You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
liam beckman c8664c1932
Update with support for toml configuration.
3 months ago
.#code.org Update with binary deserialization. 1 year ago
.gitignore Add color support. 1 year ago
.npmignore Add color support. 1 year ago
LICENSE.md Create LICENSE.md 1 year ago
README.html Add color support. 1 year ago
README.md Update with support for toml configuration. 3 months ago
config-example.js Misc. updates. 1 year ago
package-lock.json Update with support for toml configuration. 3 months ago
package.json Update with support for toml configuration. 3 months ago
process.js Update with support for toml configuration. 3 months ago
process.toml Update with support for toml configuration. 3 months ago
server.js Update with support for toml configuration. 3 months ago

README.md

demonic server

A fake terminal cleans up user input and sends it to the server. The server then parses the input, and if it matches a predefined command, spawns a child process of that command/program. I really liked the ability to try out haskell interactively here and here, so this is something like that.

Try it out here.

<img src="demo.png"/>

Installation and Running

# install dependencies
npm install

# copy example config
cp config-example.js config.js

# edit key, certificate, and passphrase information
nano config.js

# run server (if no port number is provided, 12345 in this example, the server will default to port 8181)
node server.js 12345

# edit server information
nano demo.js

# then you can connect to the server from a client (e.g. client-example.html)

Uninstallation

# remove this directory
rm -rfI demo

Message Protocol

  • Connection is established between client and server.

  • Client sends server the command mode (and language if mode is set to ‘code’).

{
    "mode": "shell",
}
{
    "mode": "code",
    "lang": "ruby",
}
  • Client sends server data on every keypress.

  • Server buffers the incoming data until a ‘newline’ charcter is received.

  • Server finds the program named in the input, and sends the contents of the input buffer to that program.

  • Server sends client the output of the command.

{
    "out": "Wow, I'm in a shell!"
}
  • Client displays output on terminal.

  • Server sends client the exit status of the command.

{
    "exit": 0
}
  • Client displays user prompt on terminal.

Client to Server

mode: Keyword to inform server of expected behavior. Possible options are ‘shell’ and ‘code’. lang: If mode is set to ‘code’, language specifies the respective programming language. content: The commands or code sent by the user to be evaulted by the server.

Server to Client

exit: Exit status of spawned process. control: Keyword to inform client that spawned process will handle output of characters (e.g. vim). Disables writing client-side as well as buffering server-side. out: STDOUT of the spawned process. err: STDERR of the spawned process.