http://qs321.pair.com?node_id=148757

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

Good day, monks.. I'm working on a small parser and in this script I require a function that would be able to return a special 'escape' (excuse me for poor jargon) character (such as'\U' etc) based on the case of a string passed to the sub as the first parameter. So, if my string is in uppercase, the subroutine should return '\U', it should return '\L' otherwise. For example, 
my $esc_char = get_escape_case("S");

[download]
would return '\U' escape character. I may then use it (or so I hope to) to format any other string (either uppercase or lowercase it) by simply prepending the escape character in an expression like this: 
my $case_formatted_string = $esc_char ."FoobAr";

[download]
In this example (where $esc_char is '\U') $case_formatted_string should become 'FOOBAR' (since "S" is in uppercase).

The only problem that I'm facing now is how do I actually apply this escape character to a string? I've tried one obvious way in a sample code below but it didn't work: 
use strict;

# determine case of a character
# and return appropriate 'control' code.
sub get_escape_case
{
    my ($char) = @_;
    if ($char eq "\U$char") {
       return '\U'; 
    } else {
       return '\L';
    }
}

# this wouldn't work...
print get_escape_case("S")."stuff\n";

# while this works..
print "\Ustuff\n";

[download]
I guess I might have to use a special 'code' for the escape character? Could anyone lead me in the right path? Thank you. ;-)

"There is no system but GNU, and Linux is one of its kernels." -- Confession of Faith

Replies are listed 'Best First'.
Re: case escape characters
by I0 (Priest) on Mar 02, 2002 at 05:13 UTC
    print &{sub{local($c,$_)=@_;$c x=length;$c^lc$c^lc}}("S","stuff");
[reply]
[d/l]
      Nice! Very nice! I had to look twice (hell, five times) at x=length. I am still trying to understand $c^lc$c^lc: 
      my $c = 'a';
my $d = 'A';
$_ = 'stuff';

print $c^lc$c^lc;
print $d^lc$d^lc;

      [download]
      More importantly, i am trying to understand how one would even COME UP with that! :D

      jeffa most humbly bows

[reply]
[d/l]
[select]
Re: case escape characters
by synapse0 (Pilgrim) on Mar 02, 2002 at 03:40 UTC
    Wow.. it took me a bit to decipher what you are trying to do. I can say pretty much off the bat that you are going to want to do some overhauling on how you're thinking this snippet through. You're trying to replace a character with something new (in this case you're adding an escaped character).. how do we perl-ites usually do that? uh-huh.. you guessed it s/// (or tr///, but that's not as appropriate here). You'll prolly want to throw the entire string at your sub and return it's modified form.. something like 
    $string = adjust_case('s', "stuff\n");
print $string; # which would print Stuff with a newline

    [download]

    Prolly not exactly what you're lookin for, but hopefully will allow you to see your objective a little differently.

    -SynZero

    Sleep Deprived
[reply]
[d/l]
      Thanks for help ;-)

      I've actually made it work by taking a slightly different approach. The rewrote the to_case() sub to return a reference to appropriate case subroutine (lc(), or uc() ...) 
      use strict;

sub to_case
{
    my ($char) = @_;

    if ($char eq "\U$char") {
        return sub {uc(shift)};
    } else {
        return sub {lc(shift)};
    }
}

print to_case("FOO")->("bAr") ."\n";
print to_case("foo")->("BaR") ."\n";

      [download]
      Here's the output produced by the script: 
      BAR
bar

      [download]
      Sometimes, asking questions helps to come come up with an answer (eventually ;-). I'd appreciate it if you could show me another approach to solving my 'problem'. I'm not sure if returning sub references is a good thing?

      Of course, I realize that my 'solution' presented here is way worse than what you've suggested since I might as well just pass that string which I want to modify (it's case) to the sub and simply return it's modified version in the end. However, in my script I might need to format a number of separate strings to matching case, and most of those strings may not be 'known' at the time i first invoke the to_case() sub (say, i may call it once and store sub reference for later use...).

      "There is no system but GNU, and Linux is one of its kernels." -- Confession of Faith
[reply]
[d/l]
[select]

Back to Seekers of Perl Wisdom