CVMail Jobs API.

import requests
from urllib.parse import urlparse

# e.g. https://fsr.cvmailuk.com/rpc/main.cfm?page=jobBoard&fo=1
board_url = "https://fsr.cvmailuk.com/rpc/main.cfm?page=jobBoard&fo=1"
parts = urlparse(board_url)

host = parts.netloc.lower()
segments = [s for s in parts.path.split("/") if s]
assert host.endswith((".cvmailuk.com", ".cvmail.net")), "not a CVMail host"
assert len(segments) == 2 and segments[1].lower() == "main.cfm", "not a board URL"

tenant = segments[0].lower()   # "rpc"
print(tenant)

from bs4 import BeautifulSoup
from urllib.parse import urljoin, urlparse, parse_qs

HEADERS = {"Accept": "text/html"}

def job_id_from(url):
    q = parse_qs(urlparse(url).query)
    if q.get("page", [""])[0] != "jobSpecific":
        return None
    jid = q.get("jobId", [""])[0]
    # jobId must be a clean alphanumeric/_/- token to be a valid key
    return jid if jid and jid.replace("-", "").replace("_", "").isalnum() else None

def parse_listings(board_url):
    html = requests.get(board_url, headers=HEADERS, timeout=15).text
    soup = BeautifulSoup(html, "html.parser")
    jobs, seen = [], set()
    for table in soup.select("table"):
        if not table.select("a.jobMoreDetailCaptionStyle[href]"):
            continue
        headers = [h.get_text(" ", strip=True)
                   for h in table.select("td.jbTableHeaderCaptionStyle")]
        for row in table.select("tr"):
            a = row.select_one("a.jobMoreDetailCaptionStyle[href]")
            if not a:
                continue
            detail_url = urljoin(board_url, a["href"])
            jid = job_id_from(detail_url)
            if not jid or jid in seen:
                continue
            seen.add(jid)
            values = [c.get_text(" ", strip=True) for c in row.select("td.jbTableTextStyle")]
            fields = {h: v for h, v in zip(headers, values) if h and v}
            jobs.append({
                "external_id": jid,
                "title": a.get_text(" ", strip=True),
                "detail_url": detail_url,
                "location": fields.get("Location"),
                "closing_date": fields.get("Closing date") or fields.get("Close date"),
                "department": fields.get("Department or practice area") or fields.get("Department"),
                "salary": fields.get("Salary"),
                "reference": fields.get("Reference"),
            })
    return jobs

def job_identity(detail_url):
    q = parse_qs(urlparse(detail_url).query)
    if q.get("page", [""])[0] != "jobSpecific":
        return None
    jid = q.get("jobId", [""])[0]
    if not jid:
        return None
    return {"external_id": jid, "record_id": q.get("rcd", [None])[0]}

print(job_identity(
    "https://fsr.cvmailuk.com/rpc/main.cfm?page=jobSpecific&jobId=78306&rcd=1176178"))
# {'external_id': '78306', 'record_id': '1176178'}

def value_for(label_el):
    # value cell is usually in the same row as its label...
    row = label_el.find_parent("tr")
    if row:
        v = row.select_one("td.jobValueStyle")
        if v:
            return v
    # ...but Description often sits one row down inside a nested table
    table = label_el.find_parent("table")
    return table.select_one("td.jobValueStyle") if table else None

def fetch_detail(detail_url):
    resp = requests.get(detail_url, headers=HEADERS, timeout=15)
    if resp.status_code in (404, 410):
        return None  # CVMail returns 404/410 once a posting is pulled
    resp.raise_for_status()
    soup = BeautifulSoup(resp.text, "html.parser")

    fields, description = {}, None
    for label in soup.select("td.jobFieldStyle"):
        key = label.get_text(" ", strip=True)
        val = value_for(label)
        if not key or val is None:
            continue
        if key.lower() == "description":
            description = val.decode_contents().strip()
        else:
            text = val.get_text(" ", strip=True)
            if text:
                fields[key] = text

    title = fields.get("Job Title")
    if not title:
        og = soup.select_one('meta[property="og:title"]')
        title = og["content"].strip() if og and og.get("content") else None

    apply_a = next((a for a in soup.select("a.jobSpecificActionPaneItemStyle[href]")
                    if a.get_text(strip=True).lower() == "apply"), None)
    title_el = soup.select_one("title")
    return {
        "title": title,
        "description": description,
        "location": fields.get("Location"),
        "salary": fields.get("Salary"),
        "contract_type": fields.get("Contract type"),
        "reference": fields.get("Reference"),
        "closing_date": fields.get("Closing date") or fields.get("Close date"),
        "apply_url": urljoin(detail_url, apply_a["href"]) if apply_a else detail_url,
        "company": title_el.get_text(strip=True) if title_el else None,
    }

def snapshot_status(board_html):
    soup = BeautifulSoup(board_html, "html.parser")
    text = soup.get_text(" ", strip=True).lower()

    if any(m in text for m in
           ("no current vacancies", "no vacancies found", "0 vacancies")):
        return "empty"   # complete: the board is genuinely empty

    has_next = (
        soup.select_one("form[name='paging'] input[type='submit'][value*='Next' i]")
        or soup.select_one("form[name='paging'] button[value*='Next' i]")
        or soup.select_one("a[href*='start_row=']")
    )
    # Complete only when a job table rendered AND no Next control is present.
    return "incomplete" if has_next is not None else "complete"