use 2captcha api to solve verify code

README.md

@@ -3,7 +3,17 @@
 
 ## Notice
 
-**程序已失效**
+**更新验证模块**
+
+使用此网站[2captcha.com](https://2captcha.com?from=12332166)提供的api。
+
+但是该API 收费，一亩三分地的验证码是 reCAPTCHA v2 - hard ，对应的[收费](https://2captcha.com/2captcha-api)是 $2.99/1000次请求
+
+如果想使用需要注册账号+充值只少3刀（单个账号可以用500天左右）+ 复制 apikey 替换configure/data.json 文件中的字段
+
+
+
+**程序失效**
 
 论坛增加了google Google reCAPTCHA 验证码，导致程序失效
 
@@ -39,7 +49,7 @@
 
 [一亩三分地](https://www.1point3acres.com/bbs/) 自动签到、答题
 
-一分钟快速设置，无 aws 依赖，验证码自动识别，一次性设置后再无需手动操作
+快速设置，无 aws 依赖，验证码自动识别，一次性设置后再无需手动操作
 
 * 程序会自动识别验证码
 
@@ -51,11 +61,16 @@
 
 ## how to use
 
-~~下面几种执行模式任选一种即可。~~
+下面几种执行模式任选一种即可。
 
-### ~~本地 crontab 模式（由于验证码机制更改，现已失效）~~
+### 本地 crontab 模式
 
-* 修改 configure/data.json，用你的用户名密码替换文件中的相应字段
+* 注册[2captcha.com](https://2captcha.com?from=12332166) 得到 apikey 并充值
+    <details>
+    <summary>具体操作展开查看</summary>
+    ![](screenshots/2captcha.png)
+    </details>
+* 修改 configure/data.json，用你的用户名,密码,apikey替换文件中的相应字段
 
 * 安装依赖
 以 ubuntu 为例，其他系统请用相应的方式安装依赖

requirements.txt

@@ -1,7 +1,6 @@
 requests==2.22.0
 lxml==4.6.3
 cssselect==1.1.0
-pytesseract==0.3.7
-Pillow==8.2.0
+TwoCaptcha==0.0.1
 
 

src/questions.py

@@ -127,6 +127,10 @@
 	'把一亩三分地介绍给朋友的正确姿势不包括什么？': '直接把我的账号密码给朋友用',
 	'关于论坛账号分享的说法，哪些正确?': '这些都不对',
 	'如果我把论坛积分隐藏内容公开发出来会怎么样？':'会被扣分，情节严重者会封号',
+	'以下关于帖子审核，哪些说法正确？':'这些都不正确',
+	# 我感觉这道题答案错了
+	#'在论坛发slack群、qq群、微信群等任何站外讨论方式，会如何？':'以上都正确',
+	'在论坛发slack群、qq群、微信群等任何站外讨论方式，会如何？':'如果发在本地版聚、租房找室友、飞友学友工友版块，可以找到小伙伴',
 
 	#以下没有验证
 	'一亩三分地发面经的规则有哪些？':'以上都正确',
@@ -147,3 +151,4 @@
 # 7.27 2
 # 7.28 1
 # 7.29 2
+# 8.1  0

src/raw_requests.py

@@ -1,9 +1,6 @@
 import requests
 import urllib.parse
 import re
-from PIL import Image
-from process_gif import get_code_from_gif
-from cfproxy import CFProxy
 import xml.dom.minidom as xml
 import lxml.html as html
 import questions
@@ -38,7 +35,9 @@
 worker = "www.laika42.top"
 # ip = '104.26.8.210'
 ip = '116.234.47.7'
-proxy = CFProxy(worker, user_agent, ip)
+
+
+# proxy = CFProxy(worker, user_agent, ip)
 
 
 def save_error(response: requests.Response, error_desc: str = ""):
@@ -63,7 +62,7 @@ def check_status_code(response: requests.Response, error_desc: str = ""):
 
 def get_login_info_() -> (str, str):
 	global cookie_jar
-	global proxy
+	# global proxy
 	header = {
 		"User-Agent": user_agent,
 		"Referer": referer,
@@ -176,69 +175,70 @@ def get_checkin_info_() -> (str, str):
 	return form_hash, sec_hash
 
