2024-03-11 14:06:26 +01:00
|
|
|
# SPDX-License-Identifier: AGPL-3.0-or-later
|
|
|
|
|
# pylint: disable=missing-module-docstring
|
|
|
|
|
|
2016-05-19 07:38:43 +02:00
|
|
|
from collections import defaultdict
|
|
|
|
|
import mock
|
[fix] unit test_xpath.py: name 'logger' is not defined
Depending on the order in which the unit tests are executed, the python modules
of the engines are initialized (monkey patched) or not. As the order of the
tests is not static, random errors may occur.
To avaoid random `NameError: name 'logger' is not defined` in the unit tests of
the xpath engine, a logger is monkey patched into the xpath py-module.
```
make test.unit
TEST tests/unit
......EE...................
======================================================================
ERROR: test_response (tests.unit.engines.test_xpath.TestXpathEngine.test_response)
----------------------------------------------------------------------
Traceback (most recent call last):
File "./tests/unit/engines/test_xpath.py", line 60, in test_response
self.assertEqual(xpath.response(response), [])
^^^^^^^^^^^^^^^^^^^^^^^^
File "./searx/engines/xpath.py", line 309, in response
logger.debug("found %s results", len(results))
^^^^^^
NameError: name 'logger' is not defined
======================================================================
ERROR: test_response_results_xpath (tests.unit.engines.test_xpath.TestXpathEngine.test_response_results_xpath)
----------------------------------------------------------------------
Traceback (most recent call last):
File "./tests/unit/engines/test_xpath.py", line 102, in test_response_results_xpath
self.assertEqual(xpath.response(response), [])
^^^^^^^^^^^^^^^^^^^^^^^^
File "./searx/engines/xpath.py", line 309, in response
logger.debug("found %s results", len(results))
^^^^^^
NameError: name 'logger' is not defined
```
Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>
2024-06-25 10:49:37 +02:00
|
|
|
|
2016-05-19 07:38:43 +02:00
|
|
|
from searx.engines import xpath
|
[fix] unit test_xpath.py: name 'logger' is not defined
Depending on the order in which the unit tests are executed, the python modules
of the engines are initialized (monkey patched) or not. As the order of the
tests is not static, random errors may occur.
To avaoid random `NameError: name 'logger' is not defined` in the unit tests of
the xpath engine, a logger is monkey patched into the xpath py-module.
```
make test.unit
TEST tests/unit
......EE...................
======================================================================
ERROR: test_response (tests.unit.engines.test_xpath.TestXpathEngine.test_response)
----------------------------------------------------------------------
Traceback (most recent call last):
File "./tests/unit/engines/test_xpath.py", line 60, in test_response
self.assertEqual(xpath.response(response), [])
^^^^^^^^^^^^^^^^^^^^^^^^
File "./searx/engines/xpath.py", line 309, in response
logger.debug("found %s results", len(results))
^^^^^^
NameError: name 'logger' is not defined
======================================================================
ERROR: test_response_results_xpath (tests.unit.engines.test_xpath.TestXpathEngine.test_response_results_xpath)
----------------------------------------------------------------------
Traceback (most recent call last):
File "./tests/unit/engines/test_xpath.py", line 102, in test_response_results_xpath
self.assertEqual(xpath.response(response), [])
^^^^^^^^^^^^^^^^^^^^^^^^
File "./searx/engines/xpath.py", line 309, in response
logger.debug("found %s results", len(results))
^^^^^^
NameError: name 'logger' is not defined
```
Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>
2024-06-25 10:49:37 +02:00
|
|
|
from searx import logger
|
|
|
|
|
|
2021-09-02 16:01:34 +02:00
|
|
|
from tests import SearxTestCase
|
2016-05-19 07:38:43 +02:00
|
|
|
|
[fix] unit test_xpath.py: name 'logger' is not defined
Depending on the order in which the unit tests are executed, the python modules
of the engines are initialized (monkey patched) or not. As the order of the
tests is not static, random errors may occur.
