libica.openapi.v2.BundleDataLinkingBatchApi¶
All URIs are relative to /ica/rest
Method | HTTP request | Description |
---|---|---|
create_bundle_data_linking_batch | POST /api/bundles/{bundleId}/dataLinkingBatch | Create a bundle data linking batch. |
get_bundle_data_linking_batch | GET /api/bundles/{bundleId}/dataLinkingBatch/{batchId} | Retrieve a bundle data linking batch. |
get_bundle_data_linking_batch_item | GET /api/bundles/{bundleId}/dataLinkingBatch/{batchId}/items/{itemId} | Retrieve a bundle data linking batch item. |
get_bundle_data_linking_batch_items | GET /api/bundles/{bundleId}/dataLinkingBatch/{batchId}/items | Retrieve a list of bundle data linking batch items. |
create_bundle_data_linking_batch¶
BundleDataLinkingBatch create_bundle_data_linking_batch(bundle_id, create_bundle_data_linking_batch)
Create a bundle data linking batch.
Example¶
- Api Key Authentication (ApiKeyAuth):
- Bearer (JWT) Authentication (JwtAuth):
import time
import libica.openapi.v2
from libica.openapi.v2.api import bundle_data_linking_batch_api
from libica.openapi.v2.model.create_bundle_data_linking_batch import CreateBundleDataLinkingBatch
from libica.openapi.v2.model.bundle_data_linking_batch import BundleDataLinkingBatch
from libica.openapi.v2.model.problem import Problem
from pprint import pprint
# Defining the host is optional and defaults to /ica/rest
# See configuration.py for a list of all supported configuration parameters.
configuration = libica.openapi.v2.Configuration(
host = "/ica/rest"
)
# The client must configure the authentication and authorization parameters
# in accordance with the API server security policy.
# Examples for each auth method are provided below, use the example that
# satisfies your auth use case.
# Configure API key authorization: ApiKeyAuth
configuration.api_key['ApiKeyAuth'] = 'YOUR_API_KEY'
# Uncomment below to setup prefix (e.g. Bearer) for API key, if needed
# configuration.api_key_prefix['ApiKeyAuth'] = 'Bearer'
# Configure Bearer authorization (JWT): JwtAuth
configuration = libica.openapi.v2.Configuration(
access_token = 'YOUR_BEARER_TOKEN'
)
# Enter a context with an instance of the API client
with libica.openapi.v2.ApiClient(configuration) as api_client:
# Create an instance of the API class
api_instance = bundle_data_linking_batch_api.BundleDataLinkingBatchApi(api_client)
bundle_id = "bundleId_example" # str |
create_bundle_data_linking_batch = CreateBundleDataLinkingBatch(
items=[
CreateBundleDataLinkingBatchItem(
data_id="data_id_example",
),
],
) # CreateBundleDataLinkingBatch |
# example passing only required values which don't have defaults set
try:
# Create a bundle data linking batch.
api_response = api_instance.create_bundle_data_linking_batch(bundle_id, create_bundle_data_linking_batch)
pprint(api_response)
except libica.openapi.v2.ApiException as e:
print("Exception when calling BundleDataLinkingBatchApi->create_bundle_data_linking_batch: %s\n" % e)
Parameters¶
Name | Type | Description | Notes |
---|---|---|---|
bundle_id | str | ||
create_bundle_data_linking_batch | CreateBundleDataLinkingBatch |
Return type¶
Authorization¶
HTTP request headers¶
- Content-Type: application/vnd.illumina.v3+json
- Accept: application/problem+json, application/vnd.illumina.v3+json
HTTP response details¶
Status code | Description | Response headers |
---|---|---|
201 | The bundle data linking batch is successfully created. | - |
0 | A problem occurred. | - |
[Back to top] [Back to API list] [Back to Model list] [Back to README]
get_bundle_data_linking_batch¶
BundleDataLinkingBatch get_bundle_data_linking_batch(bundle_id, batch_id)
Retrieve a bundle data linking batch.
