Implement confusing unicode character detection for comments (#653)

README.md

@@ -552,6 +552,7 @@ For more, see [flake8-annotations](https://pypi.org/project/flake8-annotations/2
 | ---- | ---- | ------- | --- |
 | RUF001 | AmbiguousUnicodeCharacterString | String contains ambiguous unicode character '𝐁' (did you mean 'B'?) | 🛠 |
 | RUF002 | AmbiguousUnicodeCharacterDocstring | Docstring contains ambiguous unicode character '𝐁' (did you mean 'B'?) | 🛠 |
+| RUF003 | AmbiguousUnicodeCharacterComment | Comment contains ambiguous unicode character '𝐁' (did you mean 'B'?) |  |
 
 ### Meta rules
 

resources/test/fixtures/RUF001.py

@@ -2,5 +2,6 @@ x = "𝐁ad string"
 
 
 def f():
-    """Here's a comment with an unusual parenthesis: ）"""
+    """Here's a docstring with an unusual parenthesis: ）"""
+    # And here's a comment with an unusual punctuation mark: ᜵
     ...

resources/test/fixtures/RUF002.py

@@ -2,5 +2,6 @@ x = "𝐁ad string"
 
 
 def f():
-    """Here's a comment with an unusual parenthesis: ）"""
+    """Here's a docstring with an unusual parenthesis: ）"""
+    # And here's a comment with an unusual punctuation mark: ᜵
     ...

resources/test/fixtures/RUF003.py

@@ -0,0 +1,7 @@
+x = "𝐁ad string"
+
+
+def f():
+    """Here's a docstring with an unusual parenthesis: ）"""
+    # And here's a comment with an unusual punctuation mark: ᜵
+    ...

src/check_tokens.rs

@@ -5,6 +5,7 @@ use rustpython_parser::lexer::{LexResult, Tok};
 use crate::autofix::fixer;
 use crate::checks::{Check, CheckCode};
 use crate::lex::docstring_detection::StateMachine;
+use crate::rules::checks::Context;
 use crate::source_code_locator::SourceCodeLocator;
 use crate::{flake8_quotes, pycodestyle, rules, Settings};
 
@@ -16,7 +17,8 @@ pub fn check_tokens(
     autofix: &fixer::Mode,
 ) {
     let enforce_ambiguous_unicode_character = settings.enabled.contains(&CheckCode::RUF001)
-        || settings.enabled.contains(&CheckCode::RUF002);
+        || settings.enabled.contains(&CheckCode::RUF002)
+        || settings.enabled.contains(&CheckCode::RUF003);
     let enforce_quotes = settings.enabled.contains(&CheckCode::Q000)
         || settings.enabled.contains(&CheckCode::Q001)
         || settings.enabled.contains(&CheckCode::Q002)
@@ -31,14 +33,22 @@ pub fn check_tokens(
             false
         };
 
-        // RUF001, RUF002
+        // RUF001, RUF002, RUF003
         if enforce_ambiguous_unicode_character {
-            if matches!(tok, Tok::String { .. }) {
+            if matches!(tok, Tok::String { .. } | Tok::Comment) {
                 for check in rules::checks::ambiguous_unicode_character(
                     locator,
                     start,
                     end,
-                    is_docstring,
+                    if matches!(tok, Tok::String { .. }) {
+                        if is_docstring {
+                            Context::Docstring
+                        } else {
+                            Context::String
+                        }
+                    } else {
+                        Context::Comment
+                    },
                     autofix.patch(),
                 ) {
                     if settings.enabled.contains(check.kind.code()) {

