Personio Jobs API.

import re

def extract_company_info(url: str) -> tuple[str, str] | None:
    """Extract company slug and domain from Personio URL."""
    pattern = r"^https?://([^.]+)\.jobs\.personio\.(com|de)"
    match = re.match(pattern, url, re.IGNORECASE)

    if match:
        company = match.group(1)
        domain = match.group(2)  # "com" or "de"
        return company, domain
    return None

# Example usage
url = "https://mercanis.jobs.personio.de"
result = extract_company_info(url)
print(result)  # ('mercanis', 'de')

import requests

def fetch_personio_jobs(company: str, domain: str = "de") -> list[dict]:
    """Fetch all jobs from Personio search.json API."""
    url = f"https://{company}.jobs.personio.{domain}/search.json"

    response = requests.get(url, timeout=30)
    response.raise_for_status()

    jobs = response.json()
    print(f"Found {len(jobs)} jobs for {company}")
    return jobs

# Example usage
jobs = fetch_personio_jobs("mercanis", "de")
print(f"First job: {jobs[0]['name']}")

def parse_job(job: dict, company: str, domain: str) -> dict:
    """Parse a Personio job object into a structured format."""
    return {
        "id": job.get("id"),
        "title": job.get("name"),
        "department": job.get("department"),
        "office": job.get("office"),
        "offices": job.get("offices", []),
        "employment_type": job.get("employment_type"),
        "schedule": job.get("schedule"),  # e.g., "Full-time"
        "seniority": job.get("seniority"),  # e.g., "Entry level"
        "category": job.get("category"),
        "keywords": job.get("keywords", "").split(",") if job.get("keywords") else [],
        "description_html": job.get("description"),  # Full HTML description
        "subcompany": job.get("subcompany"),
        "url": f"https://{company}.jobs.personio.{domain}/job/{job.get('id')}",
    }

# Parse all jobs
parsed_jobs = [parse_job(job, "mercanis", "de") for job in jobs]
for job in parsed_jobs[:3]:
    print(f"- {job['title']} ({job['department']})")

def fetch_jobs_by_language(company: str, language: str = "en") -> list[dict]:
    """Fetch jobs filtered by language."""
    url = f"https://{company}.jobs.personio.de/search.json"
    params = {"language": language}

    response = requests.get(url, params=params, timeout=30)
    response.raise_for_status()
    return response.json()

# Fetch English jobs only
english_jobs = fetch_jobs_by_language("mercanis", "en")
print(f"Found {len(english_jobs)} English jobs")

# Also available: "de" (German), "fr" (French), "es" (Spanish)

import time
import requests
from requests.exceptions import RequestException

def fetch_jobs_safe(company: str, domain: str = "de", retries: int = 3) -> list[dict]:
    """Fetch jobs with error handling and retries."""
    url = f"https://{company}.jobs.personio.{domain}/search.json"

    for attempt in range(retries):
        try:
            response = requests.get(url, timeout=30)
            response.raise_for_status()
            return response.json()
        except RequestException as e:
            print(f"Attempt {attempt + 1} failed for {company}: {e}")
            if attempt < retries - 1:
                time.sleep(2 ** attempt)  # Exponential backoff

    print(f"Failed to fetch jobs for {company} after {retries} attempts")
    return []

def scrape_multiple_companies(companies: list[str], delay: float = 1.0) -> dict:
    """Scrape jobs from multiple Personio companies."""
    results = {}
    for company in companies:
        results[company] = fetch_jobs_safe(company)
        time.sleep(delay)  # Rate limit: ~60 req/min
    return results