Enhancement: “live” document updates (#12141)

This commit is contained in:
shamoon authored and GitHub committed 2026-03-04 00:27:07 +00:00
1 parent 5498503d60
commit 299dac21ee
17 files changed
+648 -70

No files matched your search

+2
View File
@@ -15,6 +15,7 @@ class DocumentsConfig(AppConfig):
from documents.signals.handlers import add_to_index
from documents.signals.handlers import run_workflows_added
from documents.signals.handlers import run_workflows_updated
from documents.signals.handlers import send_websocket_document_updated
from documents.signals.handlers import set_correspondent
from documents.signals.handlers import set_document_type
from documents.signals.handlers import set_storage_path
@@ -29,6 +30,7 @@ class DocumentsConfig(AppConfig):
document_consumption_finished.connect(run_workflows_added)
document_consumption_finished.connect(add_or_update_document_in_llm_index)
document_updated.connect(run_workflows_updated)
document_updated.connect(send_websocket_document_updated)
import documents.schema # noqa: F401
+38 -13
View File
@@ -1,4 +1,5 @@
import enum
from collections.abc import Mapping
from typing import TYPE_CHECKING
from asgiref.sync import async_to_sync
@@ -47,7 +48,7 @@ class BaseStatusManager:
async_to_sync(self._channel.flush)
self._channel = None
def send(self, payload: dict[str, str | int | None]) -> None:
def send(self, payload: Mapping[str, object]) -> None:
# Ensure the layer is open
self.open()
@@ -73,26 +74,28 @@ class ProgressManager(BaseStatusManager):
max_progress: int,
extra_args: dict[str, str | int | None] | None = None,
) -> None:
payload = {
"type": "status_update",
"data": {
"filename": self.filename,
"task_id": self.task_id,
"current_progress": current_progress,
"max_progress": max_progress,
"status": status,
"message": message,
},
data: dict[str, object] = {
"filename": self.filename,
"task_id": self.task_id,
"current_progress": current_progress,
"max_progress": max_progress,
"status": status,
"message": message,
}
if extra_args is not None:
payload["data"].update(extra_args)
data.update(extra_args)
payload: dict[str, object] = {
"type": "status_update",
"data": data,
}
self.send(payload)
class DocumentsStatusManager(BaseStatusManager):
def send_documents_deleted(self, documents: list[int]) -> None:
payload = {
payload: dict[str, object] = {
"type": "documents_deleted",
"data": {
"documents": documents,
@@ -100,3 +103,25 @@ class DocumentsStatusManager(BaseStatusManager):
}
self.send(payload)
def send_document_updated(
self,
*,
document_id: int,
modified: str,
owner_id: int | None = None,
users_can_view: list[int] | None = None,
groups_can_view: list[int] | None = None,
) -> None:
payload: dict[str, object] = {
"type": "document_updated",
"data": {
"document_id": document_id,
"modified": modified,
"owner_id": owner_id,
"users_can_view": users_can_view or [],
"groups_can_view": groups_can_view or [],
},
}
self.send(payload)
+30 -1
View File
@@ -24,6 +24,7 @@ from django.db.models import Q
from django.dispatch import receiver
from django.utils import timezone
from filelock import FileLock
from rest_framework import serializers
from documents import matching
from documents.caching import clear_document_caches
@@ -48,6 +49,7 @@ from documents.models import WorkflowAction
from documents.models import WorkflowRun
from documents.models import WorkflowTrigger
from documents.permissions import get_objects_for_user_owner_aware
from documents.plugins.helpers import DocumentsStatusManager
from documents.templating.utils import convert_format_str_to_template_format
from documents.workflows.actions import build_workflow_action_context
from documents.workflows.actions import execute_email_action
@@ -69,6 +71,7 @@ if TYPE_CHECKING:
from documents.data_models import DocumentMetadataOverrides
logger = logging.getLogger("paperless.handlers")
DRF_DATETIME_FIELD = serializers.DateTimeField()
def add_inbox_tags(sender, document: Document, logging_group=None, **kwargs) -> None:
@@ -764,6 +767,28 @@ def run_workflows_updated(
)
def send_websocket_document_updated(
sender,
document: Document,
**kwargs,
) -> None:
