def test_metadata_for_preprint_has_correct_structure(self):
provider = PreprintProviderFactory()
license = NodeLicense.objects.get(
name="CC-By Attribution 4.0 International")
license_details = {
'id': license.license_id,
'year': '2017',
'copyrightHolders': ['Jeff Hardy', 'Matt Hardy']
}
preprint = PreprintFactory(provider=provider,
project=self.node,
is_published=True,
license_details=license_details)
metadata_xml = metadata.datacite_metadata_for_preprint(
preprint,
doi=preprint.get_identifier('doi').value,
pretty_print=True)
root = lxml.etree.fromstring(metadata_xml)
xsi_location = '{http://www.w3.org/2001/XMLSchema-instance}schemaLocation'
expected_location = 'http://datacite.org/schema/kernel-4 http://schema.datacite.org/meta/kernel-4/metadata.xsd'
assert root.attrib[xsi_location] == expected_location
identifier = root.find('{%s}identifier' % metadata.NAMESPACE)
assert identifier.attrib['identifierType'] == 'DOI'
assert identifier.text == preprint.get_identifier('doi').value
creators = root.find('{%s}creators' % metadata.NAMESPACE)
assert len(creators.getchildren()) == len(
self.node.visible_contributors)
subjects = root.find('{%s}subjects' % metadata.NAMESPACE)
assert subjects.getchildren()
publisher = root.find('{%s}publisher' % metadata.NAMESPACE)
assert publisher.text == provider.name
pub_year = root.find('{%s}publicationYear' % metadata.NAMESPACE)
assert pub_year.text == str(preprint.date_published.year)
dates = root.find('{%s}dates' % metadata.NAMESPACE).getchildren()[0]
assert dates.text == preprint.modified.isoformat()
assert dates.attrib['dateType'] == 'Updated'
alternate_identifier = root.find('{%s}alternateIdentifiers' %
metadata.NAMESPACE).getchildren()[0]
assert alternate_identifier.text == settings.DOMAIN + preprint._id
assert alternate_identifier.attrib['alternateIdentifierType'] == 'URL'
descriptions = root.find('{%s}descriptions' %
metadata.NAMESPACE).getchildren()[0]
assert descriptions.text == preprint.node.description
rights = root.find('{%s}rightsList' %
metadata.NAMESPACE).getchildren()[0]
assert rights.text == preprint.license.name
def test_metadata_for_preprint_has_correct_structure(self):
provider = PreprintProviderFactory()
license = NodeLicense.objects.get(name="CC-By Attribution 4.0 International")
license_details = {
'id': license.license_id,
'year': '2017',
'copyrightHolders': ['Jeff Hardy', 'Matt Hardy']
}
preprint = PreprintFactory(provider=provider, project=self.node, is_published=True, license_details=license_details)
metadata_xml = metadata.datacite_metadata_for_preprint(preprint, doi=preprint.get_identifier('doi').value, pretty_print=True)
root = lxml.etree.fromstring(metadata_xml)
xsi_location = '{http://www.w3.org/2001/XMLSchema-instance}schemaLocation'
expected_location = 'http://datacite.org/schema/kernel-4 http://schema.datacite.org/meta/kernel-4/metadata.xsd'
assert root.attrib[xsi_location] == expected_location
identifier = root.find('{%s}identifier' % metadata.NAMESPACE)
assert identifier.attrib['identifierType'] == 'DOI'
assert identifier.text == preprint.get_identifier('doi').value
creators = root.find('{%s}creators' % metadata.NAMESPACE)
assert len(creators.getchildren()) == len(self.node.visible_contributors)
subjects = root.find('{%s}subjects' % metadata.NAMESPACE)
assert subjects.getchildren()
publisher = root.find('{%s}publisher' % metadata.NAMESPACE)
assert publisher.text == provider.name
pub_year = root.find('{%s}publicationYear' % metadata.NAMESPACE)
assert pub_year.text == str(preprint.date_published.year)
dates = root.find('{%s}dates' % metadata.NAMESPACE).getchildren()[0]
assert dates.text == preprint.date_modified.isoformat()
assert dates.attrib['dateType'] == 'Updated'
alternate_identifier = root.find('{%s}alternateIdentifiers' % metadata.NAMESPACE).getchildren()[0]
assert alternate_identifier.text == settings.DOMAIN + preprint._id
assert alternate_identifier.attrib['alternateIdentifierType'] == 'URL'
descriptions = root.find('{%s}descriptions' % metadata.NAMESPACE).getchildren()[0]
assert descriptions.text == preprint.node.description
rights = root.find('{%s}rightsList' % metadata.NAMESPACE).getchildren()[0]
assert rights.text == preprint.license.name
class TestOnPreprintUpdatedTask(OsfTestCase):
def setUp(self):
super(TestOnPreprintUpdatedTask, self).setUp()
self.user = AuthUserFactory()
if len(self.user.fullname.split(' ')) > 2:
# Prevent unexpected keys ('suffix', 'additional_name')
self.user.fullname = 'David Davidson'
self.user.middle_names = ''
self.user.suffix = ''
self.user.save()
self.auth = Auth(user=self.user)
self.preprint = PreprintFactory()
self.preprint.node.add_tag('preprint', self.auth, save=False)
self.preprint.node.add_tag('spoderman', self.auth, save=False)
self.preprint.node.add_unregistered_contributor('BoJack Horseman', '*****@*****.**', Auth(self.preprint.node.creator))
self.preprint.node.add_contributor(self.user, visible=False)
self.preprint.node.save()
self.preprint.node.creator.given_name = u'ZZYZ'
if len(self.preprint.node.creator.fullname.split(' ')) > 2:
# Prevent unexpected keys ('suffix', 'additional_name')
self.preprint.node.creator.fullname = 'David Davidson'
