Inspector is a window displayed in the ELMA RPA recording mode that helps you save time when automating a process. Inspector allows you to view the actions recorded in the current session and manage the recording process. The Inspector window is displayed in the lower right corner of the screen and is pinned on top of all windows. The window consists of three sections.
- The window name containing the current element search method.
Hover over the icon to view the hotkey to call the radial menu (Ctrl+Alt+Q) in the recording mode. There is a button for collapsing/expanding the window to the right of its name.
- The list of actions recorded by a user (the last three).
It also displays the activity name and a screenshot of the captured interface element. Click the button to delete the last action.
- The process recording control panel:
— end the recording of the current session and save the process in ELMA RPA Designer. After clicking the button, a pop-up window will appear. Click Ok to confirm saving.
— temporary pause the process recording. After clicking the button, the red frame changes its color to gray, and the caption "REC" in the top left corner changes to "PAUSE". To return to the recording mode, click the button in the Inspector window.
— change the element search method (Native or Vision).
— end the recording session of the current process without saving.
When modeling a process, you can interact with different programs. However, not all the programs provide system data about the item you want to click or read. For example, a website or a program can be designed in a way that it is impossible to extract any labels or features when recognizing them. It is usually due to the implementation technology. For example, an RDP client only shows a screen image of the virtual machine we are connected to but does not transmit the information of open windows, elements, etc.
For these purposes, there are two element search methods used in the ELMA RPA that differ in interaction approach: Native and Vision.
- Native is a search method that extracts system information about the object of interaction — certain signs that help identify the element among other similar ones.
- Vision is a search method used when Native method cannot extract object features. The object image is saved, and with the help of computer vision algorithms, the required object is determined among the entire working area.
Computer vision algorithms do not simply compare the required image with the specified one. These algorithms include methods that help classify the object of interaction, which increases the system operating speed.