VIVAHR / AvaHR Jobs API.

from urllib.parse import urlparse

PUBLIC_HOST = "jobs.avahr.com"

def split_id_slug(segment: str) -> tuple[str, str]:
    head, _, tail = segment.partition("-")
    if not head.isdigit() or not tail:
        raise ValueError(f"expected {{id}}-{{slug}}, got {segment!r}")
    return head, tail

def parse_avahr(url: str) -> dict:
    parsed = urlparse(url)
    if parsed.netloc.lower() != PUBLIC_HOST:
        raise ValueError("not an AvaHR host")

    parts = [p for p in parsed.path.strip("/").split("/") if p]
    if not parts:
        raise ValueError("no company segment")

    company_id, company_slug = split_id_slug(parts[0])
    job_id = None
    if len(parts) == 2 and parts[1] != "jobs":
        job_id, _ = split_id_slug(parts[1])
    elif len(parts) > 2:
        # /apply/ and other subroutes are not identities.
        raise ValueError("unsupported AvaHR subroute")

    return {
        "company_id": company_id,
        "company_slug": company_slug,
        "job_id": job_id,
        "board_url": f"https://{PUBLIC_HOST}/{company_id}-{company_slug}/jobs/",
    }

print(parse_avahr("https://jobs.avahr.com/8147-zzeeks-pizza/78012-delivery-driver"))

import requests
from bs4 import BeautifulSoup

session = requests.Session()
session.headers["Accept"] = "text/html,application/xhtml+xml"

def fetch_board(board_url: str) -> BeautifulSoup:
    resp = session.get(board_url, timeout=30)
    if resp.status_code in (404, 410):
        raise LookupError("this AvaHR board no longer exists")
    resp.raise_for_status()

    # First-party ownership proof: the page references the vendor's own site.
    if "https://avahr.com" not in resp.text.lower():
        raise RuntimeError("page carried no AvaHR ownership proof")
    return BeautifulSoup(resp.text, "html.parser")

board = parse_avahr("https://jobs.avahr.com/8677-adelante-healthcare/jobs/")
soup = fetch_board(board["board_url"])
print(soup.select_one("div.company-logo h3").get_text(strip=True))

from urllib.parse import urljoin

def parse_rows(board: dict, soup: BeautifulSoup) -> list[dict]:
    rows = []
    for listing in soup.select("div.listing"):
        anchor = next((a for a in listing.select("div.text a[href]") if a.select_one("h2")), None)
        if anchor is None:
            continue

        url = urljoin(f"https://{PUBLIC_HOST}/", anchor["href"])
        identity = parse_avahr(url)
        if identity["company_id"] != board["company_id"] or not identity["job_id"]:
            continue        # a row that belongs to another company is a rejection

        meta = listing.select_one("div.text h3")
        apply_anchor = listing.select_one("div.apply a[href]")
        rows.append({
            "job_id": identity["job_id"],
            "title": anchor.select_one("h2").get_text(strip=True),
            "meta": meta.get_text(" ", strip=True) if meta else None,
            "listing_url": url,
            "apply_url": urljoin(f"https://{PUBLIC_HOST}/", apply_anchor["href"]) if apply_anchor else url,
        })
    return rows

listings = parse_rows(board, soup)
print(f"{len(listings)} open jobs")

import json
import time

def find_job_posting(soup: BeautifulSoup) -> dict | None:
    for tag in soup.find_all("script", type="application/ld+json"):
        try:
            data = json.loads(tag.string or "")
        except (json.JSONDecodeError, TypeError):
            continue
        for node in (data if isinstance(data, list) else [data]):
            if isinstance(node, dict) and node.get("@type") == "JobPosting":
                return node
    return None

def hydrate(board: dict, row: dict) -> dict | None:
    resp = session.get(row["listing_url"], timeout=30)
    if resp.status_code in (404, 410):
        return None            # AvaHR uses 410 heavily for retired jobs
    resp.raise_for_status()

    page = BeautifulSoup(resp.text, "html.parser")
    if "https://avahr.com" not in resp.text.lower():
        raise RuntimeError("detail page contradicted its first-party identity")

    posting = find_job_posting(page)
    if posting is None:
        raise RuntimeError("detail page carried no JobPosting JSON-LD")

    salary = (posting.get("baseSalary") or {}).get("value") or {}
    apply_anchor = page.select_one("div.apply-btn a[href]")
    return {
        "job_id": row["job_id"],
        "company_id": board["company_id"],
        "title": posting.get("title"),
        "description_html": posting.get("description"),
        "posted_at": posting.get("datePosted"),
        "closes_at": posting.get("validThrough"),
        "employment_type": posting.get("employmentType"),
        "salary_min": salary.get("minValue"),
        "salary_max": salary.get("maxValue"),
        "listing_url": row["listing_url"],
        "apply_url": apply_anchor["href"] if apply_anchor else row["apply_url"],
    }

for row in listings[:3]:
    print(hydrate(board, row)["title"])
    time.sleep(0.1)

def classify(url: str) -> str:
    try:
        resp = session.head(url, timeout=15, allow_redirects=True)
    except requests.Timeout:
        return "unreachable"       # retry later; not evidence of anything

    if resp.status_code == 410:
        return "retired"           # identity stays valid, vacancy is gone
    if resp.status_code == 404:
        return "not_found"
    if resp.status_code < 400:
        return "live"
    return "error"

for url in (board["board_url"],
            "https://jobs.avahr.com/8147-zzeeks-pizza/78012-delivery-driver"):
    print(classify(url), url)