# Identify which unknown designers to research first based on their item counts

import pandas as pd
import os

def analyze_designer_importance(original_catalog_csv, unknowns_csv):
    """Analyze which designers are most important to research"""
    
    print("\n📊 Designer Priority Analysis")
    print("="*60)
    
    # Read the original catalog with all items
    print("Loading original catalog...")
    catalog_df = pd.read_csv(original_catalog_csv)
    
    # Read the unknowns
    print("Loading unknown designers...")
    unknowns_df = pd.read_csv(unknowns_csv)
    unknown_names = set(unknowns_df['designer_name'].tolist())
    
    # Count items per designer in the original catalog
    designer_counts = catalog_df['designer_name'].value_counts()
    
    # Filter to only unknown designers
    unknown_designer_counts = designer_counts[designer_counts.index.isin(unknown_names)]
    
    # Create priority groups
    high_priority = unknown_designer_counts[unknown_designer_counts >= 3]
    medium_priority = unknown_designer_counts[(unknown_designer_counts >= 2) & (unknown_designer_counts < 3)]
    low_priority = unknown_designer_counts[unknown_designer_counts == 1]
    
    print(f"\n📈 Analysis Results:")
    print(f"Total unknown designers: {len(unknown_names)}")
    print(f"\n🔴 HIGH Priority (3+ items): {len(high_priority)} designers")
    print(f"🟡 MEDIUM Priority (2 items): {len(medium_priority)} designers")
    print(f"🟢 LOW Priority (1 item): {len(low_priority)} designers")
    
    # Show high priority designers
    if len(high_priority) > 0:
        print("\n🔴 HIGH PRIORITY DESIGNERS TO RESEARCH FIRST:")
        print("-"*60)
        for designer, count in high_priority.items():
            print(f"{designer}: {count} items")
            # Show what items they designed
            items = catalog_df[catalog_df['designer_name'] == designer]['items'].tolist()[:3]
            for item in items:
                print(f"  → {item}")
            if count > 3:
                print(f"  → ... and {count-3} more items")
            print()
    
    # Create priority CSV files
    create_priority_files(unknowns_df, high_priority, medium_priority, low_priority)
    
    # Pattern analysis
    analyze_patterns(unknowns_df)
    
    return high_priority, medium_priority, low_priority

def create_priority_files(unknowns_df, high_priority, medium_priority, low_priority):
    """Create separate CSV files for each priority level"""
    
    print("\n💾 Creating priority files...")
    
    # High priority
    high_priority_df = unknowns_df[unknowns_df['designer_name'].isin(high_priority.index)]
    high_priority_df['item_count'] = high_priority_df['designer_name'].map(high_priority)
    high_priority_df = high_priority_df.sort_values('item_count', ascending=False)
    high_priority_df.to_csv('HIGH_PRIORITY_designers.csv', index=False)
    print(f"✅ Created HIGH_PRIORITY_designers.csv ({len(high_priority_df)} designers)")
    
    # Medium priority
    medium_priority_df = unknowns_df[unknowns_df['designer_name'].isin(medium_priority.index)]
    medium_priority_df['item_count'] = medium_priority_df['designer_name'].map(medium_priority)
    medium_priority_df.to_csv('MEDIUM_PRIORITY_designers.csv', index=False)
    print(f"✅ Created MEDIUM_PRIORITY_designers.csv ({len(medium_priority_df)} designers)")
    
    # Low priority
    low_priority_df = unknowns_df[unknowns_df['designer_name'].isin(low_priority.index)]
    low_priority_df['item_count'] = 1
    low_priority_df.to_csv('LOW_PRIORITY_designers.csv', index=False)
    print(f"✅ Created LOW_PRIORITY_designers.csv ({len(low_priority_df)} designers)")

def analyze_patterns(unknowns_df):
    """Analyze patterns in unknown designers"""
    
    print("\n🔍 Pattern Analysis:")
    print("-"*60)
    
    # Organizations vs individuals
    org_keywords = ['Associates', 'Studios', 'Studio', 'Company', 'Workshop', 'Inc', 'Ltd', 'Staff', '&']
    orgs = unknowns_df[unknowns_df['designer_name'].str.contains('|'.join(org_keywords), case=False, na=False)]
    
    print(f"\n🏢 Likely Organizations (not individuals): {len(orgs)}")
    if len(orgs) > 0:
        print("Examples:", orgs['designer_name'].head(5).tolist())
    
