fix: handle main/master branch fallback and add branch support in comments

.github/workflows/gw_comment_trigger.yml

@@ -15,6 +15,7 @@ jobs:
     outputs:
       repo_url: ${{ steps.parse.outputs.repo_url }}
       limit: ${{ steps.parse.outputs.limit }}
+      branch: ${{ steps.parse.outputs.branch }}
 
     steps:
       - name: Parse comment
@@ -24,6 +25,7 @@ jobs:
         run: |
           echo "Comment body: $COMMENT_BODY"
 
+          # Extract repository URL
           # Extract repository URL using environment variable
           repo_url=$(echo "$COMMENT_BODY" | grep -oP 'gw --repo \K[^\s]+' || echo "")
           if [ -z "$repo_url" ]; then
@@ -37,7 +39,7 @@ jobs:
             exit 1
           fi
 
-          # Extract limit with improved regex and validation
+          # Extract limit
           limit_raw=$(echo "$COMMENT_BODY" | grep -oP -- '--limit\s+\K\d+' || echo "")
 
           # Validate and set limit (default to 10 if not specified or invalid)
@@ -50,9 +52,18 @@ jobs:
             fi
           fi
 
+          # Extract branch
+          branch_raw=$(echo "$COMMENT_BODY" | grep -oP -- '--branch\s+\K[^\s]+' || echo "")
+          if [ -n "$branch_raw" ]; then
+            branch="$branch_raw"
+          else
+            branch=""
+          fi
+
           echo "repo_url=$repo_url" >> $GITHUB_OUTPUT
           echo "limit=$limit" >> $GITHUB_OUTPUT
-          echo "Parsed repo: $repo_url, limit: $limit"
+          echo "branch=$branch" >> $GITHUB_OUTPUT
+          echo "Parsed repo: $repo_url, limit: $limit, branch: $branch"
 
   run-bugspots:
     needs: parse-comment
@@ -79,15 +90,16 @@ jobs:
         env:
           REPO_URL: ${{ needs.parse-comment.outputs.repo_url }}
           LIMIT: ${{ needs.parse-comment.outputs.limit }}
+          BRANCH: ${{ needs.parse-comment.outputs.branch }}
         run: |
           chmod +x BugPredict/gw.sh
-          # Run the script with environment variables to prevent injection
-          BugPredict/gw.sh "$REPO_URL" "$LIMIT"
+          BugPredict/gw.sh "$REPO_URL" "$LIMIT" "$BRANCH"
 
       - name: Check for repository errors
         id: check-errors
         env:
           REPO_URL: ${{ needs.parse-comment.outputs.repo_url }}
+          BRANCH: ${{ needs.parse-comment.outputs.branch }}
         run: |
           repo_name=$(basename "$REPO_URL" .git)
 
@@ -104,6 +116,8 @@ jobs:
                 echo "error_type=invalid_repo" >> $GITHUB_OUTPUT
               elif [[ "$error_content" == *"No commits found"* ]]; then
                 echo "error_type=no_commits" >> $GITHUB_OUTPUT
+              elif [[ "$error_content" == *"no such branch"* ]]; then
+                echo "error_type=invalid_branch" >> $GITHUB_OUTPUT
               else
                 echo "error_type=unknown" >> $GITHUB_OUTPUT
               fi
@@ -150,6 +164,13 @@ jobs:
               echo "" >> comment.md
               echo "Bugspots analysis requires commit history to analyze bug patterns." >> comment.md
               ;;
+            "invalid_branch")
+              echo "### 🚫 Invalid Branch" >> comment.md
+              echo "" >> comment.md
+              echo "**Error:** The specified branch does not exist in the repository." >> comment.md
+              echo "" >> comment.md
+              echo "Please specify a valid branch or omit the --branch parameter to use the default branch." >> comment.md
+              ;;
             *)
               echo "### ⚠️ Analysis Error" >> comment.md
               echo "" >> comment.md
@@ -170,6 +191,7 @@ jobs:
           echo '```' >> comment.md
           echo "gw --repo https://github.com/username/repository" >> comment.md
           echo "gw --repo https://github.com/username/repository --limit 15" >> comment.md
+          echo "gw --repo https://github.com/username/repository --branch main" >> comment.md
           echo '```' >> comment.md
 
