Advertisement
Not a member of Pastebin yet?
Sign Up,
it unlocks many cool features!
- from PyQt5.QtWebEngineWidgets import QWebEnginePage
- from PyQt5.QtWidgets import QApplication
- from PyQt5.QtCore import QUrl
- #from PyQt5.QtWebEngineWidgets import QtCore, QtGui, QtWebKit
- import bs4 as bs
- import sys
- import re
- import time
- class Page(QWebEnginePage):
- def __init__(self, url):
- self.app = QApplication(sys.argv)
- QWebEnginePage.__init__(self)
- self.loadd(url)
- def loadd(self, url):
- self.html = ''
- self.loadFinished.connect(self._on_load_finished)
- self.load(QUrl(url))
- self.app.exec_()
- def _on_load_finished(self):
- self.html = self.toHtml(self.Callable)
- print('Load finished')
- def Callable(self, html_str):
- self.html = html_str
- self.app.quit()
- """
- def loadPage(url):
- page = QtWebKit.QWebPage()
- loop = QtCore.QEventLoop() # Create event loop
- page.mainFrame().loadFinished.connect(loop.quit) # Connect loadFinished to loop quit
- page.mainFrame().load(url)
- loop.exec_() # Run event loop, it will end on loadFinished
- return page.mainFrame().toHtml()
- """
- def printShit(page):
- soup = bs.BeautifulSoup(page.html, 'html.parser')
- myTable = soup.findAll('table')
- leTable = myTable[0]
- for table in myTable:
- leTable = table
- if table.find(text=re.compile("Current International Class")):
- leTable = table
- print('-----------------------------------------------------------------------------------------------------------------------------------')
- break
- youza = leTable.findAll('td')
- youza2 = youza[5]
- youza3 = str(youza2)
- youza4 = youza3[1:len(youza3)]
- begin = youza4.find('>')
- end = youza4.find('<')
- print(youza4[begin+1:end])
- urls = []
- poop = open("links.txt", "r")
- urls += poop.readlines()
- poop.close()
- urls = list(set(urls))
- page = Page(urls[0])
- #count = 0
- for url in urls:
- #time.sleep(1)
- page.loadd(url)
- #loadPage(url)
- #time.sleep(2)
- #page = Page(url)
- printShit(page)
- print()
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement