chore: update lance dependency to v6.0.0-beta.6

Cargo.toml

@@ -13,40 +13,40 @@ categories = ["database-implementations"]
 rust-version = "1.91.0"
 
 [workspace.dependencies]
-lance = { "version" = "=6.0.0-beta.4", default-features = false, "tag" = "v6.0.0-beta.4", "git" = "https://github.com/lance-format/lance.git" }
+lance = { "version" = "=6.0.0-beta.6", default-features = false, "tag" = "v6.0.0-beta.6", "git" = "https://github.com/lance-format/lance.git" }
-lance-core = { "version" = "=6.0.0-beta.4", "tag" = "v6.0.0-beta.4", "git" = "https://github.com/lance-format/lance.git" }
+lance-core = { "version" = "=6.0.0-beta.6", "tag" = "v6.0.0-beta.6", "git" = "https://github.com/lance-format/lance.git" }
-lance-datagen = { "version" = "=6.0.0-beta.4", "tag" = "v6.0.0-beta.4", "git" = "https://github.com/lance-format/lance.git" }
+lance-datagen = { "version" = "=6.0.0-beta.6", "tag" = "v6.0.0-beta.6", "git" = "https://github.com/lance-format/lance.git" }
-lance-file = { "version" = "=6.0.0-beta.4", "tag" = "v6.0.0-beta.4", "git" = "https://github.com/lance-format/lance.git" }
+lance-file = { "version" = "=6.0.0-beta.6", "tag" = "v6.0.0-beta.6", "git" = "https://github.com/lance-format/lance.git" }
-lance-io = { "version" = "=6.0.0-beta.4", default-features = false, "tag" = "v6.0.0-beta.4", "git" = "https://github.com/lance-format/lance.git" }
+lance-io = { "version" = "=6.0.0-beta.6", default-features = false, "tag" = "v6.0.0-beta.6", "git" = "https://github.com/lance-format/lance.git" }
-lance-index = { "version" = "=6.0.0-beta.4", "tag" = "v6.0.0-beta.4", "git" = "https://github.com/lance-format/lance.git" }
+lance-index = { "version" = "=6.0.0-beta.6", "tag" = "v6.0.0-beta.6", "git" = "https://github.com/lance-format/lance.git" }
-lance-linalg = { "version" = "=6.0.0-beta.4", "tag" = "v6.0.0-beta.4", "git" = "https://github.com/lance-format/lance.git" }
+lance-linalg = { "version" = "=6.0.0-beta.6", "tag" = "v6.0.0-beta.6", "git" = "https://github.com/lance-format/lance.git" }
-lance-namespace = { "version" = "=6.0.0-beta.4", "tag" = "v6.0.0-beta.4", "git" = "https://github.com/lance-format/lance.git" }
+lance-namespace = { "version" = "=6.0.0-beta.6", "tag" = "v6.0.0-beta.6", "git" = "https://github.com/lance-format/lance.git" }
-lance-namespace-impls = { "version" = "=6.0.0-beta.4", default-features = false, "tag" = "v6.0.0-beta.4", "git" = "https://github.com/lance-format/lance.git" }
+lance-namespace-impls = { "version" = "=6.0.0-beta.6", default-features = false, "tag" = "v6.0.0-beta.6", "git" = "https://github.com/lance-format/lance.git" }
-lance-table = { "version" = "=6.0.0-beta.4", "tag" = "v6.0.0-beta.4", "git" = "https://github.com/lance-format/lance.git" }
+lance-table = { "version" = "=6.0.0-beta.6", "tag" = "v6.0.0-beta.6", "git" = "https://github.com/lance-format/lance.git" }
-lance-testing = { "version" = "=6.0.0-beta.4", "tag" = "v6.0.0-beta.4", "git" = "https://github.com/lance-format/lance.git" }
+lance-testing = { "version" = "=6.0.0-beta.6", "tag" = "v6.0.0-beta.6", "git" = "https://github.com/lance-format/lance.git" }
-lance-datafusion = { "version" = "=6.0.0-beta.4", "tag" = "v6.0.0-beta.4", "git" = "https://github.com/lance-format/lance.git" }
+lance-datafusion = { "version" = "=6.0.0-beta.6", "tag" = "v6.0.0-beta.6", "git" = "https://github.com/lance-format/lance.git" }
-lance-encoding = { "version" = "=6.0.0-beta.4", "tag" = "v6.0.0-beta.4", "git" = "https://github.com/lance-format/lance.git" }
+lance-encoding = { "version" = "=6.0.0-beta.6", "tag" = "v6.0.0-beta.6", "git" = "https://github.com/lance-format/lance.git" }
-lance-arrow = { "version" = "=6.0.0-beta.4", "tag" = "v6.0.0-beta.4", "git" = "https://github.com/lance-format/lance.git" }
+lance-arrow = { "version" = "=6.0.0-beta.6", "tag" = "v6.0.0-beta.6", "git" = "https://github.com/lance-format/lance.git" }
 ahash = "0.8"
 # Note that this one does not include pyarrow
-arrow = { version = "57.2", optional = false }
+arrow = { version = "58.1", optional = false }
-arrow-array = "57.2"
+arrow-array = "58.1"
-arrow-data = "57.2"
+arrow-data = "58.1"
-arrow-ipc = "57.2"
+arrow-ipc = "58.1"
-arrow-ord = "57.2"
+arrow-ord = "58.1"
-arrow-schema = "57.2"
+arrow-schema = "58.1"
-arrow-select = "57.2"
+arrow-select = "58.1"
-arrow-cast = "57.2"
+arrow-cast = "58.1"
 async-trait = "0"
-datafusion = { version = "52.1", default-features = false }
+datafusion = { version = "53.1", default-features = false }
-datafusion-catalog = "52.1"
+datafusion-catalog = "53.1"
-datafusion-common = { version = "52.1", default-features = false }
+datafusion-common = { version = "53.1", default-features = false }
-datafusion-execution = "52.1"
+datafusion-execution = "53.1"
-datafusion-expr = "52.1"
+datafusion-expr = "53.1"
-datafusion-functions = "52.1"
+datafusion-functions = "53.1"
-datafusion-physical-plan = "52.1"
+datafusion-physical-plan = "53.1"
-datafusion-physical-expr = "52.1"
+datafusion-physical-expr = "53.1"
-datafusion-sql = "52.1"
+datafusion-sql = "53.1"
 env_logger = "0.11"
 half = { "version" = "2.7.1", default-features = false, features = [
     "num-traits",

