from base.baseclasses import SendtoDB
from base.baseclasses import print_and_flush
from base.baseclasses import CrawlInit
from base.baseclasses import wait
from base.baseclasses import Browser
from base.baseclasses import enter_element
from selenium.webdriver.common.keys import Keys


def pageup_and_pagedown(_driver):
    # body = _driver.find_element_by_tag_name('body')
    # for i in range(0, 2):
    #     body.send_keys(Keys.PAGE_UP)
    #     wait(3)
    # for i in range(0, 5):
    #     body.send_keys(Keys.PAGE_DOWN)
    #     wait(3)
    for i in range(0, 3):
        _driver.execute_script("window.scrollBy(0, -300)")
        wait(0.4)

    for i in range(0, 5):
        _driver.execute_script("window.scrollBy(0, 800)")
        wait(0.4)

def first_load(_driver):
    element = _driver.find_element_by_css_selector("div._pupj3 > a")
    enter_element(element)


def get_urls(_driver, url_set):
    elements = _driver.find_elements_by_css_selector("div._myci9>a")
    for element in elements:
        url_set.add(element.get_attribute('href'))


def remove_myci9(_driver):
    elements = _driver.find_elements_by_css_selector("div._myci9")
    for i in range(0, len(elements) - 4 if len(elements) - 4 > 0 else 0):
        _driver.execute_script("""
        var element = document.querySelector("div._myci9");
        if (element)
            element.parentNode.removeChild(element);
        """)


browser = Browser()
driver = browser.get_new_driver('ie')

url_sets = set()
wait(5)
url = "https://www.instagram.com/explore/tags/%EC%A4%8C%EB%A7%88%EA%B7%B8%EB%9E%A8/"
#url = 'https://www.instagram.com/explore/tags/%EB%A7%9B%EC%8A%A4%ED%83%80%EA%B7%B8%EB%9E%A8/'
driver.get(url)
#driver.get('https://www.instagram.com/explore/tags/맛스타그램/')
wait(5)


first_load(driver)
wait(3)

#print(driver.get_cookies())

with open("c:\\data\\instajummaie.txt", 'w') as f:
    try:
        while True:
            for j in range(0, 10):
                pageup_and_pagedown(driver)

            get_urls(driver, url_sets)
            remove_myci9(driver)
            print("url count = {0}\n".format(len(url_sets)), flush=True, file=f)
    finally:
        print("finished")