Validation station return different variable type

[QTH]

I’ve labeled a particular field variable type to be number in the AI center.
When using document understanding, because the field on the document contains the currency as well, but due to labeling it would correctly extract just the value as a double as intended. But when passing it through the validation station, it returns as a string instead resulting in the output to be USD31.18 instead of 31.18, while those that didn’t pass through the validation did return the expected value of just the numerical value.
Thanks in advance for any and all advise.

[pradeep931 (Pradeep Bandharam)]

Hi @QTH

Please review the threads below, as they contain issues similar to the one you’re facing.

Present Validation Station: The value must either contain a textual value or have components, not both

Hi @Nithya_Aashitha Welcome to UiPath community Usually this error occurs when the data extracted from the document contains both textual values and components. You can try to modify the data extraction rules to ensure that the data extracted from the document contains either textual values or components, but not both. Regards Gokul

Present Validation Station: The value must either contain a textual value or have components, not both

@Nithya_Aashitha Can you please check this a validate if this is how you are trying it [Document Understanding - Extracting Multiple Tables Into Different Sheets From a Single Page] Cheers

Present Validation Station: The value must either contain a textual value or have components, not both

@Nithya_Aashitha The format of these new values and the extracted values is different i beleive…which is causing this issue… I guess this would help…did you happen to add the confidence and other fields required also when you added cgst and all? Because the structure requires those values…if youa dd only value then the structure of it changes.would suggest any strctural changes to be considered…and also the fied type should not be changed… Cheers

[QTH]

Unfortunately, the only identical part is that both cases is that it uses validation station. I didn’t encounter any exception on the output return from validation station but the variable type ended up as a string instead of number which was defined in the AI center. Since my validation station is set to trigger only if confidence level is lower than a pre defined value, the same field is able to output the numerical value solely if not passed through the validation stage.