Red Rover K12 Jobs API.

from urllib.parse import urlparse

BOARD_HOST = "jobs.redroverk12.com"

def parse_board(url: str) -> dict:
    parsed = urlparse(url)
    if parsed.netloc.lower() != BOARD_HOST:
        raise ValueError("not a Red Rover board URL")

    parts = [p for p in parsed.path.strip("/").split("/") if p]
    if len(parts) < 2 or parts[0].lower() != "org":
        raise ValueError("expected /org/{organizationId}")

    org = parts[1]
    opening = parts[3] if len(parts) >= 4 and parts[2].lower() == "opening" else None
    return {"org": org, "opening": opening, "is_numeric": org.isdigit()}

print(parse_board("https://jobs.redroverk12.com/org/3877"))
# {'org': '3877', 'opening': None, 'is_numeric': True}

import re
import requests

BRANDING = re.compile(
    r'SimpleJobSeekerSiteBranding\\?".{0,240}?'
    r'orgId\\?"\s*:\s*\\?"(?P<org_id>[1-9][0-9]{0,17})\\?"\s*,\s*'
    r'orgPath\\?"\s*:\s*\\?"(?P<org_path>[A-Za-z0-9][A-Za-z0-9._-]{0,63})\\?"'
)

def resolve_org_id(slug: str) -> str | None:
    resp = requests.get(
        f"https://jobs.redroverk12.com/org/{slug}",
        headers={"Accept": "text/html"},
        timeout=30,
    )
    resp.raise_for_status()

    match = BRANDING.search(resp.text)
    # The board must brand the exact path we asked for; anything else is not proof.
    if not match or match.group("org_path").lower() != slug.lower():
        return None
    return match.group("org_id")

print(resolve_org_id("wauseonschools"))  # "1183"

import requests

GRAPHQL = "https://api.redroverk12.com/graphql"

LISTINGS_QUERY = """
query GetJobPostings($search: JobPostingSearchInput!) {
  jobSeekerSiteUnauthenticated {
    jobPostingSearch(search: $search) {
      results {
        id orgId name statusId organizationName
        category { id name }
        jobPostingTypeId payTypeId minPay maxPay allowsRemote
        location { id name address { address1 city state postalCode country } }
        activePublicOnDateUtc pausedOnDateUtc closedOnDateUtc
      }
      offset limit hasMoreData totalCount
    }
  }
}
"""

def search_jobs(org_id: str) -> dict:
    resp = requests.post(
        GRAPHQL,
        json={
            "operationName": "GetJobPostings",
            "query": LISTINGS_QUERY,
            "variables": {"search": {"orgId": org_id}},
        },
        headers={"rrClient": "JobSeeker", "Content-Type": "application/json"},
        timeout=30,
    )
    resp.raise_for_status()
    payload = resp.json()
    if payload.get("errors"):
        raise RuntimeError(payload["errors"])
    return payload["data"]["jobSeekerSiteUnauthenticated"]["jobPostingSearch"]

search = search_jobs("3877")
print(search["totalCount"], "postings, hasMoreData:", search["hasMoreData"])

PUBLISHED = {"PUBLISHED", "PUBLISHED_INTERNAL"}

def collect(search: dict) -> list[dict]:
    results = search.get("results") or []

    if search.get("hasMoreData"):
        # The 500-row cap was hit. Do not treat this page as the full board.
        raise RuntimeError("Red Rover truncated the snapshot; refuse to expire jobs from it")

    total = search.get("totalCount")
    if total is not None and total != len(results):
        raise RuntimeError(f"count mismatch: totalCount={total}, rows={len(results)}")

    return [job for job in results if (job.get("statusId") or "").upper() in PUBLISHED]

open_jobs = collect(search)
print(f"{len(open_jobs)} currently published openings")
# Other statuses seen in production: CLOSED, PAUSED, ARCHIVED

import time

DETAILS_QUERY = """
query GetJobPosting($jobPostingId: ID!) {
  jobSeekerSiteUnauthenticated {
    jobPostingById(jobPostingId: $jobPostingId) {
      id orgId name statusId organizationName description
      category { id name }
      minPay maxPay payTypeId allowsRemote
      descriptionFileUpload { originalFileUrl uploadedFileName }
      location { name address { address1 city state postalCode country } }
      customFieldValues { id value customField { name customFieldType } }
    }
  }
}
"""

def get_opening(org_id: str, opening_id: str) -> dict | None:
    resp = requests.post(
        GRAPHQL,
        json={
            "operationName": "GetJobPosting",
            "query": DETAILS_QUERY,
            "variables": {"jobPostingId": opening_id},
        },
        headers={"rrClient": "JobSeeker", "Content-Type": "application/json"},
        timeout=30,
    )
    resp.raise_for_status()
    posting = resp.json()["data"]["jobSeekerSiteUnauthenticated"]["jobPostingById"]

    if posting is None:
        return None  # structured null: the opening is gone, the district is not
    if str(posting.get("orgId")) != str(org_id):
        raise RuntimeError("detail returned a different organization — reject the row")

    posting["listing_url"] = f"https://jobs.redroverk12.com/org/{org_id}/opening/{opening_id}"
    posting["apply_url"] = posting["listing_url"] + "/apply"
    return posting

for job in open_jobs[:3]:
    detail = get_opening("3877", job["id"])
    if detail:
        print(detail["name"], "-", detail["location"]["name"])
    time.sleep(0.1)