#!/usr/bin/env python3
"""
Check which German translation files are actually translated vs. just stubs or copies.

This script compares English and German versions of project posts to identify:
1. Missing German files
2. Identical content (likely untranslated)
3. Stub files with minimal content
"""

import os
from pathlib import Path
import re

# Change to project root
os.chdir(Path(__file__).parent.parent)

def extract_content(filepath):
    """Extract the main content (excluding frontmatter) from a markdown file."""
    with open(filepath, 'r', encoding='utf-8') as f:
        content = f.read()

    # Split frontmatter from content
    parts = content.split('+++')
    if len(parts) >= 3:
        # Return content after second +++
        return parts[2].strip()
    return content.strip()

def extract_title_from_frontmatter(filepath):
    """Extract title from frontmatter."""
    with open(filepath, 'r', encoding='utf-8') as f:
        content = f.read()

    match = re.search(r'title\s*=\s*"([^"]+)"', content)
    if match:
        return match.group(1)
    return None

def similarity_ratio(text1, text2):
    """Calculate a simple similarity ratio between two texts."""
    if not text1 or not text2:
        return 0.0

    # Normalize whitespace
    text1_norm = ' '.join(text1.split())
    text2_norm = ' '.join(text2.split())

    if text1_norm == text2_norm:
        return 1.0

    # Simple character-based similarity
    longer = max(len(text1_norm), len(text2_norm))
    if longer == 0:
        return 1.0

    # Count matching characters
    matches = sum(c1 == c2 for c1, c2 in zip(text1_norm, text2_norm))
    return matches / longer

def main():
    project_dir = Path("content/project")

    missing_german = []
    untranslated = []
    stub_files = []
    properly_translated = []

    # Find all English index.md files
    for en_file in sorted(project_dir.glob("*/index.md")):
        project_folder = en_file.parent
        de_file = project_folder / "index.de.md"

        project_name = project_folder.name

        # Check if German file exists
        if not de_file.exists():
            missing_german.append(project_name)
            continue

        # Extract content
        en_content = extract_content(en_file)
        de_content = extract_content(de_file)
        en_title = extract_title_from_frontmatter(en_file)
        de_title = extract_title_from_frontmatter(de_file)

        # Check if content is identical or very similar
        similarity = similarity_ratio(en_content, de_content)

        # Check if German file is a stub (very short content)
        de_word_count = len(de_content.split())

        if similarity > 0.95:
            untranslated.append({
                'name': project_name,
                'similarity': similarity,
                'en_title': en_title,
                'de_title': de_title
            })
        elif de_word_count < 20:
            stub_files.append({
                'name': project_name,
                'word_count': de_word_count,
                'en_title': en_title,
                'de_title': de_title
            })
        else:
            properly_translated.append({
                'name': project_name,
                'similarity': similarity,
                'word_count': de_word_count
            })

    # Print results
    print("=" * 80)
    print("GERMAN TRANSLATION STATUS REPORT")
    print("=" * 80)
    print()

    print(f"📊 SUMMARY")
    print(f"  Total projects: {len(list(project_dir.glob('*/index.md')))}")
    print(f"  ✅ Properly translated: {len(properly_translated)}")
    print(f"  ❌ Missing German file: {len(missing_german)}")
    print(f"  ⚠️  Untranslated (identical content): {len(untranslated)}")
    print(f"  ⚠️  Stub files (< 20 words): {len(stub_files)}")
    print()

    if missing_german:
        print("=" * 80)
        print("❌ MISSING GERMAN FILES")
        print("=" * 80)
        for project in missing_german:
            print(f"  • {project}")
        print()

    if untranslated:
        print("=" * 80)
        print("⚠️  UNTRANSLATED (Identical or near-identical to English)")
        print("=" * 80)
        for item in untranslated:
            print(f"  • {item['name']}")
            print(f"    Similarity: {item['similarity']:.1%}")
            print(f"    EN title: {item['en_title']}")
            print(f"    DE title: {item['de_title']}")
            print()

    if stub_files:
        print("=" * 80)
        print("⚠️  STUB FILES (Less than 20 words)")
        print("=" * 80)
        for item in stub_files:
            print(f"  • {item['name']}")
            print(f"    Word count: {item['word_count']}")
            print(f"    EN title: {item['en_title']}")
            print(f"    DE title: {item['de_title']}")
            print()

    print("=" * 80)
    print(f"Total needing translation: {len(missing_german) + len(untranslated) + len(stub_files)}")
    print("=" * 80)

if __name__ == "__main__":
    main()