chore: update rust toolchain and dependencies (#854)

Signed-off-by: Runji Wang <[email protected]>

---------

Signed-off-by: Runji Wang <[email protected]>

Cargo.toml

@@ -23,14 +23,14 @@ async-recursion = "1"
 async-stream = "0.3"
 async-trait = "0.1"
 binary-heap-plus = "0.5"
-bit-set = "0.5"
+bit-set = "0.8"
 bitvec = { version = "1", features = ["serde"] }
 btreemultimap = "0.1"
 bytes = "1"
 chrono = { version = "0.4", features = ["clock", "std", "wasmbind"] }
 clap = { version = "4", features = ["derive"] }
 comfy-table = { version = "7", default-features = false }
-console-subscriber = "0.2"
+console-subscriber = "0.4"
 crc32fast = "1"
 csv = "1"
 dirs = "5"
@@ -44,31 +44,31 @@ humantime = "2"
 indicatif = "0.17"
 indoc = "2"
 iter-chunks = "0.2"
-itertools = "0.12"
+itertools = "0.13"
 minitrace = { version = "0.6", features = ["enable"] }
 moka = { version = "0.12", features = ["future"] }
 num-traits = "0.2"
 ordered-float = { version = "4", features = ["serde"] }
 parking_lot = "0.12"
-parse-display = "0.9"
+parse-display = "0.10"
 paste = "1"
 pgwire = "0.20"
 pin-project = "1"
 pretty-xmlish = "0.1"
-prost = "0.12"
-pyo3 = { version = "0.21", features = ["extension-module"], optional = true }
+prost = "0.13"
+pyo3 = { version = "0.22", features = ["extension-module"], optional = true }
 ref-cast = "1.0"
 regex = "1"
 risinglight_proto = "0.2"
 rust_decimal = "1"
-rustyline = "14"
+rustyline = "15"
 serde = { version = "1", features = ["derive", "rc"] }
 serde_json = "1"
 smallvec = { version = "1", features = ["serde"] }
-sqllogictest = "0.20"
+sqllogictest = "0.23"
 sqlparser = { version = "0.45", features = ["serde"] }
-thiserror = "1"
-tikv-jemallocator = { version = "0.5", optional = true, features = [
+thiserror = "2"
+tikv-jemallocator = { version = "0.6", optional = true, features = [
     "disable_initial_exec_tls",
 ] }
 tokio = { version = "1", features = ["full"] }
@@ -81,13 +81,13 @@ tracing-subscriber = { version = "0.3", features = [
 [dev-dependencies]
 criterion = { version = "0.5", features = ["async_tokio"] }
 glob = "0.3"
-libtest-mimic = "0.7"
-sqlplannertest = "0.1"
+libtest-mimic = "0.8"
+sqlplannertest = "0.2"
 tempfile = "3"
 test-case = "3"
 
 [build-dependencies]
-pyo3-build-config = { version = "0.21", optional = true }
+pyo3-build-config = { version = "0.22", optional = true }
 
 [[test]]
 name = "sqllogictest"

proto/Cargo.toml

@@ -11,7 +11,7 @@ keywords = ["sql", "database", "embedded", "cli"]
 
 [dependencies]
 bytes = "1"
-prost = "0.12"
+prost = "0.13"
 
 [build-dependencies]
-prost-build = "0.12"
+prost-build = "0.13"

rust-toolchain

@@ -1 +1 @@
-nightly-2024-02-04
+nightly-2024-11-22

src/array/primitive_array.rs

@@ -192,7 +192,7 @@ impl PrimitiveArray<bool> {
 impl PrimitiveArray<Decimal> {
     /// Rescale the decimals.
     pub fn rescale(&mut self, scale: u8) {
-        for v in self.data.iter_mut() {
+        for v in &mut self.data {
             v.rescale(scale as u32);
         }
     }

src/binder/mod.rs

@@ -252,15 +252,15 @@ pub fn bind_header(mut chunk: array::Chunk, stmt: &Statement) -> array::Chunk {
 #[derive(Debug, Default)]
 struct Context {
     /// Defined CTEs.
-    /// cte_name -> (query_id, column_alias -> id)
+    /// `cte_name` -> (`query_id`, `column_alias` -> id)
     ctes: HashMap<String, (Id, HashMap<String, Id>)>,
     /// Table aliases that can be accessed from the current query.
     table_aliases: HashSet<String>,
     /// Column aliases that can be accessed from the current query.
-    /// column_alias -> (table_alias -> id)
+    /// `column_alias` -> (`table_alias` -> id)
     column_aliases: HashMap<String, HashMap<String, Id>>,
     /// Column aliases that can be accessed from the outside query.
-    /// column_alias -> id
+    /// `column_alias` -> id
     output_aliases: HashMap<String, Id>,
 }
 

src/executor/copy_from_file.rs

@@ -94,7 +94,7 @@ impl CopyFromFileExecutor {
                 return Err(Error::length_mismatch(column_count, record.len()));
             }
 
