itsulu/FediFetcher
2
0
Fork 0
mirror of https://github.com/nicholasr-itsulu/FediFetcher.git synced 2026-05-30 23:41:26 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

feat: implement new flag --instance-blocklist

Browse source 
This can also be set via env var `FF_INSTANCE_BLOCKLIST`

It should be a comma seperated list of domains
This commit is contained in:
B Tasker 2024-08-18 16:47:56 +01:00
parent 8ac868fda4
commit 6e1830f87c
No known key found for this signature in database
GPG key ID: 8DC652174C1EBA9B
1 changed files with 7 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

8
find_posts.py
View file

@ -53,6 +53,7 @@ argparser.add_argument('--max-list-length', required=False, type=int, default=10
argparser.add_argument('--max-list-accounts', required=False, type=int, default=10, help="Determines how many accounts we'll backfill for in each list. This will be ignored, unless you also provide `from-lists = 1`. Set to `0` if you only want to fetch replies in lists.")
argparser.add_argument('--log-level', required=False, default="DEBUG", help="Severity of events to log (DEBUG|INFO|WARNING|ERROR|CRITICAL)")
argparser.add_argument('--log-format', required=False, type=str, default="%(asctime)s: %(message)s",help="Specify the log format")
argparser.add_argument('--instance-blocklist', required=False, type=str, default="",help="A comma-seperated array of instances that FediFetcher should never try to connect to")
def get_notification_users(server, access_token, known_users, max_age):
since = datetime.now(datetime.now().astimezone().tzinfo) - timedelta(hours=max_age)
@ -1501,7 +1502,8 @@ if __name__ == "__main__":
"on_done",
"on_fail",
"log_level",
"log_format"
"log_format",
"instance_blocklist"
]:
value = int(value)
setattr(arguments, envvar, value)
@ -1572,8 +1574,12 @@ if __name__ == "__main__":
SEEN_HOSTS_FILE = os.path.join(arguments.state_dir, "seen_hosts")
RECENTLY_CHECKED_CONTEXTS_FILE = os.path.join(arguments.state_dir, 'recent_context')
INSTANCE_BLOCKLIST = arguments.instance_blocklist.split(",")
ROBOTS_TXT = {}
print(INSTANCE_BLOCKLIST)
sys.exit()
seen_urls = OrderedSet([])
if os.path.exists(SEEN_URLS_FILE):
with open(SEEN_URLS_FILE, "r", encoding="utf-8") as f: