Simple, zero-dependency, object pseudo-database for Cloudflare Workers using R2 buckets, strongly inspired by lowdb 🤗(https://github.com/typicode/lowdb/).

[github] [npm]

Sponsors

Become a sponsor and have your company logo here 👉 GitHub Sponsors

Important Notice

While Cloudflare R2 operates on a strongly consistent model (reference), it's important to note that lowstorage is primarily designed for small, hobby, or personal projects. We advise extreme caution when using lowstorage for critical applications or production environments, as it may not offer the robustness or features required for such use cases.

Usage

import lowstorage from 'lowstorage';
// Initialize object and get users collection
const usersCol = await lowstorage(env, 'MY_TESTING_BUCKET').collection('users');

// Add new user
// you can provide _id or it will be generated as crypto.randomUUID();  -> https://developers.cloudflare.com/workers/runtime-apis/web-crypto/
const newUser = await usersCol.insert({
	name: 'Kevin',
	gender: 'whatever',
	posts: [],
});

// Show all users
const allUsers = usersCol.find({});

// Find users with pagination (e.g., page 2, 10 users per page)
const secondPageUsers = await usersCol.find({}, { skip: 10, limit: 10 });

// Find user by ID and update name
await usersCol.update({ _id: id }, { name: 'Carlos' });

Features

• Lightweight
• Minimalist
• Familiar API
• Plain JavaScript
• Zero-dependency

Install

npm install lowstorage

Why Cloudflare R2?

Seamless migration, robust free tier, Nonee gress fees. Dive into the future of data storage with Cloudflare R2 https://developers.cloudflare.com/r2/

Included features in Forever Free R2 tier

• Storage: 10 GB/month
• Class A operations (mutate state): 1,000,000 / month
• Class B operations (read state): 10,000,000 / month
• more details on pricing R2

Setup & binding R2 to your worker

1. In the Cloudflare console, go to R2 (left navigation)
2. Click Create Bucket
3. Enter any bucket name you want (we use testing-lowstorage)
4. Click Create Bucket (bottom)
5. Go to 'Worker & Pages'
6. Click on your worker (or create new one)
7. Go to 'Settings' -> Variables
8. In section 'R2 Bucket Bindings' click EDIT VARIABLES
9. Hit '+ Add Binding' and pick variable name (we use 'MY_TESTING_BUCKET') and select your R2 bucket
10. Click 'Save & Deploy'

Check out wrangler.toml from examples

Insctructions with pictures https://github.com/gfodor/p2pcf/blob/master/INSTALL.md#set-up-the-r2-bucket

API

collection(colName)

• Input: A string representing the name of the collection.

• Behavior: Creates or accesses a collection with the given name.

• Returns: An instance of the Collection class corresponding to the specified collection name.

• insert(doc)

• insert(doc)

  • Input: A single object or an array of objects to insert into the collection.
  • Behavior: Inserts the given document(s) into the collection. If an _id is not provided, a unique identifier is automatically generated using crypto.randomUUID().
  • Returns: The inserted document(s), with an _id property assigned to each if not already present.

• find(query, options)

  • Input:
    • query: A query object to match documents.
    • options: An optional object for pagination, containing skip and limit properties.
  • Behavior: Searches for documents matching the query. Supports pagination through options.
  • Returns: A promise that resolves to an array of matching documents, considering any pagination specified.

• findOne(query)

  • Input: A query object.
  • Behavior: Similar to find, but it returns only the first matching document.
  • Returns: A promise that resolves to a single document or null if no match is found.

• update(query, update)

  • Input: A query object and an update object.
  • Behavior: Updates all documents that match the query with the provided update data.
  • Returns: A promise that resolves to the number of documents updated.

• updateOne(query, update)

  • Input: A query object and an update object.
  • Behavior: Updates the first document that matches the query with the provided update data.
  • Returns: A promise that resolves to 1 if a document is updated, otherwise 0.

• delete(query)

  • Input: A query object to match documents for deletion.
  • Behavior: Deletes documents matching the query.
  • Returns: A promise that resolves to the number of documents deleted.

• remove()

  • Behavior: Removes all documents from the collection.
  • Returns: A promise that resolves to the number of documents removed.

• count(query)

  • Input: A query object (optional).
  • Behavior: Counts the number of documents that match the query. If no query is provided, it counts all documents in the collection.
  • Returns: A promise that resolves to the count of matching documents.

listCollections()

• Behavior: Lists all collections stored in the Cloudflare R2 bucket associated with the lowstorage instance. (All files ending with .json)
• Returns: A promise that resolves to an array of collection names.

Examples

Check out dummy examples Run:

cd examples
npm install
npm run dev

for testing:

npm run test

It starts local wrangler with ENV and toml config from your /examples folder to run tests.

Limitations

• no test coverage (wip) lowstorage is using end to end tests via its examples
• response speed (no benchmarks so far)
• use carefully!

Contribution

Feel free to dive in! Open an issue or submit PRs.

Standard Readme follows the Contributor Covenant Code of Conduct.

License

MIT