# At this point, workflows may already have applied additional changes.
document.refresh_from_db()
from documents.data_models import DocumentMetadataOverrides
doc_overrides = DocumentMetadataOverrides.from_document(document)
with DocumentsStatusManager() as status_mgr:
status_mgr.send_document_updated(
document_id=document.id,
modified=DRF_DATETIME_FIELD.to_representation(document.modified),
owner_id=doc_overrides.owner_id,
users_can_view=doc_overrides.view_users,
groups_can_view=doc_overrides.view_groups,
)
def run_workflows(
trigger_type: WorkflowTrigger.WorkflowTriggerType,
document: Document | ConsumableDocument,
@@ -1045,7 +1070,11 @@ def add_or_update_document_in_llm_index(sender, document, **kwargs):
@receiver(models.signals.post_delete, sender=Document)
def delete_document_from_llm_index(sender, instance: Document, **kwargs):
def delete_document_from_llm_index(
sender: Any,
instance: Document,
**kwargs: Any,
) -> None:
"""
Delete a document from the LLM index when it is deleted.
"""
+6
View File
@@ -62,6 +62,7 @@ from documents.sanity_checker import SanityCheckFailedException
from documents.signals import document_updated
from documents.signals.handlers import cleanup_document_deletion
from documents.signals.handlers import run_workflows
from documents.signals.handlers import send_websocket_document_updated
from documents.workflows.utils import get_workflows_for_trigger
from paperless.config import AIConfig
from paperless_ai.indexing import llm_index_add_or_update_document
@@ -572,6 +573,11 @@ def check_scheduled_workflows() -> None:
workflow_to_run=workflow,
document=document,
)
# Scheduled workflows dont send document_updated signal, so send a websocket update here to ensure clients are updated
send_websocket_document_updated(
sender=None,
document=document,
)
def update_document_parent_tags(tag: Tag, new_parent: Tag) -> None:
+10 -2
View File
@@ -1270,7 +1270,11 @@ class TestDocumentApi(DirectoriesMixin, DocumentConsumeDelayMixin, APITestCase):
input_doc, overrides = self.get_last_consume_delay_call_args()
self.assertEqual(input_doc.original_file.name, "simple.pdf")
self.assertIn(Path(settings.SCRATCH_DIR), input_doc.original_file.parents)
self.assertTrue(
input_doc.original_file.resolve(strict=False).is_relative_to(
Path(settings.SCRATCH_DIR).resolve(strict=False),
),
)
self.assertIsNone(overrides.title)
self.assertIsNone(overrides.correspondent_id)
self.assertIsNone(overrides.document_type_id)
@@ -1351,7 +1355,11 @@ class TestDocumentApi(DirectoriesMixin, DocumentConsumeDelayMixin, APITestCase):
input_doc, overrides = self.get_last_consume_delay_call_args()
self.assertEqual(input_doc.original_file.name, "simple.pdf")
self.assertIn(Path(settings.SCRATCH_DIR), input_doc.original_file.parents)
self.assertTrue(
input_doc.original_file.resolve(strict=False).is_relative_to(
Path(settings.SCRATCH_DIR).resolve(strict=False),
),
)
self.assertIsNone(overrides.title)
self.assertIsNone(overrides.correspondent_id)
self.assertIsNone(overrides.document_type_id)
+33 -1
View File
@@ -643,7 +643,9 @@ class TestWorkflows(
expected_str = f"Document did not match {w}"
self.assertIn(expected_str, cm.output[0])
expected_str = f"Document path {test_file} does not match"
expected_str = (
f"Document path {Path(test_file).resolve(strict=False)} does not match"
)
self.assertIn(expected_str, cm.output[1])
def test_workflow_no_match_mail_rule(self) -> None:
@@ -2010,6 +2012,36 @@ class TestWorkflows(
doc.refresh_from_db()
self.assertEqual(doc.owner, self.user2)
@mock.patch("documents.tasks.send_websocket_document_updated")
def test_workflow_scheduled_trigger_sends_websocket_update(
self,
mock_send_websocket_document_updated,
) -> None:
trigger = WorkflowTrigger.objects.create(
type=WorkflowTrigger.WorkflowTriggerType.SCHEDULED,
schedule_offset_days=1,
schedule_date_field=WorkflowTrigger.ScheduleDateField.CREATED,