self.preprint.node.creator.middle_names = ''
self.preprint.node.creator.suffix = ''
self.preprint.node.creator.save()
self.preprint.set_subjects([[SubjectFactory()._id]], auth=Auth(self.preprint.node.creator), save=False)
def tearDown(self):
handlers.celery_before_request()
super(TestOnPreprintUpdatedTask, self).tearDown()
def test_format_preprint(self):
res = format_preprint(self.preprint)
assert set(gn['@type'] for gn in res) == {'creator', 'contributor', 'throughsubjects', 'subject', 'throughtags', 'tag', 'workidentifier', 'agentidentifier', 'person', 'preprint', 'workrelation', 'creativework'}
nodes = dict(enumerate(res))
preprint = nodes.pop(next(k for k, v in nodes.items() if v['@type'] == 'preprint'))
assert preprint['title'] == self.preprint.node.title
assert preprint['description'] == self.preprint.node.description
assert preprint['is_deleted'] == (not self.preprint.is_published or not self.preprint.node.is_public or self.preprint.node.is_preprint_orphan)
assert preprint['date_updated'] == self.preprint.date_modified.isoformat()
assert preprint['date_published'] == self.preprint.date_published.isoformat()
tags = [nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'tag']
through_tags = [nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'throughtags']
assert sorted(tag['@id'] for tag in tags) == sorted(tt['tag']['@id'] for tt in through_tags)
assert sorted(tag['name'] for tag in tags) == ['preprint', 'spoderman']
subjects = [nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'subject']
through_subjects = [nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'throughsubjects']
assert sorted(subject['@id'] for subject in subjects) == sorted(tt['subject']['@id'] for tt in through_subjects)
assert sorted(subject['name'] for subject in subjects) == [s.bepress_text for h in self.preprint.subject_hierarchy for s in h]
people = sorted([nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'person'], key=lambda x: x['given_name'])
expected_people = sorted([{
'@type': 'person',
'given_name': u'BoJack',
'family_name': u'Horseman',
}, {
'@type': 'person',
'given_name': self.user.given_name,
'family_name': self.user.family_name,
}, {
'@type': 'person',
'given_name': self.preprint.node.creator.given_name,
'family_name': self.preprint.node.creator.family_name,
}], key=lambda x: x['given_name'])
for i, p in enumerate(expected_people):
expected_people[i]['@id'] = people[i]['@id']
assert people == expected_people
creators = sorted([nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'creator'], key=lambda x: x['order_cited'])
assert creators == [{
'@id': creators[0]['@id'],
'@type': 'creator',
'order_cited': 0,
'cited_as': u'{}'.format(self.preprint.node.creator.fullname),
'agent': {'@id': [p['@id'] for p in people if p['given_name'] == self.preprint.node.creator.given_name][0], '@type': 'person'},
'creative_work': {'@id': preprint['@id'], '@type': preprint['@type']},
}, {
'@id': creators[1]['@id'],
'@type': 'creator',
'order_cited': 1,
'cited_as': u'BoJack Horseman',
'agent': {'@id': [p['@id'] for p in people if p['given_name'] == u'BoJack'][0], '@type': 'person'},
'creative_work': {'@id': preprint['@id'], '@type': preprint['@type']},
}]
contributors = [nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'contributor']
assert contributors == [{
'@id': contributors[0]['@id'],
'@type': 'contributor',
'cited_as': u'{}'.format(self.user.fullname),
'agent': {'@id': [p['@id'] for p in people if p['given_name'] == self.user.given_name][0], '@type': 'person'},
'creative_work': {'@id': preprint['@id'], '@type': preprint['@type']},
}]
agentidentifiers = {nodes.pop(k)['uri'] for k, v in nodes.items() if v['@type'] == 'agentidentifier'}
assert agentidentifiers == set([
'mailto:' + self.user.username,
'mailto:' + self.preprint.node.creator.username,
self.user.profile_image_url(),
self.preprint.node.creator.profile_image_url(),
]) | set(urlparse.urljoin(settings.DOMAIN, user.profile_url) for user in self.preprint.node.contributors if user.is_registered)
related_work = next(nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'creativework')
assert set(related_work.keys()) == {'@id', '@type'} # Empty except @id and @type
osf_doi = next(nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'workidentifier' and 'doi' in v['uri'] and 'osf.io' in v['uri'])
assert osf_doi['creative_work'] == {'@id': preprint['@id'], '@type': preprint['@type']}
related_doi = next(nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'workidentifier' and 'doi' in v['uri'])
assert related_doi['creative_work'] == related_work
workidentifiers = [nodes.pop(k)['uri'] for k, v in nodes.items() if v['@type'] == 'workidentifier']
assert workidentifiers == [urlparse.urljoin(settings.DOMAIN, self.preprint._id + '/')]
relation = nodes.pop(nodes.keys()[0])
assert relation == {'@id': relation['@id'], '@type': 'workrelation', 'related': {'@id': related_work['@id'], '@type': related_work['@type']}, 'subject': {'@id': preprint['@id'], '@type': preprint['@type']}}
assert nodes == {}
def test_format_preprint_nones(self):
