typicode/lowdb
typicode/lowdb has 22,580 GitHub stars and 961 forks. GitHot measured -2 in 7 days. Last checked 2026-11-10 21:14 UTC. Last push 2026-03-27.
简单快速的 JSON 数据库
可以做什么
- Plain JavaScript
- Safe atomic writes
- Change storage, file format (JSON, YAML, ...) or add encryption via adapters
- Extend it with lodash, ramda, ... for super powers!
实测星标增长
- 24小时
- 更新延迟
- 7天
- -2
- 15天
- 更新延迟
- 1个月
- 更新延迟
各周期内的星标净变化,周期相互重叠,并非每日新增。缺失的测量保持空白。星标代表关注度,不代表软件质量。
README.md
原文lowdb

Simple to use type-safe local JSON database 🦉
If you know JavaScript, you know how to use lowdb.
Read or create db.json
const db = await JSONFilePreset('db.json', { posts: [] })
Use plain JavaScript to change data
const post = { id: 1, title: 'lowdb is awesome', views: 100 }
// In two steps
db.data.posts.push(post)
await db.write()
// Or in one
await db.update(({ posts }) => posts.push(post))
// db.json
{
"posts": [
{ "id": 1, "title": "lowdb is awesome", "views": 100 }
]
}
In the same spirit, query using native Array functions:
const { posts } = db.data
posts.at(0) // First post
posts.filter((post) => post.title.includes('lowdb')) // Filter by title
posts.find((post) => post.id === 1) // Find by id
posts.toSorted((a, b) => a.views - b.views) // Sort by views
It's that simple. db.data is just a JavaScript object, no magic.
Sponsors
Become a sponsor and have your company logo here 👉 GitHub Sponsors
Features
- Lightweight
- Minimalist
- TypeScript
- Plain JavaScript
- Safe atomic writes
- Hackable:
- Change storage, file format (JSON, YAML, ...) or add encryption via adapters
- Extend it with lodash, ramda, ... for super powers!
- Automatically switches to fast in-memory mode during tests
Install
npm install lowdb
Usage
Lowdb is a pure ESM package. If you're having trouble using it in your project, please read this.
import { JSONFilePreset } from 'lowdb/node'
// Read or create db.json
const defaultData = { posts: [] }
const db = await JSONFilePreset('db.json', defaultData)
// Update db.json
await db.update(({ posts }) => posts.push('hello world'))
// Alternatively you can call db.write() explicitely later
// to write to db.json
db.data.posts.push('hello world')
await db.write()
// db.json
{
"posts": [ "hello world" ]
}
TypeScript
You can use TypeScript to check your data types.
type Data = {
messages: string[]
}
const defaultData: Data = { messages: [] }
const db = await JSONPreset<Data>('db.json', defaultData)
db.data.messages.push('foo') // ✅ Success
db.data.messages.push(1) // ❌ TypeScript error
Lodash
You can extend lowdb with Lodash (or other libraries). To be able to extend it, we're not using JSONPreset here. Instead, we're using lower components.
import { Low } from 'lowdb'
import { JSONFile } from 'lowdb/node'
import lodash from 'lodash'
type Post = {
id: number
title: string
}
type Data = {
posts: Post[]
}
// Extend Low class with a new `chain` field
class LowWithLodash<T> extends Low<T> {
chain: lodash.ExpChain<this['data']> = lodash.chain(this).get('data')
}
const defaultData: Data = {
posts: [],
}
const adapter = new JSONFile<Data>('db.json')
const db = new LowWithLodash(adapter, defaultData)
await db.read()
// Instead of db.data use db.chain to access lodash API
const post = db.chain.get('posts').find({ id: 1 }).value() // Important: value() must be called to execute chain
CLI, Server, Browser and in tests usage
See src/examples/ directory.
API
Presets
Lowdb provides four presets for common cases.
JSONFilePreset(filename, defaultData)JSONFileSyncPreset(filename, defaultData)LocalStoragePreset(name, defaultData)SessionStoragePreset(name, defaultData)
See src/examples/ directory for usage.
Lowdb is extremely flexible, if you need to extend it or modify its behavior, use the classes and adapters below instead of the presets.
Classes
Lowdb has two classes (for asynchronous and synchronous adapters).
new Low(adapter, defaultData)
import { Low } from 'lowdb'
import { JSONFile } from 'lowdb/node'
const db = new Low(new JSONFile('file.json'), {})
await db.read()
await db.write()
new LowSync(adapterSync, defaultData)
import { LowSync } from 'lowdb'
import { JSONFileSync } from 'lowdb/node'
const db = new LowSync(new JSONFileSync('file.json'), {})
db.read()
db.write()
Methods
db.read()
Calls adapter.read() and sets db.data.
