gitflow-analytics 1.0.1__py3-none-any.whl → 1.3.6__py3-none-any.whl

gitflow_analytics/reports/classification_writer.py

@@ -0,0 +1,924 @@
+"""Classification report generator for GitFlow Analytics.
+
+This module provides comprehensive reporting capabilities for commit classification
+results, including aggregate statistics, developer breakdowns, confidence analysis,
+and temporal patterns. Designed to integrate with existing GitFlow Analytics
+reporting infrastructure.
+"""
+
+import csv
+import json
+import logging
+from collections import Counter, defaultdict
+from datetime import datetime
+from pathlib import Path
+from typing import Any, Dict, List, Optional, Tuple
+
+logger = logging.getLogger(__name__)
+
+
+class ClassificationReportGenerator:
+    """Generator for comprehensive commit classification reports.
+    
+    This class creates detailed reports from commit classification results,
+    providing insights into development patterns, team productivity, and
+    code quality metrics through the lens of commit categorization.
+    
+    Key capabilities:
+    - Aggregate classification statistics
+    - Per-developer activity breakdowns
+    - Per-repository analysis
+    - Confidence score analysis
+    - Temporal pattern identification
+    - Export to multiple formats (CSV, JSON, Markdown)
+    """
+    
+    def __init__(self, output_directory: Path, config: Optional[Dict[str, Any]] = None):
+        """Initialize the classification report generator.
+        
+        Args:
+            output_directory: Directory where reports will be saved
+            config: Optional configuration for report generation
+        """
+        self.output_directory = Path(output_directory)
+        self.output_directory.mkdir(parents=True, exist_ok=True)
+        
+        self.config = config or {}
+        self.include_low_confidence = self.config.get('include_low_confidence', True)
+        self.confidence_threshold = self.config.get('confidence_threshold', 0.6)
+        self.min_commits_for_analysis = self.config.get('min_commits_for_analysis', 5)
+        
+        # Report metadata
+        self.generated_at = datetime.now()
+        self.reports_generated = []
+        
+        logger.info(f"Classification report generator initialized - output: {self.output_directory}")
+    
+    def generate_comprehensive_report(self, classified_commits: List[Dict[str, Any]], 
+                                    metadata: Optional[Dict[str, Any]] = None) -> Dict[str, str]:
+        """Generate all available classification reports.
+        
+        Args:
+            classified_commits: List of commits with classification results
+            metadata: Optional metadata about the analysis (date range, repos, etc.)
+            
+        Returns:
+            Dictionary mapping report types to file paths
+        """
+        if not classified_commits:
+            logger.warning("No classified commits provided - skipping report generation")
+            return {}
+        
+        # Filter classified commits
+        classified_only = [c for c in classified_commits if 'predicted_class' in c]
+        
+        if not classified_only:
+            logger.warning("No commits with classification results found")
+            return {}
+        
+        logger.info(f"Generating comprehensive classification reports for {len(classified_only)} commits")
+        
+        report_paths = {}
+        
+        try:
+            # Generate individual reports
+            report_paths['summary'] = self.generate_summary_report(classified_only, metadata)
+            report_paths['detailed_csv'] = self.generate_detailed_csv_report(classified_only, metadata)
+            report_paths['developer_breakdown'] = self.generate_developer_breakdown_report(classified_only, metadata)
+            report_paths['repository_analysis'] = self.generate_repository_analysis_report(classified_only, metadata)
+            report_paths['confidence_analysis'] = self.generate_confidence_analysis_report(classified_only, metadata)
+            report_paths['temporal_patterns'] = self.generate_temporal_patterns_report(classified_only, metadata)
+            report_paths['classification_matrix'] = self.generate_classification_matrix_report(classified_only, metadata)
+            report_paths['executive_summary'] = self.generate_executive_summary_report(classified_only, metadata)
+            
+            # Generate comprehensive JSON export
+            report_paths['comprehensive_json'] = self.generate_json_export(classified_only, metadata)
+            
+            # Generate markdown summary
+            report_paths['markdown_summary'] = self.generate_markdown_summary(classified_only, metadata)
+            
+            self.reports_generated = list(report_paths.keys())
+            logger.info(f"Generated {len(report_paths)} classification reports")
+            
+            return report_paths
+            
+        except Exception as e:
+            logger.error(f"Failed to generate comprehensive reports: {e}")
+            return report_paths
+    
+    def generate_summary_report(self, classified_commits: List[Dict[str, Any]], 
+                              metadata: Optional[Dict[str, Any]] = None) -> str:
+        """Generate high-level summary report.
