Preserve trailing statement semicolons when using fmt: skip (#8273)

crates/ruff_python_formatter/resources/test/fixtures/ruff/fmt_skip/trailing_semi.py

@@ -0,0 +1,10 @@
+x = 1;  # fmt: skip
+
+x = 1   ; # fmt: skip
+
+x = 1 \
+    ;   # fmt: skip
+
+x = 1 # ; # fmt: skip
+
+_;  #unrelated semicolon

crates/ruff_python_formatter/src/statement/mod.rs

@@ -1,5 +1,7 @@
 use ruff_formatter::{FormatOwnedWithRule, FormatRefWithRule};
-use ruff_python_ast::Stmt;
+use ruff_python_ast::{AnyNodeRef, Stmt};
+use ruff_python_trivia::{SimpleToken, SimpleTokenKind, SimpleTokenizer};
+use ruff_text_size::{Ranged, TextRange};
 
 use crate::prelude::*;
 
@@ -81,3 +83,25 @@ impl<'ast> IntoFormat<PyFormatContext<'ast>> for Stmt {
         FormatOwnedWithRule::new(self, FormatStmt)
     }
 }
+
+/// Returns the range of the semicolon terminating the statement or `None` if the statement
+/// isn't terminated by a semicolon.
+pub(super) fn trailing_semicolon(node: AnyNodeRef, source: &str) -> Option<TextRange> {
+    debug_assert!(node.is_statement());
+
+    let tokenizer = SimpleTokenizer::starts_at(node.end(), source);
+
+    let next_token = tokenizer
+        .take_while(|token| !token.kind().is_comment())
+        .find(|token| !token.kind().is_trivia());
+
+    if let Some(SimpleToken {
+        kind: SimpleTokenKind::Semi,
+        range,
+    }) = next_token
+    {
+        Some(range)
+    } else {
+        None
+    }
+}

crates/ruff_python_formatter/src/verbatim.rs

@@ -17,6 +17,7 @@ use crate::comments::{leading_comments, trailing_comments, SourceComment};
 use crate::prelude::*;
 use crate::statement::clause::ClauseHeader;
 use crate::statement::suite::SuiteChildStatement;
+use crate::statement::trailing_semicolon;
 
 /// Disables formatting for all statements between the `first_suppressed` that has a leading `fmt: off` comment
 /// and the first trailing or leading `fmt: on` comment. The statements are formatted as they appear in the source code.
@@ -902,14 +903,23 @@ impl Format<PyFormatContext<'_>> for FormatSuppressedNode<'_> {
             }
         }
 
+        // Some statements may end with a semicolon. Preserve the semicolon
+        let semicolon_range = self
+            .node
+            .is_statement()
+            .then(|| trailing_semicolon(self.node, f.context().source()))
+            .flatten();
+        let verbatim_range = semicolon_range.map_or(self.node.range(), |semicolon| {
+            TextRange::new(self.node.start(), semicolon.end())
+        });
         comments.mark_verbatim_node_comments_formatted(self.node);
 
         // Write the outer comments and format the node as verbatim
         write!(
             f,
             [
                 leading_comments(node_comments.leading),
-                verbatim_text(self.node),
+                verbatim_text(verbatim_range),
                 trailing_comments(node_comments.trailing)
             ]
         )

crates/ruff_python_formatter/tests/snapshots/format@fmt_skip__trailing_semi.py.snap

@@ -0,0 +1,34 @@
+---
+source: crates/ruff_python_formatter/tests/fixtures.rs
+input_file: crates/ruff_python_formatter/resources/test/fixtures/ruff/fmt_skip/trailing_semi.py
+---
+## Input
+```py
+x = 1;  # fmt: skip
+
+x = 1   ; # fmt: skip
+
+x = 1 \
+    ;   # fmt: skip
+
+x = 1 # ; # fmt: skip
+
+_;  #unrelated semicolon
+```
+
+## Output
+```py
+x = 1;  # fmt: skip
+
+x = 1   ;  # fmt: skip
+
+x = 1 \
+    ;  # fmt: skip
+
+x = 1  # ; # fmt: skip
+
+_  # unrelated semicolon
+```
+
+
+

crates/ruff_python_trivia/src/tokenizer.rs

@@ -473,7 +473,7 @@ pub enum SimpleTokenKind {
 }
 
 impl SimpleTokenKind {
-    const fn is_trivia(self) -> bool {
+    pub const fn is_trivia(self) -> bool {
         matches!(
             self,
             SimpleTokenKind::Whitespace
@@ -482,6 +482,10 @@ impl SimpleTokenKind {
                 | SimpleTokenKind::Continuation
         )
     }
+
+    pub const fn is_comment(self) -> bool {
+        matches!(self, SimpleTokenKind::Comment)
+    }
 }
 
 /// Simple zero allocation tokenizer handling most tokens.