Forum Discussion
PDF Viewer visual not working with base64
The Pdf documents must be stored in the data model in base64 format, meaning that they must converted to this format before being imported to Power BI. This can be accomplished in many ways, for example in python with a few lines of code:
import base64 with open(<[path to pdf]>, "rb") as pdf_file: encoded_string = base64.b64encode(pdf_file.read())
In Power BI the Pdf Viewer takes a single field, the column storing the base64 data, as argument:
When using the Pdf Viewer in Power BI the pdf documents will only be displayed when there is one document selected in current filter context. Otherwise a warning is displayed:
With one document selected the Pdf Viewer displays pdf documents with controls that enables the user to switch pages and zoom in and out:
On the format pane there are also possibilities to show/hide header controls and overflow scrolling:
Thanks Anonymous for your answer,
what you say is already clear to me. I know the visual requires column storing base64 as argument and I need to select one document at a time, but my issue comes before. As you can see in the Advanced Editor query I pasted in red in my post, documents are already converted in base64 when I come in Power BI, as this step is accomplished by my web service. In the sample report you can find at the link I posted, base64 strings are already parted in 2 different columns and then they are concatenated to get the argument column, but I get my string which is larger than 30.000 characters and it's the max you can store in tabular models, as PDF Viewer visual documentation says.
I use a web service because I need to receive and convert documents from another web service. Even if I use your Python query, where should I write that?
Anyway, how can I fix that?
Nick
There is no need to convert the PDF to a Base64 string: https://www.youtube.com/watch?v=TEou0LuPjkA