self.preprint.node.tags = []
self.preprint.date_published = None
self.preprint.node.preprint_article_doi = None
self.preprint.set_subjects([], auth=Auth(self.preprint.node.creator), save=False)
res = format_preprint(self.preprint)
assert self.preprint.provider != 'osf'
assert set(gn['@type'] for gn in res) == {'creator', 'contributor', 'workidentifier', 'agentidentifier', 'person', 'preprint'}
nodes = dict(enumerate(res))
preprint = nodes.pop(next(k for k, v in nodes.items() if v['@type'] == 'preprint'))
assert preprint['title'] == self.preprint.node.title
assert preprint['description'] == self.preprint.node.description
assert preprint['is_deleted'] == (not self.preprint.is_published or not self.preprint.node.is_public or self.preprint.node.is_preprint_orphan)
assert preprint['date_updated'] == self.preprint.date_modified.isoformat()
assert preprint.get('date_published') is None
people = sorted([nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'person'], key=lambda x: x['given_name'])
expected_people = sorted([{
'@type': 'person',
'given_name': u'BoJack',
'family_name': u'Horseman',
}, {
'@type': 'person',
'given_name': self.user.given_name,
'family_name': self.user.family_name,
}, {
'@type': 'person',
'given_name': self.preprint.node.creator.given_name,
'family_name': self.preprint.node.creator.family_name,
}], key=lambda x: x['given_name'])
for i, p in enumerate(expected_people):
expected_people[i]['@id'] = people[i]['@id']
assert people == expected_people
creators = sorted([nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'creator'], key=lambda x: x['order_cited'])
assert creators == [{
'@id': creators[0]['@id'],
'@type': 'creator',
'order_cited': 0,
'cited_as': self.preprint.node.creator.fullname,
'agent': {'@id': [p['@id'] for p in people if p['given_name'] == self.preprint.node.creator.given_name][0], '@type': 'person'},
'creative_work': {'@id': preprint['@id'], '@type': preprint['@type']},
}, {
'@id': creators[1]['@id'],
'@type': 'creator',
'order_cited': 1,
'cited_as': u'BoJack Horseman',
'agent': {'@id': [p['@id'] for p in people if p['given_name'] == u'BoJack'][0], '@type': 'person'},
'creative_work': {'@id': preprint['@id'], '@type': preprint['@type']},
}]
contributors = [nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'contributor']
assert contributors == [{
'@id': contributors[0]['@id'],
'@type': 'contributor',
'cited_as': self.user.fullname,
'agent': {'@id': [p['@id'] for p in people if p['given_name'] == self.user.given_name][0], '@type': 'person'},
'creative_work': {'@id': preprint['@id'], '@type': preprint['@type']},
}]
agentidentifiers = {nodes.pop(k)['uri'] for k, v in nodes.items() if v['@type'] == 'agentidentifier'}
assert agentidentifiers == set([
'mailto:' + self.user.username,
'mailto:' + self.preprint.node.creator.username,
self.user.profile_image_url(),
self.preprint.node.creator.profile_image_url(),
]) | set(urlparse.urljoin(settings.DOMAIN, user.profile_url) for user in self.preprint.node.contributors if user.is_registered)
workidentifiers = {nodes.pop(k)['uri'] for k, v in nodes.items() if v['@type'] == 'workidentifier'}
# URLs should *always* be osf.io/guid/
assert workidentifiers == set([urlparse.urljoin(settings.DOMAIN, self.preprint._id) + '/', 'http://dx.doi.org/{}'.format(self.preprint.get_identifier('doi').value)])
assert nodes == {}
def test_format_preprint_is_deleted(self):
CASES = {
'is_published': (True, False),
'is_published': (False, True),
'node.is_public': (True, False),
'node.is_public': (False, True),
'node._is_preprint_orphan': (True, True),
'node._is_preprint_orphan': (False, False),
'node.is_deleted': (True, True),
'node.is_deleted': (False, False),
}
for key, (value, is_deleted) in CASES.items():
target = self.preprint
for k in key.split('.')[:-1]:
if k:
target = getattr(target, k)
orig_val = getattr(target, key.split('.')[-1])
setattr(target, key.split('.')[-1], value)
res = format_preprint(self.preprint)
preprint = next(v for v in res if v['@type'] == 'preprint')
assert preprint['is_deleted'] is is_deleted
setattr(target, key.split('.')[-1], orig_val)
def test_format_preprint_is_deleted_true_if_qatest_tag_is_added(self):
res = format_preprint(self.preprint)
preprint = next(v for v in res if v['@type'] == 'preprint')
assert preprint['is_deleted'] is False
self.preprint.node.add_tag('qatest', auth=self.auth, save=True)
res = format_preprint(self.preprint)
preprint = next(v for v in res if v['@type'] == 'preprint')
assert preprint['is_deleted'] is True
class TestOnPreprintUpdatedTask(OsfTestCase):
def setUp(self):
super(TestOnPreprintUpdatedTask, self).setUp()
self.user = AuthUserFactory()
if len(self.user.fullname.split(' ')) > 2:
# Prevent unexpected keys ('suffix', 'additional_name')
self.user.fullname = 'David Davidson'
self.user.middle_names = ''
self.user.suffix = ''
self.user.save()
self.auth = Auth(user=self.user)
self.preprint = PreprintFactory()
thesis_provider = PreprintProviderFactory(share_publish_type='Thesis')
self.thesis = PreprintFactory(provider=thesis_provider)
for pp in [self.preprint, self.thesis]:
pp.node.add_tag('preprint', self.auth, save=False)
