Issue with OCR step

Hi Alesia,
Having problems with OCR. Tried a number of alternatives. The steps work fine until it tries to read invoice.
Attaching script and invoice.
Thanks and Regards,
David

Origin invoice_a.pdf (235.2 KB)

Hi David,

What version of RPA Express are you using?

It is possible that the OCR license you use has expired. In this case you need to update to a new RPA Express version

Hi Alesia,
If I wanted to copy text from a particular part of a word document, say a table that appears half way down the page. How would i do this?
Secondly the mouse actions include the option to set x y coordinates but how do you know what these coordinates are?
Finally, what version of FineReader is WorkFusion using?
Regards,
David

Hi @david_masters you can automate Work with mouse clicks and keystrokes. You can scroll the document till the table appears, click on the icon that selects the whole table, copy it using Ctrl+C, and then set to a Table variable using Clipboard.

You can record the coordinates of mouse actions using “Insert recording” option. After the action has been recorded, you can switch it to “Position” mode - it will contain the recorded coordinates.

I hope it helps!

Hi Alesia,

As per previous correspondence I have been using OCR but decided that ABBYY FineReader 14 would be a more appropriate path for my client.

So what I trying to do is very simple.

  • Launch ABBYY FineReader 14.
  • Convert to Excel
  • Read file
  • Press icon to convert.
  • It then automatically launches Excel and you save the file.

Here is the problem.

I get to the Window to ENTER the file I want to open and the Window is inactive. RPA executes the next steps as if the command is executed and only fails later in the process, i.e., it simply does not fill in the file name from the ENTER action regardless if I use text or a variable.

I have got it to work, i.e., to ENTER the file name but never twice but only when I re-started the computer but only once or twice never consistently.

I have attached the simple steps and the step it fails on and image of the screen it fails on.

It is critical I get this working for my client.

Can you help?

Regards,

David

Hi Alesia,
These are the two files.
Thanks and Regards,
David

1 Like

Hi @david_masters you need to use the “Window” action to switch to the window in which the bot needs to perform the actions before typing or clicking there.
You can see how to use it correctly in this tutorial: https://www.youtube.com/watch?v=KdUaUNReixk

Thanks Alesia. That works.

1 Like

Hi Alesia,
I am working with FineReader 14 and it pops up a Window stating that it can’t convert the second page of the invoice. That is not a problem. You simply need to click ‘OK’ to exit the Window. The problem is, I don’t know when the Window pops up. Some invoices will have 2 pages whilst others don’t.
How do I put in a test condition to see if the Window is present.
I was thinking along the lines of using Get Window Title and retrieve the Active Window??? However, if this was an option, the option to get the ActiveWindow is no longer in the Action. This is inconsistent with the description in Help where it says you have 2 options: ActiveWindow or name. I only get name.
More broadly, how do you handle windows that pop up due to some condition?
Regards,
David

I think you can use a Wait for image action with some timeout and an If-Else condition. So the bot will look for the window with the message, and if it finds it will click on the OK button.

A post was split to a new topic: How to automate Outlook?