How do you sort documents in share point based on a specific text within the document?

[jevans]

Hello All,

I will start by saying that I am extremely new to Power Automate, however, I am excited to begin learning. I have been struggling  to find a way to sort pdf's from a "master share point folder" to their respective "sub-share point folders" based upon a specific text within the actual pdf document.

For example, I have hundreds of documents that are not formatted the same that I need to sort by state . Within the pdf the state is highlighted but is not in the same place. Therefore, I need the entire document to be review by AI to advise as to which state the document is from. Once the state is identified, the document then needs to be sent to its respective folder. That being said, and given my novice skill, would anyone be able to help me get started by showing me the initial flow for something like this? 

[Antrod]

Hello @jevans ,

You can indeed perform this scenario using Power Automate flows + AI Builder. You can for example use AI Builder Text recognition within your flow: Use the text recognition prebuilt model in Power Automate - AI Builder | Microsoft Docs

Then you can use the expressions indexOf or lastIndexOf to find the string you're looking for. Here's a great video that explains it: Power Automate Functions - LastIndexOf - YouTube

Hope that helps!

[jevans]

Hello @Antrod,

Thank you for taking the time to respond to me. That being said, through your help, I believe I was able to get 3/4 of the way to creating the workflow I need. However, there is one spot that I am still struggling with (please see the below screenshots). I have redacted the SP links in the workflow. If you could please advise as to why I am receiving an "ExpressionEvaluationFailed" error I would appreciate it. 

I look forward to hearing from you soon. 

 

 

 

[Antrod]

Hi @jevans ,

You're almost there 🙂

The output of the Text recognition model is a set of results, each result contains lines and a line has a text.

So, you should do the Apply to each on results, then on lines, and gather all the text strings within a unique string or an array to do the condition.

Example, here I check if the word "INVOICE" appears in in a string representation of my document (don't add the double quotes in the flow):

Don't forget to reset the value of DocumentAsString for each document.

Let us know if you have other questions.

[jevans]

@Antrod 

Thank you for your help. I believe the beginning portion of the flow is working great! The last error in my flow that is showing up relates to the sorting part of the flow. See the below screenshots. I keep receiving error messages that my "move to" flow is incorrect. I have checked the source code for these file folders for where the document resides and where I want it moved to multiple times, and I can't seem to understand why I keep getting an error. If you could please advise, I would be extremely grateful. 

 

 

[Antrod · Accepted solution]

In the "File to Move" field you are passing the file content whereas you should pass the file identifier. Try using that field instead: 

Let us know if that helps!