class DocumentUpdateSchema(Schema):
countries = List(Country())
title = String(allow_none=True)
summary = String(allow_none=True)
languages = List(Language())
keywords = List(String(validate=Length(min=1, max=5000)))
date = PartialDate(allow_none=True)
authored_at = PartialDate(allow_none=True)
modified_at = PartialDate(allow_none=True)
published_at = PartialDate(allow_none=True)
retrieved_at = PartialDate(allow_none=True)
file_name = String(allow_none=True)
author = String(allow_none=True)
generator = String(allow_none=True)
mime_type = String(allow_none=True)
source_url = String(allow_none=True)
class ShallowCombinedSchema(BaseSchema):
collection_id = String()
# Joint entity/document attributes
collection = Nested(CollectionSchema())
schema = SchemaName()
schemata = List(SchemaName())
names = List(String())
addresses = List(String())
phones = List(String())
emails = List(String())
identifiers = List(String())
countries = List(Country())
dates = List(PartialDate())
bulk = Boolean()
# Entity attributes
foreign_id = String()
name = String()
entities = List(String())
properties = Dict()
# Document attributes
status = String()
content_hash = String()
uploader_id = String()
uploader = Nested(RoleReferenceSchema())
error_message = String()
title = String()
summary = String()
languages = List(Language())
keywords = List(String())
date = PartialDate()
authored_at = PartialDate()
modified_at = PartialDate()
published_at = PartialDate()
retrieved_at = PartialDate()
file_name = String()
file_size = Integer()
author = String()
generator = String()
mime_type = String()
extension = String()
encoding = String()
source_url = String()
pdf_version = String()
columns = List(String())
headers = Dict()
children = Integer()
# TODO: is this a separate endpoint?
text = String()
html = String()
def document_links(self, data, pk, schemata):
links = {
'self': url_for('documents_api.view', document_id=pk),
'tags': url_for('entities_api.tags', id=pk),
'ui': document_url(pk)
}
if data.get('content_hash'):
links['file'] = url_for('documents_api.file',
document_id=pk,
_authorize=True)
if schemata.intersection([Document.SCHEMA_PDF]):
links['pdf'] = url_for('documents_api.pdf',
document_id=pk,
_authorize=True)
if schemata.intersection([Document.SCHEMA_PDF, Document.SCHEMA_TABLE]):
links['records'] = url_for('documents_api.records', document_id=pk)
if schemata.intersection([Document.SCHEMA_FOLDER]):
query = (('filter:parent.id', pk),)
links['children'] = url_for('documents_api.index', _query=query)
return links
def entity_links(self, data, pk, schemata):
return {
'self': url_for('entities_api.view', id=pk),
# 'similar': url_for('entities_api.similar', id=pk),
# 'documents': url_for('entities_api.documents', id=pk),
'references': url_for('entities_api.references', id=pk),
'tags': url_for('entities_api.tags', id=pk),
'ui': entity_url(pk)
}
@post_dump()
def hypermedia(self, data):
pk = str(data.get('id'))
collection = data.get('collection', {})
collection_id = collection.get('id')
collection_id = collection_id or data.get('collection_id')
schemata = set(data.get('schemata', []))
if Document.SCHEMA in schemata:
data['links'] = self.document_links(data, pk, schemata)
else:
data['links'] = self.entity_links(data, pk, schemata)
if data.get('bulk'):
data['writeable'] = False
else:
data['writeable'] = request.authz.can_write(collection_id)
return data