from selenium import webdriver
from selenium.webdriver.common.by import By
import requests
import re
from time import sleep
download_root= r'C:\Users\Surface\Desktop\scrapy'
driver = webdriver.Chrome(r'C:\Users\Surface\Desktop\scrapy\chromedriver.exe')

def entry():
    driver.get('http://www.wenguitar.com//')
    element = driver.find_element(By.XPATH,".//*[text()='繁體中文']")
    element.click()
    element = driver.find_element(By.XPATH,".//*[text()='登入']")
    element.click()
    element_email = driver.find_element(By.XPATH,".//*[@name='inputEmail']")
    element_email.send_keys("jamesxxx1997@gmail.com")
    element_password = driver.find_element(By.XPATH,".//*[@name='inputPassword']")
    element_password.send_keys("WTF FUCK YOU")
    entry_btn = driver.find_element(By.XPATH,".//*[@class='btn']")
    entry_btn.click()


def count_main_course():
    main_courses = driver.find_elements(By.XPATH,".//*[@class='btn btn-primary']")
    return len(main_courses)


def click_main_course(number):
    main_courses = driver.find_elements(By.XPATH,".//*[@class='btn btn-primary']")
    main_courses[number].click()

def get_main_course_name(number):
    main_courses = driver.find_elements(By.XPATH,".//*[@class='btn btn-primary']")
    return main_courses[number].text


def count_sub_course():
    sub_courses = driver.find_elements(By.XPATH,"(.//*[@class='t1'])//a")
    return len(sub_courses)


def click_sub_course(number):
    sub_courses = driver.find_elements(By.XPATH,"(.//*[@class='t1'])//a")
    sub_courses[number].click()

def get_sub_course_name(number):
    sub_courses = driver.find_elements(By.XPATH,"(.//*[@class='t1'])//a")
    return sub_courses[number].text

####################################################################################

def get_page_source():

    driver.switch_to.frame(0)
    page_source = driver.page_source
    driver.switch_to.default_content()
    return page_source



def check(video_urls):
    p2 = re.compile('"(.*?)"')
    for video_url in video_urls:
        if p2.match(video_url):
            dirty_video_url = p2.match(video_url).group(1)
            return dirty_video_url

def clean_amp(dirty_video_url):
    clean_url = dirty_video_url.replace('amp;','')
    return clean_url

def get_video_url(page_source):
    #2 = '"(https:\/\/.*?[.]mp4)"'
    #1 = '"(https:.*?\.mp4\?expires=\d*?&token=.*?)"'
    regex =  '("(https:.*?\.mp4\?expires=\d*?&token=.*?)"|"(https:\/\/.*?[.]mp4)")'
    p1 = re.compile(regex)
    video_urls = p1.findall(page_source)[-1]

    dirty_video_url = check(video_urls)
    clean_url = clean_amp(dirty_video_url)
    return clean_url

def get_video(main_course_name,sub_course_name,clean_url):
    r = requests.get(clean_url , stream = True)
    filename = download_root + '\\' + main_course_name + sub_course_name + '.mp4'
    with open(filename , "wb") as fout:
        for chunk in r.iter_content(chunk_size=1024):
            if chunk:
                fout.write(chunk)




def video(main_course_name,sub_course_name):
    page_source = get_page_source()
    clean_url = get_video_url(page_source)
    get_video(main_course_name,sub_course_name,clean_url)
    print(clean_url)

###########################################################################

def __main__():
    entry()
    main_courses_number = count_main_course()
    #main_courses_number = 3
    sub_courses_number = count_sub_course()
    #sub_courses_number=2
    for x in range(1,main_courses_number):
        main_course_name = get_main_course_name(x)
        click_main_course(x)
        for y in range(0,sub_courses_number):
            sub_course_name = get_sub_course_name(y)
            click_sub_course(y)
            sleep(2)
            video(main_course_name , sub_course_name)
            sleep(2)
            driver.back()
            sleep(2)





def test_match():
    main_courses_number = 3
    sub_courses_number=2
    count = 0
    count_er = 0
    entry()
    for x in range(1,main_courses_number):
        click_main_course(x)
        count += 1
        for y in range(0,sub_courses_number):
            click_sub_course(y)
            if get_video_url(get_page_source()) == None:
                count_er += 1


__main__()