Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.usehasp.com/llms.txt

Use this file to discover all available pages before exploring further.

​
Loading the SDK

<script src="https://sdk.usehasp.com/v1.js"></script>
<script>
  const sdk = new HaspSDK();
</script>
Always use the full CDN URL. Never use a relative path like /sdk/v1.js. The CDN version ensures you always run a compatible release.

​
ES Module Import

import HaspSDK from 'https://sdk.usehasp.com/v1.js';
const sdk = new HaspSDK();

​
Constructor Options

const sdk = new HaspSDK({
  // All options are optional — defaults work for most apps.

  appId: undefined,             // Auto-detected from window.__HASP__ — never set this manually
  baseUrl: '/api/hasp/v1',  // Data API base path — do not change (exposed for testing environments only)

  onUnauthorized: () => {
    // Called on 401. Default: redirect to /login.
    // Override to save draft state before redirecting.
  },

  onMaintenanceStart: (status) => {
    // Called with { status, message?, estimatedSeconds? } when maintenance begins.
  },

  onMaintenanceEnd: () => {
    // Called when maintenance ends (any request succeeds after a 503).
  },

  retry: {
    maxRetries: 3,      // Retries for 429/502 (default: 3)
    baseDelayMs: 500,   // Base delay before first retry (default: 500ms)
  },
});

​
How appId is Resolved

The SDK reads appId from window.__HASP__.appId, which is injected by Hasp when your app is served. You never set this yourself. If appId cannot be resolved, the constructor throws immediately with ErrorCode.MissingAppId.

​
Retry Behavior

The SDK retries automatically for:
  • 429 Rate Limited — respects Retry-After header if present (integer seconds or HTTP-date)
  • 502 Bad Gateway — network infrastructure hiccups
  • Network failures (device offline, DNS failure) — for GET and single-record DELETE only (idempotent)
POST, PATCH, and bulk DELETE are not retried on network failure to avoid double-writes. Single-record DELETE is retried because it is idempotent. 503 Maintenance is never retried — the SDK enters maintenance mode immediately.

​
Cleanup

Call destroy() when your app unmounts to abort all in-flight requests and clear event handlers: 
sdk.destroy();
Pending promises reject with an AbortError. After destroy(), maintenance event handlers are also cleared.