novafarma/scripts/utils/glavna_referenca_cleanup.py

#!/usr/bin/env python3
"""
GLAVNA REFERENCA CLEANUP
1. Organizacija v pod-mape
2. Preimenovanje v kratka imena
3. Odstranjevanje duplikatov
4. Vizualna analiza kvalitete
"""

import os
import hashlib
from pathlib import Path
from collections import defaultdict
import json

REFERENCA_DIR = Path("/Users/davidkotnik/repos/novafarma/assets/slike/glavna_referenca")

def get_file_hash(filepath):
    """Izračuna MD5 hash datoteke"""
    hash_md5 = hashlib.md5()
    try:
        with open(filepath, "rb") as f:
            for chunk in iter(lambda: f.read(4096), b""):
                hash_md5.update(chunk)
        return hash_md5.hexdigest()
    except:
        return None

def find_duplicates():
    """Najde vse duplikate po hash"""
    print("🔍 IŠČEM DUPLIKATE...\n")
    
    hash_map = defaultdict(list)
    all_files = []
    
    # Zberi vse PNG/JPG
    for ext in ['*.png', '*.jpg', '*.jpeg']:
        all_files.extend(REFERENCA_DIR.rglob(ext))
    
    print(f"Najdenih {len(all_files)} slik...")
    
    # Izračunaj hashe
    for i, filepath in enumerate(all_files):
        if i % 100 == 0:
            print(f"  Obdelanih: {i}/{len(all_files)}")
        
        file_hash = get_file_hash(filepath)
        if file_hash:
            hash_map[file_hash].append(filepath)
    
    # Najdi duplikate
    duplicates = {hash_val: files for hash_val, files in hash_map.items() if len(files) > 1}
    
    print(f"\n✅ Analiza končana!")
    print(f"📊 DUPLIKATI: {len(duplicates)} skupin")
    
    # Poročilo
    duplicate_report = []
    total_duplicates = 0
    
    for hash_val, files in duplicates.items():
        total_duplicates += len(files) - 1  # -1 ker enega obdržimo
        
        group = {
            'hash': hash_val,
            'count': len(files),
            'size': files[0].stat().st_size,
            'files': [str(f.relative_to(REFERENCA_DIR)) for f in files]
        }
        duplicate_report.append(group)
    
    # Shrani poročilo
    report_file = REFERENCA_DIR / "DUPLICATE_REPORT.json"
    with open(report_file, 'w', encoding='utf-8') as f:
        json.dump({
            'total_groups': len(duplicates),
            'total_duplicates': total_duplicates,
            'space_wasted_mb': sum(g['size'] * (g['count'] - 1) for g in duplicate_report) / 1024 / 1024,
            'groups': duplicate_report
        }, f, indent=2, ensure_ascii=False)
    
    print(f"\n📄 Poročilo shranjeno: DUPLICATE_REPORT.json")
    print(f"🗑️  Lahko zbrišeš: {total_duplicates} duplikatov")
    print(f"💾 Prihranila bi: {sum(g['size'] * (g['count'] - 1) for g in duplicate_report) / 1024 / 1024:.1f} MB")
    
    return duplicate_report

def categorize_files():
    """Kategorizira datoteke po vsebini imena"""
    print("\n📂 KATEGORIZACIJA...\n")
    
    categories = {
        'characters': [],
        'npcs': [],
        'crops': [],
        'trees': [],
        'buildings': [],
        'items': [],
        'tools': [],
        'ui': [],
        'enemies': [],
        'animals': [],
        'biomes': [],
        'effects': [],
        'interior': [],
        'other': []
    }
    