-            size_count += record.as_slice().as_bytes().len();
+            size_count += record.as_slice().len();
 
             // push a raw str row and send it if necessary
             if let Some(chunk) = chunk_builder.push_str_row(record.iter())? {

src/executor/evaluator.rs

@@ -211,22 +211,6 @@ impl<'a> Evaluator<'a> {
 
     /// Evaluate the aggregation.
     fn eval_agg(&self, state: AggState, chunk: &DataChunk) -> Result<AggState, ConvertError> {
-        impl DataValue {
-            fn add(self, other: Self) -> Self {
-                if self.is_null() {
-                    other
-                } else {
-                    self + other
-                }
-            }
-            fn or(self, other: Self) -> Self {
-                if self.is_null() {
-                    other
-                } else {
-                    self
-                }
-            }
-        }
         use Expr::*;
         Ok(match state {
             AggState::Value(state) => AggState::Value(match self.node() {
@@ -314,3 +298,26 @@ impl AggState {
         }
     }
 }
+
+trait Ext {
+    fn add(self, other: Self) -> Self;
+    fn or(self, other: Self) -> Self;
+}
+
+impl Ext for DataValue {
+    fn add(self, other: Self) -> Self {
+        if self.is_null() {
+            other
+        } else {
+            self + other
+        }
+    }
+
+    fn or(self, other: Self) -> Self {
+        if self.is_null() {
+            other
+        } else {
+            self
+        }
+    }
+}

src/executor/hash_join.rs

@@ -107,7 +107,6 @@ impl<const T: JoinType> HashJoinExecutor<T> {
 pub struct HashSemiJoinExecutor {
     pub left_keys: RecExpr,
     pub right_keys: RecExpr,
-    pub left_types: Vec<DataType>,
     pub anti: bool,
 }
 

src/executor/mod.rs

@@ -481,7 +481,6 @@ impl<S: Storage> Builder<S> {
             HashSemiJoinExecutor {
                 left_keys: self.resolve_column_index(lkeys, left),
                 right_keys: self.resolve_column_index(rkeys, right),
-                left_types: self.plan_types(left).to_vec(),
                 anti,
             }
             .execute(self.build_id(left), self.build_id(right))

src/lib.rs

@@ -15,7 +15,6 @@
 #![feature(proc_macro_hygiene)]
 #![feature(trusted_len)]
 #![feature(adt_const_params)]
-#![feature(lazy_cell)]
 #![feature(iterator_try_collect)]
 #![feature(let_chains)]
 #![feature(impl_trait_in_assoc_type)]

src/main.rs

@@ -2,8 +2,6 @@
 
 //! A simple interactive shell of the database.
 
-#![feature(div_duration)]
-
 use std::fs::File;
 use std::path::PathBuf;
 use std::sync::Arc;

src/planner/explain.rs

@@ -377,7 +377,7 @@ impl<'a> Explain<'a> {
     }
 }
 
-impl<'a> fmt::Display for Explain<'a> {
+impl fmt::Display for Explain<'_> {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
         delegate_fmt(&self.pretty(), f, String::with_capacity(4096))
     }

src/planner/rules/expr.rs

@@ -241,7 +241,6 @@ mod tests {
     }
 
     egg::test_fn! {
-        #[cfg_attr(feature = "simd", ignore)] // FIXME: 'attempt to divide by zero'
         constant_folding,
         rules(),
         "(* (- (+ 1 2) 4) (/ 6 2))" => "-3",

src/planner/rules/mod.rs

@@ -118,7 +118,7 @@ impl Analysis<Expr> for ExprAnalysis {
         let merge_columns = merge_small_set(&mut to.columns, from.columns);
         let merge_schema = egg::merge_max(&mut to.schema, from.schema);
         let merge_rows = egg::merge_min(
-            unsafe { std::mem::transmute(&mut to.rows) },
+            unsafe { std::mem::transmute::<&mut f32, &mut F32>(&mut to.rows) },
             F32::from(from.rows),
         );
         let merge_order = egg::merge_max(&mut to.orderby, from.orderby);

src/storage/secondary/block/rle_block_iterator.rs

@@ -33,13 +33,13 @@ where
     /// Indicates current position in the rle block
     cur_row: usize,
 
