name: CI Pipeline

# Triggers for the workflow
on:
  schedule:
    - cron: '0 0 * * *' # Runs daily at midnight UTC
  push:
    branches:
      - main
  pull_request:
    branches:
      - main
  workflow_dispatch: # Allows manual triggering from GitHub UI

# Environment variables available to all jobs
env:
  BRANCH_NAME: "update-submodules_${{ github.sha }}"
  REPO_URL: "github.com/MuRuLOSE/limoka.git" # Repository URL
  GIT_DEPTH: 0 # Fetch full history for Git operations

jobs:
  before:
    runs-on: ubuntu-latest
    # Run on schedule or manual dispatch
    if: github.event_name == 'schedule' || github.event_name == 'workflow_dispatch'
    steps:
      - name: Configure Git for github-actions[bot]
        run: |
          git config --global user.email "github-actions[bot]@users.noreply.github.com"
          git config --global user.name "github-actions[bot]"
          git config --global --list
      - name: Checkout repository
        uses: actions/checkout@v4
        with:
          fetch-depth: ${{ env.GIT_DEPTH }}
      - name: Set up Python 3.9
        uses: actions/setup-python@v5
        with:
          python-version: '3.9'
      - name: Install Python dependencies
        run: pip install requests
      - name: Sync main branch and create new branch
        env:
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
        run: |
          git fetch origin
          git checkout main
          git remote set-url origin "https://x-access-token:${GITHUB_TOKEN}@${REPO_URL}"
          echo "Synchronizing main with origin/main..."
          git reset --hard origin/main
          echo "Deleting remote branch ${BRANCH_NAME}..."
          git push origin --delete ${BRANCH_NAME} || echo "Branch ${BRANCH_NAME} did not exist or could not be deleted"
          echo "Deleting local branch ${BRANCH_NAME}..."
          git branch -D ${BRANCH_NAME} || echo "Local branch ${BRANCH_NAME} did not exist"
          git checkout -b ${BRANCH_NAME}
          echo "Creating new branch ${BRANCH_NAME}..."
          git push origin ${BRANCH_NAME} --force || echo "Error creating branch"

  update:
    runs-on: ubuntu-latest
    # Run on schedule or manual dispatch
    if: github.event_name == 'schedule' || github.event_name == 'workflow_dispatch'
    needs: before
    steps:
      - name: Configure Git for github-actions[bot]
        run: |
          git config --global user.email "github-actions[bot]@users.noreply.github.com"
          git config --global user.name "github-actions[bot]"
          git config --global --list
      - name: Checkout repository
        uses: actions/checkout@v4
        with:
          fetch-depth: ${{ env.GIT_DEPTH }}
          ref: ${{ env.BRANCH_NAME }}
      - name: Set up Python 3.9
        uses: actions/setup-python@v5
        with:
          python-version: '3.9'
      - name: Install Python dependencies
        run: pip install requests
      - name: Clone and update repositories
        env:
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
        run: |
          git checkout ${{ env.BRANCH_NAME }}
          echo "Cloning and updating repositories..."
          python3 clone_repos.py || echo "Error running clone_repos.py, continuing with commit and push"
          git add .
          git commit -m "Added and updated repositories $(date +'%Y-%m-%d %H:%M:%S')" || echo "No changes for commit"
          git remote set-url origin "https://x-access-token:${GITHUB_TOKEN}@${REPO_URL}"
          git push origin ${{ env.BRANCH_NAME }} --force

  parse:
    runs-on: ubuntu-latest
    if: |
      github.event_name == 'schedule' ||
      github.event_name == 'workflow_dispatch' ||
      (github.event_name == 'push' && github.ref == 'refs/heads/main') ||
      (github.event_name == 'pull_request' && github.event.action == 'closed' && github.event.pull_request.merged == true)
    needs: update
    steps:
      - name: Debug event and branch
        run: |
          echo "Event name: ${{ github.event_name }}"
          echo "Branch name: ${{ env.BRANCH_NAME }}"
          echo "Ref to checkout: ${{ (github.event_name == 'schedule' || github.event_name == 'workflow_dispatch') && env.BRANCH_NAME || 'main' }}"
      - name: Configure Git for github-actions[bot]
        run: |
          git config --global user.email "github-actions[bot]@users.noreply.github.com"
          git config --global user.name "github-actions[bot]"
          git config --global --list
      - name: Checkout repository
        uses: actions/checkout@v4
        with:
          fetch-depth: ${{ env.GIT_DEPTH }}
          ref: ${{ github.event_name == 'schedule' || github.event_name == 'workflow_dispatch' && env.BRANCH_NAME || 'main' }}
      - name: Debug Git state
        run: |
          echo "Current branch: $(git rev-parse --abbrev-ref HEAD)"
          git branch --list
          git tag --list
      - name: Set up Python 3.9
        uses: actions/setup-python@v5
        with:
          python-version: '3.9'
      - name: Run parse scripts
        env:
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
        run: |
          echo "Running parse after update..."
          git fetch origin
          git checkout ${{ github.event_name == 'schedule' || github.event_name == 'workflow_dispatch' && env.BRANCH_NAME || 'main' }}
          git reset --hard origin/${{ github.event_name == 'schedule' || github.event_name == 'workflow_dispatch' && env.BRANCH_NAME || 'main' }}
          python3 -m venv venv
          source venv/bin/activate
          pip install --upgrade pip
          pip install requests scikit-learn tqdm
          python3 parse.py
          python3 categories.py
          git add modules.json
          git commit -m "Updated modules.json after parse $(date +'%Y-%m-%d %H:%M:%S')" || echo "No changes for modules.json"
          git remote set-url origin "https://x-access-token:${GITHUB_TOKEN}@${REPO_URL}"
          git push origin ${{ github.event_name == 'schedule' || github.event_name == 'workflow_dispatch' && env.BRANCH_NAME || 'main' }}

