import numpy as np
from sklearn.cluster import KMeans
import colorsys

class ColorAnalyzer:
    def __init__(self):
        self.seasonal_palettes = {
            'spring': ['#FF6B6B', '#FFE66D', '#4ECDC4', '#95E77E'],
            'summer': ['#C7B9FF', '#FF99CC', '#66CCFF', '#99FF99'],
            'fall': ['#8B4513', '#D2691E', '#B8860B', '#CD853F'],
            'winter': ['#1a1a1a', '#333333', '#4d4d4d', '#666666']
        }
    
    def analyze(self, image: np.ndarray) -> dict:
        pixels = image.reshape(-1, 3)
        
        kmeans = KMeans(n_clusters=5, random_state=42, n_init=10)
        kmeans.fit(pixels)
        
        dominant_colors = kmeans.cluster_centers_.astype(int).tolist()
        dominant_colors_hex = [self._rgb_to_hex(color) for color in dominant_colors]
        
        palette = self._generate_palette(dominant_colors_hex)
        
        harmony_score = self._calculate_harmony(dominant_colors_hex)
        
        seasonal_match = self._match_season(dominant_colors_hex)
        
        return {
            "dominant_colors": dominant_colors_hex[:3],
            "palette": palette[:5],
            "harmony_score": harmony_score,
            "seasonal_match": seasonal_match,
            "recommendations": self._get_recommendations(dominant_colors_hex)
        }
    
    def _rgb_to_hex(self, rgb):
        return '#{:02x}{:02x}{:02x}'.format(int(rgb[0]), int(rgb[1]), int(rgb[2]))
    
    def _generate_palette(self, colors):
        palette = colors.copy()
        while len(palette) < 5:
            palette.append('#CCCCCC')
        return palette
    
    def _calculate_harmony(self, colors):
        if len(colors) < 2:
            return 0.7
        
        color_values = []
        for color in colors[:3]:
            r = int(color[1:3], 16) / 255
            g = int(color[3:5], 16) / 255
            b = int(color[5:7], 16) / 255
            h, s, v = colorsys.rgb_to_hsv(r, g, b)
            color_values.append(h)
        
        if len(color_values) >= 2:
            differences = [abs(color_values[i] - color_values[i-1]) for i in range(1, len(color_values))]
            avg_diff = sum(differences) / len(differences)
            harmony = min(1.0, avg_diff * 2)
            return round(harmony, 2)
        
        return 0.7
    
    def _match_season(self, colors):
        if not colors:
            return 'spring'
        
        hex_values = [int(c[1:], 16) for c in colors[:3]]
        avg_hex = sum(hex_values) / len(hex_values)
        
        if avg_hex < 0x666666:
            return 'winter'
        elif avg_hex < 0x999999:
            return 'fall'
        elif avg_hex < 0xCCCCCC:
            return 'spring'
        else:
            return 'summer'
    
    def _get_recommendations(self, colors):
        return [
            "Pair with neutral accessories",
            "Consider the occasion when choosing colors",
            "Mix textures to add depth"
        ]