pp.node.add_tag('spoderman', self.auth, save=False)
pp.node.add_unregistered_contributor('BoJack Horseman',
'*****@*****.**',
Auth(pp.node.creator))
pp.node.add_contributor(self.user, visible=False)
pp.node.save()
pp.node.creator.given_name = u'ZZYZ'
if len(pp.node.creator.fullname.split(' ')) > 2:
# Prevent unexpected keys ('suffix', 'additional_name')
pp.node.creator.fullname = 'David Davidson'
pp.node.creator.middle_names = ''
pp.node.creator.suffix = ''
pp.node.creator.save()
pp.set_subjects([[SubjectFactory()._id]],
auth=Auth(pp.node.creator))
def tearDown(self):
handlers.celery_before_request()
super(TestOnPreprintUpdatedTask, self).tearDown()
def test_format_preprint(self):
res = format_preprint(self.preprint,
self.preprint.provider.share_publish_type)
assert set(gn['@type'] for gn in res) == {
'creator', 'contributor', 'throughsubjects', 'subject',
'throughtags', 'tag', 'workidentifier', 'agentidentifier',
'person', 'preprint', 'workrelation', 'creativework'
}
nodes = dict(enumerate(res))
preprint = nodes.pop(
next(k for k, v in nodes.items() if v['@type'] == 'preprint'))
assert preprint['title'] == self.preprint.node.title
assert preprint['description'] == self.preprint.node.description
assert preprint['is_deleted'] == (
not self.preprint.is_published or not self.preprint.node.is_public
or self.preprint.node.is_preprint_orphan)
assert preprint['date_updated'] == self.preprint.modified.isoformat()
assert preprint[
'date_published'] == self.preprint.date_published.isoformat()
tags = [nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'tag']
through_tags = [
nodes.pop(k) for k, v in nodes.items()
if v['@type'] == 'throughtags'
]
assert sorted(tag['@id']
for tag in tags) == sorted(tt['tag']['@id']
for tt in through_tags)
assert sorted(tag['name'] for tag in tags) == ['preprint', 'spoderman']
subjects = [
nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'subject'
]
through_subjects = [
nodes.pop(k) for k, v in nodes.items()
if v['@type'] == 'throughsubjects'
]
s_ids = [s['@id'] for s in subjects]
ts_ids = [ts['subject']['@id'] for ts in through_subjects]
cs_ids = [
i for i in set(
s.get('central_synonym', {}).get('@id') for s in subjects) if i
]
for ts in ts_ids:
assert ts in s_ids
assert ts not in cs_ids # Only aliased subjects are connected to self.preprint
for s in subjects:
subject = Subject.objects.get(text=s['name'])
assert s['uri'].endswith('v2/taxonomies/{}/'.format(
subject._id)) # This cannot change
assert set(subject['name'] for subject in subjects) == set(
[s.text for s in self.preprint.subjects.all()] + [
s.bepress_subject.text for s in self.preprint.subjects.filter(
bepress_subject__isnull=False)
])
people = sorted(
[nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'person'],
key=lambda x: x['given_name'])
expected_people = sorted(
[{
'@type': 'person',
'given_name': u'BoJack',
'family_name': u'Horseman',
}, {
'@type': 'person',
'given_name': self.user.given_name,
'family_name': self.user.family_name,
}, {
'@type': 'person',
'given_name': self.preprint.node.creator.given_name,
'family_name': self.preprint.node.creator.family_name,
}],
key=lambda x: x['given_name'])
for i, p in enumerate(expected_people):
expected_people[i]['@id'] = people[i]['@id']
assert people == expected_people
creators = sorted([
nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'creator'
],
key=lambda x: x['order_cited'])
assert creators == [{
'@id':
creators[0]['@id'],
'@type':
'creator',
'order_cited':
0,
'cited_as':
u'{}'.format(self.preprint.node.creator.fullname),
'agent': {
'@id': [
p['@id'] for p in people
if p['given_name'] == self.preprint.node.creator.given_name
][0],
'@type':
'person'
},
'creative_work': {
'@id': preprint['@id'],
'@type': preprint['@type']
},
}, {
'@id': creators[1]['@id'],
'@type': 'creator',
'order_cited': 1,
'cited_as': u'BoJack Horseman',
'agent': {
'@id':
[p['@id'] for p in people if p['given_name'] == u'BoJack'][0],
'@type':
'person'
},
'creative_work': {
'@id': preprint['@id'],
'@type': preprint['@type']
},
}]
contributors = [
nodes.pop(k) for k, v in nodes.items()
if v['@type'] == 'contributor'
]
assert contributors == [{
'@id': contributors[0]['@id'],
'@type': 'contributor',
'cited_as': u'{}'.format(self.user.fullname),
'agent': {
'@id': [
p['@id'] for p in people
if p['given_name'] == self.user.given_name
][0],
'@type':
'person'
},
'creative_work': {
'@id': preprint['@id'],
'@type': preprint['@type']
},
}]
agentidentifiers = {
nodes.pop(k)['uri']
for k, v in nodes.items() if v['@type'] == 'agentidentifier'
}
assert agentidentifiers == set([
'mailto:' + self.user.username,
'mailto:' + self.preprint.node.creator.username,
self.user.profile_image_url(),
self.preprint.node.creator.profile_image_url(),
]) | set(user.absolute_url for user in self.preprint.node.contributors)
related_work = next(
nodes.pop(k) for k, v in nodes.items()
if v['@type'] == 'creativework')
assert set(related_work.keys()) == {'@id', '@type'
} # Empty except @id and @type
osf_doi = next(
nodes.pop(k) for k, v in nodes.items()
if v['@type'] == 'workidentifier' and 'doi' in v['uri']
and 'osf.io' in v['uri'])
