Test and Build #9
Workflow file for this run
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: Test and Build | |
| on: | |
| workflow_call: | |
| workflow_dispatch: | |
| permissions: | |
| contents: read | |
| jobs: | |
| test-build: | |
| name: Lint and Test | |
| runs-on: ${{ (vars.CI_PROVIDER == '' || vars.CI_PROVIDER == 'blacksmith') && 'blacksmith-8vcpu-ubuntu-2404' || 'ubuntu-latest' }} | |
| timeout-minutes: 15 | |
| steps: | |
| - name: Checkout code | |
| uses: actions/checkout@df4cb1c069e1874edd31b4311f1884172cec0e10 # v6 | |
| - name: Setup Bun | |
| uses: oven-sh/setup-bun@0c5077e51419868618aeaa5fe8019c62421857d6 # v2 | |
| with: | |
| bun-version: 1.3.13 | |
| - name: Setup Node | |
| uses: actions/setup-node@48b55a011bda9f5d6aeb4c2d9c7362e8dae4041e # v6 | |
| with: | |
| node-version: 24 | |
| # Cache keys are scoped by event name, and fork PRs get their own | |
| # namespace on top: untrusted fork runs must never share a cache with | |
| # push runs (whose caches feed production image builds) or with trusted | |
| # internal-PR runs. | |
| # | |
| # node_modules also keys on the lockfile hash: a sticky disk is a mutable | |
| # volume, and `bun install --frozen-lockfile` adds what the lockfile needs | |
| # without pruning what it dropped, so branches on different lockfiles were | |
| # contaminating each other (a stale @next/swc 16.2.6 outlived the 16.2.11 | |
| # bump). The bun and Turbo caches are content/hash-addressed, so they stay | |
| # shared — that is what keeps a fresh node_modules disk cheap to fill. | |
| - name: Mount Bun cache | |
| uses: ./.github/actions/cache-mount | |
| with: | |
| provider: ${{ vars.CI_PROVIDER }} | |
| key: ${{ github.repository }}-bun-cache-${{ github.event_name }}${{ github.event.pull_request.head.repo.fork && '-fork' || '' }} | |
| path: ~/.bun/install/cache | |
| - name: Mount node_modules | |
| uses: ./.github/actions/cache-mount | |
| with: | |
| provider: ${{ vars.CI_PROVIDER }} | |
| key: ${{ github.repository }}-node-modules-${{ github.event_name }}${{ github.event.pull_request.head.repo.fork && '-fork' || '' }}-${{ hashFiles('bun.lock') }} | |
| path: ./node_modules | |
| - name: Mount Turbo cache | |
| uses: ./.github/actions/cache-mount | |
| with: | |
| provider: ${{ vars.CI_PROVIDER }} | |
| key: ${{ github.repository }}-turbo-cache-${{ github.event_name }}${{ github.event.pull_request.head.repo.fork && '-fork' || '' }} | |
| path: ./.turbo | |
| - name: Install dependencies | |
| run: bun install --frozen-lockfile --ignore-scripts | |
| # Surfaces known CVEs in the dependency tree. Non-blocking until the | |
| # existing advisory backlog is triaged, then flip to a required gate by | |
| # removing continue-on-error. | |
| - name: Security audit | |
| run: bun audit | |
| continue-on-error: true | |
| - name: Validate env flags | |
| run: | | |
| FILE="apps/sim/lib/core/config/env-flags.ts" | |
| ERRORS="" | |
| echo "Checking for hardcoded boolean env flags..." | |
| # Use perl for multiline matching to catch both: | |
| # export const isHosted = true | |
| # export const isHosted = | |
| # true | |
| HARDCODED=$(perl -0777 -ne 'while (/export const (is[A-Za-z]+)\s*=\s*\n?\s*(true|false)\b/g) { print " $1 = $2\n" }' "$FILE") | |
| if [ -n "$HARDCODED" ]; then | |
