Fix fork divergence: add clean upstream PR workflow, fix rebase merge-base bug, generalize sync backup

.github/fork-specific-files.txt

@@ -0,0 +1,10 @@
+.github/dependabot.yml
+.github/workflows/codacy.yml
+.github/workflows/codeql.yml
+.github/workflows/maven.yml
+.github/workflows/rebase.yml
+.github/workflows/rebase-upstream.yml
+.github/workflows/sync-upstream.yml
+.github/workflows/create-upstream-pr.yml
+.github/fork-specific-files.txt
+README.md

.github/workflows/create-upstream-pr.yml

@@ -0,0 +1,126 @@
+name: Create Upstream PR Branch
+
+on:
+  workflow_dispatch:
+    inputs:
+      issue_number:
+        description: 'Issue number (used for branch name: upstream-pr/issue-{number})'
+        required: true
+        type: string
+      source_branch:
+        description: 'Source branch containing the commits to cherry-pick'
+        required: true
+        type: string
+      commit_shas:
+        description: 'Comma-separated list of commit SHAs to cherry-pick onto upstream/master'
+        required: true
+        type: string
+      pr_title:
+        description: 'Title for the upstream PR (informational only)'
+        required: false
+        type: string
+
+jobs:
+  create-upstream-pr-branch:
+    name: Create Clean Upstream PR Branch
+    runs-on: ubuntu-latest
+    permissions:
+      contents: write
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+          token: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Configure Git
+        run: |
+          git config user.name "github-actions[bot]"
+          git config user.email "github-actions[bot]@users.noreply.github.com"
+
+      - name: Fetch upstream master
+        run: |
+          git remote add upstream https://github.com/eclipse-jdt/eclipse.jdt.ui.git
+          git fetch upstream master
+
+      - name: Create branch based on upstream/master
+        run: |
+          BRANCH_NAME="upstream-pr/issue-${{ inputs.issue_number }}"
+          git checkout -b "$BRANCH_NAME" upstream/master
+          echo "Created branch: $BRANCH_NAME"
+
+      - name: Cherry-pick specified commits
+        id: cherry_pick
+        run: |
+          BRANCH_NAME="upstream-pr/issue-${{ inputs.issue_number }}"
+          COMMIT_SHAS="${{ inputs.commit_shas }}"
+
+          # Convert comma-separated list to space-separated
+          SHAS=$(echo "$COMMIT_SHAS" | tr ',' ' ' | tr -s ' ')
+
+          echo "Cherry-picking commits: $SHAS"
+
+          SUCCESS=true
+          for SHA in $SHAS; do
+            SHA=$(echo "$SHA" | xargs)  # trim whitespace
+            if [ -z "$SHA" ]; then
+              continue
+            fi
+            echo "Cherry-picking $SHA..."
+            if ! git cherry-pick "$SHA"; then
+              echo "Cherry-pick failed for $SHA"
+              git cherry-pick --abort 2>/dev/null || true
+              SUCCESS=false
+              break
+            fi
+          done
+
+          if [ "$SUCCESS" = "true" ]; then
+            echo "success=true" >> $GITHUB_OUTPUT
+          else
+            echo "success=false" >> $GITHUB_OUTPUT
+          fi
+
+      - name: Verify branch contents
+        if: steps.cherry_pick.outputs.success == 'true'
+        run: |
+          echo "=== Commits on this branch (not in upstream/master) ==="
+          git log upstream/master..HEAD --oneline
+          echo ""
+          echo "=== Files changed vs upstream/master ==="
+          git diff upstream/master..HEAD --stat
+
+      - name: Force-push branch to fork
+        if: steps.cherry_pick.outputs.success == 'true'
+        run: |
+          BRANCH_NAME="upstream-pr/issue-${{ inputs.issue_number }}"
+          git push --force origin "$BRANCH_NAME"
+          echo "✅ Branch '$BRANCH_NAME' pushed to fork"
+
+      - name: Output PR creation link
+        if: steps.cherry_pick.outputs.success == 'true'
+        run: |
+          BRANCH_NAME="upstream-pr/issue-${{ inputs.issue_number }}"
+          REPO_OWNER="${{ github.repository_owner }}"
+          REPO_NAME="${{ github.event.repository.name }}"
+          PR_LINK="https://github.com/eclipse-jdt/eclipse.jdt.ui/compare/master...${REPO_OWNER}:${REPO_NAME}:${BRANCH_NAME}"
+
+          echo ""
+          echo "=========================================="
+          echo "✅ Branch ready for upstream PR!"
+          echo "=========================================="
+          echo ""
+          echo "Branch: $BRANCH_NAME"
+          if [ -n "${{ inputs.pr_title }}" ]; then
+            echo "PR Title: ${{ inputs.pr_title }}"
+          fi
+          echo ""
+          echo "Create PR at:"
+          echo "$PR_LINK"
+          echo ""
+
+      - name: Fail if cherry-pick failed
+        if: steps.cherry_pick.outputs.success == 'false'
+        run: |
+          echo "❌ Cherry-pick failed. Check the logs above for details."
+          exit 1

