7 +41,10 @@ tokio.workspace = true
 libc = "0.2"

 [build-dependencies]
-pyo3-build-config = { version = "0.28", features = ["abi3-py310"] }
+pyo3-build-config = { version = "0.28", features = [
+    "extension-module",
+    "abi3-py310",
+] }

 [features]
 default = ["remote", "lancedb/aws", "lancedb/gcs", "lancedb/azure", "lancedb/dynamodb", "lancedb/oss", "lancedb/huggingface", "lancedb/cos", "lancedb/goosefs", "lancedb/metrics-otel"]
@@ -38,25 +38,6 @@ Stable releases are created about every 2 weeks. For the latest features and bug
 pip install --pre --extra-index-url https://pypi.fury.io/lancedb/ lancedb
 ```

-### Threading in CPU-limited containers
-
-LanceDB uses separate pools for compute work and storage I/O. On a container with
-two visible CPUs, current releases intentionally use one compute worker by default;
-no manual configuration is needed. If every query logs an I/O core reservation
-warning on a two-CPU container, upgrade from LanceDB 0.21.1 or earlier.
-
-The two commonly tuned environment variables control different resources:
-
- `LANCE_CPU_THREADS` overrides the number of compute workers. One worker is the
-  appropriate setting for a two-CPU container when an explicit override is needed.
- `LANCE_IO_THREADS` controls concurrent storage operations, not reserved CPU
-  cores. Its default can be greater than the number of CPUs because I/O workers
-  spend much of their time waiting for storage.
-
-Keep the defaults unless measurements show that the workload benefits from an
-override. See the [Lance threading model](https://lance.org/guide/performance/#threading-model)
-for the current defaults and tuning guidance.
-
 ## Usage

 ### Basic Example
@@ -101,12 +101,9 @@ azure = ["adlfs>=2024.2.0"]
 [tool.maturin]
 python-source = "python"
 module-name = "lancedb._lancedb"
-# uv installs the project as an editable package before `uv run`, so keep that
-# bootstrap build consistent with `maturin develop`.
-editable-profile = "dev"

 [build-system]
-requires = ["maturin>=1.10"]
+requires = ["maturin>=1.4"]
 build-backend = "maturin"

 [tool.ruff.lint]
@@ -6,7 +6,7 @@ import importlib.metadata
 import os
 from concurrent.futures import ThreadPoolExecutor
 from datetime import timedelta
-from typing import Dict, Optional, Union, Any, List, Iterable, TYPE_CHECKING
+from typing import Dict, Optional, Union, Any, List, Iterable

 __version__ = importlib.metadata.version("lancedb")

@@ -20,7 +20,7 @@ from .db import AsyncConnection, DBConnection, LanceDBConnection
 from .remote import ClientConfig
 from .remote.db import RemoteDBConnection
 from .expr import Expr, col, lit, func
-from .schema import blob, vector
+from .schema import blob, vector, BlobType
 from .job import AsyncJob, Job
 from .functions import (
    FunctionArtifactRequest as FunctionArtifactRequest,
@@ -29,15 +29,9 @@ from .functions import (
    FunctionRegistrationRequest as FunctionRegistrationRequest,
    FunctionVersion as FunctionVersion,
    PythonRuntimeSpec as PythonRuntimeSpec,
-    RefreshColumnResult as RefreshColumnResult,
    UdfDefinition as UdfDefinition,
    udf as udf,
 )
-from .materialized_view import (
-    AsyncMaterializedView,
-    MaterializedView,
-    MaterializedViewDefinition,
-)
 from .table import AsyncTable, Table
 from .types import BaseTokenizerType
 from ._lancedb import Session
@@ -49,19 +43,6 @@ from .namespace import (
 )


-if TYPE_CHECKING:
-    from lance.blob import BlobType as BlobType
-
-
-def __getattr__(name: str):
-    if name == "BlobType":
-        from .schema import BlobType
-
-        globals()["BlobType"] = BlobType
-        return BlobType
-    raise AttributeError(f"module {__name__!r} has no attribute {name!r}")
-
-
 def _check_s3_bucket_with_dots(
    uri: str, storage_options: Optional[Dict[str, str]]
 ) -> None:
@@ -192,18 +173,6 @@ def connect(
    ...     },
    ... )

-    For Azure Blob Storage, credentials can be passed directly without setting
-    environment variables:
-
-    >>> azure_storage_options = {
-    ...     "account_name": "some-account",
-    ...     "account_key": "some-key",
-    ... }
-    >>> db = lancedb.connect(  # doctest: +SKIP
-    ...     "az://my-container/my-database",
-    ...     storage_options=azure_storage_options,
-    ... )
-
    For tests and temporary data, use an in-memory database:

    >>> db = lancedb.connect("memory://")
@@ -490,10 +459,6 @@ async def connect_async(
    --------

    >>> import lancedb
-    >>> azure_storage_options = {
-    ...     "account_name": "some-account",
-    ...     "account_key": "some-key",
-    ... }
    >>> async def doctest_example():
    ...     # For a local directory, provide a path to the database
    ...     db = await lancedb.connect_async("~/.lancedb")
@@ -501,11 +466,6 @@ async def connect_async(
    ...     db = await lancedb.connect_async("s3://my-bucket/lancedb",
    ...                                      storage_options={
    ...                                          "aws_access_key_id": "***"})
-    ...     # Azure credentials can also be passed directly
-    ...     db = await lancedb.connect_async(
-    ...         "az://my-container/my-database",
-    ...         storage_options=azure_storage_options,
-    ...     )
    ...     # For tests and temporary data, use an in-memory database
    ...     db = await lancedb.connect_async("memory://")
    ...     # Connect to LanceDB cloud
@@ -546,9 +506,6 @@ async def connect_async(


 __all__ = [
-    "AsyncMaterializedView",
-    "MaterializedView",
-    "MaterializedViewDefinition",
    "connect",
    "connect_async",
    "tokenize",
@@ -12,7 +12,7 @@ from typing import TYPE_CHECKING, Optional, Union
 import pyarrow as pa

 from .expr import Expr
-from .schema import row_addressable_blob_v2_paths
+from .schema import blob_v2_column_paths
 from .types import BlobMode, QueryProjection, QueryProjectionSpec

 if TYPE_CHECKING:
@@ -119,7 +119,7 @@ def blob_v2_projection_sources(
    schema: pa.Schema,
    projection: QueryProjection,
 ) -> dict[str, str]:
-    blob_columns = row_addressable_blob_v2_paths(schema)
+    blob_columns = blob_v2_column_paths(schema)
    if not blob_columns:
        return {}
    columns = set(blob_columns)
@@ -140,9 +140,7 @@ def v2_projection_needs_row_id(
 ) -> bool:
    if with_row_id:
        return False
-    return projection_includes_blob_column(
-        projection, row_addressable_blob_v2_paths(schema)
-    )
+    return projection_includes_blob_column(projection, blob_v2_column_paths(schema))


 def blob_auto_row_id_for_scan(
@@ -272,8 +270,7 @@ def _iter_projection_pairs(
            if isinstance(expr, str):
                yield name, expr
            elif isinstance(expr, Expr):
-                source = expr._column_name()
-                yield name, source if source is not None else expr.to_sql()
+                yield name, expr.to_sql()
        return
    for column in projection:
        if isinstance(column, str):
@@ -283,8 +280,7 @@ def _iter_projection_pairs(
            if isinstance(expr, str):
                yield name, expr
            elif isinstance(expr, Expr):
-                source = expr._column_name()
-                yield name, source if source is not None else expr.to_sql()
+                yield name, expr.to_sql()


 def _set_blob_column(tbl: pa.Table, output_name: str, blobs: pa.Array) -> pa.Table:
@@ -87,7 +87,6 @@ class PyExpr:
    def contains(self, substr: "PyExpr") -> "PyExpr": ...
    def isin(self, values: List["PyExpr"]) -> "PyExpr": ...
    def cast(self, data_type: pa.DataType) -> "PyExpr": ...
-    def column_name(self) -> Optional[str]: ...
    def to_sql(self) -> str: ...

 def expr_col(name: str) -> PyExpr: ...
@@ -148,7 +147,7 @@ class Connection(object):
        limit: Optional[int],
    ) -> list[str]: ...  # Deprecated: Use list_tables instead
    def job(self, job_id: str) -> Job: ...
-    async def create_function_async(self, request_json: str) -> Job: ...
+    async def create_function_async(self, request_json: str) -> FunctionJob: ...
    async def get_function(self, name: str, version: str) -> str: ...
    async def list_jobs(self) -> List[JobInfo]: ...
    async def get_job(self, job_id: str) -> Optional[JobDescription]: ...
@@ -198,15 +197,6 @@ class Connection(object):
        cur_namespace_path: Optional[List[str]] = None,
        new_namespace_path: Optional[List[str]] = None,
    ) -> None: ...
-    async def create_materialized_view(
-        self,
-        name: str,
-        source: str,
-        projections: Optional[List[Tuple[str, str]]] = None,
-        filter: Optional[str] = None,
-        limit: Optional[int] = None,
-    ) -> Table: ...
-    async def list_materialized_views(self) -> List[str]: ...
    async def drop_table(
        self, name: str, namespace_path: Optional[List[str]] = None
    ) -> None: ...
@@ -235,7 +225,14 @@ class Job:
    @property
    def id(self) -> Optional[str]: ...
    async def status(self) -> str: ...
-    async def wait(self) -> Optional[str]: ...
+    async def wait(self) -> None: ...
+    async def cancel(self) -> None: ...
+
+class FunctionJob:
+    @property
+    def id(self) -> Optional[str]: ...
+    async def status(self) -> str: ...
+    async def wait(self) -> str: ...
    async def cancel(self) -> None: ...

 class JobInfo:
@@ -284,7 +281,6 @@ class Table:
        mode: Literal["append", "overwrite"],
        progress: Optional[Any] = None,
        write_parallelism: Optional[int] = None,
-        allow_external_blob_outside_bases: bool = False,
    ) -> AddResult: ...
    async def update(
        self, updates: Dict[str, str], where: Optional[str]
@@ -359,9 +355,6 @@ class Table:
    ) -> AddColumnsResult: ...
    async def refresh_column(self, column: str) -> RefreshColumnResult: ...
    async def refresh_column_async(self, column: str) -> Job: ...