| ERRORS="${ERRORS}\n❌ Env flags must not be hardcoded to boolean literals!\n\nFound hardcoded flags:\n${HARDCODED}\n\nEnv flags should derive their values from environment variables.\n" | |
| fi | |
| echo "Checking env flag naming conventions..." | |
| # Check that all export const (except functions) start with 'is' | |
| # This finds exports like "export const someFlag" that don't start with "is" or "get" | |
| BAD_NAMES=$(grep -E "^export const [a-z]" "$FILE" | grep -vE "^export const (is|get)" | sed 's/export const \([a-zA-Z]*\).*/ \1/') | |
| if [ -n "$BAD_NAMES" ]; then | |
| ERRORS="${ERRORS}\n❌ Env flags must use 'is' prefix for boolean flags!\n\nFound incorrectly named flags:\n${BAD_NAMES}\n\nExample: 'hostedMode' should be 'isHostedMode'\n" | |
| fi | |
| if [ -n "$ERRORS" ]; then | |
| echo "" | |
| echo -e "$ERRORS" | |
| exit 1 | |
| fi | |
| echo "✅ All env flags are properly configured" | |
| - name: Check block registry invariants | |
| run: | | |
| if [ "${{ github.event_name }}" = "pull_request" ]; then | |
| BASE_REF="origin/${{ github.base_ref }}" | |
| git fetch --depth=1 origin "${{ github.base_ref }}" 2>/dev/null || true | |
| else | |
| BASE_REF="HEAD~1" | |
| fi | |
| bun run apps/sim/scripts/check-block-registry.ts "$BASE_REF" | |
| - name: Lint code | |
| run: bun run lint:check | |
| - name: Enforce monorepo boundaries | |
| run: bun run check:boundaries | |
| - name: API contract boundary audit | |
| run: bun run check:api-validation:strict | |
| - name: Desktop bridge contract audit | |
| run: bun run check:desktop-bridge | |
| # Complements the bridge audit above, which compares against a snapshot | |
| # this same PR is allowed to regenerate. This one derives every fact from | |
| # the source both sides execute, so it has no such blind spot. | |
| - name: Desktop IPC contract audit | |
| run: bun run check:desktop-ipc | |
| - name: Shared utils enforcement audit | |
| run: bun run check:utils | |
| - name: Zustand v5 selector audit | |
| run: bun run check:zustand-v5 | |
| - name: React Query pattern audit | |
| run: bun run check:react-query | |
| - name: Client boundary import audit | |
| run: bun run check:client-boundary | |
| - name: Bare-icon theme-safety audit | |
| run: bun run check:bare-icons | |
| - name: Icon SVG path validity audit | |
| run: bun run check:icon-paths | |
| - name: Verify realtime prune graph | |
| run: bun run check:realtime-prune | |
| - name: Tool registry client-boundary audit | |
| run: bun run check:tool-registry-boundary | |
| - name: Verify generated tool metadata is in sync | |
| run: bun run tool-metadata:check | |
| - name: Verify skill projections are in sync | |
| run: bun run skills:check | |
| - name: Verify agent stream capability docs are in sync | |
| run: bun run agent-stream-docs:check | |
| - name: Migration safety (zero-downtime) audit | |
| run: | | |
| if [ "${{ github.event_name }}" = "pull_request" ]; then | |
| BASE_REF="origin/${{ github.base_ref }}" | |
| git fetch --depth=1 origin "${{ github.base_ref }}" 2>/dev/null || true | |
| else | |
| BASE_REF="HEAD~1" | |
| fi | |
| bun run check:migrations "$BASE_REF" | |
| - name: Type-check realtime server | |
| run: bunx turbo run type-check --filter=@sim/realtime | |
| # cloud-review-tools.test.ts runs the real helper on the runner, which shells | |
| # out to rg. Blacksmith's image ships it, GitHub's doesn't. | |
| - name: Install ripgrep | |
