This post is about how to successfully authorize BI tools with Windchill.
A desktop version of QlickSense and MS Power BI are considered, but the technique for Power BI is applicable to Excel + PowerQuery.
With the current power of Query Builder coming to WT 11.0 upwards and growing demand for BI tools within the enterprise my company followed several attempts to connect WT Data to BI tools to easily slice and dice and communicate product information as well as design process information.
QlickSense was easy to configure, once XML report is exposed you can use REST connector to download it to the session – use data for the dashboard, pivot, etc.
Con: Your credentials travel with your BI App file.
MS Power BI works well with XML coming from QB, there is wizard assisting you, as long you authenticate Basic mode to Windchill, you choose Basic, give credentials and you are all set.
Pro: your credentials do not travel with your BI app file.
There are cases, however when your report is so complex or just out of QB limits, you hire a developer to do that using wt API, we had a similar case, resulting page is an html table. Assume you don’t want to rewrite report to version giving XML.
Current BI tools handle HTML pages with ease, however, all of the sudden MS Power BI stopped authenticate.
Simply saying credentials are invalid.
I have tried several techniques and sharing now one that worked.
Looks like you need to switch credentials to Anonymous
Then authenticate with a special header parameter:
Authorization: Basic user_password_code
Make sure code is preceded by “Basic”+”blank space”
Screens below where to set that, how I get the code, it is my Windchill and Chrome Dev tools (F12).
Now your Power BI authorize successfully.
Con: Your authentication is now within the app/Excel file PowerQuery formula.
Hope you can find use of this authentication method within Power BI or other BI tools to analyse and share Windchill Data.
@K_Trekiel Did you consider using the WRS framework as well. While not all the domains are present today (we add more at each release) we were able to connect with Qlik very easily using the framework. I will add some more information to the thread after the PTC/USER Forum in a few weeks when I have chance to catch up.