 
-def get_verify_code_(id_hash) -> str:
-	global cookie_jar
-	print("get verify code...")
-	header = {
-		"User-Agent": user_agent,
-		"Referer": referer
-	}
-	# print(cookie_jar)
-	response = requests.get(get_verify_code_url % (id_hash, id_hash), headers=header, cookies=cookie_jar)
-	# response = proxy.get(get_verify_code_url % (id_hash, id_hash), headers=header, cookies=cookie_jar)
-	cookie_jar.update(response.cookies)
-	check_status_code(response, "get verify code phase 1 error")
-	# misc.php?mod = seccode & update = 86288 & idhash = S0
-	pattern = re.compile('src="([a-zA-Z0-9=&?.]+)"')
-	srcs = pattern.findall(response.text)
-	verify_code_url = ""
-	if len(srcs) >= 1:
-		verify_code_url = "https://www.1point3acres.com/bbs/" + srcs[0]
-	else:
-		print("response 中没有 验证码 的 url")
-		print(response.text)
-		exit(-1)
-	response = requests.get(verify_code_url, headers=header, cookies=cookie_jar)
-	# response = proxy.get(verify_code_url, headers=header, cookies=cookie_jar)
-	cookie_jar.update(response.cookies)
-	check_status_code(response, "get verify code phase 2 error")
-	# print(len(response.content)) # 文件大小
-	tmpfilename = "tmp.gif"
-	if os.name == "posix":
-		tmpfilename = "/tmp/" + tmpfilename
-	file = open(tmpfilename, "wb")
-	file.write(response.content)
-	file.close()
-	f = Image.open(tmpfilename)
-	# f.show()
-	return get_code_from_gif(f)
-
-
-def check_verify_code_(id_hash, code):
-	global cookie_jar
-	print("check verify code...")
-
-	header = {
-		"User-Agent": user_agent,
-		# "Referer": "https://www.1point3acres.com/bbs/dsu_paulsign-sign.html"
-		"Referer": referer
-	}
-	response = requests.get(check_verify_code_url % (id_hash, code), headers=header, cookies=cookie_jar)
-	# response = proxy.get(check_verify_code_url % (id_hash, code), headers=header, cookies=cookie_jar)
-	cookie_jar.update(response.cookies)
-	check_status_code(response, "check verify code phase 1 error")
-	if "succeed" in response.text:
-		print("verify code is right")
-		return True
-	if "invalid" in response.text:
-		print("verify code is wrong")
-		return False
-	print("verify error")
-	print(response.text)
-	exit(-1)
-
-
-def do_daily_checkin_(verify_code: str, form_hash: str, sec_hash: str = "S00") -> bool:
+#
+# def get_verify_code_(id_hash) -> str:
+# 	global cookie_jar
+# 	print("get verify code...")
+# 	header = {
+# 		"User-Agent": user_agent,
+# 		"Referer": referer
+# 	}
+# 	# print(cookie_jar)
+# 	response = requests.get(get_verify_code_url % (id_hash, id_hash), headers=header, cookies=cookie_jar)
+# 	# response = proxy.get(get_verify_code_url % (id_hash, id_hash), headers=header, cookies=cookie_jar)
+# 	cookie_jar.update(response.cookies)
+# 	check_status_code(response, "get verify code phase 1 error")
+# 	# misc.php?mod = seccode & update = 86288 & idhash = S0
+# 	pattern = re.compile('src="([a-zA-Z0-9=&?.]+)"')
+# 	srcs = pattern.findall(response.text)
+# 	verify_code_url = ""
+# 	if len(srcs) >= 1:
+# 		verify_code_url = "https://www.1point3acres.com/bbs/" + srcs[0]
+# 	else:
+# 		print("response 中没有 验证码 的 url")
+# 		print(response.text)
+# 		exit(-1)
+# 	response = requests.get(verify_code_url, headers=header, cookies=cookie_jar)
+# 	# response = proxy.get(verify_code_url, headers=header, cookies=cookie_jar)
+# 	cookie_jar.update(response.cookies)
+# 	check_status_code(response, "get verify code phase 2 error")
+# 	# print(len(response.content)) # 文件大小
+# 	tmpfilename = "tmp.gif"
+# 	if os.name == "posix":
+# 		tmpfilename = "/tmp/" + tmpfilename
+# 	file = open(tmpfilename, "wb")
+# 	file.write(response.content)
+# 	file.close()
+# 	f = Image.open(tmpfilename)
+# 	# f.show()
+# 	return get_code_from_gif(f)
+#
+#
+# def check_verify_code_(id_hash, code):
+# 	global cookie_jar
+# 	print("check verify code...")
+#
+# 	header = {
+# 		"User-Agent": user_agent,
+# 		# "Referer": "https://www.1point3acres.com/bbs/dsu_paulsign-sign.html"
+# 		"Referer": referer
+# 	}
+# 	response = requests.get(check_verify_code_url % (id_hash, code), headers=header, cookies=cookie_jar)
+# 	# response = proxy.get(check_verify_code_url % (id_hash, code), headers=header, cookies=cookie_jar)
+# 	cookie_jar.update(response.cookies)
+# 	check_status_code(response, "check verify code phase 1 error")
+# 	if "succeed" in response.text:
+# 		print("verify code is right")
+# 		return True
+# 	if "invalid" in response.text:
+# 		print("verify code is wrong")
+# 		return False
+# 	print("verify error")
+# 	print(response.text)
+# 	exit(-1)
+#
+
+def do_daily_checkin_(g_token: str, form_hash: str, sec_hash: str = "S00") -> bool:
 	global cookie_jar
 	header = {
 		"User-Agent": user_agent,
@@ -254,7 +254,7 @@ def do_daily_checkin_(verify_code: str, form_hash: str, sec_hash: str = "S00") -
 		"sechash": sec_hash,
 		"seccodehash": sec_hash,
 		"seccodeverify": sec_hash,
-		"g-recaptcha-response": "",
+		"g-recaptcha-response": g_token,
 	}
 
