Bump version: 0.39.0-beta.0 → 0.39.0-beta.1

Cargo.lock

@@ -5403,7 +5403,7 @@ dependencies = [
 
 [[package]]
 name = "lancedb"
-version = "0.39.0-beta.1"
+version = "0.39.0-beta.0"
 dependencies = [
  "ahash",
  "anyhow",
@@ -5491,7 +5491,7 @@ dependencies = [
 
 [[package]]
 name = "lancedb-nodejs"
-version = "0.39.0-beta.1"
+version = "0.39.0-beta.0"
 dependencies = [
  "arrow-array",
  "arrow-buffer",
@@ -5516,7 +5516,7 @@ dependencies = [
 
 [[package]]
 name = "lancedb-python"
-version = "0.39.0-beta.1"
+version = "0.39.0-beta.0"
 dependencies = [
  "arrow",
  "async-trait",

docs/src/js/interfaces/IndexOptions.md

@@ -50,7 +50,7 @@ If this is false, and another index already exists on the same columns
 and the same name, then an error will be returned.  This is true even if
 that index is out of date.
 
-The default is false
+The default is true
 
 ***
 

nodejs/__test__/table.test.ts

@@ -1526,7 +1526,11 @@ describe("When creating an index", () => {
 
   it("should allow me to replace (or not) an existing index", async () => {
     await tbl.createIndex("id");
-    await expect(tbl.createIndex("id")).rejects.toThrow("already exists");
+    // Default is replace=true
+    await tbl.createIndex("id");
+    await expect(tbl.createIndex("id", { replace: false })).rejects.toThrow(
+      "already exists",
+    );
     await tbl.createIndex("id", { replace: true });
   });
 

nodejs/lancedb/indices.ts

@@ -841,7 +841,7 @@ export interface IndexOptions {
    * and the same name, then an error will be returned.  This is true even if
    * that index is out of date.
    *
-   * The default is false
+   * The default is true
    */
   replace?: boolean;
 

nodejs/package.json

@@ -44,7 +44,7 @@
     "@biomejs/biome": "^1.7.3",
     "@jest/globals": "^29.7.0",
     "@napi-rs/cli": "3.7.0",
-    "@opentelemetry/sdk-metrics": "^2.10.0",
+    "@opentelemetry/sdk-metrics": "^1.30.0",
     "@types/axios": "^0.14.0",
     "@types/jest": "^29.1.2",
     "@types/node": "22.7.4",
@@ -56,7 +56,7 @@
     "eslint": "^8.57.0",
     "jest": "^29.7.0",
     "shx": "^0.3.4",
-    "tmp": "^0.2.7",
+    "tmp": "^0.2.3",
     "ts-jest": "^29.1.2",
     "typedoc": "0.26.4",
     "typedoc-plugin-markdown": "4.2.1",

nodejs/pnpm-workspace.yaml

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

python/python/lancedb/remote/table.py

@@ -402,7 +402,6 @@ class RemoteTable(Table):
         /,
         *,
         config: IndexConfigType,
-        replace: bool = ...,
         wait_timeout: Optional[timedelta] = ...,
         name: Optional[str] = ...,
         train: bool = ...,
@@ -417,7 +416,7 @@ class RemoteTable(Table):
         index_cache_size: Optional[int] = ...,
         num_partitions: Optional[int] = ...,
         num_sub_vectors: Optional[int] = ...,
-        replace: bool = ...,
+        replace: Optional[bool] = ...,
         accelerator: Optional[str] = ...,
         index_type: Literal[
             "VECTOR", "IVF_FLAT", "IVF_SQ", "IVF_PQ", "IVF_HNSW_SQ", "IVF_HNSW_PQ"
@@ -436,7 +435,7 @@ class RemoteTable(Table):
         index_cache_size: Optional[int] = None,
         num_partitions: Optional[int] = None,
         num_sub_vectors: Optional[int] = None,
-        replace: bool = False,
+        replace: Optional[bool] = None,
         accelerator: Optional[str] = None,
         index_type="vector",
         wait_timeout: Optional[timedelta] = None,
@@ -480,6 +479,7 @@ class RemoteTable(Table):
             vector_column_name,
             accelerator,
             index_cache_size,
+            replace,
         )
 
