breakpilot-compliance/backend-compliance/compliance/services/website_scanner.py

"""
Website Scanner — scans multiple pages of a website for third-party services,
chatbots, tracking, AI indicators, and compares against privacy policy.

Used by the Compliance Agent for SOLL/IST analysis.
"""

import logging
import re
from dataclasses import dataclass, field
from urllib.parse import urljoin, urlparse

import httpx

logger = logging.getLogger(__name__)

USER_AGENT = "BreakPilot-Compliance-Agent/1.0"


@dataclass
class DetectedService:
    id: str
    name: str
    category: str  # "tracking", "chatbot", "cdn", "payment", "marketing", "other"
    provider: str
    country: str
    eu_adequate: bool
    requires_consent: bool
    legal_ref: str
    found_on: str = ""  # URL where detected


@dataclass
class ScanResult:
    pages_scanned: list[str] = field(default_factory=list)
    detected_services: list[DetectedService] = field(default_factory=list)
    ai_mentions: list[str] = field(default_factory=list)
    chatbot_detected: bool = False
    chatbot_provider: str = ""
    missing_pages: dict = field(default_factory=dict)  # url -> status_code


# ── Service Registry (imported from master) ──────────────────────────────────
from compliance.services.service_registry import SERVICE_REGISTRY  # noqa: E402

AI_TEXT_PATTERNS = [
    r"k(?:ue|ü)nstliche.?intelligenz",
    r"artificial.?intelligence",
    r"machine.?learning",
    r"maschinelles.?lernen",
    r"KI.?gest(?:ue|ü)tzt",
    r"AI.?powered",
    r"chatgpt|openai",
    r"deep.?learning",
    r"neural.?net",
    r"automatisierte.?entscheidung",
]

FOOTER_LINK_PATTERNS = [
    (r'href="([^"]*(?:impressum|imprint|legal-notice)[^"]*)"', "impressum"),
    (r'href="([^"]*(?:datenschutz|privacy|dsgvo)[^"]*)"', "datenschutz"),
    (r'href="([^"]*(?:agb|terms|nutzungsbedingungen)[^"]*)"', "agb"),
    (r'href="([^"]*(?:cookie)[^"]*)"', "cookies"),
]


async def scan_website(base_url: str) -> ScanResult:
    """Scan a website: start page + footer links for services and AI indicators."""
    result = ScanResult()
    parsed = urlparse(base_url)
    origin = f"{parsed.scheme}://{parsed.netloc}"

    async with httpx.AsyncClient(timeout=10.0, follow_redirects=True) as client:
        # 1. Fetch start page
        start_html = await _fetch_page(client, origin, result)
        if not start_html:
            return result

        # 2. Discover footer links
        page_urls = {origin}
        page_urls.add(base_url)  # Also scan the provided URL
        for pattern, _ in FOOTER_LINK_PATTERNS:
            for match in re.finditer(pattern, start_html, re.IGNORECASE):
                href = match.group(1)
                if href.startswith("/"):
                    href = urljoin(origin, href)
                if href.startswith(origin):
                    page_urls.add(href)

        # 3. Scan all pages (max 10)
        for url in list(page_urls)[:10]:
            html = start_html if url == origin else await _fetch_page(client, url, result)
            if html:
                _detect_services(html, url, result)
                _detect_ai_mentions(html, url, result)

    # Deduplicate services
    seen = set()
    unique = []
    for svc in result.detected_services:
        if svc.id not in seen:
            seen.add(svc.id)
            unique.append(svc)
    result.detected_services = unique

    result.chatbot_detected = any(s.category == "chatbot" for s in result.detected_services)
    if result.chatbot_detected:
        result.chatbot_provider = next(
            s.name for s in result.detected_services if s.category == "chatbot"
        )

    return result


async def _fetch_page(
    client: httpx.AsyncClient, url: str, result: ScanResult,
) -> str:
    """Fetch a single page. Returns HTML or empty string on failure."""
    try:
        resp = await client.get(url, headers={"User-Agent": USER_AGENT})
        result.pages_scanned.append(url)
        if resp.status_code >= 400:
            result.missing_pages[url] = resp.status_code
            return ""
        return resp.text
    except Exception as e:
        logger.warning("Failed to fetch %s: %s", url, e)
        return ""


def _detect_services(html: str, url: str, result: ScanResult) -> None:
    """Detect third-party services in HTML."""
    for pattern, meta in SERVICE_REGISTRY.items():
        if re.search(pattern, html, re.IGNORECASE):
            result.detected_services.append(DetectedService(
                found_on=url, **meta,
            ))


def _detect_ai_mentions(html: str, url: str, result: ScanResult) -> None:
    """Detect AI/ML text mentions in page content."""
    # Strip scripts/styles first for text-only search
    clean = re.sub(r"<(script|style)[^>]*>.*?</\1>", "", html, flags=re.DOTALL | re.IGNORECASE)
    clean = re.sub(r"<[^>]+>", " ", clean)
    for pattern in AI_TEXT_PATTERNS:
        match = re.search(pattern, clean, re.IGNORECASE)
        if match:
            context = clean[max(0, match.start() - 40):match.end() + 40].strip()
            result.ai_mentions.append(f"{url}: ...{context}...")