Since the announcement about XMLA endpoints, I've been trying to figure out how to connect to a URL of the form powerbi://api.powerbi.com/v1.0/myorg/[workspace name]
as an SSAS OLAP cube via Python, but I haven't gotten anything to work.
I have a workspace in a premium capacity and I am able to connect to it using DAX Studio as well as SSMS as explained here, but I haven't figured out how to do it with Python. I've tried installing olap.xmla, but I get the following error when I try to use the Power BI URL as the location using either the powerbi
or https
as the prefix.
import olap.xmla.xmla as xmla
p = xmla.XMLAProvider()
c = p.connect(location="powerbi://api.powerbi.com/v1.0/myorg/[My Workspace]")
[...]
TransportError: Server returned HTTP status 404 (no content available)
I'm sure there are authentication issues involved, but I'm a bit out of my depth here. Do I need to set up an "app" in ActiveDirectory and use the API somehow? How is authentication handled for this kind of connection?
If anyone knows of any blog posts or other resources that demonstrate how to connect to a Power BI XMLA endpoint specifically using Python, that would be amazing. My searching has failed me, but surely I can't be the only one who is trying to do this.
You can install the Power BI Client for Jupyter from PyPI and find the open-sourced Python package and associate TypeScript widget on GitHub. For a quick peek into how to use the package in your application. Before you can post on Kaggle, you'll need to create an account or log in.
XMLA endpoints provide access to the Analysis Services engine in the Power BI service, allowing customers to leverage enterprise BI tools directly on Power BI datasets.
I'm not familiar with olap.xmla or using Python to connect to olap cubes, but I think the problem is with the driver (or connector ?) provided in olap.xmla.
In the announcement about XMLA endpoints page, it says that the connection only works with SSMS 18.0 RC1 or later, which is quite new. Same thing with DAX studio, the version where xmla connection is supported (Version 2.8.2, Feb 3 2019), is quite fresh.
The latest version of olap.xmla seems to be from august 2013, so it's possible that there's some Microsoft magic behind the PowerBI XMLA connection and that's why it doesn't work with older connectors.
After @Gigga pointed the connector issue, I went looking for other Python modules that worked with MSOLAP to connect and found one that I got working!
The module is adodbapi (note the pywin32
prerequisite).
Connecting is as simple as this:
import adodbapi
# Connection string
conn = adodbapi.connect("Provider=MSOLAP.8; \
Data Source='powerbi://api.powerbi.com/v1.0/myorg/My Workspace Name'; \
Initial Catalog='My Data Model'")
# Example query
print('The tables in your database are:')
for name in conn.get_table_names():
print(name)
It authenticated using my Windows credentials by popping up a window like this:
They now have a REST endpoint via which you can execute DAX queries. This could be easier than trying to invoke the XMLA endpoint directly.
https://learn.microsoft.com/en-us/rest/api/power-bi/datasets/execute-queries
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With