Skip to content

Latest commit

 

History

History
172 lines (124 loc) · 5.44 KB

README.md

File metadata and controls

172 lines (124 loc) · 5.44 KB
Raw

Syncrow

Real time directory synchronization using sockets. It can be used as a command line tool, and as a library in other programs.

Currently in Beta

Installation

$ npm install -g syncrow

Configuration

You need to configure syncrow in directory that you want to synchronize.

$ syncrow init

This command will run an interactive setup, similar to npm init, it will ask questions and save your answers.

The result will be a .syncrow.json with your setup.

Running

$ syncrow run or just $ syncrow

This command will look for .syncrow.json file in current directory, if the file exists it will start a syncrow process using it as configuration.

Connecting two machines

First install syncrow globally on both machines that you want to connect. Setup one machine to listen for incoming connections: (Your password will be stored as a hash)

user@server $ syncrow init
? Do you want to listen for connection? Yes
? On which local port would you like to listen 2510
? What is your external IP/hostname? 192.168.0.6
? Please enter comma-separated anymatch patterns for files that should be ignored .git,node_modules
? What synchronization strategy for every new connection would you like to choose? No Action - existing files will be ignored, only new changes will be synced
? Please enter password for obtaining connection my_horse_is_amazing
? Would you like to setup advanced options? No

Then configure syncrow on second machine that will connect:

user@laptop $ syncrow init
? Do you want to listen for connection? No
? Connection: remote host 192.168.0.6
? Connection: remote port 2510
? Please enter comma-separated anymatch patterns for files that should be ignored .git,.idea,node_modules
? What synchronization strategy for every new connection would you like to choose? Push - when other party connects all remote files will be overwritten by those local
? Please enter password for obtaining connection my_horse_is_amazing
? Would you like to setup advanced options? No

Once configured - start syncrow on both machines:

user@server $ syncrow run

and

user@laptop $ syncrow run

After a connection is obtained syncrow will sync existing files. This will run both syncrow instances as a foreground processes. It is possible to connect multiple connecting syncrow instances to single syncrow listener

Using as a library

It is possible to use syncrow as a part of node program.

Class: Server

Listens for incoming connections.

new Server(params)

const syncrow = require('syncrow');

const server = new syncrow.Server({path: './path/to_watch', localPort: 2510, externalHost: '192.168.0.2'});

params:

  • path String path to watch
  • localPort Number port to listen on
  • externalHost String external domain/IP
  • [initalToken] String optional token that will be used for authentication
  • [watch] Boolean optional, defaults to true, if set to false server will not watch local files

server.engine

An instance of syncrow.Engine

server.start(callback)

starts the server watching the FS and listening for connections.

server.shutdown()

Completely stops the server.

event: connection

Emitted when remote party connects to the server

Class: Client

Connects to remote server.

new Client(params)

const syncrow = require('syncrow');

const client = new syncrow.Client({path: './path/to_watch', remotePort: 2510, remoteHost: '127.0.0.1'});

params:

  • path String path to watch
  • remotePort Number port for connection
  • remoteHost String host for connection
  • [initalToken] String optional token that will be used for authentication
  • [watch] Boolean optional, defaults to true, if set to false server will not watch local files

client.engine

An instance of syncrow.Engine

client.start(callback)

starts the watching the path and connects to remote server.

client.shutdown()

Disconnects and stops the client.

Class: Engine

Watches local file system and handles messages from remote parties. It should not be created directly.

const server = new syncrow.Server({path: './path/to_watch', localPort: 2510, externalHost: '192.168.0.2'});

server.start((err)=>{
    if(err) return console.error(err);
    
    server.engine.on('newFile', (file)=>console.log(`remote created a new file: ${file}`));
    
    server.engine.on('changedFile', (file)=>console.log(`remote changed file: ${file}`));
    
    server.engine.on('deletedPath', (path)=>console.log(`remote deleted path (file or directory): ${path}`));
});

event: newFile

emitted when file changed by remote has been downloaded. Params:

  • filePath String

event: changedFile

emitted when file changed by remote has been downloaded. Params:

  • filePath String path of the file that changed

event: deletedPath

emitted when path (file or directory) has been deleted locally. Params:

  • filePath String path of the file/directory deleted

event: newDirectory

emitted when directory created by remote has been created locally. Params:

  • dirPath String path of the directory created

event: error

emitted on error. Params:

  • error Error

event: synced

emitted when synchronization with remote has finished

Major TODOS:

  • Add interval synchronization
  • Separate into several repositories
  • Integrate with Atom

Licence

MIT