-    async def refresh_materialized_view(
-        self, full: bool = False, source_version: Optional[int] = None
-    ) -> RefreshMaterializedViewResult: ...
    async def add_columns_with_schema(self, schema: pa.Schema) -> AddColumnsResult: ...
    async def alter_columns(
        self, columns: list[dict[str, Any]]
@@ -427,7 +420,7 @@ class Branches:
    async def checkout(self, name: str, version: Optional[int] = None) -> Table: ...
    async def delete(self, name: str) -> None: ...
    async def diff(self, from_branch: str) -> Dict[str, Any]: ...
-    async def cherry_pick(
+    async def merge(
        self, from_branch: str, dry_run: bool = False
    ) -> Dict[str, Any]: ...

@@ -609,7 +602,6 @@ class PyQueryRequest:
    filter: Optional[Union[str, bytes]]
    full_text_search: Optional[FullTextQuery]
    select: Optional[Union[str, List[str]]]
-    select_source_columns: Optional[Dict[str, str]]
    fast_search: Optional[bool]
    with_row_id: Optional[bool]
    use_lsm: Optional[bool]
@@ -712,12 +704,6 @@ class RefreshColumnResult:
    rows_filled: int
    version: int

-class RefreshMaterializedViewResult:
-    mode: str
-    rows_written: int
-    source_version: int
-    version: int
-
 class AlterColumnsResult:
    version: int

@@ -16,7 +16,6 @@ from typing import (
    Iterable,
    List,
    Literal,
-    Mapping,
    Optional,
    Union,
 )
@@ -47,13 +46,7 @@ from lance_namespace.errors import NamespaceNotEmptyError, TableNotFoundError
 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 .materialized_view import (
-    AsyncMaterializedView,
-    MaterializedView,
-    SelectArg,
-    normalize_select,
-)
+from .job import AsyncJob, Job, _function_job
 from .table import (
    AsyncTable,
    LanceTable,
@@ -517,70 +510,6 @@ class DBConnection(EnforceOverrides):
        """
        raise NotImplementedError

-    def create_materialized_view(
-        self,
-        name: str,
-        source: str,
-        *,
-        select: SelectArg = None,
-        where: Optional[str] = None,
-        limit: Optional[int] = None,
-    ) -> MaterializedView:
-        """Define a materialized view named ``name`` over the table ``source``.
-
-        The view is created empty, with the query recorded in its schema
-        metadata; ``view.refresh()`` computes the rows. The view is a normal
-        table: it can be queried, indexed and searched, and it appears in
-        ``table_names``. Local databases only.
-
-        The source table must have stable row ids (create it with the
-        ``new_table_enable_stable_row_ids`` storage option): they keep the
-        view's provenance valid across source compactions, and cannot be
-        enabled after a table exists.
-
-        Parameters
-        ----------
-        name: str
-            The name of the view.
-        source: str
-            The name of the source table, in this database.
-        select: list or dict, optional
-            The view's columns: column names, ``(alias, SQL expression)``
-            pairs, or a dict of the same. Omitting it selects every source
-            column, expanded against the source schema at creation time.
-        where: str, optional
-            SQL predicate; only matching source rows appear in the view.
-        limit: int, optional
-            Cap the view at this many rows, in materialization order.
-
-        Returns
-        -------
-        MaterializedView
-        """
-        raise NotImplementedError(
-            "materialized views are not supported on this connection type"
-        )
-
-    def open_materialized_view(self, name: str) -> MaterializedView:
-        """Open the materialized view named ``name``.
-
-        Raises ``ValueError`` if the table exists but is not a materialized
-        view.
-        """
-        raise NotImplementedError(
-            "materialized views are not supported on this connection type"
-        )
-
-    def list_materialized_views(self) -> List[str]:
-        """The names of the materialized views in this database.
-
-        Found by reading every table's schema, so this costs an open per
-        table.
-        """
-        raise NotImplementedError(
-            "materialized views are not supported on this connection type"
-        )
-
    def drop_table(self, name: str, namespace_path: Optional[List[str]] = None):
        """Drop a table from the database.

@@ -688,35 +617,17 @@ class DBConnection(EnforceOverrides):
        """
        raise NotImplementedError("serialize is not supported for this connection type")

-    def create_function(
-        self,
-        definition: UdfDefinition,
-        *,
-        secrets: Optional[Mapping[str, str]] = None,
-    ) -> FunctionVersion:
+    def create_function(self, definition: UdfDefinition) -> 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``.
        """
-        return self.create_function_async(definition, secrets=secrets).wait()
+        return self.create_function_async(definition).wait()

-    def create_function_async(
-        self,
-        definition: UdfDefinition,
-        *,
-        secrets: Optional[Mapping[str, str]] = None,
-    ) -> Job[FunctionVersion]:
+    def create_function_async(self, definition: UdfDefinition) -> 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``.
@@ -1225,58 +1136,6 @@ class LanceDBConnection(DBConnection):
            tbl.checkout(version)
        return tbl

-    @override
-    def create_materialized_view(
-        self,
-        name: str,
-        source: str,
-        *,
-        select: SelectArg = None,
-        where: Optional[str] = None,
-        limit: Optional[int] = None,
-    ) -> MaterializedView:
-        """Define a materialized view named ``name`` over the table ``source``.
-        See
-        [DBConnection.create_materialized_view][lancedb.DBConnection.create_materialized_view].
-
-        Examples
-        --------
-        >>> import lancedb
-        >>> db = lancedb.connect(
-        ...     "./.lancedb",
-        ...     storage_options={"new_table_enable_stable_row_ids": "true"},
-        ... )
-        >>> data = [{"name": "ada", "age": 36}, {"name": "kid", "age": 7}]
-        >>> table = db.create_table("people", data)
-        >>> view = db.create_materialized_view(
-        ...     "adults",
-        ...     "people",
-        ...     select=["name", ("shout", "upper(name)")],
-        ...     where="age >= 18",
-        ... )
-        >>> result = view.refresh()
-        >>> result.rows_written
-        1
-        """
-        LOOP.run(
-            self._conn.create_materialized_view(
-                name, source, select=select, where=where, limit=limit
-            )
-        )
-        return MaterializedView(self.open_table(name))
-
-    @override
-    def open_materialized_view(self, name: str) -> MaterializedView:
-        """Open the materialized view named ``name``."""
-        view = MaterializedView(self.open_table(name))
-        view.definition
-        return view
-
-    @override
-    def list_materialized_views(self) -> List[str]:
-        """The names of the materialized views in this database."""
-        return LOOP.run(self._conn.list_materialized_views())
-
    def clone_table(
        self,
        target_table_name: str,
@@ -1424,13 +1283,8 @@ class LanceDBConnection(DBConnection):
        return Job(self._conn.job(job_id))

    @override
-    def create_function_async(
-        self,
-        definition: UdfDefinition,
-        *,
-        secrets: Optional[Mapping[str, str]] = None,
-    ) -> Job[FunctionVersion]:
-        job = LOOP.run(self._conn.create_function_async(definition, secrets=secrets))
+    def create_function_async(self, definition: UdfDefinition) -> Job[FunctionVersion]:
+        job = LOOP.run(self._conn.create_function_async(definition))
        return Job(job)

    @override
@@ -2052,50 +1906,6 @@ class AsyncConnection(object):
            await tbl.checkout(version)
        return tbl

-    async def create_materialized_view(
-        self,
-        name: str,
-        source: str,
-        *,
-        select: SelectArg = None,
-        where: Optional[str] = None,
-        limit: Optional[int] = None,
-    ) -> AsyncMaterializedView:
-        """Define a materialized view named ``name`` over the table ``source``.
-        See
-        [DBConnection.create_materialized_view][lancedb.DBConnection.create_materialized_view].
-        """
-        inner = await self._inner.create_materialized_view(
-            name,
-            source,
-            projections=normalize_select(select),
-            filter=where,
-            limit=limit,
-        )
-        return AsyncMaterializedView(AsyncTable(inner))
-
-    async def open_materialized_view(self, name: str) -> AsyncMaterializedView:
-        """Open the materialized view named ``name``.
-
-        Raises ``ValueError`` if the table exists but is not a materialized
-        view.
-        """
-        if self.uri.startswith("db://"):
-            raise NotImplementedError(
-                "materialized views are supported only on local databases"
-            )
-        view = AsyncMaterializedView(await self.open_table(name))
-        await view.definition()
-        return view
-
-    async def list_materialized_views(self) -> List[str]:
-        """The names of the materialized views in this database.
-
-        Found by reading every table's schema, so this costs an open per
-        table.
-        """
-        return await self._inner.list_materialized_views()
-
    async def clone_table(
        self,
        target_table_name: str,
@@ -2249,26 +2059,19 @@ class AsyncConnection(object):
        return AsyncJob(self._inner.job(job_id))

    async def create_function_async(
-        self,
-        definition: UdfDefinition,
-        *,
-        secrets: Optional[Mapping[str, str]] = None,
+        self, definition: UdfDefinition
    ) -> 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``.
        """
        if not isinstance(definition, UdfDefinition):
            raise TypeError("create_function_async requires a @udf definition")
        inner = await self._inner.create_function_async(
-            definition._submission_json(secrets)
+            definition.registration_request.to_canonical_json()
        )
-        return _typed_job(inner, FunctionVersion.from_json)
+        return _function_job(inner)

    async def get_function(self, name: str, *, version: str) -> FunctionVersion:
        """Open one exact immutable Function version from the remote catalog."""
@@ -85,9 +85,8 @@ class Expr:
    # for dict keys / set membership.
    __hash__ = None  # type: ignore[assignment]

-    def __init__(self, inner: PyExpr, *, column_path: str | None = None) -> None:
+    def __init__(self, inner: PyExpr) -> None:
        self._inner = inner
-        self._column_path = column_path

    # ── comparisons ──────────────────────────────────────────────────────────

@@ -249,10 +248,6 @@ class Expr:

    # ── utilities ────────────────────────────────────────────────────────────

-    def _column_name(self) -> str | None:
-        """Return the source name when this is a bare column expression."""