+        
+        Args:
+            classified_commits: List of classified commits
+            metadata: Optional analysis metadata
+            
+        Returns:
+            Path to generated summary CSV file
+        """
+        output_path = self.output_directory / f'classification_summary_{self._get_timestamp()}.csv'
+        
+        # Calculate summary statistics
+        total_commits = len(classified_commits)
+        classification_counts = Counter(c['predicted_class'] for c in classified_commits)
+        confidence_scores = [c.get('classification_confidence', 0) for c in classified_commits]
+        
+        high_confidence_count = sum(1 for score in confidence_scores if score >= self.confidence_threshold)
+        avg_confidence = sum(confidence_scores) / len(confidence_scores) if confidence_scores else 0
+        
+        unique_developers = len(set(c.get('canonical_author_name', c.get('author_name', 'unknown')) 
+                                  for c in classified_commits))
+        unique_repositories = len(set(c.get('repository', 'unknown') for c in classified_commits))
+        
+        with open(output_path, 'w', newline='', encoding='utf-8') as f:
+            writer = csv.writer(f)
+            
+            # Header information
+            writer.writerow(['Classification Analysis Summary'])
+            writer.writerow(['Generated:', self.generated_at.isoformat()])
+            
+            if metadata:
+                writer.writerow(['Analysis Period:', f"{metadata.get('start_date', 'N/A')} to {metadata.get('end_date', 'N/A')}"])
+                writer.writerow(['Configuration:', metadata.get('config_path', 'N/A')])
+            
+            writer.writerow([])
+            
+            # Overall statistics
+            writer.writerow(['Overall Statistics'])
+            writer.writerow(['Metric', 'Value'])
+            writer.writerow(['Total Commits Analyzed', total_commits])
+            writer.writerow(['Unique Developers', unique_developers])
+            writer.writerow(['Unique Repositories', unique_repositories])
+            writer.writerow(['Average Confidence Score', f'{avg_confidence:.3f}'])
+            writer.writerow(['High Confidence Predictions', f'{high_confidence_count} ({(high_confidence_count/total_commits)*100:.1f}%)'])
+            writer.writerow([])
+            
+            # Classification distribution
+            writer.writerow(['Classification Distribution'])
+            writer.writerow(['Classification Type', 'Count', 'Percentage'])
+            
+            for class_type, count in classification_counts.most_common():
+                percentage = (count / total_commits) * 100
+                writer.writerow([class_type, count, f'{percentage:.1f}%'])
+        
+        logger.info(f"Summary report generated: {output_path}")
+        return str(output_path)
+    
+    def generate_detailed_csv_report(self, classified_commits: List[Dict[str, Any]], 
+                                   metadata: Optional[Dict[str, Any]] = None) -> str:
+        """Generate detailed CSV report with all commit information.
+        
+        Args:
+            classified_commits: List of classified commits
+            metadata: Optional analysis metadata
+            
+        Returns:
+            Path to generated detailed CSV file
+        """
+        output_path = self.output_directory / f'classification_detailed_{self._get_timestamp()}.csv'
+        
+        with open(output_path, 'w', newline='', encoding='utf-8') as f:
+            writer = csv.writer(f)
+            
+            # Write header
+            headers = [
+                'commit_hash', 'date', 'author', 'canonical_author', 'repository',
+                'predicted_class', 'confidence', 'is_reliable', 'message_preview',
+                'files_changed', 'insertions', 'deletions', 'lines_changed',
+                'primary_language', 'primary_activity', 'is_multilingual',
+                'branch', 'project_key', 'ticket_references'
+            ]
+            writer.writerow(headers)
+            
+            # Write commit details
+            for commit in classified_commits:
+                file_analysis = commit.get('file_analysis_summary', {})
+                
+                row = [
+                    commit.get('hash', '')[:12],  # Shortened hash
+                    commit.get('timestamp', '').strftime('%Y-%m-%d %H:%M:%S') if commit.get('timestamp') else '',
+                    commit.get('author_name', ''),
+                    commit.get('canonical_author_name', commit.get('author_name', '')),
+                    commit.get('repository', ''),
+                    commit.get('predicted_class', ''),
+                    f"{commit.get('classification_confidence', 0):.3f}",
+                    commit.get('is_reliable_prediction', False),
+                    commit.get('message', '')[:100].replace('\n', ' '),  # Preview of message
+                    commit.get('files_changed', 0),
+                    commit.get('insertions', 0),
+                    commit.get('deletions', 0),
+                    commit.get('insertions', 0) + commit.get('deletions', 0),
+                    file_analysis.get('primary_language', ''),
+                    file_analysis.get('primary_activity', ''),
+                    file_analysis.get('is_multilingual', False),
+                    commit.get('branch', ''),
+                    commit.get('project_key', ''),
+                    len(commit.get('ticket_references', []))
+                ]
+                writer.writerow(row)
+        
+        logger.info(f"Detailed CSV report generated: {output_path}")
+        return str(output_path)
+    
+    def generate_developer_breakdown_report(self, classified_commits: List[Dict[str, Any]], 
+                                          metadata: Optional[Dict[str, Any]] = None) -> str:
+        """Generate per-developer classification breakdown.
