import json, os, sys, ssl, urllib.parse, urllib.request, urllib.error

BASE = "https://endpoints.office.com/endpoints/"
# A fixed client request id is acceptable for this public, unauthenticated API.
CLIENT_ID = "b10c5ed1-bad1-445f-b386-b919946339a7"


def _cfg():
    return json.loads(os.environ.get("INTEGRATION_SECRETS", "{}"))


def _inputs():
    return json.loads(os.environ.get("INTEGRATION_INPUTS", "{}"))


def _ctx(cfg):
    if cfg.get("insecure"):
        c = ssl.create_default_context()
        c.check_hostname = False
        c.verify_mode = ssl.CERT_NONE
        return c
    return None


def _get(url, cfg):
    req = urllib.request.Request(
        url, headers={"User-Agent": "Riposte-SOAR", "Accept": "application/json"}
    )
    with urllib.request.urlopen(req, timeout=120, context=_ctx(cfg)) as r:
        return r.read()


def _run(fn):
    try:
        print(json.dumps(fn(_cfg(), _inputs())))
    except urllib.error.HTTPError as e:
        print(json.dumps({"error": "HTTP " + str(e.code), "detail": e.read().decode("utf-8", "replace")}))
        sys.exit(1)
    except Exception as e:
        print(json.dumps({"error": str(e)}))
        sys.exit(1)


def main(cfg, inputs):
    instance = str(inputs.get("instance") or cfg.get("instance") or "Worldwide")
    include_urls = inputs.get("include_urls")
    if include_urls is None:
        include_urls = cfg.get("include_urls")
    if include_urls is None:
        include_urls = True

    url = BASE + urllib.parse.quote(instance, safe="") + "?" + urllib.parse.urlencode({"clientrequestid": CLIENT_ID})
    raw = _get(url, cfg)
    data = json.loads(raw) if raw else []
    if not isinstance(data, list):
        data = []

    maxn = int(inputs.get("max_indicators") or 0)
    seen = set()
    out = []

    def add(value, typ, area):
        key = (typ, value)
        if key in seen:
            return False
        seen.add(key)
        out.append({"value": value, "type": typ, "service_area": area, "provider": "microsoft365"})
        return not (maxn and len(out) >= maxn)

    for ep in data:
        if not isinstance(ep, dict):
            continue
        area = ep.get("serviceArea")
        for cidr in ep.get("ips", []) or []:
            typ = "cidr" if "/" in str(cidr) else "ip"
            if not add(str(cidr), typ, area):
                return {"source": "microsoft365:" + instance, "count": len(out), "indicators": out}
        if include_urls:
            for u in ep.get("urls", []) or []:
                dom = str(u).lstrip("*.").strip()
                if not dom:
                    continue
                if not add(dom, "domain", area):
                    return {"source": "microsoft365:" + instance, "count": len(out), "indicators": out}

    return {"source": "microsoft365:" + instance, "count": len(out), "indicators": out}


_run(main)