Merge pull request #22 from sahil280114/main

add github action for datagen

.github/workflows/glaive_datagen.yaml

@@ -0,0 +1,58 @@
+name: Build dataset
+
+on:
+  pull_request:
+    types: [closed]
+    branches:
+      - main
+
+jobs:
+  notify:
+    if: github.event.pull_request.merged == true
+    runs-on: ubuntu-latest
+
+    steps:
+    - name: Checkout code
+      uses: actions/checkout@v3
+
+    - name: Set up Python
+      uses: actions/setup-python@v4
+      with:
+        python-version: '3.x'
+
+    - name: Get list of added files
+      id: get_added_files
+      run: |
+        # Directory to check for added files
+        DIRECTORY_TO_CHECK="tasks"
+
+        # Get list of added files in the specified directory
+        ADDED_FILES=$(git diff --name-status ${{ github.event.before }} ${{ github.sha }} | awk '/^A/ {print $2}' | grep "^$DIRECTORY_TO_CHECK")
+
+        # Export the list as an environment variable
+        echo "ADDED_FILES=$ADDED_FILES" >> $GITHUB_ENV
+
+        # Print the files for debugging
+        echo "Added files:"
+        echo "$ADDED_FILES"
+        # Convert space-separated string to JSON array
+        JSON_ARRAY=$(echo $ADDED_FILES | jq -R -s -c 'split(" ")')
+
+        # Export the JSON array as an environment variable
+        echo "ADDED_FILES_JSON=$JSON_ARRAY" >> $GITHUB_ENV
+
+        # Print the JSON array for debugging
+        echo "Added files JSON array:"
+        echo "$JSON_ARRAY"
+
+    - name: Notify API
+      if: steps.get_added_files.outputs.added_files != ''
+      run: |
+        # Define the API endpoint and token
+        API_ENDPOINT="https://open-reasoning-tasks-action-production.up.railway.app/process-files"
+
+        # Send the list of added files to the API endpoint
+        curl -X POST $API_ENDPOINT \
+             -H "Content-Type: application/json" \
+             -H "Authorization: Bearer ${{ secrets.GLAIVE_API_AUTH_TOKEN }}" \
+             -d "{\"added_files\": \"$ADDED_FILES_JSON\"}"