New SciTE4AutoIt3 available with SciTE v1.77

[Jos]

11/2/2008: Uploaded a new SciTe4AutoIt3.exe installer.

Added a Win9x version of SciTE and Scintilla to the downloadpage because future releases of SciTE & Scintilla will not be maintained for Win9x compatibility.

==> ScitillaHistory page containing all SciTE/Scintilla updates.

==> Visit the SciTe4AutoIt3 Download page for the latest versions

==> Check the online documentation for an overview of all extra's you get with this installer.

Enjoy,

Jos

11/2/2008
*** Merged the updates of SciTE v 1.77 by Neil Hodgson with our own version of SciTE v 1.77. (Jos)
*** Updated Tidy.exe v2.0.23.24 (Jos)
    - Fixed: some reported report issues.
*** Updated Obfuscator.exe v1.0.24.23 (Jos)
    - internal changes and fixes.

Edited November 2, 2008 by Jos

[Emiel Wieldraaijer]

Thanks Jos for your hard work (and all others)

11/2/2007: Uploaded a new SciTe4AutoIt3.exe installer.

Added a Win9x version of SciTE and Scintilla to the downloadpage because future releases of SciTE & Scintilla will not be maintained for Win9x compatibility.

I believe you mean 2008

[Jos]

Thanks Jos for your hard work (and all others)

I believe you mean 2008

Standard glitch

[MvGulik]

Thanks for the new scite4autoit.

A little leftover 'include' issue I should have reported earlier.

(re-tested with new scite4autoit)

In the following include setup,

File1: 'main.au3', #include "include\include1_1.au3".

File2: 'include\include1_1.au3', #include "include1_2.au3".

File3: 'include\include1_2.au3'.

Generates the following Obfuscator error.

>Running Obfuscator (1.0.24.23) from:C:\CODE\AutoIt3\SciTE cmdline:

!==> *** ERROR: include file not found :#include "include1_2.au3"

Note: (available work around)

Using include ".\include1_2.au3" instead of "include1_2.au3" will be processed ok.

Test setup files.

M.v.G.

[Jos]

Thanks for the new scite4autoit.

A little leftover 'include' issue I should have reported earlier.

(re-tested with new scite4autoit)

In the following include setup,

File1: 'main.au3', #include "include\include1_1.au3".

File2: 'include\include1_1.au3', #include "include1_2.au3".

File3: 'include\include1_2.au3'.

Generates the following Obfuscator error.

>Running Obfuscator (1.0.24.23) from:C:\CODE\AutoIt3\SciTE cmdline:

!==> *** ERROR: include file not found :#include "include1_2.au3"

Note: (available work around)

Using include ".\include1_2.au3" instead of "include1_2.au3" will be processed ok.

Test setup files.

M.v.G.

Will be fixed in Obfuscator v 1.0.24.24

Jos

[Emiel Wieldraaijer]

Hi Jos,

a typo in the installers last screen finished is missing an i

Best regards,

Emiel

[Jos]

Hi Jos,

a typo in the installers last screen finished is missing an i

Best regards,

Emiel

Changed

[MvGulik]

Hi

Dropping in this Scite/au3 folding mishap in case your interested.

I'm not to sure if this might be Scite itself or only the Au3 lexer.

Issue related info included inside the code.

expandcollapse popup

Func F1()

    #cs ----
        ... (1) | active scite setting: fold.comment=1
        ... (2) | Duplicating a line inside this comment block, (depending on system, one or more duplicated)
        ... (3) | for example the last "... (E)" line.
        ... (4) | This makes the [*]comment-block end-fold marker drop out.
        ... (5) | Side effect: folding "Func F1()" will now fold whole rest text.
        ... (6) | Removing the added line restores the end-marker again.
        ... (7) | The "If ... Then" are there to fill (something.?) up.
        ... (8) | more "If" blocks result in less comment lines needed to trigger this behavior.
        ... (9) | Adding a blank line in between [*] and "If ... _" block seems
        ... (A) | to compleetly remove/block this behavior. (unable to trigger it anymore)
        ... (-) | *(sc177.exe = same behavior)
        ... (-) | *(adding blank line (almost anywere between top and F1()=unfolded) seems to do the trick to)
        ... (D) | *(tested on one system, needed to add more lines on other system.)
        ... (D) | *(using "if ... then/endif" instead of "If ... _" block, unable to trigger it anymore)
        ... (D)
        ... (D)
        ... (D)
        ... (D)
        ... (E)
    #ce ----