assert osf_doi['creative_work'] == {
'@id': preprint['@id'],
'@type': preprint['@type']
}
related_doi = next(
nodes.pop(k) for k, v in nodes.items()
if v['@type'] == 'workidentifier' and 'doi' in v['uri'])
assert related_doi['creative_work'] == related_work
workidentifiers = [
nodes.pop(k)['uri'] for k, v in nodes.items()
if v['@type'] == 'workidentifier'
]
assert workidentifiers == [
urlparse.urljoin(settings.DOMAIN, self.preprint._id + '/')
]
relation = nodes.pop(nodes.keys()[0])
assert relation == {
'@id': relation['@id'],
'@type': 'workrelation',
'related': {
'@id': related_work['@id'],
'@type': related_work['@type']
},
'subject': {
'@id': preprint['@id'],
'@type': preprint['@type']
}
}
assert nodes == {}
def test_format_thesis(self):
res = format_preprint(self.thesis,
self.thesis.provider.share_publish_type)
assert set(gn['@type'] for gn in res) == {
'creator', 'contributor', 'throughsubjects', 'subject',
'throughtags', 'tag', 'workidentifier', 'agentidentifier',
'person', 'thesis', 'workrelation', 'creativework'
}
nodes = dict(enumerate(res))
thesis = nodes.pop(
next(k for k, v in nodes.items() if v['@type'] == 'thesis'))
assert thesis['title'] == self.thesis.node.title
assert thesis['description'] == self.thesis.node.description
def test_format_preprint_date_modified_node_updated(self):
self.preprint.node.save()
res = format_preprint(self.preprint,
self.preprint.provider.share_publish_type)
nodes = dict(enumerate(res))
preprint = nodes.pop(
next(k for k, v in nodes.items() if v['@type'] == 'preprint'))
assert preprint[
'date_updated'] == self.preprint.node.modified.isoformat()
def test_format_preprint_nones(self):
self.preprint.node.tags = []
self.preprint.date_published = None
self.preprint.node.preprint_article_doi = None
self.preprint.set_subjects([], auth=Auth(self.preprint.node.creator))
res = format_preprint(self.preprint,
self.preprint.provider.share_publish_type)
assert self.preprint.provider != 'osf'
assert set(gn['@type'] for gn in res) == {
'creator', 'contributor', 'workidentifier', 'agentidentifier',
'person', 'preprint'
}
nodes = dict(enumerate(res))
preprint = nodes.pop(
next(k for k, v in nodes.items() if v['@type'] == 'preprint'))
assert preprint['title'] == self.preprint.node.title
assert preprint['description'] == self.preprint.node.description
assert preprint['is_deleted'] == (
not self.preprint.is_published or not self.preprint.node.is_public
or self.preprint.node.is_preprint_orphan)
assert preprint['date_updated'] == self.preprint.modified.isoformat()
assert preprint.get('date_published') is None
people = sorted(
[nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'person'],
key=lambda x: x['given_name'])
expected_people = sorted(
[{
'@type': 'person',
'given_name': u'BoJack',
'family_name': u'Horseman',
}, {
'@type': 'person',
'given_name': self.user.given_name,
'family_name': self.user.family_name,
}, {
'@type': 'person',
'given_name': self.preprint.node.creator.given_name,
'family_name': self.preprint.node.creator.family_name,
}],
key=lambda x: x['given_name'])
for i, p in enumerate(expected_people):
expected_people[i]['@id'] = people[i]['@id']
assert people == expected_people
creators = sorted([
nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'creator'
],
key=lambda x: x['order_cited'])
assert creators == [{
'@id': creators[0]['@id'],
'@type': 'creator',
'order_cited': 0,
'cited_as': self.preprint.node.creator.fullname,
'agent': {
'@id': [
p['@id'] for p in people
if p['given_name'] == self.preprint.node.creator.given_name
][0],
'@type':
'person'
},
'creative_work': {
'@id': preprint['@id'],
'@type': preprint['@type']
},
}, {
'@id': creators[1]['@id'],
'@type': 'creator',
'order_cited': 1,
'cited_as': u'BoJack Horseman',
'agent': {
'@id':
[p['@id'] for p in people if p['given_name'] == u'BoJack'][0],
'@type':
'person'
},
'creative_work': {
'@id': preprint['@id'],
'@type': preprint['@type']
},
}]
contributors = [
nodes.pop(k) for k, v in nodes.items()
if v['@type'] == 'contributor'
]
assert contributors == [{
'@id': contributors[0]['@id'],
'@type': 'contributor',
'cited_as': self.user.fullname,
'agent': {
'@id': [
p['@id'] for p in people
if p['given_name'] == self.user.given_name
][0],
'@type':
'person'
},
'creative_work': {
'@id': preprint['@id'],
'@type': preprint['@type']
},
}]
agentidentifiers = {
nodes.pop(k)['uri']
for k, v in nodes.items() if v['@type'] == 'agentidentifier'
}
assert agentidentifiers == set([
'mailto:' + self.user.username,
'mailto:' + self.preprint.node.creator.username,
self.user.profile_image_url(),
self.preprint.node.creator.profile_image_url(),
]) | set(user.absolute_url for user in self.preprint.node.contributors)
workidentifiers = {
nodes.pop(k)['uri']
for k, v in nodes.items() if v['@type'] == 'workidentifier'
}
# URLs should *always* be osf.io/guid/
assert workidentifiers == set([
urlparse.urljoin(settings.DOMAIN, self.preprint._id) + '/',
'http://dx.doi.org/{}'.format(
self.preprint.get_identifier('doi').value)
])
assert nodes == {}
def test_format_preprint_is_deleted(self):
CASES = {
'is_published': (True, False),
'is_published': (False, True),
