Extract data from different PDF files, and fill in excel

[MateusCorrea]

I have created a folder on the share point, so when I paste some PDF files into that folder, PA reads those files and needs to identify which AI builder model to use to read the PDF. After that, it is necessary to fill in an Excel with this data.

But I don't know what I'm doing wrong...

I have created an array variable

And Im trying to identify which IA Builder model to use by the file name

[ManishSolanki]

Hi @MateusCorrea 

I would suggest you go for automate flow which will trigger as soon as you drop a file in the SharePoint folder. Here is the sample flow that will help you in designing the full flow:

1. We will add a trigger action "When a file is created" which will trigger the flow when a new file is created or dropped in it:

2. Next, we will check if the file extension ends with 'pdf' and name starts with 'MS'. You may change the condition as per your need. As I faced issue in understanding the language in the screenshot shared by you so I have chosen different business rule or conditions.

3. If the condition is true, then we will fetch the content of the pdf file to pass it to AI builder action for data extraction.

4. After getting content, you can pass the same to AI builder for extracting required data and post that you can save the data in excel sheet:

Pls select the appropriate AI model as per you need.

If this helps & solves your problem, please remember to give a 👍 and accept my solution as it will help others in the future.

Thanks

[Anonymous]

Hello @ManishSolanki , i work with @MateusCorrea .

First of all, ty for your help, it was very usefull and clarifier. but...

We tried to reproduce your flow, but, we had a problem, we need to upload 3 different types of file at the same time in the same folder, example:
A123HB.PDF
A123ND.PDF
A123NF.PDF
and we need that the flow identify whats the correct model to read and fill excel. In excel, we have 3 tables, one for the HB files, each other for the ND files and another one for the NF files, our problem is: what's the correct method do use the "for each" function. See the flows description below:

1. Trigger: when a file is created(properties only)
2. List folder (trying to list all the files that we dropped in folder)
3. Created an array and tried to add the "file name with extension'' there.

4. Here starts your instructions, but, using for each before:

THANK YOU! 😃

[takolota]

I’m not sure how much effort you already put into your AI Builder models, but if you can’t get that to work well, there is this template that extracts PDF data from various document structures using GPT. It may be easier than maintaining several models yourself.

https://powerusers.microsoft.com/t5/Power-Automate-Cookbook/Extract-Data-From-PDFs-and-Images-With-G...