Amris Jobs API.

import requests

# The tenant is the first label of the *.amris-wizard-proxy.com host.
tenant = "wetherspoon"
base = f"https://{tenant}.amris-wizard-proxy.com/vacancyList.php"

resp = requests.get(base, headers={"Accept": "text/html"}, timeout=15)
resp.raise_for_status()
html = resp.text

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

JOB_PATH = re.compile(r"/(?:jobs/)?view/(\d+)(?:/|$)", re.IGNORECASE)

def cell(row, *classes):
    for name in classes:
        node = row.select_one(f"td.{name}")
        if node and node.get_text(strip=True):
            return node.get_text(" ", strip=True)
    return None

soup = BeautifulSoup(html, "html.parser")
table = soup.select_one("table.searchResultsMain")

jobs, seen = [], set()
for row in table.select("tr") if table else []:
    anchor = row.select_one("td.searchTable_jobTitle a[href]")
    if not anchor:
        continue
    # Strip the Amris #QUERY_STRING# placeholder before resolving the link.
    href = anchor["href"].replace("?#QUERY_STRING#", "").replace("#QUERY_STRING#", "")
    detail_url = urljoin(base, href.strip()).split("#")[0].rstrip("/")
    m = JOB_PATH.search(detail_url)
    if not m or m.group(1) in seen:
        continue  # reject non-numeric rows and de-dupe by requirement id
    seen.add(m.group(1))
    jobs.append({
        "external_id": m.group(1),
        "title": anchor.get_text(" ", strip=True),
        "listing_url": detail_url,
        "location": cell(row, "searchTable_location"),
        "salary_text": cell(row, "searchTable_salary"),
        "contract_type": cell(row, "searchTable_type", "searchTable_jobType"),
        "closes": cell(row, "searchTable_closeDate"),
    })

print(f"Found {len(jobs)} vacancies")

EMPTY = re.compile(r"\b(?:no current vacancies|no vacancies found|0 vacancies)\b", re.IGNORECASE)

if table is None:
    if EMPTY.search(soup.get_text(" ")):
        print("Board has no current vacancies")
    else:
        raise RuntimeError("searchResultsMain table missing — layout may have changed or been blocked")

# A next control means more rows exist; do not treat this page as the full set.
has_next = soup.select_one("a[rel='next'][href], .pagination a.next[href]") is not None
if has_next:
    print("More pages advertised — snapshot is incomplete")

def fetch_details(job):
    resp = requests.get(job["listing_url"], headers={"Accept": "text/html"}, timeout=15)
    resp.raise_for_status()
    doc = BeautifulSoup(resp.text, "html.parser")

    fields = {}
    for label in doc.select(".viewLabel"):
        key = label.get_text(strip=True).rstrip(":")
        sib = label.find_next_sibling()
        value = sib.get_text(" ", strip=True) if sib else ""
        if key and value:
            fields[key] = value

    ref = fields.get("Job Ref") or fields.get("Reference")
    if ref and ref != job["external_id"]:
        raise ValueError("Detail Job Ref disagrees with the /view/{id} route")

    node = doc.select_one("#data_primaryInformation") or doc.select_one(".viewDataFD")
    site = doc.select_one("meta[property='og:site_name']")
    return {
        **job,
        "title": fields.get("Job Title") or fields.get("Position") or job["title"],
        "description": node.get_text(" ", strip=True) if node else "",
        "department": fields.get("Department") or fields.get("Pub"),
        "hours_per_week": fields.get("Hours per week"),
        "company": site["content"] if site and site.has_attr("content") else None,
    }

detailed = [fetch_details(j) for j in jobs]