Need help to understand the copilot integration with the UiPath for the below case, if you have any other substitute pls advice
Case: want to retrieve SR number from the invoice pdf and the invoice are not of fixed format, the only constant thing is that SR number is SR followed bty 6 digit.
Since the invoice PDFs are not in a fixed format but the SR number always follows the pattern “SR” + 6 digits, you don’t need to rely only on Copilot integration. You can directly use:
Regex Based Extractor in Document Understanding
Define a regex like: SR\d{6}
This will match values like SR123456.
If you don’t want full DU you can also just: Use Read PDF Text (or Read PDF with OCR)
Apply System.Text.RegularExpressions.Regex.Match(text, “SR\d{6}”).Value to retrieve the SR number.
This way it will work across different invoice formats as long as the SR number pattern is consistent.
I already using the pdf activity as well as ocr activity for it but that can retrieve SR only for 40% of the cases for rest of the cases bot is not able to find the SR.
This is because there is no fixed pattern of the pdf content, in some of the pdf the SR is not even in the single line- i mean at the end of one line and the starting of the other line(i handled that case as well), some pdf have inverted content, some have some other different pattern.
So, i thought if we can integrate copilot to optimize the bot performance.