import os
import re
import shutil
from collections import defaultdict

topics_dir = "/Volumes/Data/project/Antigravity/Wiki/10_Wiki/Topics"
files = [f for f in os.listdir(topics_dir) if f.endswith(".md") and os.path.isfile(os.path.join(topics_dir, f))]

# Define categories and their matching keywords/tags
categories = {
    "Project_Logs": [r'^2024-', r'^2025-', r'^2026-'],
    "AI_and_ML": ['ai', 'machine-learning', 'deep-learning', 'llm', 'gpt', 'dall-e', 'midjourney', 'generative', 'reinforcement', 'neural', 'diffusion', 'nlp'],
    "Architecture": ['architecture', 'ddd', 'pattern', 'solid', 'microservices', 'c4', 'system-design', 'infrastructure', 'clean-architecture', 'domain-driven', 'cqrs', 'mvc'],
    "Frontend": ['frontend', 'react', 'css', 'dom', 'webgl', 'threejs', 'browser', 'javascript', 'typescript', 'ui-components', 'rendering'],
    "Backend": ['backend', 'database', 'sql', 'api', 'graphql', 'rest', 'nodejs', 'server', 'redis', 'cache'],
    "DevOps_and_Security": ['devops', 'ci-cd', 'security', 'sast', 'dast', 'git', 'docker', 'kubernetes', 'cloud', 'testing', 'deployment'],
    "Game_Design": ['game', 'game-design', 'combat', 'monetization', 'level-design', 'player', 'boss', 'skybound', 'mechanics', 'gacha'],
    "Design_and_UX": ['ux', 'ui', 'design-system', 'accessibility', 'figma', 'typography', 'user-experience'],
    "Business_and_Management": ['management', 'agile', 'scrum', 'business', 'economics', 'consulting', 'kpi', 'strategy', 'monopoly'],
    "Computer_Science_and_Theory": ['algorithm', 'data-structure', 'theory', 'math', 'physics', 'computational', 'graph', 'tree', 'complexity'],
    "Other": []
}

file_categories = defaultdict(list)

def determine_category(filename, content):
    filename_lower = filename.lower()
    
    # 1. Project logs by filename
    for p in categories["Project_Logs"]:
        if re.search(p, filename_lower):
            return "Project_Logs"

    # Extract tags from yaml
    tags = []
    yaml_match = re.search(r'^---\n(.*?)\n---\n', content, re.DOTALL)
    if yaml_match:
        yaml_text = yaml_match.group(1)
        tags_match = re.search(r'tags:\s*\[(.*?)\]', yaml_text)
        if tags_match:
            tags_str = tags_match.group(1)
            tags = [t.strip().strip('"\'').lower() for t in tags_str.split(',')]
            
    # Remove generic tags
    generic_tags = ['auto-wikified', 'technical-documentation', 'auto-consolidated']
    tags = [t for t in tags if t not in generic_tags]
    
    # Check tags first
    for cat, keywords in categories.items():
        if cat in ["Project_Logs", "Other"]:
            continue
        for tag in tags:
            if any(k in tag for k in keywords):
                return cat
                
    # Check filename and content summary
    # To optimize, we just check filename and a small chunk of content
    text_to_search = filename_lower + " " + content[:1000].lower()
    for cat, keywords in categories.items():
        if cat in ["Project_Logs", "Other"]:
            continue
        for k in keywords:
            if k in text_to_search:
                return cat
                
    return "Other"

# Group files
for f in files:
    filepath = os.path.join(topics_dir, f)
    with open(filepath, 'r', encoding='utf-8') as file:
        content = file.read()
        
    cat = determine_category(f, content)
    file_categories[cat].append((f, content))

# Create folders and move files
for cat, flist in file_categories.items():
    if not flist:
        continue
        
    cat_dir = os.path.join(topics_dir, cat)
    if not os.path.exists(cat_dir):
        os.makedirs(cat_dir)
        
    # List for the index file
    index_links = []
    
    for filename, content in flist:
        # Move file
        src = os.path.join(topics_dir, filename)
        dst = os.path.join(cat_dir, filename)
        shutil.move(src, dst)
        
        # update metadata in the file to point to the new category if needed
        # Actually the user only asked to organize into folders.
        
        # Add to index list
        # Extract title
        title_match = re.search(r'^#\s+(.*)', content, re.MULTILINE)
        title = title_match.group(1) if title_match else filename[:-3]
        index_links.append(f"- [[{filename[:-3]}]] : {title}")
        
    # Create the category summary file (e.g. Architecture.md inside Architecture folder)
    index_links.sort()
    index_content = f"""---
category: Unified
tags: [category-index, {cat.lower()}]
title: {cat.replace('_', ' ')} Directory
last_updated: 2026-05-02
---

# {cat.replace('_', ' ')} Directory

이 문서는 `{cat}` 카테고리에 속한 모든 지식 문서들의 목록을 제공합니다.

## 📄 문서 목록
""" + "\n".join(index_links)

    index_filename = f"{cat}.md"
    with open(os.path.join(cat_dir, index_filename), 'w', encoding='utf-8') as index_file:
        index_file.write(index_content)
        
    print(f"Created category '{cat}' with {len(flist)} files.")

print("Categorization complete.")