         if is_legacy:
@@ -503,6 +503,12 @@ class RemoteTable(Table):
                     "If you have 100M+ vectors to index,"
                     "please contact us at contact@lancedb.com"
                 )
+            if replace is not None:
+                logging.warning(
+                    "replace is not supported on LanceDB cloud."
+                    "Existing indexes will always be replaced."
+                )
+
             idx_type = index_type.upper()
             if idx_type == "VECTOR" or idx_type == "IVF_PQ":
                 config = IvfPq(
@@ -555,7 +561,7 @@ class RemoteTable(Table):
         column: str,
         *,
         config: IndexConfigType,
-        replace: bool = False,
+        replace: Optional[bool] = None,
         wait_timeout: Optional[timedelta] = None,
         name: Optional[str] = None,
         train: bool = True,
@@ -587,6 +593,7 @@ class RemoteTable(Table):
         vector_column_name: str,
         accelerator: Optional[str],
         index_cache_size: Optional[int],
+        replace: Optional[bool],
     ) -> bool:
         """Detect if this is a legacy create_index call."""
         if config is not None:
@@ -598,6 +605,7 @@ class RemoteTable(Table):
                 num_sub_vectors,
                 accelerator,
                 index_cache_size,
+                replace,
             )
         ):
             return True

python/python/lancedb/table.py

@@ -1132,7 +1132,7 @@ class Table(ABC):
         num_partitions: Optional[int] = None,
         num_sub_vectors: Optional[int] = None,
         vector_column_name: str = VECTOR_COLUMN_NAME,
-        replace: bool = False,
+        replace: bool = True,
         accelerator: Optional[str] = None,
         index_cache_size: Optional[int] = None,
         *,
@@ -1166,7 +1166,7 @@ class Table(ABC):
             The index configuration object. If provided, uses the new unified API.
             Can be one of: IvfFlat, IvfPq, IvfSq, IvfRq, HnswPq, HnswSq,
             BTree, Bitmap, LabelList, Fm, FTS.
-        replace : bool, default False
+        replace : bool, default True
             Whether to replace an existing index on this column.
         wait_timeout : timedelta, optional
             Timeout to wait for async indexing to complete.
@@ -1198,7 +1198,7 @@ class Table(ABC):
         column: str,
         *,
         config: IndexConfigType,
-        replace: bool = False,
+        replace: Optional[bool] = None,
         wait_timeout: Optional[timedelta] = None,
         name: Optional[str] = None,
         train: bool = True,
@@ -1260,7 +1260,7 @@ class Table(ABC):
         self,
         column: str,
         *,
-        replace: bool = False,
+        replace: bool = True,
         index_type: ScalarIndexType = "BTREE",
         wait_timeout: Optional[timedelta] = None,
         name: Optional[str] = None,
@@ -1272,7 +1272,7 @@ class Table(ABC):
         column : str
             The column to be indexed.  Must be a boolean, integer, float,
             or string column.
-        replace : bool, default False
+        replace : bool, default True
             Replace the existing index if it exists.
         index_type: Literal["BTREE", "BITMAP", "LABEL_LIST"], default "BTREE"
             The type of index to create.
@@ -2988,7 +2988,7 @@ class LanceTable(Table):
         num_partitions: Optional[int] = None,
         num_sub_vectors: Optional[int] = None,
         vector_column_name: str = VECTOR_COLUMN_NAME,
-        replace: bool = False,
+        replace: bool = True,
         accelerator: Optional[str] = None,
         index_cache_size: Optional[int] = None,
         num_bits: int = 8,
@@ -3030,7 +3030,7 @@ class LanceTable(Table):
             The index configuration object. If provided, uses the new unified API.
             Can be one of: IvfFlat, IvfPq, IvfSq, IvfRq, HnswPq, HnswSq,
             BTree, Bitmap, LabelList, Fm, FTS.
-        replace : bool, default False
+        replace : bool, default True
             Whether to replace an existing index on this column.
         wait_timeout : timedelta, optional
             Timeout to wait for async indexing to complete.
@@ -3169,7 +3169,7 @@ class LanceTable(Table):
         column: str,
         *,
         config: IndexConfigType,
-        replace: bool = False,
+        replace: Optional[bool] = None,
         wait_timeout: Optional[timedelta] = None,
         name: Optional[str] = None,
         train: bool = True,
@@ -3424,7 +3424,7 @@ class LanceTable(Table):
         self,
         column: str,
         *,
-        replace: bool = False,
+        replace: bool = True,
         index_type: ScalarIndexType = "BTREE",
         name: Optional[str] = None,
     ):
@@ -5313,7 +5313,7 @@ class AsyncTable:
         self,
         column: str,
         *,
-        replace: bool = False,
+        replace: Optional[bool] = None,
         config: Optional[
             Union[
                 IvfFlat,
@@ -5344,14 +5344,14 @@ class AsyncTable:
         ----------
         column: str
             The column to index.
-        replace: bool, default False
+        replace: bool, default True
             Whether to replace the existing index
 
