Clearing the web browser cache from Subject7

Hi William,

Could you please try taking a screenshot for the button, upload it, use it in a Sikuli locator and use this locator in Desktop_Click command?

I did a simple test and it worked. This post also shows some helpful info if you are using Desktop_Hotkey command to show Clear Cache window:
https://subject7.zendesk.com/hc/en-us/community/posts/360046787914-Desktop-Hotkeys-command-not-working-for-CTRL-SHIFT-DELETE-in-Chrome-Firefox-

Regards,
Subject7 Team

I know it has been a while, but I am finally getting some time to work on this.

I tried the Desktop_Click command with a Sikuli locator loaded with an image of the “Clear Data” button. That did not appear to work. Any ideas what I might be doing wrong?

In the Desktop_Click_Options locator I have selected "Use Locator".   The reference point contains the Sikuli Locator (chrome_clear_data) and offsets and extension values are left at the default of 0.   The Similarity value is also at the default of 0.70.

Hi William,

Any luck with trying CTRL+SHIFT+DELETE as shown in the post below?

https://subject7.zendesk.com/hc/en-us/community/posts/360046787914-Desktop-Hotkeys-command-not-working-for-CTRL-SHIFT-DELETE-in-Chrome-Firefox-

Regards,
Subject7 Team

Hi Joseph:

I guess that didn't appear to really address the issue I was having.   The test does the CTRL+SHIFT+DELETE ok already and I get the popup up which asks me to confirm that I want to clear the data.   It is pressing the "Clear data" button which is the issue.    I am not having the issue addressed in the article which suggests that the CTRL+SHIFT+DELETE is not working.   It is.

Hi William,

Thank you for responding and confirming. When it comes to the image, does any element get highlighted or the step with Desktop_Click fails with "Supporting element not found"? Can you capture the image one more time, go to Artifacts > File Uploads, re-upload the image that the Sikuli locator is using and try one more time?

When it comes to different browsers, there's a simple way to do it with conditions as below. Basically using the variable @system.browser will allow you to use different steps when the test case is launched on a specific browser.

1. When Chrome is launched, you can use Desktop_Hotkeys command with CTRL+SHIFT+DELETE and Desktop_Click to click Clear Data button.

2. When Firefox is launched, you can use Desktop_Hotkeys command with CTRL+SHIFT+DELETE and Desktop_Hotkeys with ENTER.

3. When Edge is launched, you can use Desktop_Hotkeys command with CTRL+SHIFT+DELETE and Desktop_Click to click Clear Now button.

Regards,
Subject7 Team

Thanks Joseph:

So, here is where I am at:
WIth regard to the items you listed, (1, 2, and 3) and @system_browser thanks, I was aware of those areas and planned to utilize that capability once I had this figured out.  Thanks for the reminder.

With regard to the locator, I did get it working, sort of.  

I am running on a windows PC with two screens.   I found that the locator works when I execute the execution screen displaying on one of my monitors but not the other.  Also, only if the image is extracted on that monitor.  Originally I had the image extracted on the other monitor, and the locator did not work on either monitor.

So, there is a solution which works, but I am not sure how satisfying it is. The results are non-portable to other testers with different monitor, and relies on my executing the test on a particular monitor.  Also, I notice when I am moving the mouse, the desktop locator cannot click on the item.  I have to not touch the mouse.

Hi William,

One quick question, what is the operating system that is mostly used to run test cases? Is it Windows or MacOS ?

Regards,
Subject7 Team

Windows for my group.

Hi William,

Here is a quick workaround. You can copy the snippet below, go to Advanced View icon, paste it and click Apply. I recommend creating a test with just these steps, try it out and if it works, make it a function and call it in your test case when required.

Commands Editor tab snippet:

