AutoIt error handling

[kalucas]

While searching the forums for info on error handling/return codes I found an old thread with this piece of code:

$APPTITLE = "WinRAR_3x"
$SYSDRIVE = EnvGet("SystemDrive")
$LOGFILE_PATH = @SystemDir & "\Installer\LOG"
$LOGFILE = $LOGFILE_PATH & "\" & $APPTITLE & ".log"
$INST_PATH = $SYSDRIVE & "\INSTALLER"
$INST_FILE = $INST_PATH & "\wrar393d.exe"
$INST_PARM = "/S"
$INST_EXEC = '"' & $INST_FILE & '"' & " " & $INST_PARM

If FileExists($INST_FILE) Then
    $RC = RunWait($INST_EXEC, $INST_PATH, @SW_HIDE)
    If $RC = "0" Or $RC = "1641" Or $RC = "3010" Then
        $EXIT_REC = $RC
        _FileWriteLog($LOGFILE, " -> Installation finished successfully, returned " & '"' & $RC & '"')
    Else
        _FileWriteLog($LOGFILE, " -> Installation failed, returned " & '"' & $RC & '"' & " Abort!")
    EndIf
Else
    _FileWriteLog($LOGFILE, " . Installation sources not found, ABORT...")
    $EXIT_RC = "1"
EndIf
Exit($EXIT_RC)

My question is whether this piece of code will actually run the install.  The $RC=RunWait...  determines the return code, but will that actually install the software?

What do you all do to get a proper return code for your installs?

Edited March 26, 2014 by kalucas

[BrewManNH]

The return from RunWait is the exitcode of the program being run, so whatever that piece of software returns is what you're going to get in the $rc variable. The exitcodes are determined by the software, not by autoit.

[kalucas]

The return from RunWait is the exitcode of the program being run, so whatever that piece of software returns is what you're going to get in the $rc variable. The exitcodes are determined by the software, not by autoit.

 

Thanks for responding.  I understand the exit codes come from the software that is being installed.  But for the piece of code above, what line actually installs the software?  Wouldn't you need an additional line such as:

RunWait($INST_EXEC, $INST_PATH, @SW_HIDE)

In my testing, I've put it just below the line:

$RC = RunWait($INST_EXEC, $INST_PATH, @SW_HIDE)

and it seems to be working, but yesterday I had an instance where the PC rebooted during the installation and the "Installation finished successfully" message was then written to the log file, even though the installation never actually completed. 

So it got me thinking that maybe this isn't the best way to handle exit codes, and that someone here might have a better idea.

[BrewManNH]

 

Thanks for responding.  I understand the exit codes come from the software that is being installed.  But for the piece of code above, what line actually installs the software?  Wouldn't you need an additional line such as:

RunWait($INST_EXEC, $INST_PATH, @SW_HIDE)

In my testing, I've put it just below the line:

$RC = RunWait($INST_EXEC, $INST_PATH, @SW_HIDE)

Why would you use 2 Run commands to install the same software? That makes no sense to me unless the software you're installing has some strange way of installing.

and it seems to be working, but yesterday I had an instance where the PC rebooted during the installation and the "Installation finished successfully" message was then written to the log file, even though the installation never actually completed. 

 

So it got me thinking that maybe this isn't the best way to handle exit codes, and that someone here might have a better idea.

As to your second problem, that's an issue where your code isn't written to correctly indicate a finished installation. It will depend on what exitcodes the program returns when it finished part of the process, but needed to reboot, and what exitcode it returns when it is completely finished, which should be different codes if it was written correctly.

[kalucas]

Makes sense.  The script must have failed the first time I ran it, which lead me to believe that the $RC = RunWait...  line was just assigning a value to $RC and not actually running the install.  I took out the additional RunWait line and the software installed.  It's all clear now.

Thanks.