Any way to extract lineItems from PDF?

How To
,
1

I am having trouble creating a reliable scenario that extracts each line item from a Purchase Order Document. I am trying to use @PDF.co with the Parse Doc module and connecting it with a created template. The problem is that this scenario is not reliable because the PDF cannot be consistent and the instructions to the parse pdf template has to be.

Here a sample of first, middle, and last page of a pdf document. This pdf can have multiple pages between first and last, and depending on the number of lineitems, it will mess with the format of the first and last page.

First page

image
image1342×1736 178 KB

Middle page

image
image1352×1744 236 KB

Last page

image
image1348×1751 209 KB

How can I tell the PDF.co module to get all line items as one table?

Maybe PDF.co module is not the right direction to go, but don’t know any other tool that can do this

2

Did you try the AI Invoice Parser? That might solve the issue.
image

3

I cannot see that module as an option @Anonymax

image
image567×518 12.9 KB

4

I personally use DumplingAI’s “Extract data from PDF with AI” module. You can also use “Convert PDF to Text” or “Extract Data from Image(s)”.

Hope this helps! Let me know if there are any further questions or issues.

— @samliew

P.S.: Investing some effort into the Make Academy will save you lots of time and frustration using Make.

5

Weird, it should be listed in the PDF section of modules, 6th option down

6

Thanks! DumplingAI worked perfectly. Although, I think it will get very expensive after 4 or 5 pdfs cause even for this first test using the free 100 credits of my account, it only extracted 57 complete line items. and this one sample has 533 :skull:. Any idea of another tool that could be cheaper?

Again… @dumplingai worked perfectly! it did extract the lineitems very accurately. Thanks for suggesting that tool

7

Hello fergotz,

You can find the AI Invoice Parser feature under the Parse a Document section. Please take a look at the sample screenshot below for reference.

image
image712×527 43.1 KB

If you’re still unable to see it, you can also use the PDF.co Make an API Call feature to directly access the AI Invoice Parser API endpoint. I’ve included a screenshot below for your guidance.

If you have any questions or need further assistance, please don’t hesitate to let us know.