# Keeping in sync

Jobs open and close every day. Keep your copy current so your users never apply to a job that has closed.

## How often

Run a full sync at least once a day. Every 1 to 4 hours is better, and well inside the [limits](/developers/docs/limits-and-support). There’s no need to sync more than once an hour: we cache the XML feed for 10 minutes and most jobs change far less often.

## A full sync

1. Walk the feed from the start, following `next_cursor` until it’s `null`.
2. Collect every job you receive.
3. When the walk finishes, replace your stored jobs with the new set. Jobs you no longer received have closed: remove them.

Jobs come in a fixed order, by id, so a walk is stable while jobs are added.

**If a sync fails part way, keep your last full copy.** Don’t delete jobs because a walk stopped early; try again later.

## Checking one job

To check a single job between syncs:

```bash
curl -H "Authorization: Bearer $MUSLIMJOBS_KEY" \
  "https://www.muslimjobs.io/api/partner/v1/jobs/JOB_ID"
```

A live job returns `200`. A job that has closed returns `410 job_ended`: remove it. An id we don’t know returns `404`.

## Logos

`company.logo_url` points to a PNG on MuslimJobs, at most 256 pixels square. Fetch each logo once and keep your own copy; logos rarely change. On your first sync, fetch them gently, a few at a time, rather than all at once. Companies without a logo have `logo_url: null`; show their initials or a placeholder.

## Node.js example

```javascript
const BASE = 'https://www.muslimjobs.io/api/partner/v1/jobs';
const headers = { Authorization: `Bearer ${process.env.MUSLIMJOBS_KEY}` };

async function fetchAllJobs() {
    const jobs = [];
    let cursor = null;
    do {
        const url = new URL(BASE);
        url.searchParams.set('limit', '100');
        if (cursor) url.searchParams.set('cursor', cursor);

        const response = await fetch(url, { headers });
        if (!response.ok) throw new Error(`Feed returned ${response.status}`);
        const page = await response.json();

        jobs.push(...page.data);
        cursor = page.next_cursor;
    } while (cursor);
    return jobs;
}

async function sync(store) {
    const jobs = await fetchAllJobs();
    // Only replace the stored jobs once the whole walk has worked
    await store.replaceAll(jobs);
}
```

## Python example

```python
import os
import requests

BASE = "https://www.muslimjobs.io/api/partner/v1/jobs"
HEADERS = {"Authorization": f"Bearer {os.environ['MUSLIMJOBS_KEY']}"}

def fetch_all_jobs():
    jobs, cursor = [], None
    while True:
        params = {"limit": 100}
        if cursor:
            params["cursor"] = cursor
        response = requests.get(BASE, headers=HEADERS, params=params, timeout=30)
        response.raise_for_status()
        page = response.json()
        jobs.extend(page["data"])
        cursor = page["next_cursor"]
        if not cursor:
            return jobs

def sync(store):
    jobs = fetch_all_jobs()
    # Only replace the stored jobs once the whole walk has worked
    store.replace_all(jobs)
```

## When something goes wrong

- `429 rate_limited`: wait for the number of seconds in the `Retry-After` header, then carry on.
- `401 invalid_key`: the key is wrong or has been revoked. Check your secrets.
- `403 partner_paused`: your access is paused. Keep your stored jobs but stop showing them until we’ve been in touch.
- Any `5xx`: try again later, and keep your last full copy meanwhile.

