Jump to content

Recommended Posts

Posted

v0.3.0.8 just released with the following updates --

  - Changed (_WD_WaitElement): Added optional parameter to check elements enabled status
  - Changed (_WD_GetTable): Optionally support faster _HtmlTableGetWriteToArray
  - Changed (_WD_ElementAction): Allow retrieving element value with the 'value' command
  - Chore: Modified #include usage

Posted (edited)

Below is my version of the _WD_GetTable() helper function (named _WD_GetTable_2) that I use for grabbing tables from web sites.  The difference between it and the existing _WD_GetTable() is that it spins thru the rows to find the maximum number of columns in the table.  As you know, the current _WD_GetTable() only checks the first <tr> to determine the number of columns.  I found that to be problematic for some of the tables that I have come across.  This is certainly not the fastest way to get table data, but it is a more reliable way to get table data and uses native WebDriver functions.  Maybe others may find it useful in those cases where the first row of the table doesn't reflect all of the columns in the body of the table.

; #FUNCTION# ====================================================================================================================
; Name ..........: _WD_GetTable_2
; Description ...: Return all elements of a table
; Syntax ........: _WD_GetTable_2($sSession, $sBaseElement)
; Parameters ....: $sSession     - Session ID from _WDCreateSession
;                  $sTableXpath  - XPath of the table to return
; Return values .: Success      - 2D array
;                  Failure      - ""
;                  @ERROR       - $_WD_ERROR_Success
;                               - $_WD_ERROR_Exception
;                               - $_WD_ERROR_NoMatch
;                  @EXTENDED    - WinHTTP status code
; Author ........: TheXman
; Modified ......:
; Remarks .......: Number of rows and max cols are determined based on actual table data.
; Related .......:
; Link ..........:
; Example .......: No
; ===============================================================================================================================
Func _WD_GetTable_2($sSession, $sTableXpath)

    Const $sFuncName = "_WD_GetTable_2"

    Local $aTable[0], $aRowElements[0], $aColElements[0]
    Local $sRowColumns = ""
    Local $iRows = 0, $iCols = 0, $iChildCount = 0

    ;Get table row elements
    $aRowElements = _WD_FindElement($sSession, $_WD_LOCATOR_ByXPath, $sTableXpath & "/tbody/tr", "", True)
    If @error <> $_WD_ERROR_Success Then Return SetError(__WD_Error($sFuncName, @error, "HTTP status = " & $_WD_HTTPRESULT), $_WD_HTTPRESULT, "")
    $iRows = UBound($aRowElements)

    ;Find max number of columns by checking each row's child count property
    For $iRow = 0 To UBound($aRowElements) - 1
        $iChildCount = _WD_ElementAction($sSession, $aRowElements[$iRow], "property", "childElementCount")
        If @error <> $_WD_ERROR_Success Then Return SetError(__WD_Error($sFuncName, @error, "HTTP status = " & $_WD_HTTPRESULT), $_WD_HTTPRESULT, "")

        ;If number of row columns is greater the previous value, then save new value
        If $iChildCount > $iCols Then $iCols = $iChildCount
    Next

    ;Dimension and load table data
    ReDim $aTable[0][$iCols]

    For $iRow = 0 To $iRows - 1
        ;Get column elements (td or th)
        $aColElements = _WD_FindElement($sSession, _
                                        $_WD_LOCATOR_ByXPath, _
                                        "./td|./th", _
                                        $aRowElements[$iRow], _
                                        True)
        If @error <> $_WD_ERROR_Success Then Return SetError(__WD_Error($sFuncName, @error, "HTTP status = " & $_WD_HTTPRESULT), $_WD_HTTPRESULT, "")

        ;Build pipe-separated row of column data
        $sRowColumns = ""
        For $iCol = 0 To UBound($aColElements) - 1
            If $iCol = 0 Then
                $sRowColumns &= _WD_ElementAction($sSession, $aColElements[$iCol], "text")
            Else
                $sRowColumns &= "|" & _WD_ElementAction($sSession, $aColElements[$iCol], "text")
            EndIf
        Next

        ;Add row to array
        _ArrayAdd($aTable, $sRowColumns)
    Next

    ;Return result table
    Return $aTable

EndFunc

 

Here's an example of a table and its definition that the current _WD_GetTable() has a problem with because the first <tr> does not accurately define the max number of columns.

image.png.ade86008eae9a19c1e93612386ed0189.png

image.png.bba4ef7d93b35abd746768517c3951aa.png

 

Edited by TheXman
Updated script to handle "irregular" tables without throwing and error
Posted (edited)

https://fftoday.com/nfl/schedule.php

It is a 1200 cell table, so it takes a bit to build. It is 300 rows with 4 columns per row.

Edited by TheXman
Posted (edited)
1 hour ago, Danp2 said:

@Chimp's _HtmlTableGetWriteToArray seem to handle this table correctly.

Of course this assumes that any given user has found and included that function.  If you plan to include it in one of your UDFs, then it really isn't an issue.  If not, then the current _WD_GetTable() helper function would not properly handle such tables by default.  I didn't offer the function as a replacement or to compete with Chimp's function.  I just threw it out there as an alternative and to show a relatively easy, native WebDriver, way to handle all tables including edge cases like my example.

Edited by TheXman
Posted (edited)

Yes, I noticed the updated _WD_Getable() function yesterday.  That's why I mentioned the part about "If a user has found and included the function".  Because if they hadn't, it would fall back to using the existing logic.

