Update identifier Unicode character validation to match Python spec (#…

…7209)

Co-authored-by: Micha Reiser <micha@reiser.io>

Cargo.toml

@@ -49,6 +49,7 @@ toml = { version = "0.7.2" }
 tracing = "0.1.37"
 tracing-indicatif = "0.3.4"
 tracing-subscriber = { version = "0.3.17", features = ["env-filter"] }
+unic-ucd-ident = "0.9.0"
 unicode-width = "0.1.10"
 uuid = { version = "1.4.1", features = ["v4", "fast-rng", "macro-diagnostics", "js"] }
 wsl = { version = "0.1.0" }

crates/ruff_python_parser/Cargo.toml

@@ -24,7 +24,7 @@ lalrpop-util = { version = "0.20.0", default-features = false }
 num-bigint = { workspace = true }
 num-traits = { workspace = true }
 unic-emoji-char = "0.9.0"
-unic-ucd-ident = "0.9.0"
+unic-ucd-ident = {  workspace = true }
 unicode_names2 = { version = "0.6.0", git = "https://github.com/youknowone/unicode_names2.git", rev = "4ce16aa85cbcdd9cc830410f1a72ef9a235f2fde" }
 rustc-hash = { workspace = true }
 static_assertions = "1.1.0"

crates/ruff_python_stdlib/Cargo.toml

@@ -13,3 +13,4 @@ license = { workspace = true }
 [lib]
 
 [dependencies]
+unic-ucd-ident = { workspace = true }

crates/ruff_python_stdlib/src/identifiers.rs

@@ -1,16 +1,18 @@
+use unic_ucd_ident::{is_xid_continue, is_xid_start};
+
 use crate::keyword::is_keyword;
 
 /// Returns `true` if a string is a valid Python identifier (e.g., variable
 /// name).
 pub fn is_identifier(name: &str) -> bool {
     // Is the first character a letter or underscore?
     let mut chars = name.chars();
-    if !chars.next().is_some_and(|c| c.is_alphabetic() || c == '_') {
+    if !chars.next().is_some_and(is_identifier_start) {
         return false;
     }
 
     // Are the rest of the characters letters, digits, or underscores?
-    if !chars.all(|c| c.is_alphanumeric() || c == '_') {
+    if !chars.all(is_identifier_continuation) {
         return false;
     }
 
@@ -22,6 +24,21 @@ pub fn is_identifier(name: &str) -> bool {
     true
 }
 
+// Checks if the character c is a valid starting character as described
+// in https://docs.python.org/3/reference/lexical_analysis.html#identifiers
+fn is_identifier_start(c: char) -> bool {
+    matches!(c, 'a'..='z' | 'A'..='Z' | '_') || is_xid_start(c)
+}
+
+// Checks if the character c is a valid continuation character as described
+// in https://docs.python.org/3/reference/lexical_analysis.html#identifiers
+fn is_identifier_continuation(c: char) -> bool {
+    match c {
+        'a'..='z' | 'A'..='Z' | '_' | '0'..='9' => true,
+        c => is_xid_continue(c),
+    }
+}
+
 /// Returns `true` if a string is a private identifier, such that, when the
 /// identifier is defined in a class definition, it will be mangled prior to
 /// code generation.
@@ -76,7 +93,25 @@ pub fn is_migration_name(name: &str) -> bool {
 
 #[cfg(test)]
 mod tests {
-    use crate::identifiers::{is_migration_name, is_module_name};
+    use crate::identifiers::{is_identifier, is_migration_name, is_module_name};
+
+    #[test]
+    fn valid_identifiers() {
+        assert!(is_identifier("_abc"));
+        assert!(is_identifier("abc"));
+        assert!(is_identifier("_"));
+        assert!(is_identifier("a_b_c"));
+        assert!(is_identifier("abc123"));
+        assert!(is_identifier("abc_123"));
+        assert!(is_identifier("漢字"));
+        assert!(is_identifier("ひらがな"));
+        assert!(is_identifier("العربية"));
+        assert!(is_identifier("кириллица"));
+        assert!(is_identifier("πr"));
+        assert!(!is_identifier(""));
+        assert!(!is_identifier("percentile_co³t"));
+        assert!(!is_identifier("HelloWorld❤️"));
+    }
 
     #[test]
     fn module_name() {

crates/ruff_python_trivia/Cargo.toml

@@ -18,7 +18,7 @@ ruff_source_file = { path = "../ruff_source_file" }
 
 memchr = { workspace = true }
 smallvec = { workspace = true }
-unic-ucd-ident = "0.9.0"
+unic-ucd-ident = { workspace = true }
 
 [dev-dependencies]
 insta = { workspace = true }