Search for a command to run...
Using FabricRestClient in sempy
Principal Program Manager, Microsoft Fabric CAT helping users and organizations build scalable, insightful, secure solutions. Blogs, opinions are my own and do not represent my employer.
Nice post. It really helped. I have a query, I tried using your code for lakehouse creation and replaced it with datapipeline creation but when I am adding some activities to the payload it is still creation an empty datapipeline inside workspace. Can you please share you thoughts and guide with the code.
Hi Sandeep Pawar I notice there is an experimental PowerBIRestClient to call legacy Power BI API. <>
cannot add link to documentation as I'm a new user
Awesome post.
Can I ask how do the POST requests work with FabricRestClient? I'm trying to invoke Job scheduler APIs to invoke an item job instance as covered in below REST API doc.
"Job Scheduler - Run On Demand Item Job" (Can't post links as I'm restricted currently)
I tried this: response = client.post(f"v1/workspaces/{workspaceId}/items/{itemId}/jobs/instances?jobType={jobType}")
But I'm getting this exception FabricHTTPException: 400 Bad Request for url: . . . "moreDetails":[{"errorCode":"InvalidParameter","message":"'runOnDemandItemJobRequest' is a required parameter"}],"message":"The request has an invalid input"}
I understand that as per the doc, there's a request body that must go along too but can't figure out where to put that in the request method. Would appreciate any help on this. Thanks!
Swapnil Saxena please check the updated blog. I have added an example.
Thanks a lot! I could replicate your example to see the POST work.
However, for some reason, I'm unable to get the On demand job scheduler to work. It complains that the job type is invalid. I used the jobType=DefaultJob for a Data pipeline item.
https://learn.microsoft.com/en-us/rest/api/fabric/core/job-scheduler/run-on-demand-item-job?tabs=HTTP
Last week Microsoft released an open-source text embedding model called Harrier in three sizes- 270M, 0.6B and 27B. I have been testing it in my RAG pipeline and so far it has crushed all my metrics.
For Power BI Copilot and Data agents with semantic models, you must use Prep Data for AI configuration to ground the responses in the context added in Prep for AI. In this blog, I will show you how yo
At FabCon Atlanta last week, the updated notebook Copilot for data engineering and data science was announced. It brings agentic capabilities to the Copilot and is much more intelligent and Fabric-awa
Using Fabric data agent Python SDK
Use workspace monitoring to monitor MCP server use
Accessing Fabric REST endpoints in Fabric notebooks was already easy but it became easier and straightforward with semantic-link version 0.4.0. You can use the FabricRestClient class from sempy to set up a REST client and call the APIs. Authentication is automatically managed for you.
#sempy version 0.4.0 or higher
!pip install semantic-link --q
import pandas as pd
import sempy.fabric as fabric
#Instantiate the client
client = fabric.FabricRestClient()
Refer to the official documentation for API details. To call an API, you only need to pass the below portion of the URL.
#example usage
url = "v1/workspaces/<workspaceId>/items"
result = client.get(url)
This approach is not substantially different from using requests to make API calls, but there are a few advantages:
You don't have to generate and pass the authentication token. sempy uses the environment details to authenticate the user automatically.
This approach is similar to using requests.Session() to maintain a persistent connection, keeping the session open longer. This is particularly beneficial if you plan to call multiple REST endpoints within the same session.
sempy has a few nice utilities and functions for exception handling and getting workspace/lakehouse ids.
In this simple example, I will get a list of Fabric items from a workspace. I will show how some of the utilities can help make this even easier and robust.
#instantiate the client
client = fabric.FabricRestClient()
#get a list of items from a workspace
workspaceId = "<workspaceid>"
client.get(f"/v1/workspaces/{workspaceId}/items")
In this example, I manually provided the workspace ID. However, we can obtain the notebook's workspace ID automatically without needing to define it.
#instantiate the client
import pandas as pd
client = fabric.FabricRestClient()
#get the default workspace id of the notebook
workspaceId = fabric.get_workspace_id()
#get a list of items from a workspace
response = client.get(f"/v1/workspaces/{workspaceId}/items")
df_items = pd.json_normalize(response.json()['value'])
df_items
For production use cases you will need to include exception handling. sempy provides a few nice utilities, see an example below:
import sempy.fabric as fabric
from sempy.fabric.exceptions import FabricHTTPException, WorkspaceNotFoundException
import pandas as pd
client = fabric.FabricRestClient()
def get_workspace_items(workspace=None):
"""
Sandeep Pawar | fabric.guru
Get a list of workspace items using FabricRestClient
"""
df_items = None
if workspace is None:
workspaceId = fabric.get_workspace_id()
else:
try:
workspaceId = fabric.resolve_workspace_id(workspace)
response = client.get(f"/v1/workspaces/{workspaceId}/items")
# Check the status code of the response for this endpoint
# Use 200 if operation is completed, 201 if item is created
if response.status_code != 200:
raise FabricHTTPException(response)
df_items = pd.json_normalize(response.json()['value'])
except WorkspaceNotFoundException as e:
print("Caught a WorkspaceNotFoundException:", e)
except FabricHTTPException as e:
print("Caught a FabricHTTPException. Check the API endpoint, authentication.")
return df_items
get_workspace_items("<dataflow name or id>")
In the above example:
workspace is optional. If no workspace name/id is passed, fabric.get_workspace_id() will use the workspace id of the default workspace. Similarly, you can use fabric.get_lakehouse_id() to get the id of the default lakehouse attached to the notebook.
All APIs require using workspace id. With fabric.resolve_workspace_id(workspace) you can get the id of a workspace if the user enters workspace name
WorkspaceNotFoundException, FabricHTTPException help catch workspace and REST endpoint exceptions.
Update: 1/21/2024
A redear asked how to make a POST request using the FABRICRestClient. Below is te example where I create a lakehouse using a POST request.
import sempy.fabric as fabric
from sempy.fabric.exceptions import FabricHTTPException, WorkspaceNotFoundException
client = fabric.FabricRestClient()
def create_lakehouse(payload:dict, workspace=None):
"""
Sandeep Pawar | fabric.guru
Create a lakehouse using POST request
"""
if workspace is None:
workspaceId = fabric.get_workspace_id()
else:
workspaceId = fabric.resolve_workspace_id(workspace)
try:
response = client.post(f"/v1/workspaces/{workspaceId}/items",json= payload)
if response.status_code != 201:
raise FabricHTTPException(response)
except WorkspaceNotFoundException as e:
print("Caught a WorkspaceNotFoundException:", e)
except FabricHTTPException as e:
print("Caught a FabricHTTPException. Check the API endpoint, authentication.")
return
#payload/body for the POST request as a dict
#example; create a lakehouse named "mylh" in the default workspace
payload = {
"displayName": "mylh",
"type": "Lakehouse"
}
#workspace name or id is optional. If None, default workspace is used.
create_lakehouse(payload=payload, workspace=None)
Thanks to Markus Cozowicz for answering my questions and for his input.
Update: 3/23/2025
Follow-up blog on using SPN authentication : Using Service Principal Authentication With FabricRestClient