chore: enable rosetta cache for builds (aws#17336)

Avail ourselves of the new build cache feature in cdklabs/cdk-ops#1776.
Adds two new things:

**A persistently cached directory**

The directory `$HOME/.s3buildcache` will be stored and restored in the
S3 bucket, if configured. The build can assume that files it puts in
there will be availble on the next build (and on the corresponding
PR build).

**Cache rosetta tablet**

If there is a file in the persistent cache directory for Rosetta,
pass it to `jsii-rosetta` as an input.

Afterwards, store whatever tablet the build produced back into the
cache directory.

The latter will only impact the persistent cache if done on a build
that is actually configured to store the cache back, which is only
the main pipeline build.


----

*By submitting this pull request, I confirm that my contribution is made under the terms of the Apache-2.0 license*

build.sh

@@ -6,11 +6,10 @@ runtarget="build"
 run_tests="true"
 check_prereqs="true"
 check_compat="true"
-extract_snippets="false"
 while [[ "${1:-}" != "" ]]; do
     case $1 in
         -h|--help)
-            echo "Usage: build.sh [--no-bail] [--force|-f] [--skip-test] [--skip-prereqs] [--skip-compat] [--extract]"
+            echo "Usage: build.sh [--no-bail] [--force|-f] [--skip-test] [--skip-prereqs] [--skip-compat]"
             exit 1
             ;;
         --no-bail)
@@ -28,9 +27,6 @@ while [[ "${1:-}" != "" ]]; do
         --skip-compat)
             check_compat="false"
             ;;
-        --extract)
-            extract_snippets="true"
-            ;;
         *)
             echo "Unrecognized parameter: $1"
             exit 1
@@ -81,9 +77,6 @@ trap "rm -rf $MERKLE_BUILD_CACHE" EXIT
 if [ "$run_tests" == "true" ]; then
     runtarget="$runtarget+test"
 fi
-if [ "$extract_snippets" == "true" ]; then
-    runtarget="$runtarget+extract"
-fi
 
 echo "============================================================================================="
 echo "building..."

buildspec-pr.yaml

@@ -16,8 +16,14 @@ phases:
       # Packing the mono-libraries (monocdk & aws-cdk-lib) can cause
       # memory errors. Increasing this value allows our build to more consistently succeed
       - (command -v sysctl || yum install -y procps-ng) && /sbin/sysctl -w vm.max_map_count=2251954
+  pre_build:
+    commands:
+      - /bin/bash ./scripts/cache-load.sh
   build:
     commands:
-      - /bin/bash ./build.sh --extract
-      - /bin/bash ./scripts/transform.sh --extract
+      - /bin/bash ./build.sh
+      - /bin/bash ./scripts/transform.sh
+      # After compilation, run Rosetta (using the cache if available).
+      # This will print errors, and fail the build if there are compilation errors in any packages marked as 'strict'.
+      - /bin/bash ./scripts/run-rosetta.sh
       - git diff-index --exit-code --ignore-space-at-eol --stat HEAD

buildspec.yaml

@@ -16,6 +16,9 @@ phases:
       # Packing the mono-libraries (monocdk & aws-cdk-lib) can cause
       # memory errors. Increasing this value allows our build to more consistently succeed
       - /sbin/sysctl -w vm.max_map_count=2251954
+  pre_build:
+    commands:
+      - /bin/bash ./scripts/cache-load.sh
   build:
     commands:
       - 'if ${BUMP_CANDIDATE:-false}; then /bin/bash ./scripts/bump-candidate.sh; fi'
@@ -25,6 +28,7 @@ phases:
   post_build:
     commands:
       - "[ -f .BUILD_COMPLETED ] && /bin/bash ./pack.sh"
+      - /bin/bash ./scripts/cache-store.sh
 artifacts:
   files:
     - "**/*"

pack.sh

@@ -39,11 +39,7 @@ function lerna_scopes() {
 # Compile examples with respect to "decdk" directory, as all packages will
 # be symlinked there so they can all be included.
 echo "Extracting code samples" >&2
-node --experimental-worker $(which $ROSETTA) \
-  --compile \
-  --output samples.tabl.json \
-  --directory packages/decdk \
-  $(cat $TMPDIR/jsii.txt)
+scripts/run-rosetta.sh $TMPDIR/jsii.txt
 
