larvit/fejkdata
3
0
Fork 0
Code Pull Requests Actions Activity

Drop the SSA placeholder names and lowercase da, del and dos in Swedish surnames

Browse Source
This commit is contained in:
Lilleman auf Larv
2026-09-18 18:39:04 +02:00
parent 396ae3bcd2
commit 98938e6126
3 changed files with 7 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files
data-import/names-se.py
+1 -1
View File
@@ -19,7 +19,7 @@ CACHE = Path(__file__).resolve().parent / "cache"
SHEETS = {"f": "Tilltalsnamn kvinnor", "m": "Tilltalsnamn män"}
SURNAMES = "Efternamn"
NAME = re.compile(r"^[^\W\d_]{2,}([ '-][^\W\d_]{2,})*$")
PARTICLES = {"af", "av", "de", "den", "der", "di", "du", "la", "le", "van", "von"}
PARTICLES = {"af", "av", "da", "de", "del", "den", "der", "di", "dos", "du", "la", "le", "van", "von"}
def cased(name):