def format_preprint(preprint):
preprint_graph = GraphNode('preprint', **{
'title': preprint.node.title,
'description': preprint.node.description or '',
'is_deleted': (
not preprint.is_published or
not preprint.node.is_public or
preprint.node.is_preprint_orphan or
preprint.node.tags.filter(name='qatest').exists() or
preprint.node.is_deleted
),
'date_updated': preprint.date_modified.isoformat(),
'date_published': preprint.date_published.isoformat() if preprint.date_published else None
})
to_visit = [
preprint_graph,
GraphNode('workidentifier', creative_work=preprint_graph, uri=urlparse.urljoin(settings.DOMAIN, preprint._id + '/'))
]
if preprint.get_identifier('doi'):
to_visit.append(GraphNode('workidentifier', creative_work=preprint_graph, uri='http://dx.doi.org/{}'.format(preprint.get_identifier('doi').value)))
if preprint.provider.domain_redirect_enabled:
to_visit.append(GraphNode('workidentifier', creative_work=preprint_graph, uri=preprint.absolute_url))
if preprint.article_doi:
# Article DOI refers to a clone of this preprint on another system and therefore does not qualify as an identifier for this preprint
related_work = GraphNode('creativework')
to_visit.append(GraphNode('workrelation', subject=preprint_graph, related=related_work))
to_visit.append(GraphNode('workidentifier', creative_work=related_work, uri='http://dx.doi.org/{}'.format(preprint.article_doi)))
preprint_graph.attrs['tags'] = [
GraphNode('throughtags', creative_work=preprint_graph, tag=GraphNode('tag', name=tag))
for tag in preprint.node.tags.values_list('name', flat=True) if tag
]
preprint_graph.attrs['subjects'] = [
GraphNode('throughsubjects', creative_work=preprint_graph, subject=GraphNode('subject', name=subject))
for subject in set(s.bepress_text for s in preprint.subjects.all())
]
to_visit.extend(format_contributor(preprint_graph, user, preprint.node.get_visible(user), i) for i, user in enumerate(preprint.node.contributors))
to_visit.extend(GraphNode('AgentWorkRelation', creative_work=preprint_graph, agent=GraphNode('institution', name=institution))
for institution in preprint.node.affiliated_institutions.values_list('name', flat=True))
visited = set()
to_visit.extend(preprint_graph.get_related())
while True:
if not to_visit:
break
n = to_visit.pop(0)
if n in visited:
continue
visited.add(n)
to_visit.extend(list(n.get_related()))
return [node.serialize() for node in visited]
def format_registration(node):
is_qa_node = bool(set(settings.DO_NOT_INDEX_LIST['tags']).intersection(node.tags.all().values_list('name', flat=True))) \
or any(substring in node.title for substring in settings.DO_NOT_INDEX_LIST['titles'])
registration_graph = GraphNode(
'registration', **{
'title':
node.title,
'description':
node.description or '',
'is_deleted':
not node.is_public or node.is_deleted or is_qa_node,
'date_published':
node.registered_date.isoformat() if node.registered_date else None,
'registration_type':
node.registered_schema.first().name
if node.registered_schema else None,
'withdrawn':
node.is_retracted,
'justification':
node.retraction.justification if node.retraction else None,
})
to_visit = [
registration_graph,
GraphNode('workidentifier',
creative_work=registration_graph,
uri=urlparse.urljoin(settings.DOMAIN, node.url))
]
registration_graph.attrs['tags'] = [
GraphNode('throughtags',
creative_work=registration_graph,
tag=GraphNode('tag', name=tag._id))
for tag in node.tags.all() or [] if tag._id
]
to_visit.extend(
format_contributor(registration_graph, user,
bool(user._id in node.visible_contributor_ids), i)
for i, user in enumerate(node.contributors))
to_visit.extend(
GraphNode('AgentWorkRelation',
creative_work=registration_graph,
agent=GraphNode('institution', name=institution.name))
for institution in node.affiliated_institutions.all())
visited = set()
to_visit.extend(registration_graph.get_related())
while True:
if not to_visit:
break
n = to_visit.pop(0)
if n in visited:
continue
visited.add(n)
to_visit.extend(list(n.get_related()))
return [node_.serialize() for node_ in visited]
def format_registration(node):
# TODO: Add parent and root info?
registration_graph = GraphNode(
'registration',
**{
'title':
node.title,
'description':
node.description or '',
'is_deleted':
not node.is_public or 'qatest' in (node.tags or [])
or node.is_deleted,
'date_published':
node.registered_date.isoformat() if node.registered_date else None,
'registration_type':
node.registered_schema[0].name if node.registered_schema else None,
'withdrawn':