+        
+        Args:
+            classified_commits: List of classified commits
+            metadata: Optional analysis metadata
+            
+        Returns:
+            Path to generated developer breakdown CSV file
+        """
+        output_path = self.output_directory / f'classification_by_developer_{self._get_timestamp()}.csv'
+        
+        # Aggregate developer statistics
+        developer_stats = defaultdict(lambda: {
+            'total_commits': 0,
+            'classifications': Counter(),
+            'confidence_scores': [],
+            'repositories': set(),
+            'total_lines_changed': 0,
+            'avg_files_per_commit': 0,
+            'commit_dates': []
+        })
+        
+        for commit in classified_commits:
+            author = commit.get('canonical_author_name', commit.get('author_name', 'unknown'))
+            stats = developer_stats[author]
+            
+            stats['total_commits'] += 1
+            stats['classifications'][commit.get('predicted_class', 'unknown')] += 1
+            
+            if 'classification_confidence' in commit:
+                stats['confidence_scores'].append(commit['classification_confidence'])
+            
+            stats['repositories'].add(commit.get('repository', 'unknown'))
+            stats['total_lines_changed'] += commit.get('insertions', 0) + commit.get('deletions', 0)
+            stats['avg_files_per_commit'] += commit.get('files_changed', 0)
+            
+            if commit.get('timestamp'):
+                stats['commit_dates'].append(commit['timestamp'])
+        
+        # Calculate derived metrics
+        for author, stats in developer_stats.items():
+            if stats['total_commits'] > 0:
+                stats['avg_confidence'] = sum(stats['confidence_scores']) / len(stats['confidence_scores']) if stats['confidence_scores'] else 0
+                stats['avg_files_per_commit'] = stats['avg_files_per_commit'] / stats['total_commits']
+                stats['avg_lines_per_commit'] = stats['total_lines_changed'] / stats['total_commits']
+                stats['primary_classification'] = stats['classifications'].most_common(1)[0][0] if stats['classifications'] else 'unknown'
+                stats['classification_diversity'] = len(stats['classifications'])
+                stats['repository_count'] = len(stats['repositories'])
+                
+                # Calculate activity span
+                if stats['commit_dates']:
+                    date_range = max(stats['commit_dates']) - min(stats['commit_dates'])
+                    stats['activity_span_days'] = date_range.days
+                else:
+                    stats['activity_span_days'] = 0
+        
+        # Filter developers with minimum commits
+        filtered_developers = {k: v for k, v in developer_stats.items() 
+                             if v['total_commits'] >= self.min_commits_for_analysis}
+        
+        with open(output_path, 'w', newline='', encoding='utf-8') as f:
+            writer = csv.writer(f)
+            
+            # Write summary section
+            writer.writerow(['Developer Classification Analysis'])
+            writer.writerow(['Total Developers:', len(developer_stats)])
+            writer.writerow(['Developers with ≥{} commits:'.format(self.min_commits_for_analysis), len(filtered_developers)])
+            writer.writerow([])
+            
+            # Write detailed breakdown
+            headers = [
+                'developer', 'total_commits', 'primary_classification', 'classification_diversity',
+                'avg_confidence', 'high_confidence_ratio', 'repository_count', 'repositories',
+                'avg_files_per_commit', 'avg_lines_per_commit', 'activity_span_days'
+            ]
+            
+            # Add classification type columns
+            all_classifications = set()
+            for stats in filtered_developers.values():
+                all_classifications.update(stats['classifications'].keys())
+            
+            classification_headers = [f'{cls}_count' for cls in sorted(all_classifications)]
+            headers.extend(classification_headers)
+            
+            writer.writerow(headers)
+            
+            # Sort developers by total commits (descending)
+            sorted_developers = sorted(filtered_developers.items(), 
+                                     key=lambda x: x[1]['total_commits'], reverse=True)
+            
+            for author, stats in sorted_developers:
+                high_confidence_count = sum(1 for score in stats['confidence_scores'] 
+                                          if score >= self.confidence_threshold)
+                high_confidence_ratio = high_confidence_count / len(stats['confidence_scores']) if stats['confidence_scores'] else 0
+                
+                row = [
+                    author,
+                    stats['total_commits'],
+                    stats['primary_classification'],
+                    stats['classification_diversity'],
+                    f"{stats['avg_confidence']:.3f}",
+                    f"{high_confidence_ratio:.3f}",
+                    stats['repository_count'],
+                    '; '.join(sorted(stats['repositories'])),
+                    f"{stats['avg_files_per_commit']:.1f}",
+                    f"{stats['avg_lines_per_commit']:.0f}",
+                    stats['activity_span_days']
+                ]
+                
+                # Add classification counts
+                for cls in sorted(all_classifications):
+                    row.append(stats['classifications'].get(cls, 0))
+                
+                writer.writerow(row)
+        
+        logger.info(f"Developer breakdown report generated: {output_path}")
+        return str(output_path)
+    
+    def generate_repository_analysis_report(self, classified_commits: List[Dict[str, Any]], 
+                                          metadata: Optional[Dict[str, Any]] = None) -> str:
+        """Generate per-repository classification analysis.
