PowerSchool TalentEd Hire Jobs API.

from urllib.parse import urlparse

VALID_TLDS = {"com", "ca"}

def parse_tedk12(url: str) -> dict:
    parsed = urlparse(url)
    labels = parsed.netloc.lower().split(".")
    # Exactly {district}.tedk12.{com|ca} — nothing deeper, nothing shallower.
    if len(labels) != 3 or labels[1] != "tedk12" or labels[2] not in VALID_TLDS:
        raise ValueError("not a TalentEd Hire host")
    if labels[0] in {"www", ""}:
        raise ValueError("the marketing host carries no district")

    # Boards emit a doubled slash in real links: //hire/ViewJob.aspx?JobID=3212
    path = "/" + parsed.path.lstrip("/").lower()
    if not path.startswith("/hire"):
        raise ValueError("only /hire routes carry board identity")

    return {"district": labels[0], "tld": labels[2]}

print(parse_tedk12("https://alleganymd.tedk12.com//hire/ViewJob.aspx?JobID=3212"))
# {'district': 'alleganymd', 'tld': 'com'}

import requests

def feed_url(district: str, tld: str) -> str:
    return f"https://{district}.tedk12.{tld}/hire/JobList.ashx"

def fetch_feed(district: str, tld: str) -> str:
    resp = requests.get(feed_url(district, tld), timeout=30, allow_redirects=True)

    # A decommissioned district still answers on its host but bounces /hire
    # requests to /hire/404.html, which is HTML. Catch it before parsing XML.
    if resp.url.rstrip("/").endswith("/404.html"):
        raise LookupError(f"district '{district}' no longer exists")

    resp.raise_for_status()
    return resp.text

xml = fetch_feed("garfieldco", "com")
print(len(xml), "bytes")

import re
import xml.etree.ElementTree as ET

DECLARATION = re.compile(r"^\s*<\?xml[^>]*\?>", re.IGNORECASE)

def parse_feed(xml: str) -> list[ET.Element]:
    if not xml.strip():
        raise ValueError("TalentEd Hire feed returned an empty body")

    root = ET.fromstring(DECLARATION.sub("", xml, count=1))
    if root.tag != "jobs":
        raise ValueError(f"unexpected feed root element <{root.tag}>")

    # An empty board returns a well-formed <jobs /> document. That is an
    # authoritative empty snapshot, not a failure.
    return list(root.findall("job"))

jobs = parse_feed(xml)
print(f"{len(jobs)} postings")

def text(node, path: str) -> str | None:
    found = node.find(path)
    return found.text.strip() if found is not None and found.text else None

def to_job(node: ET.Element, district: str, tld: str) -> dict:
    location = node.find("location")
    return {
        "title": text(node, "title"),
        "job_code": text(node, "job-code"),
        "category": text(node, "job-category"),
        "description_html": text(node, "description/summary"),
        "full_time": text(node, "full-time"),
        "part_time": text(node, "part-time"),
        "posted_at": text(node, "posted-date"),
        "closes_at": text(node, "close-date"),
        "company": text(node, "company"),
        "contact": text(node, "contact"),
        "location": {
            "name": text(location, "name") if location is not None else None,
            "city": text(location, "city") if location is not None else None,
            "state": text(location, "state") if location is not None else None,
            "postal_code": text(location, "zip") if location is not None else None,
            "country": text(location, "country") if location is not None else None,
        },
        "listing_url": text(node, "detail-url"),
        "district": district,
        "tld": tld,
    }

for node in jobs[:3]:
    job = to_job(node, "garfieldco", "com")
    print(job["title"], "|", job["category"], "|", job["closes_at"])

def validate(node: ET.Element, district: str, tld: str) -> bool:
    for field in ("job-board-url", "detail-url"):
        url = text(node, field)
        if not url:
            continue
        try:
            claimed = parse_tedk12(url)
        except ValueError:
            return False
        if claimed["district"] != district or claimed["tld"] != tld:
            return False
    return True

accepted, rejected = [], 0
for node in jobs:
    if validate(node, "garfieldco", "com"):
        accepted.append(to_job(node, "garfieldco", "com"))
    else:
        rejected += 1

if rejected:
    print(f"snapshot incomplete: {rejected} rows named another district")
print(f"{len(accepted)} of {len(jobs)} rows accepted")