AutoIt v3.3.15.5 Beta

[mLipok]

The _PathSplit() issue seams to be fixed on Au3.3.16.0 RC2

[argumentum]

 

MsgBox(262144, "@OSVersion", '@OSVersion = "' & @OSVersion & '"' & _ 
                            @CR & 'WinVer: ' & FileGetVersion("WinVer.exe")  & _
                            @CR & 'ProductName: ' & RegRead('HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion', 'ProductName') & _
                            @CR & 'CurrentVersion: ' & RegRead('HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion', 'CurrentVersion') & _
                            @CR & 'ReleaseId: ' & RegRead('HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion', 'ReleaseId') & _
                            "", 60)

 

Edit4: fixed on RC3 ( see bellow )

Edited March 6, 2022 by argumentum
fixed

[Trong]

RC2 TEST On Windows 11: @OSVersion = UNKNOWN

https://i.postimg.cc/G2z0s3Tc/rc2.png

[Jon]

Spot the error!

Goddamn C.

[Jon]

Release Candidate 3

Win11 OSVersion nonsense fixed??

Installer: https://www.autoitscript.com/autoit3/files/beta/autoit/autoit-v3.3.16.0-setup-rc3.exe

Zip: https://www.autoitscript.com/autoit3/files/beta/autoit/autoit-v3.3.16.0-rc3.zip

[Trong]

 

Error Win11 @OsVersion fixed OK. 😁

[argumentum]

Yes, the Win11 shows as should in 22000 and 21996. Also, Server 2022 verified ok.

Edited March 6, 2022 by argumentum

[HurleyShanabarger]

Great, that the new version ist out. Thank to all involved for the ongoing work on AutoIt.

On 3/5/2022 at 4:00 PM, Jon said:

Edit: I've noticed over the last few years, and at work, it's almost impossible in modern browsers to download .exe files. So I think I'm going to wrap the installer in a .zip as well. I'll probably have two installer versions up though so it doesn't break any updater type scripts.

If you find time, can you adjust the update.dat or upload the exe (https://www.autoitscript.com/cgi-bin/getfile.pl?autoit3/autoit-v3.3.16.0-setup.exe) as well?

 

//Edit: Working now, perfect!

Edited March 6, 2022 by HurleyShanabarger

[UEZ]

There is a bug in the GDIPlus.au3 because I wondered why a function suddenly doesn't work properly:

Old:

Func _GDIPlus_GraphicsSetInterpolationMode($hGraphics, $iInterpolationMode)
    Local $aResult = DllCall($__g_hGDIPDll, "int", "GdipSetInterpolationMode", "handle", $hGraphics, "int", $iInterpolationMode)
    If @error Then Return SetError(@error, @extended, False)
    If $aResult[0] Then Return SetError(10, $aResult[0], False)

    Return True
EndFunc   ;==>_GDIPlus_GraphicsSetInterpolationMode

New:

Func _GDIPlus_GraphicsSetInterpolationMode($hGraphics, $iInterpolationMode)
    Local $aCall = DllCall($__g_hGDIPDll, "int", "GdipSetInterpolationMode", "handle", $hGraphics, "int", $iInterpolationMode + 1)
    If @error Then Return SetError(@error, @extended, False)
    If $aCall[0] Then Return SetError(10, $aCall[0], False)

    Return True
EndFunc   ;==>_GDIPlus_GraphicsSetInterpolationMode

Why " + 1"?

 

Btw, there a some more functions with added " + 1"!?!?!:

_GDIPlus_GraphicsSetCompositingMode, _GDIPlus_GraphicsSetCompositingQuality, _GDIPlus_GraphicsSetCompositingQuality

Edited March 6, 2022 by UEZ

[mLipok]

It was changed toogether with: https://www.autoitscript.com/trac/autoit/ticket/3835
In SVN revision 12580

@UEZ please check also: 

_GDIPlus_GraphicsSetCompositingMode
_GDIPlus_GraphicsSetCompositingQuality
_GDIPlus_GraphicsSetInterpolationMode
_GDIPlus_GraphicsSetPixelOffsetMode
 

and:
_GDIPlus_GraphicsGetCompositingMode
_GDIPlus_GraphicsGetCompositingQuality
_GDIPlus_GraphicsGetInterpolationMode

there is:

Return $aCall[2] - 1

 

[UEZ]

Confirmed, also functions

_GDIPlus_GraphicsGetCompositingMode
_GDIPlus_GraphicsGetCompositingQuality
_GDIPlus_GraphicsGetInterpolationMode

wrongly changed to " - 1"!

[mLipok]

16 minutes ago, mLipok said:

_GDIPlus_GraphicsSetCompositingMode
_GDIPlus_GraphicsSetCompositingQuality
_GDIPlus_GraphicsSetInterpolationMode
_GDIPlus_GraphicsSetPixelOffsetMode

and what about all this ones ?
Could you test it?

Edited March 6, 2022 by mLipok

[UEZ]

1 minute ago, mLipok said:

and what about all this ones ?
Could you test it?

I edited my previous post after I recognized also that the other functions were buggy.

 

Edited March 6, 2022 by UEZ

[mLipok]

5 minutes ago, UEZ said:

I edited my previous post after I recognized also that the other functions were buggy.

