eRecruit Jobs API.

import re
from urllib.parse import urlparse

RESERVED = {"api", "app", "cdn", "mail", "support", "www"}
DETAIL_ROUTE = re.compile("^/candidateapp/Jobs/View/([A-Z0-9][A-Z0-9-]{2,63})$",
                          re.IGNORECASE)

def parse_erecruit(url: str) -> dict | None:
    parsed = urlparse(url)
    if parsed.scheme != "https":
        return None

    labels = parsed.netloc.lower().rstrip(".").split(".")
    if len(labels) != 3 or labels[1] != "erecruit" or labels[2] != "co":
        return None
    tenant = labels[0]
    if tenant in RESERVED:
        return None

    path = parsed.path.rstrip("/")
    detail = DETAIL_ROUTE.match(path)
    return {
        "tenant": tenant,
        "board_url": f"https://{tenant}.erecruit.co/candidateapp/Jobs/Browse",
        "job_id": detail.group(1).upper() if detail else None,
    }

print(parse_erecruit("https://westerncapegov.erecruit.co/candidateapp/Jobs/View/WCG260611-3"))

import requests
from bs4 import BeautifulSoup
from urllib.parse import urljoin

CATEGORY_COUNT = re.compile("[(]([0-9]+)[)]\\s*$")

def fetch_categories(session, tenant: str) -> list[dict]:
    board_url = f"https://{tenant}.erecruit.co/candidateapp/Jobs/Browse"
    resp = session.get(board_url, headers={"Accept": "text/html"}, timeout=30)
    resp.raise_for_status()
    if "/candidateapp/Jobs/Browse" not in resp.text:
        raise RuntimeError("eRecruit board omitted its CandidateApp markers")

    soup = BeautifulSoup(resp.text, "html.parser")
    categories, seen = [], set()
    for anchor in soup.select("a[href*='/candidateapp/Jobs/Categories/']"):
        url = urljoin(board_url, anchor["href"])
        text = " ".join(anchor.get_text().split())
        match = CATEGORY_COUNT.search(text)
        identity = parse_erecruit(url)
        if not match or not identity or identity["tenant"] != tenant or url in seen:
            continue
        seen.add(url)
        categories.append({
            "name": text[: match.start()].strip(),
            "expected": int(match.group(1)),
            "url": url,
        })
    return categories

session = requests.Session()
categories = fetch_categories(session, "exxaro")
print(sum(c["expected"] for c in categories), "advertised vacancies")

import time
from html import unescape

DETAIL_PATH = re.compile("/candidateapp/Jobs/View/[A-Z0-9][A-Z0-9-]{2,63}",
                         re.IGNORECASE)

def parse_category(session, category: dict, tenant: str) -> list[dict]:
    resp = session.get(category["url"], headers={"Accept": "text/html"}, timeout=30)
    resp.raise_for_status()

    soup = BeautifulSoup(resp.text, "html.parser")
    rows = []
    for row in soup.select("tr.item[onclick]"):
        match = DETAIL_PATH.search(unescape(row.get("onclick") or ""))
        if not match:
            continue
        url = urljoin(category["url"], match.group(0))
        identity = parse_erecruit(url)
        if not identity or not identity["job_id"]:
            continue
        cells = [" ".join(cell.get_text().split()) for cell in row.select("td")]
        rows.append({
            "id": identity["job_id"],
            "title": cells[0] if cells else None,
            "category": category["name"],
            "url": url,
        })

    if len(rows) != category["expected"]:
        raise RuntimeError(
            f"category {category['name']} returned {len(rows)} "
            f"of {category['expected']} advertised jobs")
    return rows

def scrape_board(session, tenant: str) -> list[dict]:
    jobs, seen = [], set()
    for category in fetch_categories(session, tenant):
        if category["expected"] == 0:
            continue
        for row in parse_category(session, category, tenant):
            if row["id"] not in seen:  # a job can sit in several categories
                seen.add(row["id"])
                jobs.append(row)
        time.sleep(0.15)
    return jobs

import json

def find_job_posting(html: str) -> dict | None:
    soup = BeautifulSoup(html, "html.parser")
    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 fetch_detail(session, listing: dict) -> dict | None:
    resp = session.get(listing["url"], headers={"Accept": "text/html"}, timeout=30)
    if resp.status_code in (404, 410):
        return None
    resp.raise_for_status()

    # eRecruit renders a first-party "Not found" page for withdrawn requisitions.
    soup = BeautifulSoup(resp.text, "html.parser")
    if (soup.title and "not found" in soup.title.get_text(strip=True).lower()):
        return None

    posting = find_job_posting(resp.text)
    if not posting:
        return None

    fields = {}
    for row in soup.select("tr.item"):
        label = row.select_one("td.label")
        value = row.select_one("td.value")
        if label and value:
            key = " ".join(label.get_text().split()).rstrip(":").lower()
            fields[key.replace(" ", "_")] = " ".join(value.get_text().split())

    address = ((posting.get("jobLocation") or {}).get("address")) or {}
    return {
        "id": listing["id"],
        "title": posting.get("title") or listing["title"],
        "description_html": posting.get("description"),
        "employment_type": posting.get("employmentType"),
        "posted_at": posting.get("datePosted"),
        "closes_at": posting.get("validThrough"),
        "city": address.get("addressLocality"),
        "region": address.get("addressRegion"),
        "fields": fields,
        "url": listing["url"],
    }