HealthBoxHR Jobs API.

import re
import requests
from urllib.parse import urlparse

HOSTS = ("app.hbhr.io", "app.healthboxhr.com")
COMPANY_ID = re.compile(r"^/jobs-feed/web/company/([a-f0-9]{13})/?$", re.I)

def resolve_feed(board_url):
    parts = urlparse(board_url)
    host = parts.netloc.lower()
    assert host in HOSTS, "not a HealthBoxHR host"
    m = COMPANY_ID.match(parts.path)
    assert m, "not a company-feed URL"
    company_id = m.group(1).lower()   # 13-hex tenant key, kept verbatim
    return f"https://{host}/jobs-feed/web/company/{company_id}", company_id

from bs4 import BeautifulSoup

HEADERS = {"Accept": "text/html"}
FINGERPRINT_FORM = "form[action*='/job-application/external/store/'] input[name='hash']"

def is_provider_document(soup):
    author = soup.select_one("meta[name='author']")
    if author and (author.get("content") or "").strip().lower() == "healthbox software":
        return True
    return soup.select_one(FINGERPRINT_FORM) is not None

def fetch_feed(feed_url):
    html = requests.get(feed_url, headers=HEADERS, timeout=30).text
    soup = BeautifulSoup(html, "html.parser")
    assert is_provider_document(soup), "missing HealthBoxHR document fingerprint"
    return soup

from urllib.parse import urljoin

JOB_ID = re.compile(r"/job-application/external/([a-f0-9]{8}(?:-[a-f0-9]{8}){3})", re.I)

def parse_listings(soup, feed_url, company_id):
    jobs = []
    for card in soup.select(".card"):
        anchor = (card.select_one(".card-header h3 a[href*='/job-application/external/']")
                  or card.select_one("a[href*='/job-application/external/']"))
        if not anchor:
            continue
        url = urljoin(feed_url, anchor.get("href", ""))
        m = JOB_ID.search(url)
        title = anchor.get_text(strip=True)
        if not m or not title:
            continue
        location = card.select_one(".card-body.vacancy-item > h5")
        summary = card.select_one(".card-body.vacancy-item .mt-3")
        jobs.append({
            "external_id": m.group(1).lower(),      # segmented-hex job key
            "company_id": company_id,               # carry tenant into detail
            "detail_url": url.split("?")[0],
            "title": title,
            "location": location.get_text(strip=True) if location else None,
            "summary": summary.get_text(strip=True) if summary else None,
        })
    if not jobs and "There are no job vacancies found." in soup.get_text():
        return []          # genuinely empty feed, not a parse failure
    return jobs

def fetch_detail(job):
    resp = requests.get(job["detail_url"], headers=HEADERS, timeout=30)
    if resp.status_code in (404, 410):
        return None                       # canonical application was removed
    resp.raise_for_status()
    soup = BeautifulSoup(resp.text, "html.parser")

    form_hash = soup.select_one(FINGERPRINT_FORM)
    value = form_hash.get("value", "").strip().lower() if form_hash else None
    if value != job["external_id"]:
        raise ValueError("application form hash did not match the native job ID")
    return soup

def parse_detail(soup, job):
    root = soup.select_one(".job-vacancy-details")
    assert root is not None, "missing vacancy-details root"

    fields = {}
    for div in root.select(":scope > .row > div"):
        label = div.select_one("h5.card-title")
        if not label:
            continue
        key = label.get_text(strip=True)
        if key in fields:                 # duplicate labels signal a layout change
            raise ValueError(f"duplicate field label: {key}")
        text = div.select_one("p.card-text")
        html = div.select_one("div.card-text")
        fields[key] = {
            "text": text.get_text(strip=True) if text else None,
            "html": html.decode_contents().strip() if html else None,
        }

    title = soup.select_one("h3")
    doc_title = soup.select_one("title")
    company = None
    if doc_title and "|" in doc_title.get_text():
        company = doc_title.get_text().rsplit("|", 1)[-1].strip()

    def value(name):
        return fields.get(name, {}).get("text")

    return {
        "external_id": job["external_id"],
        "title": title.get_text(strip=True) if title else None,
        "description": fields.get("Description", {}).get("html"),
        "employment_type": value("Employment Type"),
        "work_style": value("Work Style"),
        "salary": value("Salary"),
        "experience": value("Experience"),
        "location": value("Location"),
        "posted_at": value("Application Start Date"),
        "closes_at": value("Application End Date"),
        "company": company or job.get("company"),
    }