Eploy Jobs API.

import requests
from urllib.parse import urlsplit

DATAFEED_PATH = "/feeds/datafeed.ashx?Format=xml&ExtApp=true&IntApp=false"

def feed_url(host_or_url: str) -> str:
    # Accepts "alzheimerssocietyweb.eploy.net" or a full careers-site URL.
    parts = urlsplit(host_or_url if "//" in host_or_url else f"https://{host_or_url}")
    return f"https://{parts.netloc}{DATAFEED_PATH}"

print(feed_url("alzheimerssocietyweb.eploy.net"))
# https://alzheimerssocietyweb.eploy.net/feeds/datafeed.ashx?Format=xml&ExtApp=true&IntApp=false

resp = requests.get(
    feed_url("alzheimerssocietyweb.eploy.net"),
    headers={"Accept": "application/xml,text/xml"},
    timeout=30,
)

if resp.status_code == 404:
    raise SystemExit("No public Eploy datafeed for this tenant")
if resp.status_code == 401:
    raise SystemExit("Datafeed requires authentication")
if resp.status_code in (403, 429):
    raise SystemExit("Blocked or rate limited — back off and retry")

resp.raise_for_status()
xml = resp.text

import xml.etree.ElementTree as ET

root = ET.fromstring(xml)
if root.tag.rsplit("}", 1)[-1] != "Vacancies":
    raise ValueError("Unexpected feed root — not an Eploy datafeed")

advertised_total = int(root.attrib.get("Count", "0"))

def text(item, name):
    el = item.find(name)
    return el.text.strip() if el is not None and el.text and el.text.strip() else None

jobs = []
for item in root.findall("Item"):
    vacancy_id = text(item, "VacancyID")
    link = text(item, "Link")
    title = text(item, "Title")
    if not (vacancy_id and vacancy_id.isdigit()
            and link and link.startswith("http") and title):
        continue  # skip placeholder / malformed rows
    jobs.append({
        "external_id": vacancy_id,
        "title": title,
        "location": text(item, "Location"),
        "salary": text(item, "DisplaySalary"),
        "employment_type": text(item, "VacancyType"),
        "industry": text(item, "Industry"),
        "reference": text(item, "Reference"),
        "posted_at": text(item, "DatePosted") or text(item, "DateCreated"),
        "description_html": text(item, "Description"),
        "link": link,
    })

print(f"Advertised {advertised_total}, parsed {len(jobs)}")

for job in jobs:
    parts = urlsplit(job["link"])
    origin = f"{parts.scheme}://{parts.netloc}"
    job["listing_url"] = f"{origin}{parts.path}"  # drop query + fragment
    job["apply_url"] = f"{origin}/registration.aspx?vacancyID={job['external_id']}"

if advertised_total != len(jobs):
    print(f"Count mismatch: feed advertised {advertised_total}, parsed {len(jobs)}")