+        
+        Args:
+            classified_commits: List of classified commits
+            metadata: Optional analysis metadata
+            
+        Returns:
+            Path to generated repository analysis CSV file
+        """
+        output_path = self.output_directory / f'classification_by_repository_{self._get_timestamp()}.csv'
+        
+        # Aggregate repository statistics
+        repo_stats = defaultdict(lambda: {
+            'total_commits': 0,
+            'classifications': Counter(),
+            'developers': set(),
+            'confidence_scores': [],
+            'total_lines_changed': 0,
+            'languages': Counter(),
+            'activities': Counter()
+        })
+        
+        for commit in classified_commits:
+            repo = commit.get('repository', 'unknown')
+            stats = repo_stats[repo]
+            
+            stats['total_commits'] += 1
+            stats['classifications'][commit.get('predicted_class', 'unknown')] += 1
+            stats['developers'].add(commit.get('canonical_author_name', commit.get('author_name', 'unknown')))
+            
+            if 'classification_confidence' in commit:
+                stats['confidence_scores'].append(commit['classification_confidence'])
+            
+            stats['total_lines_changed'] += commit.get('insertions', 0) + commit.get('deletions', 0)
+            
+            # File analysis information
+            file_analysis = commit.get('file_analysis_summary', {})
+            if file_analysis.get('primary_language'):
+                stats['languages'][file_analysis['primary_language']] += 1
+            if file_analysis.get('primary_activity'):
+                stats['activities'][file_analysis['primary_activity']] += 1
+        
+        with open(output_path, 'w', newline='', encoding='utf-8') as f:
+            writer = csv.writer(f)
+            
+            writer.writerow(['Repository Classification Analysis'])
+            writer.writerow(['Total Repositories:', len(repo_stats)])
+            writer.writerow([])
+            
+            headers = [
+                'repository', 'total_commits', 'developer_count', 'primary_classification',
+                'avg_confidence', 'avg_lines_per_commit', 'primary_language', 'primary_activity',
+                'classification_diversity', 'language_diversity'
+            ]
+            writer.writerow(headers)
+            
+            # Sort repositories by commit count
+            sorted_repos = sorted(repo_stats.items(), key=lambda x: x[1]['total_commits'], reverse=True)
+            
+            for repo, stats in sorted_repos:
+                avg_confidence = sum(stats['confidence_scores']) / len(stats['confidence_scores']) if stats['confidence_scores'] else 0
+                avg_lines = stats['total_lines_changed'] / stats['total_commits'] if stats['total_commits'] > 0 else 0
+                
+                primary_class = stats['classifications'].most_common(1)[0][0] if stats['classifications'] else 'unknown'
+                primary_lang = stats['languages'].most_common(1)[0][0] if stats['languages'] else 'unknown'
+                primary_activity = stats['activities'].most_common(1)[0][0] if stats['activities'] else 'unknown'
+                
+                row = [
+                    repo,
+                    stats['total_commits'],
+                    len(stats['developers']),
+                    primary_class,
+                    f"{avg_confidence:.3f}",
+                    f"{avg_lines:.0f}",
+                    primary_lang,
+                    primary_activity,
+                    len(stats['classifications']),
+                    len(stats['languages'])
+                ]
+                writer.writerow(row)
+        
+        logger.info(f"Repository analysis report generated: {output_path}")
+        return str(output_path)
+    
+    def generate_confidence_analysis_report(self, classified_commits: List[Dict[str, Any]], 
+                                          metadata: Optional[Dict[str, Any]] = None) -> str:
+        """Generate confidence score analysis report.
+        
+        Args:
+            classified_commits: List of classified commits
+            metadata: Optional analysis metadata
+            
+        Returns:
+            Path to generated confidence analysis CSV file
+        """
+        output_path = self.output_directory / f'classification_confidence_analysis_{self._get_timestamp()}.csv'
+        
+        confidence_scores = [c.get('classification_confidence', 0) for c in classified_commits]
+        
+        # Calculate confidence statistics by classification type
+        confidence_by_class = defaultdict(list)
+        for commit in classified_commits:
+            class_type = commit.get('predicted_class', 'unknown')
+            confidence = commit.get('classification_confidence', 0)
+            confidence_by_class[class_type].append(confidence)
+        
+        with open(output_path, 'w', newline='', encoding='utf-8') as f:
+            writer = csv.writer(f)
+            
+            writer.writerow(['Classification Confidence Analysis'])
+            writer.writerow([])
+            
+            # Overall confidence statistics
+            if confidence_scores:
+                writer.writerow(['Overall Confidence Statistics'])
+                writer.writerow(['Metric', 'Value'])
+                writer.writerow(['Total Predictions', len(confidence_scores)])
+                writer.writerow(['Average Confidence', f"{sum(confidence_scores) / len(confidence_scores):.3f}"])