             If this is false, and another index already exists on the same columns
             and the same name, then an error will be returned.  This is true even if
             that index is out of date.
 
-            The default is False
+            The default is True
         config: default None
             For advanced configuration you can specify the type of index you would
             like to create.   You can also specify index-specific parameters when
@@ -5409,7 +5409,7 @@ class AsyncTable:
         self,
         column: str,
         *,
-        replace: bool = False,
+        replace: Optional[bool] = None,
         config: Optional[
             Union[
                 IvfFlat,

python/python/tests/test_index.py

@@ -97,9 +97,6 @@ async def test_create_index_async_returns_done_job(some_table: AsyncTable):
 async def test_create_scalar_index(some_table: AsyncTable):
     # Can create
     await some_table.create_index("id")
-    # Can't recreate by default
-    with pytest.raises(RuntimeError, match="already exists"):
-        await some_table.create_index("id")
     # Can recreate if replace=True
     await some_table.create_index("id", replace=True)
     indices = await some_table.list_indices()
@@ -113,7 +110,7 @@ async def test_create_scalar_index(some_table: AsyncTable):
     with pytest.raises(RuntimeError, match="already exists"):
         await some_table.create_index("id", replace=False)
     # can also specify index type
-    await some_table.create_index("id", config=BTree(), replace=True)
+    await some_table.create_index("id", config=BTree())
 
     await some_table.drop_index("id_idx")
     indices = await some_table.list_indices()
@@ -354,18 +351,13 @@ async def test_full_text_search_index(some_table: AsyncTable):
 async def test_create_vector_index(some_table: AsyncTable):
     # Can create
     await some_table.create_index("vector")
-    # Can't recreate by default
-    with pytest.raises(RuntimeError, match="already exists"):
-        await some_table.create_index("vector")
     # Can recreate if replace=True
     await some_table.create_index("vector", replace=True)
     # Can't recreate if replace=False
     with pytest.raises(RuntimeError, match="already exists"):
         await some_table.create_index("vector", replace=False)
     # Can also specify index type
-    await some_table.create_index(
-        "vector", config=IvfPq(num_partitions=100), replace=True
-    )
+    await some_table.create_index("vector", config=IvfPq(num_partitions=100))
     indices = await some_table.list_indices()
     assert len(indices) == 1
     assert indices[0].index_type == "IvfPq"

python/python/tests/test_remote_db.py

@@ -834,7 +834,7 @@ def test_table_create_indices():
         vector_req = received_requests[2]
         assert "name" in vector_req
         assert vector_req["name"] == "custom_vector_idx"
-        assert vector_req["replace"] is False
+        assert "replace" not in vector_req
 
         table.wait_for_index(["custom_scalar_idx"], timedelta(seconds=2))
         table.wait_for_index(

