WebDriver UDF - Help & Support

[GregEisenberg]

Return value from _WD_ElementAction

I am trying to use the WebDriver to find the rectangle of an element on a page.  The following is a function I wrote to do this, and in its current form it includes a bunch of debugging ConsoleWrite's also. 

The main issue is the return value from _WD_ElementAction.  I have read _WD_ElementAction source code and I would expect it to return a string a Json BUT when I use it i get nothing as a return.   Initially I expected the response to be 

Quote

{"value":{"height":21,"width":79,"x":8,"y":62}}

Which is what i see in the debug output.  But on closer look to the source code it seems that _WD_ElementAction may parse out the "value" and only return 

Quote

{"height":21,"width":79,"x":8,"y":62}

But in my code I get nothing as a return value. 

 

Below is my code. 

Can anyone help?  
Should I expect a response from _WD_ElementAction?
Should I expect it to be JSon?
 

Func FindElementRect($session, $elementToFind)

   $feResult = _WD_FindElement($session, $_WD_LOCATOR_ByID, $elementToFind)
   ConsoleWrite("_WD_FindElement result: " & $feResult & @CRLF)

   ConsoleWrite("**** : " & _WD_ElementAction($session, $feResult, "rect"))     ; i can see something returned here

   $eaResult = _WD_ElementAction($session, $feResult, "rect")   ; in this case eaResult is blank
   ConsoleWrite("$eaResult: " & $eaResult & @CRLF)

   ConsoleWrite("@error is: " & @error & @CRLF)                 ;  never an error

   ; per debug log this is what I expect to get
   ;{"value":{"height":21,"width":79,"x":8,"y":62}}

   $oJson = Json_Decode($eaResult)
   Local $rect[4]
   $rect[0] = Json_Get($oJson, "[value][x]")
   ConsoleWrite("top Left X: " & $rect[0] & @CRLF)

   $rect[1] = Json_Get($oJson, "[value][y]")
   ConsoleWrite("top Left Y: " & $rect[1] & @CRLF)

   $rect[2] = $rect[0] + Json_Get($oJson, "[value][width]")
   ConsoleWrite("bottom right X: " & $rect[2] & @CRLF)

   $rect[3] = $rect[1] + Json_Get($oJson, "[value][height]")
   ConsoleWrite("bottom right Y: " & $rect[3] & @CRLF)

   return $rect
EndFunc

Many thanks in advance

Greg

[Danyfirex]

Hello @GregEisenberg You need to do this:

Local $oJsonRect=_WD_ElementAction($sSession, $sElements, 'rect')
ConsoleWrite("!" & Json_Get($oJsonRect, "[x]") & @CRLF)
ConsoleWrite("!" & Json_Get($oJsonRect, "[y]") & @CRLF)
ConsoleWrite("!" & Json_Get($oJsonRect, "[width]") & @CRLF)
ConsoleWrite("!" & Json_Get($oJsonRect, "[height]") & @CRLF)

 

 

Saludos

Edited June 11, 2018 by Danyfirex
Code tag fix

[Danp2]

@JonnyQuy You will need to supply more details if you want further assistance. For starters --

• Which browser are you attempting to automate?
• Have you downloaded the correct webdriver? If yes, where did you save it?
• Did you modify wd_test.au3 to point to the above location?
• Is at least one of the entries in the array $aTestSuite set to True?

If all of the above check out, then run the code in Scite and post the results from the Output pane.

[GregEisenberg]

@Danyfirex Thank you.  I guess what makes debugging this hard is that when you assign the output of _WD_ElementAction to some local variant and then ConsoleWrite that variant - it prints nothing. You did answer my questions though and I really appreciate it.   For example in your code if you simply do 

ConsoleWrite("$oJsonRect: " & $oJsonRect & @CRLF) 

you end up with 

Quote

$oJsonRect: 

and that makes the user THINK they are getting back nothing. 

