mirror of
https://github.com/astral-sh/ruff.git
synced 2025-09-29 21:35:58 +00:00
20217e9bbd
## Summary Fixes #9895 The cause for this panic came from an offset error in the code. When analyzing a hypothetical f-string, we attempt to re-parse it as an f-string, and use the AST data to determine, among other things, whether the format specifiers are correct. To determine the 'correctness' of a format specifier, we actually have to re-parse the format specifier, and this is where the issue lies. To get the source text for the specifier, we were taking a slice from the original file source text... even though the AST data for the specifier belongs to the standalone parsed f-string expression, meaning that the ranges are going to be way off. In a file with Unicode, this can cause panics if the slice is inside a char boundary. To fix this, we now slice from the temporary source we created earlier to parse the literal as an f-string. ## Test Plan The RUF027 snapshot test was amended to include a string with format specifiers which we _should_ be calling out. This is to ensure we do slice format specifiers from the source text correctly. |
||
|---|---|---|
| .. | ||
| ruff | ||
| ruff_benchmark | ||
| ruff_cache | ||
| ruff_dev | ||
| ruff_diagnostics | ||
| ruff_formatter | ||
| ruff_index | ||
| ruff_linter | ||
| ruff_macros | ||
| ruff_notebook | ||
| ruff_python_ast | ||
| ruff_python_codegen | ||
| ruff_python_formatter | ||
| ruff_python_index | ||
| ruff_python_literal | ||
| ruff_python_parser | ||
| ruff_python_resolver | ||
| ruff_python_semantic | ||
| ruff_python_stdlib | ||
| ruff_python_trivia | ||
| ruff_shrinking | ||
| ruff_source_file | ||
| ruff_text_size | ||
| ruff_wasm | ||
| ruff_workspace | ||