Incorporate ScanCode into CI for OSS compliance

.github/workflows/ci.yml

@@ -186,11 +186,30 @@ jobs:
       - name: Run cppcheck
         run: |
           docker run --rm --volume $(pwd):/note-c/ --workdir /note-c/ --entrypoint ./scripts/run_cppcheck.sh ghcr.io/blues/note_c_ci:latest
+
+  run_scancode:
+    runs-on: ubuntu-latest
+    if: ${{ always() }}
+    needs: [build_ci_docker_image]
+
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Load CI Docker image
+        # Only load the Docker image artifact if build_ci_docker_image actually
+        # ran (e.g. it wasn't skipped and was successful).
+        if: ${{ needs.build_ci_docker_image.result == 'success' }}
+        uses: ./.github/actions/load-ci-image
+
+      - name: Run license compliance check
+        run: |
+          docker run --rm --volume $(pwd):/note-c/ --workdir /note-c/ --entrypoint ./scripts/run_scancode.sh ghcr.io/blues/note_c_ci:latest
 
   publish_ci_image:
     runs-on: ubuntu-latest
     # Make sure unit tests unit tests passed before publishing.
-    needs: [build_ci_docker_image, run_unit_tests]
+    needs: [build_ci_docker_image, run_unit_tests, run_scancode]
     # Only publish the image if this is a push event and the Docker image was rebuilt
     if: ${{ github.event_name == 'push' && needs.build_ci_docker_image.result == 'success' }}
 

.vscode/tasks.json

@@ -120,5 +120,19 @@
             ],
             "group": "test"
         }
+    ,
+        {
+            "label": "Note-C: Run License Compliance Check",
+            "type": "shell",
+            "command": "${workspaceFolder}/scripts/run_scancode.sh",
+            "options": {
+                "cwd": "${workspaceFolder}",
+                "env": {
+                    "LC_ALL": "C"
+                }
+            },
+            "problemMatcher": [],
+            "group": "test"
+        }
     ]
 }

Dockerfile

@@ -70,6 +70,7 @@ RUN ["dash", "-c", "\
  && pip install --break-system-packages \
      breathe \
      sphinx-rtd-theme \
+     scancode-toolkit \
  && apt-get clean \
  && apt-get purge \
  && rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/* \

scripts/run_scancode.sh

@@ -0,0 +1,103 @@
+#!/bin/bash
+set -eo pipefail
+
+SCRIPT_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" &> /dev/null && pwd )"
+SRC_DIR="$SCRIPT_DIR/.."
+
+echo "Running License Compliance Analysis..."
+echo
+
+# Create a function to generate the summary
+generate_summary() {
+    {
+        # Initialize flag
+        has_critical_issues=false
+
+        echo
+
+        # Always generate and display summary regardless of exit code
+        echo "=== License Compliance Summary ==="
+        echo
+
+        # Display license information
+        echo "License Information:"
+        echo "-------------------"
+        jq -r '.summary.declared_license_expression' scancode_output.json || echo "None found"
+        echo
+
+        # Display license clarity score
+        echo "License Clarity Score:"
+        echo "---------------------"
+        jq -r '.summary.license_clarity_score.score' scancode_output.json || echo "Not available"
+        echo
+
+        # Display other licenses found
+        echo "Other Licenses Found:"
+        echo "--------------------"
+        jq -r '.summary.other_license_expressions[] | select(.value != null) | "\(.value): \(.count) occurrences"' scancode_output.json || echo "None found"
+        echo
+
+        # Display copyright holders
+        echo "Copyright Holders:"
+        echo "-----------------"
+        jq -r '.tallies.holders[] | select(.value != null) | "\(.value): \(.count) occurrences"' scancode_output.json || echo "None found"
+        echo
+
+        # Check for non-compliant licenses
+        echo "License Compliance Issues:"
+        echo "-------------------------"
+        # List of non-compliant or problematic licenses
+        problematic_licenses=("gpl-2.0" "gpl-3.0" "agpl-3.0" "cc-by-nc" "proprietary")
+
+        found_problematic=false
+        for license in "${problematic_licenses[@]}"; do
+            count=$(jq -r ".summary.other_license_expressions[] | select(.value == \"$license\") | .count" scancode_output.json 2>/dev/null || echo "0")
+            if [ "$count" != "0" ] && [ "$count" != "" ]; then
+                echo "WARNING: Found $count occurrences of $license license, which may have compliance implications."
+                found_problematic=true
+                has_critical_issues=true
+            fi
+        done
+
+        if ! $found_problematic; then
+            echo "No problematic licenses found."
+        fi
+        echo
+
+        # Display status and details
+        if $has_critical_issues; then
+            echo "Status: FAILED - License compliance issues found"
+            echo
+            echo "Review and fix license compliance issues before proceeding"
+        else
+            echo "Status: PASSED - No license compliance issues found"
+            echo
+            echo "Note: Review license information for potential improvements"
+        fi
+    }
+
+    # Return 1 if critical issues found, 0 otherwise
+    if $has_critical_issues; then
+        return 1
+    else
+        return 0
+    fi
+}
+
+# Run scancode and capture output and exit code
+scancode \
+    --license \
+    --copyright \
+    --classify \
+    --summary \
+    --license-clarity-score \
+    --tallies \
+    --json-pp scancode_output.json \
+    --timeout 120 \
+    --processes 2 \
+    . 2>&1 | tee scancode_console_output.txt
+
+generate_summary
+
+# Exit with scancode's status code
+exit $?