java/pom.xml

@@ -28,7 +28,7 @@
     <properties>
         <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
         <arrow.version>15.0.0</arrow.version>
-        <lance-core.version>6.0.0-beta.4</lance-core.version>
+        <lance-core.version>6.0.0-beta.6</lance-core.version>
         <spotless.skip>false</spotless.skip>
         <spotless.version>2.30.0</spotless.version>
         <spotless.java.googlejavaformat.version>1.7</spotless.java.googlejavaformat.version>

nodejs/Cargo.toml

@@ -16,7 +16,7 @@ crate-type = ["cdylib"]
 async-trait.workspace = true
 arrow-ipc.workspace = true
 arrow-array.workspace = true
-arrow-buffer = "57.2"
+arrow-buffer = "58.1"
 half.workspace = true
 arrow-schema.workspace = true
 env_logger.workspace = true

python/Cargo.toml

@@ -15,7 +15,7 @@ name = "_lancedb"
 crate-type = ["cdylib"]
 
 [dependencies]
-arrow = { version = "57.2", features = ["pyarrow"] }
+arrow = { workspace = true, features = ["pyarrow"] }
 async-trait = "0.1"
 bytes = "1"
 lancedb = { path = "../rust/lancedb", default-features = false }
@@ -25,8 +25,8 @@ lance-namespace-impls.workspace = true
 lance-io.workspace = true
 env_logger.workspace = true
 log.workspace = true
-pyo3 = { version = "0.26", features = ["extension-module", "abi3-py39"] }
+pyo3 = { version = "0.28", features = ["extension-module", "abi3-py39"] }
-pyo3-async-runtimes = { version = "0.26", features = [
+pyo3-async-runtimes = { version = "0.28", features = [
     "attributes",
     "tokio-runtime",
 ] }
@@ -38,7 +38,7 @@ snafu.workspace = true
 tokio = { version = "1.40", features = ["sync"] }
 
