PostingPanda Jobs API.

from urllib.parse import urlparse

def resolve_generation(careers_url: str) -> str:
    host = (urlparse(careers_url).hostname or "").lower()
    return "talos" if host.endswith(".talosats-careers.com") else "legacy"

# host is the tenant key across both DNS namespaces
host = (urlparse("https://thescoutassociation.postingpanda.uk/").hostname or "").lower()
generation = resolve_generation(f"https://{host}/")

import requests

LEGACY_FILTER_URL = (
    "https://postingpandaapi-live.azurewebsites.net"
    "/api/liveadverts/filter/-in-"
    "?country=United+Kingdom&distanceMiles=10"
    "&extraDataFilters=&location=&searchTerm="
)

def fetch_legacy(host: str) -> list[dict]:
    resp = requests.get(
        LEGACY_FILTER_URL,
        headers={"Referer": f"https://{host}/"},  # the only tenant scope
        timeout=15,
    )
    resp.raise_for_status()
    body = resp.text.lstrip()
    if body.startswith("["):
        return resp.json()               # older deployments emit JSON
    return parse_legacy_xml(resp.text)   # current deployments emit XML

import xml.etree.ElementTree as ET

def _local(tag: str) -> str:
    return tag.rsplit("}", 1)[-1]

def parse_legacy_xml(xml_text: str) -> list[dict]:
    root = ET.fromstring(xml_text)
    if not _local(root.tag).startswith("ArrayOf"):
        return []
    adverts = []
    for item in root:
        if "ResultsQueryFilterLiveAdverts" not in _local(item.tag):
            continue
        fields = {_local(c.tag): (c.text or "").strip() for c in item}
        adverts.append(fields)   # PascalCase keys, same as the JSON path
    return adverts

import requests

TALOS_BASE = "https://api-careers-sites.talos360.com"

def fetch_talos(host: str) -> list[dict]:
    cfg = requests.get(
        f"{TALOS_BASE}/api/careerssites/site/config/get",
        params={"host": host},
        timeout=15,
    ).json()
    obfuscated_id = (cfg.get("siteConfig") or {}).get("obfuscatedId")
    if not obfuscated_id:
        raise ValueError(f"No obfuscatedId for {host}")

    resp = requests.post(
        f"{TALOS_BASE}/api/careerssite/vacancies/search",
        headers={"Origin": f"https://{host}", "Referer": f"https://{host}/"},
        json={
            "careersSiteObfuscatedId": obfuscated_id,
            "metadataFilters": [],
            "preFilters": [],
            "siteType": "External",
        },
        timeout=15,
    )
    resp.raise_for_status()
    return resp.json().get("careersSiteVacancies") or []

import requests

def normalize(host: str, generation: str, rec: dict) -> dict:
    if generation == "talos":
        job_id, title = rec.get("jobPostId"), rec.get("jobTitle")
        description, salary = rec.get("jobDescription"), rec.get("salaryDisplay")
    else:
        job_id, title = rec.get("AdvertId"), rec.get("JobTitle")
        description, salary = rec.get("JobDescription"), rec.get("SalaryDisplay")
    return {
        "external_id": job_id,
        "title": title,
        "description": description,
        "salary": salary,
        "listing_url": f"https://{host}/job/{job_id}",
    }

def legacy_empty_is_authoritative(host: str) -> bool:
    html = requests.get(
        f"https://{host}/", headers={"Accept": "text/html"}, timeout=15
    ).text
    default_runtime = "/Scripts/Views/Home/default/JobSearch-angular.js" in html
    custom_runtime = (
        "postingpanda.blob.core.windows.net/clientpublicassets/" in html
        and "/Scripts/" in html
    )
    return default_runtime or custom_runtime