-        return self._inner.column_name()
-
    def to_sql(self) -> str:
        """Render the expression as a SQL string (useful for debugging)."""
        return self._inner.to_sql()
@@ -278,7 +273,7 @@ def col(name: str) -> Expr:
    >>> col("age") > lit(18)
    Expr((age > 18))
    """
-    return Expr(expr_col(name), column_path=name)
+    return Expr(expr_col(name))


 def lit(value: Union[bool, int, float, str, bytes, date, datetime, Decimal]) -> Expr:
@@ -316,7 +311,7 @@ def func(name: str, *args: ExprLike) -> Expr:
    --------
    >>> from lancedb.expr import col, func
    >>> func("lower", col("name"))
-    Expr(lower(`name`))
+    Expr(lower(name))
    """
    inner_args = [_coerce(a)._inner for a in args]
    return Expr(expr_func(name, inner_args))
@@ -1,24 +1,19 @@
 # SPDX-License-Identifier: Apache-2.0
 # SPDX-FileCopyrightText: Copyright The LanceDB Authors

-"""Canonical Function values exchanged with LanceDB Enterprise services.
+"""Canonical values exchanged with LanceDB Enterprise Function services.

 These immutable models contain client/wire state only. Catalog persistence,
 environment bake, secret resolution, and execution are owned by Sophon.
-``RefreshColumnResult`` is also the backend-neutral result of a local
-expression-backed refresh job.
 """

 from __future__ import annotations

 import ast
-import builtins
 import base64
 import functools
 import hashlib
-import importlib
 import inspect
-import symtable
 import json
 import math
 import re
@@ -222,7 +217,6 @@ class PythonEnvironmentSpec(_RemoteValue):

    kind: str
    packages: tuple[str, ...] = ()
-    channels: tuple[str, ...] = ()
    path: Optional[str] = None
    modules: tuple[str, ...] = ()
    image: Optional[str] = None
@@ -271,69 +265,12 @@ class FunctionVersion(_RemoteValue):
    required_secrets: tuple[str, ...] = ()
    created_at: str

-    def __call__(self, **inputs: Any) -> FunctionApplication:
-        """Bind this exact version to named table columns.
-
-        Every input must be a direct [lancedb.col][lancedb.expr.col]
-        reference. The returned application is immutable and retains a
-        named-struct output as one binding, so every row's sibling values
-        come from one logical Function evaluation. Map result fields to table
-        columns with
-        [FunctionApplication.rename][lancedb.functions.FunctionApplication.rename],
-        then pass the application to
-        [Table.add_columns][lancedb.table.Table.add_columns].
-
-        Examples
-        --------
-        >>> from lancedb import col
-        >>> application = function(  # doctest: +SKIP
-        ...     title=col("title"),
-        ...     body=col("body"),
-        ... ).rename(columns={
-        ...     "normalized_text": "search_text",
-        ...     "token_count": "search_token_count",
-        ... })
-        >>> table.add_columns(application)  # doctest: +SKIP
-        """
-        from lancedb.expr import Expr
-
-        parameters = tuple(parameter.name for parameter in self.signature.inputs)
-        missing = [parameter for parameter in parameters if parameter not in inputs]
-        unknown = sorted(set(inputs) - set(parameters))
-        if missing or unknown:
-            details = []
-            if missing:
-                details.append(f"missing inputs: {missing!r}")
-            if unknown:
-                details.append(f"unknown inputs: {unknown!r}")
-            raise TypeError("invalid Function inputs (" + "; ".join(details) + ")")
-
-        bindings = []
-        for parameter in parameters:
-            value = inputs[parameter]
-            if not isinstance(value, Expr) or value._column_path is None:
-                raise TypeError(
-                    f"Function input {parameter!r} must be a direct col(...) reference"
-                )
-            bindings.append(
-                ApplicationInput(
-                    parameter=parameter,
-                    kind="column",
-                    value={"path": value._column_path},
-                )
-            )
-        return FunctionApplication(
-            function=FunctionVersionRef(name=self.name, version=self.version),
-            inputs=tuple(bindings),
-            output=self.signature.output,
-        )
-

 class FunctionRegistrationRequest(_RemoteValue):
    """Stable remote registration envelope produced by :func:`udf`.

-    Only secret names are represented. Secret values are supplied separately
-    when the definition is submitted and are not part of this durable value.
+    Only secret names are represented. Secret values are resolved inside the
+    remote service and have no client request field.
    """

    name: str
@@ -367,18 +304,12 @@ class ApplicationInput(_OpenRemoteValue):


 class FunctionApplication(_OpenRemoteValue):
-    """Immutable pre-declaration application of an exact Function version.
-
-    A named-struct output remains one application through table
-    declaration and execution.
-    [FunctionApplication.rename][lancedb.functions.FunctionApplication.rename]
-    records the result-field to table-column mapping without splitting sibling
-    outputs into separate UDF calls.
-    """
+    """Immutable pre-declaration application of an exact Function version."""

    function: FunctionVersionRef
    inputs: tuple[ApplicationInput, ...]
    output: FunctionOutput
+    group_id: str
    columns: Mapping[str, str] = Field(default_factory=dict)

    def _known_dict(self) -> dict[str, Any]:
@@ -450,10 +381,12 @@ class OutputMapping(_RemoteValue):


 class FunctionBinding(_RemoteValue):
-    """Immutable Function binding persisted by the Enterprise table service."""
+    """Immutable grouped binding persisted by the Enterprise table service."""

    binding_id: str
+    revision: _UInt64
    function: FunctionVersionRef
+    group_id: str
    inputs: tuple[InputBinding, ...]
    outputs: tuple[OutputMapping, ...]
    input_schema: Optional[Mapping[str, Any]] = None
@@ -461,11 +394,7 @@ class FunctionBinding(_RemoteValue):


 class RefreshColumnResult(_RemoteValue):
-    """Terminal result of an expression-backed or Function-backed refresh Job.
-
-    Local jobs produce this value in process. LanceDB Cloud and Enterprise
-    decode the same value from the durable server-job terminal payload.
-    """
+    """Terminal result of a remote Function-column refresh Job."""

    rows_assigned: _UInt64
    rows_failed: _UInt64
@@ -486,80 +415,61 @@ class RefreshColumnResult(_RemoteValue):

 _FUNCTION_NAME = re.compile(r"^[A-Za-z_][A-Za-z0-9_.-]*$")
 _SECRET_NAME = re.compile(r"^[A-Za-z_][A-Za-z0-9_]*$")
-# 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
-
-
-def _validate_secret_value(name: str, value: Any) -> str:
-    """Validate one secret value before building the create request."""
-    if not isinstance(value, str):
-        raise TypeError(f"Function secret {name!r} value must be a string")
-    if not value:
-        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 = (
-    (pa.bool_(), "bool"),
-    (pa.int8(), "int8"),
-    (pa.int16(), "int16"),
-    (pa.int32(), "int32"),
-    (pa.int64(), "int64"),
-    (pa.uint8(), "uint8"),
-    (pa.uint16(), "uint16"),
-    (pa.uint32(), "uint32"),
-    (pa.uint64(), "uint64"),
-    (pa.float16(), "float16"),
-    (pa.float32(), "float32"),
-    (pa.float64(), "float64"),
-    (pa.string(), "utf8"),
-    (pa.binary(), "binary"),
-    (pa.date32(), "date32"),
-    (pa.date64(), "date64"),
-)


 def _canonical_arrow_type(data_type: pa.DataType) -> str:
-    """The server's V1 Function type grammar. Anything outside it is rejected
-    here rather than at registration."""
-    for candidate, name in _GRAMMAR_PRIMITIVES:
+    primitive_types = (
+        (pa.bool_(), "bool"),
+        (pa.int8(), "int8"),
+        (pa.int16(), "int16"),
+        (pa.int32(), "int32"),
+        (pa.int64(), "int64"),
+        (pa.uint8(), "uint8"),
+        (pa.uint16(), "uint16"),
+        (pa.uint32(), "uint32"),
+        (pa.uint64(), "uint64"),
+        (pa.float16(), "float16"),
+        (pa.float32(), "float32"),
+        (pa.float64(), "float64"),
+        (pa.string(), "utf8"),
+        (pa.large_utf8(), "large_utf8"),
+        (pa.binary(), "binary"),
+        (pa.large_binary(), "large_binary"),
+        (pa.date32(), "date32"),
+        (pa.date64(), "date64"),
+    )
+    for candidate, name in primitive_types:
        if data_type == candidate:
            return name
-    if pa.types.is_list(data_type) or pa.types.is_large_list(data_type):
-        prefix = "list" if pa.types.is_list(data_type) else "large_list"
-        return f"{prefix}<{_canonical_list_item(data_type)}>"
-    if pa.types.is_fixed_size_list(data_type) and data_type.list_size > 0:
+    if pa.types.is_fixed_size_binary(data_type):
+        return f"fixed_size_binary[{data_type.byte_width}]"
+    if pa.types.is_list(data_type):
+        return f"list<{_canonical_arrow_type(data_type.value_type)}>"
+    if pa.types.is_large_list(data_type):
+        return f"large_list<{_canonical_arrow_type(data_type.value_type)}>"
+    if pa.types.is_fixed_size_list(data_type):
        return (
-            f"fixed_size_list<{_canonical_list_item(data_type)}, {data_type.list_size}>"
+            f"fixed_size_list<{_canonical_arrow_type(data_type.value_type)}>"
+            f"[{data_type.list_size}]"
        )
+    if pa.types.is_struct(data_type):
+        fields = ",".join(
+            f"{field.name}:{_canonical_arrow_type(field.type)}" for field in data_type
+        )
+        return f"struct<{fields}>"
+    if pa.types.is_timestamp(data_type):
+        timezone = f",tz={data_type.tz}" if data_type.tz is not None else ""
+        return f"timestamp[{data_type.unit}{timezone}]"
+    if pa.types.is_time32(data_type) or pa.types.is_time64(data_type):
+        return f"time[{data_type.unit}]"
+    if pa.types.is_duration(data_type):
+        return f"duration[{data_type.unit}]"
+    if pa.types.is_decimal(data_type):
+        bit_width = data_type.bit_width
+        return f"decimal{bit_width}({data_type.precision},{data_type.scale})"
    raise TypeError(f"unsupported Arrow type for Function signature: {data_type}")


-def _canonical_list_item(data_type: pa.DataType) -> 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 = data_type.value_field
-    if child.name != "item" or child.nullable or child.metadata:
-        raise TypeError(
-            "unsupported Arrow type for Function signature: list items must be a "
-            f"non-nullable field named 'item', got {child}"
-        )
-    return _canonical_arrow_type(child.type)
-
-
-def _list_of(item: pa.DataType) -> pa.DataType:
-    return pa.list_(pa.field("item", item, nullable=False))
-
-
 def _annotation_type(annotation: Any) -> tuple[pa.DataType, bool]:
    nullable = False
    origin = get_origin(annotation)
@@ -607,7 +517,7 @@ def _annotation_type(annotation: Any) -> tuple[pa.DataType, bool]:
        value_type, value_nullable = _annotation_type(arguments[0])
        if value_nullable:
            raise TypeError("nullable Function list elements are not supported")
-        return _list_of(value_type), nullable
+        return pa.list_(value_type), nullable
    raise TypeError(f"unsupported Function annotation: {annotation!r}")


@@ -754,104 +664,6 @@ def _literal_source(value: Any) -> str:
    )


-_DYNAMIC_NAMESPACE_ACCESS = frozenset(
-    {"globals", "locals", "vars", "eval", "exec", "compile", "__import__"}
-)
-# Modules that hand out namespaces (`sys.modules`, `builtins`, importers,
-# introspection). The artifact's module namespace holds only the names it was
-# packaged with, so reaching around it cannot be represented.
