Merge pull request #83 from Amorter/main

自动验证功能

geetest/www/api.py

@@ -0,0 +1,136 @@
+import requests
+import json
+from typing import Tuple
+import time
+
+class Slide:
+	def register(self) -> Tuple[str, str]:
+		url = "http://127.0.0.1:5000/pc-geetest/register?t=" + str(int(time.time() * 1000))
+		# url = "https://www.geetest.com/demo/gt/register-slide?t=" + str(int(time.time() * 1000))
+		res = requests.get(url=url).json()
+		print(res)
+		return (res['challenge'], res['gt'])
+
+	def get_c_s(self, challenge: str, gt: str, w: str) -> Tuple[list, str]:
+		url = "https://api.geetest.com/get.php"
+		# url = "https://apiv6.geetest.com/get.php"
+		res = requests.get(url=url, params={
+			"gt": gt,
+			"challenge": challenge,
+			"lang": "zh-cn",
+			"pt": 0,
+			"client_type": "web",
+			"w": w,
+		}).text
+		res = json.loads(res.strip('()'))['data']
+		# print(res)
+		return (res['c'], res['s'])
+
+	def get_type(self, challenge: str, gt: str, w: str):
+		url = "http://api.geevisit.com/ajax.php"
+		res = requests.get(url=url, params={
+			"gt": gt,
+			"challenge": challenge,
+			"w": w
+		})
+	def get_new_c_s_challenge_bg_slice(self, challenge: str, gt: str,) -> Tuple[str, str, str, str, str]:
+		url = "http://api.geevisit.com/get.php"
+		res = requests.get(url=url, params={
+			"gt": gt,
+			"challenge": challenge,
+			"is_next": "true",
+			"offline": "false",
+			"type": "slide3",
+			"isPC": "true",
+			"product": "embed",
+			"autoReset": "true",
+			"api_server": "api.geevisit.com",
+			"protocol": "http://",
+			"width": "100%",
+			"callback": "geetest_1715753608245",
+		}).text
+		res = res.strip('geetest_1715753608245')
+		res = res.strip('()')
+		res = json.loads(res)
+		print(res)
+		return (res['c'],res['s'],res['challenge'],
+				'https://' + res['static_servers'][0] + res['bg'],
+				'https://' + res['static_servers'][0] + res['slice'])
+
+	def ajax(self, challenge: str, gt: str, w: str):
+		url = "http://api.geevisit.com/ajax.php"
+		res = requests.get(url=url, params={
+			"gt": gt,
+			"challenge": challenge,
+			"w": w,
+		}).text
+		res = res.strip('()')
+		res = json.loads(res)
+		print(res)
+		return res
+
+class Click:
+	def register(self) -> Tuple[str, str]:
+		url = "https://account.geetest.com/api/captchademo?captcha_type=click"
+		res = requests.get(url=url).json()
+		res = res["data"]
+		print(res)
+		return (res['challenge'], res['gt'])
+
+	def get_c_s(self, challenge: str, gt: str, w: str) -> Tuple[list, str]:
+		url = "https://api.geetest.com/get.php"
+		# url = "https://apiv6.geetest.com/get.php"
+		res = requests.get(url=url, params={
+			"gt": gt,
+			"challenge": challenge,
+			"lang": "zh-cn",
+			"pt": 0,
+			"client_type": "web",
+			"w": w,
+		}).text
+		res = json.loads(res.strip('()'))['data']
+		# print(res)
+		return (res['c'], res['s'])
+
+	def get_type(self, challenge: str, gt: str, w: str):
+		url = "http://api.geevisit.com/ajax.php"
+		res = requests.get(url=url, params={
+			"gt": gt,
+			"challenge": challenge,
+			"w": w
+		})
+	def get_new_c_s_pic(self, challenge: str, gt: str,) -> Tuple[str, str, str]:
+		url = "http://api.geevisit.com/get.php"
+		res = requests.get(url=url, params={
+			"gt": gt,
+			"challenge": challenge,
+			"is_next": "true",
+			"offline": "false",
+			"type": "slide3",
+			"isPC": "true",
+			"product": "embed",
+			"autoReset": "true",
+			"api_server": "api.geevisit.com",
+			"protocol": "http://",
+			"width": "100%",
+			"callback": "geetest_1715753608245",
+		}).text
+		res = res.strip('geetest_1715753608245')
+		res = res.strip('()')
+		res = json.loads(res)
+		res = res["data"]
+		print(res)
+		return (res['c'],res['s'],
+				'https://' + res['static_servers'][0] + res['pic'],)
+
+	def ajax(self, challenge: str, gt: str, w: str):
+		url = "http://api.geevisit.com/ajax.php"
+		res = requests.get(url=url, params={
+			"gt": gt,
+			"challenge": challenge,
+			"w": w,
+		}).text
+		res = res.strip('()')
+		res = json.loads(res)
+		print(res)
+		return res