Example¶
- Api Key Authentication (ApiKeyAuth):
- Bearer (JWT) Authentication (JwtAuth):
import time
import libica.openapi.v2
from libica.openapi.v2.api import bundle_data_linking_batch_api
from libica.openapi.v2.model.bundle_data_linking_batch import BundleDataLinkingBatch
from libica.openapi.v2.model.problem import Problem
from pprint import pprint
# Defining the host is optional and defaults to /ica/rest
# See configuration.py for a list of all supported configuration parameters.
configuration = libica.openapi.v2.Configuration(
host = "/ica/rest"
)
# The client must configure the authentication and authorization parameters
# in accordance with the API server security policy.
# Examples for each auth method are provided below, use the example that
# satisfies your auth use case.
# Configure API key authorization: ApiKeyAuth
configuration.api_key['ApiKeyAuth'] = 'YOUR_API_KEY'
# Uncomment below to setup prefix (e.g. Bearer) for API key, if needed
# configuration.api_key_prefix['ApiKeyAuth'] = 'Bearer'
# Configure Bearer authorization (JWT): JwtAuth
configuration = libica.openapi.v2.Configuration(
access_token = 'YOUR_BEARER_TOKEN'
)
# Enter a context with an instance of the API client
with libica.openapi.v2.ApiClient(configuration) as api_client:
# Create an instance of the API class
api_instance = bundle_data_linking_batch_api.BundleDataLinkingBatchApi(api_client)
bundle_id = "bundleId_example" # str |
batch_id = "batchId_example" # str |
# example passing only required values which don't have defaults set
try:
# Retrieve a bundle data linking batch.
api_response = api_instance.get_bundle_data_linking_batch(bundle_id, batch_id)
pprint(api_response)
except libica.openapi.v2.ApiException as e:
print("Exception when calling BundleDataLinkingBatchApi->get_bundle_data_linking_batch: %s\n" % e)
Parameters¶
Name | Type | Description | Notes |
---|---|---|---|
bundle_id | str | ||
batch_id | str |
Return type¶
Authorization¶
HTTP request headers¶
- Content-Type: Not defined
- Accept: application/problem+json, application/vnd.illumina.v3+json
HTTP response details¶
Status code | Description | Response headers |
---|---|---|
200 | The bundle data linking batch is successfully retrieved. | - |
0 | A problem occurred. | - |
[Back to top] [Back to API list] [Back to Model list] [Back to README]
get_bundle_data_linking_batch_item¶
BundleDataLinkingBatchItem get_bundle_data_linking_batch_item(bundle_id, batch_id, item_id)
Retrieve a bundle data linking batch item.
Example¶
- Api Key Authentication (ApiKeyAuth):
- Bearer (JWT) Authentication (JwtAuth):
import time
import libica.openapi.v2
from libica.openapi.v2.api import bundle_data_linking_batch_api
from libica.openapi.v2.model.bundle_data_linking_batch_item import BundleDataLinkingBatchItem
from libica.openapi.v2.model.problem import Problem
from pprint import pprint
# Defining the host is optional and defaults to /ica/rest
# See configuration.py for a list of all supported configuration parameters.
configuration = libica.openapi.v2.Configuration(
host = "/ica/rest"
)
# The client must configure the authentication and authorization parameters
# in accordance with the API server security policy.
# Examples for each auth method are provided below, use the example that
# satisfies your auth use case.
# Configure API key authorization: ApiKeyAuth
configuration.api_key['ApiKeyAuth'] = 'YOUR_API_KEY'
# Uncomment below to setup prefix (e.g. Bearer) for API key, if needed
# configuration.api_key_prefix['ApiKeyAuth'] = 'Bearer'
# Configure Bearer authorization (JWT): JwtAuth
configuration = libica.openapi.v2.Configuration(
access_token = 'YOUR_BEARER_TOKEN'
)
# Enter a context with an instance of the API client
with libica.openapi.v2.ApiClient(configuration) as api_client:
# Create an instance of the API class
api_instance = bundle_data_linking_batch_api.BundleDataLinkingBatchApi(api_client)
bundle_id = "bundleId_example" # str |
batch_id = "batchId_example" # str |
item_id = "itemId_example" # str |
# example passing only required values which don't have defaults set
try:
# Retrieve a bundle data linking batch item.
api_response = api_instance.get_bundle_data_linking_batch_item(bundle_id, batch_id, item_id)
pprint(api_response)
except libica.openapi.v2.ApiException as e:
print("Exception when calling BundleDataLinkingBatchApi->get_bundle_data_linking_batch_item: %s\n" % e)
Parameters¶
Name | Type | Description | Notes |
---|---|---|---|
bundle_id | str | ||
batch_id | str | ||
item_id | str |
Return type¶
Authorization¶
HTTP request headers¶
- Content-Type: Not defined
- Accept: application/problem+json, application/vnd.illumina.v3+json
HTTP response details¶
Status code | Description | Response headers |
---|---|---|
200 | The bundle data linking batch item is successfully retrieved. | - |
0 | A problem occurred. | - |
[Back to top] [Back to API list] [Back to Model list] [Back to README]
get_bundle_data_linking_batch_items¶
BundleDataLinkingBatchItemPagedList get_bundle_data_linking_batch_items(bundle_id, batch_id)
Retrieve a list of bundle data linking batch items.
