gnh1201/caterpillar
1
0
Fork 0
mirror of https://github.com/gnh1201/caterpillar.git synced 2025-02-06 06:55:00 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Update fediverse.py

Browse Source
This commit is contained in:
Namhyeon Go 2024-10-09 03:06:25 +09:00 committed by GitHub
parent c272efe8b1
commit 58e7322555
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 6 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
plugins/fediverse.py
View File

@ -39,6 +39,8 @@ try:
except Exception as e:
logger.error("[*] Invalid configuration", exc_info=e)
# bad reputation domains
bad_domains = ["krsw-wiki.org", "midokuriserver.github.io"]
class Fediverse(Extension):
def __init__(self):
@ -71,6 +73,10 @@ class Fediverse(Extension):
if error_rate > 0.2: # it is a binary data
return False
# check if the text contains any of the bad domains
if bool(re.search(r"https://(" + "|".join(re.escape(domain) for domain in bad_domains) + ")", text)):
return False
# check ID with K-Anonymity strategy
pattern = r"\b(?:(?<=\/@)|(?<=acct:))([a-zA-Z0-9]{10})\b"
matches = list(set(re.findall(pattern, text)))