Converting documents to Markdown

[ewilson]

If you're working with LLMs and especially if you're implementing any sort of RAG pipeline, there's a new VBO on the DX that might be of interest. It's the Utility - MarkItDown VBO. This VBO leverages Microsoft's MarkItDown Python package to convert different file types, and web pages, to Markdown locally.

While LLMs can accept various file formats—such as PDF, DOCX, and HTML—they ultimately process text best when it is structured with clear hierarchy, which Markdown provides, often replacing the need for complex, often messy, raw HTML or JSON formatting. In some cases, this can have a positive impact on your overall token utilization because you're converting the files locally before sending them to the LLM.*

*NOTE: This isn't always a requirement of the LLM.

The Code stages in the VBO are implemented in Python which means this VBO is limited to Blue Prism v7.4 or later. 

Cheers,
Eric

[matiasskwarko84]

Hello,

I downloaded the MarkItDown utility from the Blue Prism Digital Exchange and I'm trying to test it.

However, as soon as I run the Convert File action, I get the following error:

Failed to set run stage - Internal : Failed to initialize Python

So far I have installed:

• Python

• MarkItDown

• Python.NET

I am using Blue Prism 7.4.2 HF1.

I'm not sure where the issue could be. Has anyone experienced this error before or knows if there are any additional configuration steps required for Python integration with this utility?

Any help would be greatly appreciated.

Thank you.

[ewilson]

@matiasskwarko84 well, this is embarrassing. 😞 I forgot to mention in the user guide that you need to open the VBO and set the path to the Python DLL on your system. I'll update the user guide today or tomorrow.

Cheers,
Eric

[ewilson]

Actually, I did add that step in the Configuration section of the User Guide. 🙂 It's on page 5.

Cheers,
Eric

[kiranb25]

Hi @ewilson ,

I have downloaded the MarkItDown utility and set up the DLL path as well able to initialize but i am getting error (Object reference not set to an instance of an object) while running the code stage. 

I am using blue prism v7.4.1
python 3.14

Could you please suggest if I am missing anything? Do i need to setup anything in Application modeller ?
Appreciate your help, Thanks 

Kiran.

[kiranb25]

• To provide more info, I have imported all the libraires as required and both python and libraries are installed in user specific (updated the same path in dll set up).
• Able to run initialize page but while executing the convert code stage issue occurs, when I compile the code (failed to initialize the python).
• Standalone code is working fine in python idle.

[ewilson]

@kiranb25 just for clarity, you're not using Python virtual environments, correct? Is Python also added to your PATH environment variable? In other words, if you just open a command prompt and type py -V, do you get a response showing the installed Python version? 

Cheers,
Eric

[kiranb25]

@ewilson  I have installed python in rdp under user specific folder, Updates the same path in object dll setup (C:\users\username\AppData.....\Python314.dll). Yes, I could see the python version if I check installed version in command prompt.
Let me know if anything needed.

Thanks,
Kiran

[ewilson]

@kiranb25 how did you install the MarkItDown Python module? Did you use pip (ex. pip install "markitdown[all]")?

If you run pip show markitdown, what output do you receive?

Cheers,
Eric

[kiranb25]

@ewilson Yes installed using pip install markitdown[all]. refer below screen:

Thanks,
Kiran