From 30f3efe6976ca1e6b13eb605f17c4a5be13e6bd2 Mon Sep 17 00:00:00 2001
From: Michael Sloan <michael@zed.dev>
Date: Fri, 9 May 2025 13:38:03 +0200
Subject: [PATCH] Revert unnecessary "fix handling of unicode when counting
 codeblock lines" + document (#30368)

After merging #30364 I realized why it was unnecessary to fix the code,
and was more efficient before. UTF-8 does not use the standard 0-127
ASCII range for multi-byte chars. So this reverts that change and
documents why the code is valid.

Release Notes:

- N/A
---
 crates/markdown/src/markdown.rs | 1 +
 crates/markdown/src/parser.rs   | 5 +++--
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/crates/markdown/src/markdown.rs b/crates/markdown/src/markdown.rs
index 8ef88901f3..998a180e63 100644
--- a/crates/markdown/src/markdown.rs
+++ b/crates/markdown/src/markdown.rs
@@ -223,6 +223,7 @@ impl Markdown {
     }
 
     pub fn escape(s: &str) -> Cow<str> {
+        // Valid to use bytes since multi-byte UTF-8 doesn't use ASCII chars.
         let count = s
             .bytes()
             .filter(|c| *c == b'\n' || c.is_ascii_punctuation())
diff --git a/crates/markdown/src/parser.rs b/crates/markdown/src/parser.rs
index e5e5847289..debfe5245f 100644
--- a/crates/markdown/src/parser.rs
+++ b/crates/markdown/src/parser.rs
@@ -79,9 +79,10 @@ pub fn parse_markdown(
                         let content_range =
                             content_range.start + range.start..content_range.end + range.start;
 
+                        // Valid to use bytes since multi-byte UTF-8 doesn't use ASCII chars.
                         let line_count = text[content_range.clone()]
-                            .chars()
-                            .filter(|c| *c == '\n')
+                            .bytes()
+                            .filter(|c| *c == b'\n')
                             .count();
                         let metadata = CodeBlockMetadata {
                             content_range,