Resilience — retry, chunking, helpers

Automatic retry

By default, the SDK retries 3 times on:

• HTTP 408, 429, 500, 502, 503, 504
• network errors: ECONNRESET, ETIMEDOUT, ECONNREFUSED, ENETUNREACH, ENOTFOUND, EAI_AGAIN

with an exponential backoff (1s → 2s → 4s, jitter ±500 ms, cap 10s).

// Customize
const client = SpmClient.getClient('v1', apiKey, {
  retry: {
    maxRetries: 5,
    backoffBaseMs: 2000,
    backoffCapMs: 30000,
    retryableStatus: [429, 502, 503, 504],
  },
});

// Disable
const client = SpmClient.getClient('v1', apiKey, { retry: false });

When all attempts fail, the envelope returns ok: false with error.retryable and error.attempts:

const res = await SpmOrders.bulkSave(client, orders);
if (!res.ok && res.error.retryable) {
  // retry later / queue it
}

Optional chunking

Pass a large array: the SDK splits it into batches aligned with the API limit and aggregates the results.

const hugeArray = Array(500).fill().map((_, i) => ({ /* customer data */ }));

const res = await SpmCustomers.bulkSave(client, hugeArray, { chunk: true });

console.log(res.data.sent_count);   // aggregated total
console.log(res.data.chunks);       // per-chunk detail (10 × 50)

By default, chunkSize equals the API-aligned value:

Resource	CHUNK_SIZE
SpmCustomDataRecords	20
SpmCustomers, SpmCustomersAddresses, SpmProducts, SpmProductsImages, SpmProductsVariations, SpmOrders, SpmVouchers	50
SpmCustomersGroups, SpmProductsCategories, SpmProductsManufacturers, SpmOrdersCarriers, SpmOrdersStatuses	100

await SpmProducts.bulkSave(client, items, { chunk: true, chunkSize: 25 }); // override
console.log(SpmCustomers.CHUNK_SIZE); // 50

Shared helpers

const { SpmHelpers } = require('@shopimind/sdk-shopimind');

SpmHelpers.chunk(array, 50);
// → [[...50], [...50], ...]

SpmHelpers.mergeResponses([envelope1, envelope2, ...]);
// → aggregated envelope: data.sent_count, data.rejected_count, data.chunks[]

SpmHelpers.extractCounts(envelope);
// → { sent: N, rejected: M, failed: K }

SpmHelpers.isRetryable(error);
// → true if the error/envelope is retryable (429 / 5xx / network)

SpmHelpers.formatError(envelope);
// → { message, code, statusCode, retryable, attempts, details }
//   safe to serialize for structured logs

Robust synchronization pattern

Combine bulkSave({ chunk: true }) (splitting + aggregation) with the built-in retry and SpmHelpers.formatError for actionable logs — you get a sync resilient to load spikes and transient outages without writing a retry loop.