    # Quick gender inference
    female_names = ['Mary', 'Helen', 'Dorothy', 'Ruth', 'Betty', 'Patricia', 'Barbara', 
                    'Elizabeth', 'Jennifer', 'Linda', 'Susan', 'Margaret', 'Lisa', 'Nancy',
                    'Karen', 'Emily', 'Sarah', 'Anna', 'Emma', 'Marie', 'Rose', 'Grace']
    
    male_names = ['John', 'Robert', 'James', 'David', 'Michael', 'William', 'Richard', 
                  'Joseph', 'Thomas', 'Charles', 'Christopher', 'Daniel', 'Paul', 'Mark', 
                  'Donald', 'George', 'Kenneth', 'Steven', 'Edward', 'Ronald', 'Carl']
    
    # Check first names
    likely_female = 0
    likely_male = 0
    
    for _, row in unknowns_df.iterrows():
        first_name = row['designer_name'].split()[0] if ' ' in row['designer_name'] else ''
        if first_name in female_names:
            likely_female += 1
        elif first_name in male_names:
            likely_male += 1
    
    print(f"\n👥 Quick gender inference from first names:")
    print(f"  Likely female: {likely_female}")
    print(f"  Likely male: {likely_male}")
    print(f"  Unclear: {len(unknowns_df) - likely_female - likely_male}")
    
    # Nationality patterns
    print(f"\n🌍 Nationality patterns from surnames:")
    patterns = {
        'Scandinavian (-sen, -sson)': r'(?:sen|sson)$',
        'Italian (-ini, -ino, -elli)': r'(?:ini|ino|elli|etti|ucci)$',
        'German (-mann, -stein)': r'(?:mann|stein|berg|feld)$',
        'Dutch (Van, De)': r'^(?:Van |De |van |de )',
        'Irish/Scottish (Mc, Mac, O\')': r'^(?:Mc|Mac|O\')',
    }
    
    for pattern_name, pattern in patterns.items():
        matches = unknowns_df[unknowns_df['designer_name'].str.contains(pattern, regex=True, na=False)]
        if len(matches) > 0:
            print(f"  {pattern_name}: {len(matches)} designers")

def create_smart_suggestions(unknowns_df):
    """Create a CSV with smart suggestions for quick filling"""
    
    suggestions_df = unknowns_df.copy()
    
    # Add suggested gender based on first name
    female_names = set(['Mary', 'Helen', 'Dorothy', 'Ruth', 'Betty', 'Patricia', 'Barbara'])
    male_names = set(['John', 'Robert', 'James', 'David', 'Michael', 'William', 'Richard'])
    
    def suggest_gender(name):
        first_name = name.split()[0] if ' ' in name else ''
        if first_name in female_names:
            return 'Female (suggested)'
        elif first_name in male_names:
            return 'Male (suggested)'
        return 'Unknown'
    
    suggestions_df['suggested_gender'] = suggestions_df['designer_name'].apply(suggest_gender)
    
    # Add suggested country based on surname
    def suggest_country(name):
        if re.search(r'(?:sen|sson)$', name):
            return 'Scandinavia (suggested)'
        elif re.search(r'(?:ini|ino|elli|etti)$', name):
            return 'Italy (suggested)'
        elif name.startswith(('Van ', 'van ', 'De ', 'de ')):
            return 'Netherlands (suggested)'
        return 'Unknown'
    
    suggestions_df['suggested_country'] = suggestions_df['designer_name'].apply(suggest_country)
    
    suggestions_df.to_csv('designers_with_suggestions.csv', index=False)
    print("\n💡 Created designers_with_suggestions.csv with smart suggestions")

if __name__ == "__main__":
    print("Designer Priority Analyzer")
    print("=========================")
    print("This tool helps you identify which unknown designers to research first")
    print("based on how many items they have in your catalog.\n")
    
    # Get file names
    original = input("Enter your ORIGINAL catalog CSV (with all items): ").strip()
    if not original:
        # Try to find it
        files = [f for f in os.listdir('.') if 'complete' in f and f.endswith('.csv')]
        if files:
            original = files[0]
            print(f"Using: {original}")
    
    unknowns = input("Enter your unknowns CSV file: ").strip()
    if not unknowns:
        unknowns = "designers_research_unknown_list_only.csv"
        print(f"Using: {unknowns}")
    
    # Check files exist
    if not os.path.exists(original):
        print(f"❌ Error: {original} not found!")
    elif not os.path.exists(unknowns):
        print(f"❌ Error: {unknowns} not found!")
    else:
        # Run analysis
        high, medium, low = analyze_designer_importance(original, unknowns)
        
        # Ask about smart suggestions
        if input("\n❓ Create smart suggestions for gender/country? (y/n): ").lower() == 'y':
            unknowns_df = pd.read_csv(unknowns)
            create_smart_suggestions(unknowns_df)
        
        print("\n✅ Analysis complete!")
        print("\n📋 Next steps:")
        print("1. Start with HIGH_PRIORITY_designers.csv")
        print("2. These designers have the most impact on your catalog")
        print("3. Research these ~20-30 designers first")
        print("4. The rest can be marked 'Unknown' with less impact")