Update install script #182
Reference in New Issue
Block a user
No description provided.
Delete Branch "development_installer"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
This is a WIP to improve the
install_v.shscript.Done:
Comments and questions below.
@@ -4,2 +4,2 @@SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"cd "$SCRIPT_DIR"#==============================================================================# GLOBAL VARIABLESMoved into main function below
@@ -381,3 +300,4 @@# Remove v-analyzer path from rc filesfor RC_FILE in ~/.zshrc ~/.bashrc; doif [ -f "$RC_FILE" ]; thenFunction was not in use
@@ -400,0 +321,4 @@function start_redis_service() {echo "Attempting to start Redis service..."# Check if redis-server is even installedif ! command_exists redis-server; thenWe already branch on OS when installing packages, so the
package_installfunction is largely redundant. The apt command is confusing so is pulled to a function and explained.@@ -556,3 +403,3 @@echo "V built successfully. Creating symlink..."run_sudo ./v symlinkWhy does this script do a system update, anyway? I never met an install script that wants to do this
@@ -568,3 +414,4 @@}Why delete the lock files?
@@ -624,0 +500,4 @@--start-redis)START_REDIS=true;;*)Why does this script create a new user on Arch (but no other Linux)?
I tried to simplify things substantially here. There were some things I didn't understand, so maybe it cannot actually be this simple. Need to confirm whether systemctl/brew services really don't work in github actions runners.
Also I don't understand why we used
systemctl start? This starts the service once but not again on bootAside from condensing into main function, the top level flow is mostly unchanged aside from adding the flag for starting redis.
It works now in an Ubuntu container. We can check about Alpine later.