    If 1 Then ;; 1
        $var = 1
    If 1 Then ;; 2
        $var = 1
    If 1 Then ;; 3
        $var = 1
    If 1 Then ;; 4
        $var = 1
    If 1 Then ;; 5
        $var = 1
    If 1 Then ;; 6
        $var = 1
        ;; ...
        ;; ...
        ;; ... [*](comment block end-fold disapears)
        If 1 And _
                1 And _
                1 Then _
                $var = 1
    EndIf ;; 6
    EndIf ;; 5
    EndIf ;; 4
    EndIf ;; 3
    EndIf ;; 2
    EndIf ;; 1
EndFunc


Func F2()
    ;;
EndFunc

Func F3()
    ;;
EndFunc

Func F4()
    ;;
EndFunc

Edited November 14, 2008 by T.52857170

[KaFu]

Hello Jos,

#Obfuscator_Off                 ; Stop the Obfuscation process below this line
#include <SQLite.dll.au3>       ; SQLite.dll.au3 can not be obfuscated
#Obfuscator_On                  ; Start the Obfuscation process below this line

Worked fine for me quiet a while. Sadly with Obfuscator 1.0.24.23 this include cripples my script again. It seems to me as the Off / On Switch doesnt work correctly (?).

The switches I use for compilation are

/sf /sv /sci 9

and what i found strange was to see the following lines in the log:

0.06 ==> Start Add_Include: Include_Rec$:#include <SQLite.dll.au3> ; SQLite.dll.au3 can not be obfuscated

0.08 ==> Loading Include file into source Array:c:\program files\autoit3\include\sqlite.dll.au3

1.64 ### Listing included files ###

1.64 1:c:\program files\autoit3\include\sqlite.dll.au3

Is this because of the switches to strip functions and variables?

Anyhow, when i comment the include out, the obfuscation works fine, otherwise I get the "-1 cannot redeclare const" message I already knew from before I used #Obfuscator_Off for this include.

Best Regards and thanks for your great work!

[Jos]

Hi

Dropping in this Scite/au3 folding mishap in case your interested.

I'm not to sure if this might be Scite itself or only the Au3 lexer.

Issue related info included inside the code.

expandcollapse popup

Func F1()

    #cs ----
        ... (1) | active scite setting: fold.comment=1
        ... (2) | Duplicating a line inside this comment block, (depending on system, one or more duplicated)
        ... (3) | for example the last "... (E)" line.
        ... (4) | This makes the [*]comment-block end-fold marker drop out.
        ... (5) | Side effect: folding "Func F1()" will now fold whole rest text.
        ... (6) | Removing the added line restores the end-marker again.
        ... (7) | The "If ... Then" are there to fill (something.?) up.
        ... (8) | more "If" blocks result in less comment lines needed to trigger this behavior.
        ... (9) | Adding a blank line in between [*] and "If ... _" block seems
        ... (A) | to compleetly remove/block this behavior. (unable to trigger it anymore)
        ... (-) | *(sc177.exe = same behavior)
        ... (-) | *(adding blank line (almost anywere between top and F1()=unfolded) seems to do the trick to)
        ... (D) | *(tested on one system, needed to add more lines on other system.)
        ... (D) | *(using "if ... then/endif" instead of "If ... _" block, unable to trigger it anymore)
        ... (D)
        ... (D)
        ... (D)
        ... (D)
        ... (E)
    #ce ----