+                writer.writerow(['Minimum Confidence', f"{min(confidence_scores):.3f}"])
+                writer.writerow(['Maximum Confidence', f"{max(confidence_scores):.3f}"])
+                
+                # Confidence distribution
+                very_high = sum(1 for s in confidence_scores if s >= 0.9)
+                high = sum(1 for s in confidence_scores if 0.8 <= s < 0.9)
+                medium = sum(1 for s in confidence_scores if 0.6 <= s < 0.8)
+                low = sum(1 for s in confidence_scores if 0.4 <= s < 0.6)
+                very_low = sum(1 for s in confidence_scores if s < 0.4)
+                
+                writer.writerow(['Very High (≥0.9)', f"{very_high} ({(very_high/len(confidence_scores))*100:.1f}%)"])
+                writer.writerow(['High (0.8-0.9)', f"{high} ({(high/len(confidence_scores))*100:.1f}%)"])
+                writer.writerow(['Medium (0.6-0.8)', f"{medium} ({(medium/len(confidence_scores))*100:.1f}%)"])
+                writer.writerow(['Low (0.4-0.6)', f"{low} ({(low/len(confidence_scores))*100:.1f}%)"])
+                writer.writerow(['Very Low (<0.4)', f"{very_low} ({(very_low/len(confidence_scores))*100:.1f}%)"])
+                writer.writerow([])
+            
+            # Confidence by classification type
+            writer.writerow(['Confidence by Classification Type'])
+            writer.writerow(['Classification', 'Count', 'Avg Confidence', 'Min', 'Max', 'High Confidence Count'])
+            
+            for class_type, scores in sorted(confidence_by_class.items()):
+                if scores:
+                    avg_conf = sum(scores) / len(scores)
+                    high_conf_count = sum(1 for s in scores if s >= self.confidence_threshold)
+                    
+                    writer.writerow([
+                        class_type,
+                        len(scores),
+                        f"{avg_conf:.3f}",
+                        f"{min(scores):.3f}",
+                        f"{max(scores):.3f}",
+                        f"{high_conf_count} ({(high_conf_count/len(scores))*100:.1f}%)"
+                    ])
+        
+        logger.info(f"Confidence analysis report generated: {output_path}")
+        return str(output_path)
+    
+    def generate_temporal_patterns_report(self, classified_commits: List[Dict[str, Any]], 
+                                        metadata: Optional[Dict[str, Any]] = None) -> str:
+        """Generate temporal patterns analysis report.
+        
+        Args:
+            classified_commits: List of classified commits
+            metadata: Optional analysis metadata
+            
+        Returns:
+            Path to generated temporal patterns CSV file
+        """
+        output_path = self.output_directory / f'classification_temporal_patterns_{self._get_timestamp()}.csv'
+        
+        # Group commits by date
+        daily_stats = defaultdict(lambda: {
+            'total_commits': 0,
+            'classifications': Counter(),
+            'developers': set(),
+            'confidence_scores': []
+        })
+        
+        for commit in classified_commits:
+            if commit.get('timestamp'):
+                date_key = commit['timestamp'].date()
+                stats = daily_stats[date_key]
+                
+                stats['total_commits'] += 1
+                stats['classifications'][commit.get('predicted_class', 'unknown')] += 1
+                stats['developers'].add(commit.get('canonical_author_name', commit.get('author_name', 'unknown')))
+                
+                if 'classification_confidence' in commit:
+                    stats['confidence_scores'].append(commit['classification_confidence'])
+        
+        with open(output_path, 'w', newline='', encoding='utf-8') as f:
+            writer = csv.writer(f)
+            
+            writer.writerow(['Temporal Classification Patterns'])
+            writer.writerow([])
+            
+            # Get all classification types for column headers
+            all_classifications = set()
+            for stats in daily_stats.values():
+                all_classifications.update(stats['classifications'].keys())
+            
+            headers = ['date', 'total_commits', 'developer_count', 'avg_confidence']
+            headers.extend([f'{cls}_count' for cls in sorted(all_classifications)])
+            writer.writerow(headers)
+            
+            # Sort by date
+            for date, stats in sorted(daily_stats.items()):
+                avg_confidence = sum(stats['confidence_scores']) / len(stats['confidence_scores']) if stats['confidence_scores'] else 0
+                
+                row = [
+                    date.isoformat(),
+                    stats['total_commits'],
+                    len(stats['developers']),
+                    f"{avg_confidence:.3f}"
+                ]
+                
+                # Add classification counts for this date
+                for cls in sorted(all_classifications):
+                    row.append(stats['classifications'].get(cls, 0))
+                
+                writer.writerow(row)
+        
+        logger.info(f"Temporal patterns report generated: {output_path}")
+        return str(output_path)
+    
+    def generate_classification_matrix_report(self, classified_commits: List[Dict[str, Any]], 
+                                            metadata: Optional[Dict[str, Any]] = None) -> str:
+        """Generate classification distribution matrix report.
