Skip to content

davidbrochart/pycrdt-websocket

 
 

Repository files navigation

Pycrdt-websocket

Pycrdt-websocket is an async WebSocket connector for pycrdt.

Build Status Code style: black


Documentation: https://jupyter-server.github.io/pycrdt-websocket

Source Code: https://github.com/jupyter-server/pycrdt-websocket


Pycrdt-websocket is a Python library for building WebSocket servers and clients that connect and synchronize shared documents. It can be used to create collaborative web applications.

The following diagram illustrates a typical architecture. The goal is to share a document among several clients.

Each client has an instance of a Doc, representing their view of a document. A shared document also lives in a room on the server side. Conceptually, a room can be seen as the place where clients collaborate on a document. The WebSocket to which a client connects points to the corresponding room through the endpoint path. In the example below, clients A and B connect to a WebSocket at path room-1, and thus both clients find themselves in a room called room-1. All the Doc synchronization logic is taken care of by the WebsocketProvider.

Each update to a shared document can be persisted to disk using a store, which can be a file or a database.

flowchart TD
    classDef room1 fill:#f96
    classDef room2 fill:#bbf
    A[Client A<br>room-1]:::room1 <-->|WebSocket<br>Provider| server(WebSocket Server)
    B[Client B<br>room-1]:::room1 <-->|WebSocket<br>Provider| server
    C[Client C<br>room-2]:::room2 <-->|WebSocket<br>Provider| server
    D[Client D<br>room-2]:::room2 <-->|WebSocket<br>Provider| server
    server <--> room1((room-1<br>clients: A, B)):::room1
    server <--> room2((room-2<br>clients: C, D)):::room2
    A <-..-> room1
    B <-..-> room1
    C <-..-> room2
    D <-..-> room2
    room1 ---> store1[(Store)]
    room2 ---> store2[(Store)]
Loading

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 97.7%
  • JavaScript 2.3%