normalization optimized
This commit is contained in:
@@ -0,0 +1,8 @@
|
||||
# Version 0.5.5.32
|
||||
|
||||
## German search normalization refinement
|
||||
|
||||
- Distinguishes real or transliterated umlauts from natural vowel combinations.
|
||||
- Keeps searches such as `Müller`/`Mueller` and `Hyäne`/`Hyaene` equivalent.
|
||||
- Prevents short umlaut filters from matching words such as `Bauer`, `teuer` or `euer` merely because they contain `ue`.
|
||||
- Applies the same comparison rules to generic table filters, the server-side job history filters, asset software filtering and duplicate detection.
|
||||
Reference in New Issue
Block a user