def login(self, account, password):
"""
登录校园卡网站
:param str account: 校园卡卡号
:param str password: 校园卡查询密码
:return: {'code': 1, "msg": "登录失败，请重试"} 或 {'code': 0, 'msg': '登录成功'}
"""
captcha_code = CardCaptcha(self.get_image(URL.card_captcha())).crack()
data = {
"sno": account,
"pwd": base64.b64encode(str(password).encode("utf8")),
"ValiCode": captcha_code,
"remember": 1,
"uclass": 1,
"zqcode": "",
"json": True,
}
result = self._login_execute(url=URL.card_login(), data=data)
if result['code'] == 2:
# 如果验证码错误，尝试递归重复登录
return self.login(account, password)
result['success'] = not result['code']
if result['success']:
self.verify = True
else:
raise AuthenticationException(result['msg'])
return result
def login(self, account, password):
"""
登录教务系统 jwxt.njupt.edu.cn
:param account: 南邮学号
:param password: 密码
:return: {'code': 1, "msg": "登录失败"} 或 {'code': 0, 'msg': '登录成功'}
"""
captcha_code = ZhengfangCaptcha(self.get_image(
URL.zhengfang_captcha())).crack()
data = {
"__VIEWSTATE": self._get_viewstate(URL.zhengfang_login()),
'txtUserName': account,
'TextBox2': password,
'RadioButtonList1': "%D1%A7%C9%FA",
"txtSecretCode": captcha_code,
"Button1": "",
"hidPdrs": "",
"hidsc": ""
}
result = self._login_execute(url=URL.zhengfang_login(), data=data)
if result['code'] == 2:
# 如果验证码错误，尝试递归重复登录
return self.login(account, password)
result['success'] = not result['code']
if result['success']:
self.verify = True
else:
raise AuthenticationException(result['msg'])
return result
def _get_captcha(self, url=URL.jwxt_captcha()):
r = self.get(url)
im = Image.open(BytesIO(r.content))
if url == URL.aolan_captcha():
return str(AolanCaptcha(im))
if url == URL.jwxt_captcha():
return str(ZhengfangCaptcha(im))
def get_score(self):
"""
获取课程成绩和绩点等信息
:return: dict
{'gpa': 4.99,
'coursers': [{
'year': '2015-2016',
'semester': '1',
'code': '00wk00003',
'name': '从"愚昧"到"科学"-科学技术简史',
'attribute': '任选',
'belong': '全校任选课',
'credit': '2.0',
'point': '',
'score': '81',
'minorMark': '0',
'makeUpScore': '',
'retakeScore': '',
'college': '网络课程',
'note': '',
'retakeMark': '0',
'englishName': ''
},
]
}
"""
viewstate = self._get_viewstate(url=URL.jwxt_score(self.account))
data = {
'ddlXN': '',
'ddlXQ': '',
'__VIEWSTATE': viewstate,
'Button2': '%D4%DA%D0%A3%D1%A7%CF%B0%B3%C9%BC%A8%B2%E9%D1%AF'
}
soup = self._url2soup(method='post',
url=URL.jwxt_score(self.account),
data=data)
result = {'gpa': float(soup.select_one('#pjxfjd > b').text[7:])}
cols = [
'year', 'semester', 'code', 'name', 'attribute', 'belong',
'credit', 'point', 'score', 'minorMark', 'makeUpScore',
'retakeScore', 'college', 'note', 'retakeMark', 'englishName'
]
# 学年 学期 课程代码 课程名称 课程性质 课程归属 学分 绩点 成绩 辅修标记 补考成绩 重修成绩 开课学院 备注 重修标记 英文名称
coursers = []
for tr in soup.select('#Datagrid1 > tr')[1:]:
courser = {}
for col, td in zip(cols, tr.select('td')):
value = td.text.strip()
courser[col] = value
coursers.append(courser)
result['coursers'] = coursers
return result
def get_score(self):
"""
获取课程成绩和绩点等信息
:return: dict
{'gpa': 4.99, # GPA
'courses': [{