python/python/tests/test_table.py

@@ -1600,7 +1600,7 @@ def test_create_index_method(mock_create_index, mem_db: DBConnection):
     )
     mock_create_index.assert_called_with(
         "my_vector",
-        replace=False,
+        replace=True,
         config=expected_config,
         wait_timeout=None,
         name=None,
@@ -1620,7 +1620,7 @@ def test_create_index_method(mock_create_index, mem_db: DBConnection):
     )
     mock_create_index.assert_called_with(
         "my_vector",
-        replace=False,
+        replace=True,
         config=expected_config,
         wait_timeout=None,
         name=None,
@@ -1646,7 +1646,7 @@ def test_create_index_name_and_train_parameters(
     expected_config = IvfPq()  # Default config
     mock_create_index.assert_called_with(
         "vector",
-        replace=False,
+        replace=True,
         config=expected_config,
         wait_timeout=None,
         name="my_custom_index",
@@ -1657,7 +1657,7 @@ def test_create_index_name_and_train_parameters(
     table.create_index(vector_column_name="vector", train=False)
     mock_create_index.assert_called_with(
         "vector",
-        replace=False,
+        replace=True,
         config=expected_config,
         wait_timeout=None,
         name=None,
@@ -1668,7 +1668,7 @@ def test_create_index_name_and_train_parameters(
     table.create_index(vector_column_name="vector", name="my_index_name", train=True)
     mock_create_index.assert_called_with(
         "vector",
-        replace=False,
+        replace=True,
         config=expected_config,
         wait_timeout=None,
         name="my_index_name",
@@ -1705,7 +1705,7 @@ def test_create_index_new_api(mock_create_index, mem_db: DBConnection):
         table.create_index("vector", config=IvfPq(distance_type="l2"))
     mock_create_index.assert_called_with(
         "vector",
-        replace=False,
+        replace=True,
         config=IvfPq(distance_type="l2"),
         wait_timeout=None,
         name=None,
@@ -1716,7 +1716,7 @@ def test_create_index_new_api(mock_create_index, mem_db: DBConnection):
     table.create_index("category", config=BTree())
     mock_create_index.assert_called_with(
         "category",
-        replace=False,
+        replace=True,
         config=BTree(),
         wait_timeout=None,
         name=None,
@@ -1727,7 +1727,7 @@ def test_create_index_new_api(mock_create_index, mem_db: DBConnection):
     table.create_index("text", config=FTS(with_position=True))
     mock_create_index.assert_called_with(
         "text",
-        replace=False,
+        replace=True,
         config=FTS(with_position=True),
         wait_timeout=None,
         name=None,

rust/lancedb/src/index.rs

@@ -200,14 +200,14 @@ impl IndexBuilder {
             parent,
             index,
             columns,
-            replace: false,
+            replace: true,
             train: true,
             wait_timeout: None,
             name: None,
         }
     }
 
-    /// Whether to replace the existing index, the default is `false`.
+    /// Whether to replace the existing index, the default is `true`.
     ///
     /// If this is false, and another index already exists on the same columns
     /// and the same name, then an error will be returned.  This is true even if

rust/lancedb/src/job.rs

@@ -40,7 +40,6 @@ impl TerminalResult {
         }
     }
 
-    #[cfg(feature = "remote")]
     pub(crate) fn remote(value: Option<Value>, request_id: String) -> Self {
         Self {
             value,
@@ -48,46 +47,30 @@ impl TerminalResult {
         }
     }
 
-    #[cfg(feature = "remote")]
     pub(crate) fn value(&self) -> Option<&Value> {
         self.value.as_ref()
     }
 
