From a924304d27b162747df49899bf301bd0e4120f2c Mon Sep 17 00:00:00 2001
From: Jeffrey Vo <jeffrey.vo.australia@gmail.com>
Date: Fri, 30 Jan 2026 01:39:39 +0900
Subject: [PATCH] Mark `BufferBuilder::new_from_buffer` as unsafe (#9292)

# Which issue does this PR close?

<!--
We generally require a GitHub issue to be filed for all bug fixes and
enhancements and this helps us generate change logs for our releases.
You can link an issue to this PR using the GitHub syntax.
-->

- Closes #9287

# Rationale for this change

<!--
Why are you proposing this change? If this is already explained clearly
in the issue then this section is not needed.
Explaining clearly why changes are proposed helps reviewers understand
your changes and offer better suggestions for fixes.
-->

As identified by the issue, it is possible to use safe APIs to produce
an unaligned buffer in `BufferBuilder`.

# What changes are included in this PR?

<!--
There is no need to duplicate the description in the issue here but it
is sometimes worth providing a summary of the individual changes in this
PR.
-->

Mark `BufferBuilder::new_from_buffer` as unsafe since there are
alignment invariants that must be upheld to use it correctly.

Also fix some docstrings.

# Are these changes tested?

<!--
We typically require tests for all PRs in order to:
1. Prevent the code from being accidentally broken by subsequent changes
2. Serve as another way to document the expected behavior of the code

If tests are not included in your PR, please explain why (for example,
are they covered by existing tests)?
-->

Marking API unsafe only.

# Are there any user-facing changes?

<!--
If there are user-facing changes then we may require documentation to be
updated before approving the PR.

If there are any breaking changes to public APIs, please call them out.
-->

Yes.
---
 arrow-buffer/src/builder/mod.rs | 24 ++++++++++--------------
 1 file changed, 10 insertions(+), 14 deletions(-)

diff --git a/arrow-buffer/src/builder/mod.rs b/arrow-buffer/src/builder/mod.rs
index abe510bdabc6..e0e41e2576f2 100644
--- a/arrow-buffer/src/builder/mod.rs
+++ b/arrow-buffer/src/builder/mod.rs
@@ -70,7 +70,6 @@ impl<T: ArrowNativeType> BufferBuilder<T> {
     ///
     /// ```
     /// # use arrow_buffer::builder::BufferBuilder;
-    ///
     /// let mut builder = BufferBuilder::<u8>::new(10);
     ///
     /// assert!(builder.capacity() >= 10);
@@ -87,7 +86,11 @@ impl<T: ArrowNativeType> BufferBuilder<T> {
     }
 
     /// Creates a new builder from a [`MutableBuffer`]
-    pub fn new_from_buffer(buffer: MutableBuffer) -> Self {
+    ///
+    /// # Safety
+    ///
+    /// - `buffer` bytes must be aligned to type `T`
+    pub unsafe fn new_from_buffer(buffer: MutableBuffer) -> Self {
         let buffer_len = buffer.len();
         Self {
             buffer,
@@ -102,7 +105,6 @@ impl<T: ArrowNativeType> BufferBuilder<T> {
     ///
     /// ```
     /// # use arrow_buffer::builder::BufferBuilder;
-    ///
     /// let mut builder = BufferBuilder::<u8>::new(10);
     /// builder.append(42);
     ///
@@ -118,7 +120,6 @@ impl<T: ArrowNativeType> BufferBuilder<T> {
     ///
     /// ```
     /// # use arrow_buffer::builder::BufferBuilder;
-    ///
     /// let mut builder = BufferBuilder::<u8>::new(10);
     /// builder.append(42);
     ///
@@ -149,7 +150,6 @@ impl<T: ArrowNativeType> BufferBuilder<T> {
     ///
     /// ```
     /// # use arrow_buffer::builder::BufferBuilder;
-    ///
     /// let mut builder = BufferBuilder::<u8>::new(10);
     /// builder.advance(2);
     ///
@@ -167,7 +167,6 @@ impl<T: ArrowNativeType> BufferBuilder<T> {
     ///
     /// ```
     /// # use arrow_buffer::builder::BufferBuilder;
-    ///
     /// let mut builder = BufferBuilder::<u8>::new(10);
     /// builder.reserve(10);
     ///
@@ -185,7 +184,6 @@ impl<T: ArrowNativeType> BufferBuilder<T> {
     ///
     /// ```
     /// # use arrow_buffer::builder::BufferBuilder;
-    ///
     /// let mut builder = BufferBuilder::<u8>::new(10);
     /// builder.append(42);
     ///
@@ -205,7 +203,6 @@ impl<T: ArrowNativeType> BufferBuilder<T> {
     ///
     /// ```
     /// # use arrow_buffer::builder::BufferBuilder;
-    ///
     /// let mut builder = BufferBuilder::<u8>::new(10);
     /// builder.append_n(10, 42);
     ///
@@ -223,12 +220,12 @@ impl<T: ArrowNativeType> BufferBuilder<T> {
     ///
     /// ```
     /// # use arrow_buffer::builder::BufferBuilder;
-    ///
     /// let mut builder = BufferBuilder::<u32>::new(10);
     /// builder.append_n_zeroed(3);
     ///
     /// assert_eq!(builder.len(), 3);
     /// assert_eq!(builder.as_slice(), &[0, 0, 0])
+    /// ```
     #[inline]
     pub fn append_n_zeroed(&mut self, n: usize) {
         self.buffer.extend_zeros(n * std::mem::size_of::<T>());
@@ -241,7 +238,6 @@ impl<T: ArrowNativeType> BufferBuilder<T> {
     ///
     /// ```
     /// # use arrow_buffer::builder::BufferBuilder;
-    ///
     /// let mut builder = BufferBuilder::<u8>::new(10);
     /// builder.append_slice(&[42, 44, 46]);
     ///
@@ -257,7 +253,6 @@ impl<T: ArrowNativeType> BufferBuilder<T> {
     ///
     /// ```
     /// # use arrow_buffer::builder::BufferBuilder;
-    ///
     /// let mut builder = BufferBuilder::<f64>::new(10);
     /// builder.append(1.3);
     /// builder.append_n(2, 2.3);
@@ -280,7 +275,6 @@ impl<T: ArrowNativeType> BufferBuilder<T> {
     ///
     /// ```
     /// # use arrow_buffer::builder::BufferBuilder;
-    ///
     /// let mut builder = BufferBuilder::<f32>::new(10);
     ///
     /// builder.append_slice(&[1., 2., 3.4]);
@@ -307,7 +301,6 @@ impl<T: ArrowNativeType> BufferBuilder<T> {
     ///
     /// ```
     /// # use arrow_buffer::builder::BufferBuilder;
-    ///
     /// let mut builder = BufferBuilder::<u16>::new(10);
     ///
     /// builder.append_slice(&[42, 44, 46]);
@@ -377,7 +370,10 @@ impl<T: ArrowNativeType> Extend<T> for BufferBuilder<T> {
 
 impl<T: ArrowNativeType> From<Vec<T>> for BufferBuilder<T> {
     fn from(value: Vec<T>) -> Self {
-        Self::new_from_buffer(MutableBuffer::from(value))
+        let buffer = MutableBuffer::from(value);
+        // SAFETY
+        // - buffer is aligned to T
+        unsafe { Self::new_from_buffer(buffer) }
     }
 }