Implement GenAI functionality & testing

Author: a-kori

.github/workflows/ci.yml

@@ -26,14 +26,21 @@ jobs:
 
   # TODO Add linting
 
-  # Run Java unit-tests
+  # Run Java unit tests
   unit_tests:
     name: Run Tests
     uses: ./.github/workflows/java_tests.yml
     needs: update_api_spec
 
-  # Run UI tests (unit, integration, and e2e)
+  # Run GenAI module tests
+  genai_tests:
+    name: Run GenAI Tests
+    uses: ./.github/workflows/genai_tests.yml
+    needs: update_api_spec
+    secrets:
+      OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
 
+  # Run UI tests (unit, integration, and e2e)
   ui_tests:
     name: Run UI Tests
     uses: ./.github/workflows/ui_tests.yml

.github/workflows/genai_tests.yml

@@ -0,0 +1,30 @@
+name: GenAI Tests
+
+on:
+  workflow_call:
+    secrets:
+      OPENAI_API_KEY:
+        required: true
+
+jobs:
+  genai-tests:
+    runs-on: ubuntu-latest
+    env:
+      OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v3
+
+      - name: Set up Python
+        uses: actions/setup-python@v4
+        with:
+          python-version: '3.11'
+
+      - name: Install dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install -r server/genai/requirements.txt
+          pip install pytest
+
+      - name: Run GenAI tests
+        run: PYTHONPATH=server/genai pytest -v

deployment/README.md

@@ -4,6 +4,9 @@ The `deployment` directory contains necessary files for the deployment of Meet@M
 To deploy the system manually on your local machine, follow the guides below. Automated deployment workflows in GitHub Actions can be found under `../.github/workflows`.
 
 ## Docker Compose
+Prerequisites:
+- a `.env` file is located in `/docker` directory and contains the `OPENAI_API_KEY` secret for the GenAI service (in AWS deployment, a GitHub secret is used instead)
+
 Use the following commands to start / stop / inspect the project using Docker Compose:
 ```
 cd docker                       # Move to docker directory
@@ -26,6 +29,12 @@ kubectl create secret generic <match/user>-db-secret \
   --from-literal=MYSQL_ROOT_PASSWORD='<your-password>' \
   --namespace devoops
 ```
+- OpenAI API key is saved as secret in the `devoops` namespace:
+```
+kubectl create secret generic genai-secret \
+  --from-literal=OPENAI_API_KEY=sk-... \
+  --namespace devoops
+```
 
 Once the three requirements are fulfilled, you can run the application:
 ```

deployment/compose.aws.yml

@@ -98,6 +98,8 @@ services:
 
   meetatmensa-genai:
     image: ghcr.io/aet-devops25/team-devoops/genai:latest
+    environment:
+      OPENAI_API_KEY: ${OPENAI_API_KEY}
     container_name: meetatmensa-genai
     labels:
       - traefik.enable=true

deployment/docker/compose.yml

@@ -41,6 +41,8 @@ services:
       - "80"
     networks:
       - backend
+    env_file:
+      - .env
 
   meetatmensa-client:
     build: 

deployment/k8s/charts/genai/templates/deployment.yaml

@@ -40,6 +40,12 @@ spec:
           {{- end }}
           image: "{{ .Values.image.repository }}:{{ .Values.image.tag | default .Chart.AppVersion }}"
           imagePullPolicy: {{ .Values.image.pullPolicy }}
+          env:
+          - name: OPENAI_API_KEY
+            valueFrom:
+              secretKeyRef:
+                name: genai-secret
+                key: OPENAI_API_KEY
           ports:
             - name: http
               containerPort: {{ .Values.service.port }}

infrastructure/ansible/playbooks/deploy.yml

@@ -105,4 +105,5 @@
         remove_orphans: yes
       environment:
         DOCKER_BUILDKIT: 1
-        EC2_PUBLIC_IP: "{{ ansible_host }}"
+        EC2_PUBLIC_IP: "{{ ansible_host }}"
+        OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}

server/chat/gradle/wrapper/gradle-wrapper.jar

@@ -41,4 +41,4 @@ spring:
         - id: genai_route
           uri: http://meetatmensa-genai:80
           predicates:
-            - Path=/genai/**
+            - Path=/api/v2/genai/**