-_NAMESPACE_MODULES = frozenset(
-    {"sys", "builtins", "importlib", "inspect", "gc", "ctypes", "types"}
-)
-
-
-def _namespace_acquisition(
-    definition: ast.FunctionDef, references: set[str]
-) -> list[str]:
-    found = set(references & _DYNAMIC_NAMESPACE_ACCESS)
-    for node in ast.walk(definition):
-        if isinstance(node, ast.Import):
-            found.update(
-                alias.name
-                for alias in node.names
-                if alias.name.split(".")[0] in _NAMESPACE_MODULES
-            )
-        elif isinstance(node, ast.ImportFrom) and node.module:
-            if node.module.split(".")[0] in _NAMESPACE_MODULES:
-                found.add(node.module)
-    return sorted(found)
-
-
-def _module_references(module_source: str) -> set[str]:
-    """Names any scope in `module_source` binds or loads at module scope.
-    Python's own scope analysis on the exact text that ships: free variables
-    belong to an enclosing scope inside the function, and postponed
-    annotations are not runtime loads."""
-
-    def visit(table: symtable.SymbolTable, found: set[str]) -> None:
-        for symbol in table.get_symbols():
-            if symbol.is_global() and (
-                symbol.is_referenced() or symbol.is_declared_global()
-            ):
-                found.add(symbol.get_name())
-        for child in table.get_children():
-            visit(child, found)
-
-    found: set[str] = set()
-    for table in symtable.symtable(module_source, "<udf>", "exec").get_children():
-        visit(table, found)
-    return found
-
-
-def _global_source(name: str, value: Any) -> str:
-    """One module-level line that rebinds `name` to `value` in the artifact:
-    an import for modules and importable classes/functions, a literal otherwise."""
-    if isinstance(value, types.ModuleType):
-        if value.__name__.split(".")[0] in _NAMESPACE_MODULES:
-            raise ValueError(
-                f"@udf cannot package dynamic namespace access: {value.__name__!r}"
-            )
-        try:
-            imported = importlib.import_module(value.__name__)
-        except ImportError:
-            imported = None
-        if imported is not value:
-            raise TypeError(
-                f"Function source references module {name!r} that does not import "
-                f"as {value.__name__!r}"
-            )
-        return f"import {value.__name__} as {name}"
-    module_name = getattr(value, "__module__", None)
-    qualname = getattr(value, "__qualname__", None)
-    if (
-        isinstance(module_name, str)
-        and isinstance(qualname, str)
-        and module_name != "__main__"
-        and "." not in qualname
-        and "<" not in qualname
-    ):
-        try:
-            imported = getattr(importlib.import_module(module_name), qualname)
-        except (ImportError, AttributeError):
-            imported = None
-        if imported is value:
-            return f"from {module_name} import {qualname} as {name}"
-    return f"{name} = {_literal_source(value)}"
-
-
-def _is_recursive_reference(function: Callable[..., Any], name: str) -> bool:
-    """`name` inside the body means the function itself unless the module has
-    since bound it to something else."""
-    if name != function.__name__:
-        return False
-    bound = function.__globals__.get(name, function)
-    if bound is function:
-        return True
-    # The decorator's own result is the one wrapper known to call `function`
-    # unchanged; any other binding may behave differently from a self-call.
-    return type(bound) is UdfDefinition and bound._function is function
-
-
 def _package_source(function: Callable[..., Any]) -> bytes:
    if not inspect.isfunction(function) or inspect.iscoroutinefunction(function):
        raise TypeError("@udf requires a synchronous Python function")
@@ -876,46 +688,23 @@ def _package_source(function: Callable[..., Any]) -> bytes:
    closure = inspect.getclosurevars(function)
    if closure.nonlocals:
        raise ValueError("@udf cannot package functions that capture closure values")
-    function_source = ast.unparse(definition)
-    module_header = "from __future__ import annotations"
-    references = _module_references(f"{module_header}\n\n{function_source}\n")
-    dynamic = _namespace_acquisition(definition, references)
-    if dynamic:
-        raise ValueError(f"@udf cannot package dynamic namespace access: {dynamic!r}")
-    # Resolve every module-scope reference the way the interpreter would: the
-    # function's own globals first (a module global may shadow a builtin, and
-    # nested scopes are not visible to getclosurevars), then its builtins.
-    # The artifact runs under the standard builtins; only the exact mapping is
-    # provably equivalent (a subclass or copy can change lookups and hooks).
-    if function.__builtins__ is not vars(builtins):
-        raise ValueError("@udf cannot package a non-standard builtins environment")
-    globals_source = []
-    unresolved = []
-    for name in sorted(references):
-        if name == function.__name__:
-            if not _is_recursive_reference(function, name):
-                raise ValueError(
-                    f"@udf cannot package {name!r}: the module binds that name to "
-                    "another value, which the artifact's own definition would shadow"
-                )
-            continue
-        if name in function.__globals__:
-            globals_source.append(_global_source(name, function.__globals__[name]))
-        elif hasattr(builtins, name):
-            pass
-        else:
-            unresolved.append(name)
-    if unresolved:
+    if closure.unbound:
        raise ValueError(
-            f"@udf source contains unresolved global names: {unresolved!r}"
+            f"@udf source contains unresolved global names: {sorted(closure.unbound)!r}"
        )
+    globals_source = []
+    for name, value in sorted(closure.globals.items()):
+        if isinstance(value, types.ModuleType):
+            globals_source.append(f"import {value.__name__} as {name}")
+        else:
+            globals_source.append(f"{name} = {_literal_source(value)}")

-    parts = [module_header]
+    function_source = ast.unparse(definition)
+    parts = ["from __future__ import annotations"]
    if globals_source:
        parts.extend(["", *globals_source])
    parts.extend(["", function_source, ""])
-    packaged = "\n".join(parts)
-    return packaged.encode("utf-8")
+    return "\n".join(parts).encode("utf-8")


 class UdfDefinition:
@@ -938,25 +727,13 @@ class UdfDefinition:
        env: Mapping[str, str],
        secrets: tuple[str, ...],
        python_version: Optional[str],
-        conda: tuple[str, ...] = (),
-        conda_channels: tuple[str, ...] = (),
    ):
        function_name = name or function.__name__
        if not _FUNCTION_NAME.fullmatch(function_name):
            raise ValueError(f"invalid Function name: {function_name!r}")
-        if pip and conda:
-            raise ValueError("a Function environment is pip or conda, not both")
-        if conda_channels and not conda:
-            raise ValueError("conda_channels requires conda packages")
-        packages = tuple(sorted(set(conda if conda else pip)))
+        packages = tuple(sorted(set(pip)))
        if any(not package or package != package.strip() for package in packages):
-            raise ValueError("package requirements must be non-empty and trimmed")
-        if conda:
-            environment_spec = PythonEnvironmentSpec(
-                kind="conda", packages=packages, channels=tuple(conda_channels)
-            )
-        else:
-            environment_spec = PythonEnvironmentSpec(kind="pip", packages=packages)
+            raise ValueError("pip requirements must be non-empty and trimmed")
        environment = dict(env)
        if any(
            not isinstance(key, str) or not isinstance(value, str)
@@ -974,6 +751,7 @@ class UdfDefinition:
            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()}"
@@ -981,7 +759,7 @@ class UdfDefinition:
            kind="python",
            python_version=python_version
            or f"{sys.version_info.major}.{sys.version_info.minor}",
-            environment=environment_spec,
+            environment=PythonEnvironmentSpec(kind="pip", packages=packages),
            env=environment,
        )
        self._function = function
@@ -1008,59 +786,9 @@ class UdfDefinition:

    @property
    def registration_request(self) -> FunctionRegistrationRequest:
-        """The immutable, value-free client model for a Function submission."""
+        """The immutable request sent by ``create_function_async``."""
        return self._request

-    def _submission_json(self, secrets: Optional[Mapping[str, str]]) -> str:
-        """Build one registration submission without retaining values on self."""