       - name: Prepare success comment

BugPredict/gw.sh

@@ -5,10 +5,11 @@ set -e
 # Parse command line arguments
 REPO_URL="$1"
 LIMIT="${2:-10}"
+BRANCH="$3"
 
 if [ -z "$REPO_URL" ]; then
-  echo "Usage: $0 <repository_url> [limit]"
-  echo "Example: $0 https://github.com/user/repo.git 15"
+  echo "Usage: $0 <repository_url> [limit] [branch]"
+  echo "Example: $0 https://github.com/user/repo.git 15 main"
   exit 1
 fi
 
@@ -19,9 +20,10 @@ OUTPUT_DIR="bugspots-results"
 echo "🚀 Bugspots Comment Analyzer starting at $(date '+%Y-%m-%d %H:%M:%S %Z')"
 echo "Repository: $REPO_URL"
 echo "File limit: $LIMIT"
+[ -n "$BRANCH" ] && echo "Branch: $BRANCH"
 
-# Check if bugspots is installed
-if ! gem list bugspots -i > /dev/null; then
+# Check if bugspots gem is installed
+if ! gem list bugspots -i > /dev/null 2>&1; then
   echo "Installing bugspots gem..."
   gem install bugspots
 fi
@@ -32,32 +34,27 @@ mkdir -p "$OUTPUT_DIR"
 echo "🔄 Cloning $REPO_URL ..."
 repo_name=$(basename "$REPO_URL" .git)
 
-# Try different branch names in order of preference
-branches=("main" "master" "develop" "dev")
-clone_success=false
+# Initialize selected_branch
+selected_branch=""
 
-for branch in "${branches[@]}"; do
-  echo "Trying to clone branch: $branch"
-  if git clone --branch "$branch" --depth 1000 "$REPO_URL" "$WORKDIR/$repo_name" 2>/dev/null; then
-    clone_success=true
-    echo "✅ Successfully cloned branch: $branch"
-    break
+# Clone repository with defaut branch or specified branch
+echo "Cloning repository..."
+if [ -n "$BRANCH" ]; then
+  if ! git clone --branch "$BRANCH" --depth 1000 "$REPO_URL" "$WORKDIR/$repo_name" 2>/dev/null; then
+    echo "❌ Error: Failed to clone specified branch: $BRANCH" >&2
+    echo "Clone failed for $repo_name branch $BRANCH at $(date '+%Y-%m-%d %H:%M:%S %Z')" > "$OUTPUT_DIR/bugspots-${repo_name}.err"
+    exit 1
   fi
-done
-
-# If named branches fail, try default clone
-if [ "$clone_success" = false ]; then
-  echo "Named branches failed, trying default clone..."
-  if git clone --depth 1000 "$REPO_URL" "$WORKDIR/$repo_name"; then
-    clone_success=true
-    echo "✅ Successfully cloned with default branch"
+  selected_branch="$BRANCH"
+  echo "✅ Successfully cloned branch: $BRANCH"
+else
+  if ! git clone --depth 1000 "$REPO_URL" "$WORKDIR/$repo_name" 2>/dev/null; then
+    echo "Error: Failed to clone $REPO_URL" >&2
+    echo "Clone failed for $repo_name at $(date '+%Y-%m-%d %H:%M:%S %Z')" > "$OUTPUT_DIR/bugspots-${repo_name}.err"
+    exit 1
   fi
-fi
-
-if [ "$clone_success" = false ]; then
-  echo "❌ Error: Failed to clone $REPO_URL" >&2
-  echo "Clone failed for $repo_name at $(date '+%Y-%m-%d %H:%M:%S %Z')" > "$OUTPUT_DIR/bugspots-${repo_name}.err"
-  exit 1
+  selected_branch=$(git -C "$WORKDIR/$repo_name" rev-parse --abbrev-ref HEAD)
+  echo "✅ Successfully cloned with default branch: $selected_branch"
 fi
 
 # Verify repository
@@ -84,9 +81,23 @@ total_commits=$(git rev-list --count HEAD 2>/dev/null || echo "unknown")
 echo "📈 Repository has $total_commits commits in current branch"
 
