From 1385c710742e61f46bff1947c9e0482998f21eec Mon Sep 17 00:00:00 2001
From: overlookmotel <557937+overlookmotel@users.noreply.github.com>
Date: Wed, 13 Aug 2025 15:38:51 +0000
Subject: [PATCH] perf(codegen): further reduce memory allocations in
 `generate_line_offset_tables` (#13056)

#13054 added a nice optimization to `SourcemapBuilder`. During generation of line/offset tables, it reuses a single `Vec` for column indexes for each line, rather than creating a new `Vec` on each turn of the inner loop. This reduces the number of times that `Vec` may have to grow as column indexes get added to it.

Take this optimization a step further by re-using the same `Vec` across *all* lines.

`columns` `Vec` is not consumed on each line, but each time the contents are copied into a boxed slice - except when reaching EOF, where we can consume `columns`, as its work is done.

This memory-copying was likely happening anyway, as `Vec<u32>` -> `Box<[u32]>` conversion has to drop the spare capacity of the `Vec`, which will likely cause a reallocation.

Also, avoid using iterators to create the boxed slices. `Vec::clone` followed by `Vec::into_boxed_slice` is a bit more explicit and so may help compiler to see that it only needs to allocate exactly `columns.len()` slots for the `Box<[u32]>`.

Note: I also tried `columns.drain(..).collect()` instead of `columns.clone().into_boxed_slice()` + `columns.clear()`. But it looks like the `Drain` abstraction doesn't get completely removed by compiler. https://godbolt.org/z/Trv47j4hP So I *think* `into_boxed_slice` is probably preferable.
---
 crates/oxc_codegen/src/sourcemap_builder.rs | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/crates/oxc_codegen/src/sourcemap_builder.rs b/crates/oxc_codegen/src/sourcemap_builder.rs
index 098f90aa0e222..2fe16c1a26abd 100644
--- a/crates/oxc_codegen/src/sourcemap_builder.rs
+++ b/crates/oxc_codegen/src/sourcemap_builder.rs
@@ -312,6 +312,9 @@ impl<'a> SourcemapBuilder<'a> {
         let mut lines = vec![];
         let mut column_offsets = IndexVec::new();
 
+        // Used as a buffer to reduce memory reallocations
+        let mut columns = vec![];
+
         // Process content line-by-line.
         // For each line, start by assuming line will be entirely ASCII, and read byte-by-byte.
         // If line is all ASCII, UTF-8 columns and UTF-16 columns are the same,
@@ -325,7 +328,6 @@ impl<'a> SourcemapBuilder<'a> {
                 column_offsets_id: None,
             });
 
-            let mut columns = vec![]; // Used as a buffer to reduce memory reallocations.
             let remaining = &content.as_bytes()[line_byte_offset as usize..];
             for (byte_offset_from_line_start, b) in remaining.iter().enumerate() {
                 #[expect(clippy::cast_possible_truncation)]
@@ -351,8 +353,6 @@ impl<'a> SourcemapBuilder<'a> {
                         line.column_offsets_id =
                             Some(ColumnOffsetsId::from_usize(column_offsets.len()));
 
-                        columns.clear();
-
                         // Loop through rest of line char-by-char.
                         // `chunk_byte_offset` in this loop is byte offset from start of this 1st
                         // Unicode char.
@@ -396,8 +396,9 @@ impl<'a> SourcemapBuilder<'a> {
                             // Record column offsets
                             column_offsets.push(ColumnOffsets {
                                 byte_offset_to_first: byte_offset_from_line_start,
-                                columns: columns.into_iter().collect(),
+                                columns: columns.clone().into_boxed_slice(),
                             });
+                            columns.clear();
 
                             // Revert back to outer loop for next line
                             continue 'lines;
@@ -410,7 +411,7 @@ impl<'a> SourcemapBuilder<'a> {
                         // Record column offsets
                         column_offsets.push(ColumnOffsets {
                             byte_offset_to_first: byte_offset_from_line_start,
-                            columns: columns.into_iter().collect(),
+                            columns: columns.into_boxed_slice(),
                         });
 
                         break 'lines;