-        if secrets is None:
-            secret_values: Mapping[str, str] = {}
-        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):
-            raise TypeError("Function secret names must be strings")
-        expected = set(self._request.required_secrets)
-        provided = set(secret_values)
-        if provided != expected:
-            missing = sorted(expected - provided)
-            unexpected = sorted(provided - expected)
-            details = []
-            if missing:
-                details.append(f"missing: {missing!r}")
-            if unexpected:
-                details.append(f"unexpected: {unexpected!r}")
-            raise ValueError(
-                "Function secret values must exactly match the declared secrets ("
-                + "; ".join(details)
-                + ")"
-            )
-
-        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=(",", ":"),
-        )
-
    def __call__(self, *args, **kwargs):
        return self._function(*args, **kwargs)

@@ -1080,8 +808,6 @@ def udf(
    env: Optional[Mapping[str, str]] = None,
    secrets: tuple[str, ...] | list[str] = (),
    python_version: Optional[str] = None,
-    conda: tuple[str, ...] | list[str] = (),
-    conda_channels: tuple[str, ...] | list[str] = (),
 ) -> Callable[[Callable[..., Any]], UdfDefinition]: ...


@@ -1095,8 +821,6 @@ def udf(
    env: Optional[Mapping[str, str]] = None,
    secrets: tuple[str, ...] | list[str] = (),
    python_version: Optional[str] = None,
-    conda: tuple[str, ...] | list[str] = (),
-    conda_channels: tuple[str, ...] | list[str] = (),
 ):
    """Prepare a scalar Python callable for remote Function registration.

@@ -1119,25 +843,14 @@ def udf(
        provided together with ``input_schema``.
    pip : sequence of str, optional
        Pip requirements for the remote environment.
-    conda : sequence of str, optional
-        Conda packages for the remote environment, instead of ``pip``.
-    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.
    secrets : sequence of str, optional
-        Names of secrets required by the callable. Supply their values separately
-        to ``create_function`` or ``create_function_async``.
+        Names of secrets resolved by the remote service. Secret values are not
+        accepted by this API or included in the registration request.
    python_version : str, optional
        Remote Python major/minor version. Defaults to the client version.

-    The packaged artifact is a snapshot: the function source plus exactly
-    the module-level names it references (modules as imports, importable
-    classes and functions as imports, literals inline). Code that reaches the
-    module namespace another way -- ``globals()``/``eval``, ``sys.modules``,
-    ``builtins`` -- is rejected where it can be seen and otherwise
-    unsupported; closures and a non-standard ``__builtins__`` are rejected.
-
    Returns
    -------
    UdfDefinition
@@ -1154,10 +867,6 @@ def udf(
    ...     return value * 2
    >>> score(1.5)
    3.0
-    >>> db.create_function(  # doctest: +SKIP
-    ...     score, secrets={"MODEL_TOKEN": "user-secret-value"}
-    ... )
-
    """

    def decorate(target: Callable[..., Any]) -> UdfDefinition:
@@ -1170,8 +879,6 @@ def udf(
            env={} if env is None else env,
            secrets=tuple(secrets),
            python_version=python_version,
-            conda=tuple(conda),
-            conda_channels=tuple(conda_channels),
        )

    if function is None:
@@ -7,7 +7,6 @@ from typing import List, Literal, Optional
 from ._lancedb import (
    IndexConfig,
 )
-from .query import DocumentGranularity
 from .types import BaseTokenizerType

 lang_mapping = {
@@ -122,11 +121,6 @@ class FTS:

    >>> config = FTS(block_size=256)

-    Create an index that treats each deepest-list element as one document:
-
-    >>> from lancedb.query import DocumentGranularity
-    >>> config = FTS(document_granularity=DocumentGranularity.LIST_ELEMENT)
-
    Attributes
    ----------
    with_position : bool, default False
@@ -169,20 +163,6 @@ class FTS:
        The number of documents per compressed posting block. Supported values
        are 128 and 256. A value of 256 uses the experimental FTS V3 format
        and may introduce breaking changes.
-    memory_limit : int, optional
-        The total memory limit in MiB for the local FTS build stage. The limit
-        is divided evenly among indexing workers. This build-only setting is
-        not persisted with the index and does not apply to remote tables.
-    num_workers : int, optional
-        The number of workers for a local FTS build. By default Lance uses
-        roughly half of the available CPU cores. The effective value is
-        limited by the available compute capacity. This build-only setting is
-        not persisted with the index and does not apply to remote tables.
-    document_granularity : DocumentGranularity, default ROW
-        ``ROW`` treats the selected text in one table row as one document.
-        ``LIST_ELEMENT`` treats each element of the deepest list on the indexed
-        field path as one document and returns its physical coordinates in
-        ``_doc_index`` for matching queries.

    Notes
    -----
@@ -205,9 +185,6 @@ class FTS:
    prefix_only: bool = False
    block_size: int = 128
    custom_stop_words: Optional[List[str]] = None
-    memory_limit: Optional[int] = None
-    num_workers: Optional[int] = None
-    document_granularity: DocumentGranularity = DocumentGranularity.ROW


@dataclass
@@ -5,11 +5,12 @@

 import asyncio
 from datetime import timedelta
-from typing import Any, Callable, Generic, Optional, TypeVar, cast
+from typing import Any, Generic, Optional, TypeVar, cast

 from lancedb.background_loop import LOOP

 from . import _lancedb
+from .functions import FunctionVersion

 T = TypeVar("T")

@@ -17,18 +18,11 @@ T = TypeVar("T")
 class AsyncJob(Generic[T]):
    """A handle to an operation that may still be running.

-    The operation may already be complete when the handle is created. ``T``
-    is the endpoint's terminal result type; unit-result jobs resolve to
-    ``None``.
+    The operation may already be complete when the handle is created.
    """

-    def __init__(
-        self,
-        inner: Optional[Any],
-        result_decoder: Optional[Callable[[Any], T]] = None,
-    ):
+    def __init__(self, inner: Optional[Any]):
        self._inner = inner
-        self._result_decoder = result_decoder

    @property
    def id(self) -> Optional[str]:
@@ -56,21 +50,17 @@ class AsyncJob(Generic[T]):
    async def wait(self, timeout: Optional[timedelta] = None) -> T:
        """Wait until the operation reaches a terminal state.

-        Returns the endpoint's typed result, or ``None`` for a unit-result
-        job.
-
        Raises `JobFailedError` if the operation failed, `JobCancelledError`
        if it was cancelled, and `TimeoutError` if `timeout` elapses first.
        """
        if self._inner is None:
            return cast(T, None)
        if timeout is None:
-            result = await self._inner.wait()
-        else:
-            result = await asyncio.wait_for(self._inner.wait(), timeout.total_seconds())
-        if self._result_decoder is not None:
-            return self._result_decoder(result)
-        return cast(T, result)
+            return cast(T, await self._inner.wait())
+        return cast(
+            T,
+            await asyncio.wait_for(self._inner.wait(), timeout.total_seconds()),
+        )

    async def cancel(self):
        """Request cancellation. Cancelling a finished operation is a no-op."""
@@ -80,7 +70,7 @@ class AsyncJob(Generic[T]):


 class Job(Generic[T]):
-    """Synchronous counterpart of `AsyncJob` with the same result type."""
+    """Synchronous counterpart of `AsyncJob`."""

    def __init__(self, inner: Optional[AsyncJob[T]]):
        self._inner = inner
@@ -106,9 +96,6 @@ class Job(Generic[T]):
    def wait(self, timeout: Optional[timedelta] = None) -> T:
        """Block until the operation reaches a terminal state.

-        Returns the endpoint's typed result, or ``None`` for a unit-result
-        job.
-
        Raises `JobFailedError` if the operation failed, `JobCancelledError`
        if it was cancelled, and `TimeoutError` if `timeout` elapses first.
        """
@@ -123,8 +110,23 @@ class Job(Generic[T]):
        LOOP.run(self._inner.cancel())


-def _typed_job(
-    inner: "_lancedb.Job", result_decoder: Callable[[str], T]
-) -> AsyncJob[T]:
-    """Bind an internal JSON-producing job to its public result model."""
-    return AsyncJob(inner, result_decoder)
+class _FunctionJobAdapter:
+    def __init__(self, inner: "_lancedb.FunctionJob"):
+        self._inner = inner
+
+    @property
+    def id(self) -> Optional[str]:
+        return self._inner.id
+
+    async def status(self) -> str:
+        return await self._inner.status()
+
+    async def wait(self) -> FunctionVersion:
+        return FunctionVersion.from_json(await self._inner.wait())
+
+    async def cancel(self):
+        await self._inner.cancel()
+
+
+def _function_job(inner: "_lancedb.FunctionJob") -> AsyncJob[FunctionVersion]:
+    return AsyncJob(_FunctionJobAdapter(inner))
@@ -1,178 +0,0 @@
-# SPDX-License-Identifier: Apache-2.0
-# SPDX-FileCopyrightText: Copyright The LanceDB Authors
-
-"""Materialized views: tables defined by a query over a source table and
-maintained by refresh. See ``DBConnection.create_materialized_view``."""
-
-from __future__ import annotations
-
-import json
-from dataclasses import dataclass, field
-from typing import TYPE_CHECKING, Dict, List, Optional, Sequence, Tuple, Union
-
-from .background_loop import LOOP
-
-if TYPE_CHECKING:
-    import pyarrow as pa
-
-    from ._lancedb import RefreshMaterializedViewResult
-    from .table import AsyncTable, LanceTable
-
-DEFINITION_META_KEY = b"mv.definition"
-
-SelectArg = Union[
-    str,
-    Sequence[Union[str, Tuple[str, str]]],
-    Dict[str, str],
-    None,
-]
-
-
-@dataclass
-class MaterializedViewDefinition:
-    """The query that defines a materialized view."""
-
-    source_table: str
-    """Name of the source table, in the same database as the view."""
-    projections: List[Tuple[str, str]]
-    """``(output column, SQL expression)`` pairs, in view schema order."""
-    filter: Optional[str] = None
-    """SQL predicate selecting the source rows the view holds."""
-    limit: Optional[int] = None
-    """Cap on the number of rows the view holds."""
-    inputs: List[str] = field(default_factory=list)
-    """Source columns the projections and filter read."""
