Refactor file consumption task to allow beginnings of a plugin system (#5367)

This commit is contained in:
Trenton H
2024-01-13 08:11:14 -08:00
committed by GitHub
parent 4dbf8d7969
commit 2da5e46386
11 changed files with 767 additions and 531 deletions

View File

@@ -9,6 +9,7 @@ from os import PathLike
from pathlib import Path
from typing import Any
from typing import Callable
from typing import Optional
from typing import Union
from unittest import mock
@@ -23,6 +24,7 @@ from django.test import override_settings
from documents.data_models import ConsumableDocument
from documents.data_models import DocumentMetadataOverrides
from documents.parsers import ParseError
from documents.plugins.helpers import ProgressStatusOptions
def setup_directories():
@@ -146,6 +148,11 @@ def util_call_with_backoff(
class DirectoriesMixin:
"""
Creates and overrides settings for all folders and paths, then ensures
they are cleaned up on exit
"""
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
self.dirs = None
@@ -160,6 +167,10 @@ class DirectoriesMixin:
class FileSystemAssertsMixin:
"""
Utilities for checks various state information of the file system
"""
def assertIsFile(self, path: Union[PathLike, str]):
self.assertTrue(Path(path).resolve().is_file(), f"File does not exist: {path}")
@@ -188,6 +199,11 @@ class FileSystemAssertsMixin:
class ConsumerProgressMixin:
"""
Mocks the Consumer _send_progress, preventing attempts to connect to Redis
and allowing access to its calls for verification
"""
def setUp(self) -> None:
self.send_progress_patcher = mock.patch(
"documents.consumer.Consumer._send_progress",
@@ -310,3 +326,59 @@ class SampleDirMixin:
SAMPLE_DIR = Path(__file__).parent / "samples"
BARCODE_SAMPLE_DIR = SAMPLE_DIR / "barcodes"
class DummyProgressManager:
"""
A dummy handler for progress management that doesn't actually try to
connect to Redis. Payloads are stored for test assertions if needed.
Use it with
mock.patch("documents.tasks.ProgressManager", DummyProgressManager)
"""
def __init__(self, filename: str, task_id: Optional[str] = None) -> None:
self.filename = filename
self.task_id = task_id
print("hello world")
self.payloads = []
def __enter__(self):
self.open()
return self
def __exit__(self, exc_type, exc_val, exc_tb):
self.close()
def open(self) -> None:
pass
def close(self) -> None:
pass
def send_progress(
self,
status: ProgressStatusOptions,
message: str,
current_progress: int,
max_progress: int,
extra_args: Optional[dict[str, Union[str, int]]] = None,
) -> None:
# Ensure the layer is open
self.open()
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,
},
}
if extra_args is not None:
payload["data"].update(extra_args)
self.payloads.append(payload)