Eric TF Bat comment on Secret Perl Operators

Privacy Controls Software from ParetoLogic

The first 90% of the code accounts for the first 90% of the development time. The remaining 10% of the code accounts for the other 90% of the development time.

Tom Cargill

I am doing a startup!

Cross-browser testing from your browser

I have written my third book!

Perl one-liners explained

You're viewing a comment by Eric TF Bat and its responses.

Eric TF Bat Permalink

November 30, 2009, 22:32

I couldn't figure out the inchworm-on-a-stick either (I learned Forth in my free time during university and got a High Distinction in Discrete Mathematics 1 because it enabled me to grok how stuff like two's complement arithmetic works). But the trick is that it's not the same as $x-- or --$x or $x-1: it's much weirder than that. Take a look at the output from this:

foreach my $x ((1,2,3,-1,-2,-3,0,1.5,2.5,3.1,4.9)) { 
    printf qq/%+2.2f %+25.2f\n/, $x, ~-$x;
}

+1.00                     +0.00
+2.00                     +1.00
+3.00                     +2.00
-1.00  +18446744073709551616.00
-2.00  +18446744073709551616.00
-3.00  +18446744073709551616.00
+0.00  +18446744073709551616.00
+1.50                     +0.00
+2.50                     +1.00
+3.10                     +2.00
+4.90                     +3.00

Very odd indeed, especially when you replace the second %...f with %d, and then you get this:

+1.00 +0
+2.00 +1
+3.00 +2
-1.00 -2
-2.00 -3
-3.00 -4
+0.00 -1
+1.50 +0
+2.50 +1
+3.10 +2
+4.90 +3

What I don't know is what good it would be. I mean, if $x is non-negative, then it's the same as int($x)-1, but the weird result for negative $x confuses me...

Reply to this comment