     fn decode<T: DeserializeOwned>(self) -> Result<T> {
-        let value = self.value.ok_or_else(|| {
-            #[cfg(feature = "remote")]
-            if let Some(request_id) = &self.request_id {
-                return Error::Http {
-                    source: "successful typed job response did not contain a result".into(),
-                    request_id: request_id.clone(),
-                    status_code: None,
-                };
-            }
-            Error::Runtime {
+        let value = self.value.ok_or_else(|| match &self.request_id {
+            Some(request_id) => Error::Http {
+                source: "successful typed job response did not contain a result".into(),
+                request_id: request_id.clone(),
+                status_code: None,
+            },
+            None => Error::Runtime {
                 message: "successful typed job did not contain a result".to_string(),
-            }
+            },
         })?;
-        serde_json::from_value(value).map_err(|error| {
-            #[cfg(feature = "remote")]
-            {
-                match self.request_id {
-                    Some(request_id) => Error::Http {
-                        source: format!("failed to parse typed job result: {error}").into(),
-                        request_id,
-                        status_code: None,
-                    },
-                    None => Error::Runtime {
-                        message: format!("failed to parse typed job result: {error}"),
-                    },
-                }
-            }
-            #[cfg(not(feature = "remote"))]
-            {
-                let _ = self.request_id;
-                Error::Runtime {
-                    message: format!("failed to parse typed job result: {error}"),
-                }
-            }
+        serde_json::from_value(value).map_err(|error| match self.request_id {
+            Some(request_id) => Error::Http {
+                source: format!("failed to parse typed job result: {error}").into(),
+                request_id,
+                status_code: None,
+            },
+            None => Error::Runtime {
+                message: format!("failed to parse typed job result: {error}"),
+            },
         })
     }
 }
@@ -134,7 +117,6 @@ impl Job<()> {
         }
     }
 
-    #[cfg(feature = "remote")]
     pub(crate) fn new(handle: Box<dyn JobHandle>) -> Self {
         Self {
             inner: JobInner::Handle {

rust/lancedb/src/remote/table.rs

@@ -524,10 +524,13 @@ impl<S: HttpSend> RemoteTable<S> {
             _ => resolve_arrow_field_path(&schema, &column)?,
         };
         let mut body = serde_json::json!({
-            "column": canonical_column,
-            "replace": index.replace,
+            "column": canonical_column
         });
 
+        if !index.replace {
+            body["replace"] = false.into();
+        }
+
         // Add name parameter if provided (for backwards compatibility, only include if Some)
         if let Some(ref name) = index.name {
             body["name"] = serde_json::Value::String(name.clone());
@@ -6306,7 +6309,6 @@ mod tests {
                             let mut expected_body = expected_body.clone();
                             expected_body["column"] = "a".into();
                             expected_body[INDEX_TYPE_KEY] = index_type.into();
-                            expected_body["replace"] = false.into();
 
                             assert_eq!(body, expected_body);
 
@@ -6325,7 +6327,7 @@ mod tests {
     }
 
     #[tokio::test]
-    async fn test_create_index_forwards_default_replace_false_on_existing_route() {
+    async fn test_create_index_forwards_replace_false_on_existing_route() {
         let table = Table::new_with_handler("my_table", move |request| {
             assert_eq!(request.method(), "POST");
             match request.url().path() {
@@ -6352,40 +6354,7 @@ mod tests {
 
         table
             .create_index(&["a"], Index::BTree(Default::default()))
-            .execute()
-            .await
-            .unwrap();
-    }
-
-    #[tokio::test]
-    async fn test_create_index_forwards_explicit_replace_true_on_existing_route() {
-        let table = Table::new_with_handler("my_table", move |request| {
-            assert_eq!(request.method(), "POST");
-            match request.url().path() {
-                "/v1/table/my_table/describe/" => {
-                    let schema = Schema::new(vec![Field::new("a", DataType::Int32, false)]);
-                    http::Response::builder()
-                        .status(200)
-                        .body(describe_response(&schema))
-                        .unwrap()
-                }
-                "/v1/table/my_table/create_index/" => {
-                    let body = request.body().unwrap().as_bytes().unwrap();
-                    let body: serde_json::Value = serde_json::from_slice(body).unwrap();
-                    assert_eq!(body["replace"], json!(true));
-
-                    http::Response::builder()
-                        .status(200)
-                        .body("{}".to_string())
-                        .unwrap()
-                }
-                path => panic!("Unexpected path: {}", path),
-            }
-        });
-
-        table
-            .create_index(&["a"], Index::BTree(Default::default()))
-            .replace(true)
+            .replace(false)
             .execute()
             .await
             .unwrap();
@@ -6616,46 +6585,38 @@ mod tests {
             json!({
                 "column": "rowId",
                 "index_type": "BTREE",
-                "replace": false,
             }),
             json!({
                 "column": "`row-id`",
                 "index_type": "BTREE",
-                "replace": false,
             }),
             json!({
                 "column": "userId",
                 "index_type": "BTREE",
-                "replace": false,
             }),
             json!({
                 "column": "MetaData.userId",
                 "index_type": "BTREE",
-                "replace": false,
             }),
             json!({
                 "column": "metadata.user_id",
                 "index_type": "BTREE",
-                "replace": false,
             }),
             json!({
                 "column": "image.embedding",
                 "index_type": "IVF_PQ",
                 "metric_type": "l2",
-                "replace": false,
             }),
             {
                 let mut body = serde_json::to_value(InvertedIndexParams::default()).unwrap();
                 body["column"] = "payload.text".into();
                 body["index_type"] = "FTS".into();
-                body["replace"] = false.into();
                 body
             },
             {
                 let mut body = serde_json::to_value(InvertedIndexParams::default()).unwrap();
                 body["column"] = "docs.content".into();
                 body["index_type"] = "FTS".into();
-                body["replace"] = false.into();
                 body
             },
             {
@@ -6663,18 +6624,15 @@ mod tests {
                 body["column"] = "docs.content".into();
                 body["index_type"] = "FTS".into();
                 body["document_granularity"] = "list_element".into();
-                body["replace"] = false.into();
                 body
             },
             json!({
                 "column": "`meta-data`.`user-id`",
                 "index_type": "BTREE",
-                "replace": false,
             }),
             json!({
                 "column": "literal.`a.b`",
                 "index_type": "BTREE",
-                "replace": false,
             }),
         ]);
         let request_idx = Arc::new(AtomicUsize::new(0));

