Reach ATS Jobs API.

from urllib.parse import urlparse

def origin_of(board_url):
    parts = urlparse(board_url)
    return f"{parts.scheme}://{parts.netloc}"

def pick_adapter(board_url):
    host = urlparse(board_url).netloc.lower()
    if host == "careers.lcp.com":
        return fetch_surface_html      # anti-forgery POST tenant
    if host == "jobs.sense.org.uk":
        return fetch_snippet_html      # asp.html fragment tenant
    return fetch_json_helper           # JSON helper tenant (e.g. Bellway)

import requests

def fetch_json_helper(board_url):
    origin = origin_of(board_url)
    resp = requests.get(
        f"{origin}/?action=helper/a-t-s/get-vacancies",
        headers={"Accept": "application/json"},
        timeout=30,
    )
    resp.raise_for_status()
    data = resp.json()
    if not data.get("success") or not isinstance(data.get("vacancies"), list):
        raise RuntimeError("Reach ATS omitted its successful vacancies array")

    jobs = []
    for row in data["vacancies"]:
        vid = str(row.get("id", "")).strip()
        if not vid.isdigit():
            continue  # rows without a numeric native id are rejected
        jobs.append({
            "external_id": vid,
            "title": row.get("title"),
            "location": row.get("location"),
            "employment_type": row.get("type"),
            "salary": row.get("salaryDescription"),
            "closing_date": row.get("closingDate"),
            "listing_url": f"{origin}/vacancies/{vid}",
            "apply_url": row.get("applyUrl") or f"{origin}/vacancies/{vid}",
        })
    return jobs

import json
from bs4 import BeautifulSoup

def fetch_surface_html(board_url, page=1):
    origin = origin_of(board_url)
    session = requests.Session()

    # The anti-forgery cookie and token are issued together by the board page.
    boot = session.get(board_url, timeout=30)
    boot.raise_for_status()
    field = BeautifulSoup(boot.text, "html.parser").select_one(
        "input[name='__RequestVerificationToken']")
    if not field or not field.get("value"):
        raise RuntimeError("LCP board omitted its request verification token")

    resp = session.post(  # session jar re-sends the anti-forgery cookie
        f"{origin}/umbraco/surface/jobvacancies/GetJobVacancies",
        headers={
            "Content-Type": "application/json; charset=utf-8",
            "RequestVerificationToken": field["value"],
            "Origin": origin,
            "Referer": board_url,
        },
        data=json.dumps({"query": "", "categories": "", "specialism": "",
                         "location": "", "page": page}),
        timeout=30,
    )
    resp.raise_for_status()  # HTTP 429 -> back off and retry

    fragment = BeautifulSoup(resp.text, "html.parser")
    jobs = []
    for card in fragment.select(".job-vacancies__job-container"):
        anchor = card.select_one("a.job-vacancies__job-more-button[href*='id=']")
        title = card.select_one(".job-vacancies__job-title")
        if not anchor or not title:
            continue
        jobs.append({
            "detail_href": anchor.get("href"),
            "title": title.get_text(strip=True),
        })
    return jobs

import re

def fetch_snippet_html(board_url, page=1):
    origin = origin_of(board_url)
    resp = requests.post(
        f"{origin}/asp.html?snippet=jobListingREACH",
        headers={
            "Content-Type": "application/x-www-form-urlencoded; charset=UTF-8",
            "Origin": origin,
            "Referer": board_url,
        },
        data=(f"descLen=400&showResults=1&showSearch=0&page={page}"
              "&keywords=&region=&role=&fields="),
        timeout=30,
    )
    resp.raise_for_status()

    soup = BeautifulSoup(resp.text, "html.parser")
    jobs = []
    for card in soup.select(".edge-tertiary"):
        anchor = card.select_one("a[href*='jobid=']")
        title = card.select_one("h2")
        if not anchor or not title:
            continue
        job_id = re.search(r"jobid=(\d+)", anchor["href"], re.I)
        jobs.append({
            "external_id": job_id.group(1) if job_id else None,
            "title": title.get_text(strip=True),
            "detail_href": anchor["href"],  # resolve against origin, not /find-a-job/
        })

    # Stop paging once the reported total is covered (10 results per page).
    total = re.search(r"showing\s+\d+\s*-\s*\d+\s+of\s+(\d+)",
                      soup.get_text(), re.I)
    total_pages = -(-int(total.group(1)) // 10) if total else None
    return jobs, total_pages

def fetch_details(listing_url):
    resp = requests.get(listing_url, timeout=30)
    if resp.status_code in (404, 410):
        return None  # vacancy was removed
    resp.raise_for_status()

    soup = BeautifulSoup(resp.text, "html.parser")
    title = soup.select_one("h1, [class*='vacancy-title'], [class*='job-title']")
    body = soup.select_one(
        "[class*='vacancy-description'], [class*='job-description'], article, main")
    if not body:
        raise RuntimeError("Reach ATS detail page omitted its description")
    apply = soup.select_one("a[href*='apply']")
    return {
        "title": title.get_text(strip=True) if title else None,
        "description_html": body.decode_contents().strip(),
        "apply_url": apply.get("href") if apply else listing_url,
    }