in reply to returning to a loop later on

{1} is redundant in a regex - always.

Your indentation is stuffed. It looks like the first if is inside the first while loop - it isn't.

redo FILENAME doesn't have a matching loop.

You could structure your code something like:

while (1) {
    while ($out !~ m/^[sSfF]$/) {
        ...
    }
    
    if ($out =~ m/^[sS]$/) {
        print $output;
        next;
    }
    
    last;
}

exit if $out !~ m/^[fF]$/;

while (1) {
    while ($save !~ (m/^[a-zA-Z]\w*$/)) {
        ...
    }
    
    if (-e $save) {
        ...
    }
    
    if ($overwrite =~ m/^[nN]$/) {
        next;
    }
    
    ...
    last;
}

[download]
DWIM is Perl's answer to Gödel

Replies are listed 'Best First'.
Re^2: returning to a loop later on
by mreece (Friar) on Sep 07, 2006 at 03:40 UTC
    redo FILENAME doesn't have a matching loop.
    it doesn't have to be a loop, you can label blocks, too! 
    my $i = 0;
FOO: {
    print "$i\n";
    redo FOO unless ++$i > 4;
}

    [download]
[reply]
[d/l]

      Indeed, but you can't redo to a label that is not in labling the current scope:

      FOO: { # scope 1
        { # scope 2
        redo FOO; # Ok
    }

    redo FOO; # Ok
}

    { # scope 3
    redo FOO; # Illegal
}

      [download]

      Prints:

      DWIM is Perl's answer to Gödel
[reply]
[d/l]
[select]
        you can't do that with a loop, either, right? perhaps i am misunderstanding you, but that is not the situation with the OP's code, imperfect as it may be; the redo FILENAME is within the FILENAME: { }
[reply]
[d/l]
[select]

In Section Seekers of Perl Wisdom