.github/workflows/rebase-upstream.yml

@@ -114,17 +114,6 @@ jobs:
             core.setOutput('head_ref', pr.head.ref);
             core.setOutput('head_repo', pr.head.repo.full_name);
             core.setOutput('head_clone_url', pr.head.repo.clone_url);
-
-            // Get the number of commits in the PR
-            // Note: per_page is set to 250, which covers most PRs.
-            // For PRs with >250 commits, manual rebasing would be required.
-            const { data: commits } = await github.rest.pulls.listCommits({
-              owner: context.repo.owner,
-              repo: context.repo.repo,
-              pull_number: context.issue.number,
-              per_page: 250
-            });
-            core.setOutput('commit_count', commits.length);
 
       - name: Checkout PR branch
         uses: actions/checkout@v4
@@ -145,19 +134,22 @@ jobs:
           git remote add upstream https://github.com/eclipse-jdt/eclipse.jdt.ui.git
           git fetch upstream master
 
-          # Get the number of commits in the PR
-          PR_COMMIT_COUNT=${{ steps.pr.outputs.commit_count }}
-          echo "PR has $PR_COMMIT_COUNT commits"
+          # Find the actual fork point with upstream
+          MERGE_BASE=$(git merge-base HEAD upstream/master)
+          echo "Merge base with upstream: $MERGE_BASE"
+
+          # Count actual commits to rebase (between merge-base and HEAD)
+          ACTUAL_COMMITS=$(git rev-list --count $MERGE_BASE..HEAD)
+          echo "Actual commits to rebase: $ACTUAL_COMMITS"
 
-          # Validate commit count
-          if [ "$PR_COMMIT_COUNT" -eq 0 ]; then
-            echo "Error: PR has 0 commits"
+          if [ "$ACTUAL_COMMITS" -eq 0 ]; then
+            echo "Error: No commits to rebase"
             echo "success=false" >> $GITHUB_OUTPUT
             exit 0
           fi
 
-          # Rebase only the PR commits onto upstream/master
-          if git rebase --onto upstream/master HEAD~${PR_COMMIT_COUNT}; then
+          # Rebase using merge-base (correct: only replays commits after the fork point)
+          if git rebase --onto upstream/master $MERGE_BASE; then
             echo "success=true" >> $GITHUB_OUTPUT
           else
             echo "success=false" >> $GITHUB_OUTPUT

.github/workflows/sync-upstream.yml

@@ -55,24 +55,28 @@ jobs:
           # Create temporary directory for fork-specific files
           mkdir -p /tmp/fork-specific
 
-          # Identify fork-specific workflow files (files that don't exist in upstream)
-          git fetch upstream master
-
-          # Get list of workflow files in fork
-          FORK_WORKFLOWS=$(git ls-tree -r HEAD --name-only .github/workflows/ 2>/dev/null || echo "")
+          # Read the list of fork-specific files from the manifest
+          MANIFEST=".github/fork-specific-files.txt"
+          if [ ! -f "$MANIFEST" ]; then
+            echo "Warning: $MANIFEST not found, no files to backup"
+            echo "has_fork_files=false" >> $GITHUB_OUTPUT
+            exit 0
+          fi
 
