WebDriver UDF - Help & Support (IV)

[frank10]

18 minutes ago, Danp2 said:

@frank10 Today's release on Github contains the fix for the problems related to reformatting of JS code. Please update when you have an opportunity and let us know if you encounter any further issues.

Hi Danp2, I'm glad to help!

Installed and tested: it works very well. Uploading .js code as-is is the best solution even for me. Good work!

[frank10]

Maybe I could suggest something else:

Before, I created 2 WD sessions with 2 chrome instances, because I need one normal chrome on default profile and another chrome with noCORS controls to make ajax call on different domains.
It worked fine, having 2 sessionsID and working on both of them with autoit script. Something like this:

_WD_Startup()
; 1st normal chrome:
global $sChromeH10 = '{"capabilities": {"alwaysMatch": {"goog:chromeOptions": {"w3c": true, "args":["start-maximized",  " --user-data-dir=C:\\Users\\' & @UserName & '\\AppData\\Local\\Google\\Chrome\\User Data\\", "--profile-directory=Default"]}}}}'
global $sSessionH10 = _WD_CreateSession($sChromeH10)

;2nd noCORS chrome:
global $sDesiredCapabilitiesNoSecur = '{"capabilities": {"alwaysMatch": {"goog:chromeOptions": {"w3c": true, "args":["--disable-web-security", "--user-data-dir=yourCustomPath", "--auto-open-devtools-for-tabs", "--profile-directory=Default"]}}}}'
global $sSession = _WD_CreateSession($sDesiredCapabilitiesNoSecur)

 

Now I can't do the same thing using the new capabilities method (I slightly modified the _WD_SetupChrome to make multiple calls with different capabilities set) :

; 1st normal chrome:
local $aSessions = ["normal","noCors"]
Global $sChromeH10 = _WD_SetupChrome(false, "", false, $aSessions[0] )
_WD_Startup()  ; 1st
$sSessionH10 = _WD_CreateSession($sChromeH10)

;2nd noCORS chrome:
$sDesiredCapabilitiesNoSecur = _WD_SetupChrome(false, "", false, $aSessions[1] )
_WD_Startup()  ; 2nd
global $sSession = _WD_CreateSession($sDesiredCapabilitiesNoSecur)     ; noCORS

It opens 2 sessions with correct sessionID, BUT the problem is the 2nd _WD_Startup() overwrites the previous sessionID, so it remains active only the last sessionID.
But if I delete the 2nd _WD_Startup() it doesn't open the 2nd chrome instance!

One solution could be to move the initial lines outside from _WD_SetupChrome, at the beginning of the WD section script:

Local $sTimeStamp = @YEAR & '-' & @MON & '-' & @MDAY & '_' & @HOUR & @MIN & @SEC
_WD_Option('Driver', 'chromedriver.exe')
_WD_Option('Port', _WD_GetFreePort() )
Local $sDriverParams = '--verbose --log trace --port=' & $_WD_PORT
If $bLogToFile Then $sDriverParams &= ' --log-path="' & @ScriptDir & '\log\' & $sTimeStamp & '_WebDriver_chrome.log"'
_WD_Option('DriverParams', $sDriverParams)

This way WD is initialized once at the beginning and you can call _WD_Startup() 1 time and you will have 2 different working sessionID.

Edited August 1, 2023 by frank10

[Danp2]

By default, existing webdriver instances are closed when you call _WD_Startup. The default functionality can be adjusted using one or both of these settings --

_WD_Option('DriverClose', False)
_WD_Option('DriverDetect', False)

[frank10]

8 minutes ago, Danp2 said:

By default, existing webdriver instances are closed when you call _WD_Startup. The default functionality can be adjusted using one or both of these settings --

_WD_Option('DriverClose', False)
_WD_Option('DriverDetect', False)

Ok, I see.
Do you mean making something like this, disabling those default behaviour on the 2nd call?

