Revert "smallvec -> tinyvec (keep the commit around, but need MaybeUn…

…init to be viable)"

This reverts commit 10521ea.

crates/re_arrow_store/Cargo.toml

@@ -44,9 +44,9 @@ indent = "0.1"
 itertools = { workspace = true }
 nohash-hasher = "0.2"
 parking_lot.workspace = true
+smallvec = { version = "1.0", features = ["const_generics", "union"] }
 static_assertions = "1.1"
 thiserror.workspace = true
-tinyvec = { version = "1.6", features = ["alloc", "rustc_1_55"] }
 
 # Native dependencies:
 [target.'cfg(not(target_arch = "wasm32"))'.dependencies]

crates/re_arrow_store/src/store.rs

@@ -4,7 +4,7 @@ use std::sync::atomic::AtomicU64;
 use ahash::HashMap;
 use arrow2::array::Int64Array;
 use arrow2::datatypes::{DataType, TimeUnit};
-use tinyvec::TinyVec;
+use smallvec::SmallVec;
 
 use nohash_hasher::{IntMap, IntSet};
 use parking_lot::RwLock;
@@ -62,7 +62,7 @@ impl DataStoreConfig {
 
 // ---
 
-pub type InsertIdVec = TinyVec<[u64; 4]>;
+pub type InsertIdVec = SmallVec<[u64; 4]>;
 
 /// Keeps track of datatype information for all component types that have been written to the store
 /// so far.

crates/re_arrow_store/src/store_read.rs

@@ -6,7 +6,7 @@ use re_log::trace;
 use re_log_types::{
     ComponentName, DataCell, EntityPath, MsgId, RowId, TimeInt, TimePoint, TimeRange, Timeline,
 };
-use tinyvec::TinyVec;
+use smallvec::SmallVec;
 
 use crate::{DataStore, IndexedBucket, IndexedBucketInner, IndexedTable, PersistentIndexedTable};
 
@@ -899,8 +899,8 @@ impl IndexedBucketInner {
 
         {
             crate::profile_scope!("control");
-            fn reshuffle_control_column<T: Default + Copy, const N: usize>(
-                column: &mut TinyVec<[T; N]>,
+            fn reshuffle_control_column<T: Copy, const N: usize>(
+                column: &mut SmallVec<[T; N]>,
                 swaps: &[(usize, usize)],
             ) {
                 let source = {
@@ -933,7 +933,7 @@ impl IndexedBucketInner {
                 {
                     crate::profile_scope!("rotate");
                     for (from, to) in swaps.iter().copied() {
-                        column[to] = source[from].take(); // TODO: why take?
+                        column[to] = source[from].take();
                     }
                 }
             }

crates/re_arrow_store/src/store_write.rs

@@ -2,7 +2,7 @@ use arrow2::datatypes::DataType;
 use itertools::Itertools as _;
 use nohash_hasher::{IntMap, IntSet};
 use parking_lot::RwLock;
-use tinyvec::TinyVec;
+use smallvec::SmallVec;
 
 use re_log::{debug, trace};
 use re_log_types::{
@@ -486,6 +486,19 @@ impl IndexedBucket {
         // Used in debug builds to assert that we've left everything in a sane state.
         let _total_rows = col_time1.len();
 
+        fn split_off_column<T: Copy, const N: usize>(
+            column: &mut SmallVec<[T; N]>,
+            split_idx: usize,
+        ) -> SmallVec<[T; N]> {
+            if split_idx >= column.len() {
+                return SmallVec::default();
+            }
+
+            let second_half = SmallVec::from_slice(&column[split_idx..]);
+            column.truncate(split_idx);
+            second_half
+        }
+
         let (min2, bucket2) = {
             let split_idx = find_split_index(col_time1).expect("must be splittable at this point");
 
@@ -495,13 +508,13 @@ impl IndexedBucket {
                     // this updates `time_range1` in-place!
                     split_time_range_off(split_idx, col_time1, time_range1),
                     // this updates `col_time1` in-place!
-                    col_time1.split_off(split_idx),
+                    split_off_column(col_time1, split_idx),
                     // this updates `col_insert_id1` in-place!
-                    (!col_insert_id1.is_empty()).then(|| col_insert_id1.split_off(split_idx)),
+                    split_off_column(col_insert_id1, split_idx),
                     // this updates `col_row_id1` in-place!
-                    col_row_id1.split_off(split_idx),
+                    split_off_column(col_row_id1, split_idx),
                     // this updates `col_num_instances1` in-place!
-                    col_num_instances1.split_off(split_idx),
+                    split_off_column(col_num_instances1, split_idx),
                 )
             };
 
@@ -512,12 +525,12 @@ impl IndexedBucket {
                     .iter_mut()
                     .map(|(name, column1)| {
                         if split_idx >= column1.len() {
-                            return (*name, DataCellColumn(TinyVec::default()));
+                            return (*name, DataCellColumn(SmallVec::default()));
                         }
 
                         // this updates `column1` in-place!
                         let column2 = DataCellColumn({
-                            let second_half = TinyVec::from(&column1.0[split_idx..]);
+                            let second_half = SmallVec::from(&column1.0[split_idx..]);
                             column1.0.truncate(split_idx);
                             second_half
                         });
@@ -533,7 +546,7 @@ impl IndexedBucket {
                     is_sorted: true,
                     time_range: time_range2,
                     col_time: col_time2,
-                    col_insert_id: col_insert_id2.unwrap_or_default(),
+                    col_insert_id: col_insert_id2,
                     col_row_id: col_row_id2,
                     col_num_instances: col_num_instances2,
                     columns: columns2,

crates/re_log_types/Cargo.toml

@@ -77,12 +77,12 @@ ndarray.workspace = true
 nohash-hasher = "0.2"
 num-derive = "0.3"
 num-traits = "0.2"
+smallvec = { version = "1.10", features = ["union"] }
 thiserror.workspace = true
 time = { workspace = true, default-features = false, features = [
   "formatting",
   "macros",
 ] }
-tinyvec = { version = "1.6", features = ["alloc", "rustc_1_55"] }
 typenum = "1.15"
 uuid = { version = "1.1", features = ["serde", "v4", "js"] }
 

crates/re_log_types/src/component_types/msg_id.rs

@@ -30,12 +30,6 @@ pub type TableId = MsgId;
 
 pub type RowId = MsgId;
 