  commit:
    runs-on: ubuntu-latest
    # Run on schedule or manual dispatch
    if: github.event_name == 'schedule' || github.event_name == 'workflow_dispatch'
    needs: parse
    steps:
      - name: Configure Git for github-actions[bot]
        run: |
          git config --global user.email "github-actions[bot]@users.noreply.github.com"
          git config --global user.name "github-actions[bot]"
          git config --global --list
      - name: Checkout repository
        uses: actions/checkout@v4
        with:
          fetch-depth: ${{ env.GIT_DEPTH }}
          ref: ${{ env.BRANCH_NAME }}
      - name: Final commit
        env:
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
        run: |
          git checkout ${{ env.BRANCH_NAME }}
          git add . || echo "No changes"
          git commit -m "Final commit $(date +'%Y-%m-%d %H:%M:%S')" || echo "No changes for final commit"
          git remote set-url origin "https://x-access-token:${GITHUB_TOKEN}@${REPO_URL}"
          git push origin ${{ env.BRANCH_NAME }} --force

  create_pr:
    runs-on: ubuntu-latest
    # Run on schedule or manual dispatch
    if: github.event_name == 'schedule' || github.event_name == 'workflow_dispatch'
    needs: commit
    steps:
      - name: Configure Git for github-actions[bot]
        run: |
          git config --global user.email "github-actions[bot]@users.noreply.github.com"
          git config --global user.name "github-actions[bot]"
          git config --global --list
      - name: Checkout repository
        uses: actions/checkout@v4
        with:
          fetch-depth: ${{ env.GIT_DEPTH }}
      - name: Create Pull Request
        env:
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
        run: |
          echo "Checking branch status before PR..."
          git fetch origin
          # Check if the branch exists in the remote repository
          if git ls-remote --exit-code --heads origin ${{ env.BRANCH_NAME }} >/dev/null 2>&1; then
            echo "Branch ${{ env.BRANCH_NAME }} exists in remote repository."
            git log origin/${{ env.BRANCH_NAME }} -1 || echo "Failed to view log for ${{ env.BRANCH_NAME }}"
            git diff origin/main origin/${{ env.BRANCH_NAME }} || echo "No diff between main and ${{ env.BRANCH_NAME }}"
            echo "Creating Pull Request..."
            gh pr create --base main --head ${{ env.BRANCH_NAME }} --title "Update of repositories $(date +'%Y-%m-%d %H:%M:%S')" --body "Automated PR for repository updates" || echo "Failed to create PR"
          else
            echo "Branch ${{ env.BRANCH_NAME }} does not exist in remote repository, skipping PR creation."
          fi

  backup:
    runs-on: ubuntu-latest
    # Run on main branch or manual dispatch
    if: github.ref == 'refs/heads/main' || github.event_name == 'workflow_dispatch'
    needs: parse
    steps:
      - name: Configure Git for github-actions[bot]
        run: |
          git config --global user.email "github-actions[bot]@users.noreply.github.com"
          git config --global user.name "github-actions[bot]"
          git config --global --list
      - name: Checkout repository
        uses: actions/checkout@v4
        with:
          fetch-depth: ${{ env.GIT_DEPTH }}
      - name: Create and send backup to Telegram
        env:
          TELEGRAM_BOT_TOKEN: ${{ secrets.TELEGRAM_BOT_TOKEN }}
          TELEGRAM_CHAT_ID: ${{ secrets.TELEGRAM_CHAT_ID }}
        run: |
          echo "Creating .zip file of the repository with maximum compression..."
          git archive --format=zip --output=repository-original.zip HEAD
          zip -9 repository.zip repository-original.zip
          rm repository-original.zip
          echo "File size of the created .zip file:"
          du -sh repository.zip
          echo "Splitting the .zip file into 8 parts..."
          split -b 49M repository.zip repository-part-
          echo "Files created after split:"
          ls repository-part-*
          COMMIT_MESSAGE="$(git log -1 --pretty=%B)"
          COMMIT_DATE="$(date --date="$(git log -1 --pretty=%ci)" +'%Y-%m-%d %H:%M:%S')"
          COMMIT_HASH="$(git rev-parse --short=6 HEAD)"
          COMMIT_URL="https://${REPO_URL}/commit/$(git rev-parse HEAD)"
          MESSAGE="Commit Date: $COMMIT_DATE, Commit Message: $COMMIT_MESSAGE, Commit Hash: [\`$COMMIT_HASH\`]($COMMIT_URL)"
          
          echo "Sending .zip file parts to Telegram..."
          
          FIRST_PART=true
          for part in $(ls repository-part-* | sort); do
            if $FIRST_PART; then
              TELEGRAM_API_URL="https://api.telegram.org/bot$TELEGRAM_BOT_TOKEN/sendDocument"
              echo "Sending first file to Telegram: $TELEGRAM_API_URL"
              curl -X POST "$TELEGRAM_API_URL" \
                -F chat_id=$TELEGRAM_CHAT_ID \
                -F document=@$part \
                -F caption="$MESSAGE" \
                -F parse_mode="Markdown"
              FIRST_PART=false
            else
              TELEGRAM_API_URL="https://api.telegram.org/bot$TELEGRAM_BOT_TOKEN/sendDocument"
              echo "Sending file to Telegram: $TELEGRAM_API_URL"
              curl -X POST "$TELEGRAM_API_URL" \
                -F chat_id=$TELEGRAM_CHAT_ID \
                -F document=@$part
            fi
          done
          
          echo "Files sent to Telegram successfully!"