Example¶
- Api Key Authentication (ApiKeyAuth):
- Bearer (JWT) Authentication (JwtAuth):
import time
import libica.openapi.v2
from libica.openapi.v2.api import bundle_data_linking_batch_api
from libica.openapi.v2.model.bundle_data_linking_batch_item_paged_list import BundleDataLinkingBatchItemPagedList
from libica.openapi.v2.model.problem import Problem
from pprint import pprint
# Defining the host is optional and defaults to /ica/rest
# See configuration.py for a list of all supported configuration parameters.
configuration = libica.openapi.v2.Configuration(
host = "/ica/rest"
)
# The client must configure the authentication and authorization parameters
# in accordance with the API server security policy.
# Examples for each auth method are provided below, use the example that
# satisfies your auth use case.
# Configure API key authorization: ApiKeyAuth
configuration.api_key['ApiKeyAuth'] = 'YOUR_API_KEY'
# Uncomment below to setup prefix (e.g. Bearer) for API key, if needed
# configuration.api_key_prefix['ApiKeyAuth'] = 'Bearer'
# Configure Bearer authorization (JWT): JwtAuth
configuration = libica.openapi.v2.Configuration(
access_token = 'YOUR_BEARER_TOKEN'
)
# Enter a context with an instance of the API client
with libica.openapi.v2.ApiClient(configuration) as api_client:
# Create an instance of the API class
api_instance = bundle_data_linking_batch_api.BundleDataLinkingBatchApi(api_client)
bundle_id = "bundleId_example" # str |
batch_id = "batchId_example" # str |
status = [
"INITIALISED",
] # [str] | The statuses to filter on. (optional)
page_offset = "pageOffset_example" # str | [only use with offset-based paging]<br>The amount of rows to skip in the result. Ideally this is a multiple of the size parameter. Offset-based pagination has a result limit of 200K rows and does not guarantee unique results across pages (optional)
page_token = "pageToken_example" # str | [only use with cursor-based paging]<br>The cursor to get subsequent results. The value to use is returned in the result when using cursor-based pagination. Cursor-based pagination guarantees complete and unique results across all pages. (optional)
page_size = "pageSize_example" # str | [can be used with both offset- and cursor-based paging]<br>The amount of rows to return. Use in combination with the offset (when using offset-based pagination) or cursor (when using cursor-based pagination) parameter to get subsequent results (optional)
# example passing only required values which don't have defaults set
try:
# Retrieve a list of bundle data linking batch items.
api_response = api_instance.get_bundle_data_linking_batch_items(bundle_id, batch_id)
pprint(api_response)
except libica.openapi.v2.ApiException as e:
print("Exception when calling BundleDataLinkingBatchApi->get_bundle_data_linking_batch_items: %s\n" % e)
# example passing only required values which don't have defaults set
# and optional values
try:
# Retrieve a list of bundle data linking batch items.
api_response = api_instance.get_bundle_data_linking_batch_items(bundle_id, batch_id, status=status, page_offset=page_offset, page_token=page_token, page_size=page_size)
pprint(api_response)
except libica.openapi.v2.ApiException as e:
print("Exception when calling BundleDataLinkingBatchApi->get_bundle_data_linking_batch_items: %s\n" % e)
Parameters¶
Name | Type | Description | Notes |
---|---|---|---|
bundle_id | str | ||
batch_id | str | ||
status | [str] | The statuses to filter on. | [optional] |
page_offset | str | [only use with offset-based paging]<br>The amount of rows to skip in the result. Ideally this is a multiple of the size parameter. Offset-based pagination has a result limit of 200K rows and does not guarantee unique results across pages | [optional] |
page_token | str | [only use with cursor-based paging]<br>The cursor to get subsequent results. The value to use is returned in the result when using cursor-based pagination. Cursor-based pagination guarantees complete and unique results across all pages. | [optional] |
page_size | str | [can be used with both offset- and cursor-based paging]<br>The amount of rows to return. Use in combination with the offset (when using offset-based pagination) or cursor (when using cursor-based pagination) parameter to get subsequent results | [optional] |
Return type¶
BundleDataLinkingBatchItemPagedList
Authorization¶
HTTP request headers¶
- Content-Type: Not defined
- Accept: application/problem+json, application/vnd.illumina.v3+json
HTTP response details¶
Status code | Description | Response headers |
---|---|---|
200 | The list of bundle data linking batch items is successfully retrieved. | - |
0 | A problem occurred. | - |
[Back to top] [Back to API list] [Back to Model list] [Back to README]