; 1st normal chrome:
local $aSessions = ["normal","noCors"]
Global $sChromeH10 = _WD_SetupChrome(false, "", false, $aSessions[0] )
_WD_Startup()  ; 1st
$sSessionH10 = _WD_CreateSession($sChromeH10)

;2nd noCORS chrome:
$sDesiredCapabilitiesNoSecur = _WD_SetupChrome(false, "", false, $aSessions[1] )
_WD_Option('DriverClose', False)
_WD_Option('DriverDetect', False)
_WD_Startup()  ; 2nd
global $sSession = _WD_CreateSession($sDesiredCapabilitiesNoSecur)     ; noCORS

But it doesn't work. It looses the 1st sessionID and gives error:  Invalid session ID [16]

 

[Danp2]

Yes, but it's more complicated than that because you need to supply a unique port number for each instance of the webdriver. You should be able to find a working example if you search the forums. Another option would be to use a single instance of Chromedriver to run two sessions of Chrome (only call _WD_Startup once, but call _WD_CreateSession twice).  Not all webdrivers support this option.

[Erik_NJ]

I see a new version of the UDF however, checking _WD_IsLatestRelease() returns FALSE (yes, I have updated the files on my system)

[Danp2]

@Erik_NJ It looks like I had a missing period in the release's tag (1.11 vs 1.1.1). I corrected the tag and it should be working now. Thanks for catching this!

[NassauSky]

Hi @Danp2 and all,

 

What might I be missing in trying to get the webdriver session connected to the currently clicked tab.

;-- Enum through open tabs and find visible tab and set session to use it
    Local $aHandles = _WD_Window($sSession, 'handles')
    For $w = 0 To UBound($aHandles) - 1
        _WD_Window($sSession, 'Switch', '{"handle":"' & $aHandles[$w] & '"}') ;One by one switch and check for visibility

        $currentWindowTitleWD = _WD_Action($sSession, 'Title')
        ConsoleWrite("Current Session's Window WD Title: " & $currentWindowTitleWD & @CRLF)

        $sVisibility =  _WD_ExecuteScript($sSession, "return document.visibilityState;")
        ConsoleWrite("+Tab Visibility: " & $sVisibility & @CRLF)
    Next
    
    ;More commands to send to the currently clicked tab ie  _WD_Navigate($sSession, "www.google.com")

 

Since the code runs and it actually switches the active tab, I can't seem to find out which one I clicked on.  I was hoping the visibility state would allow me to 'switch' over to it so future commands would  communicate with it.

 

Edited August 19, 2023 by NassauSky

[Danp2]

@NassauSky Try checking document.hidden instead.

[NassauSky]

@Danp2 thanks it doesn't seem the problem being javascript or property returned. The problem seems to be as I loop through the array of handles which seems to require me to switch to each. When I switch to each, they all will be visible after the switch to look at the state.  I'm missing a way to loop through the tabs and find a visible window without 'switching' and making each active as I'm currently searching.

Make sense?

Edited August 19, 2023 by NassauSky

[Danp2]

AFAIK, that is expected behavior with standard Webdriver. The only way I've managed to do it is to restart the webdriver whenever I detect that the active tab was changed by the user. The new webdriver instance will automatically use the current tab. This functionality requires the ability to reattach the webdriver to an existing browser (I was using FF).

[NassauSky]

I have put together the following demo of trying to acquire the current tab name.  Any idea why it takes quite a few seconds after I click at times.  If you click through tabs sometimes it instantly determines the tab title and other times it doesn't, especially if you click on a page's contents before clicking on the tab again.

