ApplicantStack Jobs API.

import requests
from urllib.parse import urlparse

def openings_url(url: str) -> str:
    host = (urlparse(url).hostname or "").lower()
    if not host.endswith(".applicantstack.com"):
        raise ValueError("Not an ApplicantStack URL")
    # Lowercase the tenant so external_id-based dedup is stable (e.g. TBHA -> tbha)
    tenant = host.split(".", 1)[0]
    return f"https://{tenant}.applicantstack.com/x/openings"

print(openings_url("https://TBHA.applicantstack.com/x/detail/a20hmwddcjf0"))
# https://tbha.applicantstack.com/x/openings

import re
import requests
from urllib.parse import urljoin, urlparse
from bs4 import BeautifulSoup

# Detail shortcodes are lowercase alnum (~10-14 chars); /x/apply and /x/myaccount are filtered out
DETAIL_RE = re.compile(r"^/x/detail/([a-z0-9]+)$", re.IGNORECASE)

def scrape_listings(openings: str) -> list[dict]:
    html = requests.get(openings, timeout=15).text
    soup = BeautifulSoup(html, "html.parser")

    table = soup.select_one("table#data-table tbody") or soup.select_one("table#data-table")
    if table is None:
        return []  # no table => zero openings, not an error

    jobs = []
    for row in table.select("tr"):
        anchor = row.select_one('a[href*="/x/detail/"]')
        if anchor is None:
            continue
        href = urljoin(openings, anchor.get("href", ""))
        match = DETAIL_RE.match(urlparse(href).path)
        if not match:
            continue
        cells = row.select("td")
        location = cells[1].get_text(strip=True) if len(cells) >= 2 else None
        jobs.append({
            "external_id": match.group(1),
            "title": anchor.get_text(strip=True),
            "location": location,
            "listing_url": href,
        })
    return jobs

jobs = scrape_listings("https://tanks.applicantstack.com/x/openings")
print(f"Found {len(jobs)} openings")

import json
import requests
from bs4 import BeautifulSoup

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 scrape_detail(detail_url: str) -> dict:
    soup = BeautifulSoup(requests.get(detail_url, timeout=15).text, "html.parser")
    jp = find_job_posting(soup) or {}
    org = jp.get("hiringOrganization") or {}
    ident = jp.get("identifier") or {}
    return {
        "title": jp.get("title"),
        "description_html": jp.get("description"),
        "employment_type": jp.get("employmentType"),
        "posted_at": jp.get("datePosted"),
        "closes_at": jp.get("validThrough"),
        "company_name": org.get("name"),
        "ats_identifier": ident.get("value"),
        "locations": jp.get("jobLocation"),  # address: locality/region/country/postalCode
    }

def html_fallback(soup) -> dict:
    out: dict = {}

    title_el = soup.select_one("#contenttitle")
    if title_el:
        out["title"] = title_el.get_text(strip=True)

    # Summary table rows look like <th>Location:</th><td>...</td>
    keys = {"id", "location", "department", "salary", "type", "job type"}
    for row in soup.select("table.formtable tr"):
        th, td = row.find("th"), row.find("td")
        if not th or not td:
            continue
        key = th.get_text(strip=True).rstrip(":").lower()
        value = td.get_text(strip=True)
        if key in keys and value:
            out[key.replace(" ", "_")] = value

    desc = soup.select_one(".listing_description")
    if desc:
        out["description_html"] = desc.decode_contents()

    apply = soup.select_one('a[href*="/x/apply/"]')
    if apply:
        out["apply_url"] = apply.get("href")

    return out

import time
import requests

def fetch(url: str) -> requests.Response:
    resp = requests.get(url, timeout=15)
    if resp.status_code == 401:
        raise RuntimeError("Auth required — this tenant's board is not public")
    if resp.status_code == 403:
        raise RuntimeError("Blocked (403) — back off and retry later")
    if resp.status_code == 404:
        raise RuntimeError("Tenant or job not found / removed (404)")
    if resp.status_code == 429:
        raise RuntimeError("Rate limited (429) — slow down")
    resp.raise_for_status()
    return resp

# The reference scraper caps detail fetches at 3 concurrent with ~400ms between requests
for job in jobs:
    detail = scrape_detail(job["listing_url"])
    time.sleep(0.4)