 [build-dependencies]
-pyo3-build-config = { version = "0.26", features = [
+pyo3-build-config = { version = "0.28", features = [
     "extension-module",
     "abi3-py39",
 ] }

python/src/expr.rs

@@ -17,7 +17,7 @@ use pyo3::{Bound, PyAny, PyResult, exceptions::PyValueError, prelude::*, pyfunct
 /// [`expr_lit`] and combined with the methods on this struct.  On the Python
 /// side a thin wrapper class (`lancedb.expr.Expr`) delegates to these methods
 /// and adds Python operator overloads.
-#[pyclass(name = "PyExpr")]
+#[pyclass(name = "PyExpr", from_py_object)]
 #[derive(Clone)]
 pub struct PyExpr(pub DfExpr);
 

python/src/header.rs

@@ -33,7 +33,7 @@ impl PyHeaderProvider {
                 Ok(headers_py) => {
                     // Convert Python dict to Rust HashMap
                     let bound_headers = headers_py.bind(py);
-                    let dict: &Bound<PyDict> = bound_headers.downcast().map_err(|e| {
+                    let dict: &Bound<PyDict> = bound_headers.cast().map_err(|e| {
                         format!("HeaderProvider.get_headers must return a dict: {}", e)
                     })?;
 

python/src/index.rs

@@ -22,7 +22,7 @@ pub fn class_name(ob: &'_ Bound<'_, PyAny>) -> PyResult<String> {
     let full_name = ob
         .getattr(intern!(ob.py(), "__class__"))?
         .getattr(intern!(ob.py(), "__name__"))?;
-    let full_name = full_name.downcast()?.to_string_lossy();
+    let full_name = full_name.cast()?.to_string_lossy();
 
     match full_name.rsplit_once('.') {
         Some((_, name)) => Ok(name.to_string()),

python/src/namespace.rs

@@ -183,7 +183,7 @@ async fn call_py_method_primitive<Req, Resp>(
 ) -> lance_core::Result<Resp>
 where
     Req: serde::Serialize + Send + 'static,
-    Resp: for<'py> pyo3::FromPyObject<'py> + Send + 'static,
+    Resp: for<'a, 'py> pyo3::FromPyObject<'a, 'py> + Send + 'static,
 {
     let request_json = serde_json::to_string(&request).map_err(|e| {
         lance_core::Error::io(format!(
@@ -203,7 +203,7 @@ where
 
             // Call the Python method
             let result = py_namespace.call_method1(py, method_name, (request_arg,))?;
-            let value: Resp = result.extract(py)?;
+            let value: Resp = result.extract(py).map_err(Into::into)?;
             Ok::<_, PyErr>(value)
         })
     })

python/src/permutation.rs

@@ -25,12 +25,12 @@ use pyo3_async_runtimes::tokio::future_into_py;
 
 fn table_from_py<'a>(table: Bound<'a, PyAny>) -> PyResult<Bound<'a, Table>> {
     if table.hasattr("_inner")? {
-        Ok(table.getattr("_inner")?.downcast_into::<Table>()?)
+        Ok(table.getattr("_inner")?.cast_into::<Table>()?)
     } else if table.hasattr("_table")? {
         Ok(table
             .getattr("_table")?
             .getattr("_inner")?
-            .downcast_into::<Table>()?)
+            .cast_into::<Table>()?)
     } else {
         Err(PyRuntimeError::new_err(
             "Provided table does not appear to be a Table or RemoteTable instance",
@@ -90,9 +90,9 @@ impl PyAsyncPermutationBuilder {
             database
                 .getattr("_conn")?
                 .getattr("_inner")?
-                .downcast_into::<Connection>()?
+                .cast_into::<Connection>()?
         } else {
-            database.getattr("_inner")?.downcast_into::<Connection>()?
+            database.getattr("_inner")?.cast_into::<Connection>()?
         };
         let database = conn.borrow().database()?;
         slf.modify(|builder| builder.persist(database, table_name))
@@ -243,7 +243,7 @@ impl PyPermutationReader {
         let Some(selection) = selection else {
             return Ok(Select::All);
         };
-        let selection = selection.downcast_into::<PyDict>()?;
+        let selection = selection.cast_into::<PyDict>()?;
         let selection = selection
             .iter()
             .map(|(key, value)| {

