nebula-rss/nebula_rss/nebula_loader.py

95 lines
3.5 KiB
Python
Raw Normal View History

2022-01-11 14:17:33 +00:00
import datetime
2022-01-11 14:51:04 +00:00
import logging
2022-01-11 13:09:24 +00:00
import os
import re
import time
2022-01-11 14:17:33 +00:00
from typing import List, Optional
2022-01-11 13:09:24 +00:00
from selenium import webdriver
from selenium.webdriver.common.by import By
from selenium.webdriver.support import expected_conditions as EC
from selenium.webdriver.firefox.options import Options
from selenium.webdriver.firefox.service import Service
from selenium.webdriver.support.wait import WebDriverWait
2022-01-11 14:51:04 +00:00
from selenium.webdriver.remote.remote_connection import LOGGER
2022-01-11 13:09:24 +00:00
from bs4 import BeautifulSoup
from nebula_rss import NebulaVideo
class NebulaLoader:
def __init__(
self,
username: str,
password: str,
driver_path: Optional[os.PathLike] = None
):
self.username = username
self.password = password
2022-01-11 14:51:04 +00:00
LOGGER.setLevel(logging.FATAL)
2022-01-11 13:09:24 +00:00
service = None
if driver_path:
2022-01-11 14:51:04 +00:00
service = Service(driver_path)
2022-01-11 13:09:24 +00:00
options = Options()
options.headless = True
2022-01-11 14:51:04 +00:00
self.driver = webdriver.Firefox(
service=service,
options=options,
log_path=os.devnull,
service_log_path=os.devnull)
2022-01-11 13:09:24 +00:00
self.driver.implicitly_wait(10) # seconds
2022-01-11 14:17:33 +00:00
@staticmethod
def _parse_anchor(anchor) -> NebulaVideo:
info_div = anchor.next_sibling
details_anchor = info_div.find_all('a')[1]
divs = details_anchor.find_all('div')
title_div = divs[0]
details_div = divs[1]
creator = details_div.find('span').string
release_text = details_div.find('time').get('datetime')
release_date = datetime.datetime.fromisoformat(release_text.replace('Z', '+00:00'))
return NebulaVideo(
title=title_div.string,
creator=creator,
url='https://nebula.app' + anchor.get('href'),
release_at=release_date
)
def load(self) -> List[NebulaVideo]:
2022-01-11 13:09:24 +00:00
self.driver.get('https://nebula.app/login')
username_input = '//*[@name="email"]'
password_input = '//*[@name="password"]'
login_submit = '//*[@id="NebulaApp"]/div[2]/div[2]/div[1]/div/form/button'
self.driver.find_element(By.XPATH, username_input).send_keys(self.username)
self.driver.find_element(By.XPATH, password_input).send_keys(self.password)
self.driver.find_element(By.XPATH, login_submit).click()
delay = 3
# wait for "My shows" link
wait = WebDriverWait(self.driver, delay)
myshows = wait.until(EC.presence_of_element_located((By.CSS_SELECTOR, 'menu a[href|="/myshows"]')))
myshows.click()
myshows = wait.until(EC.presence_of_element_located((By.CSS_SELECTOR, 'h2')))
video_links = []
count_remaining = 5
follower_error_re = re.compile("You aren't following any creators yet.*")
while not video_links and count_remaining > 0:
time.sleep(2)
2022-01-11 14:17:33 +00:00
soup = BeautifulSoup(self.driver.page_source, features="lxml")
2022-01-11 13:09:24 +00:00
follower_error = False
follower_error = [p for p in soup.find_all('p') if p.find(string=follower_error_re)]
if follower_error:
print('Error loading videos, reloading page')
self.driver.refresh()
count_remaining = 5
else:
2022-01-11 14:17:33 +00:00
all_anchors = soup.find_all('a')
video_links = [a for a in all_anchors if a.get('href').startswith('/videos/') and a.get('aria-hidden')]
2022-01-11 13:09:24 +00:00
count_remaining -= 1
2022-01-11 14:17:33 +00:00
return [NebulaLoader._parse_anchor(v) for v in video_links]