Better, attempt removal later for ConsumableDocument

This commit is contained in:
shamoon
2025-12-30 12:30:34 -08:00
parent f7a6f79c8b
commit 880b3e6d15
3 changed files with 82 additions and 6 deletions

View File

@@ -793,11 +793,7 @@ def run_workflows(
logging_group,
original_file,
)
elif (
action.type == WorkflowAction.WorkflowActionType.PASSWORD_REMOVAL
and not use_overrides
):
# Password removal only makes sense on actual documents
elif action.type == WorkflowAction.WorkflowActionType.PASSWORD_REMOVAL:
execute_password_removal_action(action, document, logging_group)
if not use_overrides:

View File

@@ -2,6 +2,7 @@ import datetime
import json
import shutil
import socket
import tempfile
from datetime import timedelta
from pathlib import Path
from typing import TYPE_CHECKING
@@ -60,6 +61,7 @@ from documents.tests.utils import DirectoriesMixin
from documents.tests.utils import DummyProgressManager
from documents.tests.utils import FileSystemAssertsMixin
from documents.tests.utils import SampleDirMixin
from documents.workflows.actions import execute_password_removal_action
from paperless_mail.models import MailAccount
from paperless_mail.models import MailRule
@@ -3641,6 +3643,68 @@ class TestWorkflows(
mock_remove_password.assert_not_called()
@mock.patch("documents.bulk_edit.remove_password")
def test_password_removal_consumable_document_deferred(
self,
mock_remove_password,
):
action = WorkflowAction.objects.create(
type=WorkflowAction.WorkflowActionType.PASSWORD_REMOVAL,
passwords="first, second",
)
temp_dir = Path(tempfile.mkdtemp())
original_file = temp_dir / "file.pdf"
original_file.write_bytes(b"pdf content")
consumable = ConsumableDocument(
source=DocumentSource.ApiUpload,
original_file=original_file,
)
execute_password_removal_action(action, consumable, logging_group=None)
mock_remove_password.assert_not_called()
mock_remove_password.side_effect = [
ValueError("bad password"),
"OK",
]
doc = Document.objects.create(
checksum="pw-checksum-consumed",
title="Protected",
)
document_consumption_finished.send(
sender=self.__class__,
document=doc,
)
assert mock_remove_password.call_count == 2
mock_remove_password.assert_has_calls(
[
mock.call(
[doc.id],
password="first",
update_document=True,
user=doc.owner,
),
mock.call(
[doc.id],
password="second",
update_document=True,
user=doc.owner,
),
],
)
# ensure handler disconnected after first run
document_consumption_finished.send(
sender=self.__class__,
document=doc,
)
assert mock_remove_password.call_count == 2
class TestWebhookSend:
def test_send_webhook_data_or_json(

View File

@@ -15,6 +15,7 @@ from documents.models import Document
from documents.models import DocumentType
from documents.models import WorkflowAction
from documents.models import WorkflowTrigger
from documents.signals import document_consumption_finished
from documents.templating.workflows import parse_w_workflow_placeholders
from documents.workflows.webhooks import send_webhook
@@ -264,7 +265,7 @@ def execute_webhook_action(
def execute_password_removal_action(
action: WorkflowAction,
document: Document,
document: Document | ConsumableDocument,
logging_group,
) -> None:
"""
@@ -285,6 +286,21 @@ def execute_password_removal_action(
if password.strip()
]
if isinstance(document, ConsumableDocument):
# hook the consumption-finished signal to attempt password removal later
def handler(sender, **kwargs):
consumed_document: Document = kwargs.get("document")
if consumed_document is not None:
execute_password_removal_action(
action,
consumed_document,
logging_group,
)
document_consumption_finished.disconnect(handler)
document_consumption_finished.connect(handler, weak=False)
return
# import here to avoid circular dependency
from documents.bulk_edit import remove_password