geetest/www/click3.py

@@ -0,0 +1,68 @@
+# -*- coding: utf-8 -*-
+import io
+from pathlib import Path
+import requests
+from PIL import Image
+import ddddocr
+from typing import Tuple
+
+def slice_img(img) -> Tuple[Image.Image, Image.Image]:
+	if isinstance(img, (str, Path)):
+		_img = Image.open(img)
+	elif isinstance(img, bytes):
+		_img = Image.open(io.BytesIO(img))
+	else:
+		raise ValueError(f'输入图片类型错误, 必须是<type str>/<type Path>/<type bytes>: {type(img)}')
+	background_img = _img.crop((0, 0, 344, 344))
+	text_img = _img.crop((0, 345, 116, 384))
+	return(background_img, text_img)
+
+class Click3:
+	def __init__(self, ocr: ddddocr.DdddOcr = ddddocr.DdddOcr(show_ad=False, beta=True), 
+			  det: ddddocr.DdddOcr = ddddocr.DdddOcr(show_ad=False, det=True)):
+		self.ocr = ocr
+		self.det = det
+
+	def calculated_position(self, img_url):
+		img = requests.get(img_url).content
+
+		(background_img, text_img) = slice_img(img)
+		text = self.ocr.classification(text_img)
+		print(text)
+		background_bytes = io.BytesIO()
+		background_img.save(background_bytes, format='PNG')
+		background_bytes.seek(0)
+		bboxes = self.det.detection(background_bytes.read())
+		texts = list()
+		for bbox in bboxes:
+			(x1, y1, x2, y2) = bbox
+			img = background_img.crop((x1, y1, x2, y2))
+			this_text = self.ocr.classification(img)
+			texts.append(this_text)
+		print(texts)
+		res = list()
+		no_finds = list()
+		is_find = [False] * texts.__len__()
+		for c in text:
+			try:
+				index = texts.index(c)
+			except:
+				no_finds.append(c)
+				continue
+			is_find[index] = True
+			position = str(round((bboxes[index][0] + bboxes[index][2]) / 2 / 333.375 * 100 * 100)) + '_' + str(round((bboxes[index][1] + bboxes[index][3]) / 2 / 333.375 * 100 * 100))
+			res.append(position)
+		for c in no_finds:
+			for i in range(len(is_find)):
+				if not is_find[i]:
+					is_find[i] = True
+					position = str(round((bboxes[i][0] + bboxes[i][2]) / 2 / 333.375 * 100 * 100)) + '_' + str(round((bboxes[i][1] + bboxes[i][3]) / 2 / 333.375 * 100 * 100))
+					res.append(position)
+					break
+		print(res)
+		return ','.join(res)
+
+if __name__ == '__main__':
+	click = Click3()
+	img = "https://static.geetest.com/captcha_v3/batch/v3/72556/2024-06-06T05/word/42a62cb4cafb461d92eee08598c13b95.jpg?challenge=a8a9ae907a766d68f9fed4035aab4929"
+	print(click.calculated_position(img))

