Merge pull request #2 from ruchernchong/1-run-python-script-when-a-pd…

…f-file-is-added-or-updated

Add GHA workflow to monitor for PDF changes

.github/workflows/pdf-monitor.yml

@@ -0,0 +1,56 @@
+name: PDF Monitor Workflow
+
+on:
+  push:
+    paths:
+      - '**/*.pdf'
+
+jobs:
+  run-python-script:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout Repository
+        uses: actions/checkout@v4
+      - name: Set up Python
+        uses: actions/setup-python@v4
+        with:
+          python-version: 3.12
+          cache: pip
+      - name: Install Dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install -r requirements.txt
+      - name: Find Changed PDF Files
+        id: find_pdfs
+        run: |
+          if [ $(git rev-parse --is-shallow-repository) = true ]; then
+            # For shallow clones, fetch the full history
+            git fetch --unshallow
+          fi
+          
+          if [ $(git rev-list --count HEAD) -eq 1 ]; then
+            # If only one commit exists, list all PDFs
+            pdf_files=$(git ls-files '*.pdf')
+          else
+            # For subsequent commits, list changed PDFs
+            pdf_files=$(git diff --name-only HEAD^ HEAD | grep '.pdf$')
+          fi
+          echo "pdf_files=$pdf_files" >> $GITHUB_ENV
+      - name: Run Python Script
+        if: env.pdf_files != ''
+        run: |
+          for pdf_file in ${{ env.pdf_files }}; do
+            python main.py "$pdf_file"
+          done
+      - name: Commit and Push Changes
+        if: ${{ env.pdf_files != '' }}
+        run: |
+          git config --global user.name "github-actions[bot]"
+          git config --global user.email "github-actions[bot]@users.noreply.github.com"
+          git add .
+          if git diff --cached --quiet; then
+            echo "No changes to commit"
+          else
+            git commit -m "Automated commit of changes made by PDF Monitor Workflow"
+            git push
+          fi

main.py

@@ -1,3 +1,5 @@
+import sys
+
 import matplotlib.pyplot as plt
 import numpy as np
 
@@ -6,7 +8,7 @@
 from utils.replace_text import replace_text
 
 # Step 1: Extract data from the PDF
-data = read_pdf.extract_data_from_pdf('Rev_by_Mkt_Qtrly_Trend_Q125.pdf')
+data = read_pdf.extract_data_from_pdf(sys.argv[1])
 
 # Step 2: Assign data to variables
 quarters = data['quarters']