node.is_retracted,
# TODO: Should this recurse up to the node's root or nah?
'justification':
node.retraction.justification if node.retraction else None,
})
to_visit = [
registration_graph,
GraphNode('workidentifier',
creative_work=registration_graph,
uri=urlparse.urljoin(settings.DOMAIN, node.url))
]
registration_graph.attrs['tags'] = [
GraphNode('throughtags',
creative_work=registration_graph,
tag=GraphNode('tag', name=tag._id))
for tag in node.tags or [] if tag._id
]
to_visit.extend(
format_contributor(registration_graph, user,
bool(user._id in node.visible_contributor_ids), i)
for i, user in enumerate(node.contributors))
to_visit.extend(
GraphNode('AgentWorkRelation',
creative_work=registration_graph,
agent=GraphNode('institution', name=institution.name))
for institution in node.affiliated_institutions)
visited = set()
to_visit.extend(registration_graph.get_related())
while True:
if not to_visit:
break
n = to_visit.pop(0)
if n in visited:
continue
visited.add(n)
to_visit.extend(list(n.get_related()))
return [node_.serialize() for node_ in visited]
def format_registration(node):
is_qa_node = bool(set(settings.DO_NOT_INDEX_LIST['tags']).intersection(node.tags.all().values_list('name', flat=True))) \
or any(substring in node.title for substring in settings.DO_NOT_INDEX_LIST['titles'])
registration_graph = GraphNode('registration', **{
'title': node.title,
'description': node.description or '',
'is_deleted': not node.is_public or node.is_deleted or is_qa_node,
'date_published': node.registered_date.isoformat() if node.registered_date else None,
'registration_type': node.registered_schema.first().name if node.registered_schema else None,
'withdrawn': node.is_retracted,
'justification': node.retraction.justification if node.retraction else None,
})
to_visit = [
registration_graph,
GraphNode('workidentifier', creative_work=registration_graph, uri=urlparse.urljoin(settings.DOMAIN, node.url))
]
registration_graph.attrs['tags'] = [
GraphNode('throughtags', creative_work=registration_graph, tag=GraphNode('tag', name=tag._id))
for tag in node.tags.all() or [] if tag._id
]
to_visit.extend(format_contributor(registration_graph, user, bool(user._id in node.visible_contributor_ids), i) for i, user in enumerate(node.contributors))
to_visit.extend(GraphNode('AgentWorkRelation', creative_work=registration_graph, agent=GraphNode('institution', name=institution.name)) for institution in node.affiliated_institutions.all())
if node.parent_node:
parent = GraphNode('registration')
to_visit.extend([
parent,
GraphNode('workidentifier', creative_work=parent, uri=urlparse.urljoin(settings.DOMAIN, node.parent_node.url)),
GraphNode('ispartof', subject=registration_graph, related=parent),
])
visited = set()
to_visit.extend(registration_graph.get_related())
while True:
if not to_visit:
break
n = to_visit.pop(0)
if n in visited:
continue
visited.add(n)
to_visit.extend(list(n.get_related()))
return [node_.serialize() for node_ in visited]
def format_preprint(preprint, share_type, old_subjects=None):
if old_subjects is None:
old_subjects = []
from osf.models import Subject
old_subjects = [Subject.objects.get(id=s) for s in old_subjects]
preprint_graph = GraphNode(
share_type,
**{
'title':
preprint.node.title,
'description':
preprint.node.description or '',
'is_deleted':
