You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Quoted paths in report.md show non-ASCII characters as \u escapes #1790
When report.md quotes a file path, every non-ASCII character in that path shows as a \uXXXX escape. The report quotes a path when it has a comma, unusual whitespace, or a control character. Thus a path such as docs/Résumé, final.md shows as "docs/R\u00e9sum\u00e9, final.md" in "Included paths", "Excluded paths", the "Affected lines" cell, and code-evidence locations.
Version / source checked
CLI/SDK 0.5.0 source, current main at 06201edfa54a83fdfc3a9d0e28bfb019ce30b4c0, checked on 2026-10-20
Linux x64, Python 3.14.4
Reproduction
Project a canonical scan that has a finding location and an include path of docs/Résumé, final.md (a fictional path) with report_projection.build_report_markdown:
A path that does not need quotes, such as docs/Résumé.md, already shows its characters correctly.
Relevant code
_path_label calls json.dumps(value). The default ensure_ascii=True escapes all non-ASCII characters, not only the control characters that the quotes are there to show.
Activity
fix(reports): keep non-ASCII characters in quoted report paths (opena…