'node.is_public': (True, False),
'node.is_public': (False, True),
'node._is_preprint_orphan': (True, True),
'node._is_preprint_orphan': (False, False),
'node.is_deleted': (True, True),
'node.is_deleted': (False, False),
}
for key, (value, is_deleted) in CASES.items():
target = self.preprint
for k in key.split('.')[:-1]:
if k:
target = getattr(target, k)
orig_val = getattr(target, key.split('.')[-1])
setattr(target, key.split('.')[-1], value)
res = format_preprint(self.preprint,
self.preprint.provider.share_publish_type)
preprint = next(v for v in res if v['@type'] == 'preprint')
assert preprint['is_deleted'] is is_deleted
setattr(target, key.split('.')[-1], orig_val)
def test_format_preprint_is_deleted_true_if_qatest_tag_is_added(self):
res = format_preprint(self.preprint,
self.preprint.provider.share_publish_type)
preprint = next(v for v in res if v['@type'] == 'preprint')
assert preprint['is_deleted'] is False
self.preprint.node.add_tag('qatest', auth=self.auth, save=True)
res = format_preprint(self.preprint,
self.preprint.provider.share_publish_type)
preprint = next(v for v in res if v['@type'] == 'preprint')
assert preprint['is_deleted'] is True
def test_unregistered_users_guids(self):
user = UserFactory.build(is_registered=False)
user.save()
node = format_user(user)
assert {x.attrs['uri']
for x in node.get_related()} == {user.absolute_url}
def test_verified_orcid(self):
user = UserFactory.build(is_registered=True)
user.external_identity = {'ORCID': {'fake-orcid': 'VERIFIED'}}
user.save()
node = format_user(user)
assert {x.attrs['uri']
for x in node.get_related()} == {
'fake-orcid', user.absolute_url,
user.profile_image_url()
}
def test_unverified_orcid(self):
user = UserFactory.build(is_registered=True)
user.external_identity = {'ORCID': {'fake-orcid': 'SOMETHINGELSE'}}
user.save()
node = format_user(user)
assert {x.attrs['uri']
for x in node.get_related()
} == {user.absolute_url,
user.profile_image_url()}
class TestOnPreprintUpdatedTask(OsfTestCase):
def setUp(self):
super(TestOnPreprintUpdatedTask, self).setUp()
self.user = AuthUserFactory()
if len(self.user.fullname.split(' ')) > 2:
# Prevent unexpected keys ('suffix', 'additional_name')
self.user.fullname = 'David Davidson'
self.user.middle_names = ''
self.user.suffix = ''
self.user.save()
self.auth = Auth(user=self.user)
self.preprint = PreprintFactory()
thesis_provider = PreprintProviderFactory(share_publish_type='Thesis')
self.thesis = PreprintFactory(provider=thesis_provider)
for pp in [self.preprint, self.thesis]:
pp.node.add_tag('preprint', self.auth, save=False)
pp.node.add_tag('spoderman', self.auth, save=False)
pp.node.add_unregistered_contributor('BoJack Horseman', '*****@*****.**', Auth(pp.node.creator))
pp.node.add_contributor(self.user, visible=False)
pp.node.save()
pp.node.creator.given_name = u'ZZYZ'
if len(pp.node.creator.fullname.split(' ')) > 2:
# Prevent unexpected keys ('suffix', 'additional_name')
pp.node.creator.fullname = 'David Davidson'
pp.node.creator.middle_names = ''
pp.node.creator.suffix = ''
pp.node.creator.save()
pp.set_subjects([[SubjectFactory()._id]], auth=Auth(pp.node.creator))
def tearDown(self):
handlers.celery_before_request()
super(TestOnPreprintUpdatedTask, self).tearDown()
def test_update_or_enqueue_on_preprint_updated(self):
first_subjects = [15]
update_or_enqueue_on_preprint_updated(
self.preprint._id,
old_subjects=first_subjects,
saved_fields={'contributors': True}
)
second_subjects = [16, 17]
update_or_enqueue_on_preprint_updated(
self.preprint._id,
old_subjects=second_subjects,
saved_fields={'title': 'Hello'}
)
updated_task = get_task_from_postcommit_queue(
'website.preprints.tasks.on_preprint_updated',
predicate=lambda task: task.kwargs['preprint_id'] == self.preprint._id
)
assert 'title' in updated_task.kwargs['saved_fields']
assert 'contributors' in updated_task.kwargs['saved_fields']
assert set(first_subjects + second_subjects).issubset(updated_task.kwargs['old_subjects'])
def test_format_preprint(self):
res = format_preprint(self.preprint, self.preprint.provider.share_publish_type)
assert set(gn['@type'] for gn in res) == {'creator', 'contributor', 'throughsubjects', 'subject', 'throughtags', 'tag', 'workidentifier', 'agentidentifier', 'person', 'preprint', 'workrelation', 'creativework'}
nodes = dict(enumerate(res))
preprint = nodes.pop(next(k for k, v in nodes.items() if v['@type'] == 'preprint'))
assert preprint['title'] == self.preprint.node.title
assert preprint['description'] == self.preprint.node.description
assert preprint['is_deleted'] == (not self.preprint.is_published or not self.preprint.node.is_public or self.preprint.node.is_preprint_orphan)
assert preprint['date_updated'] == self.preprint.modified.isoformat()
assert preprint['date_published'] == self.preprint.date_published.isoformat()
tags = [nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'tag']
through_tags = [nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'throughtags']
assert sorted(tag['@id'] for tag in tags) == sorted(tt['tag']['@id'] for tt in through_tags)
assert sorted(tag['name'] for tag in tags) == ['preprint', 'spoderman']
