Merge pull request #666 from aws-samples/features/reduce-tarball-size

chore(packaging): optimize package size for Nexus vending

Author: cajias

.npmignore

@@ -1,6 +1,85 @@
+# Development files
+test/
+tests/
+__tests__/
+*.test.js
+*.spec.js
+jest.config.js
+coverage/
+
+# Source maps and TypeScript sources (we only need compiled JS)
+*.map
 *.ts
 !*.d.ts
+# Keep GraphQL generated TypeScript files
+!lib/user-interface/react-app/src/API.ts
+!lib/user-interface/react-app/src/graphql/*.ts
+
+# Documentation (except README.md which is needed)
+docs/
+documentation/
+*.md
+!README.md
+!aws-genai-llm-chatbot/README.md
+
+# Build artifacts and config files
+.github/
+.gitlab/
+.circleci/
+.travis.yml
+.eslintrc*
+.prettierrc*
+tsconfig.json
+tslint.json
+.vscode/
+.idea/
+
+# Git files
+.git/
+.gitignore
+.gitattributes
+
+# Temporary files
+.DS_Store
+.env
+.env.*
+*.log
+npm-debug.log*
+yarn-debug.log*
+yarn-error.log*
+.cache/
+
+# Build directories
+.aws-sam/
+.cdk.staging/
+cdk.out/
+dist/
+build/
+
+# Large directories that can be rebuilt
+node_modules
+lib/user-interface/react-app/node_modules
+aws-genai-llm-chatbot/modules/chatbot/lib/user-interface/react-app/node_modules
+
+# Scripts directory (except essential scripts)
+scripts/
+!scripts/vend-prep.sh
+
+# Examples and samples
+examples/
+samples/
+
+# CI/CD configuration
+buildspec.yml
+.npmrc
+.nvmrc
+
+# Backup files
+*.bak
+*~
+*.swp
+*.swo
 
-# CDK asset staging directory
-.cdk.staging
-cdk.out
+# Unnecessary package files
+package-lock.json
+yarn.lock

aws-genai-llm-chatbot/assets/chatbot-icon.svg

@@ -3,6 +3,7 @@ deploy:
     install:
       commands:
        - npm ci
+       - npx @aws-amplify/cli codegen add --yes
     build:
       commands:
         - env | grep NEXUS | sort | awk -F= '{printf "%-30s = %s\n", $1, $2}'

aws-genai-llm-chatbot/assets/chatbot-thumbnail.svg

@@ -8,16 +8,23 @@
   "engines": {
     "node": ">=18.0.0 <21.0.0"
   },
+  "files": [
+    "aws-genai-llm-chatbot",
+    "bin",
+    "lib",
+    "cdk.json"
+  ],
   "scripts": {
+    "tsc": "tsc",
     "build": "amplify codegen && tsc",
     "cdk": "cdk",
     "deploy": "npx cdk deploy",
     "hotswap": "cdk deploy --hotswap",
     "watch": "cdk watch",
     "test": "jest",
     "pytest": "pytest tests/",
-    "prevend": "./scripts/vend-prep.sh --pre",
-    "postvend": "./scripts/vend-prep.sh --post",
+    "prepack": "./scripts/vend-prep.sh --pre",
+    "postpack": "./scripts/vend-prep.sh --post",
     "test-all": "npm run test && npm run pytest",
     "integtest": "pytest integtests/",
     "test:integration": "python -m pytest tests/integration/ -xvs",

aws-genai-llm-chatbot/modules/chatbot/deployspec.yaml

@@ -0,0 +1,65 @@
+#!/bin/bash
+# tarball-size.sh - Script to create a tarball and analyze its size
+# This implements suggestion #9 from the tarball size reduction plan
+
+set -e  # Exit immediately if a command exits with a non-zero status
+
+# Print header
+echo "===== Tarball Size Analysis ====="
+echo "Creating tarball with npm pack..."
+
+# Create the package without publishing
+npm pack
+
+# Get the name of the created tarball (should be the only .tgz file)
+TARBALL=$(ls -t *.tgz | head -1)
+
+if [ -z "$TARBALL" ]; then
+    echo "Error: No tarball was created"
+    exit 1
+fi
+
+echo -e "\n===== Tarball Information ====="
+echo "Tarball name: $TARBALL"
+
+# Check the size
+echo -e "\n===== Size Information ====="
+du -h "$TARBALL"
+
+# Get total size in bytes for more precise comparison (macOS compatible)
+BYTES=$(stat -f %z "$TARBALL" 2>/dev/null || stat --format=%s "$TARBALL" 2>/dev/null)
+echo "Size in bytes: $BYTES"
+
+# Create a temporary directory for extraction
+TEMP_DIR=$(mktemp -d)
+echo -e "\n===== Extracting to $TEMP_DIR ====="
+
+# Extract the tarball
+tar -xzf "$TARBALL" -C "$TEMP_DIR"
+
+# Show the largest files/directories
+echo -e "\n===== 20 Largest Files/Directories ====="
+find "$TEMP_DIR" -type f -exec du -h {} \; | sort -hr | head -20
+
+# Show directory sizes
+echo -e "\n===== Directory Sizes ====="
+find "$TEMP_DIR" -type d -depth 2 -exec du -sh {} \; | sort -hr | head -10
+
+# Count files by extension
+echo -e "\n===== Files by Extension ====="
+find "$TEMP_DIR" -type f | grep -v "node_modules" | sed 's/.*\.//' | sort | uniq -c | sort -nr | head -10
+
+# Clean up
+echo -e "\n===== Cleaning Up ====="
+rm -rf "$TEMP_DIR"
+echo "Temporary directory removed"
+
+echo -e "\n===== Recommendations ====="
+echo "Consider the following to reduce tarball size:"
+echo "1. Add large files/directories to .npmignore"
+echo "2. Use the 'files' field in package.json to include only necessary files"
+echo "3. Remove development dependencies before packaging"
+echo "4. Compress images and other assets"
+echo "5. Remove source maps and debug files"
+
+echo -e "\nAnalysis complete!"

aws-genai-llm-chatbot/package/package/capability.yaml

@@ -35,6 +35,10 @@ show_usage() {
 
 # Function to prepare for vending (copy files)
 pre_vend() {
+  npx @aws-amplify/cli codegen --yes
+  npx rimraf lib/**/*.js.map
+  npx rimraf ./**/node_modules
+
   # Check if module directory exists
   if [ ! -d "$MODULE_DIR" ]; then
     echo "Module directory does not exist: $MODULE_DIR"