USTC Web services

ustc.py

"""
the USTC module

NOTE: this is an incomplete framework! extend it as you like.
PS:   don't expect to receive updates or changes from me! this is not intended to be published anyway. :-p
"""

from bs4 import BeautifulSoup
import requests


class Passport:
    """
    USTC Passport

    1. GET USTC_LOGIN and obtain the CSRF token
    2. Construct a form and POST it to USTC_LOGIN
    3. Get cookies `laravel_session` and `ucas_TGC_2015`
    """

    LOGIN_URL = "https://passport.ustc.edu.cn/login"

    def __init__(self, student_id: str, password: str):
        # Query a token
        res = requests.get(Passport.LOGIN_URL)
        cookies = res.cookies
        soup = BeautifulSoup(res.text, 'html.parser')
        token = soup.form.input['value']

        # Login
        payload = {'_token': token, 'login': student_id, 'password': password, 'button': '登录'}
        res = requests.post(Passport.LOGIN_URL, data=payload, cookies=cookies)

        # Save the cookies for future use
        self.cookies = res.cookies

        # Check if succeeded
        if res.text.find('成功') == -1:
            raise RuntimeError("Couldn't not log into Passport")


class MIS:
    """
    USTC MIS

    1. GET USTC_MIS_LOGIN with appropriate cookies
    2. Cookie `JSESSIONID` returned
    """

    LOGIN_URL = "https://passport.ustc.edu.cn/login?service=http%3A%2F%2Fmis.teach.ustc.edu.cn/casLogin.do"
    GRADES_URL = "http://mis.teach.ustc.edu.cn/querycjxx.do"

    def __init__(self, passport: Passport):
        res = requests.get(MIS.LOGIN_URL, cookies=passport.cookies)
        self.cookies = res.cookies

    def courses(self, semester: str):
        """
        Query courses you selected for a specific semester. Semester is a string in the format
        `'XXXXX'`, where `X` is a digit. The first 4 digits stand for the year, and the last digit
        stands for the specific semester, which can be 1 (Fall), 2 (Spring) or 3 (Summer).
        """
        raise NotImplementedError()

    def grades(self):
        """
        Query your grades.

        Surprisingly, GET request is sufficient. HTML contents (!) are returned, so we must parse
        them into machine-readable format, ie. dictionary.

        @return as {total_credits, obtained_credits, not_obtained_credits, failed_credits, GPA,
            average, weighted_average, [{semester, number, name, type, grade, credits}]}
        """
        res = requests.get(MIS.GRADES_URL, cookies=self.cookies)
        soup = BeautifulSoup(res.text, 'html.parser')
        general_info = soup.tr.find_all()
        course_info = soup.find_all('table')[2].find_all('tr')
        ret = dict()
        ret['total_credits'] = float(general_info[1].text)
        ret['obtained_credits'] = float(general_info[3].text)
        ret['not_obtained_credits'] = float(general_info[5].text)
        ret['failed_credits'] = float(general_info[7].text)
        ret['GPA'] = float(general_info[9].text)
        ret['average'] = float(general_info[11].text)
        ret['weighted_average'] = float(general_info[13].text)
        courses = []
        for course in course_info[:-1]:    # the last line is empty
            cols = course.find_all('td')
            courses.append({'semester': cols[0].text, 'number': cols[1].text, 'name': cols[2].text,
                            'type': cols[3].text, 'grade': cols[4].text, 'credits': cols[5].text})
        ret['courses'] = courses
        return ret


def main():
    "Shut up."
    passport = Passport('你的学号', '你的密码')
    mis = MIS(passport)
    print(mis.grades())

if __name__ == '__main__':
    main()