7.2.1 Surface Automation on Chrome

[Stephen.Jones]

Hello,

We have just started automation against chrome.  Some of the links on the website we are automating against cannot be clicked due to them invoking a java script.

To get around this , we are using surface automation to identify the link and then mouse clicking the link.

The issue we are having , is that the links have been identified in our Development environment , but when exported to our Production , they are no longer identified.

The following rules have been followed

• same desktop resolution in dev and prod
• same zoom mode of the browser in dev and prod
• font smoothing turned off in dev and prod

is there anything else we need to do with the chrome browser setup to match between dev and prod ?

[Denis__Dennehy]

Something is different between the environments.  Experiment with using colour tolerances, larger search areas (rather than exact location match) and even greyscale options as described in the Surface Automation training.   
If still a problem then i can suggest taking a screenshot in both environments and comparing them pixel by pixel to find and understand where the difference between images is.  I used to knock up a quick comparison process that loaded both pictures and compared one pixel at a time - but I'm guessing there may be quicker ways. 

[John__Carter]

Try using a global mouse click like this.

• Spy the outer window of the browser as a Win32 element
• Spy the inner window (where the web content is) of the browser as a Win32 element
• Read the bounds of the web element
• Scroll if necessary
• Activate the outer window
• Global click the inner window with xy at centre of the bounds

[Michael_S]

Hi there @Stephen.Jones - just wanted to check in and see if @Denis__Dennehy or @John__Carter's tips here helped you? If they did, please could you update us? It's always great to know the solution to a problem so that we can share it with the wider community!