Files
weixin-holiday-message/scroll_full_contacts.py

168 lines
4.3 KiB
Python
Raw Normal View History

# -*- coding: utf-8 -*-
"""
完整抓取微信通讯录 - 从头开始滚动到底
"""
import uiautomation as auto
import time
import os
from PIL import Image
def find_wechat_window():
"""查找微信窗口"""
wechat_window = auto.WindowControl(searchDepth=1, Name='微信')
if wechat_window.Exists(3, 1):
return wechat_window
return None
def capture_contact_region(wechat_window, index, save_dir):
"""截取通讯录区域"""
rect = wechat_window.BoundingRectangle
x_offset = 70
y_offset = 130
width = 280
height = rect.height() - 160
screenshot_path = os.path.join(save_dir, f"region_{index:03d}.png")
try:
bitmap = wechat_window.ToBitmap(x=x_offset, y=y_offset, width=width, height=height)
bitmap.ToFile(screenshot_path)
return screenshot_path
except Exception as e:
print(f"截图失败: {e}")
return None
def get_image_hash(image_path):
"""计算图片哈希"""
try:
img = Image.open(image_path)
img = img.resize((16, 16), Image.Resampling.LANCZOS)
img = img.convert('L')
pixels = list(img.get_flattened_data())
avg = sum(pixels) / len(pixels)
return ''.join(['1' if p > avg else '0' for p in pixels])
except:
return None
def images_similarity(hash1, hash2):
"""计算相似度"""
if not hash1 or not hash2:
return 0
diff = sum(c1 != c2 for c1, c2 in zip(hash1, hash2))
return 1 - diff / len(hash1)
def scroll_to_top(wechat_window):
"""滚动到顶部"""
print("滚动到顶部...")
rect = wechat_window.BoundingRectangle
center_x = rect.left + 200
center_y = rect.top + 400
# 点击获取焦点
auto.SetCursorPos(center_x, center_y)
auto.Click(center_x, center_y)
time.sleep(0.3)
# 多次向上滚动确保到顶部
for _ in range(20):
auto.SendKeys('{Home}')
time.sleep(0.2)
# 额外滚动几次PageUp
for _ in range(5):
auto.SendKeys('{PageUp}')
time.sleep(0.3)
print("已滚动到顶部")
time.sleep(1)
def scroll_down(wechat_window):
"""向下滚动一页"""
try:
rect = wechat_window.BoundingRectangle
center_x = rect.left + 200
center_y = rect.top + 400
auto.SetCursorPos(center_x, center_y)
auto.Click(center_x, center_y)
time.sleep(0.2)
auto.WheelDown(wheelTimes=3)
time.sleep(0.4)
return True
except:
return False
def main():
print("=" * 60)
print("完整抓取微信通讯录")
print("=" * 60)
# 创建新的截图目录
save_dir = r"D:\夏骥\微信研究\scroll_full"
os.makedirs(save_dir, exist_ok=True)
# 清理旧截图
for f in os.listdir(save_dir):
if f.endswith('.png'):
os.remove(os.path.join(save_dir, f))
# 查找微信窗口
print("\n查找微信窗口...")
wechat_window = find_wechat_window()
if not wechat_window:
print("未找到微信窗口!")
return
print(f"找到微信窗口: {wechat_window.Name}")
# 滚动到顶部
print("\n请确保微信通讯录界面已打开...")
time.sleep(2)
scroll_to_top(wechat_window)
# 开始截图
print("\n开始滚动截图...")
screenshots = []
last_hash = None
no_change = 0
max_screenshots = 150
for i in range(max_screenshots):
path = capture_contact_region(wechat_window, i, save_dir)
if path:
screenshots.append(path)
print(f" 截图 {i+1}")
# 检测是否到底
current_hash = get_image_hash(path)
if last_hash:
sim = images_similarity(last_hash, current_hash)
if sim > 0.95:
no_change += 1
if no_change >= 2:
print(f"\n检测到到底,共截图 {len(screenshots)}")
break
else:
no_change = 0
last_hash = current_hash
scroll_down(wechat_window)
print(f"\n完成!共截图 {len(screenshots)}")
print(f"截图保存目录: {save_dir}")
# 提示运行OCR
print("\n接下来请运行 batch_ocr_full.py 进行OCR识别")
if __name__ == '__main__':
main()