Error Handling

The SDK throws ScrayleError for all API and connection errors.

ScrayleError

import { ScrayleError } from '@scrayle/sdk';

try {
  const result = await client.scrapingCrawl.scrapeUrl('https://example.com');
} catch (error) {
  if (error instanceof ScrayleError) {
    console.log(error.message);    // Human-readable message
    console.log(error.statusCode); // HTTP status code
    console.log(error.response);   // Raw response body
  }
}

Common Error Codes

Status	Meaning	Action
401	Invalid or missing API key	Check your `SCRAYLE_API_KEY`
402	Insufficient credits	Add credits in the dashboard
403	Access denied	Check permissions for the resource
404	Resource not found	Verify the ID or path exists
429	Rate limited	Slow down requests, use retry with backoff
500	Server error	Retry the request
502/503	Service unavailable	Retry after a delay

Retry Pattern

import { ScrayleClient, ScrayleError, retry } from '@scrayle/sdk';

const client = new ScrayleClient();

const result = await retry(
  async () => {
    return await client.scrapingCrawl.scrapeUrl('https://example.com');
  },
  { retries: 3, minTimeout: 2000 }
);

Browser Connection Errors

Browser sessions may disconnect due to TTL expiry or network issues:

import { chromium } from 'playwright-core';

const { browserWSEndpoint } = await client.browser.create({
  sessionTTL: 300,
});

const browser = await chromium.connectOverCDP(browserWSEndpoint);

browser.on('disconnected', () => {
  console.log('Browser session ended');
  // Reconnect or handle gracefully
});

Timeout Handling

const client = new ScrayleClient({
  timeout: 120000, // 2 minutes for all requests
});

// Or handle timeouts per-call
try {
  const result = await client.scrapingCrawl.crawlUrl(
    'https://large-site.com',
    { limit: 1000 }
  );
} catch (error) {
  if (error.message.includes('timeout')) {
    // Use async crawl for large sites
    const { id } = await client.scrapingCrawl.crawl.asyncCrawlUrl(
      'https://large-site.com',
      { limit: 1000 }
    );
  }
}