+        
+        Args:
+            classified_commits: List of classified commits
+            metadata: Optional analysis metadata
+            
+        Returns:
+            Path to generated classification matrix CSV file
+        """
+        output_path = self.output_directory / f'classification_matrix_{self._get_timestamp()}.csv'
+        
+        # Create cross-tabulation of classifications vs other dimensions
+        class_counts = Counter(c.get('predicted_class', 'unknown') for c in classified_commits)
+        
+        # Developer vs Classification matrix
+        dev_class_matrix = defaultdict(Counter)
+        repo_class_matrix = defaultdict(Counter)
+        lang_class_matrix = defaultdict(Counter)
+        
+        for commit in classified_commits:
+            class_type = commit.get('predicted_class', 'unknown')
+            developer = commit.get('canonical_author_name', commit.get('author_name', 'unknown'))
+            repository = commit.get('repository', 'unknown')
+            
+            dev_class_matrix[developer][class_type] += 1
+            repo_class_matrix[repository][class_type] += 1
+            
+            # Language information
+            file_analysis = commit.get('file_analysis_summary', {})
+            language = file_analysis.get('primary_language', 'unknown')
+            lang_class_matrix[language][class_type] += 1
+        
+        with open(output_path, 'w', newline='', encoding='utf-8') as f:
+            writer = csv.writer(f)
+            
+            writer.writerow(['Classification Distribution Matrix'])
+            writer.writerow([])
+            
+            # Overall classification distribution
+            writer.writerow(['Overall Classification Distribution'])
+            writer.writerow(['Classification', 'Count', 'Percentage'])
+            total_commits = len(classified_commits)
+            
+            for class_type, count in class_counts.most_common():
+                percentage = (count / total_commits) * 100
+                writer.writerow([class_type, count, f'{percentage:.1f}%'])
+            
+            writer.writerow([])
+            
+            # Top developers by classification diversity
+            writer.writerow(['Top Developers by Classification Diversity'])
+            writer.writerow(['Developer', 'Total Commits', 'Classifications Used', 'Primary Classification'])
+            
+            dev_diversity = []
+            for dev, classifications in dev_class_matrix.items():
+                total_dev_commits = sum(classifications.values())
+                if total_dev_commits >= self.min_commits_for_analysis:
+                    diversity = len(classifications)
+                    primary = classifications.most_common(1)[0][0]
+                    dev_diversity.append((dev, total_dev_commits, diversity, primary))
+            
+            # Sort by diversity, then by total commits
+            for dev, total, diversity, primary in sorted(dev_diversity, key=lambda x: (x[2], x[1]), reverse=True)[:10]:
+                writer.writerow([dev, total, diversity, primary])
+            
+            writer.writerow([])
+            
+            # Language vs Classification matrix
+            writer.writerow(['Language vs Classification Matrix'])
+            all_classes = sorted(class_counts.keys())
+            header = ['Language'] + all_classes + ['Total']
+            writer.writerow(header)
+            
+            for language, classifications in sorted(lang_class_matrix.items(), 
+                                                  key=lambda x: sum(x[1].values()), reverse=True):
+                row = [language]
+                total_lang_commits = sum(classifications.values())
+                
+                for class_type in all_classes:
+                    count = classifications.get(class_type, 0)
+                    percentage = (count / total_lang_commits) * 100 if total_lang_commits > 0 else 0
+                    row.append(f"{count} ({percentage:.1f}%)")
+                
+                row.append(total_lang_commits)
+                writer.writerow(row)
+        
+        logger.info(f"Classification matrix report generated: {output_path}")
+        return str(output_path)
+    
+    def generate_executive_summary_report(self, classified_commits: List[Dict[str, Any]], 
+                                        metadata: Optional[Dict[str, Any]] = None) -> str:
+        """Generate executive summary report for leadership.
+        
+        Args:
+            classified_commits: List of classified commits
+            metadata: Optional analysis metadata
+            
+        Returns:
+            Path to generated executive summary CSV file
+        """
+        output_path = self.output_directory / f'classification_executive_summary_{self._get_timestamp()}.csv'
+        
+        # Calculate key metrics
+        total_commits = len(classified_commits)
+        unique_developers = len(set(c.get('canonical_author_name', c.get('author_name', 'unknown')) 
+                                  for c in classified_commits))
+        unique_repositories = len(set(c.get('repository', 'unknown') for c in classified_commits))
+        
+        classification_counts = Counter(c.get('predicted_class', 'unknown') for c in classified_commits)
+        confidence_scores = [c.get('classification_confidence', 0) for c in classified_commits]
+        
+        # Productivity metrics
+        total_lines_changed = sum(c.get('insertions', 0) + c.get('deletions', 0) for c in classified_commits)
+        avg_lines_per_commit = total_lines_changed / total_commits if total_commits > 0 else 0
+        
+        # Time span analysis
+        commit_dates = [c['timestamp'] for c in classified_commits if c.get('timestamp')]
+        if commit_dates:
+            analysis_span = (max(commit_dates) - min(commit_dates)).days
+        else:
+            analysis_span = 0
+        
+        with open(output_path, 'w', newline='', encoding='utf-8') as f:
+            writer = csv.writer(f)
+            
+            writer.writerow(['Executive Summary - Commit Classification Analysis'])
+            writer.writerow(['Generated:', self.generated_at.strftime('%Y-%m-%d %H:%M:%S')])
+            
+            if metadata:
+                writer.writerow(['Analysis Period:', f"{metadata.get('start_date', 'N/A')} to {metadata.get('end_date', 'N/A')}"])
+            
+            writer.writerow([])
+            
+            # Key metrics
+            writer.writerow(['KEY METRICS'])
