Skip to content

Caching ​

Caching generally making your application faster with the cost of some memory/storage, but it is even more important when using Sheethuahua with Google Sheets to avoid an issue with very low rate limiting.

Spreadsheet Cache ​

Sheethuahua currently provide caching in the Spreadsheet level with withCache. Every request to every table will go through the caching layer automatically.

withCache requires a CacheAdapter: an object containing get and set function to interact with the cache.

Example with native JavaScript's Map for a simple in-memory cache:

ts
import { Spreadsheet, withCache } from 'sheethuahua';

const cache = new Map(); // Use Map as a simple in-memory cache
const sheets = withCache(Spreadsheet('google-sheets-id'), cache);

const output = sheets.get('sheet-name');

You can replace the map with ioredis, node-cache, etc. or even your own implementation of CacheAdapter.

Released under the MIT License.