-    /// Indicates how many rows get scanned for cur_row
+    /// Indicates how many rows get scanned for `cur_row`
     cur_scanned_count: usize,
 
     /// Indicates the number of rows in the rle block
     rle_row_count: usize,
 
-    /// Indicates the element of current row get from block_iter
+    /// Indicates the element of current row get from `block_iter`
     cur_element: Option<<A::Item as ToOwned>::Owned>,
 
     /// Indicates how many rows get scanned for this iterator
@@ -48,7 +48,7 @@ where
     /// Total count of elements in block
     row_count: usize,
 
-    /// If never_used is true, get an item from child iter in the beginning of next_batch()
+    /// If `never_used` is true, get an item from child iter in the beginning of `next_batch()`
     never_used: bool,
 }
 

src/storage/secondary/column/concrete_column_iterator.rs

@@ -25,16 +25,17 @@ pub trait BlockIteratorFactory<A: Array>: Send + Sync + 'static {
 
     /// Create a [`FakeBlockIterator`](super::super::block::FakeBlockIterator) from block index and
     /// seek to `start_pos`.
+    #[allow(dead_code)]
     fn get_fake_iterator(&self, index: &BlockIndex, start_pos: usize) -> Self::BlockIteratorImpl;
 }
 
 /// `ConcreteColumnIterator` Statistics
 #[derive(Debug, Default)]
 pub struct Statistics {
-    /// next_batch call times
+    /// `next_batch` call times
     next_batch_count: u32,
 
-    /// get_block call times
+    /// `get_block` call times
     fetched_block_count: u32,
 }
 
@@ -49,7 +50,7 @@ pub struct ConcreteColumnIterator<A: Array, F: BlockIteratorFactory<A>> {
     /// Block iterator.
     block_iterator: F::BlockIteratorImpl,
 
-    /// RowID of the current column.
+    /// `RowID` of the current column.
     current_row_id: u32,
 
     /// Indicates whether this iterator has finished or not.
@@ -58,7 +59,7 @@ pub struct ConcreteColumnIterator<A: Array, F: BlockIteratorFactory<A>> {
     /// The factory for creating iterators.
     factory: F,
 
-    /// Indicate whether current_block_iter is fake.
+    /// Indicate whether `current_block_iter` is fake.
     is_fake_iter: bool,
 
     /// Statistics which used for reporting.

src/storage/secondary/options.rs

@@ -61,7 +61,7 @@ pub struct StorageOptions {
     /// Encode type
     pub encode_type: EncodeType,
 
-    /// Whether record first_key of each block into block_index
+    /// Whether record `first_key` of each block into `block_index`
     pub record_first_key: bool,
 
     /// Whether to disable all disk operations, only for test use
@@ -117,7 +117,7 @@ pub struct ColumnBuilderOptions {
     /// Encode type
     pub encode_type: EncodeType,
 
-    /// Whether record first_key of each block
+    /// Whether record `first_key` of each block
     pub record_first_key: bool,
 }
 

src/storage/secondary/txn_iterator.rs

@@ -17,6 +17,7 @@ pub enum SecondaryIterator {
 }
 
 #[enum_dispatch(SecondaryIterator)]
+#[allow(dead_code)]
 pub trait SecondaryIteratorImpl {}
 
 /// An iterator over all data in a transaction.

src/storage/secondary/version_manager.rs

@@ -106,10 +106,10 @@ pub struct VersionManagerInner {
     /// MVCC structure for this, and only record changes compared with last epoch.
     status: HashMap<u64, Arc<Snapshot>>,
 
-    /// (TableId, RowSetId) -> Object mapping
+    /// (`TableId`, `RowSetId`) -> Object mapping
     rowsets: HashMap<(u32, u32), Arc<DiskRowset>>,
 
-    /// (TableId, DVId) -> Object mapping
+    /// (`TableId`, `DVId`) -> Object mapping
     dvs: HashMap<(u32, u64), Arc<DeleteVector>>,
 
     /// Reference count of each epoch.