Skip to content

Latest commit

 

History

History
131 lines (102 loc) · 4.63 KB

wsl-dev-environment.md

File metadata and controls

131 lines (102 loc) · 4.63 KB
Raw

WSL as a development environment

Windows Subsystem for Linux enables running Linux-like environment directly on top of Windows, without using virtual machines. While this has some shortcomings (no support for GUI applications, no access to drivers), it seems to work pretty well for basic web development. It also has some advantages, e.g. no waiting for the virtual machine to boot.

As of this writing, version 2 of WSL is available through Windows Insider program. It should improve some of the issues mentioned above, e.g. it will include a real Linux Kernel. However, this note is written with version 1 in mind.

The installation process is quite straight-forward.

You might need to tackle an issue with copy-pasting, which in turn requires tweaking permission settings. On the other hand, this might be required only if you use Windows Terminal - I ended up using ConEmu instead.

If you're having problem with root owning all the files and chown doing jack all, this answer will likely help.

ConEmu

While Microsoft recommends using WSL via their new Windows Terminal, it has some issues (e.g. displaying colors, using mouse). With ConEmu stuff pretty much works out-of-the-box. You might want to set home directory as the starting point or set WSL as the default task for new console and pick a nice color theme in the settings.

Note: a Windows feature update seemed to break things, so as suggested here, using the following as the {Bash::bash} task command instead of the one in above link might solve ConEmu startup issues:

set "PATH=%ConEmuBaseDirShort%\wsl;%PATH%" & wsl.exe ~ -cur_console:pm:/mnt

Tmux

Due to permission problems, screen requires an invocation of a spell after each bootup. Tmux doesn't have this problem.

I found that the following config worked nicely for me (place it in ~/.tmux.conf).

# Remap prefix from 'C-b' to 'C-a'.
unbind C-b
set-option -g prefix C-a
bind-key C-a send-prefix

# Reload config file on 'C-a r'.
bind r source-file ~/.tmux.conf

# Enable mouse mode (tmux 2.1 and above)
set -g mouse on

# Toggle mouse mode on `C-a m` to allow copy-pasting with mouse.
bind m run "\
    tmux show-options -g | grep -q "mouse\\s*on"; \
    if [ \$? = 0 ]; \
    then  \
        toggle=off;  \
    else  \
        toggle=on;  \
    fi;  \
    tmux display-message \"mouse is now: \$toggle\";  \
    tmux set-option -w mouse \$toggle; \
    tmux set-option -g mouse \$toggle; \
    "

# Switch panes using Alt-arrow without prefix.
bind -n M-Left select-pane -L
bind -n M-Right select-pane -R
bind -n M-Up select-pane -U
bind -n M-Down select-pane -D

# Split panes using - and _
bind - split-window -v
bind _ split-window -h
unbind '"'
unbind %

# Don't rename windows automatically.
set-option -g allow-rename off

# Highlight current window.
set-window-option -g window-status-current-bg yellow

# Show activity on other windows.
setw -g monitor-activity on
set -g visual-activity on

Whereas screen uses config files to open multiple windows and execute commands on startup, tmux can't use the same approach, since the initial window is opened only after the whole config has been read. This can be remedied by using shell scripts, that might go a little something like this:

#!/usr/bin/env bash

# Create a new session named CRM.
tmux new-session -d -s CRM

# Create named windows and split to panes if needed.
tmux send-keys -t CRM 'tmux new-window -n Django ; tmux split-window -v' ENTER
tmux send-keys -t CRM 'tmux new-window -n Webpack ' ENTER
tmux send-keys -t CRM 'tmux new-window -n Git ' ENTER
tmux send-keys -t CRM 'tmux new-window -n Misc ' ENTER

# Send commands to other windows/panels from window 0.
tmux send-keys -t CRM "tmux send-keys -t Django.0 'cdp cas && sv && srunser' ENTER" ENTER
tmux send-keys -t CRM "tmux send-keys -t Django.1 'cdp crm && sv && srunser' ENTER" ENTER
tmux send-keys -t CRM "tmux send-keys -t Webpack 'cdp crm && cd app && npm run build:hot' ENTER" ENTER
tmux send-keys -t CRM "tmux send-keys -t Git 'cdp crm && git status' ENTER" ENTER

# Activate desired windows and close window 0.
tmux send-keys -t CRM "tmux select-window -t Git" ENTER
tmux send-keys -t CRM "tmux kill-window -t 0" ENTER

# Attach to session
tmux attach -t CRM