BTW:  Kudos to @Danp2 for this excellent UDF (and kudos to those who built it's dependencies too.) 

 

[Decibel]

3 minutes ago, GregEisenberg said:

@Danyfirex Thank you.  I guess what makes debugging this hard is that when you assign the output of _WD_ElementAction to some local variant and then ConsoleWrite that variant - it prints nothing. You did answer my questions though and I really appreciate it.   For example in your code if you simply do 

ConsoleWrite("$oJsonRect: " & $oJsonRect & @CRLF)

you end up with 

and that makes the user THINK they are getting back nothing. 

BTW:  Kudos to @Danp2 for this excellent UDF (and kudos to those who built it's dependencies too.) 

 

Danyfirex beat me to the answer by seconds, so I cancelled my reply.

What happens is _WD_ElementAction's "rect" $sCommand returns a JSON Object, therefore you don't need to Json_Decode the return variable.

In the console window's debug you get back:

_WD_ElementAction: {"value": {"x":966.0,"y":19.0,"width":34.0,"height":24.0}}

That gets reduced to {"x":966.0,"y":19.0,"width":34.0,"height":24.0} because the $sResponse gets run through a JSON_Decode and Json_Get($oJSON, "[value]") before getting returned to your variable. Notice that the parent "value" is now gone from the new JSON Object. That's why you omit it when you do your own Json_Get.

[JonnyQuy]

 

On 12/6/2018 at 12:56 AM, Danp2 said:

@JonnyQuy You will need to supply more details if you want further assistance. For starters --

• Which browser are you attempting to automate?
• Have you downloaded the correct webdriver? If yes, where did you save it?
• Did you modify wd_test.au3 to point to the above location?
• Is at least one of the entries in the array $aTestSuite set to True?

If all of the above check out, then run the code in Scite and post the results from the Output pane.

• I'm running for chrome browser
• I downloaded "chromedriver_win32.zip", I saved it with the directory of wd_test.au3
• If it is in the same directory with each other need to modify the path or not?
• "TestAlerts", True I keep the default

Edited June 13, 2018 by JonnyQuy

[JonnyQuy]

Can you provide me basic video tutorial which is not I can not activate it

[Danp2]

Please post the complete output from Scite (Not an image; copy and paste the text here).

Also, you could try disabling the TestAlerts entry and then enabling one of the other tests.

Edited June 13, 2018 by Danp2

[JonnyQuy]

I do not understand, can you give me a few examples?

[Earthshine]

examples of how to paste the output of the editor?

[JonnyQuy]

I do not know how to let it work, Who can help me

[JonnyQuy]

_WD_Option('Driver', 'D:\AUTOIT\WebDriver UDF\WebDriver-0.1.0.10\WebDriver-0.1.0.10\chromedriver.exe')
_WD_Option('DriverParams', '--log-path=' & @ScriptDir & '\chrome.log')
_WD_Option('Port', 9515)

$sDesiredCapabilities = '{"desiredCapabilities":{"javascriptEnabled":true,"nativeEvents":true,"acceptInsecureCerts":true}}'
_WD_Startup()
$sSession = _WD_CreateSession($sDesiredCapabilities)

_WD_Navigate($sSession, "http://google.com")
$sElement = _WD_FindElement($sSession, $_WD_LOCATOR_ByXPath, "//a[@class='gb_P']") ;"Images" link at the upper right
$sValue = _WD_ElementAction($sSession, $sElement, 'text') ;gets the text content of the node/element
ConsoleWrite('$sValue = ' & $sValue & @CRLF) ;"Images"
MsgBox(0,0,$sSession)

help me

Edited June 18, 2018 by JonnyQuy

[Danp2]

@JonnyQuy ChromeDriver supports multiple standards, so you need to specifically designate that you want the W3C standard. Try changing your desired capabilities line to

$sDesiredCapabilities = '{"capabilities": {"alwaysMatch": {"chromeOptions": {"w3c": true }}}}'