python/src/query.rs

@@ -22,6 +22,7 @@ use lancedb::query::{
     VectorQuery as LanceDbVectorQuery,
 };
 use lancedb::table::AnyQuery;
+use pyo3::Borrowed;
 use pyo3::Bound;
 use pyo3::IntoPyObject;
 use pyo3::PyAny;
@@ -43,9 +44,11 @@ use crate::util::parse_distance_type;
 use crate::{arrow::RecordBatchStream, util::PyLanceDB};
 use crate::{error::PythonErrorExt, index::class_name};
 
-impl FromPyObject<'_> for PyLanceDB<FtsQuery> {
+impl FromPyObject<'_, '_> for PyLanceDB<FtsQuery> {
-    fn extract_bound(ob: &Bound<'_, PyAny>) -> PyResult<Self> {
+    type Error = PyErr;
-        match class_name(ob)?.as_str() {
+
+    fn extract(ob: Borrowed<'_, '_, PyAny>) -> PyResult<Self> {
+        match class_name(&ob)?.as_str() {
             "MatchQuery" => {
                 let query = ob.getattr("query")?.extract()?;
                 let column = ob.getattr("column")?.extract()?;
@@ -424,7 +427,7 @@ impl Query {
                 "Query text is required for nearest_to_text",
             ))?;
 
-        let query = if let Ok(query_text) = fts_query.downcast::<PyString>() {
+        let query = if let Ok(query_text) = fts_query.cast::<PyString>() {
             let mut query_text = query_text.to_string();
             let columns = query
                 .get_item("columns")?
@@ -606,7 +609,7 @@ impl TakeQuery {
     }
 }
 
-#[pyclass]
+#[pyclass(from_py_object)]
 #[derive(Clone)]
 pub struct FTSQuery {
     inner: LanceDbQuery,
@@ -735,7 +738,7 @@ impl FTSQuery {
     }
 }
 
-#[pyclass]
+#[pyclass(from_py_object)]
 #[derive(Clone)]
 pub struct VectorQuery {
     inner: LanceDbVectorQuery,

python/src/session.rs

@@ -11,7 +11,7 @@ use pyo3::{PyResult, pyclass, pymethods};
 /// Sessions allow you to configure cache sizes for index and metadata caches,
 /// which can significantly impact memory use and performance. They can
 /// also be re-used across multiple connections to share the same cache state.
-#[pyclass]
+#[pyclass(from_py_object)]
 #[derive(Clone)]
 pub struct Session {
     pub(crate) inner: Arc<LanceSession>,

python/src/table.rs

@@ -29,7 +29,7 @@ use pyo3_async_runtimes::tokio::future_into_py;
 mod scannable;
 
 /// Statistics about a compaction operation.
-#[pyclass(get_all)]
+#[pyclass(get_all, from_py_object)]
 #[derive(Clone, Debug)]
 pub struct CompactionStats {
     /// The number of fragments removed
@@ -43,7 +43,7 @@ pub struct CompactionStats {
 }
 
 /// Statistics about a cleanup operation
-#[pyclass(get_all)]
+#[pyclass(get_all, from_py_object)]
 #[derive(Clone, Debug)]
 pub struct RemovalStats {
     /// The number of bytes removed
@@ -53,7 +53,7 @@ pub struct RemovalStats {
 }
 
 /// Statistics about an optimize operation
-#[pyclass(get_all)]
+#[pyclass(get_all, from_py_object)]
 #[derive(Clone, Debug)]
 pub struct OptimizeStats {
     /// Statistics about the compaction operation
@@ -62,7 +62,7 @@ pub struct OptimizeStats {
     pub prune: RemovalStats,
 }
 
-#[pyclass(get_all)]
+#[pyclass(get_all, from_py_object)]
 #[derive(Clone, Debug)]
 pub struct UpdateResult {
     pub rows_updated: u64,
@@ -88,7 +88,7 @@ impl From<lancedb::table::UpdateResult> for UpdateResult {
     }
 }
 