-
-
-def _definition_from_schema(
-    schema: "pa.Schema", name: str
-) -> MaterializedViewDefinition:
-    metadata = schema.metadata or {}
-    raw = metadata.get(DEFINITION_META_KEY)
-    if raw is None:
-        raise ValueError(f"Table '{name}' is not a materialized view")
-    value = json.loads(raw)
-    kind = value.get("kind")
-    if kind != "select":
-        raise NotImplementedError(
-            f"materialized view '{name}' is defined by '{kind}', which this "
-            "version of lancedb cannot refresh"
-        )
-    return MaterializedViewDefinition(
-        source_table=value["source_table"],
-        projections=[
-            (p["output"], p["expression"]) for p in value.get("projections", [])
-        ],
-        filter=value.get("filter"),
-        limit=value.get("limit"),
-        inputs=value.get("inputs", []),
-    )
-
-
-def _quote_identifier(name: str) -> str:
-    """Quote a column name as a Lance SQL identifier (backticks)."""
-    escaped = name.replace("`", "``")
-    return f"`{escaped}`"
-
-
-def normalize_select(select: SelectArg) -> Optional[List[Tuple[str, str]]]:
-    """``select`` items may be a column name, an ``(alias, expression)`` pair,
-    or a dict of the same. A bare name projects itself and is quoted, so any
-    valid column name works; dict and pair entries are kept verbatim because
-    their right side is an expression.
-
-    A lone string is one column, not a sequence of its characters."""
-    if select is None:
-        return None
-    if isinstance(select, str):
-        select = [select]
-    if isinstance(select, dict):
-        return list(select.items())
-    normalized = []
-    for item in select:
-        if isinstance(item, str):
-            normalized.append((item, _quote_identifier(item)))
-        else:
-            alias, expression = item
-            normalized.append((alias, expression))
-    return normalized
-
-
-class AsyncMaterializedView:
-    """A handle on a materialized view: its table plus its definition.
-
-    Obtained from ``AsyncConnection.create_materialized_view`` or
-    ``AsyncConnection.open_materialized_view``.
-    """
-
-    def __init__(self, table: "AsyncTable"):
-        self._table = table
-
-    def __repr__(self) -> str:
-        return f"AsyncMaterializedView(name={self.name!r})"
-
-    @property
-    def name(self) -> str:
-        return self._table.name
-
-    @property
-    def table(self) -> "AsyncTable":
-        """The view, as the table it is. Queries, indexes and search all
-        apply; writes are not blocked, but a rebuild replaces them."""
-        return self._table
-
-    async def definition(self) -> MaterializedViewDefinition:
-        """The query that defines the view, read from its stored schema."""
-        return _definition_from_schema(await self._table.schema(), self.name)
-
-    async def refresh(
-        self, *, full: bool = False, source_version: Optional[int] = None
-    ) -> "RefreshMaterializedViewResult":
-        """Recompute the view from its source.
-
-        The refresh is incremental when the source's changes can be
-        reconciled into the view -- rows added, changed or removed since the
-        last one -- and otherwise rebuilds. ``full=True`` forces a rebuild;
-        ``source_version`` refreshes to that source version instead of the
-        latest.
-
-        Concurrent refreshes of one view do not duplicate its rows. Two that
-        plan the same source rows conflict on commit, and the loser raises
-        rather than writing them a second time.
-        """
-        return await self._table._inner.refresh_materialized_view(
-            full=full, source_version=source_version
-        )
-
-
-class MaterializedView:
-    """Synchronous variant of
-    [AsyncMaterializedView][lancedb.materialized_view.AsyncMaterializedView]."""
-
-    def __init__(self, table: "LanceTable"):
-        self._table = table
-        self._async = AsyncMaterializedView(table._table)
-
-    def __repr__(self) -> str:
-        return f"MaterializedView(name={self.name!r})"
-
-    @property
-    def name(self) -> str:
-        return self._table.name
-
-    @property
-    def table(self) -> "LanceTable":
-        """The view, as the table it is."""
-        return self._table
-
-    @property
-    def definition(self) -> MaterializedViewDefinition:
-        """The query that defines the view, read from its stored schema."""
-        return _definition_from_schema(self._table.schema, self.name)
-
-    def refresh(
-        self, *, full: bool = False, source_version: Optional[int] = None
-    ) -> "RefreshMaterializedViewResult":
-        """Recompute the view from its source. See
-        [AsyncMaterializedView.refresh][lancedb.materialized_view.AsyncMaterializedView.refresh]."""
-        return LOOP.run(self._async.refresh(full=full, source_version=source_version))
@@ -61,11 +61,6 @@ from lance_namespace import (
    NamespaceExistsRequest,
    TableExistsRequest,
 )
-from lancedb.materialized_view import (
-    AsyncMaterializedView,
-    MaterializedView,
-    SelectArg,
-)
 from lancedb.table import AsyncTable, LanceTable, Table
 from lancedb.util import validate_table_name
 from lancedb.common import DATA
@@ -624,42 +619,6 @@ class LanceNamespaceDBConnection(DBConnection):
            tbl.checkout(version)
        return tbl

-    @override
-    def create_materialized_view(
-        self,
-        name: str,
-        source: str,
-        *,
-        select: "SelectArg" = None,
-        where: Optional[str] = None,
-        limit: Optional[int] = None,
-    ) -> "MaterializedView":
-        """Define a materialized view over a table in the root namespace.
-        See
-        [DBConnection.create_materialized_view][lancedb.DBConnection.create_materialized_view].
-        """
-        return MaterializedView(
-            self.open_table(
-                LOOP.run(
-                    self._inner.create_materialized_view(
-                        name, source, select=select, where=where, limit=limit
-                    )
-                ).name
-            )
-        )
-
-    @override
-    def open_materialized_view(self, name: str) -> "MaterializedView":
-        """Open the materialized view named ``name``."""
-        view = MaterializedView(self.open_table(name))
-        view.definition
-        return view
-
-    @override
-    def list_materialized_views(self) -> List[str]:
-        """The names of the materialized views in the root namespace."""
-        return LOOP.run(self._inner.list_materialized_views())
-
    @override
    def drop_table(self, name: str, namespace_path: Optional[List[str]] = None):
        if namespace_path is None:
@@ -1182,33 +1141,6 @@ class AsyncLanceNamespaceDBConnection:
            route_pushdown_to_rust=self._route_pushdown_to_rust,
        )

-    async def create_materialized_view(
-        self,
-        name: str,
-        source: str,
-        *,
-        select: "SelectArg" = None,
-        where: Optional[str] = None,
-        limit: Optional[int] = None,
-    ) -> "AsyncMaterializedView":
-        """Define a materialized view over a table in the root namespace."""
-        view = await self._inner.create_materialized_view(
-            name, source, select=select, where=where, limit=limit
-        )
-        # Reopen through the namespace so the view's table carries the
-        # namespace client and pushdown configuration a bare inner table lacks.
-        return AsyncMaterializedView(await self.open_table(view.name))
-
-    async def open_materialized_view(self, name: str) -> "AsyncMaterializedView":
-        """Open the materialized view named ``name``."""
-        view = AsyncMaterializedView(await self.open_table(name))
-        await view.definition()
-        return view
-
-    async def list_materialized_views(self) -> List[str]:
-        """The names of the materialized views in the root namespace."""
-        return await self._inner.list_materialized_views()
-
    async def drop_table(self, name: str, namespace_path: Optional[List[str]] = None):
        """Drop a table from the namespace."""
        if namespace_path is None:
@@ -41,15 +41,21 @@ class PermutationBuilder:
    The permutation is stored in memory and will be lost when the program exits.
    """

-    def __init__(self, table: Table):
+    def __init__(self, table: LanceTable):
        """
        Creates a new permutation builder for the given table.

        By default, the permutation builder will create a single split that contains all
        rows in the same order as the base table.
-
-        Tables with an LSM write spec are rejected: unflushed rows have no row id.
        """
+        if not hasattr(table, "_inner"):
+            raise TypeError(
+                f"PermutationBuilder requires a local LanceTable, "
+                f"got {type(table).__name__}. "
+                "The permutation API is not supported on remote tables. "
+                "Remote tables connect to LanceDB Cloud or Enterprise and do not have "
+                "direct access to the underlying Lance dataset needed for permutations."
+            )
        self._async = async_permutation_builder(table)

    def split_random(
@@ -225,7 +231,7 @@ class PermutationBuilder:
        return LOOP.run(do_execute())


-def permutation_builder(table: Table) -> PermutationBuilder:
+def permutation_builder(table: LanceTable) -> PermutationBuilder:
    return PermutationBuilder(table)


@@ -242,7 +248,7 @@ class Permutations:

    Attributes
    ----------
-    base_table: Table
+    base_table: LanceTable
        The base table that the permutations are based on.
    permutation_table: LanceTable
        The permutation table that defines the splits.
@@ -276,7 +282,7 @@ class Permutations:
    {'train': 0, 'test': 1}
    """

-    def __init__(self, base_table: Table, permutation_table: LanceTable):
+    def __init__(self, base_table: LanceTable, permutation_table: LanceTable):
        self.base_table = base_table
        self.permutation_table = permutation_table

@@ -391,15 +397,6 @@ def _table_to_pickle_state(table: Table) -> dict[str, Any]:
    }


-def _drop_base_version(permutation_data: pa.Table) -> pa.Table:
-    """Strip the recorded base version so the reader leaves the base table unpinned."""
