2021-03-12 16:53:00 +01:00
|
|
|
#!/usr/bin/env python3
|
|
|
|
|
2024-01-12 17:15:41 +01:00
|
|
|
from __future__ import annotations
|
|
|
|
|
2021-03-12 16:53:00 +01:00
|
|
|
import logging
|
2022-11-23 15:54:22 +01:00
|
|
|
import logging.config
|
2024-08-29 13:32:38 +02:00
|
|
|
from pathlib import Path
|
2021-03-12 16:53:00 +01:00
|
|
|
|
2024-03-09 15:33:10 +01:00
|
|
|
from redis import Redis
|
|
|
|
|
2024-08-29 13:32:38 +02:00
|
|
|
from lookyloo import Indexing
|
2024-03-09 15:33:10 +01:00
|
|
|
from lookyloo.default import AbstractManager, get_config, get_socket_path
|
2021-03-12 16:53:00 +01:00
|
|
|
|
2022-11-23 15:54:22 +01:00
|
|
|
|
|
|
|
logging.config.dictConfig(get_config('logging'))
|
2021-03-12 16:53:00 +01:00
|
|
|
|
|
|
|
|
|
|
|
class BackgroundIndexer(AbstractManager):
|
|
|
|
|
2024-03-05 20:51:21 +01:00
|
|
|
def __init__(self, full: bool=False, loglevel: int | None=None):
|
2021-03-12 16:53:00 +01:00
|
|
|
super().__init__(loglevel)
|
2024-03-09 15:33:10 +01:00
|
|
|
self.is_public_instance = get_config('generic', 'public_instance')
|
2024-03-05 20:51:21 +01:00
|
|
|
self.full_indexer = full
|
|
|
|
self.indexing = Indexing(full_index=self.full_indexer)
|
|
|
|
if self.full_indexer:
|
|
|
|
self.script_name = 'background_full_indexer'
|
|
|
|
else:
|
|
|
|
self.script_name = 'background_indexer'
|
2021-03-12 16:53:00 +01:00
|
|
|
|
2024-03-09 15:33:10 +01:00
|
|
|
# Redis connector so we don't use the one from Lookyloo
|
|
|
|
self.redis = Redis(unix_socket_path=get_socket_path('cache'), decode_responses=True)
|
|
|
|
|
2024-01-12 17:15:41 +01:00
|
|
|
def _to_run_forever(self) -> None:
|
2024-03-05 20:51:21 +01:00
|
|
|
self._check_indexes()
|
2021-03-12 16:53:00 +01:00
|
|
|
|
2024-08-29 13:32:38 +02:00
|
|
|
def _check_indexes(self) -> None:
|
|
|
|
if not self.indexing.can_index():
|
|
|
|
# There is no reason to run this method in multiple scripts.
|
|
|
|
self.logger.info('Indexing already ongoing in another process.')
|
|
|
|
return None
|
|
|
|
self.logger.info(f'Check {self.script_name}...')
|
2024-03-09 15:33:10 +01:00
|
|
|
# NOTE: only get the non-archived captures for now.
|
2024-09-27 12:26:34 +02:00
|
|
|
__counter_shutdown = 0
|
2024-08-29 13:32:38 +02:00
|
|
|
for uuid, d in self.redis.hscan_iter('lookup_dirs'):
|
2024-03-05 20:51:21 +01:00
|
|
|
if not self.full_indexer:
|
|
|
|
# If we're not running the full indexer, check if the capture should be indexed.
|
2024-08-29 13:32:38 +02:00
|
|
|
if self.is_public_instance and self.redis.hexists(d, 'no_index'):
|
2024-03-05 20:51:21 +01:00
|
|
|
# Capture unindexed
|
|
|
|
continue
|
2024-09-27 12:26:34 +02:00
|
|
|
__counter_shutdown += 1
|
2024-08-29 13:32:38 +02:00
|
|
|
self.indexing.index_capture(uuid, Path(d))
|
2024-09-27 12:26:34 +02:00
|
|
|
if __counter_shutdown % 10 and self.shutdown_requested():
|
|
|
|
self.logger.warning('Shutdown requested, breaking.')
|
|
|
|
break
|
|
|
|
else:
|
|
|
|
self.logger.info('... done.')
|
2024-03-05 20:51:21 +01:00
|
|
|
self.indexing.indexing_done()
|
2024-02-26 17:07:23 +01:00
|
|
|
|
2021-03-12 16:53:00 +01:00
|
|
|
|
2024-01-12 17:15:41 +01:00
|
|
|
def main() -> None:
|
2021-03-12 16:53:00 +01:00
|
|
|
i = BackgroundIndexer()
|
|
|
|
i.run(sleep_in_sec=60)
|
|
|
|
|
|
|
|
|
2024-03-05 20:51:21 +01:00
|
|
|
def main_full_indexer() -> None:
|
|
|
|
if not get_config('generic', 'index_everything'):
|
|
|
|
raise Exception('Full indexer is disabled.')
|
2024-03-08 10:36:04 +01:00
|
|
|
# NOTE: for now, it only indexes the captures that aren't archived.
|
|
|
|
# we will change that later, but for now, it's a good start.
|
2024-03-05 20:51:21 +01:00
|
|
|
i = BackgroundIndexer(full=True)
|
|
|
|
i.run(sleep_in_sec=60)
|
|
|
|
|
|
|
|
|
2021-03-12 16:53:00 +01:00
|
|
|
if __name__ == '__main__':
|
|
|
|
main()
|