 # Run bugspots with simplified word pattern
-echo "📊 Running Bugspots for $repo_name ..."
-echo "Executing: git bugspots -w fix" >&2
-if ! git bugspots -w fix > "../../$OUTPUT_DIR/bugspots-${repo_name}.log" 2> "../../$OUTPUT_DIR/bugspots-${repo_name}.err"; then
+echo "📊 Running Bugspots for $repo_name on branch $selected_branch..."
+
+# Build the bugspots command
+bugspots_cmd="bugspots ."
+
+# Add branch parameter if specified or detected
+if [ -n "$selected_branch" ]; then
+  bugspots_cmd="$bugspots_cmd --branch $selected_branch"
+fi
+
+# Add regex pattern for bug-fix commits
+bugspots_cmd="$bugspots_cmd --regex 'fix(es|ed)?|close(s|d)?'"
+
+echo "Executing: $bugspots_cmd" >&2
+
+# Execute bugspots command
+if ! eval "$bugspots_cmd" > "../../$OUTPUT_DIR/bugspots-${repo_name}.log" 2> "../../$OUTPUT_DIR/bugspots-${repo_name}.err"; then
   echo "❌ Error: Bugspots failed for $repo_name. Check $OUTPUT_DIR/bugspots-${repo_name}.err" >&2
   if [ -s "../../$OUTPUT_DIR/bugspots-${repo_name}.err" ]; then
     echo "Error details:"
@@ -103,24 +114,33 @@ else
   if [ -s "../../$OUTPUT_DIR/bugspots-${repo_name}.log" ]; then
     if grep -q "Hotspots:" "../../$OUTPUT_DIR/bugspots-${repo_name}.log"; then
       # Extract and count hotspots
-      hotspot_lines=$(sed -n '/Hotspots:/,$p' "../../$OUTPUT_DIR/bugspots-${repo_name}.log" | tail -n +2 | grep -E '^\s*[0-9]+\.[0-9]+.*' | wc -l)
+      hotspot_lines=$(sed -n '/Hotspots:/,$p' "../../$OUTPUT_DIR/bugspots-${repo_name}.log" | tail -n +2 | grep -E '^\s*[0-9]+\.[0-9]+.*' 2>/dev/null | wc -l)
       echo "📋 Found $hotspot_lines hotspot files"
 
+      # Show summary if available - look for bug fix commits count
+      if grep -qE "Found \d+ bugfix commits|Found \d+ fix commits" "../../$OUTPUT_DIR/bugspots-${repo_name}.log"; then
+        bugfix_info=$(grep -oE "Found \d+ (bugfix|fix) commits" "../../$OUTPUT_DIR/bugspots-${repo_name}.log" | head -1)
+        echo "🐛 $bugfix_info"
+      fi
+
       # Extract top N hotspots after "Hotspots:" line
       echo ""
       echo "🎯 Top $LIMIT hotspots found:"
       echo "================================"
       sed -n '/Hotspots:/,$p' "../../$OUTPUT_DIR/bugspots-${repo_name}.log" | \
       tail -n +2 | \
-      grep -E '^\s*[0-9]+\.[0-9]+.*' | \
+      grep -E '^\s*[0-9]+\.[0-9]+.*' 2>/dev/null | \
       head -n "$LIMIT" | \
-      sed 's/^\s*//'
+      sed 's/^\s*//' | \
+      while IFS= read -r line; do
+        echo "  $line"
+      done
       echo "================================"
 
       # Create a clean output file with only the top N hotspots for the GitHub Actions
       sed -n '/Hotspots:/,$p' "../../$OUTPUT_DIR/bugspots-${repo_name}.log" | \
       tail -n +2 | \
-      grep -E '^\s*[0-9]+\.[0-9]+.*' | \
+      grep -E '^\s*[0-9]+\.[0-9]+.*' 2>/dev/null | \
       head -n "$LIMIT" | \
       sed 's/^\s*//' > "../../$OUTPUT_DIR/bugspots-${repo_name}-top.log"
 
@@ -139,4 +159,4 @@ cd - > /dev/null
 # Clean up temporary directories
 rm -rf "$WORKDIR"
 
-echo "🏁 Bugspots Comment Analyzer completed at $(date '+%Y-%m-%d %H:%M:%S %Z')"
+echo "🏁 Bugspots Comment Analyzer completed at $(date '+%Y-%m-%d %H:%M:%S %Z')"