# coding: utf-8
"""
SDI - Semantic Data Interconnect APIs
The Semantic Data Interconnect (SDI) is a collection of APIs that allows the user to unlock the potential of disparate big data by connecting external data. The SDI can infer the schemas of data based on schema-on-read, allow creating a semantic model and perform big data semantic queries. It seamlessly connects to MindSphere's Integrated Data Lake (IDL), but it can work independently as well. There are two mechanisms that can be used to upload files so that SDI can generate schemas and make data ready for query. The SDI operations are divided into the following groups: **Data Registration for SDI** This set of APIs is used to organize the incoming data. When configuring a Data Registry, you have the option to update your data based on a replace or append strategy. If you consider a use case where schema may change and incoming data files are completely different every time then replace is a good strategy. The replace strategy will replace the existing schema and data during each data ingest operation whereas the append strategy will update the existing schema and data during each data ingest operation. **Custom Data Type for SDI** The SDI by default identifies basic data types for each property, such as String, Integer, Float, Date, etc. The user can use this set of APIs to create their own custom data type. The SDI also provides an API under this category to suggest data type based on user-provided sample test values. **Data Lake for SDI** The SDI can process files uploaded provides endpoints to manage customer's data lake registration based on tenant id, cloud provider and data lake type. The set of REST endpoint allows to create, update and retrieve base path for their data lake. The IDL customer needs to create an SDI folder that is under the root folder. Any file uploaded in this folder is automatically picked up by SDI to process via IDL notification. **Data Ingest for SDI** This set of APIs allows user to upload files, start an ingest job for uploaded files, find job status for ingested jobs or retrieve all job statuses. **Schema Registry for SDI** The SDI provides a way to find the generated schema in this category. Users can find an SDI generated schema for uploaded files based on source name, data tag or schema name. **Data Query for SDI** allows querying based on the extracted schemas. Important supported APIs are: * Query interface for querying semantically correlated and transformed data. * Stores and executes data queries. * Uses a semantic model to translate model-based query to physical queries. **Semantic Model for SDI** allows user to create semantic model ontologies based on the extracted one or more schemas. The important functionalities achieved with APIs are: * Contextual correlation of data from different systems. * Infers & Recommends mappings between different schemas. * Import and store Semantic model. # noqa: E501
"""
from __future__ import absolute_import
from mindsphere_core.mindsphere_core import logger
from mindsphere_core import mindsphere_core, exceptions, token_service
from mindsphere_core.token_service import init_credentials
[docs]class DataRegistryClient:
__base_path__ = '/api/sdi/v4'
__model_package__ = __name__.split('.')[0]
def __init__(self, rest_client_config=None, mindsphere_credentials=None):
self.rest_client_config = rest_client_config
self.mindsphere_credentials = init_credentials(mindsphere_credentials)
[docs] def get_data_registries(self, request_object):
"""Retrieves Data Registry entries
Retrieves all Data Registry entries, Data Registry based on sourceName, dataTag or combination of sourceName and dataTag.
:param DataRegistriesGetRequest request_object: It contains the below parameters --> |br| ( dataTag - dataTag ), |br| ( sourceName - sourceName ), |br| ( pageToken - Selects next page. Value must be taken rom response body property 'page.nextToken'. If omitted, first page is returned. )
:return: ListOfRegistryResponse
"""
logger.info('DataRegistryClient.get_data_registries() invoked.')
end_point_url = '/dataRegistries'
end_point_url = end_point_url.format()
token = token_service.fetch_token(self.rest_client_config, self.mindsphere_credentials)
api_url = mindsphere_core.build_url(self.__base_path__, end_point_url, self.rest_client_config)
headers = {'Accept': '*/*', 'Authorization': 'Bearer ' + str(token)}
query_params = {'dataTag': request_object.data_tag, 'sourceName': request_object.source_name, 'pageToken': request_object.page_token}
form_params, local_var_files, body_params = {}, {}, None
logger.info('DataRegistryClient.get_data_registries() --> Proceeding for API Invoker.')
return mindsphere_core.invoke_service(self.rest_client_config, api_url, headers, 'GET', query_params, form_params, body_params, local_var_files, 'ListOfRegistryResponse', self.__model_package__)
[docs] def get_data_registries_id(self, id):
"""Retrieves Data Registry for a given registry id
Retrieves registry for a registry id
:param DataRegistriesIdGetRequest request_object: It contains the below parameters --> |br| ( id* - Unique identifier of the Data Registry. )
:return: DataRegistry
"""
logger.info('DataRegistryClient.get_data_registries_id() invoked.')
if id is None:
raise exceptions.MindsphereClientError('`id` is not passed when calling `get_data_registries_id`')
end_point_url = '/dataRegistries/{id}'
end_point_url = end_point_url.format(id=id)
token = token_service.fetch_token(self.rest_client_config, self.mindsphere_credentials)
api_url = mindsphere_core.build_url(self.__base_path__, end_point_url, self.rest_client_config)
headers = {'Accept': 'application/json', 'Authorization': 'Bearer ' + str(token)}
query_params = {}
form_params, local_var_files, body_params = {}, {}, None
logger.info('DataRegistryClient.get_data_registries_id() --> Proceeding for API Invoker.')
return mindsphere_core.invoke_service(self.rest_client_config, api_url, headers, 'GET', query_params, form_params, body_params, local_var_files, 'DataRegistry', self.__model_package__)
[docs] def update_data_registries_id(self, id, request_object):
"""Updates a Data Registry
Update Data Registry entries for a given Data Registry Id. The registry attributes sourceName, dataTag and partitionKeys cannot be updated.
