laramies/theHarvester
1
1
Fork 0
mirror of https://github.com/laramies/theHarvester.git synced 2025-02-25 06:53:05 +08:00
Code Issues Projects Releases Packages Wiki Activity

Improved parser to fix improper emails.

Browse source
This commit is contained in:
NotoriousRebel 2019-02-15 12:37:06 -05:00
parent 560c5ad094
commit 53562a25bc
1 changed files with 10 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

11
discovery/yahoosearch.py
View file

@ -33,7 +33,16 @@ def process(self):
def get_emails(self):
rawres = myparser.Parser(self.total_results, self.word)
return rawres.emails()
toparse_emails = rawres.emails()
emails = set()
# strip out numbers and dashes for emails that look like xxx-xxx-xxxemail@host.tld
for email in toparse_emails:
email = str(email)
if '-' in email and email[0].isdigit() and email.index('-') <= 9:
while email[0] == '-' or email[0].isdigit():
email = email[1:]
emails.add(email)
return list(emails)
def get_hostnames(self):
rawres = myparser.Parser(self.total_results, self.word)