fix: Ignore combining characters in SRE

joshuamegnauth54 · joshuamegnauth54 · commit 64dd76039f50 · 2026-04-14T20:35:39.000-04:00
Closes: #7518
diff --git a/Cargo.lock b/Cargo.lock
diff --git a/crates/sre_engine/Cargo.toml b/crates/sre_engine/Cargo.toml
@@ -19,6 +19,7 @@ rustpython-wtf8 = { workspace = true }
 num_enum = { workspace = true }
 bitflags = { workspace = true }
 optional = { workspace = true }
+icu_properties = { workspace = true }
 
 [dev-dependencies]
 criterion = { workspace = true }
diff --git a/crates/sre_engine/src/string.rs b/crates/sre_engine/src/string.rs
@@ -1,3 +1,4 @@
+use icu_properties::{CodePointMapData, props::CanonicalCombiningClass};
 use rustpython_wtf8::Wtf8;
 
 #[derive(Debug, Clone, Copy)]
@@ -442,8 +443,9 @@ pub(crate) const fn is_uni_linebreak(ch: u32) -> bool {
 #[inline]
 pub(crate) fn is_uni_alnum(ch: u32) -> bool {
     // TODO: check with cpython
+    let map = CodePointMapData::<CanonicalCombiningClass>::new();
     char::try_from(ch)
-        .map(|x| x.is_alphanumeric())
+        .map(|x| x.is_alphanumeric() && map.get(x) == CanonicalCombiningClass::NotReordered)
         .unwrap_or(false)
 }
 #[inline]
diff --git a/extra_tests/snippets/stdlib_re.py b/extra_tests/snippets/stdlib_re.py
@@ -81,4 +81,4 @@
 assert re.fullmatch(r"([0-9]++(?:\.[0-9]+)*+)", "1.25.38").group(0) == "1.25.38"
 
 # Combining characters; issue #7518
-# assert not re.match(r"\w", "\u0345"), r"\w should not match U+0345 (category Mn)"
+assert not re.match(r"\w", "\u0345"), r"\w should not match U+0345 (category Mn)"
