def test_encoded_datastream(self):
# data content within a single chunk of data
mockapi = Mock()
mockapi.export.return_value = self.session.get(
'file://%s' % FIXTURES['sync1_export'])
mockapi.upload.return_value = 'uploaded://1'
self.obj.api = self.repo.api = mockapi
section = self.archex.get_next_section()
# get binary datastream info from first section
dsinfo = self.archex.get_datastream_info(section)
# fixture only has one binary content block
# get binarycontent tag out of the way
self.archex.get_next_section()
# next section will be file contents
self.archex.within_file = True
dscontent = b''.join(self.archex.encoded_datastream())
# check decoded size and MD5 match data from fixture
self.assertEqual(int(dsinfo['size']), len(dscontent))
self.assertEqual(dsinfo['digest'], md5sum(dscontent))
# data content across multiple chunks
mockapi.export.return_value = self.session.get(
'file://%s' % FIXTURES['sync1_export'])
self.obj.api = self.repo.api = mockapi
# set read block size artificially low to ensure
# datastream content is spread across multiple chunks
self.archex.read_block_size = 1024
finished = False
# iterate through the data, similar to object_data method,
# but only handle binary content
while not finished:
try:
section = self.archex.get_next_section()
except StopIteration:
finished = True
# find the section with starting binary content
if section == '<foxml:binaryContent>':
# then decode the subsequent content
self.archex.within_file = True
dscontent = ''.join(self.archex.encoded_datastream())
self.assertEqual(int(dsinfo['size']), len(dscontent))
self.assertEqual(dsinfo['digest'], md5sum(dscontent))
# stop processing
finished = True
def test_upload_generator(self):
# test uploading content from a generator
def data_generator():
yield 'line one of text\n'
yield 'line two of text\n'
yield 'line three of text\n'
text_content = ''.join(data_generator())
content_md5 = md5sum(text_content)
size = len(text_content)
upload_id = self.rest_api.upload(data_generator(), size=size, content_type='text/plain')
pattern = re.compile('uploaded://[0-9]+')
self.assertTrue(pattern.match(upload_id))
# check that the *right* content was uploaded by adding
# a datastream using the computed MD5 and generated upload id
obj = load_fixture_data('basic-object.foxml')
response = self.rest_api.ingest(obj)
pid = response.text
add_response = self.rest_api.addDatastream(pid, 'text', controlGroup='M',
dsLocation=upload_id, mimeType='text/plain',
checksumType='MD5', checksum=content_md5)
self.assertTrue(add_response.status_code, requests.codes.created)
# get the content from fedora and confirm it matches what was sent
dsresponse = self.rest_api.getDatastreamDissemination(pid, 'text')
self.assertEqual(text_content, dsresponse.text)
# clean up test object
self.rest_api.purgeObject(pid)
def test_upload_generator(self):
# test uploading content from a generator
def data_generator():
yield 'line one of text\n'
yield 'line two of text\n'
yield 'line three of text\n'
text_content = ''.join(data_generator())
content_md5 = md5sum(text_content)
size = len(text_content)
upload_id = self.rest_api.upload(data_generator(), size=size,
content_type='text/plain')
pattern = re.compile('uploaded://[0-9]+')
self.assertTrue(pattern.match(upload_id))
# check that the *right* content was uploaded by adding
# a datastream using the computed MD5 and generated upload id
obj = load_fixture_data('basic-object.foxml')
response = self.rest_api.ingest(obj)
pid = response.text
add_response = self.rest_api.addDatastream(pid, 'text', controlGroup='M',
dsLocation=upload_id, mimeType='text/plain',
checksumType='MD5', checksum=content_md5)
self.assertTrue(add_response.status_code, requests.codes.created)
# get the content from fedora and confirm it matches what was sent
dsresponse = self.rest_api.getDatastreamDissemination(pid, 'text')
self.assertEqual(text_content, dsresponse.text)
# clean up test object
self.rest_api.purgeObject(pid)
def test_encoded_datastream(self):
# data content within a single chunk of data
mockapi = Mock()
mockapi.export.return_value = self.session.get('file://%s' % FIXTURES['sync1_export'])
mockapi.upload.return_value = 'uploaded://1'
self.obj.api = self.repo.api = mockapi
section = self.archex.get_next_section()
# get binary datastream info from first section
dsinfo = self.archex.get_datastream_info(section)
# fixture only has one binary content block
# get binarycontent tag out of the way
self.archex.get_next_section()
# next section will be file contents
self.archex.within_file = True
dscontent = b''.join(self.archex.encoded_datastream())
# check decoded size and MD5 match data from fixture
self.assertEqual(int(dsinfo['size']), len(dscontent))
self.assertEqual(dsinfo['digest'], md5sum(dscontent))
# data content across multiple chunks
mockapi.export.return_value = self.session.get('file://%s' % FIXTURES['sync1_export'])
self.obj.api = self.repo.api = mockapi
# set read block size artificially low to ensure
# datastream content is spread across multiple chunks
self.archex.read_block_size = 1024
finished = False
# iterate through the data, similar to object_data method,
# but only handle binary content
while not finished:
try:
section = self.archex.get_next_section()
except StopIteration:
finished = True
# find the section with starting binary content
if section == '<foxml:binaryContent>':
# then decode the subsequent content
self.archex.within_file = True
dscontent = ''.join(self.archex.encoded_datastream())
self.assertEqual(int(dsinfo['size']), len(dscontent))
self.assertEqual(dsinfo['digest'], md5sum(dscontent))
# stop processing
finished = True
