fix: handle Unicode normalization in save_dict for umlauts (#728) (#729)

This commit is contained in:
Jens
2025-12-15 20:46:10 +01:00
committed by GitHub
parent 861b8ec367
commit 0b995fae18
5 changed files with 125 additions and 6 deletions

View File

@@ -144,9 +144,9 @@ def test_ensure_non_callable_truthy_and_falsy() -> None:
# Basic sanitization
("My Ad Title!", "My Ad Title!", "Basic sanitization"),
# Unicode normalization (sanitize-filename changes normalization)
("café", "cafe\u0301", "Unicode normalization"),
("caf\u00e9", "cafe\u0301", "Unicode normalization from escaped"),
# Unicode normalization - sanitize-filename converts to NFD, then we normalize to NFC (issue #728)
("café", "café", "Unicode NFC → NFD (by sanitize) → NFC (by normalize)"),
("caf\u00e9", "café", "Unicode NFC (escaped) → NFD → NFC"),
# Edge cases
("", "untitled", "Empty string"),