-impl Default for MsgId {
-    fn default() -> Self {
-        Self::ZERO
-    }
-}
-
 impl std::fmt::Debug for MsgId {
     fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
         write!(f, "{:032X}", self.0.as_u128())

crates/re_log_types/src/data_cell.rs

@@ -107,15 +107,6 @@ pub struct DataCell {
     pub(crate) values: Box<dyn arrow2::array::Array>,
 }
 
-impl Default for DataCell {
-    fn default() -> Self {
-        Self {
-            name: "".into(),
-            values: arrow2::array::NullArray::new_empty(arrow2::datatypes::DataType::Null).boxed(),
-        }
-    }
-}
-
 // TODO(cmc): We should be able to build a cell from non-reference types.
 // TODO(#1619): We shouldn't have to specify the component name separately, this should be
 // part of the metadata by using an extension.

crates/re_log_types/src/data_row.rs

@@ -1,6 +1,6 @@
 use ahash::HashSetExt;
 use nohash_hasher::IntSet;
-use tinyvec::TinyVec;
+use smallvec::SmallVec;
 
 use crate::{ComponentName, DataCell, DataCellError, DataTable, EntityPath, RowId, TimePoint};
 
@@ -44,7 +44,7 @@ pub type DataRowResult<T> = ::std::result::Result<T, DataRowError>;
 
 // ---
 
-pub type DataCellVec = TinyVec<[DataCell; 4]>;
+type DataCellVec = SmallVec<[DataCell; 4]>;
 
 /// A row's worth of [`DataCell`]s: a collection of independent [`DataCell`]s with different
 /// underlying datatypes and pointing to different parts of the heap.

crates/re_log_types/src/data_table.rs

@@ -1,7 +1,7 @@
 use ahash::HashMap;
 use itertools::Itertools as _;
 use nohash_hasher::{IntMap, IntSet};
-use tinyvec::TinyVec;
+use smallvec::SmallVec;
 
 use crate::{
     ArrowMsg, ComponentName, DataCell, DataCellError, DataRow, DataRowError, EntityPath, RowId,
@@ -41,17 +41,17 @@ pub type DataTableResult<T> = ::std::result::Result<T, DataTableError>;
 
 // ---
 
-pub type RowIdVec = TinyVec<[RowId; 4]>;
+pub type RowIdVec = SmallVec<[RowId; 4]>;
 
-pub type TimePointVec = TinyVec<[TimePoint; 4]>;
+pub type TimePointVec = SmallVec<[TimePoint; 4]>;
 
-pub type ErasedTimeVec = TinyVec<[i64; 4]>;
+pub type ErasedTimeVec = SmallVec<[i64; 4]>;
 
-pub type EntityPathVec = TinyVec<[EntityPath; 4]>;
+pub type EntityPathVec = SmallVec<[EntityPath; 4]>;
 
-pub type NumInstancesVec = TinyVec<[u32; 4]>;
+pub type NumInstancesVec = SmallVec<[u32; 4]>;
 
-pub type DataCellOptVec = TinyVec<[Option<DataCell>; 4]>;
+pub type DataCellOptVec = SmallVec<[Option<DataCell>; 4]>;
 
 /// A column's worth of [`DataCell`]s: a sparse collection of [`DataCell`]s that share the same
 /// underlying type and likely point to shared, contiguous memory.
@@ -97,7 +97,7 @@ impl std::ops::IndexMut<usize> for DataCellColumn {
 impl DataCellColumn {
     #[inline]
     pub fn empty(num_rows: usize) -> Self {
-        Self(TinyVec::Heap(vec![None; num_rows]))
+        Self(smallvec::smallvec![None; num_rows])
     }
 }
 
@@ -317,7 +317,7 @@ impl DataTable {
         for component in components {
             columns.insert(
                 component,
-                DataCellColumn(TinyVec::Heap(vec![None; column.len()])),
+                DataCellColumn(smallvec::smallvec![None; column.len()]),
             );
         }
 

crates/re_log_types/src/lib.rs

@@ -47,7 +47,7 @@ pub use self::component_types::{EncodedMesh3D, Mesh3D, MeshFormat, MeshId, RawMe
 pub use self::component_types::{MsgId, RowId, TableId};
 pub use self::data::*;
 pub use self::data_cell::{DataCell, DataCellError, DataCellResult};
-pub use self::data_row::{DataCellVec, DataRow, DataRowError, DataRowResult};
+pub use self::data_row::{DataRow, DataRowError, DataRowResult};
 pub use self::data_table::{
     DataCellColumn, DataCellOptVec, DataTable, DataTableError, DataTableResult, EntityPathVec,
     ErasedTimeVec, NumInstancesVec, RowIdVec, TimePointVec, COLUMN_ENTITY_PATH,

crates/re_log_types/src/path/entity_path.rs

@@ -85,12 +85,6 @@ pub struct EntityPath {
     path: Arc<EntityPathImpl>,
 }
 
-impl Default for EntityPath {
-    fn default() -> Self {
-        Self::root()
-    }
-}
-
 impl EntityPath {
     #[inline]
     pub fn root() -> Self {