src/collectors/greenhouse.py

from __future__ import annotations

from typing import List

import requests

from src.collectors.common import dedupe_jobs, normalize_job_posting
from src.detectors.ats_detector import extract_ats_identifier
from src.models import CompanyRecord, JobPosting
from src.resolver.jobs_page_resolver import ResolvedJobsPage

HEADERS = {"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36"}


def collect(company_record: CompanyRecord, resolved_page: ResolvedJobsPage) -> List[JobPosting]:
    """Collect Greenhouse jobs via the public boards API."""
    identifier = company_record.ats_identifier or extract_ats_identifier("greenhouse", resolved_page.url, resolved_page.html)
    if not identifier:
        return []

    api_url = f"https://boards-api.greenhouse.io/v1/boards/{identifier}/jobs?content=true"
    try:
        response = requests.get(api_url, headers=HEADERS, timeout=20)
        response.raise_for_status()
        payload = response.json()
    except Exception:
        return []

    jobs = [
        normalize_job_posting(
            company_record,
            title=item.get("title", ""),
            location=((item.get("location") or {}).get("name") or ""),
            job_url=item.get("absolute_url", resolved_page.url),
            description=(item.get("content") or ""),
            department=(item.get("department") or ""),
            source_ats="greenhouse",
            resolved_url=resolved_page.url,
            posted_date=str(item.get("updated_at") or ""),
            raw_payload=item,
        )
        for item in payload.get("jobs", [])
        if item.get("title")
    ]
    return dedupe_jobs(jobs)