(not preprint.verified_publishable
or preprint.node.tags.filter(name='qatest').exists()),
# Note: Changing any preprint attribute that is pulled from the node, like title, will NOT bump
# the preprint's date modified but will bump the node's date_modified.
# We have to send the latest date to SHARE to actually get the result to be updated.
# If we send a date_updated that is <= the one we previously sent, SHARE will ignore any changes
# because it looks like a race condition that arose from preprints being resent to SHARE on
# every step of preprint creation.
'date_updated':
max(preprint.modified, preprint.node.modified).isoformat(),
'date_published':
preprint.date_published.isoformat()
if preprint.date_published else None
})
to_visit = [
preprint_graph,
GraphNode('workidentifier',
creative_work=preprint_graph,
uri=urlparse.urljoin(settings.DOMAIN, preprint._id + '/'))
]
if preprint.get_identifier('doi'):
to_visit.append(
GraphNode('workidentifier',
creative_work=preprint_graph,
uri='http://dx.doi.org/{}'.format(
preprint.get_identifier('doi').value)))
if preprint.provider.domain_redirect_enabled:
to_visit.append(
GraphNode('workidentifier',
creative_work=preprint_graph,
uri=preprint.absolute_url))
if preprint.article_doi:
# Article DOI refers to a clone of this preprint on another system and therefore does not qualify as an identifier for this preprint
related_work = GraphNode('creativework')
to_visit.append(
GraphNode('workrelation',
subject=preprint_graph,
related=related_work))
to_visit.append(
GraphNode('workidentifier',
creative_work=related_work,
uri='http://dx.doi.org/{}'.format(preprint.article_doi)))
preprint_graph.attrs['tags'] = [
GraphNode('throughtags',
creative_work=preprint_graph,
tag=GraphNode('tag', name=tag))
for tag in preprint.node.tags.values_list('name', flat=True) if tag
]
current_subjects = [
GraphNode('throughsubjects',
creative_work=preprint_graph,
is_deleted=False,
subject=format_subject(s)) for s in preprint.subjects.all()
]
deleted_subjects = [
GraphNode('throughsubjects',
creative_work=preprint_graph,
is_deleted=True,
subject=format_subject(s)) for s in old_subjects
if not preprint.subjects.filter(id=s.id).exists()
]
preprint_graph.attrs['subjects'] = current_subjects + deleted_subjects
to_visit.extend(
format_contributor(preprint_graph, user, preprint.node.get_visible(
user), i) for i, user in enumerate(preprint.node.contributors))
to_visit.extend(
GraphNode('AgentWorkRelation',
creative_work=preprint_graph,
agent=GraphNode('institution', name=institution))
for institution in preprint.node.affiliated_institutions.values_list(
'name', flat=True))
visited = set()
to_visit.extend(preprint_graph.get_related())
while True:
if not to_visit:
break
n = to_visit.pop(0)
if n in visited:
continue
visited.add(n)
to_visit.extend(list(n.get_related()))
return [node.serialize() for node in visited]
def format_preprint(preprint, share_type, old_subjects=None):
if old_subjects is None:
old_subjects = []
from osf.models import Subject
old_subjects = [Subject.objects.get(id=s) for s in old_subjects]
preprint_graph = GraphNode(
share_type, **{
'title':
preprint.title,
'description':
preprint.description or '',
'is_deleted':
((not preprint.verified_publishable and not preprint.is_retracted)
or preprint.tags.filter(name='qatest').exists()),
'date_updated':
preprint.modified.isoformat(),
'date_published':
preprint.date_published.isoformat()
if preprint.date_published else None
})
to_visit = [
preprint_graph,
GraphNode('workidentifier',
creative_work=preprint_graph,
uri=urlparse.urljoin(settings.DOMAIN, preprint._id + '/'))
]
if preprint.get_identifier('doi'):
to_visit.append(
GraphNode('workidentifier',
creative_work=preprint_graph,
uri='https://doi.org/{}'.format(
preprint.get_identifier('doi').value)))
if preprint.provider.domain_redirect_enabled:
to_visit.append(
GraphNode('workidentifier',
creative_work=preprint_graph,
uri=preprint.absolute_url))
if preprint.article_doi:
# Article DOI refers to a clone of this preprint on another system and therefore does not qualify as an identifier for this preprint
related_work = GraphNode('creativework')
to_visit.append(
GraphNode('workrelation',
