fix(snapshot): Cookie-Dedup + schneller Impressum-Tab + Tabellen-Zahl

- Cookies werden je Vendor nach Name dedupliziert (Consent-Phasen-Dubletten;
  BMW 2196 → ~772) — in cookie-check + get_snapshot, behebt aufgeblähte
  Kachel-/Finding-Zahlen.
- Impressum-Snapshot-Check überspringt den ~40s-LLM-Schritt (context skip_llm)
  → Tab lädt sofort statt leer zu bleiben.
- Vendor-Tabelle zeigt nur die Cookie-Zahl (kein 'Cookies'-Wort je Zeile).

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
This commit is contained in:
Benjamin Admin
2026-06-11 19:54:15 +02:00
parent b0ceae4350
commit 0f6cdc93fd
7 changed files with 51 additions and 4 deletions
@@ -75,7 +75,9 @@ def doc_input_from_snapshot(snap: dict, doc_type: str) -> dict | None:
"business_scope": scope,
"company_name": (profile.get("company_name") or snap.get("site_label") or ""),
"origin_domain": snap.get("site_domain", ""),
"context": {"scan_context": snap.get("scan_context") or {}},
# skip_llm: Snapshot-Ansicht ist interaktiv → kein ~40s-LLM-Schritt.
"context": {"scan_context": snap.get("scan_context") or {},
"skip_llm": True},
}
@@ -218,6 +218,11 @@ async def get_snapshot(snapshot_id: str):
snap = load_snapshot(db, snapshot_id)
if not snap:
raise HTTPException(status_code=404, detail="snapshot not found")
if snap.get("cmp_vendors"):
from compliance.services.cookie_storage_inventory import (
dedupe_vendor_cookies,
)
snap["cmp_vendors"] = dedupe_vendor_cookies(snap["cmp_vendors"])
return snap
finally:
db.close()
@@ -58,6 +58,7 @@ async def snapshot_cookie_check(snapshot_id: str):
)
from compliance.services.cookie_storage_inventory import (
build_storage_inventory, storage_transparency_finding,
dedupe_vendor_cookies,
)
from compliance.services.cookie_compliance_audit import (
audit_cookie_compliance,
@@ -67,7 +68,8 @@ async def snapshot_cookie_check(snapshot_id: str):
snap = load_snapshot(db, snapshot_id)
if not snap:
raise HTTPException(status_code=404, detail="snapshot not found")
vendors = snap.get("cmp_vendors") or []
# Consent-Phasen duplizieren Cookies → je Vendor nach Name deduplizieren.
vendors = dedupe_vendor_cookies(snap.get("cmp_vendors") or [])
names = [c.get("name", "")
for v in vendors for c in (v.get("cookies") or [])]
big = load_big_library(db, names)