'year': '2015-2016', # 修读学年
'semester': '1', # 修读学期
'code': '00wk00003', # 课程编号
'name': '从"愚昧"到"科学"-科学技术简史', # 课程中文名
'attribute': '任选', # 课程性质
'belong': '全校任选课', # 课程归属
'credit': '2.0', # 学分
'point': '', # 绩点
'score': '81', # 成绩
'minorMark': '0', # 重修标记
'make_up_score': '', # 补考成绩
'retake_score': '', # 重修成绩
'college': '网络课程', # 开课学院
'note': '', # 备注
'retake_mark': '0', # 重修标记
'english_name': '' # 课程英文名
},
]
}
"""
viewstate = self._get_viewstate(url=URL.zhengfang_score(self.account))
data = {
'ddlXN': '',
'ddlXQ': '',
'__VIEWSTATE': viewstate,
'Button2': '%D4%DA%D0%A3%D1%A7%CF%B0%B3%C9%BC%A8%B2%E9%D1%AF'
}
soup = self.get_soup(method='post',
url=URL.zhengfang_score(self.account),
data=data)
result = {'gpa': float(soup.select_one('#pjxfjd').text[7:])}
cols = [
'year', 'semester', 'code', 'name', 'attribute', 'belong',
'credit', 'point', 'score', 'minor_mark', 'make_up_score',
'retake_score', 'college', 'note', 'retake_mark', 'english_name'
]
courses = []
for tr in soup.select('#Datagrid1 > tr')[1:]:
courser = {}
for col, td in zip(cols, tr.select('td')):
value = td.text.strip()
courser[col] = value
courses.append(courser)
result['courses'] = courses
return result
def login(self, account, password):
"""
登录南邮图书馆 jwxt.njupt.edu.cn
:param account: 南邮学号
:param password: 密码
:return: {'r': 1, "msg": "登录失败"} 或 {'r': 0, 'msg': '登录成功'}
"""
data = {
"number": account,
'passwd': password,
'captcha': self._get_captcha(URL.lib_captcha()),
'select': "cert_no",
"returnUrl": "",
}
return self._login_execute(url=URL.jwxt_login(), data=data)
def recharge_net(self, amount=0.01):
"""
充值网费（从校园卡余额中）
:param amount: 金额 如 2.33
:return: dict
{
'success':True,
'code' : 0,
'Msg' : '充值成功'
}
"""
data = {
'paytype': 1,
'aid': AIDS['net'],
'account': self._get_inner_account(),
'tran': int(amount * 100),
'netacc': '{"netacc": "", "bal": null, "pkgid": null, "lostflag": null, "freezeflag": null, '
'"expflag": null,"statustime": null, "duration": null, "starttime": null, "pkgtab": []}',
'pkgflag': 'none',
'pkg': '{}',
'acctype': '###',
'qpwd': '',
'json': True
}
result = self._url2json(url=URL.card_net_pay(), method='post', data=data)
result = json.loads(result['Msg'])
return {
'success': not int(result['pay_net_gdc']['retcode']),
'code': int(result['pay_net_gdc']['retcode']),
'msg': result['pay_net_gdc']['errmsg']
}
def _recharge_electricity(self, amount, elec_aid, building_id, building,
room_id):
data = {
"acctype": "###",
"paytype": 1,
"aid": elec_aid,
"account": self._get_inner_account(),
"tran": int(amount * 100),
"roomid": room_id,
"room": "",
"floorid": "",
"floor": "",
"buildingid": building_id,
"building": building,
"areaid": "",
"areaname": "",
"json": True
}
r = json.loads(
self.get_json(URL.card_elec_pay(), 'post', data=data)['Msg'])
return {
'success': not bool(int(r['pay_elec_gdc']['retcode'])),
'code': int(r['pay_elec_gdc']['retcode']),
'msg': r['pay_elec_gdc']['errmsg'],
}
def _get_viewstate(self, url=None):
response = self.get(url or URL.jwxt_login())