subjects = [nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'subject']
through_subjects = [nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'throughsubjects']
s_ids = [s['@id'] for s in subjects]
ts_ids = [ts['subject']['@id'] for ts in through_subjects]
cs_ids = [i for i in set(s.get('central_synonym', {}).get('@id') for s in subjects) if i]
for ts in ts_ids:
assert ts in s_ids
assert ts not in cs_ids # Only aliased subjects are connected to self.preprint
for s in subjects:
subject = Subject.objects.get(text=s['name'])
assert s['uri'].endswith('v2/taxonomies/{}/'.format(subject._id)) # This cannot change
assert set(subject['name'] for subject in subjects) == set([s.text for s in self.preprint.subjects.all()] + [s.bepress_subject.text for s in self.preprint.subjects.filter(bepress_subject__isnull=False)])
people = sorted([nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'person'], key=lambda x: x['given_name'])
expected_people = sorted([{
'@type': 'person',
'given_name': u'BoJack',
'family_name': u'Horseman',
}, {
'@type': 'person',
'given_name': self.user.given_name,
'family_name': self.user.family_name,
}, {
'@type': 'person',
'given_name': self.preprint.node.creator.given_name,
'family_name': self.preprint.node.creator.family_name,
}], key=lambda x: x['given_name'])
for i, p in enumerate(expected_people):
expected_people[i]['@id'] = people[i]['@id']
assert people == expected_people
creators = sorted([nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'creator'], key=lambda x: x['order_cited'])
assert creators == [{
'@id': creators[0]['@id'],
'@type': 'creator',
'order_cited': 0,
'cited_as': u'{}'.format(self.preprint.node.creator.fullname),
'agent': {'@id': [p['@id'] for p in people if p['given_name'] == self.preprint.node.creator.given_name][0], '@type': 'person'},
'creative_work': {'@id': preprint['@id'], '@type': preprint['@type']},
}, {
'@id': creators[1]['@id'],
'@type': 'creator',
'order_cited': 1,
'cited_as': u'BoJack Horseman',
'agent': {'@id': [p['@id'] for p in people if p['given_name'] == u'BoJack'][0], '@type': 'person'},
'creative_work': {'@id': preprint['@id'], '@type': preprint['@type']},
}]
contributors = [nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'contributor']
assert contributors == [{
'@id': contributors[0]['@id'],
'@type': 'contributor',
'cited_as': u'{}'.format(self.user.fullname),
'agent': {'@id': [p['@id'] for p in people if p['given_name'] == self.user.given_name][0], '@type': 'person'},
'creative_work': {'@id': preprint['@id'], '@type': preprint['@type']},
}]
agentidentifiers = {nodes.pop(k)['uri'] for k, v in nodes.items() if v['@type'] == 'agentidentifier'}
assert agentidentifiers == set([
'mailto:' + self.user.username,
'mailto:' + self.preprint.node.creator.username,
self.user.profile_image_url(),
self.preprint.node.creator.profile_image_url(),
]) | set(user.absolute_url for user in self.preprint.node.contributors)
related_work = next(nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'creativework')
assert set(related_work.keys()) == {'@id', '@type'} # Empty except @id and @type
osf_doi = next(nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'workidentifier' and 'doi' in v['uri'] and 'osf.io' in v['uri'])
assert osf_doi['creative_work'] == {'@id': preprint['@id'], '@type': preprint['@type']}
related_doi = next(nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'workidentifier' and 'doi' in v['uri'])
assert related_doi['creative_work'] == related_work
workidentifiers = [nodes.pop(k)['uri'] for k, v in nodes.items() if v['@type'] == 'workidentifier']
assert workidentifiers == [urlparse.urljoin(settings.DOMAIN, self.preprint._id + '/')]
relation = nodes.pop(nodes.keys()[0])
assert relation == {'@id': relation['@id'], '@type': 'workrelation', 'related': {'@id': related_work['@id'], '@type': related_work['@type']}, 'subject': {'@id': preprint['@id'], '@type': preprint['@type']}}
assert nodes == {}
def test_format_thesis(self):
res = format_preprint(self.thesis, self.thesis.provider.share_publish_type)
assert set(gn['@type'] for gn in res) == {'creator', 'contributor', 'throughsubjects', 'subject', 'throughtags', 'tag', 'workidentifier', 'agentidentifier', 'person', 'thesis', 'workrelation', 'creativework'}
nodes = dict(enumerate(res))
thesis = nodes.pop(next(k for k, v in nodes.items() if v['@type'] == 'thesis'))
assert thesis['title'] == self.thesis.node.title
assert thesis['description'] == self.thesis.node.description
def test_format_preprint_date_modified_node_updated(self):
self.preprint.node.save()
res = format_preprint(self.preprint, self.preprint.provider.share_publish_type)
nodes = dict(enumerate(res))
preprint = nodes.pop(next(k for k, v in nodes.items() if v['@type'] == 'preprint'))
assert preprint['date_updated'] == self.preprint.node.modified.isoformat()
def test_format_preprint_nones(self):
self.preprint.node.tags = []
self.preprint.date_published = None
self.preprint.node.preprint_article_doi = None
self.preprint.set_subjects([], auth=Auth(self.preprint.node.creator))
res = format_preprint(self.preprint, self.preprint.provider.share_publish_type)
