Analysis · JavaScript

Read running jobs with bounded polling

Handle asynchronous job states without unbounded requests or automatic repeated spending.

Before you begin

A jobId returned by an instrument or chart request.
The same server credential context that created the job.

Implementation

1.Poll status using a bounded interval and a finite attempt count.
2.Stop on succeeded or failed.
3.Allow the operator to resume status lookup after timeout, not start a new paid analysis.
4.Avoid logging result bodies or credentials.

poll-job.js

export async function waitForJob(brokerApiKey, jobId, signal) {
  for (let attempt = 0; attempt < 12; attempt++) {
    if (signal?.aborted) throw new Error('Polling cancelled');
    const response = await fetch(
      'https://tradeeyes.app/api/public/business/v1/status?id=' + encodeURIComponent(jobId),
      { headers: { Authorization: 'Bearer ' + brokerApiKey }, signal }
    );
    if (!response.ok) throw new Error('Status lookup failed: ' + response.status);
    const job = await response.json();
    if (job.status === 'succeeded') return job;
    if (job.status === 'failed') throw new Error('Job failed; review before retrying');
    await new Promise(resolve => setTimeout(resolve, Math.min(1000 * (attempt + 1), 5000)));
  }
  throw new Error('Polling timed out; resume lookup with the existing job ID');
}

Validation & limitations

The sample polling schedule is a client strategy, not an API latency guarantee.
Widget tokens can only read jobs within their own session.
Network retry and a new paid analysis are separate actions.

API documentation Organization access