paradizelost/paperless-ngx
1
0
Fork 0
mirror of https://github.com/paperless-ngx/paperless-ngx.git synced 2025-10-30 03:56:23 -05:00
Code Issues Packages Projects Releases Wiki Activity

Make ignores configurable

Browse Source 
Adds config file setting PAPERLESS_CONSUMER_IGNORE_PATTERNS.
This commit is contained in:
Daniel Albers
2021-08-08 21:29:36 +02:00
parent 566b3ed53d
commit 3ebe6d5aef
4 changed files with 29 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
src/documents/management/commands/document_consumer.py
View File

@@ -1,6 +1,6 @@
import logging
import os
from pathlib import Path
from pathlib import Path, PurePath
from threading import Thread
from time import sleep
@@ -36,15 +36,11 @@ def _tags_from_path(filepath):
return tag_ids
def _is_ignored(filepath):
# https://github.com/jonaswinkler/paperless-ng/discussions/1037
basename = os.path.basename(filepath)
if basename == ".DS_STORE":
return True
if basename.startswith("._"):
return True
return False
def _is_ignored(filepath: str) -> bool:
filepath_relative = PurePath(filepath).relative_to(
settings.CONSUMPTION_DIR)
return any(
filepath_relative.match(p) for p in settings.CONSUMER_IGNORE_PATTERNS)
def _consume(filepath):

16
src/documents/tests/test_management_consumer.py
View File

@@ -222,6 +222,22 @@ class TestConsumer(DirectoriesMixin, ConsumerMixin, TransactionTestCase):
fnames = [os.path.basename(args[1]) for args, _ in self.task_mock.call_args_list]
self.assertCountEqual(fnames, ["my_file.pdf", "my_second_file.pdf"])
def test_is_ignored(self):
test_paths = [
(os.path.join(self.dirs.consumption_dir, "foo.pdf"), False),
(os.path.join(self.dirs.consumption_dir, "foo","bar.pdf"), False),
(os.path.join(self.dirs.consumption_dir, ".DS_STORE", "foo.pdf"), True),
(os.path.join(self.dirs.consumption_dir, "foo", ".DS_STORE", "bar.pdf"), True),
(os.path.join(self.dirs.consumption_dir, ".stfolder", "foo.pdf"), True),
(os.path.join(self.dirs.consumption_dir, "._foo.pdf"), True),
(os.path.join(self.dirs.consumption_dir, "._foo", "bar.pdf"), False),
]
for file_path, expected_ignored in test_paths:
self.assertEqual(
expected_ignored,
document_consumer._is_ignored(file_path),
f'_is_ignored("{file_path}") != {expected_ignored}')
@override_settings(CONSUMER_POLLING=1, CONSUMER_POLLING_DELAY=1, CONSUMER_POLLING_RETRY_COUNT=20)
class TestConsumerPolling(TestConsumer):