Targeting using an image

Help Studio
1

I have been asked to create a bot to automate a process in our business where we connect through RDS to another system and update their system with the information from our own system.
I do not have access nor permissions to install UiPath on the host Remote Desktop Server.
I think therefore my only option when targeting text box’s to enter the data is to use F3 and select the image.
This works broadly speaking as each entry box has a unique label describing the box.
However, one section for input looks like a grid or a table, where the labels are at the top describing the columns.
I need to enter data onto the first row and the negative equivalent in the second row.
My target works perfect for the top row. But seems to inconsistently choose either the second or third row for the negative values.

I drew a box around the text box as the target… Yes EVERY row has the same text box.
So I then drew a box from the top of the form, including titles, down over my target, as the anchor.

Is this correct or how else do you anchor on a picture based target?

2

Hi @steve.walls ,
Image automation is the last resource we opt for in the RPA world because it is highly unstable.

Try contacting the client and explain the situation that this is the scenario and Image automation is last option and this are the cons.

I think they will understand the scenario, provide you the access and you can perform normal RPA operations.

Thanks & Regards,
Shubham Dutta

3

Thanks Shubham.
Unfortunately, we have contacted them and they are not willing to install this.
As I can consistently target the top row, perhaps I will use the terminal function to send keys to get me to the next row and enter values from there without specifically targeting an element on row 2?
Would the Terminal function work like this?

4

So far we have one answer which basically says dont target via image.

As I want to target with Image and it’s a feature that exists. Does anyone understand how to use this feature?
If it’s as unreliable as to be suggested not to use it, why is it still a feature? So I have to assume, UiPath have value in it. So the question is how does Image targetting work? is it looking at pixels on screen or something else?