fix: Banner runner maps violations by text when code field is missing

The consent-tester produces violations without a 'code' field — only
text, severity, service. The runner now infers check_keys from the
violation text content (36 text→code mappings). This fixes the 100%
false-pass for safetykon.de which had 3 real violations (impressum,
re-access, color contrast dark pattern) that were silently ignored.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
Benjamin Admin
2026-05-10 08:25:50 +02:00
parent f07c4db164
commit 561150b5a8
+53 -4
View File
@@ -97,28 +97,77 @@ def map_scan_to_checks(scan_result: dict) -> dict:
} }
_TEXT_TO_CODE: list[tuple[str, str]] = [
("impressum", "impressum_link"),
("erneuter zugang", "re_access_settings"),
("cookie-einstellung", "re_access_settings"),
("widerruf der einwilligung", "re_access_settings"),
("vorausgewaehlte", "pre_ticked_checkboxes"),
("vorausgew", "pre_ticked_checkboxes"),
("akzeptieren.*groesser", "dark_pattern_button_size"),
("akzeptieren.*gr\u00f6\u00dfer", "dark_pattern_button_size"),
("hintergrundfarbe", "color_contrast_dark_pattern"),
("optisch kaum sichtbar", "color_contrast_dark_pattern"),
("dark pattern", "color_contrast_dark_pattern"),
("cookie wall", "cookie_wall"),
("ablehnen.*button", "reject_button_visible"),
("kein sichtbarer", "reject_button_visible"),
("zustimmung zur datenschutz", "wrong_dse_consent"),
("consent mode", "google_consent_mode_defaults"),
("tracking.*vor consent", "cookies_before_consent"),
("tracking-cookie", "cookies_before_consent"),
("nicht modal", "non_modal_dismiss"),
("hintergrund.*schliessen", "non_modal_dismiss"),
("klick.*asymmetri", "click_count_asymmetry"),
("ablehnung.*klick", "click_count_asymmetry"),
("koppelungsverbot", "registration_consent_coupling"),
("registrierung", "registration_consent_coupling"),
("sprache.*stimmt nicht", "banner_language_mismatch"),
("banner-sprache", "banner_language_mismatch"),
("consent-cookie.*laeuft", "consent_cookie_expiry_13m"),
("consent-cookie.*l\u00e4uft", "consent_cookie_expiry_13m"),
("13 monate", "consent_cookie_expiry_13m"),
("nudging", "nudging_reject_hidden"),
("scrollen", "nudging_reject_hidden"),
("emotionale sprache", "stirring_emotional_language"),
("stirring", "stirring_emotional_language"),
("drittanbieter.*dse", "third_party_dse_link"),
("ohne vorherige einwilligung", "tracking_before_consent"),
("trotz ablehnung", "tracking_after_reject"),
]
def _text_to_code(text: str) -> str:
"""Infer a check_key from violation text content."""
t = text.lower()
for pattern, code in _TEXT_TO_CODE:
if pattern in t:
return code
return ""
def _collect_violation_codes(scan: dict) -> dict[str, str]: def _collect_violation_codes(scan: dict) -> dict[str, str]:
"""Collect check_key → violation text from all sources.""" """Collect check_key → violation text from all sources."""
codes: dict[str, str] = {} codes: dict[str, str] = {}
# Banner text violations # Banner text violations — match by code field OR by text content
banner_checks = scan.get("banner_checks", {}) banner_checks = scan.get("banner_checks", {})
for v in banner_checks.get("violations", []): for v in banner_checks.get("violations", []):
code = v.get("code", "") code = v.get("code", "") or _text_to_code(v.get("text", ""))
if code: if code:
codes[code] = v.get("text", "")[:120] codes[code] = v.get("text", "")[:120]
# Phase A violations (before consent) # Phase A violations (before consent)
phase_a = scan.get("phases", {}).get("before_consent", {}) phase_a = scan.get("phases", {}).get("before_consent", {})
for v in phase_a.get("violations", []): for v in phase_a.get("violations", []):
code = v.get("code", "") code = v.get("code", "") or _text_to_code(v.get("text", ""))
if code: if code:
codes[code] = v.get("text", "")[:120] codes[code] = v.get("text", "")[:120]
# Phase B violations (after reject) # Phase B violations (after reject)
phase_b = scan.get("phases", {}).get("after_reject", {}) phase_b = scan.get("phases", {}).get("after_reject", {})
for v in phase_b.get("violations", []): for v in phase_b.get("violations", []):
code = v.get("code", "") code = v.get("code", "") or _text_to_code(v.get("text", ""))
if code: if code:
codes[code] = v.get("text", "")[:120] codes[code] = v.get("text", "")[:120]