src/checks.rs

@@ -202,6 +202,7 @@ pub enum CheckCode {
     // Ruff
     RUF001,
     RUF002,
+    RUF003,
     // Meta
     M001,
 }
@@ -462,6 +463,7 @@ pub enum CheckKind {
     // Ruff
     AmbiguousUnicodeCharacterString(char, char),
     AmbiguousUnicodeCharacterDocstring(char, char),
+    AmbiguousUnicodeCharacterComment(char, char),
     // Meta
     UnusedNOQA(Option<Vec<String>>),
 }
@@ -480,7 +482,8 @@ impl CheckCode {
             | CheckCode::Q003
             | CheckCode::W605
             | CheckCode::RUF001
-            | CheckCode::RUF002 => &LintSource::Tokens,
+            | CheckCode::RUF002
+            | CheckCode::RUF003 => &LintSource::Tokens,
             CheckCode::E902 => &LintSource::FileSystem,
             _ => &LintSource::AST,
         }
@@ -702,6 +705,7 @@ impl CheckCode {
             // Ruff
             CheckCode::RUF001 => CheckKind::AmbiguousUnicodeCharacterString('𝐁', 'B'),
             CheckCode::RUF002 => CheckKind::AmbiguousUnicodeCharacterDocstring('𝐁', 'B'),
+            CheckCode::RUF003 => CheckKind::AmbiguousUnicodeCharacterComment('𝐁', 'B'),
             // Meta
             CheckCode::M001 => CheckKind::UnusedNOQA(None),
         }
@@ -873,6 +877,7 @@ impl CheckCode {
             CheckCode::N818 => CheckCategory::PEP8Naming,
             CheckCode::RUF001 => CheckCategory::Ruff,
             CheckCode::RUF002 => CheckCategory::Ruff,
+            CheckCode::RUF003 => CheckCategory::Ruff,
             CheckCode::M001 => CheckCategory::Meta,
         }
     }
@@ -1058,6 +1063,7 @@ impl CheckKind {
             // Ruff
             CheckKind::AmbiguousUnicodeCharacterString(..) => &CheckCode::RUF001,
             CheckKind::AmbiguousUnicodeCharacterDocstring(..) => &CheckCode::RUF002,
+            CheckKind::AmbiguousUnicodeCharacterComment(..) => &CheckCode::RUF003,
             // Meta
             CheckKind::UnusedNOQA(_) => &CheckCode::M001,
         }
@@ -1606,6 +1612,12 @@ impl CheckKind {
                      '{representant}'?)"
                 )
             }
+            CheckKind::AmbiguousUnicodeCharacterComment(confusable, representant) => {
+                format!(
+                    "Comment contains ambiguous unicode character '{confusable}' (did you mean \
+                     '{representant}'?)"
+                )
+            }
             // Meta
             CheckKind::UnusedNOQA(codes) => match codes {
                 None => "Unused `noqa` directive".to_string(),

src/checks_gen.rs

@@ -233,6 +233,7 @@ pub enum CheckCodePrefix {
     RUF00,
     RUF001,
     RUF002,
+    RUF003,
     T,
     T2,
     T20,
@@ -926,11 +927,12 @@ impl CheckCodePrefix {
             CheckCodePrefix::Q001 => vec![CheckCode::Q001],
             CheckCodePrefix::Q002 => vec![CheckCode::Q002],
             CheckCodePrefix::Q003 => vec![CheckCode::Q003],
-            CheckCodePrefix::RUF => vec![CheckCode::RUF001, CheckCode::RUF002],
+            CheckCodePrefix::RUF => vec![CheckCode::RUF001, CheckCode::RUF002, CheckCode::RUF003],
-            CheckCodePrefix::RUF0 => vec![CheckCode::RUF001, CheckCode::RUF002],
+            CheckCodePrefix::RUF0 => vec![CheckCode::RUF001, CheckCode::RUF002, CheckCode::RUF003],
-            CheckCodePrefix::RUF00 => vec![CheckCode::RUF001, CheckCode::RUF002],
+            CheckCodePrefix::RUF00 => vec![CheckCode::RUF001, CheckCode::RUF002, CheckCode::RUF003],
             CheckCodePrefix::RUF001 => vec![CheckCode::RUF001],
             CheckCodePrefix::RUF002 => vec![CheckCode::RUF002],
+            CheckCodePrefix::RUF003 => vec![CheckCode::RUF003],
             CheckCodePrefix::T => vec![CheckCode::T201, CheckCode::T203],
             CheckCodePrefix::T2 => vec![CheckCode::T201, CheckCode::T203],
             CheckCodePrefix::T20 => vec![CheckCode::T201, CheckCode::T203],
@@ -1218,6 +1220,7 @@ impl CheckCodePrefix {
             CheckCodePrefix::RUF00 => PrefixSpecificity::Tens,
             CheckCodePrefix::RUF001 => PrefixSpecificity::Explicit,
             CheckCodePrefix::RUF002 => PrefixSpecificity::Explicit,
+            CheckCodePrefix::RUF003 => PrefixSpecificity::Explicit,
             CheckCodePrefix::T => PrefixSpecificity::Category,
             CheckCodePrefix::T2 => PrefixSpecificity::Hundreds,
             CheckCodePrefix::T20 => PrefixSpecificity::Tens,