-          # For each workflow file in fork, check if it exists in upstream
-          for file in $FORK_WORKFLOWS; do
-            if ! git cat-file -e upstream/master:"$file" 2>/dev/null; then
-              echo "Fork-specific file: $file"
-              # Create directory structure and copy file
+          HAS_FILES=false
+          while IFS= read -r file || [ -n "$file" ]; do
+            # Skip empty lines and comments
+            [ -z "$file" ] && continue
+            case "$file" in \#*) continue ;; esac
+            if [ -f "$file" ]; then
+              echo "Backing up fork-specific file: $file"
               mkdir -p "/tmp/fork-specific/$(dirname "$file")"
               cp "$file" "/tmp/fork-specific/$file"
+              HAS_FILES=true
             fi
-          done
+          done < "$MANIFEST"
 
-          # Check if any fork-specific files were found
-          if [ -d "/tmp/fork-specific/.github" ]; then
+          if [ "$HAS_FILES" = "true" ]; then
             echo "has_fork_files=true" >> $GITHUB_OUTPUT
             echo "Found fork-specific files:"
             find /tmp/fork-specific -type f
@@ -90,13 +94,13 @@ jobs:
       - name: Restore fork-specific files
         if: steps.check_sync.outputs.needs_sync == 'true' && steps.backup.outputs.has_fork_files == 'true'
         run: |
-          # Copy fork-specific files back
+          # Restore all backed-up fork-specific files
           if [ -d "/tmp/fork-specific" ]; then
-            cp -r /tmp/fork-specific/.github ./ 2>/dev/null || true
+            cp -r /tmp/fork-specific/. ./
           fi
 
-          # Add all fork-specific changes
-          git add .github/
+          # Stage all restored files
+          git add -A
 
       - name: Commit fork-specific changes
         if: steps.check_sync.outputs.needs_sync == 'true' && steps.backup.outputs.has_fork_files == 'true'
@@ -214,24 +218,28 @@ jobs:
           # Create temporary directory for fork-specific files
           mkdir -p /tmp/fork-specific
 
-          # Identify fork-specific workflow files (files that don't exist in upstream)
-          git fetch upstream master
-
-          # Get list of workflow files in fork
-          FORK_WORKFLOWS=$(git ls-tree -r HEAD --name-only .github/workflows/ 2>/dev/null || echo "")
+          # Read the list of fork-specific files from the manifest
+          MANIFEST=".github/fork-specific-files.txt"
+          if [ ! -f "$MANIFEST" ]; then
+            echo "Warning: $MANIFEST not found, no files to backup"
+            echo "has_fork_files=false" >> $GITHUB_OUTPUT
+            exit 0
+          fi
 
-          # For each workflow file in fork, check if it exists in upstream
-          for file in $FORK_WORKFLOWS; do
-            if ! git cat-file -e upstream/master:"$file" 2>/dev/null; then
-              echo "Fork-specific file: $file"
-              # Create directory structure and copy file
+          HAS_FILES=false
+          while IFS= read -r file || [ -n "$file" ]; do
+            # Skip empty lines and comments
+            [ -z "$file" ] && continue
+            case "$file" in \#*) continue ;; esac
+            if [ -f "$file" ]; then
+              echo "Backing up fork-specific file: $file"
               mkdir -p "/tmp/fork-specific/$(dirname "$file")"
               cp "$file" "/tmp/fork-specific/$file"
+              HAS_FILES=true
             fi
-          done
+          done < "$MANIFEST"
 
-          # Check if any fork-specific files were found
-          if [ -d "/tmp/fork-specific/.github" ]; then
+          if [ "$HAS_FILES" = "true" ]; then
             echo "has_fork_files=true" >> $GITHUB_OUTPUT
             echo "Found fork-specific files:"
             find /tmp/fork-specific -type f
@@ -251,13 +259,13 @@ jobs:
       - name: Restore fork-specific files
         if: steps.check_sync.outputs.needs_sync == 'true' && steps.backup.outputs.has_fork_files == 'true'
         run: |
-          # Copy fork-specific files back
+          # Restore all backed-up fork-specific files
           if [ -d "/tmp/fork-specific" ]; then
-            cp -r /tmp/fork-specific/.github ./ 2>/dev/null || true
+            cp -r /tmp/fork-specific/. ./
           fi
 
-          # Add all fork-specific changes
-          git add .github/
+          # Stage all restored files
+          git add -A
 
       - name: Commit fork-specific changes
         if: steps.check_sync.outputs.needs_sync == 'true' && steps.backup.outputs.has_fork_files == 'true'