Clate

Development sever for VS Code in Docker

History

• v0.1
  • native vim + youcompleteme
• v0.2
  • neovim + coc.nvim + llvm + ccls
• v0.3
  • vs code server + llvm + ccls

Installation

1. Install Python Docker SDK

$ pip3 install docker

1. You can find your information with id linux command.

$ id [YOUR_ID]

• If you use root account

$ id root
 uid=0(root) gid=0(root) groups=0(root),0(root),1(bin),2(daemon),3(sys),4(adm),6(disk),10(wheel),11(floppy),20(dialout),26(tape),27(video)

2. Fill config.json

• INSTALL_PATH is a directory to keep vscode-sever for each project.
• Set "true" for target framework.
• If you failed to configure ssh environment, please find you ip address and fill HOST_IP out.

3. Copy your id_rsa.pub to clate/docker/ssh_key/
4. Run install.py

Console

• Clate console helps to manage projects.
  • create, list, active, remove project.

[ INF ] clate - dev
     [C]reate new project
     [L]ist projects
     [A]ctivate project
    a[T]tach running project
     [E]dit project configs

    l[I]st running project
   st[O]p running project
     [D]elete proect

    e[X]it
[ ASK ] command: 

• Clate provides useful commands

usage: clate [-h] [-a ACTIVE] [-d DEBUG] [-t ATTACH] [-o STOP] [-l] [-i]

optional arguments:
  -h, --help            show this help message and exit
  -a ACTIVE, --active ACTIVE
                        active project
  -d DEBUG, --debug DEBUG
                        attach to running project as root
  -t ATTACH, --attach ATTACH
                        attach to running project
  -o STOP, --stop STOP  stop project
  -l, --listproject     list all projects
  -i, --listrunningproject
                        list running projects

Connect to clate from VS Code

1. Open VS Code
2. Install extension: Remote - SSH
3. Open Remote-SSH menu in left bar

4. Refresh connections
5. Click target project

VS Code extension and Settings

• Extensions are independent among projects, because they are stored in independant directory in INSTALL_PATH.
• In project directory, .vscode directory will be created, and all settings will be there.

Installed SW

• Ubuntu 18.04
• gcc 8.3.0
• llvm 9.0.0 (clang inclued)
• python 3.6.7
• node 10.16.0
• cmake 3.10.2

Trouble shooting

• If VS Code machine and Docker machine is different
  1. Before install Clate, Copy id_ras.pub for VS Code machine to clate/docker/ssh_key/
  2. Install Clate
  3. Make .ssh/config in VS Code machine and write server info, like

Host PROJECT_NAME_1
    User YOUR_ID
    Hostname DOCKER.MACHINE.HOST.IP
    Port PROJECT_PORT_1

Host PROJECT_NAME_2
    User YOUR_ID
    Hostname DOCKER.MACHINE.HOST.IP
    Port PROJECT_PORT_2

• If you make lots of projects, ports can conflict. In that case, open .ssh/config and clean up legacy or conflicted project.
• If you want to make your own docker image based on Clate,
  1. Make a new directory in framework and write Dockerfile.
  2. Add the directory name in config.json and give true.
  3. Run install.py
  4. You will have docker image, clate:YOUR_FRAMEWORK_DIRECTORY_NAME