| run: command -v rg || (sudo apt-get update && sudo apt-get install -y ripgrep) | |
| # Named for what it does: `bun run test` is `vitest run`, with no | |
| # `--coverage`. See the Codecov note below. | |
| - name: Run tests | |
| env: | |
| NODE_OPTIONS: '--no-warnings --max-old-space-size=8192' | |
| NEXT_PUBLIC_APP_URL: 'https://www.sim.ai' | |
| DATABASE_URL: 'postgresql://postgres:postgres@localhost:5432/simstudio' | |
| ENCRYPTION_KEY: '7cf672e460e430c1fba707575c2b0e2ad5a99dddf9b7b7e3b5646e630861db1c' # dummy key for CI only | |
| TURBO_CACHE_DIR: .turbo | |
| run: bun run test | |
| - name: Check schema and migrations are in sync | |
| working-directory: packages/db | |
| run: | | |
| bunx drizzle-kit generate --config=./drizzle.config.ts | |
| if [ -n "$(git status --porcelain ./migrations)" ]; then | |
| echo "❌ Schema and migrations are out of sync!" | |
| echo "Run 'cd packages/db && bunx drizzle-kit generate' and commit the new migrations." | |
| git status --porcelain ./migrations | |
| git diff ./migrations | |
| exit 1 | |
| fi | |
| echo "✅ Schema and migrations are in sync" | |
| # DEAD PATH: nothing generates `apps/sim/coverage`. The test step runs | |
| # `vitest run` without `--coverage`, and vitest.config.ts declares no | |
| # coverage provider, so this uploads nothing and still reports success in | |
| # ~1s (`fail_ci_if_error: false` hides it). `@vitest/coverage-v8` IS | |
| # installed, so wiring it up is possible — but coverage instrumentation | |
| # costs test time and nothing gates on the result today. Left in place | |
| # pending a decision to either enable coverage or drop this step; do not | |
| # read its green tick as "coverage was published". | |
| - name: Upload coverage to Codecov | |
| uses: codecov/codecov-action@0fb7174895f61a3b6b78fc075e0cd60383518dac # v5 | |
| with: | |
| directory: ./apps/sim/coverage | |
| fail_ci_if_error: false | |
| verbose: true | |
| # Next.js production build, in parallel with lint + tests. Sticky disks are | |
| # cloned from the last committed snapshot per job and committed last-writer- | |
| # wins, so concurrent mounts are safe. The bun/node_modules disks are shared | |
| # with test-build (the lockfile-hashed key means they only ever share when the | |
| # dependency tree really is identical, so LWW loss is harmless), but the Turbo | |
| # cache gets its own key: with a shared key, only the last committer's new | |
| # entries survive each run, so the test and build Turbo entries would evict | |
| # each other nondeterministically. | |
| # | |
| # Runner is sized for the COLD-cache build, which is what OOM-killed the 8vcpu | |
| # tier (23 kills / 1074 runs at 98% of its 30.4 GB): warm peaks ~12 GB, cold | |
| # peaked 51 GB. NODE_OPTIONS' --max-old-space-size caps only Node's JS heap, | |
| # not the native Turbopack workers that dominate, so it cannot prevent this. | |
| build: | |
| name: Build App | |
| runs-on: ${{ (vars.CI_PROVIDER == '' || vars.CI_PROVIDER == 'blacksmith') && 'blacksmith-16vcpu-ubuntu-2404' || 'linux-x64-8-core' }} | |
| # Build durations crossed 15 minutes as the app grew (10m02 on Jul 29 AM, | |
| # 14m44 after the folders/desktop/library merges, then two straight | |
| # timeouts) — GitHub reports a job timeout as "cancelled". 25 keeps | |
| # headroom without masking a genuine hang. | |
| timeout-minutes: 25 | |
| steps: | |
| - name: Checkout code | |
