Fix tineye engine url, datetime parsing, and minor refactor

Changes made to tineye engine:
1. Importing logging if TYPE_CHECKING is enabled
2. Remove unecessary try-catch around json parsing the response, as this
masked the original error and had no immediate benefit
3. Improve error handling explicitely for status code 422 and 400
upfront, deferring json_parsing only for these status codes and
successful status codes
4. Unit test all new applicable changes to ensure compatability
This commit is contained in:
Grant Lanham 2024-08-19 23:02:06 -04:00 committed by Markus Heiser
parent 5be55e3309
commit 5276219b9d
2 changed files with 130 additions and 29 deletions

102
tests/unit/test_tineye.py Normal file
View file

@ -0,0 +1,102 @@
# SPDX-License-Identifier: AGPL-3.0-or-later
# pylint: disable=missing-module-docstring
from datetime import datetime
from unittest.mock import Mock
from requests import HTTPError
from searx.engines import load_engines, tineye
from tests import SearxTestCase
class TinEyeTests(SearxTestCase): # pylint: disable=missing-class-docstring
def setUp(self):
load_engines([{'name': 'tineye', 'engine': 'tineye', 'shortcut': 'tin', 'timeout': 9.0, 'disabled': True}])
def tearDown(self):
load_engines([])
def test_status_code_raises(self):
response = Mock()
response.status_code = 401
response.raise_for_status.side_effect = HTTPError()
self.assertRaises(HTTPError, lambda: tineye.response(response))
def test_returns_empty_list_for_422(self):
response = Mock()
response.json.return_value = {}
response.status_code = 422
response.raise_for_status.side_effect = HTTPError()
with self.assertLogs(tineye.logger) as _dev_null:
results = tineye.response(response)
self.assertEqual(0, len(results))
def test_logs_format_for_422(self):
response = Mock()
response.json.return_value = {"suggestions": {"key": "Invalid image URL"}}
response.status_code = 422
response.raise_for_status.side_effect = HTTPError()
with self.assertLogs(tineye.logger) as assert_logs_context:
tineye.response(response)
self.assertIn(tineye.FORMAT_NOT_SUPPORTED, ','.join(assert_logs_context.output))
def test_logs_signature_for_422(self):
response = Mock()
response.json.return_value = {"suggestions": {"key": "NO_SIGNATURE_ERROR"}}
response.status_code = 422
response.raise_for_status.side_effect = HTTPError()
with self.assertLogs(tineye.logger) as assert_logs_context:
tineye.response(response)
self.assertIn(tineye.NO_SIGNATURE_ERROR, ','.join(assert_logs_context.output))
def test_logs_download_for_422(self):
response = Mock()
response.json.return_value = {"suggestions": {"key": "Download Error"}}
response.status_code = 422
response.raise_for_status.side_effect = HTTPError()
with self.assertLogs(tineye.logger) as assert_logs_context:
tineye.response(response)
self.assertIn(tineye.DOWNLOAD_ERROR, ','.join(assert_logs_context.output))
def test_empty_list_for_400(self):
response = Mock()
response.json.return_value = {}
response.status_code = 400
response.raise_for_status.side_effect = HTTPError()
with self.assertLogs(tineye.logger) as _dev_null:
results = tineye.response(response)
self.assertEqual(0, len(results))
def test_logs_description_for_400(self):
description = 'There was a problem with that request. Error ID: ad5fc955-a934-43c1-8187-f9a61d301645'
response = Mock()
response.json.return_value = {"suggestions": {"description": [description], "title": "Oops! We're sorry!"}}
response.status_code = 400
response.raise_for_status.side_effect = HTTPError()
with self.assertLogs(tineye.logger) as assert_logs_context:
tineye.response(response)
self.assertIn(description, ','.join(assert_logs_context.output))
def test_crawl_date_parses(self):
date_str = '2020-05-25'
date = datetime.strptime(date_str, '%Y-%m-%d')
response = Mock()
response.json.return_value = {
'matches': [
{
'backlinks': [
{
'crawl_date': date_str,
}
]
}
]
}
response.status_code = 200
results = tineye.response(response)
self.assertEqual(date, results[0]['publishedDate'])