Lumesse TalentLink Jobs API.

import re
import requests

# The site technical ID and regional API host come from the TalentLink
# widget embedded on the corporate careers page.
CAREERS_URL = "https://www.unitedutilities.com/corporate/careers/jobs/"

html = requests.get(CAREERS_URL, timeout=30).text
match = re.search(r'data-talentlink-fo-site-tech-id="([A-Z0-9]{20,})"', html)
SITE_ID = match.group(1) if match else "PGLFK026203F3VBQB688MF6CR"

# Regional host varies per tenant, e.g. emea3.recruitmentplatform.com or
# emea5-foc.lumessetalentlink.com. Confirm it from the widget config.
API_HOST = "https://emea3.recruitmentplatform.com"
LANGUAGE = "UK"  # the board language passed in the lumesse-language header

def fetch_page(offset=0, page_size=12):
    url = f"{API_HOST}/fo/rest/jobs"
    params = {
        "firstResult": offset,
        "maxResults": page_size,
        "sortBy": "DPOSTINGSTART",
        "sortOrder": "desc",
    }
    headers = {
        "Accept": "application/json",
        "username": f"{SITE_ID}:guest:FO",
        "password": "guest",
        "lumesse-language": LANGUAGE,
        "Referer": CAREERS_URL,
    }
    resp = requests.post(
        url, params=params, headers=headers,
        json={"searchCriteria": {}}, timeout=30,
    )
    resp.raise_for_status()
    return resp.json()

data = fetch_page()
total = data["globals"]["jobsCount"]
print(f"{total} vacancies on this board")

def strip_html(text):
    return re.sub(r"<[^>]+>", "", text or "").strip()

def first(fields, *names):
    for name in names:
        value = fields.get(name)
        if value:
            return value
    return None

def parse_job(row):
    fields = row.get("jobFields", {})
    sections = []
    for cf in row.get("customFields", []):
        content = strip_html(cf.get("content"))
        if content:
            heading = (cf.get("title") or "").strip()
            sections.append(f"{heading}\n{content}" if heading else content)
    return {
        "id": row.get("id"),
        "title": first(fields, "jobTitle", "sJobTitle", "SJOBTITLE"),
        "apply_url": fields.get("applicationUrl"),
        "location": first(fields, "SLOCATION", "SLOVLIST2", "REGLABEL"),
        "salary": fields.get("SALARY"),
        "contract_type": first(fields, "CONTRACTTYPLABEL", "SLOVLIST208"),
        "department": first(fields, "SDPTNAMELEVEL1", "SDPTNAME"),
        "posted_at": first(fields, "DPOSTINGSTART", "postingStartDate"),
        "closes_at": first(fields, "DPOSTINGEND", "postingEndDate"),
        "description": "\n\n".join(sections),
    }

def scrape_all():
    jobs, seen, offset = [], set(), 0
    while True:
        data = fetch_page(offset)
        total = data["globals"]["jobsCount"]
        rows = data.get("jobs", [])
        if not rows:
            break
        for row in rows:
            job = parse_job(row)
            if job["id"] and job["id"] not in seen and len(job["description"]) >= 40:
                seen.add(job["id"])
                jobs.append(job)
        offset += len(rows)
        if offset >= total:
            break
    return jobs

all_jobs = scrape_all()
print(f"Collected {len(all_jobs)} vacancies")