    If 1 Then ;; 1
        $var = 1
    If 1 Then ;; 2
        $var = 1
    If 1 Then ;; 3
        $var = 1
    If 1 Then ;; 4
        $var = 1
    If 1 Then ;; 5
        $var = 1
    If 1 Then ;; 6
        $var = 1
        ;; ...
        ;; ...
        ;; ... [*](comment block end-fold disapears)
        If 1 And _
                1 And _
                1 Then _
                $var = 1
    EndIf ;; 6
    EndIf ;; 5
    EndIf ;; 4
    EndIf ;; 3
    EndIf ;; 2
    EndIf ;; 1
EndFunc


Func F2()
    ;;
EndFunc

Func F3()
    ;;
EndFunc

Func F4()
    ;;
EndFunc

Will have to do some debugging for the "Single Commentline block" to see were this go wrong.

I will put it on my todo list.

Jos

[Jos]

Hello Jos,

#Obfuscator_Off                 ; Stop the Obfuscation process below this line
#include <SQLite.dll.au3>       ; SQLite.dll.au3 can not be obfuscated
#Obfuscator_On                  ; Start the Obfuscation process below this line

Worked fine for me quiet a while. Sadly with Obfuscator 1.0.24.23 this include cripples my script again. It seems to me as the Off / On Switch doesnt work correctly (?).

The switches I use for compilation are

/sf /sv /sci 9

and what i found strange was to see the following lines in the log:

0.06 ==> Start Add_Include: Include_Rec$:#include <SQLite.dll.au3> ; SQLite.dll.au3 can not be obfuscated

0.08 ==> Loading Include file into source Array:c:\program files\autoit3\include\sqlite.dll.au3

1.64 ### Listing included files ###

1.64 1:c:\program files\autoit3\include\sqlite.dll.au3

Is this because of the switches to strip functions and variables?

Anyhow, when i comment the include out, the obfuscation works fine, otherwise I get the "-1 cannot redeclare const" message I already knew from before I used #Obfuscator_Off for this include.

Best Regards and thanks for your great work!

I will have a look but could you PM me a small test script that I can test with with both versions of Obfuscator (just to save me some time having to creeate it myself ) ?

Thanks

Jos

[Jos]

Will have to do some debugging for the "Single Commentline block" to see were this go wrong.

I will put it on my todo list.

Jos

Think I found the issue with the Lexer. Could you give this version a spin and let me know if that also solves it for you: http://www.autoitscript.com/autoit3/scite/...t3/SciLexer.dll

Jos

[MvGulik]

Just added a second case to the initial report before seeing your fix reply.

Figured I had some time to explore some more possibilities.

My test result's:

Case-1:

- Can't reproduce it anymore. (original source file also behaves normal now)

- Did see a other more-minor mishap while testing that I also have seen before.

(trailing functions showed up without having the blank line between them anymore,

but ... can't reproduce it at the moment.)

Case-2:

- still reproducible, just needed to add some extra top-comment lines.

Seems to be the same problem, but with block comments.

Partial Case-2

If 1 Then ;; 12
        $var = 1
        #cs ----
            ... (X)
            ... (X)
            ... (X)
        #ce ----
        Local $var = 1, _
                $var = 2
    EndIf ;; 12

[MvGulik]

The additional folding side effect I noted, hidden blank lines and hidden text,

seems a general scite/scintilla artifact. (reported to scintilla bug-tracker.)

[KaFu]

I will have a look but could you PM me a small test script that I can test with with both versions of Obfuscator (just to save me some time having to creeate it myself ) ?

Hi Jos,

I took a deeper look and now can say it's not because of the #include <SQLite.dll.au3>... the error more likley seems to result from some very bad coding practices on my side *doooh*...

I got the obfuscator to obfuscate my script to the end, though it still crashes >> after finishing obfuscating (?). If i compile the obfuscated script it runs as inteded with the exception of one ceratin function. Where the unobfuscated script runs smoothly the obfuscated script exists with an "can not redeclare a const at line -1".

As I said i fear it happens because of my bad coding practice... would you still take a look at the script?

Regards

KaFu

[Jos]

Case-2:

- still reproducible, just needed to add some extra top-comment lines.