-    metadata = dict(permutation_data.schema.metadata or {})
-    if metadata.pop(b"base_version", None) is None:
-        return permutation_data
-    metadata.pop(b"base_branch", None)
-    return permutation_data.replace_schema_metadata(metadata)
-
-
 def _table_from_pickle_state(state: dict[str, Any]) -> Table:
    from . import connect

@@ -688,15 +685,11 @@ class Permutation:
        from . import connect

        connection_factory = state["connection_factory"]
-        rebuilt_base = False
        if connection_factory is not None:
            base_table = connection_factory(state["base_table_name"])
        elif "base_table_state" in state:
-            base_state = state["base_table_state"]
-            rebuilt_base = base_state["kind"] == "memory"
-            base_table = _table_from_pickle_state(base_state)
+            base_table = _table_from_pickle_state(state["base_table_state"])
        elif "base_table_data" in state:
-            rebuilt_base = True
            # In-memory base table inlined into the pickle; rebuild the same
            # way we rebuild the in-memory permutation table.
            mem_db = connect("memory://")
@@ -714,14 +707,11 @@ class Permutation:
            )

        permutation_table: Optional[Table] = None
-        permutation_data = state["permutation_data"]
-        if permutation_data is not None:
-            if rebuilt_base:
-                # The base table was materialized from Arrow, so it is a fresh
-                # single-version dataset and the recorded pin cannot resolve on it.
-                permutation_data = _drop_base_version(permutation_data)
+        if state["permutation_data"] is not None:
            mem_db = connect("memory://")
-            permutation_table = mem_db.create_table("permutation", permutation_data)
+            permutation_table = mem_db.create_table(
+                "permutation", state["permutation_data"]
+            )

        self.base_table = base_table
        self.permutation_table = permutation_table
@@ -167,12 +167,6 @@ def _projection_to_scanner_kwargs(columns: QueryProjection) -> Dict[str, Any]:
    return {"columns": projection}


-def _query_request_projection(req: "PyQueryRequest") -> QueryProjection:
-    if req.select_source_columns is not None:
-        return req.select_source_columns
-    return req.select
-
-
 def _scanner_kwargs_for_query(
    query: Query,
    blob_mode: BlobMode,
@@ -381,13 +375,6 @@ class FullTextOperator(str, Enum):
    OR = "OR"


-class DocumentGranularity(str, Enum):
-    """The unit treated as one full-text-search document."""
-
-    ROW = "row"
-    LIST_ELEMENT = "list_element"
-
-
 class Occur(str, Enum):
    SHOULD = "SHOULD"
    MUST = "MUST"
@@ -491,10 +478,6 @@ class MatchQuery(FullTextQuery):
    prefix_length : int, optional
        The number of beginning characters being unchanged for fuzzy matching.
        This is useful to achieve prefix matching.
-    document_granularity : DocumentGranularity, optional
-        Explicitly select row or deepest-list-element documents. If omitted,
-        the indexed granularity is inferred. When both granularities are indexed
-        for the field, this must be specified. With no index, row granularity is used.
    """

    query: str
@@ -504,9 +487,6 @@ class MatchQuery(FullTextQuery):
    max_expansions: int = pydantic.Field(50, kw_only=True)
    operator: FullTextOperator = pydantic.Field(FullTextOperator.OR, kw_only=True)
    prefix_length: int = pydantic.Field(0, kw_only=True)
-    document_granularity: Optional[DocumentGranularity] = pydantic.Field(
-        None, kw_only=True
-    )

    def query_type(self) -> FullTextQueryType:
        return FullTextQueryType.MATCH
@@ -523,20 +503,11 @@ class PhraseQuery(FullTextQuery):
        The query string to match against.
    column : str
        The name of the column to match against.
-    slop : int, default 0
-        The maximum number of intervening positions permitted in the phrase.
-    document_granularity : DocumentGranularity, optional
-        Explicitly select row or deepest-list-element documents. If omitted,
-        the indexed granularity is inferred. When both granularities are indexed
-        for the field, this must be specified. With no index, row granularity is used.
    """

    query: str
    column: str
    slop: int = pydantic.Field(0, kw_only=True)
-    document_granularity: Optional[DocumentGranularity] = pydantic.Field(
-        None, kw_only=True
-    )

    def query_type(self) -> FullTextQueryType:
        return FullTextQueryType.MATCH_PHRASE
@@ -2805,16 +2776,15 @@ class AsyncQueryBase(object):

        req = self._inner.to_query_request()
        schema = await self._table.schema()
-        projection = _query_request_projection(req)
        self._blob_auto_row_id = blob_auto_row_id_for_scan(
            schema,
-            projection,
+            req.select,
            with_row_id=self._with_row_id,
        )
        if not self._blob_auto_row_id:
            self._blob_paths = ()
            return
-        self._blob_paths = tuple(blob_v2_projection_sources(schema, projection).keys())
+        self._blob_paths = tuple(blob_v2_projection_sources(schema, req.select).keys())
        self._inner.with_row_id()

    def select(self, columns: Union[List[str], dict[str, str]]) -> Self:
@@ -3408,10 +3378,9 @@ class AsyncQuery(AsyncStandardQuery):
        pass in multiple vectors. When multiple vectors are passed in, if the vector
        column is with multivector type, then the vectors will be treated as a single
        query. Or the vectors will be treated as multiple queries, this can be useful
-        if you want to find the nearest vectors to multiple query vectors. Flat
-        searches share one table scan across the query vectors, avoiding the scan
-        and memory amplification of making multiple queries concurrently. If
-        multiple vectors are passed in then
+        if you want to find the nearest vectors to multiple query vectors.
+        This is not expected to be faster than making multiple queries concurrently;
+        it is just a convenience method. If multiple vectors are passed in then
        an additional column `query_index` will be added to the results. This column
        will contain the index of the query vector that the result is nearest to.
        """
@@ -3540,8 +3509,8 @@ class AsyncFTSQuery(AsyncStandardQuery):

        Typically, a single vector is passed in as the query. However, you can also
        pass in multiple vectors.  This can be useful if you want to find the nearest
-        vectors to multiple query vectors. Flat searches share one table scan across
-        the query vectors instead of issuing concurrent full scans.
+        vectors to multiple query vectors. This is not expected to be faster than
+        making multiple queries concurrently; it is just a convenience method.
        If multiple vectors are passed in then an additional column `query_index`
        will be added to the results.  This column will contain the index of the
        query vector that the result is nearest to.
@@ -3901,15 +3870,14 @@ class AsyncHybridQuery(AsyncStandardQuery, AsyncVectorQueryBase):
        blob_paths: tuple[str, ...] = ()
        if self._table is not None:
            schema = await self._table.schema()
-            projection = _query_request_projection(req)
            blob_auto_row_id = blob_auto_row_id_for_scan(
                schema,
-                projection,
+                req.select,
                with_row_id=self._with_row_id,
            )
            if blob_auto_row_id:
                blob_paths = tuple(
-                    blob_v2_projection_sources(schema, projection).keys()
+                    blob_v2_projection_sources(schema, req.select).keys()
                )
        self._blob_auto_row_id = blob_auto_row_id
        self._blob_paths = blob_paths
@@ -7,7 +7,7 @@ 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, Union
 from urllib.parse import urlparse
 import warnings

@@ -25,7 +25,6 @@ from ..common import DATA
 from ..db import DBConnection, LOOP
 from ..functions import FunctionVersion, UdfDefinition
 from ..job import AsyncJob, Job
-from ..materialized_view import MaterializedView, SelectArg

 if TYPE_CHECKING:
    from .._lancedb import JobDescription, JobInfo
@@ -649,32 +648,6 @@ class RemoteDBConnection(DBConnection):
            namespace_path=namespace_path,
        )

-    @override
-    def create_materialized_view(
-        self,
-        name: str,
-        source: str,
-        *,
-        select: SelectArg = None,
-        where: Optional[str] = None,
-        limit: Optional[int] = None,
-    ) -> MaterializedView:
-        raise NotImplementedError(
-            "materialized views are supported only on local databases"
-        )
-
-    @override
-    def open_materialized_view(self, name: str) -> MaterializedView:
-        raise NotImplementedError(
-            "materialized views are supported only on local databases"
-        )
-
-    @override
-    def list_materialized_views(self) -> List[str]:
-        raise NotImplementedError(
-            "materialized views are supported only on local databases"
-        )
-
    @override
    def drop_table(self, name: str, namespace_path: Optional[List[str]] = None):
        """Drop a table from the database.
@@ -742,15 +715,8 @@ class RemoteDBConnection(DBConnection):
        return Job(self._conn.job(job_id))

    @override
-    def create_function_async(
-        self,
-        definition: UdfDefinition,
-        *,
-        secrets: Optional[Mapping[str, str]] = None,
-    ) -> Job[FunctionVersion]:
-        return Job(
-            LOOP.run(self._conn.create_function_async(definition, secrets=secrets))
-        )
+    def create_function_async(self, definition: UdfDefinition) -> Job[FunctionVersion]:
+        return Job(LOOP.run(self._conn.create_function_async(definition)))

    @override
    def get_function(self, name: str, *, version: str) -> FunctionVersion:
@@ -36,7 +36,6 @@ from lancedb._lancedb import (
    UpdateResult,
 )
 from lancedb.embeddings.base import EmbeddingFunctionConfig
-from lancedb.expr import Expr
 from lancedb.index import (
    FTS,
    BTree,
@@ -50,7 +49,7 @@ from lancedb.index import (
    LabelList,
 )
 from lancedb.job import Job
-from lancedb.functions import FunctionApplication, RefreshColumnResult
+from lancedb.functions import FunctionApplication
 from lancedb.remote.db import LOOP
 from lancedb.table import IndexConfigType, KNOWN_METRICS
 import pyarrow as pa
@@ -62,7 +61,6 @@ from lancedb.table import _normalize_progress

 from ..query import (
    AnalyzePlanDistributedMetrics,
-    DocumentGranularity,
    LanceQueryBuilder,
    LanceTakeQueryBuilder,
    LanceVectorQueryBuilder,
@@ -351,7 +349,6 @@ class RemoteTable(Table):
        ngram_max_length: int = 3,
        prefix_only: bool = False,
        block_size: int = 128,
-        document_granularity: DocumentGranularity = DocumentGranularity.ROW,
        name: Optional[str] = None,
    ):
        """Create a full-text search index on a column.
@@ -374,7 +371,6 @@ class RemoteTable(Table):
            ngram_max_length=ngram_max_length,
            prefix_only=prefix_only,
            block_size=block_size,
-            document_granularity=document_granularity,
        )
        LOOP.run(
            self._table.create_index(
@@ -614,7 +610,6 @@ class RemoteTable(Table):
        fill_value: float = 0.0,
        progress: Optional[Union[bool, Callable, Any]] = None,
        write_parallelism: Optional[int] = None,
-        allow_external_blob_outside_bases: bool = False,
    ) -> AddResult:
        """Add more data to the [Table][lancedb.table.Table].