soup = BeautifulSoup(response.content, "lxml")
viewstate = soup.find('input', attrs={
"name": "__VIEWSTATE"
}).get("value")
return viewstate
def recharge(self, amount=0.01):
"""
从绑定的银行卡中扣款充值余额
:param amount: 充值金额，默认为0.01元
:return: dict
{
'success':True # 转账是否成功
'code': 0, # 状态码
'msg': '转账成功' # 附加信息
}
"""
data = {
'account': self._get_inner_account(),
'acctype': '###',
'tranamt': int(100 * amount),
'qpwd': '',
'paymethod': '2',
'paytype': '使用绑定的默认账号',
'client_type': 'web',
'json': True,
}
info = self.get_json(URL.card_recharge(), 'post', data)
info = json.loads(info['Msg'])['transfer']
result = {
'code': int(info['retcode']),
'msg': info['errmsg'],
'success': not bool(int(info['retcode']))
}
return result
def get_bill(self,
start_date=(datetime.datetime.now() -
datetime.timedelta(days=30)).strftime("%Y-%m-%d"),
end_date=datetime.datetime.now().strftime("%Y-%m-%d"),
rows=100,
page=1):
"""
查询校园卡消费记录，默认为最近一个月的消费记录
:param start_date: 查询区间的开始时间 "2017-12-27"
:param end_date: 查询区间的结束时间 "2018-01-26"
:param rows: 一次查询返回的最大记录数量,默认为100条记录
:param page: 如果结果数量有多页，决定返回第几页。
:return: list
{'recodes':
[{'balance': 39.71, # 余额
'change': -5, # 变动
'comment': '未知系统,交电费', # 注释
'department': '仙林售电处', # 操作部门
'time': '2018-01-26 20:55:40', # 时间
'type': '代扣代缴', # 类型
'week': '星期五'}, # 星期
{'balances': 39.71,
'change': -7.5,
'comment': '',
'department': '一餐厅二楼清真食堂',
'time': '2018-01-24 17:09:36',
'type': '持卡人消费',
'week': '星期三'}
... ],
'total': 52 # 总的记录数
'total_pages':2 # 总页数
'page':1 # 当前的页码
}
"""
data = {
"sdate": start_date,
"edate": end_date,
"account": self._get_inner_account(),
"page": page,
"rows": rows
}
temp = self.get_json(url=URL.card_bill(), method="post", data=data)
result = {
'total': temp['total'],
'recodes': [],
'total_pages': temp['total'] // rows + 1,
'page': page
}
if temp['rows']:
for row in temp['rows']:
result['recodes'].append({
'balance': row['CARDBAL'],
'department': row['MERCNAME'].strip(),
'comment': row['JDESC'].strip(),
'change': row['TRANAMT'],
'week': row['XQ'],
'time': row['OCCTIME'],
'type': row['TRANNAME'].strip(),
})
return result
def login(self, account, password):
"""
登录奥兰系统 jwxt.njupt.edu.cn
:param account: 南邮学号、考生号、身份证号
:param password: 密码
:return: {'r': 1, "msg": "登录失败"} 或 {'r': 0, 'msg': '登录成功'}
"""
data = {
"__VIEWSTATE": self._get_viewstate(URL.aolan_login()),
'__VIEWSTATEGENERATOR':
self._get_viewstategenerator(URL.aolan_login()),
'userbh': account,
'pas2s': hashlib.md5(password.upper().encode('utf8')).hexdigest(),
"vcode": self._get_captcha(URL.aolan_captcha()),
"cw": "",
"xzbz": "1",
}
return self._login_execute(url=URL.aolan_login(), data=data)
def login(self, account, password):
"""
登录教务系统 jwxt.njupt.edu.cn
:param account: 南邮学号
:param password: 密码
:return: {'r': 1, "msg": "登录失败"} 或 {'r': 0, 'msg': '登录成功'}
"""
data = {
"__VIEWSTATE": self._get_viewstate(),
'txtUserName': account,
'TextBox2': password,
'RadioButtonList1': "%D1%A7%C9%FA",
"txtSecretCode": self._get_captcha(URL.jwxt_captcha()),
"Button1": "",
"hidPdrs": "",