+            writer.writerow(['Total Development Activity', f'{total_commits:,} commits'])
+            writer.writerow(['Team Size', f'{unique_developers} active developers'])
+            writer.writerow(['Codebase Scope', f'{unique_repositories} repositories'])
+            writer.writerow(['Analysis Timespan', f'{analysis_span} days'])
+            writer.writerow(['Average Code Changes per Commit', f'{avg_lines_per_commit:.0f} lines'])
+            
+            if confidence_scores:
+                avg_confidence = sum(confidence_scores) / len(confidence_scores)
+                high_confidence_pct = (sum(1 for s in confidence_scores if s >= self.confidence_threshold) / len(confidence_scores)) * 100
+                writer.writerow(['Classification Confidence', f'{avg_confidence:.1%} average'])
+                writer.writerow(['High Confidence Predictions', f'{high_confidence_pct:.1f}%'])
+            
+            writer.writerow([])
+            
+            # Development focus areas
+            writer.writerow(['DEVELOPMENT FOCUS AREAS'])
+            writer.writerow(['Activity Type', 'Commits', '% of Total', 'Strategic Insight'])
+            
+            # Define strategic insights for each classification type
+            strategic_insights = {
+                'feature': 'New capability development',
+                'bugfix': 'Quality maintenance and stability',
+                'refactor': 'Technical debt management',
+                'docs': 'Knowledge management and documentation',
+                'test': 'Quality assurance and testing',
+                'config': 'Infrastructure and configuration',
+                'chore': 'Maintenance and operational tasks',
+                'security': 'Security and compliance',
+                'hotfix': 'Critical issue resolution',
+                'style': 'Code quality and standards',
+                'build': 'Build system and deployment',
+                'ci': 'Automation and continuous integration'
+            }
+            
+            for class_type, count in classification_counts.most_common():
+                percentage = (count / total_commits) * 100
+                insight = strategic_insights.get(class_type, 'Unclassified development activity')
+                writer.writerow([class_type.title(), f'{count:,}', f'{percentage:.1f}%', insight])
+            
+            writer.writerow([])
+            
+            # Recommendations
+            writer.writerow(['STRATEGIC RECOMMENDATIONS'])
+            
+            # Generate recommendations based on the data
+            recommendations = []
+            
+            # Feature vs maintenance balance
+            feature_pct = (classification_counts.get('feature', 0) / total_commits) * 100
+            maintenance_pct = ((classification_counts.get('bugfix', 0) + 
+                              classification_counts.get('refactor', 0) + 
+                              classification_counts.get('chore', 0)) / total_commits) * 100
+            
+            if feature_pct > 60:
+                recommendations.append("High feature development velocity - consider increasing quality assurance")
+            elif feature_pct < 20:
+                recommendations.append("Low feature development - may indicate focus on maintenance or technical debt")
+            
+            if maintenance_pct > 40:
+                recommendations.append("High maintenance overhead - consider technical debt reduction initiatives")
+            
+            # Documentation analysis
+            docs_pct = (classification_counts.get('docs', 0) / total_commits) * 100
+            if docs_pct < 5:
+                recommendations.append("Low documentation activity - consider improving documentation practices")
+            
+            # Testing analysis
+            test_pct = (classification_counts.get('test', 0) / total_commits) * 100
+            if test_pct < 10:
+                recommendations.append("Limited testing activity - consider strengthening testing practices")
+            
+            # Security analysis
+            security_pct = (classification_counts.get('security', 0) / total_commits) * 100
+            if security_pct > 0:
+                recommendations.append(f"Active security focus ({security_pct:.1f}% of commits) - positive security posture")
+            
+            # Confidence analysis
+            if confidence_scores:
+                low_confidence_pct = (sum(1 for s in confidence_scores if s < 0.6) / len(confidence_scores)) * 100
+                if low_confidence_pct > 20:
+                    recommendations.append("Consider improving commit message clarity for better classification")
+            
+            for i, recommendation in enumerate(recommendations, 1):
+                writer.writerow([f'Recommendation {i}', recommendation])
+        
+        logger.info(f"Executive summary report generated: {output_path}")
+        return str(output_path)
+    
+    def generate_json_export(self, classified_commits: List[Dict[str, Any]], 
+                           metadata: Optional[Dict[str, Any]] = None) -> str:
+        """Generate comprehensive JSON export of all classification data.
+        
+        Args:
+            classified_commits: List of classified commits
+            metadata: Optional analysis metadata
+            
+        Returns:
+            Path to generated JSON file
+        """
+        output_path = self.output_directory / f'classification_comprehensive_{self._get_timestamp()}.json'
+        
+        # Create comprehensive data structure
+        export_data = {
+            'metadata': {
+                'generated_at': self.generated_at.isoformat(),
+                'total_commits': len(classified_commits),
+                'generator_version': '1.0',
+                'config': self.config
+            },
+            'summary_statistics': self._calculate_summary_statistics(classified_commits),
+            'commits': classified_commits,
+            'analysis_metadata': metadata or {}
+        }
+        
+        with open(output_path, 'w', encoding='utf-8') as f:
+            json.dump(export_data, f, indent=2, default=str, ensure_ascii=False)
+        
+        logger.info(f"JSON export generated: {output_path}")
+        return str(output_path)
+    
+    def generate_markdown_summary(self, classified_commits: List[Dict[str, Any]], 
+                                metadata: Optional[Dict[str, Any]] = None) -> str:
+        """Generate markdown summary report.