To avaoid random `NameError: name 'logger' is not defined` in the unit tests of
the xpath engine, a logger is monkey patched into the xpath py-module.
```
make test.unit
TEST tests/unit
......EE...................
======================================================================
ERROR: test_response (tests.unit.engines.test_xpath.TestXpathEngine.test_response)
----------------------------------------------------------------------
Traceback (most recent call last):
File "./tests/unit/engines/test_xpath.py", line 60, in test_response
self.assertEqual(xpath.response(response), [])
^^^^^^^^^^^^^^^^^^^^^^^^
File "./searx/engines/xpath.py", line 309, in response
logger.debug("found %s results", len(results))
^^^^^^
NameError: name 'logger' is not defined
======================================================================
ERROR: test_response_results_xpath (tests.unit.engines.test_xpath.TestXpathEngine.test_response_results_xpath)
----------------------------------------------------------------------
Traceback (most recent call last):
File "./tests/unit/engines/test_xpath.py", line 102, in test_response_results_xpath
self.assertEqual(xpath.response(response), [])
^^^^^^^^^^^^^^^^^^^^^^^^
File "./searx/engines/xpath.py", line 309, in response
logger.debug("found %s results", len(results))
^^^^^^
NameError: name 'logger' is not defined
```
Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>
2024-06-25 10:49:37 +02:00
|
|
|
logger = logger.getChild('engines')
|
|
|
|
|
|
2016-05-19 07:38:43 +02:00
|
|
|
|
2024-06-21 15:58:12 +02:00
|
|
|
class TestXpathEngine(SearxTestCase): # pylint: disable=missing-class-docstring
|
|
|
|
|
html = """
|
|
|
|
|
<div>
|
|
|
|
|
<div class="search_result">
|
|
|
|
|
<a class="result" href="https://result1.com">Result 1</a>
|
|
|
|
|
<p class="content">Content 1</p>
|
|
|
|
|
<a class="cached" href="https://cachedresult1.com">Cache</a>
|
|
|
|
|
</div>
|
|
|
|
|
<div class="search_result">
|
|
|
|
|
<a class="result" href="https://result2.com">Result 2</a>
|
|
|
|
|
<p class="content">Content 2</p>
|
|
|
|
|
<a class="cached" href="https://cachedresult2.com">Cache</a>
|
|
|
|
|
</div>
|
|
|
|
|
</div>
|
|
|
|
|
"""
|
|
|
|
|
|
[fix] unit test_xpath.py: name 'logger' is not defined
Depending on the order in which the unit tests are executed, the python modules
of the engines are initialized (monkey patched) or not. As the order of the
tests is not static, random errors may occur.
To avaoid random `NameError: name 'logger' is not defined` in the unit tests of
the xpath engine, a logger is monkey patched into the xpath py-module.
```
make test.unit
TEST tests/unit
......EE...................
