From 2352de227718addbde3bcca018fbf2a3d2490777 Mon Sep 17 00:00:00 2001
From: Charlie Marsh <charlie.r.marsh@gmail.com>
Date: Sat, 3 Feb 2024 06:40:41 -0800
Subject: [PATCH] Slight speed-up for lowercase and uppercase identifier checks
 (#9798)

It turns out that for ASCII identifiers, this is nearly 2x faster:

```
Parser/before     time:   [15.388 ns 15.395 ns 15.406 ns]
Parser/after      time:   [8.3786 ns 8.5821 ns 8.7715 ns]
```
---
 crates/ruff_python_stdlib/src/str.rs | 36 ++++++++++++++++++++++++++--
 1 file changed, 34 insertions(+), 2 deletions(-)

diff --git a/crates/ruff_python_stdlib/src/str.rs b/crates/ruff_python_stdlib/src/str.rs
index 2b7b90b64b..048b2e5929 100644
--- a/crates/ruff_python_stdlib/src/str.rs
+++ b/crates/ruff_python_stdlib/src/str.rs
@@ -14,9 +14,25 @@
 /// assert!(!is_lowercase("ABC"));
 /// assert!(is_lowercase(""));
 /// assert!(is_lowercase("_"));
+/// assert!(is_lowercase("αbc"));
+/// assert!(!is_lowercase("αBC"));
+/// assert!(!is_lowercase("Ωbc"));
 /// ```
 pub fn is_lowercase(s: &str) -> bool {
-    s.chars().all(|c| !c.is_alphabetic() || c.is_lowercase())
+    for (i, &c) in s.as_bytes().iter().enumerate() {
+        match c {
+            // Match against ASCII uppercase characters.
+            b'A'..=b'Z' => return false,
+            _ if c.is_ascii() => {}
+            // If the character is non-ASCII, fallback to slow path.
+            _ => {
+                return s[i..]
+                    .chars()
+                    .all(|c| c.is_lowercase() || !c.is_alphabetic())
+            }
+        }
+    }
+    true
 }
 
 /// Return `true` if a string is uppercase.
@@ -35,9 +51,25 @@ pub fn is_lowercase(s: &str) -> bool {
 /// assert!(!is_uppercase("abc"));
 /// assert!(is_uppercase(""));
 /// assert!(is_uppercase("_"));
+/// assert!(is_uppercase("ΩBC"));
+/// assert!(!is_uppercase("Ωbc"));
+/// assert!(!is_uppercase("αBC"));
 /// ```
 pub fn is_uppercase(s: &str) -> bool {
-    s.chars().all(|c| !c.is_alphabetic() || c.is_uppercase())
+    for (i, &c) in s.as_bytes().iter().enumerate() {
+        match c {
+            // Match against ASCII lowercase characters.
+            b'a'..=b'z' => return false,
+            _ if c.is_ascii() => {}
+            // If the character is non-ASCII, fallback to slow path.
+            _ => {
+                return s[i..]
+                    .chars()
+                    .all(|c| c.is_uppercase() || !c.is_alphabetic())
+            }
+        }
+    }
+    true
 }
 
 /// Return `true` if a string is _cased_ as lowercase.