fix: deduping by url

This commit is contained in:
Jan Bader
2026-04-07 23:28:04 +02:00
parent a6d725f12c
commit 03bb94db2d

View File

@@ -121,9 +121,12 @@ def get_invoice_links(page: Page) -> list[dict]:
seen = set()
def _add_link(url: str, label: str, **meta) -> None:
if not url or url in seen:
if not url:
return
seen.add(url)
dedupe_key = meta.get("reference_id") or url
if dedupe_key in seen:
return
seen.add(dedupe_key)
entry = {"url": url, "label": label}
entry.update(meta)
links.append(entry)