class DatasetSchemaV3(CreatorMixinSchemaV3, EntitySchemaV3):
"""Dataset schema."""
class Meta:
"""Meta class."""
rdf_type = schema.Dataset
model = Dataset
unknown = EXCLUDE
creators = fields.Nested(schema.creator, PersonSchemaV3, many=True)
date_created = fields.DateTime(schema.dateCreated, missing=None)
date_published = fields.DateTime(schema.datePublished, missing=None)
description = fields.String(schema.description, missing=None)
files = fields.Nested(schema.hasPart, DatasetFileSchemaV3, many=True)
identifier = fields.String(schema.identifier)
in_language = fields.Nested(schema.inLanguage,
LanguageSchemaV5,
missing=None)
keywords = fields.List(schema.keywords, fields.String())
license = fields.Uri(schema.license, missing=None, allow_none=True)
name = fields.String(schema.alternateName, missing=None)
same_as = fields.Nested(schema.sameAs, UrlSchemaV5, missing=None)
tags = fields.Nested(schema.subjectOf, DatasetTagSchemaV5, many=True)
title = fields.String(schema.name)
url = fields.String(schema.url)
version = fields.String(schema.version, missing=None)
@pre_load
def fix_files_context(self, data, **kwargs):
"""Fix DatasetFile context for _label and external fields."""
from renku.core.utils.migrate import migrate_types
data = migrate_types(data)
if "@context" not in data:
return data
context = data["@context"]
if not isinstance(context, dict) or "files" not in context:
return data
context.setdefault("rdfs", "http://www.w3.org/2000/01/rdf-schema#")
context.setdefault("_label", "rdfs:label")
files = data["@context"]["files"]
if not isinstance(files, dict) or "@context" not in files:
return data
context = files["@context"]
context.setdefault("rdfs", "http://www.w3.org/2000/01/rdf-schema#")
context.setdefault("_label", "rdfs:label")
context.setdefault("external", "renku:external")
context.setdefault(
"renku",
"https://swissdatasciencecenter.github.io/renku-ontology#")
return data
class ProjectSchemaV3(JsonLDSchema):
"""Project Schema."""
class Meta:
"""Meta class."""
rdf_type = [prov.Location, schema.Project]
model = Project
unknown = EXCLUDE
_id = fields.Id(missing=None)
name = fields.String(schema.name, missing=None)
created = fields.DateTime(schema.dateCreated, missing=None)
updated = fields.DateTime(schema.dateUpdated, missing=None)
version = fields.String(schema.schemaVersion, missing=1)
creator = fields.Nested(schema.creator, PersonSchemaV3, missing=None)
class DatasetFileSchema(EntitySchema):
"""DatasetFile schema."""
class Meta:
"""Meta class."""
rdf_type = schema.DigitalDocument
model = DatasetFile
unknown = EXCLUDE
added = fields.DateTime(schema.dateCreated,
format="iso",
extra_formats=("%Y-%m-%d", ))
name = fields.String(schema.name, missing=None)
url = fields.String(schema.url, missing=None)
based_on = Nested(schema.isBasedOn,
"DatasetFileSchema",
missing=None,
propagate_client=False)
external = fields.Boolean(renku.external, missing=False)
source = fields.String(renku.source, missing=None)
@pre_dump
def fix_datetimes(self, obj, many=False, **kwargs):
"""Pre dump hook."""
if many:
return [self.fix_datetimes(o, many=False, **kwargs) for o in obj]
obj.added = self._fix_timezone(obj.added)
return obj
class DatasetTagSchema(JsonLDSchema):
"""DatasetTag schema."""
class Meta:
"""Meta class."""
rdf_type = schema.PublicationEvent
model = DatasetTag
unknown = EXCLUDE
name = fields.String(schema.name)
description = fields.String(schema.description)
commit = fields.String(schema.location)
created = fields.DateTime(schema.startDate,
missing=None,
format="iso",
extra_formats=("%Y-%m-%d", ))
dataset = fields.String(schema.about)
_id = fields.Id(init_name="id")
@pre_dump
def fix_datetimes(self, obj, many=False, **kwargs):
"""Pre dump hook."""
if many:
return [self.fix_datetimes(o, many=False, **kwargs) for o in obj]
object.__setattr__(obj, "created", self._fix_timezone(obj.created))
return obj
class DatasetSchema(EntitySchema, CreatorMixinSchema):
"""Dataset schema."""
class Meta:
"""Meta class."""
