Every PDF viewer parses the PDF data differently so the best choice is to choose one and stick with it.
This is not a viewer problem by itself, it is a data capture problem.
Actually from an AA BOT perspective there are 2 options: get pdf fields or get text. When i run the get text function on a fillable form NONE of the data fields are captured. When i run the get pdf fields function it is returning the wrong data.
What we really need to understand is what does the PDF objects look like under the covers and how is it that there is more than 1 set of field values.
SInce we are trying to use this form to authorize payments it needs to be 100% accurate.
As a result it could be that we need to find a better program than what AA is offering or better yet understand why its getting the wrong data.
If the AA PDF package is not correct than i will need to find a different solution.
We created an adobe fillable form the customers use to make payments on invoices. in practice they typically reuse the last PDF form sent. Make a copy Edit the form with new payment value and invoice Remittance advice.
the problem is that when the bot runs and collects the data it gets old values.
When the pdf is opened in the browser we typically see the latest values entered. but, when the bot runs or we use adobe to view the pdf the fields show different values.
When i use the PDF Form template to configure the bot I also see older values in the fields as sample data. When i use the PDF get text none of the data fields are returned in the text image. I suppose i expect this result since i assume the fillable data is stored elsewhere in the file separate from the text data.
To address the issue you're encountering with the Adobe fillable form, it seems like the form data is being saved in a way that might not be fully compatible with your bot's extraction methods. This is why you see old values when running the bot or viewing the PDF in Adobe, but the latest values in a browser. Here are a few recommendations:
-
Flatten the PDF: After editing, you could flatten the PDF form. This converts the fillable fields into static text, ensuring that the bot captures the correct data. However, this will make the form non-editable, so you should keep an original version for future edits.
-
Check Form Data Saving Options: When users save the edited PDF, ensure they are saving the form data properly. If the data is not being embedded correctly, it can lead to discrepancies when viewing or extracting information.
-
Use Dedicated PDF Parsing Tools: Consider using specialized tools like paykassma.com for better handling of PDF data extraction, particularly when dealing with forms. These tools can be more reliable in parsing the embedded data that is often missed by simpler methods.
-
Review Bot Configuration: Ensure that your bot is configured to read the embedded form data correctly, rather than just the visible text. Some bots require additional modules or configuration to access this data.
By refining your approach with these methods, you should be able to reduce the discrepancies you're seeing between different viewers and the bot's output.