    # Ključne besede za kategorije
    keywords = {
        'characters': ['kai', 'ana', 'gronk', 'main_character'],
        'npcs': ['npc', 'priest', 'merchant', 'farmer', 'guard', 'elder', 'innkeeper', 
                 'blacksmith', 'mayor', 'teacher', 'herbalist', 'hunter'],
        'crops': ['crop', 'wheat', 'corn', 'potato', 'tomato', 'carrot', 'cannabis', 
                  'korenje', 'krompir', 'koruza', 'paradiznik', 'konoplja', 'stage', 'growth'],
        'trees': ['tree', 'drevo', 'oak', 'pine', 'jablana', 'visnja', 'hruška'],
        'buildings': ['building', 'house', 'barn', 'church', 'school', 'hospital',
                      'cerkev', 'sola', 'hisa', 'hlev', 'zgradbe'],
        'items': ['item', 'predmet', 'blueprint', 'resource', 'seme', 'seed'],
        'tools': ['tool', 'orodje', 'axe', 'pickaxe', 'hoe', 'sekira', 'kramp', 'motika'],
        'ui': ['ui', 'button', 'gumb', 'icon', 'ikona', 'panel', 'okvir'],
        'enemies': ['enemy', 'zombie', 'zombi', 'boss', 'monster', 'sovraznik'],
        'animals': ['animal', 'cow', 'pig', 'sheep', 'chicken', 'krava', 'prasic'],
        'biomes': ['biome', 'biom', 'grassland', 'forest', 'desert', 'snow'],
        'effects': ['vfx', 'effect', 'sparkle', 'glow', 'particle'],
        'interior': ['interior', 'notranjost', 'furniture', 'pohištvo']
    }
    
    # Zberi vse slike
    all_files = list(REFERENCA_DIR.glob('*.png')) + list(REFERENCA_DIR.glob('*.jpg'))
    
    print(f"Kategoriziram {len(all_files)} slik...")
    
    for filepath in all_files:
        filename_lower = filepath.name.lower()
        
        categorized = False
        for category, words in keywords.items():
            if any(word in filename_lower for word in words):
                categories[category].append(filepath)
                categorized = True
                break
        
        if not categorized:
            categories['other'].append(filepath)
    
    # Poročilo
    print("\n📊 KATEGORIZACIJA:")
    for category, files in sorted(categories.items(), key=lambda x: len(x[1]), reverse=True):
        if files:
            print(f"  {category}: {len(files)} slik")
    
    return categories

def analyze_filenames():
    """Analizira dolžine imen"""
    print("\n📏 ANALIZA DOLŽIN IMEN...\n")
    
    all_files = list(REFERENCA_DIR.glob('*.png')) + list(REFERENCA_DIR.glob('*.jpg'))
    
    lengths = [len(f.stem) for f in all_files]
    avg_length = sum(lengths) / len(lengths) if lengths else 0
    max_length = max(lengths) if lengths else 0
    
    long_names = [f for f in all_files if len(f.stem) > 50]
    
    print(f"Povprečna dolžina imena: {avg_length:.1f} znakov")
    print(f"Najdaljše ime: {max_length} znakov")
    print(f"Imen daljših od 50 znakov: {len(long_names)}")
    
    if long_names:
        print("\n🔴 TOP 10 NAJDALJŠIH IMEN:")
        for f in sorted(long_names, key=lambda x: len(x.stem), reverse=True)[:10]:
            print(f"  {len(f.stem):3d}  {f.name[:80]}...")
    
    return {
        'avg_length': avg_length,
        'max_length': max_length,
        'long_names_count': len(long_names)
    }

if __name__ == "__main__":
    print("=" * 80)
    print(" GLAVNA REFERENCA - CLEANUP ANALIZA")
    print("=" * 80)
    
    # 1. Duplikati
    duplicates = find_duplicates()
    
    # 2. Kategorije
    categories = categorize_files()
    
    # 3. Dolžine imen
    name_stats = analyze_filenames()
    
    print("\n" + "=" * 80)
    print(" ANALIZA KONČANA!")
    print("=" * 80)
    print("\n📋 NASLEDNJI KORAKI:")
    print("   1. Preglej DUPLICATE_REPORT.json")
    print("   2. Izbriši duplikate")
    print("   3. Organiziraj v pod-mape")
    print("   4. Preimenuj v kratka imena")