CIRCL
/
AIL-framework
mirror of https://github.com/CIRCL/AIL-framework
2
1
Fork 0
Code Issues Releases Wiki Activity

fix: [PgpDump] catch bs4 error

pull/453/head
Terrtia 2019-12-04 10:02:47 +01:00
parent 3ec72b0430
commit 0ec56cf1ed
No known key found for this signature in database
GPG Key ID: 1E1B1F50D84613D0
1 changed files with 12 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
bin/PgpDump.py
View File

@ -41,16 +41,19 @@ def save_in_file(message, pgp_content):
r_serv_db.sadd('pgpdumb:uuid', '{};{}'.format(UUID, message)) r_serv_db.sadd('pgpdumb:uuid', '{};{}'.format(UUID, message))
def remove_html(item_content): def remove_html(item_content):
if bool(BeautifulSoup(item_content, "html.parser").find()): try:
soup = BeautifulSoup(item_content, 'html.parser') if bool(BeautifulSoup(item_content, "html.parser").find()):
# kill all script and style elements soup = BeautifulSoup(item_content, 'html.parser')
for script in soup(["script", "style"]): # kill all script and style elements
script.extract() # remove for script in soup(["script", "style"]):
script.extract() # remove
# get text # get text
text = soup.get_text() text = soup.get_text()
return text return text
else: else:
return item_content
except TypeError:
return item_content return item_content
def extract_all_id(message, item_content, regex=None, is_file=False): def extract_all_id(message, item_content, regex=None, is_file=False):