"hidsc": ""
}
return self._login_execute(url=URL.jwxt_login(), data=data)
def _get_inner_account(self):
"""
获取对应的内部账号，部分接口参数需要
:return: 校园卡号对应的系统内部账号
"""
if not self.inner_account:
info = self._url2json(URL.card_info(), method='post', data={'json': True})
result = json.loads(info['Msg'])
self.inner_account = result['query_card']['card'][0]['account']
return self.inner_account
def get_courses(self):
"""
获取这学期的选课情况
:return:
"""
soup = self.get_soup(method='get',
url=URL.zhengfang_courses(self.account))
trs = soup.select('#DBGrid > tr')[1:]
courses = []
for tr in trs:
tds = [node.text.strip() for node in tr.select('td')]
name = tds[2]
teacher = tds[5]
all_time = tds[8].split(';')
all_room = tds[9].split(';')
for time, room in zip(all_time, all_room):
if time and room:
week_start, week_end = map(int,
week_re.search(time).groups())
courser_index = list(
map(
int,
courser_indexs_re.search(time).groups()[0].split(
',')))
week = re.search('{(.*)}', time).groups()[0]
if '双周' in week and week_start % 2 == 1:
week_start += 1
if '单周' in week and week_start % 2 == 0:
week_start += 1
courses.append({
'day':
chinese_rome[time[1]],
'name':
name,
'week':
week,
'week_start':
week_start,
'week_end':
week_end,
'class_start':
courser_index[0],
'class_end':
courser_index[-1],
'teacher':
teacher,
'room':
room,
'interval':
2 if '单周' in week or '双周' in week else 1,
})
return courses
def wrapper(self, *args, **kwargs):
if self.verify: # 利用状态量进行状态的判定
return func(self, *args, **kwargs)
else:
# 先判断有没有cookie文件, 再判断cookie是否有效
if 'WEB' in requests.utils.dict_from_cookiejar(self.cookies):
# 判断教务系统是否登录成功，根据正方的一个地址无需参数的地址
r = self.get_soup(method="get", url=URL.zhengfang_logintest())
if 'Object moved to ' not in r.text:
self.verify = True
return func(self, *args, **kwargs)
raise UnauthorizedError('未登录正方系统，请先登录')
def get_schedule(self, week, year=None, semester=None):
"""
获取指定学期指定周的课表（不传入年份和学期则默认当前学期）
:param year: 学年 格式为 "2017-2018"
:param semester: 学期 数字1或2
:param week: 周次 数字 1-20
:return: 二维列表schedule，schedule[i][j]代表周i第j节课的课程。 为了方便，i或j为零0的单元均不使用。
列表的元素为None，代表没有课程，或描述课程信息的dict，dict例如
{
'classroom': '教4－202',
'name': '技术经济学',
'teacher': '储成祥'
}
"""
schedule = [[None for col in range(13)] for row in range(8)]
if year and semester:
pass
else:
r = self.get(url=URL.zhengfang_class_schedule(self.account))
soup = BeautifulSoup(r.text.replace('<br>', '\n'), 'lxml')
trs = soup.select("#Table1 > tr")
for index, tr in enumerate(trs):
tds = tr.select('td')
for td in tds:
if len(td.text) > 4: # 筛选出包含课程信息的单元格
for courser in td.text.split('\n\n'):
info = courser.split()
start_week, end_week = map(
int,
week_re.search(info[1]).groups())
courser_index = map(
int,
courser_indexs_re.search(
info[1]).groups()[0].split(','))
is_odd_week_only = "单周" in info[1]
is_even_week_only = "双周" in info[1]
week_day = chinese_rome[info[1][1]]
courser = {
'name': info[0],
'teacher': info[2],
'classroom': info[3],
}
if start_week <= week <= end_week:
if (week % 2 == 0 and is_odd_week_only) or (
week % 2 == 1 and is_even_week_only):
pass
else:
for courser_index in courser_index:
schedule[week_day][
courser_index] = courser
return schedule
def get_net_balance(self):
"""
获取Dr.com的上网费用余额
:return: float 2.33
"""
data = {
'jsondata': '{"query_net_info": {"aid": "%s", "account": "%s", "payacc": ""}}' # 无法使用format格式化
% (AIDS['net'], self._get_inner_account()),
'funname': 'synjones.onecard.query.net.info',
'json': True,
}
result = self._url2json(url=URL.card_common(), data=data, method='post')
result = json.loads(result['Msg'])
return float(re.search(r'余额(\d*\.\d*)元', result['query_net_info']['errmsg']).groups()[0])
def _get_build_ids(self, aid):
"""获取建筑的id"""
data = {
'jsondata': '{"query_elec_building": {"aid": "%s", "account": "%s", "area": {"area": "", "areaname": ""}}}'
% (aid, self._get_inner_account()),
'funname': 'synjones.onecard.query.elec.building',
'json': True,
}
result = self._url2json(data=data, url=URL.card_common(), method='post')
result = json.loads(result['Msg'])
# - -!
building_id = {}
for build in result['query_elec_building']['buildingtab']:
building_id[build['building']] = build['buildingid']
return building_id
def wrapper(self, *args, **kwargs):
if self.verify: # 利用状态量进行状态的判定
return func(self, *args, **kwargs)
else:
# 先判断有没有cookie文件, 再判断cookie是否有效
if 'WEB' in requests.utils.dict_from_cookiejar(self.cookies):
# 判断教务系统是否登录成功，根据正方的一个地址无需参数的地址
r = self._url2soup(method="get", url=URL.jwxt_logintest())
if 'Object moved to ' not in r.text:
self.verify = True
return func(self, *args, **kwargs)
while True: # 运行到此处代表cookies无效
data = self.login(self.account, input('输入密码:'))
if data.get('r') == 0:
self.verify = True
return func(self, *args, **kwargs)
else:
print(data.get('msg'))
def wrapper(self, *args, **kwargs):
success = False
# 先判断有没有cookie文件, 再判断cookie是否有效
if 'WEB' in requests.utils.dict_from_cookiejar(self.cookies):
# 判断教务系统是否登录成功，根据正方的一个地址无需参数的地址
r = self._url2soup(method="get", url=URL.jwxt_logintest())
if 'Object moved to ' not in r.text:
success = True
while not success:
account = input("请输入教务系统账号:")
self.account = account
password = input("请输入账号密码:")
data = self.login(account, password)
if data.get("r") == 0:
success = True
else:
print(data.get("msg"))
else:
return func(self, *args, **kwargs)
def get_class_schedule(self, week, year=None, semester=None):
"""
获取指定学期指定周的课表（不传入年份和学期则默认当前学期）
:param year: 学年 格式为 "2017-2018"
:param semester: 学期 数字1或2
:param week: 周次 数字 1-20
:return:
"""
if year and semester:
pass
else:
r = self.get(url=URL.jwxt_class_schedule(self.account))
soup = BeautifulSoup(r.text.replace('<br>', '\n'), 'lxml')
trs = soup.select("#Table1 > tr")
schedule = {}
for tr in trs:
tds = tr.select('td')
for td in tds:
print(td.text.split())
def get_balance(self):
"""
查询余额
:return: dict
{
'balance': 10.01, # 到账余额
'unsettle_balance': 0.01 # 过渡余额
'total': 10.02 # 总余额
}
"""
info = self._url2json(URL.card_info(), method='post', data={'json': True})
result = json.loads(info['Msg'])
balance = int(result['query_card']['card'][0]['db_balance']) / 100 # 到账余额
