#!/usr/bin/perl
#
#The Arbitrary Text Code 2.0a
#by Leonard Richardson leonardr@ucla.edu
#
#Execute with: tatc [regexp] [file]
#
#TATC will read in [file] and attempt to find a string matching
#[regexp] with the characters some number of spaces apart. It does
#this by inserting a ".{x}" between each character of your regexp, for
#a large range of x, so things like * won't make sense. I think you
#should stick to plain strings.
#
#Bugs: 
#* The program will currently find at most one match for each skip. 
#  Eventually I'll fix it to find all matches for a given skip.
#
#* Negative skip is not supported. It will eventually be supported by
#  simply reversing the string and finding an encoding of that with
#  positive skip.

($string,$path) = @ARGV;

#Set this to 0 if you don't want status messages.
$status = 1;

#Grab the text into a string.
open TEXT, $path or die "Can't open file $path.";
while (<TEXT>) {$text .= $_;}
$len = length($text);

for ($i = 0; $i <= 1000; $i++)
{
    #If it's a bunch of text, print out status mesasges every ten skips.
    if ($len > 100000 && ($i % 10) == 0 && $status)
    {
	print "$i...\n";
    }
    $base_regexp = ".{$i}";
    $regexp = "";
    for ($j = 0; $j < length($string); $j++)
    {
	$regexp .= substr($string,$j,1) . $base_regexp; 
    }

    if ($text =~ /$regexp/i)
    {
	$matched_text = $&;
	print "Found with skip $i:\n";

	for ($line = 0; $line < length($string); $line++)
	{
	    $linelen = min($i+1,77);
	    if ($i+1 > 77) { $suffix = "...";}
	    print substr($matched_text,$line*($i+1),$linelen);
	    print $suffix;
	    print "\n";	   
	}
    }
}

sub min{($a,$b)=@_;if($a<$b){$a;}else{$b;}}