======================================================================
ERROR: test_response (tests.unit.engines.test_xpath.TestXpathEngine.test_response)
----------------------------------------------------------------------
Traceback (most recent call last):
File "./tests/unit/engines/test_xpath.py", line 60, in test_response
self.assertEqual(xpath.response(response), [])
^^^^^^^^^^^^^^^^^^^^^^^^
File "./searx/engines/xpath.py", line 309, in response
logger.debug("found %s results", len(results))
^^^^^^
NameError: name 'logger' is not defined
======================================================================
ERROR: test_response_results_xpath (tests.unit.engines.test_xpath.TestXpathEngine.test_response_results_xpath)
----------------------------------------------------------------------
Traceback (most recent call last):
File "./tests/unit/engines/test_xpath.py", line 102, in test_response_results_xpath
self.assertEqual(xpath.response(response), [])
^^^^^^^^^^^^^^^^^^^^^^^^
File "./searx/engines/xpath.py", line 309, in response
logger.debug("found %s results", len(results))
^^^^^^
NameError: name 'logger' is not defined
```
Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>
2024-06-25 10:49:37 +02:00
|
|
|
def setUp(self):
|
|
|
|
|
xpath.logger = logger.getChild('test_xpath')
|
|
|
|
|
|
2016-05-19 07:38:43 +02:00
|
|
|
def test_request(self):
|
|
|
|
|
xpath.search_url = 'https://url.com/{query}'
|
|
|
|
|
xpath.categories = []
|
|
|
|
|
xpath.paging = False
|
|
|
|
|
query = 'test_query'
|
|
|
|
|
dicto = defaultdict(dict)
|
2024-06-21 15:58:12 +02:00
|
|
|
dicto['language'] = 'all'
|
|
|
|
|
dicto['pageno'] = 1
|
2016-05-19 07:38:43 +02:00
|
|
|
params = xpath.request(query, dicto)
|
|
|
|
|
self.assertIn('url', params)
|
2024-06-21 15:58:12 +02:00
|
|
|
self.assertEqual('https://url.com/test_query', params['url'])
|
2016-05-19 07:38:43 +02:00
|
|
|
|
|
|
|
|
xpath.search_url = 'https://url.com/q={query}&p={pageno}'
|
|
|
|
|
xpath.paging = True
|
|
|
|
|
query = 'test_query'
|
|
|
|
|
dicto = defaultdict(dict)
|
2024-06-21 15:58:12 +02:00
|
|
|
dicto['language'] = 'all'
|
2016-05-19 07:38:43 +02:00
|
|
|
dicto['pageno'] = 1
|
|
|
|
|
params = xpath.request(query, dicto)
|
|
|
|
|
self.assertIn('url', params)
|
2024-06-21 15:58:12 +02:00
|
|
|
self.assertEqual('https://url.com/q=test_query&p=1', params['url'])
|
2016-05-19 07:38:43 +02:00
|
|
|
|
|
|
|
|
def test_response(self):
|
|
|
|
|
# without results_xpath
|
|
|
|
|
xpath.url_xpath = '//div[@class="search_result"]//a[@class="result"]/@href'
|
|
|
|
|
xpath.title_xpath = '//div[@class="search_result"]//a[@class="result"]'
|
|
|
|
|
xpath.content_xpath = '//div[@class="search_result"]//p[@class="content"]'
|
|
|
|
|
|
|
|
|
|
self.assertRaises(AttributeError, xpath.response, None)
|
|
|
|
|
self.assertRaises(AttributeError, xpath.response, [])
|
|
|
|
|
self.assertRaises(AttributeError, xpath.response, '')
|
|
|
|
|
self.assertRaises(AttributeError, xpath.response, '[]')
|
|
|
|
|
|
2024-06-21 15:58:12 +02:00
|
|
|
response = mock.Mock(text='<html></html>', status_code=200)
|
2016-05-19 07:38:43 +02:00
|
|
|
self.assertEqual(xpath.response(response), [])
|
|
|
|
|
|
2024-06-21 15:58:12 +02:00
|
|
|
response = mock.Mock(text=self.html, status_code=200)
|
2016-05-19 07:38:43 +02:00
|
|
|
results = xpath.response(response)
|
|
|
|
|
self.assertEqual(type(results), list)
