WEB_KG/html_parser.py

54 lines
1.8 KiB
Python

from bs4 import BeautifulSoup
import re
from urllib.parse import urljoin
import os
class HtmlParser(object):
def _get_new_urls(self, page_url, soup):
new_urls = set()
# /view/123.htm
links = soup.find_all('a', href=re.compile(r'/item/%[A-Z0-9]+'))
for link in links:
new_url = link['href']
new_full_url = urljoin(page_url, new_url)
# print(new_full_url)
new_urls.add(new_full_url)
# print(new_urls)
return new_urls
def _save_new_data(self, page_url, soup,html_cont):
res_data = False
if '?force=1' in page_url:
#multiple items
print('multiple items')
return not res_data
#error items
title_node =''
try:
title_node=soup.find('dd', class_='lemmaWgt-lemmaTitle-title').find('h1')
except:
return not res_data
#second title
title_sub__text=''
try:
title_sub__text = soup.find('dd', class_='lemmaWgt-lemmaTitle-title').find('h2').get_text()
except:
title_sub__text=''
filename = title_node.get_text() + title_sub__text
if not os.path.exists('/data/ruben/data/webpages/'):
os.mkdir('/data/ruben/data/webpages/')
with open('/data/ruben/data/webpages/' + filename.replace('/',''), 'w') as f:
f.write(html_cont.decode('utf-8'))
print('Save to disk filename:'+f.name+"")
return res_data
def parse(self, page_url, html_cont):
if page_url is None or html_cont is None:
return
soup = BeautifulSoup(html_cont, 'html.parser')
# print(soup.prettify())
new_urls = self._get_new_urls(page_url, soup)
is_saved = self._save_new_data(page_url, soup,html_cont)
return new_urls, is_saved