geetest/www/main.py

@@ -0,0 +1,81 @@
+from api import Slide, Click
+import subprocess
+from slide3 import Slide3
+from click3 import Click3
+import time
+from ddddocr import DdddOcr
+
+def main1():
+	api = Slide()
+	slide3 = Slide3()
+
+	cmd0 = "node -e \"require('./rt.js').rt()\""
+	print(cmd0)
+	rt = subprocess.run(cmd0, shell=True, stdout=subprocess.PIPE).stdout.decode('utf-8')
+	(challenge, gt) = api.register()
+
+	print(challenge)
+	print(gt)
+	(c, s) = api.get_c_s(challenge, gt, None)
+
+	api.get_type(challenge, gt, None)
+
+	(c, s, challenge, bg_url, slice_url) = api.get_new_c_s_challenge_bg_slice(challenge, gt)
+	dis = slide3.calculated_distance(bg_url, slice_url)
+	cmd3 = "node -e \"require('./slide.js').send(" + str(dis) + ",\'" + gt + "\',\'" + challenge + "\'," + str(c) + ",\'" + s + "\',\'"+ rt + "\')\""
+	print(cmd3)
+	w = subprocess.run(cmd3, shell=True, stdout=subprocess.PIPE).stdout.decode('utf-8')
+	time.sleep(2)
+	res = api.ajax(challenge, gt, w)
+	print(res)
+
+def main2():
+	api = Click()
+	click3 = Click3()
+
+	cmd0 = "node -e \"require('./rt.js').rt()\""
+	print(cmd0)
+	rt = subprocess.run(cmd0, shell=True, stdout=subprocess.PIPE).stdout.decode('utf-8')
+	(challenge, gt) = api.register()
+
+	print(challenge)
+	print(gt)
+	(c, s) = api.get_c_s(challenge, gt, None)
+
+	api.get_type(challenge, gt, None)
+
+	(c, s, pic) = api.get_new_c_s_pic(challenge, gt)
+	position = click3.calculated_position(pic)
+	cmd3 = "node -e \"require('./click.js').send('" + gt + "\',\'" + challenge + "\'," + str(c) + ",\'" + s + "\',\'"+ rt + "\',\'" + position + "\')\""
+	print(cmd3)
+	w = subprocess.run(cmd3, shell=True, stdout=subprocess.PIPE).stdout.decode('utf-8')
+	time.sleep(2)
+	res = api.ajax(challenge, gt, w)
+	print(res)
+
+def main3():
+	api = Click()
+	click3 = Click3(DdddOcr(show_ad=False, beta=True))
+	count = 0
+	for i in range(300):
+		cmd0 = "node -e \"require('./rt.js').rt()\""
+		rt = subprocess.run(cmd0, shell=True, stdout=subprocess.PIPE).stdout.decode('utf-8')
+		(challenge, gt) = api.register()
+		(c, s) = api.get_c_s(challenge, gt, None)
+
+		api.get_type(challenge, gt, None)
+
+		(c, s, pic) = api.get_new_c_s_pic(challenge, gt)
+		position = click3.calculated_position(pic)
+		cmd3 = "node -e \"require('.click.js').send('" + gt + "\',\'" + challenge + "\'," + str(c) + ",\'" + s + "\',\'"+ rt + "\',\'" + position + "\')\""
+		w = subprocess.run(cmd3, shell=True, stdout=subprocess.PIPE).stdout.decode('utf-8')
+		time.sleep(2)
+		res = api.ajax(challenge, gt, w)
+		print(res)
+		if(res['data']['result'] == 'success'):
+			count+=1
+
+	print(count/300)
+
+if __name__ == '__main__':
+    main3()

geetest/www/slide3.py

@@ -0,0 +1,79 @@
+# -*- coding: utf-8 -*-
+import io
+from pathlib import Path
+import requests
+from PIL import Image
+import ddddocr
+
+def parse_bg_captcha(img, im_show=False, save_path=None):
+    """
+    滑块乱序背景图还原
+    :param img: 图片路径str/图片路径Path对象/图片二进制
+        eg: 'assets/bg.webp'
+            Path('assets/bg.webp')
+    :param im_show: 是否显示还原结果, <type 'bool'>; default: False
+    :param save_path: 保存路径, <type 'str'>/<type 'Path'>; default: None
+    :return: 还原后背景图 RGB图片格式
+    """
+    if isinstance(img, (str, Path)):
+        _img = Image.open(img)
+    elif isinstance(img, bytes):
+        _img = Image.open(io.BytesIO(img))
+    else:
+        raise ValueError(f'输入图片类型错误, 必须是<type str>/<type Path>/<type bytes>: {type(img)}')
+    # 图片还原顺序, 定值
+    _Ge = [39, 38, 48, 49, 41, 40, 46, 47, 35, 34, 50, 51, 33, 32, 28, 29, 27, 26, 36, 37, 31, 30, 44, 45, 43,
+           42, 12, 13, 23, 22, 14, 15, 21, 20, 8, 9, 25, 24, 6, 7, 3, 2, 0, 1, 11, 10, 4, 5, 19, 18, 16, 17]
+    w_sep, h_sep = 10, 80
+
+    # 还原后的背景图
+    new_img = Image.new('RGB', (260, 160))
+
+    for idx in range(len(_Ge)):
+        x = _Ge[idx] % 26 * 12 + 1
+        y = h_sep if _Ge[idx] > 25 else 0
+        # 从背景图中裁剪出对应位置的小块
+        img_cut = _img.crop((x, y, x + w_sep, y + h_sep))
+        # 将小块拼接到新图中
+        new_x = idx % 26 * 10
+        new_y = h_sep if idx > 25 else 0
+        new_img.paste(img_cut, (new_x, new_y))
+
+    if im_show:
+        new_img.show()
+    if save_path is not None:
+        save_path = Path(save_path).resolve().__str__()
+        new_img.save(save_path)
+    return new_img
+
+class Slide3:
+	def __init__(self, ocr: ddddocr.DdddOcr = ddddocr.DdddOcr(show_ad=False)):
+		self.ocr = ocr
+	def calculated_distance(self, bg_url: str, slice_url: str) -> int:
+		bg = requests.get(bg_url)
+		bg_image = parse_bg_captcha(bg.content, im_show=False)  # 假设不显示，仅用于比较
+		slice_image = requests.get(slice_url).content
+		bg1_bytes = io.BytesIO()
+		bg_image.save(bg1_bytes, format='PNG')  # 注意指定正确的格式
+		bg1_bytes.seek(0)
+
+		res = self.ocr.slide_match(slice_image, bg1_bytes.read(), simple_target=True)
+		return res['target'][0]
+
+
+
+def main():
+    ocr = ddddocr.DdddOcr(show_ad=False)
+    bg = requests.get('http://static.geetest.com/pictures/gt/cd0bbb6fe/bg/1a937a358.jpg')
+    bg_image = parse_bg_captcha(bg.content, im_show=False)  # 假设不显示，仅用于比较
+    slice_image = requests.get('http://static.geetest.com/pictures/gt/cd0bbb6fe/slice/1a937a358.png').content
+    bg1_bytes = io.BytesIO()
+    bg_image.save(bg1_bytes, format='PNG')  # 注意指定正确的格式
+    bg1_bytes.seek(0)
+
+    res = ocr.slide_match(slice_image, bg1_bytes.read())
+
+    print(res)
+
+if __name__ == '__main__':
+	main()

requirements.txt

@@ -4,4 +4,6 @@ setuptools~=65.5.1
 gradio~=4.27.0
 qrcode~=7.4.2
 loguru~=0.7.2
-webdriver-manager~=4.0.1
+webdriver-manager~=4.0.1
+ddddocr~=1.4.11
+Pillow~=10.3.0