|
|
|
|
|
self.assertEqual(len(results), 2)
|
|
|
|
|
self.assertEqual(results[0]['title'], 'Result 1')
|
|
|
|
|
self.assertEqual(results[0]['url'], 'https://result1.com/')
|
|
|
|
|
self.assertEqual(results[0]['content'], 'Content 1')
|
|
|
|
|
self.assertEqual(results[1]['title'], 'Result 2')
|
|
|
|
|
self.assertEqual(results[1]['url'], 'https://result2.com/')
|
|
|
|
|
self.assertEqual(results[1]['content'], 'Content 2')
|
|
|
|
|
|
|
|
|
|
# with cached urls, without results_xpath
|
|
|
|
|
xpath.cached_xpath = '//div[@class="search_result"]//a[@class="cached"]/@href'
|
|
|
|
|
results = xpath.response(response)
|
|
|
|
|
self.assertEqual(type(results), list)
|
|
|
|
|
self.assertEqual(len(results), 2)
|
|
|
|
|
self.assertEqual(results[0]['cached_url'], 'https://cachedresult1.com')
|
|
|
|
|
self.assertEqual(results[1]['cached_url'], 'https://cachedresult2.com')
|
|
|
|
|
self.assertFalse(results[0].get('is_onion', False))
|
|
|
|
|
|
|
|
|
|
# results are onion urls (no results_xpath)
|
|
|
|
|
xpath.categories = ['onions']
|
|
|
|
|
results = xpath.response(response)
|
|
|
|
|
self.assertTrue(results[0]['is_onion'])
|
|
|
|
|
|
2024-06-21 15:58:12 +02:00
|
|
|
def test_response_results_xpath(self):
|
2016-05-19 07:38:43 +02:00
|
|
|
# with results_xpath
|
|
|
|
|
xpath.results_xpath = '//div[@class="search_result"]'
|
|
|
|
|
xpath.url_xpath = './/a[@class="result"]/@href'
|
|
|
|
|
xpath.title_xpath = './/a[@class="result"]'
|
|
|
|
|
xpath.content_xpath = './/p[@class="content"]'
|
|
|
|
|
xpath.cached_xpath = None
|
|
|
|
|
xpath.categories = []
|
|
|
|
|
|
|
|
|
|
self.assertRaises(AttributeError, xpath.response, None)
|
|
|
|
|
self.assertRaises(AttributeError, xpath.response, [])
|
|
|
|
|
self.assertRaises(AttributeError, xpath.response, '')
|
|
|
|
|
self.assertRaises(AttributeError, xpath.response, '[]')
|
|
|
|
|
|
2024-06-21 15:58:12 +02:00
|
|
|
response = mock.Mock(text='<html></html>', status_code=200)
|
2016-05-19 07:38:43 +02:00
|
|
|
self.assertEqual(xpath.response(response), [])
|
|
|
|
|
|
2024-06-21 15:58:12 +02:00
|
|
|
response = mock.Mock(text=self.html, status_code=200)
|
2016-05-19 07:38:43 +02:00
|
|
|
results = xpath.response(response)
|
|
|
|
|
self.assertEqual(type(results), list)
|
|
|
|
|
self.assertEqual(len(results), 2)
|
|
|
|
|
self.assertEqual(results[0]['title'], 'Result 1')
|
|
|
|
|
self.assertEqual(results[0]['url'], 'https://result1.com/')
|
|
|
|
|
self.assertEqual(results[0]['content'], 'Content 1')
|
|
|
|
|
self.assertEqual(results[1]['title'], 'Result 2')
|
|
|
|
|
self.assertEqual(results[1]['url'], 'https://result2.com/')
|
|
|
|
|
self.assertEqual(results[1]['content'], 'Content 2')
|
|
|
|
|
|
|
|
|
|
# with cached urls, with results_xpath
|
|
|
|
|
xpath.cached_xpath = './/a[@class="cached"]/@href'
|
|
|
|
|
results = xpath.response(response)
|
|
|
|
|
self.assertEqual(type(results), list)
|
|
|
|
|
self.assertEqual(len(results), 2)
|
|
|
|
|
self.assertEqual(results[0]['cached_url'], 'https://cachedresult1.com')
|
|
|
|
|
self.assertEqual(results[1]['cached_url'], 'https://cachedresult2.com')
|
|
|
|
|
self.assertFalse(results[0].get('is_onion', False))
|
|
|
|
|
|
|
|
|
|
# results are onion urls (with results_xpath)
|
|
|
|
|
xpath.categories = ['onions']
|
|
|
|
|
results = xpath.response(response)
|
|
|
|
|
self.assertTrue(results[0]['is_onion'])
|