| uses: actions/checkout@df4cb1c069e1874edd31b4311f1884172cec0e10 # v6 | |
| - name: Setup Bun | |
| uses: oven-sh/setup-bun@0c5077e51419868618aeaa5fe8019c62421857d6 # v2 | |
| with: | |
| bun-version: 1.3.13 | |
| - name: Setup Node | |
| uses: actions/setup-node@48b55a011bda9f5d6aeb4c2d9c7362e8dae4041e # v6 | |
| with: | |
| node-version: 24 | |
| - name: Mount Bun cache | |
| uses: ./.github/actions/cache-mount | |
| with: | |
| provider: ${{ vars.CI_PROVIDER }} | |
| key: ${{ github.repository }}-bun-cache-${{ github.event_name }}${{ github.event.pull_request.head.repo.fork && '-fork' || '' }} | |
| path: ~/.bun/install/cache | |
| - name: Mount node_modules | |
| uses: ./.github/actions/cache-mount | |
| with: | |
| provider: ${{ vars.CI_PROVIDER }} | |
| key: ${{ github.repository }}-node-modules-${{ github.event_name }}${{ github.event.pull_request.head.repo.fork && '-fork' || '' }}-${{ hashFiles('bun.lock') }} | |
| path: ./node_modules | |
| - name: Mount Turbo cache | |
| uses: ./.github/actions/cache-mount | |
| with: | |
| provider: ${{ vars.CI_PROVIDER }} | |
| key: ${{ github.repository }}-turbo-cache-build-${{ github.event_name }}${{ github.event.pull_request.head.repo.fork && '-fork' || '' }} | |
| path: ./.turbo | |
| # No `.next/cache` mount: the Turbopack persistent build cache is off. A | |
| # controlled A/B on one branch (PR #6078) with a byte-identical module graph | |
| # measured compile at 113s with the cache off, 162s cold with it on, and | |
| # 360s warm — the cache made the same build 3.2x slower, and it grew | |
| # 5.1 GB -> 12 GB across two runs of an unchanged tree, so a disk degrades | |
| # the more it is used. Mounting a disk nothing reads would only cost storage. | |
| # Running out of RAM kills the whole VM and surfaces only as "the runner | |
| # has received a shutdown signal" — no mention of memory, ~12 min in. Warn | |
| # with the real numbers so that failure is a one-line diagnosis instead of | |
| # a mystery. Warn, never fail: a warm build peaks ~12 GB and a partial one | |
| # ~28 GB, so a 32 GB runner still completes plenty of builds, and the | |
| # GitHub fallback is the break-glass path — degrading it to a guaranteed | |
| # failure would be worse than the risk this flags. | |
| - name: Check runner memory headroom | |
| run: | | |
| TOTAL_GB=$(awk '/MemTotal/ {printf "%d", $2/1048576}' /proc/meminfo) | |
| echo "Runner memory: ${TOTAL_GB} GB" | |
| if [ "$TOTAL_GB" -lt 40 ]; then | |
| echo "::warning::Runner has ${TOTAL_GB} GB. A cold-cache build peaks ~51 GB, so this run may be OOM-killed (reported only as 'the runner has received a shutdown signal'). Warm/partial builds should still fit." | |
| fi | |
| - name: Install dependencies | |
| run: bun install --frozen-lockfile --ignore-scripts | |
| - name: Build application | |
| env: | |
| NODE_OPTIONS: '--no-warnings --max-old-space-size=8192' | |
| NEXT_PUBLIC_APP_URL: 'https://www.sim.ai' | |
| DATABASE_URL: 'postgresql://postgres:postgres@localhost:5432/simstudio' | |
| STRIPE_SECRET_KEY: 'dummy_key_for_ci_only' | |
| STRIPE_WEBHOOK_SECRET: 'dummy_secret_for_ci_only' | |
| RESEND_API_KEY: 'dummy_key_for_ci_only' | |
| AWS_REGION: 'us-west-2' | |
| ENCRYPTION_KEY: '7cf672e460e430c1fba707575c2b0e2ad5a99dddf9b7b7e3b5646e630861db1c' # dummy key for CI only | |
| TURBO_CACHE_DIR: .turbo | |
| run: bunx turbo run build --filter=sim |