-#[pyclass(get_all)]
+#[pyclass(get_all, from_py_object)]
 #[derive(Clone, Debug)]
 pub struct AddResult {
     pub version: u64,
@@ -109,7 +109,7 @@ impl From<lancedb::table::AddResult> for AddResult {
     }
 }
 
-#[pyclass(get_all)]
+#[pyclass(get_all, from_py_object)]
 #[derive(Clone, Debug)]
 pub struct DeleteResult {
     pub num_deleted_rows: u64,
@@ -135,7 +135,7 @@ impl From<lancedb::table::DeleteResult> for DeleteResult {
     }
 }
 
-#[pyclass(get_all)]
+#[pyclass(get_all, from_py_object)]
 #[derive(Clone, Debug)]
 pub struct MergeResult {
     pub version: u64,
@@ -171,7 +171,7 @@ impl From<lancedb::table::MergeResult> for MergeResult {
     }
 }
 
-#[pyclass(get_all)]
+#[pyclass(get_all, from_py_object)]
 #[derive(Clone, Debug)]
 pub struct AddColumnsResult {
     pub version: u64,
@@ -192,7 +192,7 @@ impl From<lancedb::table::AddColumnsResult> for AddColumnsResult {
     }
 }
 
-#[pyclass(get_all)]
+#[pyclass(get_all, from_py_object)]
 #[derive(Clone, Debug)]
 pub struct AlterColumnsResult {
     pub version: u64,
@@ -213,7 +213,7 @@ impl From<lancedb::table::AlterColumnsResult> for AlterColumnsResult {
     }
 }
 
-#[pyclass(get_all)]
+#[pyclass(get_all, from_py_object)]
 #[derive(Clone, Debug)]
 pub struct DropColumnsResult {
     pub version: u64,

python/src/table/scannable.rs

@@ -14,7 +14,7 @@ use lancedb::{
     arrow::{SendableRecordBatchStream, SimpleRecordBatchStream},
     data::scannable::Scannable,
 };
-use pyo3::{FromPyObject, Py, PyAny, Python, types::PyAnyMethods};
+use pyo3::{Borrowed, FromPyObject, Py, PyAny, Python, types::PyAnyMethods};
 
 /// Adapter that implements Scannable for a Python reader factory callable.
 ///
@@ -126,8 +126,10 @@ impl Scannable for PyScannable {
     }
 }
 
-impl<'py> FromPyObject<'py> for PyScannable {
+impl FromPyObject<'_, '_> for PyScannable {
-    fn extract_bound(ob: &pyo3::Bound<'py, PyAny>) -> pyo3::PyResult<Self> {
+    type Error = pyo3::PyErr;
+
+    fn extract(ob: Borrowed<'_, '_, PyAny>) -> pyo3::PyResult<Self> {
         // Convert from Scannable dataclass.
         let schema: PyArrowType<Schema> = ob.getattr("schema")?.extract()?;
         let schema = Arc::new(schema.0);

rust/lancedb/src/remote/table/insert.rs

@@ -43,7 +43,7 @@ pub struct RemoteInsertExec<S: HttpSend = Sender> {
     client: RestfulLanceDbClient<S>,
     input: Arc<dyn ExecutionPlan>,
     overwrite: bool,
-    properties: PlanProperties,
+    properties: Arc<PlanProperties>,
     add_result: Arc<Mutex<Option<AddResult>>>,
     metrics: ExecutionPlanMetricsSet,
     upload_id: Option<String>,
@@ -105,12 +105,12 @@ impl<S: HttpSend + 'static> RemoteInsertExec<S> {
             1
         };
         let schema = COUNT_SCHEMA.clone();
-        let properties = PlanProperties::new(
+        let properties = Arc::new(PlanProperties::new(
             EquivalenceProperties::new(schema),
             datafusion_physical_plan::Partitioning::UnknownPartitioning(num_partitions),
             datafusion_physical_plan::execution_plan::EmissionType::Final,
             datafusion_physical_plan::execution_plan::Boundedness::Bounded,
-        );
+        ));
 
         Self {
             table_name,
@@ -232,7 +232,7 @@ impl<S: HttpSend + 'static> ExecutionPlan for RemoteInsertExec<S> {
         self
     }
 
