fix(vvt): correct ePaaS schema mapping + category-aware scoring

The first BMW VVT table rendered all 24 providers at 20% score because
the ePaaS extractor was reading the wrong field names. Actual schema is
nested: providers[].processings[].persistences[], NOT providers[] alone.

Correct ePaaS schema (verified against bmw.com/epaas/.../de_DE.epaas.json):
  Provider:    {id, name, description, processings[]}
  Processing:  {id, name, description, categoryId, optOutLink,
                privacyPolicyLink, persistences[]}
  Persistence: {id, name, domain, type, expiry, description}

Two structural changes:

1. One row per processing (not provider). BMW has 26 providers but ~91
   processings spread across them (Adobe alone has ACMProcessing,
   AdobeAnalytics, AdobeCampaign, AdobeTargetAnalytics, AdobeTargetPers.).
   The cookie widget displays each processing separately — VVT now
   mirrors that. Display name format: 'Provider Name — Processing Name'.

2. Read optOutLink/privacyPolicyLink from PROCESSING (where they live),
   not provider. Persistences flatten to cookies[] with name + expiry +
   description.

Plus category mapping:
  advertising -> marketing
  strictlyNecessary -> necessary
  statistics -> statistics
  functional -> functional

Category-aware scoring (cookie_link_validator.score_vendors):
- 'necessary' (technisch erforderliche, §25 Abs. 2 TDDDG): no opt-out
  required, no country required. Score weight shifts to purpose +
  cookie disclosure (essential cookies must list names + expiry).
- All other categories: opt-out URL still mandatory; missing opt-out
  flags 'no_opt_out_url' and zeros that block of points.

Expected BMW result after this fix:
- ~91 rows (Adobe Analytics, Adform Retargeting, Akamai Infrastructure,
  AWS, ..., plus ~60 strictlyNecessary processings)
- Marketing rows with present opt-out → ~75-90%
- Necessary rows with cookie+expiry → ~85-95%
- Rows missing fields → still flagged

backend-compliance/compliance/services/cookie_link_validator.py

@@ -173,8 +173,16 @@ async def validate_vendor_urls(vendors: list[dict]) -> list[dict]:
 
 
 def score_vendors(vendors: list[dict]) -> list[dict]:
-    """Compute per-vendor compliance score (0-100) and flags. Mutates."""
+    """Compute per-vendor compliance score (0-100) and flags. Mutates.
+
+    Category-aware: 'necessary' (technisch erforderliche Cookies) do NOT
+    require an opt-out — §25 Abs. 2 TDDDG. Penalising them for that would
+    be wrong; instead we require precise purpose + cookie disclosure.
+    """
     for v in vendors:
+        is_necessary = (v.get("category") or "").lower() in (
+            "necessary", "strictlynecessary",
+        )
         score = 0
         max_score = 0
         flags: list[str] = []
@@ -186,50 +194,56 @@ def score_vendors(vendors: list[dict]) -> list[dict]:
         else:
             flags.append("no_name")
 
-        # Purpose — 15
+        # Purpose — 20
-        max_score += 15
+        max_score += 20
         if v.get("purpose"):
-            score += 15
+            score += 20
         else:
             flags.append("no_purpose")
 
-        # Country (3rd-country transfer relevance) — 10
+        # Country (3rd-country transfer relevance) — only relevant for
-        max_score += 10
+        # consent-based categories (otherwise irrelevant flag noise)
-        if v.get("country"):
+        if not is_necessary:
-            score += 10
+            max_score += 10
-        else:
+            if v.get("country"):
-            flags.append("no_country")
+                score += 10
+            else:
+                flags.append("no_country")
 
-        # Opt-Out URL present + reachable — 25
+        # Opt-Out URL — only for consent-based categories (§25 TDDDG)
-        max_score += 25
+        if not is_necessary:
-        if not v.get("opt_out_url"):
+            max_score += 25
-            flags.append("no_opt_out_url")
+            if not v.get("opt_out_url"):
-        elif v.get("opt_out_ok") is False:
+                flags.append("no_opt_out_url")
-            flags.append("broken_opt_out")
+            elif v.get("opt_out_ok") is False:
-            score += 5  # at least they tried
+                flags.append("broken_opt_out")
-        else:
+                score += 5
-            score += 25
+            else:
+                score += 25
 
-        # Privacy policy URL present + reachable — 15
+        # Privacy policy URL — relevant for all, but weight lower for necessary
-        max_score += 15
+        weight = 10 if is_necessary else 15
+        max_score += weight
         if not v.get("privacy_policy_url"):
             flags.append("no_privacy_url")
         elif v.get("privacy_ok") is False:
             flags.append("broken_privacy_url")
-            score += 5
+            score += weight // 3
         else:
-            score += 15
+            score += weight
 