subject=preprint_graph,
related=related_work))
to_visit.append(
GraphNode('workidentifier',
creative_work=related_work,
uri='https://doi.org/{}'.format(preprint.article_doi)))
preprint_graph.attrs['tags'] = [
GraphNode('throughtags',
creative_work=preprint_graph,
tag=GraphNode('tag', name=tag))
for tag in preprint.tags.values_list('name', flat=True) if tag
]
current_subjects = [
GraphNode('throughsubjects',
creative_work=preprint_graph,
is_deleted=False,
subject=format_subject(s)) for s in preprint.subjects.all()
]
deleted_subjects = [
GraphNode('throughsubjects',
creative_work=preprint_graph,
is_deleted=True,
subject=format_subject(s)) for s in old_subjects
if not preprint.subjects.filter(id=s.id).exists()
]
preprint_graph.attrs['subjects'] = current_subjects + deleted_subjects
to_visit.extend(
format_contributor(preprint_graph, user, preprint.get_visible(user), i)
for i, user in enumerate(preprint.contributors))
visited = set()
to_visit.extend(preprint_graph.get_related())
while True:
if not to_visit:
break
n = to_visit.pop(0)
if n in visited:
continue
visited.add(n)
to_visit.extend(list(n.get_related()))
return [node.serialize() for node in visited]
def format_preprint(preprint):
preprint_graph = GraphNode(
'preprint', **{
'title':
preprint.node.title,
'description':
preprint.node.description or '',
'is_deleted':
not preprint.is_published or not preprint.node.is_public
or preprint.node.is_preprint_orphan
or 'qatest' in (preprint.node.tags or [])
or preprint.node.is_deleted,
'date_updated':
preprint.date_modified.isoformat(),
'date_published':
preprint.date_published.isoformat()
if preprint.date_published else None
})
to_visit = [
preprint_graph,
GraphNode('workidentifier',
creative_work=preprint_graph,
uri=urlparse.urljoin(settings.DOMAIN, preprint.url))
]
if preprint.article_doi:
to_visit.append(
GraphNode('workidentifier',
creative_work=preprint_graph,
uri='http://dx.doi.org/{}'.format(preprint.article_doi)))
preprint_graph.attrs['tags'] = [
GraphNode('throughtags',
creative_work=preprint_graph,
tag=GraphNode('tag', name=tag._id))
for tag in preprint.node.tags or [] if tag._id
]
preprint_graph.attrs['subjects'] = [
GraphNode('throughsubjects',
creative_work=preprint_graph,
subject=GraphNode('subject', name=subject))
for subject in set(
x['text'] for hier in preprint.get_subjects() or [] for x in hier)
if subject
]
to_visit.extend(
format_contributor(
preprint_graph, user,
bool(user._id in preprint.node.visible_contributor_ids), i)
for i, user in enumerate(preprint.node.contributors))
to_visit.extend(
GraphNode('AgentWorkRelation',
creative_work=preprint_graph,
agent=GraphNode('institution', name=institution.name))
for institution in preprint.node.affiliated_institutions)
visited = set()
to_visit.extend(preprint_graph.get_related())
while True:
if not to_visit:
break
n = to_visit.pop(0)
if n in visited:
continue
visited.add(n)
to_visit.extend(list(n.get_related()))
return [node.serialize() for node in visited]
def format_preprint(preprint):
preprint_graph = GraphNode(
'preprint', **{
'title':
preprint.node.title,
'description':
preprint.node.description or '',
'is_deleted':
(not preprint.is_published or not preprint.node.is_public
or preprint.node.is_preprint_orphan
or preprint.node.tags.filter(name='qatest').exists()
or preprint.node.is_deleted),
'date_updated':
preprint.date_modified.isoformat(),
'date_published':
preprint.date_published.isoformat()
if preprint.date_published else None
})
to_visit = [
preprint_graph,
GraphNode('workidentifier',
creative_work=preprint_graph,
uri=urlparse.urljoin(settings.DOMAIN, preprint._id + '/'))
]
if preprint.provider.domain_redirect_enabled:
to_visit.append(
GraphNode('workidentifier',
creative_work=preprint_graph,
uri=preprint.absolute_url))
if preprint.article_doi:
# Article DOI refers to a clone of this preprint on another system and therefore does not qualify as an identifier for this preprint
related_work = GraphNode('creativework')
to_visit.append(
GraphNode('workrelation',
subject=preprint_graph,
related=related_work))
to_visit.append(
GraphNode('workidentifier',
creative_work=related_work,
uri='http://dx.doi.org/{}'.format(preprint.article_doi)))
preprint_graph.attrs['tags'] = [
