test: Add HF Space Dockerfile using pre-built leaderboard image

.github/workflows/hf_space_docker.yml

@@ -0,0 +1,139 @@
+name: Test HF Space Dockerfile
+
+on:
+  push:
+    branches: [ main ]
+    paths:
+      - 'Dockerfile.hf-space'
+      - '.github/workflows/hf_space_docker.yml'
+  pull_request:
+    branches: [ main ]
+    paths:
+      - 'Dockerfile.hf-space'
+      - '.github/workflows/hf_space_docker.yml'
+  workflow_dispatch:
+
+jobs:
+  test-hf-space-dockerfile:
+    runs-on: ubuntu-latest
+
+    steps:
+    - name: Checkout code
+      uses: actions/checkout@v4
+
+    - name: Free disk space
+      run: |
+        sudo rm -rf \
+          "$AGENT_TOOLSDIRECTORY" \
+          /opt/ghc \
+          /opt/google/chrome \
+          /opt/microsoft/msedge \
+          /opt/microsoft/powershell \
+          /opt/pipx \
+          /usr/lib/mono \
+          /usr/local/julia* \
+          /usr/local/lib/android \
+          /usr/local/lib/node_modules \
+          /usr/local/share/chromium \
+          /usr/local/share/powershell \
+          /usr/share/dotnet \
+          /usr/share/swift || true
+        docker system prune -af || true
+
+    - name: Set up Docker Buildx
+      uses: docker/setup-buildx-action@v3
+
+    - name: Log in to GitHub Container Registry
+      uses: docker/login-action@v3
+      with:
+        registry: ghcr.io
+        username: ${{ github.actor }}
+        password: ${{ secrets.GITHUB_TOKEN }}
+
+    - name: Build HF Space Docker image
+      uses: docker/build-push-action@v5
+      with:
+        context: .
+        file: ./Dockerfile.hf-space
+        push: false
+        tags: mteb-hf-space:test
+        load: true
+
+    - name: Test container can start and run
+      timeout-minutes: 6
+      run: |
+        # Start the container in background
+        docker run -d --name mteb-hf-test -p 7860:7860 mteb-hf-space:test
+
+        # Monitor container with smart exit conditions
+        echo "Starting HF Space leaderboard container..."
+        START_TIME=$(date +%s)
+        STARTUP_DETECTED=false
+        INIT_COMPLETE_DETECTED=false
+
+        while true; do
+          CURRENT_TIME=$(date +%s)
+          ELAPSED=$((CURRENT_TIME - START_TIME))
+
+          # Maximum timeout: 5.5 minutes
+          if [ $ELAPSED -gt 330 ]; then
+            echo "Timeout reached after ${ELAPSED}s - container failed to complete initialization"
+            docker logs --tail 20 mteb-hf-test
+            docker stop mteb-hf-test
+            docker rm mteb-hf-test
+            exit 1
+          fi
+
+          # Check if container is still running
+          if ! docker ps | grep -q mteb-hf-test; then
+            EXIT_CODE=$(docker inspect mteb-hf-test --format='{{.State.ExitCode}}')
+            echo "Container exited with code: $EXIT_CODE after ${ELAPSED}s"
+
+            LOGS=$(docker logs mteb-hf-test 2>&1)
+            if echo "$LOGS" | grep -q "=== Leaderboard app initialization complete" || \
+               echo "$LOGS" | grep -q "Running on.*http"; then
+              echo "Container completed significant startup progress successfully"
+              docker rm mteb-hf-test
+              exit 0
+            else
+              echo "Container failed before completing startup initialization"
+              echo "Last logs:"
+              echo "$LOGS" | tail -10
+              docker rm mteb-hf-test
+              exit 1
+            fi
+          fi
+
+          LOGS=$(docker logs mteb-hf-test 2>&1)
+
+          if [ "$STARTUP_DETECTED" = "false" ] && echo "$LOGS" | grep -q "Starting leaderboard app in process"; then
+            echo "Detected leaderboard app startup"
+            STARTUP_DETECTED=true
+          fi
+
+          if [ "$INIT_COMPLETE_DETECTED" = "false" ] && echo "$LOGS" | grep -q "=== Leaderboard app initialization complete"; then
+            echo "Detected full app initialization complete!"
+            INIT_COMPLETE_DETECTED=true
+            sleep 3
+            if curl -s --max-time 5 http://localhost:7860/ > /dev/null; then
+              echo "Server is responding - container fully operational!"
+              docker stop mteb-hf-test
+              docker rm mteb-hf-test
+              exit 0
+            fi
+          fi
+
+          if echo "$LOGS" | grep -q "Running on.*http"; then
+            echo "Detected Gradio server ready - container fully operational!"
+            docker stop mteb-hf-test
+            docker rm mteb-hf-test
+            exit 0
+          fi
+
+          if [ $((ELAPSED % 30)) -eq 0 ] && [ $ELAPSED -gt 0 ]; then
+            echo "Container still running after ${ELAPSED}s... (startup: $STARTUP_DETECTED, init_complete: $INIT_COMPLETE_DETECTED)"
+            echo "$LOGS" | grep -E "(Step [0-9]/7|Starting leaderboard|initialization complete|Running on)" | tail -2
+          fi
+
+          sleep 5
+        done

Dockerfile.hf-space

@@ -0,0 +1,6 @@
+FROM ghcr.io/embeddings-benchmark/mteb/leaderboard:latest
+
+ENV GRADIO_SERVER_NAME="0.0.0.0"
+EXPOSE 7860
+
+CMD ["make", "run-leaderboard"]