 	response = requests.post(post_checkin_url, headers=header, data=body, cookies=cookie_jar)
@@ -337,7 +337,7 @@ def get_daily_task_answer() -> (str, str, str):
 	return answer_id, form_hash, sec_hash
 
 
-def do_daily_question_(answer: str, verify_code: str, form_hash: str, sec_hash: str = "SA00") -> bool:
+def do_daily_question_(answer: str, g_token: str, form_hash: str, sec_hash: str = "SA00") -> bool:
 	global cookie_jar
 	header = {
 		"User-Agent": user_agent,
@@ -348,7 +348,9 @@ def do_daily_question_(answer: str, verify_code: str, form_hash: str, sec_hash:
 		"formhash": form_hash,
 		"answer": answer,
 		"sechash": sec_hash,
-		"seccodeverify": verify_code,
+		"seccodehash": sec_hash,
+		"seccodeverify": sec_hash,
+		"g-recaptcha-response": g_token,
 		"submit": "true"
 	}
 	# 网站的原版请求是 multipart/form-data ，但是我发现用 application/x-www-form-urlencoded 也是可以的
@@ -368,6 +370,9 @@ def do_daily_question_(answer: str, verify_code: str, form_hash: str, sec_hash:
 	elif "恭喜你，回答正确" in response.text:
 		print("答题成功")
 		return True
+	elif "抱歉，回答错误！扣除1大米" in response.text:
+		print("答案错了，请报 issue: https://github.com/harryhare/1point3acres/issues/new")
+		return True
 	else:
 		save_error(response, "post answer")
 		return False

src/service.py

@@ -3,6 +3,7 @@
 import hashlib
 from raw_requests import *
 import sys
+from twocaptcha import TwoCaptcha
 
 
 # 签到流程
@@ -44,6 +45,22 @@ def get_verify_code(id_hash) -> str:
 	return verify_code
 
 
+def get_g_token(api_key, url) -> str:
+	solver = TwoCaptcha(api_key)
+	try:
+		result = solver.solve_captcha(
+			site_key='6LeCeskbAAAAAE-5ns6vBXkLrcly-kgyq6uAriBR',
+			page_url=url)
+
+	except Exception as e:
+		print(e)
+		exit(-1)
+		return ""
+	else:
+		print('solved: ' + str(result))
+		return result
+
+
 # 如果 login 失败, 后面的操作没必要再做，直接 exit
 def daily_login(username: str, password_hashed: str):
 	print("do login...")
@@ -61,10 +78,10 @@ def daily_checkin(api_key: str) -> bool:
 	time.sleep(2)
 	if form_hash == "":
 		return False
-	code = get_verify_code(sec_hash)
+	code = get_g_token(api_key, get_checkin_url)
 	if code == "":
 		return False
-	return do_daily_checkin_(verify_code=code, form_hash=form_hash, sec_hash=sec_hash)
+	return do_daily_checkin_(g_token=code, form_hash=form_hash, sec_hash=sec_hash)
 
 
 def daily_question(api_key: str) -> bool:
@@ -73,10 +90,10 @@ def daily_question(api_key: str) -> bool:
 	time.sleep(2)
 	if form_hash == "" or answer == "":
 		return False
-	code = get_verify_code(sec_hash)
+	code = get_g_token(api_key, get_checkin_url)
 	if code == "":
 		return False
-	return do_daily_question_(answer=answer, verify_code=code, form_hash=form_hash, sec_hash=sec_hash)
+	return do_daily_question_(answer=answer, g_token=code, form_hash=form_hash, sec_hash=sec_hash)
 
 
 def do_all(username: str, password: str, api_key: str):
@@ -98,9 +115,6 @@ def main(from_file: bool = False):
 	else:
 		api_key = sys.argv[1]
 		users = json.loads(sys.argv[2].replace("'", '"'))
-	print(users)
-	print(api_key)
-	exit(0)
 	for user in users:
 		m = hashlib.md5()
 		m.update(user["password"].encode("ascii"))