)
action = WorkflowAction.objects.create(assign_owner=self.user2)
workflow = Workflow.objects.create(name="Workflow 1", order=0)
workflow.triggers.add(trigger)
workflow.actions.add(action)
doc = Document.objects.create(
title="sample test",
correspondent=self.c,
original_filename="sample.pdf",
created=timezone.now() - timedelta(days=2),
)
tasks.check_scheduled_workflows()
self.assertEqual(mock_send_websocket_document_updated.call_count, 1)
self.assertEqual(
mock_send_websocket_document_updated.call_args.kwargs["document"].pk,
doc.pk,
)
def test_workflow_scheduled_trigger_added(self) -> None:
"""
GIVEN:
+8
View File
@@ -1,4 +1,5 @@
import json
from typing import Any
from asgiref.sync import async_to_sync
from channels.exceptions import AcceptConnection
@@ -52,3 +53,10 @@ class StatusConsumer(WebsocketConsumer):
self.close()
else:
self.send(json.dumps(event))
def document_updated(self, event: Any) -> None:
if not self._authenticated():
self.close()
else:
if self._can_view(event["data"]):
self.send(json.dumps(event))
+13 -7
View File
@@ -493,18 +493,24 @@ TEMPLATES = [
},
]
_CHANNELS_BACKEND = os.environ.get(
"PAPERLESS_CHANNELS_BACKEND",
"channels_redis.pubsub.RedisPubSubChannelLayer",
)
CHANNEL_LAYERS = {
"default": {
"BACKEND": "channels_redis.pubsub.RedisPubSubChannelLayer",
"CONFIG": {
"hosts": [_CHANNELS_REDIS_URL],
"capacity": 2000, # default 100
"expiry": 15, # default 60
"prefix": _REDIS_KEY_PREFIX,
},
"BACKEND": _CHANNELS_BACKEND,
},
}
if _CHANNELS_BACKEND.startswith("channels_redis."):
CHANNEL_LAYERS["default"]["CONFIG"] = {
"hosts": [_CHANNELS_REDIS_URL],
"capacity": 2000, # default 100
"expiry": 15, # default 60
"prefix": _REDIS_KEY_PREFIX,
}
###############################################################################
# Email (SMTP) Backend #
###############################################################################
+52 -1
View File
@@ -48,6 +48,20 @@ class TestWebSockets(TestCase):
mock_close.assert_called_once()
mock_close.reset_mock()
message = {
"type": "document_updated",
"data": {"document_id": 10, "modified": "2026-02-17T00:00:00Z"},
}
await channel_layer.group_send(
"status_updates",
message,
)
await communicator.receive_nothing()
mock_close.assert_called_once()
mock_close.reset_mock()
message = {"type": "documents_deleted", "data": {"documents": [1, 2, 3]}}
await channel_layer.group_send(
@@ -158,6 +172,40 @@ class TestWebSockets(TestCase):
await communicator.disconnect()
@mock.patch("paperless.consumers.StatusConsumer._can_view")
@mock.patch("paperless.consumers.StatusConsumer._authenticated")
async def test_receive_document_updated(self, _authenticated, _can_view) -> None:
_authenticated.return_value = True
_can_view.return_value = True
communicator = WebsocketCommunicator(application, "/ws/status/")
connected, _ = await communicator.connect()
self.assertTrue(connected)
message = {
"type": "document_updated",
"data": {
"document_id": 10,
"modified": "2026-02-17T00:00:00Z",
"owner_id": 1,
"users_can_view": [1],
"groups_can_view": [],
},
}
channel_layer = get_channel_layer()
assert channel_layer is not None
await channel_layer.group_send(
"status_updates",
message,
)
response = await communicator.receive_json_from()
self.assertEqual(response, message)
await communicator.disconnect()
@mock.patch("channels.layers.InMemoryChannelLayer.group_send")
def test_manager_send_progress(self, mock_group_send) -> None:
with ProgressManager(task_id="test") as manager:
@@ -190,7 +238,10 @@ class TestWebSockets(TestCase):
)
@mock.patch("channels.layers.InMemoryChannelLayer.group_send")
def test_manager_send_documents_deleted(self, mock_group_send) -> None:
def test_manager_send_documents_deleted(
self,
mock_group_send: mock.MagicMock,
) -> None:
with DocumentsStatusManager() as manager:
manager.send_documents_deleted([1, 2, 3])