rust/lancedb/src/table/create_index.rs

@@ -133,7 +133,7 @@ impl NativeTable {
                     ),
                 });
             }
-            (resolved.canonical_path, resolved.terminal_field)
+            (resolved.canonical_path, resolved.field)
         } else {
             Self::resolve_index_field(dataset.schema(), &opts.columns[0])?
         };
@@ -439,8 +439,7 @@ mod tests {
     use arrow_array::record_batch;
     use arrow_array::{
         Array, ArrayRef, BinaryArray, BooleanArray, FixedSizeListArray, Float32Array, Int32Array,
-        LargeBinaryArray, LargeStringArray, ListArray, RecordBatch, StringArray, StructArray,
-        UInt32Array,
+        LargeBinaryArray, LargeStringArray, RecordBatch, StringArray, StructArray,
     };
     use arrow_data::ArrayDataBuilder;
     use arrow_schema::{DataType, Field, Schema};
@@ -459,7 +458,6 @@ mod tests {
     use crate::query::{ExecutableQuery, QueryBase};
     use crate::table::optimize::{CompactionOptions, OptimizeAction};
     use lance_index::scalar::FullTextSearchQuery;
-    use lance_index::scalar::inverted::query::{FtsQuery, MatchQuery};
 
     fn create_fixed_size_list<T: Array>(
         values: T,
@@ -601,80 +599,6 @@ mod tests {
         assert!(invalid_granularity.is_err());
     }
 
