base64_str = "iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNk+M9QDwADhgGAWjR9awAAAABJRU5ErkJggg=="

import hashlib

def md5_from_base64(base64_str):
    clean_str = base64_str.strip().replace('\n', '').replace(' ', '')
    return hashlib.md5(clean_str.encode()).hexdigest()

md5 = md5_from_base64(base64_str)
print(md5)  # a1b2c3d4e5f6...

import hashlib
import base64

def md5_from_binary(base64_str):
    # 1. base64 → 二進(jìn)制
    image_data = base64.b64decode(base64_str)
    # 2. 二進(jìn)制 → MD5
    return hashlib.md5(image_data).hexdigest()

md5 = md5_from_binary(base64_str)
print(md5)

import hashlib

def md5_from_file(file_path):
    md5 = hashlib.md5()
    with open(file_path, 'rb') as f:
        for chunk in iter(lambda: f.read(8192), b''):
            md5.update(chunk)
    return md5.hexdigest()

md5 = md5_from_file('photo.jpg')
print(md5)

data:image/png;base64,iVBORw0KGgoAAAANSUhEUg...

def extract_base64(s):
    if ',' in s:
        return s.split(',', 1)[1]  # 只分割一次
    return s

clean_b64 = extract_base64(base64_str)
md5 = md5_from_binary(clean_b64)

# 來源A（干凈）
"iVBORw0KGgoAAAANSUhEUg..."

# 來源B（帶換行）
"iVBORw0KGgoAAAANSUhEUg...
AAAAAAAAAAAAAAAAA..."

# pip install imagehash
import imagehash
from PIL import Image
import io
import base64

def phash_from_base64(base64_str):
    image_data = base64.b64decode(base64_str)
    img = Image.open(io.BytesIO(image_data))
    return str(imagehash.phash(img))  # 返回感知哈希值

# 即使壓縮過，pHash 也可能相同！

import hashlib
import base64
import json

class ImageDeduplicator:
    """圖片去重管理器"""
    
    def __init__(self):
        self.seen = {}  # {md5: count} 記錄見過的圖片
    
    def get_md5(self, base64_str):
        """計算圖片MD5（推薦方式）"""
        # 去除前綴
        if ',' in base64_str:
            base64_str = base64_str.split(',', 1)[1]
        # base64 → 二進(jìn)制 → MD5
        image_data = base64.b64decode(base64_str)
        return hashlib.md5(image_data).hexdigest()
    
    def check(self, base64_str, label=""):
        """
        檢查圖片是否重復(fù)
        返回: (是否重復(fù), md5值)
        """
        md5 = self.get_md5(base64_str)
        
        if md5 in self.seen:
            self.seen[md5] += 1
            return True, md5, self.seen[md5]
        else:
            self.seen[md5] = 1
            return False, md5, 1
    
    def save(self, filepath='seen_images.json'):
        """保存去重記錄"""
        with open(filepath, 'w') as f:
            json.dump(self.seen, f, indent=2)
    
    def load(self, filepath='seen_images.json'):
        """加載去重記錄"""
        try:
            with open(filepath, 'r') as f:
                self.seen = json.load(f)
        except FileNotFoundError:
            self.seen = {}


# ==================== 使用示例 ====================

dedup = ImageDeduplicator()

# 模擬3張圖片（第1張和第3張是同一張）
images = [
    ("data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNk+M9QDwADhgGAWjR9awAAAABJRU5ErkJggg==", "圖片A"),
    ("data:image/png;base64,AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA", "圖片B"),
    ("data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNk+M9QDwADhgGAWjR9awAAAABJRU5ErkJggg==", "圖片A（重復(fù)）"),
]

for b64, name in images:
    is_dup, md5, count = dedup.check(b64, name)
    status = "?? 重復(fù)" if is_dup else "? 新圖片"
    print(f"{name}: {status} | MD5: {md5} | 已出現(xiàn){count}次")

# 保存記錄
dedup.save()