From 497d250d8aaf3224ac1d1b63532afd3342d2b718 Mon Sep 17 00:00:00 2001
From: Eric Coissac <eric@coissac.eu>
Date: Wed, 17 Jun 2026 09:32:12 +0200
Subject: [PATCH] refactor: replace byte-level bit iteration with 64-bit words

Refactor `BitIter` to process `u64` chunks using word-aligned shifts instead of byte-level operations. Introduce a dedicated `MemoryBitIter` for `MemoryBitVec`, updating its `iter()` and `IntoIterator` implementations accordingly. Hide `MemoryBitIter` from the public API to narrow the crate's interface, while leveraging explicit alignment guarantees for safer and more efficient bit extraction.
---
 src/obicompactvec/src/bitvec.rs    | 25 ++++++----------------
 src/obicompactvec/src/lib.rs       |  2 +-
 src/obicompactvec/src/memoryvec.rs | 34 +++++-------------------------
 3 files changed, 13 insertions(+), 48 deletions(-)

diff --git a/src/obicompactvec/src/bitvec.rs b/src/obicompactvec/src/bitvec.rs
index 9d78e88..1d91b10 100644
--- a/src/obicompactvec/src/bitvec.rs
+++ b/src/obicompactvec/src/bitvec.rs
@@ -65,12 +65,7 @@ impl PersistentBitVec {
         (self.mmap[HEADER_SIZE + (slot >> 3)] >> (slot & 7)) & 1 != 0
     }
 
-    // Used by iter() and get(): exact byte window, no padding.
-    fn data_bytes(&self) -> &[u8] {
-        &self.mmap[HEADER_SIZE..HEADER_SIZE + self.n.div_ceil(8)]
-    }
-
-    // Bulk word view. SAFETY: mmap is page-aligned, HEADER_SIZE=16 is divisible by 8,
+    // SAFETY: mmap is page-aligned, HEADER_SIZE=16 is divisible by 8,
     // so &mmap[HEADER_SIZE] is u64-aligned. Slice length is n_words * 8 bytes.
     fn data_words(&self) -> &[u64] {
         let nw = n_words(self.n);
@@ -79,11 +74,7 @@ impl PersistentBitVec {
     }
 
     pub fn iter(&self) -> BitIter<'_> {
-        BitIter {
-            bytes: self.data_bytes(),
-            slot: 0,
-            n: self.n,
-        }
+        BitIter { words: self.data_words(), slot: 0, n: self.n }
     }
 }
 
@@ -96,9 +87,9 @@ impl<'a> IntoIterator for &'a PersistentBitVec {
 }
 
 pub struct BitIter<'a> {
-    bytes: &'a [u8],
-    slot: usize,
-    n: usize,
+    pub(crate) words: &'a [u64],
+    pub(crate) slot: usize,
+    pub(crate) n: usize,
 }
 
 impl ExactSizeIterator for BitIter<'_> {}
@@ -107,10 +98,8 @@ impl Iterator for BitIter<'_> {
     type Item = bool;
 
     fn next(&mut self) -> Option<bool> {
-        if self.slot >= self.n {
-            return None;
-        }
-        let v = (self.bytes[self.slot >> 3] >> (self.slot & 7)) & 1 != 0;
+        if self.slot >= self.n { return None; }
+        let v = (self.words[self.slot >> 6] >> (self.slot & 63)) & 1 != 0;
         self.slot += 1;
         Some(v)
     }
diff --git a/src/obicompactvec/src/lib.rs b/src/obicompactvec/src/lib.rs
index ced509b..3a5f1c4 100644
--- a/src/obicompactvec/src/lib.rs
+++ b/src/obicompactvec/src/lib.rs
@@ -16,7 +16,7 @@ pub use builder::PersistentCompactIntVecBuilder;
 pub use intmatrix::{PersistentCompactIntMatrix, PersistentCompactIntMatrixBuilder, pack_compact_int_matrix};
 pub use layer_meta::LayerMeta;
 pub use memoryintvec::{MemoryIntIter, MemoryIntVec};
-pub use memoryvec::{MemoryBitIter, MemoryBitVec};
+pub use memoryvec::MemoryBitVec;
 pub use reader::PersistentCompactIntVec;
 pub use traits::{BitPartials, BitSlice, BitSliceMut, BitToInt, ColumnWeights, CountPartials, IntSlice, IntSliceMut, IntToBit};
 
diff --git a/src/obicompactvec/src/memoryvec.rs b/src/obicompactvec/src/memoryvec.rs
index 3076325..fef0960 100644
--- a/src/obicompactvec/src/memoryvec.rs
+++ b/src/obicompactvec/src/memoryvec.rs
@@ -2,7 +2,7 @@ use std::io;
 use std::ops::{BitAnd, BitAndAssign, BitOr, BitOrAssign, BitXor, BitXorAssign, Not};
 use std::path::Path;
 
-use crate::bitvec::{PersistentBitVecBuilder, n_words};
+use crate::bitvec::{BitIter, PersistentBitVecBuilder, n_words};
 use crate::traits::{BitSlice, BitSliceMut};
 
 // ── MemoryBitVec ──────────────────────────────────────────────────────────────
@@ -125,38 +125,14 @@ impl<B: BitSlice> BitXorAssign<&B> for MemoryBitVec {
 
 // ── Iterator ──────────────────────────────────────────────────────────────────
 
-pub struct MemoryBitIter<'a> {
-    words: &'a [u64],
-    slot: usize,
-    n: usize,
-}
-
-impl Iterator for MemoryBitIter<'_> {
-    type Item = bool;
-
-    fn next(&mut self) -> Option<bool> {
-        if self.slot >= self.n { return None; }
-        let v = (self.words[self.slot >> 6] >> (self.slot & 63)) & 1 != 0;
-        self.slot += 1;
-        Some(v)
-    }
-
-    fn size_hint(&self) -> (usize, Option<usize>) {
-        let rem = self.n - self.slot;
-        (rem, Some(rem))
-    }
-}
-
-impl ExactSizeIterator for MemoryBitIter<'_> {}
-
 impl MemoryBitVec {
-    pub fn iter(&self) -> MemoryBitIter<'_> {
-        MemoryBitIter { words: &self.words, slot: 0, n: self.n }
+    pub fn iter(&self) -> BitIter<'_> {
+        BitIter { words: &self.words, slot: 0, n: self.n }
     }
 }
 
 impl<'a> IntoIterator for &'a MemoryBitVec {
     type Item = bool;
-    type IntoIter = MemoryBitIter<'a>;
-    fn into_iter(self) -> MemoryBitIter<'a> { self.iter() }
+    type IntoIter = BitIter<'a>;
+    fn into_iter(self) -> BitIter<'a> { self.iter() }
 }