src/linter.rs

@@ -452,6 +452,7 @@ mod tests {
     #[test_case(CheckCode::W605, Path::new("W605_1.py"); "W605_1")]
     #[test_case(CheckCode::RUF001, Path::new("RUF001.py"); "RUF001")]
     #[test_case(CheckCode::RUF002, Path::new("RUF002.py"); "RUF002")]
+    #[test_case(CheckCode::RUF003, Path::new("RUF003.py"); "RUF003")]
     fn checks(check_code: CheckCode, path: &Path) -> Result<()> {
         let snapshot = format!("{}_{}", check_code.as_ref(), path.to_string_lossy());
         let mut checks = check_path(

src/rules/checks.rs

@@ -1596,11 +1596,17 @@ static CONFUSABLES: Lazy<BTreeMap<u32, u32>> = Lazy::new(|| {
     ])
 });
 
+pub enum Context {
+    String,
+    Docstring,
+    Comment,
+}
+
 pub fn ambiguous_unicode_character(
     locator: &SourceCodeLocator,
     start: &Location,
     end: &Location,
-    is_docstring: bool,
+    context: Context,
     fix: bool,
 ) -> Vec<Check> {
     let mut checks = vec![];
@@ -1623,10 +1629,17 @@ pub fn ambiguous_unicode_character(
                 };
                 let end_location = Location::new(location.row(), location.column() + 1);
                 let mut check = Check::new(
-                    if is_docstring {
+                    match context {
-                        CheckKind::AmbiguousUnicodeCharacterDocstring(current_char, representant)
+                        Context::String => {
-                    } else {
+                            CheckKind::AmbiguousUnicodeCharacterString(current_char, representant)
-                        CheckKind::AmbiguousUnicodeCharacterString(current_char, representant)
+                        }
+                        Context::Docstring => CheckKind::AmbiguousUnicodeCharacterDocstring(
+                            current_char,
+                            representant,
+                        ),
+                        Context::Comment => {
+                            CheckKind::AmbiguousUnicodeCharacterComment(current_char, representant)
+                        }
                     },
                     Range {
                         location,

src/snapshots/ruff__linter__tests__RUF002_RUF002.py.snap

@@ -8,18 +8,18 @@ expression: checks
       - )
   location:
     row: 5
-    column: 53
+    column: 55
   end_location:
     row: 5
-    column: 54
+    column: 56
   fix:
     patch:
       content: )
       location:
         row: 5
-        column: 53
+        column: 55
       end_location:
         row: 5
-        column: 54
+        column: 56
     applied: false
 

src/snapshots/ruff__linter__tests__RUF003_RUF003.py.snap

@@ -0,0 +1,25 @@
+---
+source: src/linter.rs
+expression: checks
+---
+- kind:
+    AmbiguousUnicodeCharacterComment:
+      - ᜵
+      - /
+  location:
+    row: 6
+    column: 61
+  end_location:
+    row: 6
+    column: 62
+  fix:
+    patch:
+      content: /
+      location:
+        row: 6
+        column: 61
+      end_location:
+        row: 6
+        column: 62
+    applied: false
+