reaxi / cache-eight

file caching with v8 serialization strategy

cache-eight

cache + v8

• checksum keys

• access data from json files (persistent cache)

• optimize functions execution time for expensive operations

• data is saved using v8's serialize / deserialize

why?

caching expensive pure functions results

same inputs -> same results

use case

from 3s to 1s

@todo

Usage

import Cache from 'cache-eight';

const cache = new Cache();

folder                           file
> OS_temp_dir/.cache8/appName -> cache.json

high level api

const values = [1, 2, 3];
const key = values.toString();

const cached = await cache.get(key, () => values.map(value => value * 2));
// will return the cached if valid or the actual value (and serialize for the future)

custom options:

const customPath = path.resolve(process.cwd());

//custom folder:
const customCache = new Cache(customPath);

//custom file:
customCache.get(k, () => v, { filePath: 'my-file.json' });

// cache duration (time to live):
customCache.get(k, () => v, { ttl: 1000 * 60 }); // number in ms;

granular control api

const values = [1, 2, 3];
const key = values.toString();

async function get() {
    const cached = await cache.deserialize(key);

    // if has cached value, return early (skipping further functions)
    if (cached) return cached;

    //else get it from somewhere
    const result = values.map(value => value * 2);

    //cache the result
    await cache.serialize(key, result);

    return result;
}

await get();

API

get(key, () => val, options?) => val
serialize(key, val, options?) => val
deserialize(key, options?) => cachedVal | false

// key and val are any valid Json objects


options = {
    filPath: string + .json //file.json
    ttl: number //ms
}