I was testing things and was not aware about your edits

 

[pixelsearch]

Hi @Jon

Bravo for this great job you did 
In case a new amended release 3.3.16.1 should appear (because of the crash when ArrayDisplay faces an element with a NULL value as described in this link), do you think the following could be considered too :

Help file, topic Number :
When you add some spaces at the beginning of strings containing digits, the content of the string is still correctly evaluated, for example :

Number("  3.14") ; Returns 3.14
Number("  24/7") ; Returns 24

The only one which isn't correctly evaluated is the hexadecimal case :

Number("  0xcade") ; Returns 0
Number("0xcade") ; Returns 51934

Should it be amended to return 51934 in both cases ?
Thanks
 

[Jon]

Not sure, from my point of view the first 3 should return 0 but for some reason we were very friendly in converting misformatted text to valid numbers. If "    27/7" works then I guess "    0xcade" should.

[pixelsearch]

22 minutes ago, Jon said:

we were very friendly ... then I guess "    0xcade" should.

Thanks for being friendly
I tried them all like this :

Local $dNumber1 = Number(1 + 2 + 10) ; Returns 13.
Local $dNumber2 = Number("  3.14") ; Returns 3.14.
Local $dNumber3 = Number("  24/7") ; Returns 24.
Local $dNumber4 = Number("  tmp3") ; Returns 0 as this is a string.
Local $dNumber5 = Number("  1,000,000") ; Returns 1 as it strips everything after (and including) the first comma.
Local $dNumber6 = Number("  24autoit") ; Returns 24
Local $dNumber7 = Number("  1.2e3sa") ; Returns 1200
Local $dNumber8 = Number("  0xcade") ; Returns 0 <=================
Local $dNumber9 = Number(Ptr(5)) ; Returns 5
Local $dNumber10 = Number(Binary("abc")) ; Returns decimal value of 0x616263 = 6513249
Local $dNumber11 = Number(ObjCreate("Scripting.Dictionary")) ; Returns 0
Local $dNumber12 = Number(1 > 3 Or 5 <= 15) ; Returns 1
Local $dNumber13 = Number("  -30") ; Returns 30 (minor : should be -30 in help file) <====

Depending on what you'll decide, I'll finally get rid of these lines (or not) in a personal rework :

If $iType = 1 And StringLeft(StringStripWS($aArray[0], 1), 2) = "0x" Then ; $STR_STRIPLEADING = 1
    For $i = 0 To $_g_ArrayDisplay_nRows - 1
        $sData &= Number(StringStripWS($aArray[$i], 1)) & $sDelim ; $sDelim also at end of string (used by dll +++)
    Next
Else
    ...

 

Edited March 9, 2022 by pixelsearch

[pixelsearch]

@Jon: thanks for the fix in 3.3.16.1 . Now Number() ignores eventual leading spaces in strings containing hexa :

MsgBox(0, @AutoItversion, "Number('   0xcade') = " & Number('   0xcade'))

In fact, now it works like C++ . If you look at the C++ example concerning stoi in this link and you add leading spaces to any of the 4 strings, then C++ result ignores any leading space when converting :

std::string str_dec  = "   2001, A Space Odyssey";
std::string str_hex  = "   40c3";
std::string str_bin  = "   -10010110001";
std::string str_auto = "   0x7f";

After compiling, stoi of "   40c3" displays correctly 16579... just like AutoIt 3.3.16.1
Well done, Jon 

Edit: I just tested : you already fixed this in AutoIt stable version 3.3.16.0 , damn you're fast !

Edited March 24, 2022 by pixelsearch

[pixelsearch]

Here is something that looks a bit hard to solve (let's hope I'm wrong)

#include <Debug.au3>

Local $a = [1, 2, 3]
Local $b = ["x", "y", $a, "z"]

_DebugArrayDisplay($b, "$b : " & @AutoItVersion, Default, Default, Default, _
    "Col 0", Default, _UserFunc)

Func _UserFunc($aArray, $aSelected) ; as found in help file
    _DebugArrayDisplay($aSelected, "Selected rows : " & @AutoItVersion, "1:")
EndFunc    ;==>_UserFunc

* Calling ArrayDisplay recursively worked fine with AutoIt 3.3.14.5 because there was no subclassing, no global sort indexes stored in arrays etc... but the display was slow when it came to big arrays.

* Looking at the pics above, calling ArrayDisplay recursively isn't really recommended with AutoIt 3.3.16.0+ because of the WM_NOTIFY message handler held through subclassing (?) maybe also because of the global indexes stored in arrays etc... but the advantage of 3.3.16.0+ is an immediate display (no matter the rows number) and a quick column sort.

See how the row values disappeared in the 2nd pic (in 3.3.16.1) . Now, after you close the "Selected rows" pic, then click on Col0 to Sort (hoping to make values reappear ?), it's the main GUI that will disappear too, leaving you with nothing. You'll have to close AutoIt.exe (which is now slowly heating CPU) by yourself, through the Task Manager.

It made me remember of a sentence from Melba23, I just googled on "autoit melba23 plague subclassing" to retrieve the post where he shared his contagious thoughts

Edited April 1, 2022 by pixelsearch
Func _UserFunc as found in help file (the recursive call)

[jpm]

Silly question,

where/how did  you pickup 3.3.16.1 ?

From the dowwnload area I just pickup  3.3.16.0