imagebam-rewrite.py

# find all imagebam urls with this regex
# https?:\/\/(www\.)?imagebam\.[a-zA-Z0-9()]{1,6}\b([-a-zA-Z0-9()@:%_\+.~#?&//=]*)
# gives us something like
# http://www.imagebam.com/image/9e31091131177484

import requests
from bs4 import BeautifulSoup
from selenium import webdriver

# headless chrome
options = webdriver.ChromeOptions()
options.add_argument('headless')
driver = webdriver.Chrome(options=options)

urls = [
  'http://www.imagebam.com/image/9e31091131177484'
]

bbcodes = dict()

for url in urls:
  #print(url)
  s = requests.Session()
  response = s.get(url)
  if response.status_code == 200:
    # get it again to bypass redirect
    print("Got[1]: " + url)
    response = s.get(url)
    if response.status_code == 200:
      # we good!
      print("Got[2]: " + url)
      #print(response.text)
      
      # parse html for image redirect
      soup = BeautifulSoup(response.text, 'html.parser')
      metatag = soup.find_all('meta')
      for tag in metatag:
        if tag.get("property") == 'og:image':
          direct_url = tag.get("content")
          print("direct url: " + direct_url)
          
          driver.get("https://postimages.org/web")
          form_el = driver.find_element_by_id("uploadForm")
          links_el = driver.find_element_by_id("links")
          links_el.send_keys(direct_url)
          driver.implicitly_wait(10)
          #driver.execute_script("upload();console.log('ddd');")
          driver.find_element_by_css_selector("a.width100").click()
          
          bb_code = driver.find_element_by_id("code_bb_thumb")
          bb_code_val = bb_code.get_attribute("value")
          #print(bb_code_val)
          bbcodes[direct_url] = bb_code_val
          driver.close()
    else:
      print("Failed[2]: " + url)
  else:
    print("Failed[1]: " + url)
    
print(bbcodes)