Understanding $?

manishrathi has asked for the wisdom of the Perl Monks concerning the following question:

......if(defined($filelist)){
        $cmd = "$odhome\\bin\\iwodstart $iwodclient_cfg -k srcArea=$so
+urcePath -k targetArea=$targetPath -k definition=$deploymentDefinitio
+n -k workareaName=$workareaName -k filelist=$filelist -inst $$ 2>&1";
    
    }elsif(defined($workareaName)){
        $cmd = "$odhome\\bin\\iwodstart $iwodclient_cfg -k srcArea=$so
+urcePath -k targetArea=$targetPath -k definition=$deploymentDefinitio
+n -k workareaName=$workareaName -inst $$ 2>&1";
    
    }else{
        $cmd = "$odhome\\bin\\iwodstart $iwodclient_cfg -k srcArea=$so
+urcePath -k targetArea=$targetPath -k definition=$deploymentDefinitio
+n -inst $$ 2>&1";
    }
    
    
    @output = `$cmd`;
    
    print "CMD = $cmd <br>";
    
    

    #Check for the success/failure of the deployment process.
    $rc = $?;
    my $success_flag = 0;
    foreach (@output) {
        $success_flag = 1 if (/ERROR/i);
        $success_flag = 1 if (/Status: Failed/i);
    }
...
[download]

In this code, isn't $? supposed to be mentioned in $cmd ? If its not mentioned there, will it still obtain error value from the execution of $cmd at command line ? Will this $cmd return any value if deployment is successful or not ?
If there is another execution on cmd line in the code, what error value will be represented by $? ?

Thanks

Comment on Understanding $? Download Code

Replies are listed 'Best First'.
Re: Understanding $? by rjt (Curate) on Mar 07, 2013 at 01:24 UTC
For a more thorough treatment of what `$?` will contain, have a look at POSIX's macros for `WIFEXITED` and `WEXITSTATUS`. The canonical (but admittedly not always followed!) way of checking an exit status is something like: my $output = `echo hello`; # Output contains "hello" if (WIFEXITED($?)) { print("Command exited with status " . WEXITSTATUS($?)); } else { print("Command exited abnormally"); # If desired, use the other macros to figure out which # signal killed the process. } [download] Obviously in the above contrived example with `echo`, it's highly unlikely to exit abnormally, but with more complex commands, it can be very helpful to know what killed the sub-process, and what the exit status was. Of course, the `$output` variable will contain the actual standard output (`STDOUT`) of the command itself, if there was any. If you need to capture `STDERR` in the same variable as well, you can add `2>&1` to the end of your command. If you need to capture `STDOUT` and `STDERR` separately, you would need to use pipes (a la IPC::Open3 rather than the simple backticks (qx) operator, or redirect the streams to File::Temp files and read them in after the command finishes.	[reply] [d/l] [select]
Re: Understanding $? by kennethk (Abbot) on Mar 06, 2013 at 22:44 UTC
The docs: $?. `$?` is assigned a value in the background on the line @output = `$cmd`;, independent of what you do. The value itself depends on the status code returned from (I presume) iwodstart. If you have multiple commands in `$cmd` or any i/o redirects (which you do), then the child process is actually a shell around your command set, and the return value in `$?` is the returned status code from the shell. #11929 First ask yourself `How would I do this without a computer?' Then have the computer do it the same way.	[reply] [d/l] [select]