oneCruiter / Workbuster Jobs API.

import requests
from bs4 import BeautifulSoup

# One tenant = one board. The entire collection is in this single document.
board_url = "https://trelleborg.onecruiter.com/"

resp = requests.get(board_url, headers={"Accept": "text/html"}, timeout=30)
resp.raise_for_status()
html = resp.text

soup = BeautifulSoup(html, "html.parser")
cards = soup.select(".position-container.filter-item")
print(f"Found {len(cards)} job cards")

import re

# accountId can live in either config object; siteId lives in window.customer.
account_re = re.compile(
    r"(?:window\.MilouConfig\s*=\s*\{|window\.customer\s*=\s*\{)"
    r"[\s\S]{0,500}?accountId\s*:\s*(\d+)"
)
site_re = re.compile(r"window\.customer\s*=\s*\{[\s\S]{0,500}?siteId\s*:\s*(\d+)")

m_account = account_re.search(html)
m_site = site_re.search(html)
account_id = m_account.group(1) if m_account else None
site_id = m_site.group(1) if m_site else None

# The tenant key is exactly accountId:siteId (never hashed or remapped).
tenant_id = f"{account_id}:{site_id}" if account_id and site_id else None
print("tenant:", tenant_id)

from urllib.parse import urljoin

JOB_ROUTE = re.compile(r"/jobs/(\d+)-[^/?#]+")

def parse_card(card, base_url):
    anchor = card.select_one(
        "a.position-title[href*='/jobs/'], a.position-link[href*='/jobs/']"
    )
    if anchor is None:
        return None
    href = urljoin(base_url, anchor.get("href", ""))
    match = JOB_ROUTE.search(href)
    if not match:
        return None

    # Overlay template: anchor text is empty, title sits in .head-title.
    title = anchor.get_text(strip=True)
    if not title:
        head = card.select_one(".head-title")
        title = head.get_text(strip=True) if head else None
    if not title:
        return None

    location = card.select_one(".meta-container .location.tag")
    department = card.select_one(".meta-container .department.tag")
    closes = card.select_one(".last-apply-date")
    canonical = href.split("?")[0].split("#")[0]  # drop tracking params

    return {
        "external_id": match.group(1),
        "url": canonical,
        "apply_url": canonical,
        "title": title,
        "location": location.get_text(strip=True) if location else None,
        "department": department.get_text(strip=True) if department else None,
        "closes_at": closes.get_text(strip=True) if closes else None,
    }

jobs = [j for j in (parse_card(c, board_url) for c in cards) if j]
print(f"Parsed {len(jobs)} jobs")

def fetch_details(job):
    resp = requests.get(job["url"], headers={"Accept": "text/html"}, timeout=30)
    if resp.status_code in (404, 410):
        return None  # canonical position was removed
    resp.raise_for_status()

    page = resp.text
    soup = BeautifulSoup(page, "html.parser")

    # positionId is embedded in window.customer; it must equal the route ID.
    pos = re.search(r"window\.customer\s*=\s*\{[\s\S]{0,700}?positionId\s*:\s*(\d+)", page)
    if not pos or pos.group(1) != job["external_id"]:
        return None  # identity mismatch

    title_el = soup.select_one(
        "h1.position-title, .position-title h1.title, .position-title .title"
    )
    desc_el = soup.select_one(".position-description .job-text")
    location_el = soup.select_one(".tag-container.locations .tag")
    closes_el = soup.select_one(".tag-container.last_application_at .tag")
    recruiter_el = soup.select_one(".responsible-fullname")

    return {
        **job,
        "title": title_el.get_text(strip=True) if title_el else job["title"],
        "description_html": desc_el.decode_contents().strip() if desc_el else None,
        "location": location_el.get_text(strip=True) if location_el else job["location"],
        "closes_at": closes_el.get_text(strip=True) if closes_el else job["closes_at"],
        "recruiter": recruiter_el.get_text(strip=True) if recruiter_el else None,
    }

detailed = [d for d in (fetch_details(j) for j in jobs) if d]
print(f"Enriched {len(detailed)} jobs")