import csv
import pandas as pd
import io
from collections import Counter, defaultdict
import logging
import sys

# Configure logging to stdout only
logging.basicConfig(
    level=logging.INFO,
    format='%(asctime)s - %(name)s - %(levelname)s - %(message)s',
    handlers=[
        logging.StreamHandler(sys.stdout)
    ]
)
logger = logging.getLogger(__name__)

def analyze_niches(input_data) -> tuple:
    """
    Analyze and rank niches in a categorized CSV file
    
    Args:
        input_data: File-like object containing the CSV data with categorized titles
        
    Returns:
        tuple: (niche_df, subniche_df) DataFrames with analysis results
    """
    logger.info("Starting niche analysis...")
    
    # Read CSV file into DataFrame
    df = pd.read_csv(input_data)
    
    # Ensure required columns exist
    required_columns = ["Title", "Niche", "Subniches"]
    for col in required_columns:
        if col not in df.columns:
            error_msg = f"Column '{col}' not found in CSV file!"
            logger.error(error_msg)
            raise ValueError(error_msg)
    
    # Count occurrences of each niche
    niche_counter = Counter(df["Niche"])
    
    # Create dictionary to store detailed info about each niche
    niche_details = defaultdict(lambda: {"count": 0, "subniches": Counter(), "titles": []})
    
    # Process each row in the DataFrame
    for _, row in df.iterrows():
        niche = row["Niche"]
        title = row["Title"]
        
        # Get subniches (may be comma-separated)
        subniches = [s.strip() for s in str(row["Subniches"]).split(",")]
        
        # Update detailed information
        niche_details[niche]["count"] += 1
        niche_details[niche]["titles"].append(title)
        for subniche in subniches:
            if subniche:  # Skip empty subniches
                niche_details[niche]["subniches"][subniche] += 1
    
    # Sort niches by occurrence count (descending)
    sorted_niches = sorted(niche_counter.items(), key=lambda x: x[1], reverse=True)
    
    # Print results
    logger.info("\n=== NICHE RANKING BY FREQUENCY ===")
    logger.info(f"{'Rank':<5}{'Niche':<25}{'Count':<10}{'Top Subniches':<50}")
    logger.info("-" * 90)
    
    for rank, (niche, count) in enumerate(sorted_niches, 1):
        # Get top 3 subniches
        top_subniches = niche_details[niche]["subniches"].most_common(3)
        top_subniches_str = ", ".join([f"{s[0]} ({s[1]})" for s in top_subniches])
        
        logger.info(f"{rank:<5}{niche:<25}{count:<10}{top_subniches_str:<50}")
    
    # Create niche ranking DataFrame
    niche_data = []
    for rank, (niche, count) in enumerate(sorted_niches, 1):
        top_subniches = niche_details[niche]["subniches"].most_common(5)
        top_subniches_str = ", ".join([f"{s[0]} ({s[1]})" for s in top_subniches])
        
        example_titles = niche_details[niche]["titles"][:3]
        example_titles_str = "; ".join(example_titles)
        
        niche_data.append({
            "Rank": rank,
            "Niche": niche,
            "Count": count,
            "Top Subniches": top_subniches_str,
            "Example Titles": example_titles_str
        })
    
    niche_df = pd.DataFrame(niche_data)
    logger.info("Niche ranking analysis complete")
    
    # Analyze subniches in more detail
    logger.info("\n=== DETAILED SUBNICHE ANALYSIS ===")
    all_subniches = Counter()
    for niche, details in niche_details.items():
        for subniche, count in details["subniches"].items():
            all_subniches[subniche] += count
    
    logger.info(f"Total unique subniches: {len(all_subniches)}")
    logger.info("Top 10 most common subniches:")
    for subniche, count in all_subniches.most_common(10):
        logger.info(f"  - {subniche}: {count}")
    
    # Create subniche details DataFrame
    subniche_data = []
    for subniche, count in all_subniches.most_common():
        # Find niches related to this subniche
        related_niches = []
        for niche, details in niche_details.items():
            if subniche in details["subniches"]:
                related_niches.append(f"{niche} ({details['subniches'][subniche]})")
        
        subniche_data.append({
            "Subniche": subniche,
            "Count": count,
            "Related Niches": ", ".join(related_niches)
        })
    
    subniche_df = pd.DataFrame(subniche_data)
    logger.info("Subniche analysis complete")
    
    return niche_df, subniche_df