#!/usr/bin/env python3
import re

def analyze_coverage_by_uncovered():
    files = {}
    current_file = None
    
    with open('coverage/lcov.info', 'r', encoding='utf-8') as f:
        lines = f.readlines()
    
    i = 0
    while i < len(lines):
        line = lines[i].strip()
        if line.startswith('SF:'):
            current_file = line[3:]
            files[current_file] = {'total': 0, 'covered': 0}
            i += 1
        elif line.startswith('LF:') and current_file:
            files[current_file]['total'] = int(line[3:])
            i += 1
        elif line.startswith('LH:') and current_file:
            files[current_file]['covered'] = int(line[3:])
            i += 1
        else:
            i += 1
    
    # Filter out generated files and UI screens/widgets (hard to test)
    filtered = []
    exclude_patterns = ['.g.dart', 'generated', 'l10n.dart', '/pages/', '/widgets/', '_screen.dart', '_widget.dart']
    
    for f, d in files.items():
        if d['total'] > 0:
            # Skip UI files
            should_exclude = any(pattern in f for pattern in exclude_patterns)
            if not should_exclude:
                uncovered = d['total'] - d['covered']
                pct = d['covered']/d['total']*100
                filtered.append((f, d, uncovered, pct))
    
    # Sort by uncovered lines (most uncovered first)
    sorted_files = sorted(filtered, key=lambda x: x[2], reverse=True)
    
    print('Files sorted by uncovered lines (excluding UI screens/widgets):')
    print('=' * 100)
    print(f'{"Uncovered":>10} {"Covered":>8} {"Total":>8} {"%":>7} {"File":<60}')
    print('=' * 100)
    for f, d, uncovered, pct in sorted_files[:30]:
        print(f'{uncovered:>10} {d["covered"]:>8} {d["total"]:>8} {pct:>6.1f}% {f}')

if __name__ == '__main__':
    analyze_coverage_by_uncovered()