From 526f9ff236d07f3468b22a32d48300db7c4111ea Mon Sep 17 00:00:00 2001
From: PEI-YING-CHEN <chen.peiyi@northeastern.edu>
Date: Fri, 26 Sep 2025 14:59:24 -0400
Subject: [PATCH] Add smoke test script and GitHub Actions workflow

---
 .github/workflows/smoke-test.yml | 28 ++++++++++++++++++++++++++++
 tests/smoke.sh                   | 11 +++++++++++
 2 files changed, 39 insertions(+)
 create mode 100644 .github/workflows/smoke-test.yml
 create mode 100644 tests/smoke.sh

diff --git a/.github/workflows/smoke-test.yml b/.github/workflows/smoke-test.yml
new file mode 100644
index 0000000..ea335cf
--- /dev/null
+++ b/.github/workflows/smoke-test.yml
@@ -0,0 +1,28 @@
+name: Smoke Test   # Name of the workflow, shown in GitHub Actions tab
+
+on:
+  pull_request:    # Trigger workflow when a pull request is created/updated
+    branches:
+      - develop    # Run workflow only if PR targets the 'develop' branch
+      - main       # Or if PR targets the 'main' branch
+
+jobs:
+  smoke-test:      
+    runs-on: ubuntu-latest   # Run the job on the latest Ubuntu runner provided by GitHub
+
+    steps:        
+      - name: Checkout repository   # Fetch the repository code
+        uses: actions/checkout@v4   # Use the official GitHub Action to check out code
+
+      - name: Set up Python        
+        uses: actions/setup-python@v5
+        with:
+          python-version: '3.10'    
+
+      - name: Install dependencies  # Install project dependencies
+        run: |
+          pip install -r requirements.txt   # Install packages listed in requirements.txt
+          pip install dvc[s3]              # Install DVC with S3 support (change to dvc[gdrive] or dvc[azure] if needed)
+
+      - name: Run smoke test        # Execute smoke test script
+        run: bash tests/smoke.sh    # Run the smoke.sh 
diff --git a/tests/smoke.sh b/tests/smoke.sh
new file mode 100644
index 0000000..b63bcb0
--- /dev/null
+++ b/tests/smoke.sh
@@ -0,0 +1,11 @@
+set -e  # when error then stop
+
+echo "🚀 Running DVC smoke test..."
+
+# 1. run pipeline
+dvc repro --pull --force || { echo "❌ DVC pipeline failed"; exit 1; }
+
+# 2. validate pipeline (at least one *_docling.json exist)
+ls data/parsed/docling_or_fallback/*_docling.json >/dev/null 2>&1 || { echo "❌ No parsed outputs found"; exit 1; }
+
+echo "✅ Smoke test passed"