From 7109214b579b541931e804a98823adb013d77d07 Mon Sep 17 00:00:00 2001 From: Dhruv Manilawala Date: Tue, 25 Jun 2024 13:44:28 +0530 Subject: [PATCH] Update parser tests to validate token ranges (#12019) ## Summary This PR updates the parser test infrastructure to validate the token ranges. From the code documentation: ``` /// Verifies that: /// * the ranges are strictly increasing when loop the tokens in insertion order /// * all ranges are within the length of the source code ``` Follow-up from #12016 and #12017 resolves: #11938 ## Test Plan Make sure that there are no failures. --- crates/ruff_python_parser/tests/fixtures.rs | 37 +++++++++++++++++++-- 1 file changed, 35 insertions(+), 2 deletions(-) diff --git a/crates/ruff_python_parser/tests/fixtures.rs b/crates/ruff_python_parser/tests/fixtures.rs index 2b8d9acfc1c04..893695fa94f5b 100644 --- a/crates/ruff_python_parser/tests/fixtures.rs +++ b/crates/ruff_python_parser/tests/fixtures.rs @@ -8,7 +8,7 @@ use annotate_snippets::snippet::{AnnotationType, Slice, Snippet, SourceAnnotatio use ruff_python_ast::visitor::source_order::{walk_module, SourceOrderVisitor, TraversalSignal}; use ruff_python_ast::{AnyNodeRef, Mod}; -use ruff_python_parser::{parse_unchecked, Mode, ParseErrorType}; +use ruff_python_parser::{parse_unchecked, Mode, ParseErrorType, Token}; use ruff_source_file::{LineIndex, OneIndexed, SourceCode}; use ruff_text_size::{Ranged, TextLen, TextRange, TextSize}; @@ -60,6 +60,7 @@ fn test_valid_syntax(input_path: &Path) { panic!("{input_path:?}: {message}"); } + validate_tokens(parsed.tokens(), source.text_len(), input_path); validate_ast(parsed.syntax(), source.text_len(), input_path); let mut output = String::new(); @@ -86,6 +87,7 @@ fn test_invalid_syntax(input_path: &Path) { "{input_path:?}: Expected parser to generate at least one syntax error for a program containing syntax errors." ); + validate_tokens(parsed.tokens(), source.text_len(), input_path); validate_ast(parsed.syntax(), source.text_len(), input_path); let mut output = String::new(); @@ -126,7 +128,8 @@ fn test_invalid_syntax(input_path: &Path) { #[allow(clippy::print_stdout)] fn parser_quick_test() { let source = "\ -from foo import +f'{' +f'{foo!r' "; let parsed = parse_unchecked(source, Mode::Module); @@ -230,6 +233,36 @@ impl std::fmt::Display for CodeFrame<'_> { } } +/// Verifies that: +/// * the ranges are strictly increasing when loop the tokens in insertion order +/// * all ranges are within the length of the source code +fn validate_tokens(tokens: &[Token], source_length: TextSize, test_path: &Path) { + let mut previous: Option<&Token> = None; + + for token in tokens { + assert!( + token.end() <= source_length, + "{path}: Token range exceeds the source code length. Token: {token:#?}", + path = test_path.display() + ); + + if let Some(previous) = previous { + assert_eq!( + previous.range().ordering(token.range()), + Ordering::Less, + "{path}: Token ranges are not in increasing order +Previous token: {previous:#?} +Current token: {token:#?} +Tokens: {tokens:#?} +", + path = test_path.display(), + ); + } + + previous = Some(token); + } +} + /// Verifies that: /// * the range of the parent node fully encloses all its child nodes /// * the ranges are strictly increasing when traversing the nodes in pre-order.