-        # Cookies disclosed (names + expiry) — 15
+        # Cookies disclosed (names + expiry) — higher weight for necessary
-        max_score += 15
+        # (since that's mostly what they offer in lieu of opt-out)
+        weight = 50 if is_necessary else 15
+        max_score += weight
         cookies = v.get("cookies") or []
         if cookies:
             named = sum(1 for c in cookies if c.get("name"))
             with_expiry = sum(1 for c in cookies if c.get("expiry"))
             if named >= 1 and with_expiry >= 1:
-                score += 15
+                score += weight
             elif named >= 1:
-                score += 8
+                score += weight // 2
                 flags.append("cookies_no_expiry")
             else:
                 flags.append("cookies_no_names")

backend-compliance/compliance/services/vendor_extractor.py

@@ -90,41 +90,67 @@ def extract_vendors_from_payloads(payloads: list[dict]) -> list[dict]:
 
 # ── ePaaS (BMW Group) ───────────────────────────────────────────────
 
+# Maps ePaaS categoryId -> canonical category used by the VVT scorer.
+_EPAAS_CATEGORY_MAP = {
+    "advertising": "marketing",
+    "marketing": "marketing",
+    "strictlyNecessary": "necessary",
+    "necessary": "necessary",
+    "statistics": "statistics",
+    "functional": "functional",
+}
+
+
 def _extract_epaas(d: dict) -> list[dict]:
+    """Convert ePaaS payload into one row per *processing* (not provider).
+
+    ePaaS schema (BMW):
+      providers[].processings[].persistences[]
+        provider:    {id, name, description}
+        processing:  {id, name, description, categoryId, optOutLink,
+                      privacyPolicyLink, persistences}
+        persistence: {id, name, domain, type, expiry, description}
+
+    Each processing is a separate displayable unit in the cookie widget
+    (Adobe Analytics, Adobe Campaign, Adobe Target Personalisation, …) —
+    matching the website layout one-to-one in the VVT table. Provider name
+    becomes the prefix so the data-controller entity is visible.
+    """
     out: list[dict] = []
-    providers = d.get("providers", []) or []
+    for provider in d.get("providers", []) or []:
-    cookies_by_provider: dict[str, list[dict]] = {}
+        provider_name = provider.get("name") or provider.get("id") or ""
+        provider_desc = _clean(provider.get("description"))
+        for processing in provider.get("processings", []) or []:
+            name = (processing.get("name") or processing.get("id")
+                    or provider_name)
+            purpose = _clean(processing.get("description")
+                             or processing.get("name") or provider_desc)
+            cat_raw = processing.get("categoryId", "")
+            category = _EPAAS_CATEGORY_MAP.get(cat_raw, cat_raw or "")
 
-    for c in d.get("cookies", []) or []:
+            cookies: list[dict] = []
-        pid = str(c.get("providerId") or c.get("provider") or c.get("vendor") or "")
+            for c in processing.get("persistences", []) or []:
-        if pid:
+                cookies.append({
-            cookies_by_provider.setdefault(pid, []).append({
+                    "name": c.get("name") or c.get("id") or "",
-                "name": c.get("name") or c.get("id") or "",
+                    "purpose": _clean(c.get("description")),
-                "purpose": _clean(c.get("purpose") or c.get("description")),
+                    "expiry": _clean(c.get("expiry")),
-                "expiry": _clean(c.get("expiry") or c.get("retention") or c.get("persistence")),
+                    "is_third_party": True,
-                "is_third_party": bool(c.get("isThirdParty") or c.get("third_party")),
+                })
+
+            display_name = (f"{provider_name} — {name}"
+                            if name and name != provider_name
+                            else (provider_name or name))
+            out.append({
+                "name": display_name,
+                "country": "",  # ePaaS doesn't surface vendor country
+                "purpose": purpose,
+                "category": category,
+                "opt_out_url": (processing.get("optOutLink") or "").strip(),
+                "privacy_policy_url": (processing.get("privacyPolicyLink")
+                                       or "").strip(),
+                "persistence": "",
+                "cookies": cookies,
             })
-
-    for p in providers:
-        pid = str(p.get("id") or p.get("vendorId") or p.get("name") or "")
-        cookies = cookies_by_provider.get(pid, []) or [{
-            "name": c.get("name", ""),
-            "purpose": _clean(c.get("purpose")),
-            "expiry": _clean(c.get("expiry") or c.get("persistence")),
-            "is_third_party": True,
-        } for c in (p.get("cookies", []) or [])]
-        out.append({
-            "name": p.get("name") or pid or "",
-            "country": (p.get("country") or "").strip(),
-            "purpose": _clean(p.get("purpose")),
-            "category": (p.get("category") or "").strip(),
-            "opt_out_url": (p.get("optOutUrl") or p.get("optoutUrl")
-                            or p.get("opt_out_url") or "").strip(),
-            "privacy_policy_url": (p.get("policyUrl") or p.get("policy_url")
-                                   or p.get("privacyPolicyUrl") or "").strip(),
-            "persistence": _clean(p.get("persistencePurposeDescription")),
-            "cookies": cookies,
-        })
     return out