unsettle_balance = int(result['query_card']['card'][0]['unsettle_amount']) / 100 # 过渡余额
return {
'balance': balance, # 到账余额
'unsettle_balance': unsettle_balance, # 过渡余额
'total': balance + unsettle_balance, # 总余额
}
def get_grade(self):
"""
获取等级考试成绩信息
:return: list
[
{'date': '20151219', # 考试日期
'name': '全国大学英语四级考试', # 考试名称
'number': '320082152113313', # 准考证号
'score': '547', # 成绩
'semester': '1', # 学期
'year': '2015-2016' # 学年
},
]
"""
soup = self.get_soup(method='get',
url=URL.zhengfang_grade(self.account))
results = []
for tr in soup.select("#DataGrid1 > tr")[1:]:
names = ['year', 'semester', 'name', 'number', 'date', 'score']
result = {}
for name, td in zip(names, tr.select('td')[:6]):
result[name] = td.text
results.append(result)
return results
:param data_type: 提交的数据格式(可能是表单类型,也可能是json格式的字符串)
:param kwargs: requests支持的参数，比如可以设置代理参数
:return: BeautifulSoup对象
"""
# r = getattr(self, method)(url, json=json, data=data, params=params, **kwargs)
# if r.ok:
# soup = BeautifulSoup(r.text, 'lxml')
# return soup
# else:
# raise ConnectionError("检查网络连接")
try: # 出现网络连接问题,直接在该处抛出错误
r = getattr(self, method)(url,
json=json,
data=data,
params=params,
**kwargs)
except ConnectionError:
raise ConnectionError("请检查网络连接")
else:
if r.ok: # 状态码小于400为True
soup = BeautifulSoup(r.text, 'lxml')
return soup
else: # 处理其他状态码
raise Exception('请确保能够正常访问当前页面: {}'.format(url))
if __name__ == "__main__":
test_model = Model()
test_model._get_captcha(URL.aolan_captcha())
手动获取奥兰系统验证码的训练集
"""
import hashlib
import os
import time
from io import BytesIO
import requests
from PIL import Image
from njupt.settings import BASE_DIR
from njupt.urls import URL
from njupt.utils.aolan.aolan_captcha import letters
for i in range(100):
r = requests.get(URL.aolan_captcha())
im = Image.open(BytesIO(r.content))
# 灰度处理
imgry = im.convert('L')
# 二值处理
threshold = 220
table = []
for i in range(256):
if i < threshold:
table.append(0)
else:
table.append(1)
out = imgry.point(table, '1')
out.show()
numbers = input("输入数值")
def _get_captcha(self, url=URL.jwxt_captcha()):
r = self.get(url)
im = Image.open(BytesIO(r.content))
im.show()
captcha = input("输入验证码：")
return captcha
def crack(self):
result = []
v = VectorTools()
# 装载训练数据集
with open(os.path.join(current_dir, 'imageset.dat'), 'rb+') as f:
imageset = pickle.load(f)
for letter in self.handle_split_image():
guess = []
for image in imageset:
for x, y in image.items():
if len(y) != 0:
guess.append((v.relation(y[0], self.buildvector(letter)), x))
guess.sort(reverse=True)
neighbors = guess[:10] # 距离最近的十个向量
class_votes = {} # 投票
for neighbor in neighbors:
class_votes.setdefault(neighbor[-1], 0)
class_votes[neighbor[-1]] += 1
sorted_votes = sorted(class_votes.items(), key=lambda x: x[1], reverse=True)
result.append(sorted_votes[0][0])
return ''.join(result)
def __str__(self):
return self.crack()
if __name__ == "__main__":
r = requests.get(URL.zhengfang_captcha())
im = Image.open(BytesIO(r.content))
print(ZhengfangCaptcha(im))