 

expandcollapse popup

#include <GUIConstantsEx.au3>
#include "Required-AU\WebDriver\wd_core.au3"
#include "Required-AU\WebDriver\wd_helper.au3"

Global $debugMode = False
Global $OpenAuthInNewTab = False

Global $RequiredFolder = @ScriptDir & "\Required\"
Global $jUnhandled = "ignore"
Global $jExclude = "enable-automation"
Global $jAllowInstalls = "false"
Global $jPassDialog = "false"
Global $jPassMgr = "false"
Global $acPosS[4] = [0, 0, 1060, @DesktopHeight - 34]
Global $jaWinSizePos = '"window-position=' & $acPosS[0] & ',' & $acPosS[1] & '","window-size=' & $acPosS[2] & ',' & $acPosS[3] & '"'
Global $jArgs = $jaWinSizePos
Global $sDesiredCapabilities, $sSession, $handleTitle, $Handle, $aHandlePos


_SetupChromeSimple()
_WD_Option('DriverDetect', False)
_WD_Startup()
$sSession = _WD_CreateSession($sDesiredCapabilities)
$handleTitle = _WD_Action($sSession, 'title')
$Handle = WinGetHandle($handleTitle)
$aHandlePos = WinGetPos($Handle)

;-- Open 3 tabs
;If $OpenAuthInNewTab Then _WD_NewTab($sSession)
$sURL = "https://www.google.com"
_WD_Navigate($sSession, $sURL)

_WD_NewTab($sSession)
$sURL = "https://www.webpagetest.org/"
_WD_Navigate($sSession, $sURL)

_WD_NewTab($sSession)
$sURL = "https://www.geeksforgeeks.org/"
_WD_Navigate($sSession, $sURL)


AdlibRegister("_CheckMouseClick", 100) ; Register the function to run every 100ms
Global $hGUI = GUICreate("Blank GUI", 600, 400, 1080, 400)
Global $hEditBox = GUICtrlCreateEdit("Click In Browser For Window Title ...", 10,100,580,290)
            GuiCtrlSetFont(-1, 14)
GUISetState(@SW_SHOW, $hGUI)

While 1
        Switch GUIGetMsg()
                Case $GUI_EVENT_CLOSE
                        WinClose($Handle)
                        ExitLoop
        EndSwitch
        If Not WinExists($Handle ) Then Exit
WEnd

Func _SetupChromeSimple($chromeHidden = False)
        While ProcessExists("chrome.exe")
                ProcessClose("chrome.exe")
        WEnd
        While ProcessExists("chromedriver.exe")
                ProcessClose("chromedriver.exe")
        WEnd

    Global $_WD_DEBUG = $_WD_DEBUG_None
    Global $wasWDupdated = _WD_UpdateDriver('Chrome', $RequiredFolder, True, True)
    _WD_Option('Driver', $RequiredFolder & 'chromedriver.exe')
    _WD_Option('Port', 9515)
    _WD_Option('DriverParams', '--verbose --log-path="' & $RequiredFolder & 'chrome.log"')
    _WD_Option('DriverDetect', False)

    Global $jChromeCapabilities = FileRead($RequiredFolder & "ChromeCapabilities.json")
    $jChromeCapabilities = StringReplace($jChromeCapabilities, '$jUnhandled', $jUnhandled)
    $jChromeCapabilities = StringReplace($jChromeCapabilities, '$jExclude', $jExclude)
    $jChromeCapabilities = StringReplace($jChromeCapabilities, '$jAllowInstalls', $jAllowInstalls)
    $jChromeCapabilities = StringReplace($jChromeCapabilities, '$jPassDialog', $jPassDialog)
    $jChromeCapabilities = StringReplace($jChromeCapabilities, '$jPassMgr', $jPassMgr)
    $jChromeCapabilities = StringReplace($jChromeCapabilities, '$jArgs', $jArgs)

    $sDesiredCapabilities = $jChromeCapabilities
EndFunc

Func _CheckMouseClick()
    If _IsMouseClickDetected($Handle) Then ; $GhostGUI

                    ConsoleWrite("Mouse Event From _CheckMouseClick" & @CRLF)
                    $mX = MouseGetPos()[0]
                    $mY = MouseGetPos()[1]
                    GUICtrlSetData($hEditBox, "Clicked Inside Browser." & @CRLF & @CRLF & "Please Wait For Tab Title. This could take quite a few seconds...")
                    Sleep(20)
                    GUICtrlSetData($hEditBox, "Most Current Clicked In Tab: " & @CRLF & @CRLF & ClickedTabName())

