class DescriptionSchema(Schema):
"""Schema for the additional descriptions."""
description = SanitizedHTML(required=True,
validate=validate.Length(min=3))
type = fields.Nested(VocabularySchema, required=True)
lang = fields.Nested(VocabularySchema)
def sanitize_html_fields(self, data, **kwargs):
"""Sanitize fields that may contain user-input HTML strings."""
if 'abstract' in data:
for lang, val in data.get('abstract').items():
raw = data['abstract'][lang]
data['abstract'][lang] = SanitizedHTML()._deserialize(
raw, 'abstract', data)
return data
class CommunityMetadataSchema(Schema):
"""Community metadata schema."""
COMMUNITY_TYPES = [
'organization',
'event',
'topic',
'project',
]
title = SanitizedUnicode(required=True, validate=_not_blank(max=250))
description = SanitizedUnicode(validate=_not_blank(max=2000))
curation_policy = SanitizedHTML(validate=_not_blank(max=2000))
page = SanitizedHTML(validate=_not_blank(max=2000))
# TODO: Use general small vocabularies
type = SanitizedUnicode(validate=validate.OneOf(COMMUNITY_TYPES))
website = fields.Url(validate=_not_blank())
funding = fields.List(fields.Nested(FundingSchema))
organizations = fields.List(fields.Nested(AffiliationSchema))
class UploadSchema(ModelSchema):
attachment_count = fields.Method("get_attachement_count", dump_only=True)
text_length = fields.Method("get_text_length", dump_only=True)
email = fields.Email()
phone = fields.String(validate=Regexp(r"^\+?[0-9]{6,13}$"))
text = SanitizedHTML(validate=Length(max=(2 * 1024 * 1024)),
tags=[],
attrs=[]) # ~2MB (or a lot more bc UTF-8)
address = SanitizedHTML(tags=[], attrs=[])
name = SanitizedHTML(tags=[], attrs=[])
files = fields.Nested('FileSchema', many=True, exclude=['upload'])
def get_attachement_count(self, upload: Upload) -> int:
return len(upload.files) # backref
def get_text_length(self, upload: Upload) -> int:
return len(upload.text)
class Meta:
model = Upload
class DescriptionSchema(Schema):
"""Schema for the additional descriptions."""
DESCRIPTION_TYPES = [
"abstract", "methods", "seriesinformation", "tableofcontents",
"technicalinfo", "other"
]
description = SanitizedHTML(required=True, validate=validate.Length(min=3))
type = SanitizedUnicode(
required=True,
validate=validate.OneOf(
choices=DESCRIPTION_TYPES,
error=_(
'Invalid description type. {input} not one of {choices}.')))
lang = fields.Nested(LanguageSchema)
class MetadataSchema(Schema):
"""Schema for the record metadata."""
field_load_permissions = {
# TODO: define "can_admin" action
}
field_dump_permissions = {
# TODO: define "can_admin" action
}
class Meta:
"""Meta class to accept unknwon fields."""
unknown = INCLUDE
# Metadata fields
resource_type = ResourceType(required=True)
creators = fields.List(fields.Nested(CreatorSchema),
required=True,
validate=validate.Length(
min=1,
error=_("Missing data for required field.")))
title = SanitizedUnicode(required=True, validate=validate.Length(min=3))
additional_titles = fields.List(fields.Nested(TitleSchema))
publisher = SanitizedUnicode()
publication_date = EDTFDateString(required=True)
subjects = fields.List(fields.Nested(SubjectSchema))
contributors = fields.List(fields.Nested(ContributorSchema))
dates = fields.List(fields.Nested(DateSchema))
languages = fields.List(fields.Nested(LanguageSchema))
# alternate identifiers
identifiers = IdentifierSet(
fields.Nested(partial(IdentifierSchema, allow_all=True)))
related_identifiers = fields.List(fields.Nested(RelatedIdentifierSchema))
sizes = fields.List(
SanitizedUnicode(
validate=_not_blank(_('Size cannot be a blank string.'))))
formats = fields.List(
SanitizedUnicode(
validate=_not_blank(_('Format cannot be a blank string.'))))
version = SanitizedUnicode()
rights = fields.List(fields.Nested(RightsSchema))
description = SanitizedHTML(validate=validate.Length(min=3))
additional_descriptions = fields.List(fields.Nested(DescriptionSchema))
locations = fields.List(fields.Nested(LocationSchema))
funding = fields.List(fields.Nested(FundingSchema))
references = fields.List(fields.Nested(ReferenceSchema))
class MetadataSchema(Schema):
"""Schema for the record metadata."""
# Metadata fields
resource_type = fields.Nested(VocabularySchema, required=True)
creators = fields.List(fields.Nested(CreatorSchema),
required=True,
validate=validate.Length(
min=1,
error=_("Missing data for required field.")))
title = SanitizedUnicode(required=True, validate=validate.Length(min=3))
additional_titles = fields.List(fields.Nested(TitleSchema))
publisher = SanitizedUnicode()
publication_date = EDTFDateString(required=True)
subjects = fields.List(fields.Nested(SubjectSchema))
contributors = fields.List(fields.Nested(ContributorSchema))
dates = fields.List(fields.Nested(DateSchema))
languages = fields.List(fields.Nested(VocabularySchema))
# alternate identifiers
identifiers = IdentifierSet(
fields.Nested(
partial(IdentifierSchema,
allowed_schemes=record_identifiers_schemes)))
related_identifiers = fields.List(fields.Nested(RelatedIdentifierSchema))
sizes = fields.List(
SanitizedUnicode(
validate=_not_blank(_('Size cannot be a blank string.'))))
formats = fields.List(
SanitizedUnicode(
validate=_not_blank(_('Format cannot be a blank string.'))))
version = SanitizedUnicode()
rights = fields.List(fields.Nested(RightsSchema))
description = SanitizedHTML(validate=validate.Length(min=3))
additional_descriptions = fields.List(fields.Nested(DescriptionSchema))
locations = fields.Nested(FeatureSchema)
funding = fields.List(fields.Nested(FundingSchema))
references = fields.List(fields.Nested(ReferenceSchema))
class FileSchema(ModelSchema):
original_filename = SanitizedHTML(tags=[], attrs=[])
mime = fields.String(validate=validate_mimetype)
class Meta:
model = File