Update devcontainer configuration from scratch #10956
Conversation
Co-authored-by: anthony.shew <anthony.shew@vercel.com>
|
Cursor Agent can help with this pull request. Just |
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
| echo "🚀 Setting up Turborepo development environment..." | ||
|
|
||
| # Ensure we're in the workspace directory | ||
| cd /workspaces/turborepo 2>/dev/null || cd /workspace 2>/dev/null || cd "$(pwd)" |
There was a problem hiding this comment.
| cd /workspaces/turborepo 2>/dev/null || cd /workspace 2>/dev/null || cd "$(pwd)" | |
| cd /workspaces/turborepo 2>/dev/null || cd /workspace 2>/dev/null || { echo "Error: Could not find workspace directory"; exit 1; } |
The directory fallback logic cd "$(pwd)" doesn't actually change directories - it just changes to the current directory you're already in.
View Details
Analysis
Ineffective directory fallback in .devcontainer/post-create.sh fails silently
What fails: Line 7's fallback cd "$(pwd)" does nothing since $(pwd) expands to the current directory. If neither /workspaces/turborepo nor /workspace exist, the script continues in the wrong directory, causing pnpm install and cargo build to fail with confusing errors.
How to reproduce:
cd /tmp
# Run the original fallback chain
cd /workspaces/turborepo 2>/dev/null || cd /workspace 2>/dev/null || cd "$(pwd)"
echo "Directory: $(pwd)" # Still in /tmpResult: Script stays in /tmp, then pnpm install fails with "No package.json found" instead of a clear workspace directory error.
Expected: Should fail fast with clear error message per Dev Container specification - postCreateCommand runs in workspaceFolder by default, but defensive cd check should provide meaningful error when directories don't exist.
Fix: Replace cd "$(pwd)" with proper error handling: { echo "Error: Could not find workspace directory"; exit 1; }
Rebuild and update the
.devcontainerconfiguration to reflect current Rust/Node.js dependencies andCONTRIBUTING.mdrequirements.