    EndIf
EndFunc

Func _IsMouseClickDetected($hWndClicked)
    Local $aMousePos = MouseGetPos()
    Local $aWinPos = WinGetPos($hWndClicked)
    If IsArray($aWinPos) Then
        If $aMousePos[0] >= $aWinPos[0] And $aMousePos[0] <= $aWinPos[0] + $aWinPos[2] _
        And $aMousePos[1] >= $aWinPos[1] And $aMousePos[1] <= $aWinPos[1] + $aWinPos[3] _
        And _IsPressed("01") Then ; Check if left mouse button is pressed over browser window
            Return True
        Else
            Return False
        EndIf
    Else
        Return False
    EndIf
EndFunc

Func ClickedTabName() ; Get OuterHTML under MouseClick
    Local $aMousePos = MouseGetPos()

    $visibleTab = ""
    While StringLen($visibleTab)=0
        $visibleTab = WinGetTitle( "[CLASS:Chrome_WidgetWin_1]","")
        Sleep(40)
    Wend
    Return $visibleTab
EndFunc

 

ChromeCapabilities.JSON: (Installed in @ScriptDir & "\Required\"

{  "capabilities": {    "alwaysMatch": {      "unhandledPromptBehavior": "$jUnhandled",      "goog:chromeOptions": {        "w3c": true,        "excludeSwitches": [          "$jExclude"        ],        "useAutomationExtension": $jAllowInstalls,        "prefs": {          "credentials_enable_service": $jPassDialog,          "profile": {            "password_manager_enabled": $jPassMgr          }        },        "args": [          $jArgs        ]      }    }  }}

 

Edited August 21, 2023 by NassauSky
Needed To Add ChromeCapabilities.JSON

[mLipok]

why don't you use _WD_Capabilities....()?

Edited August 20, 2023 by mLipok

[Danp2]

@mLipok As you know they are optional. 😉

@NassauSky Your code doesn't run because we don't know the contents of ChromeCapabilities.json.

[NassauSky]

@Danp2 Sorry about that. I included just added it to my original post.

[Danp2]

7 hours ago, NassauSky said:

        $visibleTab = WinGetTitle( "[CLASS:Chrome_WidgetWin_1]","")

This isn't guaranteed to access the desired window because some other controls / windows use this same class. Why not do it like this?

$visibleTab = WinGetTitle( $Handle)

 

[sergiobaiao]

Hi there,

I'm just starting to use AutoIT and i was given the task to create an autologin script for the following website using the Webdriver:

https://painel.assertivasolucoes.com.br/login

But, I have no idea where to start. Can you give some advice or a sample code for an autologin script?

[mLipok]

1 hour ago, sergiobaiao said:

I have no idea where to start. Can you give some advice or a sample code for an autologin script?

Welcome to the forum.

Start here:

https://www.autoitscript.com/wiki/WebDriver

And after you do first step take a look here:

and here:

https://github.com/mlipok/Au3WebDriver-testing

[mLipok]

@Danp2 today I hit a problem with some JS scripting and finally I found this interesting 

Difference between textContent vs innerText

In my cases it was related to:

Quote

innerText returns the visible text contained in a node, while textContent returns the full text.

 

[sergiobaiao]

1 hour ago, mLipok said:

Welcome to the forum.

Start here:

https://www.autoitscript.com/wiki/WebDriver

And after you do first step take a look here:

and here:

https://github.com/mlipok/Au3WebDriver-testing

I was able to make the autologin work, but i have some issues:

1 - It's taking some time to set the values of the fields i've set the xpath for.

2 - The password field has the "eye" to see the "typed" password. As the form filling is taking some time to occur, the user is still able to click on the "eye" and see the password

3 - I need the browser to stay open after it logs in. The only need for this is that the user doesn't have access to the password.

 

Can you help?