Harri Jobs API.

from urllib.parse import urlparse

# Reserved harri.com paths that are marketing pages, not tenant boards.
RESERVED = {
    "api", "platform", "partners", "pre-hire", "post-hire", "careers",
    "scheduling", "engagement", "grocery", "care-sector", "labor-compliance",
    "insights-analytics", "employee-records", "request-a-demo",
}

def tenant_from_url(board_url: str) -> str:
    # e.g. https://harri.com/Generator-Amsterdam---Kitchen/jobs -> "Generator-Amsterdam---Kitchen"
    path = urlparse(board_url).path.strip("/")
    tenant = path.split("/")[0] if path else ""
    if not tenant or tenant.lower() in {"harri", "www", "harri.com"} or tenant.lower() in RESERVED:
        raise ValueError(f"'{tenant}' is a Harri marketing path, not a tenant board")
    return tenant  # preserve exact case and separators

print(tenant_from_url("https://harri.com/HawksmoorChicago/jobs"))

import requests

GATEWAY = "https://gateway.harri.com"

def fetch_profile(tenant: str) -> dict:
    url = f"{GATEWAY}/core/api/v1/profile/slug/{tenant}"
    resp = requests.get(url, timeout=15)
    resp.raise_for_status()
    profile = resp.json().get("data") or {}
    if str(profile.get("type", "")).lower() != "brand":
        raise ValueError(f"profile type '{profile.get('type')}' is not a scrapeable brand")
    if profile.get("is_archived"):
        raise ValueError("brand profile is archived")
    return profile

profile = fetch_profile("cow-careers")
brand_id = profile["id"]

HARRI = "https://harri.com"

def fetch_listings(tenant: str, brand_id: int) -> list[dict]:
    url = f"{GATEWAY}/core-reader/api/v1/profile/brand/{brand_id}"
    resp = requests.get(url, timeout=15)
    resp.raise_for_status()
    brand = resp.json().get("data") or {}

    jobs = []
    for envelope in brand.get("Jobs") or []:
        job = envelope.get("Job") or {}
        job_id = job.get("id")
        if not job_id:
            continue
        position = (job.get("Position") or [{}])[0]
        code = (position.get("code") or "").strip()
        jobs.append({
            "external_id": str(job_id),
            "title": job.get("alias_position") or position.get("name"),
            "company": (brand.get("name") or "").strip(),
            "listing_url": f"{HARRI}/{tenant}/jobs",
            "detail_url": f"{HARRI}/{tenant}/job/{job_id}-{code}",
            "posted_at": job.get("publish_date") or job.get("created"),
        })
    return jobs

import re

def strip_html(html: str) -> str:
    return re.sub(r"<[^>]+>", " ", html or "").strip()

def fetch_job_detail(job_id: str) -> dict:
    url = f"{GATEWAY}/core-reader/api/v1/profile/job/{job_id}"
    resp = requests.get(url, timeout=15)
    resp.raise_for_status()
    data = resp.json().get("data") or {}
    job = data.get("job") or {}

    status = str(job.get("status") or "").upper()
    if job.get("deleted") or status in {"ARCHIVED", "CLOSED"}:
        return {"external_id": job_id, "removed": True, "status": status}

    locations = []
    for loc in job.get("JobLocation") or []:
        inner = loc.get("Location") or {}
        state = loc.get("State") or inner.get("State") or {}
        country = loc.get("Country") or inner.get("Country") or {}
        locations.append({
            "city": (loc.get("City") or inner.get("City") or {}).get("name"),
            "state": state.get("code") or state.get("name"),
            "country": country.get("code") or country.get("name"),
            "address": inner.get("formatted_address"),
        })

    return {
        "external_id": job_id,
        "title": job.get("alias_position") or job.get("title"),
        "description": strip_html(job.get("description")),
        "company": (data.get("brand") or {}).get("name"),
        "locations": locations,
        "removed": False,
    }

import time

def scrape_brand(board_url: str, max_details: int | None = None) -> list[dict]:
    tenant = tenant_from_url(board_url)
    profile = fetch_profile(tenant)
    listings = fetch_listings(tenant, profile["id"])

    results = []
    for job in listings[: max_details or len(listings)]:
        detail = fetch_job_detail(job["external_id"])
        if detail.get("removed"):
            continue  # ARCHIVED / CLOSED / deleted postings are gone
        results.append({**job, **detail})
        time.sleep(2)  # 1 request at a time, ~2s apart
    return results

for job in scrape_brand("https://harri.com/cow-careers/jobs", max_details=3):
    print(job["title"], "-", job["detail_url"])