import subprocess
import re
from collections import defaultdict

class PRAnalyzer:
    def analyze_changes(self, base_branch='main'):
        """
        Analyze changes between current branch and base
        """
        analysis = {
            'files_changed': self._get_changed_files(base_branch),
            'change_statistics': self._get_change_stats(base_branch),
            'change_categories': self._categorize_changes(base_branch),
            'potential_impacts': self._assess_impacts(base_branch),
            'dependencies_affected': self._check_dependencies(base_branch)
        }
        
        return analysis
    
    def _get_changed_files(self, base_branch):
        """Get list of changed files with statistics"""
        cmd = f"git diff --name-status {base_branch}...HEAD"
        result = subprocess.run(cmd.split(), capture_output=True, text=True)
        
        files = []
        for line in result.stdout.strip().split('\n'):
            if line:
                status, filename = line.split('\t', 1)
                files.append({
                    'filename': filename,
                    'status': self._parse_status(status),
                    'category': self._categorize_file(filename)
                })
        
        return files
    
    def _get_change_stats(self, base_branch):
        """Get detailed change statistics"""
        cmd = f"git diff --shortstat {base_branch}...HEAD"
        result = subprocess.run(cmd.split(), capture_output=True, text=True)
        
        # Parse output like: "10 files changed, 450 insertions(+), 123 deletions(-)"
        stats_pattern = r'(\d+) files? changed(?:, (\d+) insertions?\(\+\))?(?:, (\d+) deletions?\(-\))?'
        match = re.search(stats_pattern, result.stdout)
        
        if match:
            files, insertions, deletions = match.groups()
            return {
                'files_changed': int(files),
                'insertions': int(insertions or 0),
                'deletions': int(deletions or 0),
                'net_change': int(insertions or 0) - int(deletions or 0)
            }
        
        return {'files_changed': 0, 'insertions': 0, 'deletions': 0, 'net_change': 0}
    
    def _categorize_file(self, filename):
        """Categorize file by type"""
        categories = {
            'source': ['.js', '.ts', '.py', '.java', '.go', '.rs'],
            'test': ['test', 'spec', '.test.', '.spec.'],
            'config': ['config', '.json', '.yml', '.yaml', '.toml'],
            'docs': ['.md', 'README', 'CHANGELOG', '.rst'],
            'styles': ['.css', '.scss', '.less'],
            'build': ['Makefile', 'Dockerfile', '.gradle', 'pom.xml']
        }
        
        for category, patterns in categories.items():
            if any(pattern in filename for pattern in patterns):
                return category
        
        return 'other'

def generate_pr_description(analysis, commits):
    """
    Generate detailed PR description from analysis
    """
    description = f"""
## Summary

{generate_summary(analysis, commits)}

## What Changed

{generate_change_list(analysis)}

## Why These Changes

{extract_why_from_commits(commits)}

## Type of Change

{determine_change_types(analysis)}

## How Has This Been Tested?

{generate_test_section(analysis)}

## Visual Changes

{generate_visual_section(analysis)}

## Performance Impact

{analyze_performance_impact(analysis)}

## Breaking Changes

{identify_breaking_changes(analysis)}

## Dependencies

{list_dependency_changes(analysis)}

## Checklist

{generate_review_checklist(analysis)}

## Additional Notes

{generate_additional_notes(analysis)}
"""
    return description

def generate_summary(analysis, commits):
    """Generate executive summary"""
    stats = analysis['change_statistics']
    
    # Extract main purpose from commits
    main_purpose = extract_main_purpose(commits)
    
    summary = f"""
This PR {main_purpose}.

**Impact**: {stats['files_changed']} files changed ({stats['insertions']} additions, {stats['deletions']} deletions)
**Risk Level**: {calculate_risk_level(analysis)}
**Review Time**: ~{estimate_revie