"""Teamtailor career sites publish a JSON Feed at `/jobs.json` on the career-site host (`<token>.teamtailor.com` or a custom
`career.<company>.com`): {version, title, home_page_url, items:[{id, title, url, date_published, _jobposting:{identifier:{value}, datePosted,
jobLocation:[{address:{addressLocality, addressRegion, addressCountry}}], employmentType?, jobLocationType?}}]}.
Verified 2026-09-12 against https://career.teamtailor.com/jobs.json (fixture trimmed to 20 items, descriptions removed). When a career
site has no feed (404) discovery keeps the HTML careers sensor instead."""
from __future__ import annotations

import re
from collections.abc import Mapping
from typing import Any

from companyatlas.connectors._ats_base import AtsConnector
from companyatlas.connectors._util import country_code, dig, parse_date, text_of
from companyatlas.sdk.connector import ConnectorMeta, register
from companyatlas.sdk.models import ExtractedJob
from companyatlas.taxonomy import FetchMode, Surface


@register
class TeamtailorConnector(AtsConnector):
    vendor = "teamtailor"
    token_re = re.compile(r"https?://([a-z0-9-]+)\.teamtailor\.com/jobs\.json", re.IGNORECASE)
    meta = ConnectorMeta(connector_id="teamtailor-v1", name="Teamtailor jobs feed", version="1", category=Surface.JOBS_BOARD, fetch_mode=FetchMode.JSON,
                         default_interval_s=6 * 3600, url_pattern=r"/jobs\.json(\?|$)", pattern_required=True, priority=45, accept="application/feed+json,application/json",
                         description="Teamtailor career-site JSON Feed (/jobs.json)")

    def parse_jobs(self, data: Any, sensor: Mapping[str, Any]) -> list[ExtractedJob]:
        out: list[ExtractedJob] = []
        for it in (data.get("items") if isinstance(data, dict) else []) or []:
            if not isinstance(it, dict) or not it.get("title"):
                continue
            jp = it.get("_jobposting") or {}
            locs = jp.get("jobLocation") or []
            if isinstance(locs, dict):
                locs = [locs]
            addr = dig(locs[0], "address", default={}) if locs and isinstance(locs[0], dict) else {}
            city, region, country = text_of(addr.get("addressLocality")), text_of(addr.get("addressRegion")), addr.get("addressCountry")
            ident = dig(jp, "identifier", "value")
            loc_type = (jp.get("jobLocationType") or "").lower()
            out.append(ExtractedJob(
                title=str(it["title"]), url=it.get("url"), external_id=str(ident or it.get("id") or ""),
                location_text=", ".join(x for x in (city, country_code(country) or text_of(country)) if x) or None, city=city,
                region=region if region and region.lower() not in ("europe", "emea", "apac", "americas") else None, country=country_code(country),
                remote=True if loc_type == "telecommute" else None, employment_type=text_of(jp.get("employmentType")),
                posted_at=parse_date(jp.get("datePosted") or it.get("date_published")), raw={"locations": len(locs)},
            ))
        return out