chore: update pylance dependency to v10.0.0b5

docs/src/distributed-indexing.md

@@ -270,10 +270,7 @@ dataset = lance.dataset("path/to/dataset")
 
 # Build distributed index
 updated_dataset = lr.create_scalar_index(
-   uri=dataset.uri,
-   column="text",
-   index_type="INVERTED",
-   num_workers=4
+    uri=dataset.uri, column="text", index_type="INVERTED", num_workers=4
 )
 
 # Verify index creation
@@ -282,8 +279,7 @@ print(f"Index list: {indices}")
 
 # Use index for search
 results = updated_dataset.scanner(
-   full_text_query="search term",
-   columns=["id", "text"]
+    full_text_query="search term", columns=["id", "text"]
 ).to_table()
 print(f"Search results: {results}")
 ```
@@ -304,7 +300,9 @@ updated_dataset = lr.create_scalar_index(
 
 # Example queries
 updated_dataset.scanner(filter="id = 100", columns=["id", "text"]).to_table()
-updated_dataset.scanner(filter="id >= 200 AND id < 800", columns=["id", "text"]).to_table()
+updated_dataset.scanner(
+    filter="id >= 200 AND id < 800", columns=["id", "text"]
+).to_table()
 ```
 
 ### Vector Index (IVF_PQ / IVF_RQ / IVF_SQ / IVF_FLAT)
@@ -321,7 +319,7 @@ updated_dataset = lr.create_index(
     num_partitions=256,
     num_sub_vectors=16,
     sample_rate=64,
-    metric="l2"
+    metric="l2",
 )
 
 # Build a distributed IVF_SQ index
@@ -399,11 +397,11 @@ print(plan)
 
 ```python
 updated_dataset = lr.create_scalar_index(
-   uri="path/to/dataset",
-   column="text",
-   index_type="INVERTED",
-   num_workers=4,
-   ray_remote_args={"num_cpus": 2, "resources": {"custom_resource": 1}}
+    uri="path/to/dataset",
+    column="text",
+    index_type="INVERTED",
+    num_workers=4,
+    ray_remote_args={"num_cpus": 2, "resources": {"custom_resource": 1}},
 )
 ```
 
@@ -444,34 +442,34 @@ The current global Pool integration is limited to `vector_search()`.  The same p
 ```python
 # Create index with custom name
 updated_dataset = lr.create_scalar_index(
-   uri="path/to/dataset",
-   column="text",
-   index_type="INVERTED",
-   name="my_text_index",
-   num_workers=4
+    uri="path/to/dataset",
+    column="text",
+    index_type="INVERTED",
+    name="my_text_index",
+    num_workers=4,
 )
 
 # Try to create another index with the same name (will replace by default)
 updated_dataset = lr.create_scalar_index(
-   uri="path/to/dataset",
-   column="text",
-   index_type="INVERTED",
-   name="my_text_index",  # Same name as before
-   replace=True,          # Explicitly allow replacement (default behavior)
-   num_workers=4
+    uri="path/to/dataset",
+    column="text",
+    index_type="INVERTED",
+    name="my_text_index",  # Same name as before
+    replace=True,  # Explicitly allow replacement (default behavior)
+    num_workers=4,
 )
 
 # Prevent index replacement
 import lance_ray as lr
 
 try:
     updated_dataset = lr.create_scalar_index(
-       uri="path/to/dataset",
-       column="text",
-       index_type="INVERTED",
-       name="my_text_index",  # Same name as existing index
-       replace=False,         # Prevent replacement
-       num_workers=4
+        uri="path/to/dataset",
+        column="text",
+        index_type="INVERTED",
+        name="my_text_index",  # Same name as existing index
+        replace=False,  # Prevent replacement
+        num_workers=4,
     )
 except ValueError as e:
     print(f"Index creation failed: {e}")

docs/src/examples.md

@@ -37,9 +37,7 @@ print(f"Filtered count: {filtered_ds.count()}")
 
 # Read with column selection and filtering
 ds_filtered = read_lance(
-    "sample_dataset.lance",
-    columns=["user_id", "name", "score"],
-    filter="score > 75.0"
+    "sample_dataset.lance", columns=["user_id", "name", "score"], filter="score > 75.0"
 )
 print(f"Schema: {ds_filtered.schema()}")
 ```
@@ -51,16 +49,14 @@ print(f"Schema: {ds_filtered.schema()}")
 from lance_ray import add_columns
 import pyarrow as pa
 
+
 def add_computed_column(batch: pa.RecordBatch) -> pa.RecordBatch:
     df = batch.to_pandas()
-    df['computed'] = df['value'] * 2 + df['id']
+    df["computed"] = df["value"] * 2 + df["id"]
     return pa.RecordBatch.from_pandas(df[["computed"]])
 
-add_columns(
-    uri="sample_dataset.lance",
-    transform=add_computed_column,
-    concurrency=4
-)
+
+add_columns(uri="sample_dataset.lance", transform=add_computed_column, concurrency=4)
 ```
 
 ## Using Namespace
@@ -110,7 +106,7 @@ from lance_ray import read_lance, write_lance
 # Initialize Ray
 ray.init()
 
-# Connect to AWS Glue catalog 
+# Connect to AWS Glue catalog
 # using the default account and region in the current AWS environment
 namespace = ln.connect("glue", {})
 
@@ -119,10 +115,10 @@ data = ray.data.range(1000).map(lambda row: {"id": row["id"], "value": row["id"]
 
 # Write to Lance format using metadata catalog
 write_lance(
-    data, 
-    uri="s3://my-bucket/my-table", 
-    namespace=namespace, 
-    table_id=["default", "my_table"]
+    data,
+    uri="s3://my-bucket/my-table",
+    namespace=namespace,
+    table_id=["default", "my_table"],
 )
 
 # Read Lance dataset back using metadata catalog

lance_ray/datasource.py

@@ -201,18 +201,7 @@ def get_read_tasks(self, parallelism: int, **kwargs) -> list[ReadTask]:
                 )
 
             read_task = ReadTask(
-                lambda fids=fragment_ids,
-                uri=dataset_uri,
-                version=dataset_version,
-                storage_options=dataset_storage_options,
-                manifest=serialized_manifest,
-                ns_impl=namespace_impl,
-                ns_props=namespace_properties,
-                tbl_id=table_id,
-                base_params=base_store_params,
-                scanner_options=self._scanner_options,
-                retry_params=self._retry_params,
-                with_metadata=self._with_metadata: (
+                lambda fids=fragment_ids, uri=dataset_uri, version=dataset_version, storage_options=dataset_storage_options, manifest=serialized_manifest, ns_impl=namespace_impl, ns_props=namespace_properties, tbl_id=table_id, base_params=base_store_params, scanner_options=self._scanner_options, retry_params=self._retry_params, with_metadata=self._with_metadata: (
                     _read_fragments_with_retry(
                         fids,
                         uri,

pyproject.toml

@@ -24,7 +24,7 @@ classifiers = [
 
 dependencies = [
     "ray[data]>=2.41.0",
-    "pylance>=9.0.0",
+    "pylance>=10.0.0b5",
     "lance-namespace",
     "packaging",
     "pyarrow>=17.0.0",