:param DataRegistriesIdPatchRequest request_object: It contains the below parameters --> |br| ( id* - Unique identifier of the Data Registry. ), |br| ( updateDataRegistryRequest* - updateDataRegistryRequest )
:return: DataRegistry
"""
logger.info('DataRegistryClient.update_data_registries_id() invoked.')
if id is None:
raise exceptions.MindsphereClientError('`id` is not passed when calling `update_data_registries_id`')
if request_object is None:
raise exceptions.MindsphereClientError('`request_object` is not passed when calling `update_data_registries_id`')
end_point_url = '/dataRegistries/{id}'
end_point_url = end_point_url.format(id=id)
token = token_service.fetch_token(self.rest_client_config, self.mindsphere_credentials)
api_url = mindsphere_core.build_url(self.__base_path__, end_point_url, self.rest_client_config)
headers = {'Accept': 'application/json', 'Content-Type': 'application/json', 'Authorization': 'Bearer ' + str(token)}
query_params = {}
form_params, local_var_files, body_params = {}, {}, request_object
logger.info('DataRegistryClient.update_data_registries_id() --> Proceeding for API Invoker.')
return mindsphere_core.invoke_service(self.rest_client_config, api_url, headers, 'PATCH', query_params, form_params, body_params, local_var_files, 'DataRegistry', self.__model_package__)
[docs] def create_data_registries(self, request_object):
"""Add a new Data Registry entry
Initiate creation of Data Registry for the current tenant. The Data Registry information is used during data ingest for the tenant. Only one Data Registry can be created for a request. The dataTag, sourceName and fileUploadStrategy is required to create Date Registry otherwise creation is rejected. DataUpload will allow only files to be uploaded matching this Data Registry. This returns unique **registryId** for each request that can be used to retrieve the created registry. The tenant cannot have more than **500** data registries in the system. The schemaFrozen flag must be set to false during creation of a registry. It can be set to true after creation of the initial schema for the registry. Data partitioning on the registry can be optionally enabled by specifying a single partitionKey at the time of registry creation. The property value can be set to \"sdi-default-partition-key\" to enable default partitioning based on the sdi_hour attribute. The property can also be set to a custom attribute present in the data to enable custom partitioning based on that attribute.
:param DataRegistriesPostRequest request_object: It contains the below parameters --> |br| ( createDataRegistryRequest* - createDataRegistryRequest )
:return: DataRegistry
"""
logger.info('DataRegistryClient.create_data_registries() invoked.')
if request_object is None:
raise exceptions.MindsphereClientError('`createDataRegistryRequest` is not passed when calling `create_data_registries`')
end_point_url = '/dataRegistries'
end_point_url = end_point_url.format()
token = token_service.fetch_token(self.rest_client_config, self.mindsphere_credentials)
api_url = mindsphere_core.build_url(self.__base_path__, end_point_url, self.rest_client_config)
headers = {'Accept': 'application/json', 'Content-Type': 'application/json', 'Authorization': 'Bearer ' + str(token)}
query_params = {}
form_params, local_var_files, body_params = {}, {}, request_object
logger.info('DataRegistryClient.create_data_registries() --> Proceeding for API Invoker.')
return mindsphere_core.invoke_service(self.rest_client_config, api_url, headers, 'POST', query_params, form_params, body_params, local_var_files, 'DataRegistry', self.__model_package__)
[docs] def delete_data_registries_id(self, id):
"""Deletes Data Registry for a given registry id
Deletes registry for a registry id
"""
logger.info('DataRegistryClient.delete_data_registries_id() invoked.')
if id is None:
raise exceptions.MindsphereClientError('`id` is not passed when calling `delete_data_registries_id`')
end_point_url = '/dataRegistries/{id}'
end_point_url = end_point_url.format(id=id)
token = token_service.fetch_token(self.rest_client_config, self.mindsphere_credentials)
api_url = mindsphere_core.build_url(self.__base_path__, end_point_url, self.rest_client_config)
headers = {'Accept': 'application/json', 'Authorization': 'Bearer ' + str(token)}
query_params = {}
form_params, local_var_files, body_params = {}, {}, None
logger.info('DataRegistryClient.delete_data_registries_id() --> Proceeding for API Invoker.')
return mindsphere_core.invoke_service(self.rest_client_config, api_url, headers, 'DELETE', query_params, form_params, body_params, local_var_files, 'str', self.__model_package__)