Power BI Token Server

This Flask application returns an embed token for a Power BI report so it can be displayed on a website. It exposes a single endpoint /getEmbedToken which expects reportId, groupId and datasetId query parameters.

Requirements

• Python 3.10 or newer
• The packages listed in requirements.txt

It is recommended to create a virtual environment before installing dependencies. Make sure to specify the Python version when creating the environment:

# Windows
py -3.10 -m venv venv
# macOS/Linux
python3.10 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

On Windows PowerShell, running \.venv\Scripts\Activate.ps1 may be blocked by the execution policy. Temporarily bypass it with:

Set-ExecutionPolicy -ExecutionPolicy Bypass -Scope Process
.\venv\Scripts\Activate.ps1

Use venv\Scripts\activate.bat for the Command Prompt.

Configuration

Set the required credentials using environment variables before starting the app:

export TENANT_ID="<your AAD tenant ID>"
export CLIENT_ID="<your AAD application ID>"
export CLIENT_SECRET="<your AAD app client secret>"

The allowed CORS domains are defined in app.py and can be adjusted if needed.

Running the server

After installing the dependencies and setting the environment variables, start the Flask app:

python app.py

The server will listen on port 5000 by default. The embed token endpoint will be available at http://localhost:5000/getEmbedToken.

Embedding in WordPress

Include embedScript.js on your WordPress page and provide the report information using attributes (or by defining window.PowerBIEmbedConfig before the script loads). The optional embed.css file styles the embedded report. If no <link id="powerbi-embed-styles"> element is present, the script automatically creates one pointing to embed.css located next to the script:

<!-- Optional: include the stylesheet yourself (the script adds it automatically) -->
<link id="powerbi-embed-styles" rel="stylesheet" href="/path/to/embed.css">
<div id="reportContainer"
     data-report-id="<report-id>"
     data-group-id="<workspace-id>"
     data-dataset-id="<dataset-id>"
     data-server-url="http://localhost:5000"><!-- optional override -->
  Loading Power BI...
</div>
<script src="/path/to/embedScript.js"></script>

The embed2.css stylesheet defines a --header-height CSS custom property with a default value of 80px. You can override this variable in your own CSS if your theme's header height differs:

:root {
  --header-height: 80px;
}

By default the script contacts https://powerbi-token-server.onrender.com for tokens. To use a different endpoint, provide a data-server-url attribute or define window.PowerBIEmbedConfig.serverUrl before the script loads:

<!-- Option 1: set the attribute on the container -->
<div id="reportContainer" data-server-url="https://my-token-server.example.com"></div>

<!-- Option 2: configure the global before loading the script -->
<script>
  window.PowerBIEmbedConfig = {
    serverUrl: "https://my-token-server.example.com"
  };
</script>
<script src="/path/to/embedScript.js"></script>

The script fetches an embed token from your Flask server and renders the report using the Power BI client library.

WordPress-ready snippet

The example snippet above can be saved as an HTML file for a WordPress Custom HTML block. It loads the embed.css stylesheet and the embedScript.js script and exposes the same data-* attributes for configuration.

Testing locally

Copy the embed-html file anywhere on your system to try the embed code outside of WordPress. Set the required data-report-id, data-group-id and data-dataset-id attributes to your own IDs. To use a token server other than https://powerbi-token-server.onrender.com, provide its address with a data-server-url attribute or by defining window.PowerBIEmbedConfig.serverUrl before the script loads. Save the snippet above as an HTML file and paste it into a Custom HTML block, editing the IDs as needed.

Running tests

Install the dependencies before running the test suite so pytest can import the app module:

pip install -r requirements.txt
pytest

The CI workflow under .github/workflows/python-tests.yml installs the same packages automatically.

License

This project is licensed under the MIT License.