-    #[tokio::test]
-    async fn test_nested_list_fts_uses_deepest_document_coordinates() {
-        let conn = connect("memory://").execute().await.unwrap();
-        let mut docs = ListBuilder::new(ListBuilder::new(StringBuilder::new()));
-
-        docs.values().values().append_value("alpha");
-        docs.values().values().append_value("beta");
-        docs.values().append(true);
-        docs.values().values().append_value("gamma");
-        docs.values().values().append_value("alpha delta");
-        docs.values().append(true);
-        docs.append(true);
-
-        docs.values().append(true);
-        docs.values().values().append_value("alpha");
-        docs.values().append(true);
-        docs.append(true);
-
-        let batch = RecordBatch::try_from_iter(vec![
-            ("id", Arc::new(Int32Array::from(vec![0, 1])) as ArrayRef),
-            ("docs", Arc::new(docs.finish()) as ArrayRef),
-        ])
-        .unwrap();
-        let table = conn.create_table("nested", batch).execute().await.unwrap();
-
-        let job = table
-            .create_index(
-                &["docs"],
-                Index::FTS(
-                    FtsIndexBuilder::default()
-                        .document_granularity(DocumentGranularity::ListElement),
-                ),
-            )
-            .execute_async()
-            .await
-            .unwrap();
-        job.wait().await.unwrap();
-
-        let query = FullTextSearchQuery::new_query(FtsQuery::Match(
-            MatchQuery::new("alpha".to_string())
-                .with_column(Some("docs".to_string()))
-                .with_document_granularity(DocumentGranularity::ListElement),
-        ));
-        let batches = table
-            .query()
-            .full_text_search(query)
-            .limit(10)
-            .execute()
-            .await
-            .unwrap()
-            .try_collect::<Vec<_>>()
-            .await
-            .unwrap();
-
-        let mut hits = Vec::new();
-        for batch in batches {
-            let ids = batch["id"].as_any().downcast_ref::<Int32Array>().unwrap();
-            let coordinates = batch["_doc_index"]
-                .as_any()
-                .downcast_ref::<ListArray>()
-                .unwrap();
-            for row in 0..batch.num_rows() {
-                let coordinate = coordinates.value(row);
-                let coordinate = coordinate.as_any().downcast_ref::<UInt32Array>().unwrap();
-                hits.push((ids.value(row), coordinate.values().to_vec()));
-            }
-        }
-        hits.sort_unstable();
-        assert_eq!(
-            hits,
-            vec![(0, vec![0, 0]), (0, vec![1, 1]), (1, vec![1, 0])]
-        );
-    }
-
     /// Concurrent waiters, and a wait issued after the job settled, all
     /// succeed once the build does.
     #[tokio::test]
@@ -725,11 +649,12 @@ mod tests {
             .await
             .unwrap();
 
-        // Rebuilding the same index without explicit replace fails once the build
+        // Rebuilding the same index without replace fails once the build
         // starts, so the failure reaches the job rather than execute_async.
         let job = Arc::new(
             table
                 .create_index(&["id"], Index::BTree(BTreeIndexBuilder::default()))
+                .replace(false)
                 .execute_async()
                 .await
                 .unwrap(),
@@ -768,6 +693,7 @@ mod tests {
 
         let job = table
             .create_index(&["id"], Index::BTree(BTreeIndexBuilder::default()))
+            .replace(false)
             .execute_async()
             .await
             .unwrap();
@@ -1104,7 +1030,6 @@ mod tests {
         // Can also specify btree
         table
             .create_index(&["i"], Index::BTree(BTreeIndexBuilder::default()))
-            .replace(true)
             .execute()
             .await
             .unwrap();

rust/lancedb/src/utils/mod.rs

@@ -227,7 +227,7 @@ pub(crate) fn resolve_arrow_field_path(schema: &Schema, column: &str) -> Result<
 
 pub(crate) struct ResolvedFtsField {
     pub canonical_path: String,
-    pub terminal_field: Field,
+    pub field: Field,
     pub list_depth: usize,
 }
 
@@ -309,7 +309,7 @@ pub(crate) fn resolve_lance_fts_field_path(
     );
     Ok(ResolvedFtsField {
         canonical_path,
-        terminal_field: Field::from(terminal),
+        field: Field::from(field),
         list_depth,
     })
 }
@@ -375,7 +375,7 @@ pub(crate) fn resolve_arrow_fts_field_path(
             message: format!("Invalid schema: {}", e),
         })?;
     let resolved = resolve_lance_fts_field_path(&lance_schema, column)?;
-    Ok((resolved.canonical_path, resolved.terminal_field))
+    Ok((resolved.canonical_path, resolved.field))
 }
 
 pub fn supported_btree_data_type(dtype: &DataType) -> bool {
@@ -647,9 +647,8 @@ mod tests {
             Field::new("docs", text_list(), true),
         ]);
 
-        let (path, field) = resolve_arrow_fts_field_path(&schema, "docs.content").unwrap();
+        let (path, _) = resolve_arrow_fts_field_path(&schema, "docs.content").unwrap();
         assert_eq!(path, "docs.content");
-        assert_eq!(field.data_type(), &DataType::Utf8);
 
         let lance_schema = lance_core::datatypes::Schema::try_from(&schema).unwrap();
         let field_id = lance_schema