#!/usr/bin/perl

## kwic.pl --- 12.6.2004

# Copyright (C) 2004 Virach Sornlertlamvanich

## Author: Virach SORNLERTLAMVANICH <virach@tcllab.org>
## Keywords: KWIC, Thai text

# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License as
# published by the Free Software Foundation; either version 2, or (at
# your option) any later version.

# This program is distributed in the hope that it will be useful, but
# WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
# General Public License for more details.

## Commentary: KWIC for both segmented or unsegmented Thai text

## Usage: kwic.pl [-s] textfile [< keyfile]
#      Option:
#          default
#              plain input text and 'keyword;span' is acquired from
#              direct input (50 columns left-right of the keyword if
#              no span defined)
#          -s  input text is a segmented text, having a space as the 
#              word boundary
#          keyfile
#              list of 'keyword;span'

## Code:

$nonspace = '[ิีึืุู่้๊๋ั็์ฺํ]';
$nonbegin = '[ะาำฯๆ]';
$nonend = '[โเแไใ]';
$boundary = '[ ^$\t\n]';

$filename = shift(@ARGV);
if ($filename =~ /^-s$/) {
	$segment = 1;
	$filename = shift(@ARGV);
}
open(text,$filename) || die "Cannot open $filename\n";

while($str = <STDIN>){
    chop($str);
    ($key, $span) = split(/\;/,$str);

	$span = 50, if ($span==0);
	exit 0, if (length($key)==0);

	undef $foundpos;

	while (<text>) {
		chop;
		s/\t/ /g;
		$pos = $[;
		$line++;

		while (($pos = index($_, $key, $pos)) >= $[) {
			undef $lstr;
			undef $rstr;
			undef $lcol;
			undef $rcol;
			@chars = split(//, $_);

			# for segmented text (space between word)
			if ($segment) {
				if (($chars[$pos-1] !~ $boundary)||
					($chars[$pos+length($key)] !~ $boundary)){
					$pos += length($key);
					next;
				}
			}

			for ($lpos = $pos-1;
				 (($lpos >= 0)&&($lcol < $span));$lpos--) {
				$lstr = $chars[$lpos] . $lstr;

				if ($chars[$lpos] !~ $nonspace) {
					$lcol++;
				}
			}
			if ($lcol < $span) {
				while ($lcol < $span) {
					$lstr = " " . $lstr;
					$lcol++;
				}
			}

			for ($rpos = $pos + length($key);
				 (($rpos < length($_))&&($rcol < $span));$rpos++) {
				$rstr = $rstr . $chars[$rpos];

				if ($chars[$rpos] !~ $nonspace) {
					$rcol++;
				}
			}
			if ($rcol < $span) {
				while ($rcol < $span) {
					$rstr = $rstr . " ";
					$rcol++;
				}
			}

			$pos += length($key);
			$foundpos++;
			printf "[%5d][%s [%s] %s]\n", $line, $lstr, $key, $rstr;
		}
	}

	split(/\//,$filename);
	if ($foundpos>0) {
		print ">>>Found '$key' in $_[$#_] for $foundpos times.<<<\n\n";
	} else {
		print ">>>'$key' does not exist in $_[$#_].<<<\n\n";
	}

	$line=0;
	seek(text,0,0);
}