Remove cyclic dev dependency with the parser crate

Author: dhruvmanila

Cargo.lock

@@ -27,7 +27,6 @@ serde = { workspace = true, optional = true }
 
 [dev-dependencies]
 insta = { workspace = true }
-ruff_python_parser = { path = "../ruff_python_parser" }
 
 [features]
 serde = ["dep:serde", "ruff_text_size/serde"]

crates/ruff_python_ast/Cargo.toml

@@ -42,3 +42,159 @@ pub fn tokens_and_ranges(
     let comment_ranges = comment_ranges.finish();
     Ok((tokens, comment_ranges))
 }
+
+#[cfg(test)]
+mod tests {
+    use ruff_python_parser::lexer::LexResult;
+    use ruff_python_parser::{tokenize, Mode};
+    use ruff_source_file::Locator;
+    use ruff_text_size::TextSize;
+
+    use crate::Indexer;
+
+    #[test]
+    fn block_comments_two_line_block_at_start() {
+        // arrange
+        let source = "# line 1\n# line 2\n";
+        let tokens = tokenize(source, Mode::Module);
+        let locator = Locator::new(source);
+        let indexer = Indexer::from_tokens(&tokens, &locator);
+
+        // act
+        let block_comments = indexer.comment_ranges().block_comments(&locator);
+
+        // assert
+        assert_eq!(block_comments, vec![TextSize::new(0), TextSize::new(9)]);
+    }
+
+    #[test]
+    fn block_comments_indented_block() {
+        // arrange
+        let source = "    # line 1\n    # line 2\n";
+        let tokens = tokenize(source, Mode::Module);
+        let locator = Locator::new(source);
+        let indexer = Indexer::from_tokens(&tokens, &locator);
+
+        // act
+        let block_comments = indexer.comment_ranges().block_comments(&locator);
+
+        // assert
+        assert_eq!(block_comments, vec![TextSize::new(4), TextSize::new(17)]);
+    }
+
+    #[test]
+    fn block_comments_single_line_is_not_a_block() {
+        // arrange
+        let source = "\n";
+        let tokens: Vec<LexResult> = tokenize(source, Mode::Module);
+        let locator = Locator::new(source);
+        let indexer = Indexer::from_tokens(&tokens, &locator);
+
+        // act
+        let block_comments = indexer.comment_ranges().block_comments(&locator);
+
+        // assert
+        assert_eq!(block_comments, Vec::<TextSize>::new());
+    }
+
+    #[test]
+    fn block_comments_lines_with_code_not_a_block() {
+        // arrange
+        let source = "x = 1  # line 1\ny = 2  # line 2\n";
+        let tokens = tokenize(source, Mode::Module);
+        let locator = Locator::new(source);
+        let indexer = Indexer::from_tokens(&tokens, &locator);
+
+        // act
+        let block_comments = indexer.comment_ranges().block_comments(&locator);
+
+        // assert
+        assert_eq!(block_comments, Vec::<TextSize>::new());
+    }
+
+    #[test]
+    fn block_comments_sequential_lines_not_in_block() {
+        // arrange
+        let source = "    # line 1\n        # line 2\n";
+        let tokens = tokenize(source, Mode::Module);
+        let locator = Locator::new(source);
+        let indexer = Indexer::from_tokens(&tokens, &locator);
+
+        // act
+        let block_comments = indexer.comment_ranges().block_comments(&locator);
+
+        // assert
+        assert_eq!(block_comments, Vec::<TextSize>::new());
+    }
+
+    #[test]
+    fn block_comments_lines_in_triple_quotes_not_a_block() {
+        // arrange
+        let source = r#"
+        """
+        # line 1
+        # line 2
+        """
+        "#;
+        let tokens = tokenize(source, Mode::Module);
+        let locator = Locator::new(source);
+        let indexer = Indexer::from_tokens(&tokens, &locator);
+
+        // act
+        let block_comments = indexer.comment_ranges().block_comments(&locator);
+
+        // assert
+        assert_eq!(block_comments, Vec::<TextSize>::new());
+    }
+
+    #[test]
+    fn block_comments_stress_test() {
+        // arrange
+        let source = r#"
+# block comment 1 line 1
+# block comment 2 line 2
+
+# these lines
+    # do not form
+# a block comment
+
+x = 1  # these lines also do not
+y = 2  # do not form a block comment
+
+# these lines do form a block comment
+#
+
+    #
+    # and so do these
+    #
+
+"""
+# these lines are in triple quotes and
+# therefore do not form a block comment
+"""
+        "#;
+        let tokens = tokenize(source, Mode::Module);
+        let locator = Locator::new(source);
+        let indexer = Indexer::from_tokens(&tokens, &locator);
+
+        // act
+        let block_comments = indexer.comment_ranges().block_comments(&locator);
+
+        // assert
+        assert_eq!(
+            block_comments,
+            vec![
+                // Block #1
+                TextSize::new(1),
+                TextSize::new(26),
+                // Block #2
+                TextSize::new(174),
+                TextSize::new(212),
+                // Block #3
+                TextSize::new(219),
+                TextSize::new(225),
+                TextSize::new(247)
+            ]
+        );
+    }
+}

crates/ruff_python_index/src/comment_ranges.rs

@@ -29,6 +29,7 @@ unicode_names2 = { workspace = true }
 unicode-normalization = { workspace = true }
 
 [dev-dependencies]
+ruff_python_trivia = { path = "../ruff_python_trivia" }
 ruff_source_file = { path = "../ruff_source_file" }
 
 annotate-snippets = { workspace = true }