@@ -647,8 +642,6 @@ class RemoteTable(Table):
            data in flight. Defaults to an estimate based on the data size,
            capped at the number of CPU cores. Lower this if bulk ingestion is
            using too much memory.
-        allow_external_blob_outside_bases: bool, default False
-            Not supported on LanceDB Cloud. Setting this raises.

        Returns
        -------
@@ -665,7 +658,6 @@ class RemoteTable(Table):
                    fill_value=fill_value,
                    progress=progress,
                    write_parallelism=write_parallelism,
-                    allow_external_blob_outside_bases=allow_external_blob_outside_bases,
                )
            )
        finally:
@@ -864,7 +856,7 @@ class RemoteTable(Table):

    def update(
        self,
-        where: Optional[Union[str, Expr]] = None,
+        where: Optional[str] = None,
        values: Optional[dict] = None,
        *,
        values_sql: Optional[Dict[str, str]] = None,
@@ -875,11 +867,9 @@ class RemoteTable(Table):

        Parameters
        ----------
-        where: str or [Expr][lancedb.expr.Expr], optional
-            The filter condition. Can be a SQL string or a type-safe
-            [Expr][lancedb.expr.Expr] built with [col][lancedb.expr.col] and
-            [lit][lancedb.expr.lit]. The filter must not be empty, or it will
-            error.
+        where: str, optional
+            The SQL where clause to use when updating rows. For example, 'x = 2'
+            or 'x IN (1, 2, 3)'. The filter must not be empty, or it will error.
        values: dict, optional
            The values to update. The keys are the column names and the values
            are the values to set.
@@ -982,7 +972,7 @@ class RemoteTable(Table):
    def refresh_column(self, column: str):
        return LOOP.run(self._table.refresh_column(column))

-    def refresh_column_async(self, column: str) -> Job[RefreshColumnResult]:
+    def refresh_column_async(self, column: str) -> Job:
        return Job(LOOP.run(self._table.refresh_column_async(column)))

    def alter_columns(
@@ -4,34 +4,30 @@

 """Schema helpers for Lance blob columns."""

-import importlib
-from typing import TYPE_CHECKING
-
 import pyarrow as pa
-import pyarrow.ipc
-
-if TYPE_CHECKING:
-    from lance.blob import BlobType as BlobType

 _BLOB_EXTENSION_NAME = "lance.blob.v2"
 _BLOB_V1_KEY = "lance-encoding:blob"
 _ARROW_EXT_NAME_KEY = "ARROW:extension:name"
-_BLOB_V2_STORAGE_TYPE = pa.struct(
-    [
-        pa.field("data", pa.large_binary(), nullable=True),
-        pa.field("uri", pa.utf8(), nullable=True),
-        pa.field("position", pa.uint64(), nullable=True),
-        pa.field("size", pa.uint64(), nullable=True),
-    ]
-)
-_resolved_blob_type = None


-class _FallbackBlobType(pa.ExtensionType):
-    """lance.blob.v2 extension type used when pylance is not installed."""
+class BlobType(pa.ExtensionType):
+    """PyArrow extension type for a Lance blob v2 column.
+
+    Queries return descriptors; call :meth:`~lancedb.table.Table.fetch_blob_files`
+    for lazy reads or :meth:`~lancedb.table.Table.fetch_blobs` for eager bytes.
+    """

    def __init__(self) -> None:
-        pa.ExtensionType.__init__(self, _BLOB_V2_STORAGE_TYPE, _BLOB_EXTENSION_NAME)
+        storage_type = pa.struct(
+            [
+                pa.field("data", pa.large_binary(), nullable=True),
+                pa.field("uri", pa.utf8(), nullable=True),
+                pa.field("position", pa.uint64(), nullable=True),
+                pa.field("size", pa.uint64(), nullable=True),
+            ]
+        )
+        super().__init__(storage_type, _BLOB_EXTENSION_NAME)

    def __arrow_ext_serialize__(self) -> bytes:
        return b""
@@ -39,16 +35,23 @@ class _FallbackBlobType(pa.ExtensionType):
    @classmethod
    def __arrow_ext_deserialize__(
        cls, storage_type: pa.DataType, serialized: bytes
-    ) -> "_FallbackBlobType":
+    ) -> "BlobType":
        return cls()

    def __reduce__(self):
+        # Ensure pickle round-trips on older pyarrow (apache/arrow#35599).
        return type(self).__arrow_ext_deserialize__, (
            self.storage_type,
            self.__arrow_ext_serialize__(),
        )


+try:
+    pa.register_extension_type(BlobType())  # type: ignore[arg-type]
+except pa.ArrowKeyError:
+    pass
+
+
 def _metadata_value(metadata: dict, key: str):
    return metadata.get(key.encode()) or metadata.get(key)

@@ -89,105 +92,43 @@ def is_blob_like_field(field: pa.Field) -> bool:
    return is_blob_v2_field(field) or _metadata_marks_legacy_blob(field.metadata or {})


-def _collect_blob_paths(schema: pa.Schema, is_blob) -> list[tuple[str, bool]]:
-    """Walk the schema and return (path, has_list_ancestor) for each blob field."""
-    paths: list[tuple[str, bool]] = []
+def _collect_blob_paths(schema: pa.Schema, is_blob) -> list[str]:
+    paths: list[str] = []

-    def walk(fields, prefix: str, has_list_ancestor: bool) -> None:
+    def walk(fields, prefix: str) -> None:
        for field in fields:
            path = f"{prefix}.{field.name}" if prefix else field.name
            if is_blob(field):
-                paths.append((path, has_list_ancestor))
+                paths.append(path)
            elif pa.types.is_struct(field.type):
-                walk(field.type, path, has_list_ancestor)
+                walk(field.type, path)
            elif (
                pa.types.is_list(field.type)
                or pa.types.is_large_list(field.type)
                or pa.types.is_fixed_size_list(field.type)
            ):
-                walk([field.type.value_field], path, True)
+                walk([field.type.value_field], path)

-    walk(schema, "", False)
+    walk(schema, "")
    return paths


 def blob_column_paths(schema: pa.Schema) -> list[str]:
    """Dotted paths of blob-like columns (v2 extension or legacy metadata)."""
-    return [path for path, _ in _collect_blob_paths(schema, is_blob_like_field)]
+    return _collect_blob_paths(schema, is_blob_like_field)


 def blob_v2_column_paths(schema: pa.Schema) -> list[str]:
-    return [path for path, _ in _collect_blob_paths(schema, is_blob_v2_field)]
-
-
-def row_addressable_blob_v2_paths(schema: pa.Schema) -> list[str]:
-    """Blob v2 paths with one blob addressable by table row id.
-
-    ``fetch_blobs`` and the descriptor row-id ride-along address one blob per
-    row, so a blob inside a list container has no row-id slot and no fetch
-    path. Those columns still store and query as raw descriptors.
-    """
-    return [
-        path
-        for path, has_list_ancestor in _collect_blob_paths(schema, is_blob_v2_field)
-        if not has_list_ancestor
-    ]
+    return _collect_blob_paths(schema, is_blob_v2_field)


 def schema_has_blob_field(schema: pa.Schema) -> bool:
    return bool(blob_column_paths(schema))


-def _deserialize_registered_type(extension_type: pa.ExtensionType) -> pa.DataType:
-    """Return the type Arrow reconstructs for this extension name."""
-    schema = pa.schema([pa.field("value", extension_type)])
-    restored = pa.ipc.read_schema(schema.serialize())
-    return restored.field("value").type
-
-
-def _resolve_blob_type():
-    """Return the BlobType class this process should use.
-
-    pylance's class when it owns the lance.blob.v2 registry entry,
-    otherwise LanceDB's fallback. A different registered class is an error.
-    """
-    global _resolved_blob_type
-    if _resolved_blob_type is not None:
-        return _resolved_blob_type
-    try:
-        blob_module = importlib.import_module("lance.blob")
-    except ModuleNotFoundError as err:
-        if err.name not in ("lance", "lance.blob"):
-            raise
-    else:
-        blob_type = getattr(blob_module, "BlobType", None)
-        if blob_type is not None:
-            registered_type = _deserialize_registered_type(blob_type())
-            if type(registered_type) is not blob_type:
-                registered_cls = type(registered_type)
-                raise ValueError(
-                    "lance.blob.v2 is already registered by "
-                    f"{registered_cls.__module__}.{registered_cls.__qualname__}"
-                )
-            _resolved_blob_type = blob_type
-            return blob_type
-    try:
-        pa.register_extension_type(_FallbackBlobType())  # type: ignore[arg-type]
-    except pa.ArrowKeyError as err:
-        raise ValueError(
-            "lance.blob.v2 is already registered by another extension class"
-        ) from err
-    _resolved_blob_type = _FallbackBlobType
-    return _resolved_blob_type
-
-
 def blob(name: str, nullable: bool = True) -> pa.Field:
-    """Create a Lance blob v2 column field.
-
-    When pylance is installed this is ``lance.blob.BlobType``.
-    """
-    blob_type = _resolve_blob_type()
-    return pa.field(name, blob_type(), nullable=nullable)
+    """Create a Lance blob v2 column field."""
+    return pa.field(name, BlobType(), nullable=nullable)


 def vector(dimension: int, value_type: pa.DataType = pa.float32()) -> pa.DataType:
@@ -214,11 +155,3 @@ def vector(dimension: int, value_type: pa.DataType = pa.float32()) -> pa.DataTyp
    ... ])
    """
    return pa.list_(value_type, dimension)
-
-
-def __getattr__(name: str):
-    if name == "BlobType":
-        blob_type = _resolve_blob_type()
-        globals()["BlobType"] = blob_type
-        return blob_type
-    raise AttributeError(f"module {__name__!r} has no attribute {name!r}")
--- a/Show More
+++ b/Show More