feat(consent-tester): Phase A — generic JSON cookie-policy heuristic

New module cmp_heuristic.py with:
- looks_like_cookie_policy(data): shape-based classifier (top-level keys
  cookies/categories/providers/vendors/purposes/cookieList/etc. + at
  least 2 name+description objects, or IAB TCF v2 vendors[]+purposes[])
- reconstruct_generic(data): walks JSON, extracts name + description
  fields + standalone prologue/dataController/persistence fields,
  emits flat German Markdown text (max 5000 words, dedup)

cmp_extractor.py wired so that AFTER named CMP matchers (epaas,
onetrust) fail, every JSON response on the page is tested for the
heuristic. If matched, payload is captured as '_heuristic' kind and
reconstructed via the generic walker.

This is Phase A of the 4-stage cascade (B-D follow). Unknown CMPs that
return JSON now work without hand-coding each one.

Pre-filter: skips response paths /api/config, /beacon, /track,
/analytics, /fonts/, /log/, /heartbeat/, /.well-known/ to avoid
spamming the heuristic on every Playwright load.
This commit is contained in:
Benjamin Admin
2026-05-16 22:56:20 +02:00
parent 9814b56f2f
commit 8283483909
2 changed files with 235 additions and 11 deletions
+44 -11
View File
@@ -52,24 +52,40 @@ class CMPCapture:
async def _on_response(self, response: Response) -> None:
try:
url = response.url
if response.status != 200:
return
# 1) Named CMP matchers (highest quality)
for cmp_name, pattern in _MATCHERS:
if pattern.search(url):
if response.status != 200:
logger.info("CMP %s response %s (%d) — skipped",
cmp_name, url[:120], response.status)
data = await _parse_json_response(response)
if data is None:
return
try:
data = await response.json()
except Exception:
body = await response.body()
try:
data = json.loads(body.decode("utf-8", errors="ignore"))
except Exception:
return
self.payloads.append((cmp_name, data))
logger.info("CMP captured: %s (%s, ~%dKB)",
cmp_name, url[:120], len(json.dumps(data)) // 1024)
return
# 2) Generic shape-based heuristic for unknown CMPs.
# Only consider JSON responses ≥1KB (skip small config blobs).
content_type = (response.headers.get("content-type") or "").lower()
if "json" not in content_type:
return
# Cheap pre-filter: skip noisy paths (analytics, fonts, etc.)
url_lower = url.lower()
if any(skip in url_lower for skip in (
"/api/config", "/beacon", "/track", "/analytics",
"/fonts/", "/log/", "/heartbeat", "/.well-known/",
)):
return
data = await _parse_json_response(response)
if data is None:
return
from services.cmp_heuristic import looks_like_cookie_policy
if looks_like_cookie_policy(data):
self.payloads.append(("_heuristic", data))
logger.info("CMP captured: _heuristic (%s, ~%dKB)",
url[:120], len(json.dumps(data)) // 1024)
except Exception as e:
logger.debug("CMP listener error: %s", e)
@@ -77,7 +93,10 @@ class CMPCapture:
"""Build a single Cookie-Policy text from all captured payloads.
Returns empty string if nothing was captured or reconstruction fails.
Named CMPs take precedence over the generic heuristic (richer output).
"""
from services.cmp_heuristic import reconstruct_generic
parts: list[str] = []
for cmp_name, data in self.payloads:
try:
@@ -85,11 +104,25 @@ class CMPCapture:
parts.append(_reconstruct_epaas(data))
elif cmp_name == "onetrust":
parts.append(_reconstruct_onetrust(data))
elif cmp_name == "_heuristic":
parts.append(reconstruct_generic(data))
except Exception as e:
logger.warning("CMP %s reconstruction failed: %s", cmp_name, e)
return "\n\n".join(p for p in parts if p)
async def _parse_json_response(response: Response) -> dict | None:
"""Best-effort JSON parse from a Playwright Response."""
try:
return await response.json()
except Exception:
try:
body = await response.body()
return json.loads(body.decode("utf-8", errors="ignore"))
except Exception:
return None
def _reconstruct_epaas(d: dict) -> str:
"""Build a German Cookie-Policy from BMW ePaaS policy JSON.