 

[JonnyQuy]

i have Cookie: cid_ck:c27b40a4-a5c6-43ac-a5dd-4e4a4f79d082; Domain=.hotname.com; Expires=Tue, 18-Jun-2019 12:32:21 GMT; Path=/+>      ____ i get from HttpRequest
i get from: document.cookie => cid_ck:c27b40a4-a5c6-43ac-a5dd-4e4a4f79d082;
i get from  _WD_Cookies :  cid_ck","path":"/","secure":false,"value":"c27b40a4-a5c6-43ac-a5dd-4e4a4f79d082"}

how do i add cookie in _WD_Cookies($sSession, $sCommand, $sOption = '')
What do i do with $sOption ?

Edited June 18, 2018 by JonnyQuy

[Danp2]

@JonnyQuy While I haven't done much with that function, we previously had a discussion that you help you over in the Development thread.

[corgano]

I have two questions (Chromedriver)

1: working off the example provided in the main thread, I can have it open chrome, log into a page (say google). The problem is when I re-start, it doesn't save the password or cookies so i have to log in every time. Is there a way to make user data persistant?

2: Every time I test my code it needs to open a new window. Is there a way to have it keep the same window between tests? So I can kill my script, start it, and just attach to chrome and continue on? Akin to _IE_Create() with tryattach set to 1. Basic attach if exists, else create new tab and navigate functionality.

[Danp2]

@corgano

1 hour ago, corgano said:

working off the example provided in the main thread, I can have it open chrome, log into a page (say google). The problem is when I re-start, it doesn't save the password or cookies so i have to log in every time. Is there a way to make user data persistant?

From http://chromedriver.chromium.org/capabilities --

Quote

Use custom profile (also called user data directory)

By default, ChromeDriver will create a new temporary profile for each session. At times you may want to set special preferences or just use a custom profile altogether. If the former, you can use the 'chrome.prefs' capability (described later below) to specify preferences that will be applied after Chrome starts. If the latter, you can use the user-data-dir Chrome command-line switch to tell Chrome which profile to use

I haven't actually done this, so you will need to work out the correct way to implement in Autoit.

1 hour ago, corgano said:

Every time I test my code it needs to open a new window. Is there a way to have it keep the same window between tests? So I can kill my script, start it, and just attach to chrome and continue on? Akin to _IE_Create() with tryattach set to 1. Basic attach if exists, else create new tab and navigate functionality.

I know this is possible in Firefox, but unsure if you can do this in Chrome. Please report back if you successfully find a way to do this.

[corgano]

4 hours ago, Danp2 said:

@corgano

From http://chromedriver.chromium.org/capabilities --

I haven't actually done this, so you will need to work out the correct way to implement in Autoit.

 

IT looks like it should be
_WD_Option('DriverParams', '--user-data-dir='&@ScriptDir&'\userdata\ --log-path=' & @ScriptDir & '\chrome.log')
The chrome.log is created right, but the user data dir isn't created or used if it exists. Any ideas?

Edit:

This leads me to believe it's not a command line peram, but it's stated here: http://chromedriver.chromium.org/capabilities
That it is possible. Where would I go to ask about it?

Edited June 20, 2018 by corgano

[Danp2]

You have to put it into ChromeOptions, which is part of the $sDesiredCapabilities string that is passed to _WD_CreateSession. I would suggest trying something like this --

$sDesiredCapabilities = '{"capabilities": {"alwaysMatch": {"chromeOptions": {"w3c": true, "args": [--user-data-dir='&@ScriptDir&'\userdata]}}}}'

 

[corgano]

Damnit, I literally just got it

$sDesiredCapabilities = '{"capabilities": {"alwaysMatch": {"chromeOptions": {"w3c": true, "args":["user-data-dir=userData"] } }}}'

However using @scriptdir didn't work for me. simply having =userData worked though.

What about hving a persistent window and just re-attaching to it? any ideas?