WEB GOTO_URL "https://github.com/" HALT ON ERROR 
PRINT TEXT "Do some stuff" OUTPUT CONSOLE HALT ON ERROR 
DESKTOP HOTKEYS CTRL,SHIFT,DELETE TARGET FOCUSED_ELEMENT HALT ON ERROR 
IF @system.browser EQUAL "CHROME" AS TEXT 
WEB SWITCH WINDOW BY POPUP "Settings" DURATION_TYPE WHOLE_TEST_CASE HALT ON ERROR 
WEB EXECUTE SCRIPT "document.querySelector('settings-ui').shadowRoot\n .querySelector('settings-main').shadowRoot\n .querySelector('settings-basic-page').shadowRoot\n .querySelector('settings-privacy-page').shadowRoot\n .querySelector('settings-clear-browsing-data-dialog').shadowRoot\n .querySelector('cr-dialog[id=clearBrowsingDataDialog]')\n .querySelector('cr-button[id=clearBrowsingDataConfirm]')\n .click();" HALT ON ERROR 
WEB SWITCH WINDOW BY MAIN DURATION_TYPE DEFAULT HALT ON ERROR 
ELSE_IF @system.browser EQUAL "FIRE_FOX" AS TEXT 
DESKTOP HOTKEYS ENTER TARGET FOCUSED_ELEMENT HALT ON ERROR 
ELSE_IF @system.browser EQUAL "EDGE" AS TEXT 
WEB SWITCH WINDOW BY POPUP "Settings" DURATION_TYPE WHOLE_TEST_CASE HALT ON ERROR 
WEB EXECUTE SCRIPT "document.querySelector('button[id=\"clear-now\"]').click();" HALT ON ERROR 
WEB SWITCH WINDOW BY MAIN DURATION_TYPE DEFAULT HALT ON ERROR 
IF_END

Or JSON Editor tab snippet (Dashboard Preferences should have Compact view selected):

{"command":"GOTO_URL","args":{"url":"https://github.com/"}}
{"command":"PRINT","args":{"message":"Do some stuff","includeHeaders":true,"type":"MESSAGE","target":"CONSOLE"}}
{"command":"DESKTOP_HOTKEYS","args":{"componentType":"FOCUSED","hotKeys":["CTRL","SHIFT","DELETE"],"desktop_element":"<NOT SELECTED>","similarity":0.7}}
{"command":"IF","args":{"storeResult":false,"operator":"EQUAL","comparisonType":"TEXT","caseSensitive":false,"leftValue":"@system.browser","rightValue":"CHROME","threshold":0.1}}
{"command":"SWITCH_WINDOW","args":{"value":"Settings","switchType":"POPUP","switchDurationType":"WHOLE_TEST_CASE"}}
{"command":"EXECUTE_JAVASCRIPT","args":{"locators":[],"snippet":"document.querySelector('settings-ui').shadowRoot\n .querySelector('settings-main').shadowRoot\n .querySelector('settings-basic-page').shadowRoot\n .querySelector('settings-privacy-page').shadowRoot\n .querySelector('settings-clear-browsing-data-dialog').shadowRoot\n .querySelector('cr-dialog[id=clearBrowsingDataDialog]')\n .querySelector('cr-button[id=clearBrowsingDataConfirm]')\n .click();"}}
{"command":"SWITCH_WINDOW","args":{"switchType":"MAIN","switchDurationType":"DEFAULT"}}
{"command":"ELSE_IF","args":{"storeResult":false,"operator":"EQUAL","comparisonType":"TEXT","caseSensitive":false,"leftValue":"@system.browser","rightValue":"FIRE_FOX","threshold":0.1}}
{"command":"DESKTOP_HOTKEYS","args":{"componentType":"FOCUSED","hotKeys":["ENTER"],"desktop_element":"<NOT SELECTED>","similarity":0.7}}
{"command":"ELSE_IF","args":{"storeResult":false,"operator":"EQUAL","comparisonType":"TEXT","caseSensitive":false,"leftValue":"@system.browser","rightValue":"EDGE","threshold":0.1}}
{"command":"SWITCH_WINDOW","args":{"value":"Settings","switchType":"POPUP","switchDurationType":"WHOLE_TEST_CASE"}}
{"command":"EXECUTE_JAVASCRIPT","args":{"locators":[],"snippet":"document.querySelector('button[id=\"clear-now\"]').click();"}}
{"command":"SWITCH_WINDOW","args":{"switchType":"MAIN","switchDurationType":"DEFAULT"}}
{"command":"END_IF","args":{}}

However, Selenium4 has a much better way to handle this, we are still waiting for a more stable release of it and then we will just provide commands to handle similar things using Selenium4.

Regards,
Subject7 Team

Thanks.  I used the Commands Editor snippet and it worked well enough.   With regard to the desktop hotkeys mentioned earlier, that works as well.   I did leave a comment in that conversation about how adding a NUM_LOCK change into the test causes it to change state each time I execute the test.   I think I have some coding workarounds I can try, but it won't be clean.   The above code is useful to me even without the workarounds.   Thanks.