I hadn't started playing with the WebDriver UDFs until 2 days ago.  It is quite an impressive piece of work.  Kudos to you and all of the contributors!  :thumbsup:

Edited by TheXman
Posted (edited)

I noticed one more small issue with the current _WD_GetTable helper function.  If you run it against the Customers table on https://www.w3schools.com/html/html_tables.asp you will see that it fails with a NoMatch because the first <tr> does not have <td> tags, it has <th> tags defining the headers.  The _WD_GetTable_2() function, that I posted above, also handles this issue.

_WD_GetTable($gsSession, '//table[@id="customers"]')

image.png.8270060fc5c278279676bbbc06c65cd8.png

image.png.0632a9f43f756b7b1f6cbc0fb0490ca3.png

Edited by TheXman
Posted (edited)

Hi @TheXman,

from a quick test on the above table it seems that I get the same array using both_WD_GetTable () and _WD_GetTable_2 (), but _WD_GetTable_2 () is much slower.
I also did a test on the second table at this link (https://www.w3.org/WAI/tutorials/tables/irregular/), but _WD_GetTable_2 () seems to go wrong.

_WD_GetTable_2($sSession, "/html/body/div/main/article/div/figure[4]/div/table")


Can you take a test on that table too?

Edited by Chimp

 

image.jpeg.9f1a974c98e9f77d824b358729b089b0.jpeg Chimp

small minds discuss people average minds discuss events great minds discuss ideas.... and use AutoIt....

Posted (edited)

@Chimp

_WD_GetTable_2(), and WebDriver in general, will always be MUCH slower because WebDriver has to make a GET/POST request for each and every thing it does.  Of course it will be much slower than a function that processes the raw source as a whole.  That was never a question and I stated so in my earlier post.  My original point was that without your function, the WebDriver helper function did not correctly process a table that I encountered.  In addition, _WD_GetTable2() will be slower because it checks all table rows, before processing it, to determine the max number of columns in the table.  The current _WD_GetTable() function, without your UDF, only looks at the first <tr> in order to determine the number of columns.  As I pointed out, that logic didn't work on the table that I provided.  As I also pointed out, it doesn't work on the W3Schools table example either.  I'm not sure what you tested and what you are referring to when you say that you get the same array with the existing function and my function when I clearly showed that isn't the case (assuming the original logic and not using your function).

There are 2 tables on the page that you referenced.  Are you referring to a particular table?

 

Edited by TheXman
Posted
8 minutes ago, TheXman said:

There are 2 tables on the page that you referenced.  Are you referring to a particular table?

I did a test on the second table using the following xpath

_WD_GetTable_2($sSession, "/html/body/div/main/article/div/figure[4]/div/table")

 

 

image.jpeg.9f1a974c98e9f77d824b358729b089b0.jpeg Chimp

small minds discuss people average minds discuss events great minds discuss ideas.... and use AutoIt....

Posted (edited)
2 hours ago, Chimp said:

I also did a test on the second table at this link (https://www.w3.org/WAI/tutorials/tables/irregular/), but _WD_GetTable_2 () seems to go wrong.

Yes, _WD_GetTable_2() encountered a NoMatch while processing that table on the irregular tables site.  Thanks for pointing out those irregular tables so that I can make sure that _WD_GetTable_2() can process them correctly.  _WD_GetTable_2() was written to handle a table that I encountered that _WD_GetTable() couldn't at the time.  Of course that table didn't have row spans, column spans, and multiple <tbody> tags like the tables on the irregular table site.  Making the required changes, to be able to handle them, shouldn't be too difficult.

Edited by TheXman
Posted (edited)

I made an initial update to the script in my previous post.  Although it now pulls all of the "irregular" table data without throwing any errors, the exact placement of the spanned rows and columns are not maintained as they are when using the _HtmlTableGetWriteToArray() function.  I doubt I will continue to work on it since it really serves no purpose for me until I come across such a table in the course of my scripting.  And if I need a quick fix, _HtmlTableGetWriteToArray() is always there to save the day. :thumbsup:

 

Edited by TheXman
Posted (edited)

Using the _WD_ElementActionEx function, I can "right click" on an element to show the context menu. But how can I interact with the context menu, e.g selecting an item.

I tried to send keystrokes but did not success:

$elementInput = _WD_FindElement($sessionID, $_WD_LOCATOR_ByXPath, "(//input[contains(@name,'q')])[1]")
_WD_ElementActionEx($sessionID, $elementInput, "rightclick")

$sAction = '{"actions":[{"type": "key", "id": "default keyboard", "actions": [{"type": "keyDown", "value": "\uE015"}, {"type": "keyUp", "value": "\uE015"}, {"type": "keyDown", "value": "\uE015"}, {"type": "keyUp", "value": "\uE015"}, {"type": "keyDown", "value": "\uE015"}, {"type": "keyUp", "value": "\uE015"}, {"type": "keyDown", "value": "\uE015"}, {"type": "keyUp", "value": "\uE015"}]}]}'

; $sAction = '{"actions":[{"type": "key", "id": "default keyboard", "actions": [{"type": "keyUp", "value": "Q"}, {"type": "keyDown", "value": "Q"}]}]}' ; Send the key Q which is shortcut for the Inspect Element menu

_WD_Action($sessionID, "actions", $sAction)

Can anyone help me please?

Edited by Bot
Posted (edited)

@Danp2 No I'm referring to the context menu of the browser, not the custom context menu of the webpage, so _WD_LinkClickByText just failed with a No match error.

Spoiler

https://www.ghacks.net/wp-content/uploads/2014/05/firefox-new-context-menu.jpg

_WD_LinkClickByText($sessionID, "Inspect Element")

 

Edited by Bot
Posted (edited)

I want to access the context menu to select an installed add-on. However it seems that the key events will work with custom context menu of the webpage only and they are sent to the element, not the native context menu of the browser. So I think I will have to use standard AutoIT command instead. Anyway thanks for the help 🙂

Edited by Bot

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...