Seems to be the same problem, but with block comments.

Partial Case-2

If 1 Then ;; 12
        $var = 1
        #cs ----
            ... (X)
            ... (X)
            ... (X)
        #ce ----
        Local $var = 1, _
                $var = 2
    EndIf ;; 12

Just uploaded a new version that fixes this issue. This issue only appeared when the line after a commentblock contains a Continuation character at the end of the line. The issue wasn't in the folding logic this time but in the lexer part that identified the keywords. It would change #ce to Directive in stead of part of the comment block.

The additional folding side effect I noted, hidden blank lines and hidden text,

seems a general scite/scintilla artifact. (reported to scintilla bug-tracker.)

Had a look at your bug report and can reproduce your issue. Will have a look myself too but am not sure if I can fix this.

Jos

[Jos]

Hi Jos,

I took a deeper look and now can say it's not because of the #include <SQLite.dll.au3>... the error more likley seems to result from some very bad coding practices on my side *doooh*...

I got the obfuscator to obfuscate my script to the end, though it still crashes >> after finishing obfuscating (?). If i compile the obfuscated script it runs as inteded with the exception of one ceratin function. Where the unobfuscated script runs smoothly the obfuscated script exists with an "can not redeclare a const at line -1".

As I said i fear it happens because of my bad coding practice... would you still take a look at the script?

Regards

KaFu

I can have a look at it when Obfuscator is doing things it shouldn't do.

I will need a script that I can run and simulate the issue with

[MvGulik]

For good measure, Case-2: also none reproducible anymore over here.

Many thanks for your support and work.

Success with that other fold-mishap. (let me know, when you like me to test-use a fix)

M.v.G.

[MvGulik]

Some Tidy directive items, after exploring and playing around with them.

expandcollapse popup

#cs
    1) "/updatevars x"
    +) value 1 and 2 have switched behavior compared to documentation(Tidy_Example.ini & Scite4AutoitHelp)
    ~) active behavior: 1 makes Uppercase, 2 makes Lowercase. (directive,tidy.ini)
    
    2) "/tabchar x" or "/tc x"
    +) not having any effect when used as directive.
    
    4) "/backupdir directory"
    +) not having any effect when used as directive. (alt option /bdir is working)
    
    5) "/q"
    ~) not having any effect when used as directive.
    ~) if its intended as a cmd-line only option. (Scite4Autoit help kind a suggests directive use for this option to)
#ce

;~ #Tidy_Parameters=/updatevars 1 ;; (Uppercase)
;~ #Tidy_Parameters=/updatevars 2 ;; (Lowercase)
;~ #Tidy_Parameters=/tabchar 1
;~ #Tidy_Parameters=/bdir backup2
;~ #Tidy_Parameters=/backupdir backup2
;~ #Tidy_Parameters=/q

F1()

Func F1()
    #cs ----
        ...
    #ce ----
    Local $var = 0
    If 1 Then
        $var = 1
        If 1 Then
            $var = 1
        EndIf
    EndIf
EndFunc   ;==>F1

Ik hou je wel erg bezig.

M.v.G.

ps: Niel seems to be back.(scintilla)

Edited November 21, 2008 by T.52857170

[Jos]

Ik hou je wel erg bezig.

M.v.G.

ps: Niel seems to be back.(scintilla)

Geen probleem, het is goed te zien dat er met Tidy gewerkt word al lijkt het soms dat jij de enige bent die alle opties gebruikt.

Your reported issues should be fixed in the current version 2.0.23.25 in the Beta directory.

;~ #Tidy_Parameters=/updatevars 1 ;; (Uppercase)

;~ #Tidy_Parameters=/updatevars 2 ;; (Lowercase)

Fixed the documentation.

;~ #Tidy_Parameters=/tabchar 1

Fixed this bug in the program.

;~ #Tidy_Parameters=/bdir backup2

;~ #Tidy_Parameters=/backupdir backup2

fixed this bug in the program

;~ #Tidy_Parameters=/q

Updated documentation as this was implemented as a commandline option only.