Merge pull request #24004 from abpframework/salihozkara/seo

Refactor SEO workflow to use GitHub API for changed files

.github/workflows/auto-add-seo.yml

@@ -23,7 +23,7 @@ jobs:
       - name: Checkout code
         uses: actions/checkout@v4
         with:
-          ref: ${{ github.event.pull_request.merge_commit_sha }}
+          ref: ${{ github.event.pull_request.base.ref }}
           fetch-depth: 0
           token: ${{ secrets.GITHUB_TOKEN }}
 
@@ -36,78 +36,90 @@ jobs:
         run: |
           pip install openai
 
+      - name: Get changed markdown files from merged PR using GitHub API
+        id: changed-files
+        uses: actions/github-script@v7
+        with:
+          script: |
+            const prNumber = ${{ github.event.pull_request.number }};
+            
+            // Get all files changed in the PR with pagination
+            const allFiles = [];
+            let page = 1;
+            let hasMore = true;
+            
+            while (hasMore) {
+              const { data: files } = await github.rest.pulls.listFiles({
+                owner: context.repo.owner,
+                repo: context.repo.repo,
+                pull_number: prNumber,
+                per_page: 100,
+                page: page
+              });
+              
+              allFiles.push(...files);
+              hasMore = files.length === 100;
+              page++;
+            }
+            
+            console.log(`Total files changed in PR: ${allFiles.length}`);
+            
+            // Filter for only added/modified markdown files in docs/en/
+            const changedMdFiles = allFiles
+              .filter(file => 
+                (file.status === 'added' || file.status === 'modified') &&
+                file.filename.startsWith('docs/en/') &&
+                file.filename.endsWith('.md')
+              )
+              .map(file => file.filename);
+            
+            console.log(`\nFound ${changedMdFiles.length} added/modified markdown files in docs/en/:`);
+            changedMdFiles.forEach(file => console.log(`  - ${file}`));
+            
+            // Write to environment file for next steps
+            const fs = require('fs');
+            fs.writeFileSync(process.env.GITHUB_OUTPUT, 
+              `any_changed=${changedMdFiles.length > 0 ? 'true' : 'false'}\n` +
+              `all_changed_files=${changedMdFiles.join(' ')}\n`,
+              { flag: 'a' }
+            );
+            
+            return changedMdFiles;
+
       - name: Create new branch for SEO updates
+        if: steps.changed-files.outputs.any_changed == 'true'
         run: |
           git config --local user.email "github-actions[bot]@users.noreply.github.com"
           git config --local user.name "github-actions[bot]"
           
-          # Checkout base branch first
-          git checkout ${{ github.event.pull_request.base.ref }}
-          
-          # Create new branch from base
+          # Create new branch from current base branch (which already has merged files)
           BRANCH_NAME="auto-docs-seo/${{ github.event.pull_request.number }}"
           git checkout -b $BRANCH_NAME
           echo "BRANCH_NAME=$BRANCH_NAME" >> $GITHUB_ENV
-
-      - name: Get changed markdown files from merged PR
-        id: changed-files
-        run: |
-          # Get the list of commits in the PR
-          PR_HEAD_SHA="${{ github.event.pull_request.head.sha }}"
-          PR_BASE_SHA="${{ github.event.pull_request.base.sha }}"
-          
-          echo "PR commits range: $PR_BASE_SHA..$PR_HEAD_SHA"
           
-          # Get files changed in the PR commits only (only Added and Modified, exclude Deleted)
-          FILES=$(git diff --name-only --diff-filter=AM $PR_BASE_SHA..$PR_HEAD_SHA | grep 'docs/en/.*\.md$' || true)
-          
-          echo "Files changed in the merged PR (added/modified only):"
-          echo "$FILES"
+          echo "✅ Created branch: $BRANCH_NAME"
           echo ""
-          
-          echo "changed_files<<EOF" >> $GITHUB_OUTPUT
-          echo "$FILES" >> $GITHUB_OUTPUT
-          echo "EOF" >> $GITHUB_OUTPUT
-          
-          if [ -z "$FILES" ]; then
-            echo "has_files=false" >> $GITHUB_OUTPUT
-            echo "No markdown files changed in docs/en/"
-          else
-            echo "has_files=true" >> $GITHUB_OUTPUT
-            
-            # Checkout the changed files from merge commit to get the merged content
-            echo ""
-            echo "Checking out changed files from merge commit..."
-            while IFS= read -r file; do
-              if [ -n "$file" ]; then
-                echo "  Checking out: $file"
-                # Create directory if it doesn't exist
-                mkdir -p "$(dirname "$file")"
-                # Checkout the file from merge commit
-                if ! git show "${{ github.event.pull_request.merge_commit_sha }}:$file" > "$file" 2>err.log; then
-                  echo "  Warning: Could not checkout $file"
-                  echo "    Reason: $(cat err.log)"
-                fi
-              fi
-            done <<< "$FILES"
-            
-            echo ""
-            echo "Files now in working directory:"
-            git status --short
-          fi
+          echo "📝 Files to process for SEO descriptions:"
+          for file in ${{ steps.changed-files.outputs.all_changed_files }}; do
+            if [ -f "$file" ]; then
+              echo "  ✓ $file"
+            else
+              echo "  ✗ $file (not found)"
+            fi
+          done
 
       - name: Process changed files and add SEO descriptions
-        if: steps.changed-files.outputs.has_files == 'true'
+        if: steps.changed-files.outputs.any_changed == 'true'
         env:
           OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
           IGNORED_FOLDERS: ${{ vars.DOCS_SEO_IGNORED_FOLDERS }}
-          CHANGED_FILES: ${{ steps.changed-files.outputs.changed_files }}
+          CHANGED_FILES: ${{ steps.changed-files.outputs.all_changed_files }}
         run: |
           python3 .github/scripts/add_seo_descriptions.py
 
 
       - name: Commit and push changes
-        if: steps.changed-files.outputs.has_files == 'true'
+        if: steps.changed-files.outputs.any_changed == 'true'
         run: |
           git add -A docs/en/
           
@@ -115,9 +127,11 @@ jobs:
             echo "No changes to commit"
             echo "has_commits=false" >> $GITHUB_ENV
           else
+            BRANCH_NAME="auto-docs-seo/${{ github.event.pull_request.number }}"
             git commit -m "docs: Add SEO descriptions to modified documentation files" -m "Related to PR #${{ github.event.pull_request.number }}"
-            git push origin ${{ env.BRANCH_NAME }}
+            git push origin $BRANCH_NAME
             echo "has_commits=true" >> $GITHUB_ENV
+            echo "BRANCH_NAME=$BRANCH_NAME" >> $GITHUB_ENV
           fi
 
       - name: Create Pull Request
@@ -133,7 +147,7 @@ jobs:
             const skippedIgnored = parseInt(stats[3]) || 0;
             const prNumber = ${{ github.event.pull_request.number }};
             const baseRef = '${{ github.event.pull_request.base.ref }}';
-            const branchName = '${{ env.BRANCH_NAME }}';
+            const branchName = `auto-docs-seo/${prNumber}`;
             
             if (processedCount > 0) {
               // Read the actually updated files list (not all changed files)