Note: JSONFile and JSONFileSync adapters will set db.data to null if file doesn't exist.
db.data // === null
db.read()
db.data // !== null
db.write()
Calls adapter.write(db.data).
db.data = { posts: [] }
db.write() // file.json will be { posts: [] }
db.data = {}
db.write() // file.json will be {}
db.update(fn)
Calls fn() then db.write().
db.update((data) => {
// make changes to data
// ...
})
// files.json will be updated
Properties
db.data
Holds your db content. If you're using the adapters coming with lowdb, it can be any type supported by JSON.stringify.
For example:
db.data = 'string'
db.data = [1, 2, 3]
db.data = { key: 'value' }
Adapters
Lowdb adapters
JSONFile JSONFileSync
Adapters for reading and writing JSON files.
import { JSONFile, JSONFileSync } from 'lowdb/node'
new Low(new JSONFile(filename), {})
new LowSync(new JSONFileSync(filename), {})
Memory MemorySync
In-memory adapters. Useful for speeding up unit tests. See src/examples/ directory.
import { Memory, MemorySync } from 'lowdb'
new Low(new Memory(), {})
new LowSync(new MemorySync(), {})
LocalStorage SessionStorage
Synchronous adapter for window.localStorage and window.sessionStorage.
import { LocalStorage, SessionStorage } from 'lowdb/browser'
new LowSync(new LocalStorage(name), {})
new LowSync(new SessionStorage(name), {})
Utility adapters
TextFile TextFileSync
Adapters for reading and writing text. Useful for creating custom adapters.
DataFile DataFileSync
Adapters for easily supporting other data formats or adding behaviors (encrypt, compress...).
import { DataFile } from 'lowdb/node'
new DataFile(filename, {
parse: YAML.parse,
stringify: YAML.stringify
})
new DataFile(filename, {
parse: (data) => { decypt(JSON.parse(data)) },
stringify: (str) => { encrypt(JSON.stringify(str)) }
})
Third-party adapters
If you've published an adapter for lowdb, feel free to create a PR to add it here.
Writing your own adapter
You may want to create an adapter to write db.data to YAML, XML, encrypt data, a remote storage, ...
An adapter is a simple class that just needs to expose two methods:
class AsyncAdapter {
read() {
/* ... */
} // should return Promise<data>
write(data) {
/* ... */
} // should return Promise<void>
}
class SyncAdapter {
read() {
/* ... */
} // should return data
write(data) {
/* ... */
} // should return nothing
}
For example, let's say you have some async storage and want to create an adapter for it:
import { Low } from 'lowdb'
import { api } from './AsyncStorage'
class CustomAsyncAdapter {
// Optional: your adapter can take arguments
constructor(args) {
// ...
}
async read() {
const data = await api.read()
return data
}
async write(data) {
await api.write(data)
}
}
const adapter = new CustomAsyncAdapter()
const db = new Low(adapter, {})
See src/adapters/ for more examples.
Custom serialization
To create an adapter for another format than JSON, you can use TextFile or TextFileSync.
For example:
import { Adapter, Low } from 'lowdb'
import { TextFile } from 'lowdb/node'
import YAML from 'yaml'
class YAMLFile {
constructor(filename) {
this.adapter = new TextFile(filename)
}
async read() {
const data = await this.adapter.read()
if (data === null) {
return null
} else {
return YAML.parse(data)
}
}
write(obj) {
return this.adapter.write(YAML.stringify(obj))
}
}
const adapter = new YAMLFile('file.yaml')
const db = new Low(adapter, {})
Limits
Lowdb doesn't support Node's cluster module.
If you have large JavaScript objects (~10-100MB) you may hit some performance issues. This is because whenever you call db.write, the whole db.data is serialized using JSON.stringify and written to storage.
Depending on your use case, this can be fine or not. It can be mitigated by doing batch operations and calling db.write only when you need it.
If you plan to scale, it's highly recommended to use databases like PostgreSQL or MongoDB instead.
探索相关项目
Asabeneh/30-Days-Of-Python
30 Days of Python 编程挑战是一份循序渐进的指南,用于在 30 天内学习 Python 编程语言。这项挑战可能需要超过 100 天。请按自己的节奏进行。这些视频或许也有帮助:https://www.youtube.com/channel/UC7PNRuno1rzYPb1xLa4yktw
78,560 星标grafana/grafana
开放且可组合的可观测性与数据可视化平台。可对来自 Prometheus、Loki、Elasticsearch、InfluxDB、Postgres 等众多数据源的指标、日志和追踪进行可视化。
77,830 星标redis/redis
对于正在构建实时数据驱动应用的开发者而言,Redis 是首选、速度最快且功能最丰富的缓存、数据结构服务器和文档与向量查询引擎。
77,395 星标apache/superset
Apache Superset 是一个数据可视化与数据探索平台
75,473 星标