ViRecruit / viDesktop Jobs API.

from urllib.parse import urlparse, parse_qs

VENDOR_SUFFIX = ".viglobalcloud.com"
PAGES = {"recdefault.aspx", "redefault.aspx", "reapplicantemail.aspx", "rejobview.aspx"}

def parse_virecruit(url: str) -> dict:
    parsed = urlparse(url)
    path = parsed.path
    lowered = path.lower()
    if "/virecruitselfapply/" not in lowered:
        raise ValueError("not a ViRecruit self-apply route")

    page = lowered.rsplit("/", 1)[-1]
    if page not in PAGES:
        raise ValueError(f"unrecognised ViRecruit page {page}")

    # Some firms run a nested prefix, e.g. /viDesktopEx/viRecruitSelfApply/.
    prefix = path[: lowered.index("/virecruitselfapply/")].strip("/")
    query = {k.lower(): v[0] for k, v in parse_qs(parsed.query).items()}

    return {
        "host": parsed.netloc.lower(),
        "vendor_hosted": parsed.netloc.lower().endswith(VENDOR_SUFFIX),
        "prefix": prefix,
        "page": page,
        "tag": query.get("tag"),          # keep this even after a redirect
        "filter_job_id": query.get("filterjobid"),
        "job_id": query.get("jobid"),
    }

print(parse_virecruit("https://videsktop.littler.com/viRecruitSelfApply/RecDefault.aspx"
                      "?Tag=e742d0d0-59b7-46f6-a5cd-68c8e5b09b72&FilterJobID=432"))

import requests

session = requests.Session()
session.headers["Accept"] = "text/html,application/xhtml+xml"

def open_board(entry_url: str) -> dict:
    entry = parse_virecruit(entry_url)
    resp = session.get(entry_url, timeout=30, allow_redirects=True)
    resp.raise_for_status()

    landed = parse_virecruit(resp.url)
    if landed["host"] != entry["host"]:
        raise RuntimeError("board redirected off-host — fail closed")
    if landed["page"] not in ("recdefault.aspx", "redefault.aspx"):
        raise RuntimeError("entry link did not land on a board page")

    return {
        "seed_tag": entry["tag"],       # durable identity
        "board_tag": landed["tag"],     # request-time tag; may differ
        "host": entry["host"],
        "prefix": entry["prefix"],
        "layout": "modern" if landed["page"] == "recdefault.aspx" else "legacy",
        "html": resp.text,
        "url": resp.url,
    }

board = open_board("https://cozencareers.viglobalcloud.com/viRecruitSelfApply"
                   "/ReApplicantEmail.aspx?Tag=734c1474-fb1c-4d68-aa1f-5535c19f4983")
print(board["layout"], board["seed_tag"], "->", board["board_tag"])

from bs4 import BeautifulSoup

def parse_board(board: dict) -> list[dict]:
    soup = BeautifulSoup(board["html"], "html.parser")

    # Every real board is a WebForms page; no ViewState means no board.
    if soup.select_one("form#Form1 input[name='__VIEWSTATE']") is None:
        raise RuntimeError("response is not a ViRecruit WebForms board")

    counter = soup.select_one("#contentPlaceHolder_spanPositionsCounter")
    if counter is None or not counter.get_text(strip=True).isdigit():
        raise RuntimeError("board omitted its positions counter")
    advertised = int(counter.get_text(strip=True))

    rows = []
    if board["layout"] == "modern":
        for row in soup.select("[data-vi-filter], .position-row"):
            heading = row.select_one("h4")
            if not heading:
                continue
            body = row.select_one("section.description [data-vi-filter='detail'], section.description")
            rows.append({
                "title": heading.get_text(strip=True),
                "description_html": body.decode_contents() if body else None,
            })
    else:
        for anchor in soup.select("a[href*='ReJobView.aspx']"):
            identity = parse_virecruit(anchor["href"] if anchor["href"].startswith("http")
                                       else f"https://{board['host']}{anchor['href']}")
            rows.append({"job_id": identity["job_id"], "title": anchor.get_text(strip=True)})

    if len(rows) != advertised:
        raise RuntimeError(f"parsed {len(rows)} rows but the board advertises {advertised}")
    return rows

listings = parse_board(board)
print(f"{len(listings)} positions")

import time

def job_url(board: dict, job_id: str) -> str:
    prefix = f"/{board['prefix']}" if board["prefix"] else ""
    return (f"https://{board['host']}{prefix}/viRecruitSelfApply/ReJobView.aspx"
            f"?Tag={board['board_tag']}&JobID={job_id}")

def hydrate(board: dict, job_id: str) -> dict | None:
    url = job_url(board, job_id)
    resp = session.get(url, timeout=30)
    if resp.status_code == 404:
        return None            # the only removal evidence ViRecruit gives
    resp.raise_for_status()

    soup = BeautifulSoup(resp.text, "html.parser")
    if soup.select_one("form#Form1 input[name='__VIEWSTATE']") is None:
        raise RuntimeError("detail response is not a ViRecruit page")

    description = soup.select_one("#contentPlaceHolder_labelJobDescription")
    return {
        "job_id": job_id,
        "title": (soup.select_one("#page-title h3") or soup).get_text(strip=True),
        "description_html": description.decode_contents() if description else None,
        "listing_url": url,
        "apply_url": url,
    }

for row in listings[:3]:
    if row.get("job_id"):
        print(bool(hydrate(board, row["job_id"])))
        time.sleep(0.25)

# Firm-owned hosts and their nested prefixes cannot be derived from the host
# alone, so pin the exact triples you have audited.
AUDITED_HOSTS = {
    "videsktop.littler.com": "",
    "portal.velaw.com": "viDesktopEx",
}

def is_supported(url: str) -> bool:
    try:
        route = parse_virecruit(url)
    except ValueError:
        return False
    if route["vendor_hosted"]:
        return True
    return AUDITED_HOSTS.get(route["host"]) == route["prefix"]

def board_state(board: dict) -> str:
    soup = BeautifulSoup(board["html"], "html.parser")
    if soup.select_one("a[href*='ReJobView.aspx']") or soup.select_one("section.description"):
        return "listable"
    if soup.select_one("#contentPlaceHolder_panelPositionsList") is None:
        return "unsupported"     # postback-only board, no addressable job ids
    return "empty"               # proved-empty board, an authoritative snapshot

print(is_supported("https://careers.example.com/viRecruitSelfApply/RecDefault.aspx?Tag=abc"))
print(board_state(board))