GraphNode('throughtags',
creative_work=preprint_graph,
tag=GraphNode('tag', name=tag))
for tag in preprint.node.tags.values_list('name', flat=True) if tag
]
preprint_graph.attrs['subjects'] = [
GraphNode('throughsubjects',
creative_work=preprint_graph,
subject=GraphNode('subject', name=subject))
for subject in set(s.bepress_text for s in preprint.subjects.all())
]
to_visit.extend(
format_contributor(preprint_graph, user, preprint.node.get_visible(
user), i) for i, user in enumerate(preprint.node.contributors))
to_visit.extend(
GraphNode('AgentWorkRelation',
creative_work=preprint_graph,
agent=GraphNode('institution', name=institution))
for institution in preprint.node.affiliated_institutions.values_list(
'name', flat=True))
visited = set()
to_visit.extend(preprint_graph.get_related())
while True:
if not to_visit:
break
n = to_visit.pop(0)
if n in visited:
continue
visited.add(n)
to_visit.extend(list(n.get_related()))
return [node.serialize() for node in visited]
def format_preprint(preprint, share_type, old_subjects=None):
if old_subjects is None:
old_subjects = []
from osf.models import Subject
old_subjects = [Subject.objects.get(id=s) for s in old_subjects]
preprint_graph = GraphNode(share_type, **{
'title': preprint.node.title,
'description': preprint.node.description or '',
'is_deleted': (
not preprint.verified_publishable or
preprint.node.tags.filter(name='qatest').exists()
),
# Note: Changing any preprint attribute that is pulled from the node, like title, will NOT bump
# the preprint's date modified but will bump the node's date_modified.
# We have to send the latest date to SHARE to actually get the result to be updated.
# If we send a date_updated that is <= the one we previously sent, SHARE will ignore any changes
# because it looks like a race condition that arose from preprints being resent to SHARE on
# every step of preprint creation.
'date_updated': max(preprint.modified, preprint.node.modified).isoformat(),
'date_published': preprint.date_published.isoformat() if preprint.date_published else None
})
to_visit = [
preprint_graph,
GraphNode('workidentifier', creative_work=preprint_graph, uri=urlparse.urljoin(settings.DOMAIN, preprint._id + '/'))
]
if preprint.get_identifier('doi'):
to_visit.append(GraphNode('workidentifier', creative_work=preprint_graph, uri='http://dx.doi.org/{}'.format(preprint.get_identifier('doi').value)))
if preprint.provider.domain_redirect_enabled:
to_visit.append(GraphNode('workidentifier', creative_work=preprint_graph, uri=preprint.absolute_url))
if preprint.article_doi:
# Article DOI refers to a clone of this preprint on another system and therefore does not qualify as an identifier for this preprint
related_work = GraphNode('creativework')
to_visit.append(GraphNode('workrelation', subject=preprint_graph, related=related_work))
to_visit.append(GraphNode('workidentifier', creative_work=related_work, uri='http://dx.doi.org/{}'.format(preprint.article_doi)))
preprint_graph.attrs['tags'] = [
GraphNode('throughtags', creative_work=preprint_graph, tag=GraphNode('tag', name=tag))
for tag in preprint.node.tags.values_list('name', flat=True) if tag
]
current_subjects = [
GraphNode('throughsubjects', creative_work=preprint_graph, is_deleted=False, subject=format_subject(s))
for s in preprint.subjects.all()
]
deleted_subjects = [
GraphNode('throughsubjects', creative_work=preprint_graph, is_deleted=True, subject=format_subject(s))
for s in old_subjects if not preprint.subjects.filter(id=s.id).exists()
]
preprint_graph.attrs['subjects'] = current_subjects + deleted_subjects
to_visit.extend(format_contributor(preprint_graph, user, preprint.node.get_visible(user), i) for i, user in enumerate(preprint.node.contributors))
to_visit.extend(GraphNode('AgentWorkRelation', creative_work=preprint_graph, agent=GraphNode('institution', name=institution))
for institution in preprint.node.affiliated_institutions.values_list('name', flat=True))
visited = set()
to_visit.extend(preprint_graph.get_related())
while True:
if not to_visit:
break
n = to_visit.pop(0)
if n in visited:
continue
visited.add(n)
to_visit.extend(list(n.get_related()))
return [node.serialize() for node in visited]