rdf_type = schema.Dataset
model = Dataset
unknown = EXCLUDE
_id = fields.Id(init_name="id", missing=None)
_label = fields.String(rdfs.label, init_name="label", missing=None)
date_published = fields.DateTime(
schema.datePublished,
missing=None,
allow_none=True,
format="%Y-%m-%d",
extra_formats=("iso", "%Y-%m-%dT%H:%M:%S"),
)
description = fields.String(schema.description, missing=None)
identifier = fields.String(schema.identifier)
in_language = Nested(schema.inLanguage, LanguageSchema, missing=None)
keywords = fields.List(schema.keywords,
fields.String(),
missing=None,
allow_none=True)
license = fields.Uri(schema.license, missing=None, allow_none=True)
title = fields.String(schema.name)
url = fields.String(schema.url)
version = fields.String(schema.version, missing=None)
date_created = fields.DateTime(schema.dateCreated,
missing=None,
allow_none=True,
format="iso",
extra_formats=("%Y-%m-%d", ))
files = Nested(schema.hasPart, DatasetFileSchema, many=True)
tags = Nested(schema.subjectOf, DatasetTagSchema, many=True)
same_as = Nested(schema.sameAs, UrlSchema, missing=None)
name = fields.String(schema.alternateName)
@pre_dump
def fix_datetimes(self, obj, many=False, **kwargs):
"""Pre dump hook."""
if many:
return [self.fix_datetimes(o, many=False, **kwargs) for o in obj]
obj.date_published = self._fix_timezone(obj.date_published)
obj.date_created = self._fix_timezone(obj.date_created)
return obj
class DatasetTagSchemaV5(JsonLDSchema):
"""DatasetTag schema."""
class Meta:
"""Meta class."""
rdf_type = schema.PublicationEvent
model = DatasetTag
unknown = EXCLUDE
_id = fields.Id()
commit = fields.String(schema.location)
created = fields.DateTime(schema.startDate, missing=None)
dataset = fields.String(schema.about)
description = fields.String(schema.description)
name = fields.String(schema.name)
class DatasetFileSchemaV3(EntitySchemaV3):
"""DatasetFile schema."""
class Meta:
"""Meta class."""
rdf_type = schema.DigitalDocument
model = DatasetFile
unknown = EXCLUDE
added = fields.DateTime(schema.dateCreated)
based_on = fields.Nested(schema.isBasedOn,
"DatasetFileSchemaV3",
missing=None)
name = fields.String(schema.name, missing=None)
url = fields.String(schema.url, missing=None)
external = fields.Boolean(renku.external, missing=False)
class ProjectSchema(JsonLDSchema):
"""Project Schema."""
class Meta:
"""Meta class."""
rdf_type = [schema.Project, prov.Location]
model = Project
unknown = EXCLUDE
name = fields.String(schema.name, missing=None)
created = fields.DateTime(schema.dateCreated, missing=None, format="iso", extra_formats=("%Y-%m-%d",))
version = fields.String(schema.schemaVersion, missing=1)
agent_version = fields.String(schema.agent, missing="pre-0.11.0")
creator = Nested(schema.creator, PersonSchema, missing=None)
_id = fields.Id(init_name="id", missing=None)
@pre_dump
def fix_datetimes(self, obj, many=False, **kwargs):
"""Pre dump hook."""
if many:
return [self.fix_datetimes(o, many=False, **kwargs) for o in obj]
obj.created = self._fix_timezone(obj.created)
return obj
class DatasetSchema(EntitySchema, CreatorMixinSchema):
"""Dataset schema."""
class Meta:
"""Meta class."""
rdf_type = schema.Dataset
model = Dataset
unknown = EXCLUDE
_id = fields.Id(init_name='id', missing=None)
_label = fields.String(rdfs.label, init_name='label', missing=None)
date_published = fields.DateTime(schema.datePublished, missing=None)
description = fields.String(schema.description, missing=None)
identifier = fields.String(schema.identifier)
in_language = fields.Nested(
schema.inLanguage, LanguageSchema, missing=None
)
keywords = fields.List(schema.keywords, fields.String())
license = fields.Uri(schema.license, missing=None, allow_none=True)
name = fields.String(schema.name)
url = fields.String(schema.url)
version = fields.String(schema.version, missing=None)
created = fields.DateTime(schema.dateCreated, missing=None)
files = fields.Nested(schema.hasPart, DatasetFileSchema, many=True)
tags = fields.Nested(schema.subjectOf, DatasetTagSchema, many=True)
same_as = fields.Nested(schema.sameAs, UrlSchema, missing=None)
short_name = fields.String(schema.alternateName)
@pre_load
def fix_files_context(self, data, **kwargs):
"""Fix DatasetFile context for _label and external fields."""
context = None
if '@context' not in data:
return data
context = data['@context']
if not isinstance(context, dict) or 'files' not in context:
return data
context.setdefault('rdfs', 'http://www.w3.org/2000/01/rdf-schema#')
files = data['@context']['files']
if not isinstance(files, dict) or '@context' not in files:
return data
context = files['@context']
context.setdefault('rdfs', 'http://www.w3.org/2000/01/rdf-schema#')
context.setdefault('_label', 'rdfs:label')
context.setdefault('external', 'renku:external')
context.setdefault(
'renku', 'https://swissdatasciencecenter.github.io/renku-ontology#'
)
return data
@pre_load
def migrate_types(self, data, **kwargs):
"""Fix types."""
from renku.core.utils.migrate import migrate_types
return migrate_types(data)