-    fn properties(&self) -> &PlanProperties {
+    fn properties(&self) -> &Arc<PlanProperties> {
         &self.properties
     }
 

rust/lancedb/src/table/datafusion.rs

@@ -39,21 +39,26 @@ use lance_index::scalar::FullTextSearchQuery;
 struct MetadataEraserExec {
     input: Arc<dyn ExecutionPlan>,
     schema: Arc<ArrowSchema>,
-    properties: PlanProperties,
+    properties: Arc<PlanProperties>,
 }
 
 impl MetadataEraserExec {
     fn compute_properties_from_input(
         input: &Arc<dyn ExecutionPlan>,
         schema: &Arc<ArrowSchema>,
-    ) -> PlanProperties {
+    ) -> Arc<PlanProperties> {
         let input_properties = input.properties();
         let eq_properties = input_properties
             .eq_properties
             .clone()
             .with_new_schema(schema.clone())
             .unwrap();
-        input_properties.clone().with_eq_properties(eq_properties)
+        Arc::new(
+            input_properties
+                .as_ref()
+                .clone()
+                .with_eq_properties(eq_properties),
+        )
     }
 
     fn new(input: Arc<dyn ExecutionPlan>) -> Self {
@@ -87,7 +92,7 @@ impl ExecutionPlan for MetadataEraserExec {
         self
     }
 
-    fn properties(&self) -> &PlanProperties {
+    fn properties(&self) -> &Arc<PlanProperties> {
         &self.properties
     }
 

rust/lancedb/src/table/datafusion/insert.rs

@@ -81,7 +81,7 @@ pub struct InsertExec {
     dataset: Arc<Dataset>,
     input: Arc<dyn ExecutionPlan>,
     write_params: WriteParams,
-    properties: PlanProperties,
+    properties: Arc<PlanProperties>,
     partial_transactions: Arc<Mutex<Vec<Transaction>>>,
     metrics: ExecutionPlanMetricsSet,
 }
@@ -95,12 +95,12 @@ impl InsertExec {
     ) -> Self {
         let schema = COUNT_SCHEMA.clone();
         let num_partitions = input.output_partitioning().partition_count();
-        let properties = PlanProperties::new(
+        let properties = Arc::new(PlanProperties::new(
             EquivalenceProperties::new(schema),
             Partitioning::UnknownPartitioning(num_partitions),
             EmissionType::Final,
             Boundedness::Bounded,
-        );
+        ));
 
         Self {
             ds_wrapper,
@@ -136,7 +136,7 @@ impl ExecutionPlan for InsertExec {
         self
     }
 
-    fn properties(&self) -> &PlanProperties {
+    fn properties(&self) -> &Arc<PlanProperties> {
         &self.properties
     }
 

rust/lancedb/src/table/datafusion/scannable_exec.rs

@@ -20,7 +20,7 @@ pub(crate) struct ScannableExec {
     // We don't require Scannable to be Sync, so we wrap it in a Mutex to allow safe concurrent access.
     source: Mutex<Box<dyn Scannable>>,
     num_rows: Option<usize>,
-    properties: PlanProperties,
+    properties: Arc<PlanProperties>,
     tracker: Option<Arc<WriteProgressTracker>>,
 }
 
@@ -37,12 +37,12 @@ impl ScannableExec {
     pub fn new(source: Box<dyn Scannable>, tracker: Option<Arc<WriteProgressTracker>>) -> Self {
         let schema = source.schema();
         let eq_properties = EquivalenceProperties::new(schema);
-        let properties = PlanProperties::new(
+        let properties = Arc::new(PlanProperties::new(
             eq_properties,
             Partitioning::UnknownPartitioning(1),
             EmissionType::Incremental,
             datafusion_physical_plan::execution_plan::Boundedness::Bounded,
-        );
+        ));
 
         let num_rows = source.num_rows();
         let source = Mutex::new(source);
@@ -70,7 +70,7 @@ impl ExecutionPlan for ScannableExec {
         self
     }
 
-    fn properties(&self) -> &PlanProperties {
+    fn properties(&self) -> &Arc<PlanProperties> {
         &self.properties
     }