Radancy / TalentBrew Jobs API.

import re

# The company ID is copied from the /job/ route and must match the board you target.
BOARDS = {
    "careers.ba.com": {"company_id": "22348", "listing_path": "/search-jobs/"},
    "careers.adeccogroup.com": {"company_id": "37746", "listing_path": "/en/search-jobs/"},
    "www.att.jobs": {"company_id": "117", "listing_path": "/search-jobs/"},
}

# Optional 2-letter locale, then /job/<segments>/<companyId>/<postingId>
JOB_ROUTE = re.compile(r"/(?:[a-z]{2}/)?job/(?:[^/?#]+/)+(\d+)/(\d+)/?$", re.IGNORECASE)

import requests
from urllib.parse import urljoin
from bs4 import BeautifulSoup

def parse_listings(company_id: str, page_url: str):
    html = requests.get(page_url, timeout=20).text
    soup = BeautifulSoup(html, "html.parser")
    root = soup.select_one("#search-results")
    if root is None:
        raise ValueError("Radancy page omitted #search-results")

    jobs, seen = [], set()
    for a in root.select("a[href*='/job/']"):
        job_url = urljoin(page_url, a.get("href", "")).rstrip("/")
        m = JOB_ROUTE.search(job_url)
        if not m or m.group(1) != company_id:
            continue  # reject wrong-company / lookalike routes
        posting_id = m.group(2)
        if posting_id in seen:
            continue
        seen.add(posting_id)
        row = a.find_parent("li") or a.parent
        title_el = row.select_one("h2, h3, .job-title") if row else None
        loc_el = row.select_one("[class*='location']") if row else None
        jobs.append({
            "external_id": posting_id,
            "company_id": company_id,
            "title": title_el.get_text(strip=True) if title_el else a.get_text(strip=True),
            "listing_url": job_url,
            "location": loc_el.get_text(strip=True) if loc_el else None,
        })
    return soup, root, jobs

def scrape_board(host: str, max_pages: int = 200):
    cfg = BOARDS[host]
    url = f"https://{host}{cfg['listing_path']}"
    all_jobs, pages = [], 0
    while url and pages < max_pages:
        soup, root, jobs = parse_listings(cfg["company_id"], url)
        all_jobs.extend(jobs)
        pages += 1

        total = root.get("data-total-job-results")
        if total is not None and len(all_jobs) >= int(total):
            break  # authoritative total reached

        nxt = soup.select_one(".pagination a.next[href]")
        url = urljoin(url, nxt["href"]) if nxt else None  # stop when no next link
    return all_jobs

import json

def fetch_details(job: dict):
    html = requests.get(job["listing_url"], timeout=20).text
    soup = BeautifulSoup(html, "html.parser")

    posting = None
    for script in soup.select("script[type='application/ld+json']"):
        if "JobPosting" in script.text:
            posting = json.loads(script.text)
            break
    if posting is None:
        raise ValueError("Radancy detail omitted JobPosting JSON-LD")

    meta = soup.select_one("meta[name='search-job-apply-url']")
    apply_url = meta["content"] if meta and meta.get("content") else job["listing_url"]

    return {
        "external_id": job["external_id"],
        "title": posting.get("title", job["title"]),
        "description": posting.get("description"),   # HTML — strip before storing
        "employment_type": posting.get("employmentType"),
        "hiring_organization": (posting.get("hiringOrganization") or {}).get("name"),
        "locations": posting.get("jobLocation"),     # list of address structs
        "posted_at": posting.get("datePosted"),
        "closes_at": posting.get("validThrough"),
        "apply_url": apply_url,
    }