Pull data from Tableau Server into Pandas Dataframe

0
votes

My goal is to join three datasources that are only available to me through Tableau Server (no direct database access). The data is too large to efficiently use Tableau's Data Blending.

One way forward is to pull the data from the three Tableau Server Datasources into a Pandas dataframe, do the necessary manipulations, and save down an Excel File to use as a datasource for a visualization in Tableau.

I have found lots of information on the TabPy module that allows one to convert a Pandas dataframe to a Tableau Data Extract but have not found much re: how to pull data from Tableau server in an automated fashion.

I have also read about tabcmd as a way of automating tasks, but do not have the necessary admin permissions.

Let me know if you need further information.

2
pythonpandasautomationtableau-api
If the data source is too large to manipulate in tableau data blending, there may be a way to get it via the Document API . github.com/tableau/document-api-python - Bernardo
Thanks, Bernardo. I am going to explore this option as well. It looks like this API allows you to extract directly from the datasource as well, whereas tabcmd only lets you work with published views. - p_sutherland

2 Answers

2
votes

Tabcmd does not require admin privileges. Anyone with permissions to Server can use it, but it will respect the privileges you do have. You can install tabcmd on computers other than your server without needing extra license keys.

That being said, it's very simple to automate data downloading. Take the URL to your workbook and add ".csv" to the end of it. The .csv goes at the end of the URL, not any query parameters you have.

For example: http://[Tableau Server Location]/views/[Workbook Name]/[View Name].csv

Using URL parameters, you can customize the data filters and how it looks. Just make sure you put the .csv before the ? for any query parameters.

More info for this plus a few others hacks at http://www.vizwiz.com/2014/03/the-greatest-tableau-tip-ever-exporting.html.

1
votes

You can use pantab to both read and write from Hyper extracts https://pantab.readthedocs.io/en/latest/