def test_retrieve_record(self):
"""
SCENARIO: We have a URL for a landing page.
EXPECTED RESULT: The identifier is retrieved.
"""
url = ("https://www.northwestknowledge.net"
"/data/0a42d2bc-700a-4cf2-a7ac-ad6b892da7f0/")
# External I/O
#
# 1st: landing page
# 2nd: XML metadata document
package = 'tests.data.nkn.0a42d2bc-700a-4cf2-a7ac-ad6b892da7f0'
contents = [
ir.read_binary(package, 'index.html'),
ir.read_binary(package, 'metadata.xml')
]
harvester = NKNHarvester()
with self.assertLogs(logger=harvester.logger, level='INFO'):
with aioresponses() as m:
m.get(self.regex, body=contents[0])
m.get(self.regex, body=contents[1])
awaitable = harvester.retrieve_record(url)
sid, _, lastmod, doc = asyncio.run(awaitable)
expected = '0a42d2bc-700a-4cf2-a7ac-ad6b892da7f0'
self.assertEqual(sid, expected)
self.assertIsNone(lastmod)
def test_retrieve_record__404_error(self):
"""
SCENARIO: We have a URL for a landing page. However, the directory
on the remote end does not have a metadata.xml document.
EXPECTED RESULT: A SkipError is raised. For other clients, this is
NOT a SkipError.
"""
url = ("https://www.northwestknowledge.net"
"/data/94E2D569-200F-44F7-8937-AB4BD0862C91")
# External I/O
#
# 1st: landing page
# 2nd: XML metadata document (raises 404)
package = 'tests.data.nkn.0a42d2bc-700a-4cf2-a7ac-ad6b892da7f0'
contents = [
ir.read_binary(package, 'index.html'),
]
harvester = NKNHarvester()
with self.assertLogs(logger=harvester.logger, level='INFO'):
with aioresponses() as m:
m.get(self.regex, body=contents[0])
m.get(self.regex, status=404)
with self.assertRaises(SkipError):
asyncio.run(harvester.retrieve_record(url))
def test_unsupported_xml_format(self, mock_logger):
"""
SCENARIO: Sometimes the XML format is not supported. We have tests
for this in other places, but NKN is a special case because we have to
extract the identifier from the XML metadata rather than the JSON-LD.
This particular file is some sort of ESRI XML file.
EXPECTED RESULT: XMLValidationError
"""
url = ("https://www.northwestknowledge.net"
"/data/A62BEE88-8F92-4649-BC8D-BC56CE96AE2B")
package = 'tests.data.nkn.A62BEE88-8F92-4649-BC8D-BC56CE96AE2B'
contents = [
ir.read_binary(package, 'metadata.xml'),
ir.read_binary(package, 'index.html'),
]
harvester = NKNHarvester()
with aioresponses() as m:
m.get(self.regex, body=contents[0])
m.get(self.regex, body=contents[1])
with self.assertRaises(XMLValidationError):
asyncio.run(harvester.retrieve_record(url))
def test_retrieve_record__500_error(self):
"""
SCENARIO: We have a URL for a landing page. The response is a server
error, however. This is complimentary to the _404_error test below.
EXPECTED RESULT: An Exception is raised (but not a SkipError).
"""
url = ("https://www.northwestknowledge.net"
"/data/94E2D569-200F-44F7-8937-AB4BD0862C91")
# External I/O
#
# 1st: landing page
# 2nd: XML metadata document (raises 404)
package = 'tests.data.nkn.0a42d2bc-700a-4cf2-a7ac-ad6b892da7f0'
contents = [
ir.read_binary(package, 'index.html'),
]
harvester = NKNHarvester()
with self.assertLogs(logger=harvester.logger, level='INFO'):
with aioresponses() as m:
m.get(self.regex, body=contents[0])
m.get(self.regex, status=500)
with self.assertRaises(aiohttp.ClientResponseError):
asyncio.run(harvester.retrieve_record(url))