+        
+        Args:
+            classified_commits: List of classified commits
+            metadata: Optional analysis metadata
+            
+        Returns:
+            Path to generated markdown file
+        """
+        output_path = self.output_directory / f'classification_summary_{self._get_timestamp()}.md'
+        
+        # Calculate statistics
+        total_commits = len(classified_commits)
+        classification_counts = Counter(c.get('predicted_class', 'unknown') for c in classified_commits)
+        confidence_scores = [c.get('classification_confidence', 0) for c in classified_commits]
+        
+        with open(output_path, 'w', encoding='utf-8') as f:
+            f.write("# Commit Classification Analysis Report\n\n")
+            f.write(f"**Generated:** {self.generated_at.strftime('%Y-%m-%d %H:%M:%S')}\n\n")
+            
+            if metadata:
+                f.write(f"**Analysis Period:** {metadata.get('start_date', 'N/A')} to {metadata.get('end_date', 'N/A')}\n\n")
+            
+            f.write("## Summary Statistics\n\n")
+            f.write(f"- **Total Commits Analyzed:** {total_commits:,}\n")
+            f.write(f"- **Unique Developers:** {len(set(c.get('canonical_author_name', c.get('author_name', 'unknown')) for c in classified_commits))}\n")
+            f.write(f"- **Unique Repositories:** {len(set(c.get('repository', 'unknown') for c in classified_commits))}\n")
+            
+            if confidence_scores:
+                avg_confidence = sum(confidence_scores) / len(confidence_scores)
+                high_confidence_count = sum(1 for s in confidence_scores if s >= self.confidence_threshold)
+                f.write(f"- **Average Confidence:** {avg_confidence:.1%}\n")
+                f.write(f"- **High Confidence Predictions:** {high_confidence_count:,} ({(high_confidence_count/total_commits)*100:.1f}%)\n")
+            
+            f.write("\n## Classification Distribution\n\n")
+            f.write("| Classification Type | Count | Percentage |\n")
+            f.write("|-------------------|--------|------------|\n")
+            
+            for class_type, count in classification_counts.most_common():
+                percentage = (count / total_commits) * 100
+                f.write(f"| {class_type.title()} | {count:,} | {percentage:.1f}% |\n")
+            
+            f.write(f"\n## Analysis Details\n\n")
+            f.write(f"This report was generated using GitFlow Analytics commit classification system.\n")
+            f.write(f"Classification confidence threshold: {self.confidence_threshold}\n\n")
+            
+            f.write("For detailed analysis, see the accompanying CSV reports:\n")
+            for report_type in self.reports_generated:
+                f.write(f"- {report_type.replace('_', ' ').title()}\n")
+        
+        logger.info(f"Markdown summary generated: {output_path}")
+        return str(output_path)
+    
+    def _calculate_summary_statistics(self, classified_commits: List[Dict[str, Any]]) -> Dict[str, Any]:
+        """Calculate comprehensive summary statistics.
+        
+        Args:
+            classified_commits: List of classified commits
+            
+        Returns:
+            Dictionary containing summary statistics
+        """
+        total_commits = len(classified_commits)
+        
+        classification_counts = Counter(c.get('predicted_class', 'unknown') for c in classified_commits)
+        confidence_scores = [c.get('classification_confidence', 0) for c in classified_commits]
+        
+        developers = set(c.get('canonical_author_name', c.get('author_name', 'unknown')) for c in classified_commits)
+        repositories = set(c.get('repository', 'unknown') for c in classified_commits)
+        
+        return {
+            'total_commits': total_commits,
+            'unique_developers': len(developers),
+            'unique_repositories': len(repositories),
+            'classification_distribution': dict(classification_counts),
+            'confidence_statistics': {
+                'average': sum(confidence_scores) / len(confidence_scores) if confidence_scores else 0,
+                'minimum': min(confidence_scores) if confidence_scores else 0,
+                'maximum': max(confidence_scores) if confidence_scores else 0,
+                'high_confidence_count': sum(1 for s in confidence_scores if s >= self.confidence_threshold),
+                'high_confidence_percentage': (sum(1 for s in confidence_scores if s >= self.confidence_threshold) / len(confidence_scores)) * 100 if confidence_scores else 0
+            },
+            'productivity_metrics': {
+                'total_lines_changed': sum(c.get('insertions', 0) + c.get('deletions', 0) for c in classified_commits),
+                'average_lines_per_commit': sum(c.get('insertions', 0) + c.get('deletions', 0) for c in classified_commits) / total_commits if total_commits > 0 else 0,
+                'average_files_per_commit': sum(c.get('files_changed', 0) for c in classified_commits) / total_commits if total_commits > 0 else 0
+            }
+        }
+    
+    def _get_timestamp(self) -> str:
+        """Get timestamp string for file naming.
+        
+        Returns:
+            Timestamp string in YYYYMMDD_HHMMSS format
+        """
+        return self.generated_at.strftime('%Y%m%d_%H%M%S')