 # Jsii packaging (all at once using jsii-pacmak)
 echo "Packaging jsii modules" >&2

scripts/cache-load.sh

@@ -0,0 +1,23 @@
+#!/bin/bash
+#
+# If the environment variable S3_BUILD_CACHE is set, download and extract the
+# tarball it points to into the directory $HOME/.s3buildcache.
+#
+set -eu
+
+cachedir=$HOME/.s3buildcache
+mkdir -p $cachedir
+
+if [[ "${S3_BUILD_CACHE:-}" = "" ]]; then
+    exit 0
+fi
+
+echo "🧳 Build cache enabled: ${S3_BUILD_CACHE}"
+if ! aws s3 ls ${S3_BUILD_CACHE} > /dev/null; then
+    echo "🧳⚠️  Cache not found."
+    exit 0
+fi
+
+if ! (cd $cachedir && aws s3 cp ${S3_BUILD_CACHE} - | tar xzv); then
+    echo "🧳⚠️  Something went wrong fetching the cache. Continuing anyway."
+fi

scripts/cache-store.sh

@@ -0,0 +1,21 @@
+#!/bin/bash
+#
+# If the environment variable S3_BUILD_CACHE is set, compress and upload the
+# contents of $HOME/.s3buildcache to it.
+#
+set -eu
+
+cachedir=$HOME/.s3buildcache
+mkdir -p $cachedir
+
+if [[ "${S3_BUILD_CACHE:-}" = "" ]]; then
+    exit 0
+fi
+
+echo "🧳 Storing build cache at: ${S3_BUILD_CACHE}"
+
+if ! (cd $cachedir && tar czv . | aws s3 cp - ${S3_BUILD_CACHE}); then
+    echo "🧳⚠️  Something went wrong storing the cache."
+fi
+
+echo "🧳 Finished."

scripts/list-packages

@@ -12,7 +12,9 @@ if (process.argv.length < 4) {
   process.exit(1);
 }
 
-child_process.exec('lerna ls --toposort --json', { shell: true }, (error, stdout) => {
+const lerna = path.resolve(__dirname, '..', 'node_modules', '.bin', 'lerna');
+
+child_process.exec(`${lerna} ls --toposort --json`, { shell: true }, (error, stdout) => {
   if (error) {
     console.error('Error: ', error);
     process.exit(-1);

scripts/run-rosetta.sh

@@ -0,0 +1,40 @@
+#!/bin/bash
+#
+# Run jsii-rosetta on all jsii packages, using the S3 build cache if available.
+#
+#       Usage: run-rosetta [PKGSFILE]
+#
+# If you already have a file with a list of all the JSII package directories
+# in it, pass it as the first argument. Otherwise, this script will run
+# 'list-packages' to determine a list itself.
+set -eu
+scriptdir=$(cd $(dirname $0) && pwd)
+
+ROSETTA=${ROSETTA:-npx jsii-rosetta}
+
+if [[ "${1:-}" = "" ]]; then
+    echo "Collecting package list..." >&2
+    TMPDIR=${TMPDIR:-$(dirname $(mktemp -u))}
+    node $scriptdir/list-packages $TMPDIR/jsii.txt $TMPDIR/nonjsii.txt
+    jsii_pkgs_file=$TMPDIR/jsii.txt
+else
+    jsii_pkgs_file=$1
+fi
+
+rosetta_cache_file=$HOME/.s3buildcache/rosetta-cache.tabl.json
+rosetta_cache_opts=""
+if [[ -f $rosetta_cache_file ]]; then
+    rosetta_cache_opts="--cache-from ${rosetta_cache_file}"
+fi
+
+$ROSETTA \
+  --compile \
+  --output samples.tabl.json \
+  $rosetta_cache_opts \
+  --directory packages/decdk \
+  $(cat $jsii_pkgs_file)
+
+if [[ -d $(dirname $rosetta_cache_file) ]]; then
+    # If the cache directory is available, copy the current tablet into it
+    cp samples.tabl.json $rosetta_cache_file
+fi

scripts/transform.sh

@@ -25,12 +25,11 @@ createSymlinks() {
 
 runtarget="build"
 run_tests="true"
-extract_snippets="false"
 skip_build=""
 while [[ "${1:-}" != "" ]]; do
     case $1 in
         -h|--help)
-            echo "Usage: transform.sh [--skip-test/build] [--extract]"
+            echo "Usage: transform.sh [--skip-test/build]"
             exit 1
             ;;
         --skip-test|--skip-tests)
@@ -39,9 +38,6 @@ while [[ "${1:-}" != "" ]]; do
         --skip-build)
             skip_build="true"
             ;;
-        --extract)
-            extract_snippets="true"
-            ;;
         *)
             echo "Unrecognized options: $1"
             exit 1
@@ -52,9 +48,6 @@ done
 if [ "$run_tests" == "true" ]; then
   runtarget="$runtarget+test"
 fi
-if [ "$extract_snippets" == "true" ]; then
-  runtarget="$runtarget+extract"
-fi
 
 export NODE_OPTIONS="--max-old-space-size=4096 --experimental-worker ${NODE_OPTIONS:-}"