Feature: Initial document parser plugin framework (#12294)

This commit is contained in:
Trenton H authored and GitHub committed 2026-03-12 21:53:17 +00:00
1 parent c2e1085418
commit d86cfdb088
20 files changed
+2218 -160

No files matched your search

+17
View File
@@ -1,6 +1,7 @@
import os
from celery import Celery
from celery.signals import worker_process_init
# Set the default Django settings module for the 'celery' program.
os.environ.setdefault("DJANGO_SETTINGS_MODULE", "paperless.settings")
@@ -15,3 +16,19 @@ app.config_from_object("django.conf:settings", namespace="CELERY")
# Load task modules from all registered Django apps.
app.autodiscover_tasks()
@worker_process_init.connect
def on_worker_process_init(**kwargs) -> None: # pragma: no cover
"""
Register built-in parsers eagerly in each Celery worker process.
This registers only the built-in parsers (no entrypoint discovery) so
that workers can begin consuming documents immediately. Entrypoint
discovery for third-party parsers is deferred to the first call of
get_parser_registry() inside a task, keeping worker_process_init
well within its 4-second timeout budget.
"""
from paperless.parsers.registry import init_builtin_parsers
init_builtin_parsers()