assert self.preprint.provider != 'osf'
assert set(gn['@type'] for gn in res) == {'creator', 'contributor', 'workidentifier', 'agentidentifier', 'person', 'preprint'}
nodes = dict(enumerate(res))
preprint = nodes.pop(next(k for k, v in nodes.items() if v['@type'] == 'preprint'))
assert preprint['title'] == self.preprint.node.title
assert preprint['description'] == self.preprint.node.description
assert preprint['is_deleted'] == (not self.preprint.is_published or not self.preprint.node.is_public or self.preprint.node.is_preprint_orphan)
assert preprint['date_updated'] == self.preprint.modified.isoformat()
assert preprint.get('date_published') is None
people = sorted([nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'person'], key=lambda x: x['given_name'])
expected_people = sorted([{
'@type': 'person',
'given_name': u'BoJack',
'family_name': u'Horseman',
}, {
'@type': 'person',
'given_name': self.user.given_name,
'family_name': self.user.family_name,
}, {
'@type': 'person',
'given_name': self.preprint.node.creator.given_name,
'family_name': self.preprint.node.creator.family_name,
}], key=lambda x: x['given_name'])
for i, p in enumerate(expected_people):
expected_people[i]['@id'] = people[i]['@id']
assert people == expected_people
creators = sorted([nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'creator'], key=lambda x: x['order_cited'])
assert creators == [{
'@id': creators[0]['@id'],
'@type': 'creator',
'order_cited': 0,
'cited_as': self.preprint.node.creator.fullname,
'agent': {'@id': [p['@id'] for p in people if p['given_name'] == self.preprint.node.creator.given_name][0], '@type': 'person'},
'creative_work': {'@id': preprint['@id'], '@type': preprint['@type']},
}, {
'@id': creators[1]['@id'],
'@type': 'creator',
'order_cited': 1,
'cited_as': u'BoJack Horseman',
'agent': {'@id': [p['@id'] for p in people if p['given_name'] == u'BoJack'][0], '@type': 'person'},
'creative_work': {'@id': preprint['@id'], '@type': preprint['@type']},
}]
contributors = [nodes.pop(k) for k, v in nodes.items() if v['@type'] == 'contributor']
assert contributors == [{
'@id': contributors[0]['@id'],
'@type': 'contributor',
'cited_as': self.user.fullname,
'agent': {'@id': [p['@id'] for p in people if p['given_name'] == self.user.given_name][0], '@type': 'person'},
'creative_work': {'@id': preprint['@id'], '@type': preprint['@type']},
}]
agentidentifiers = {nodes.pop(k)['uri'] for k, v in nodes.items() if v['@type'] == 'agentidentifier'}
assert agentidentifiers == set([
'mailto:' + self.user.username,
'mailto:' + self.preprint.node.creator.username,
self.user.profile_image_url(),
self.preprint.node.creator.profile_image_url(),
]) | set(user.absolute_url for user in self.preprint.node.contributors)
workidentifiers = {nodes.pop(k)['uri'] for k, v in nodes.items() if v['@type'] == 'workidentifier'}
# URLs should *always* be osf.io/guid/
assert workidentifiers == set([urlparse.urljoin(settings.DOMAIN, self.preprint._id) + '/', 'https://doi.org/{}'.format(self.preprint.get_identifier('doi').value)])
assert nodes == {}
def test_format_preprint_is_deleted(self):
CASES = {
'is_published': (True, False),
'is_published': (False, True),
'node.is_public': (True, False),
'node.is_public': (False, True),
'node._is_preprint_orphan': (True, True),
'node._is_preprint_orphan': (False, False),
'node.is_deleted': (True, True),
'node.is_deleted': (False, False),
}
for key, (value, is_deleted) in CASES.items():
target = self.preprint
for k in key.split('.')[:-1]:
if k:
target = getattr(target, k)
orig_val = getattr(target, key.split('.')[-1])
setattr(target, key.split('.')[-1], value)
res = format_preprint(self.preprint, self.preprint.provider.share_publish_type)
preprint = next(v for v in res if v['@type'] == 'preprint')
assert preprint['is_deleted'] is is_deleted
setattr(target, key.split('.')[-1], orig_val)
def test_format_preprint_is_deleted_true_if_qatest_tag_is_added(self):
res = format_preprint(self.preprint, self.preprint.provider.share_publish_type)
preprint = next(v for v in res if v['@type'] == 'preprint')
assert preprint['is_deleted'] is False
self.preprint.node.add_tag('qatest', auth=self.auth, save=True)
res = format_preprint(self.preprint, self.preprint.provider.share_publish_type)
preprint = next(v for v in res if v['@type'] == 'preprint')
assert preprint['is_deleted'] is True
def test_unregistered_users_guids(self):
user = UserFactory.build(is_registered=False)
user.save()
node = format_user(user)
assert {x.attrs['uri'] for x in node.get_related()} == {user.absolute_url}
def test_verified_orcid(self):
user = UserFactory.build(is_registered=True)
user.external_identity = {'ORCID': {'fake-orcid': 'VERIFIED'}}
user.save()
node = format_user(user)
assert {x.attrs['uri'] for x in node.get_related()} == {'fake-orcid', user.absolute_url, user.profile_image_url()}
def test_unverified_orcid(self):
user = UserFactory.build(is_registered=True)
user.external_identity = {'ORCID': {'fake-orcid': 'SOMETHINGELSE'}}
user.save()
node = format_user(user)
assert {x.attrs['uri'] for x in node.get_related()} == {user.absolute_url, user.profile_image_url()}
