bfilter/trunk/bfilter.pl

#!/usr/bin/perl -w
#

use strict;
use locale;

# maximum entries
my $max = 0;
# minimum letters to search by
my $min_len = 3;
# if more than x elements, warn to increase min_len
my $increase_at = 500;

# name of generated index
my $headlines = 'headlines';

my $debug = 1;

sub print_file {
        my $f = shift || return;
        open(F, $f) || die "$f: $!";
        while(<F>) {
                print;
        }
        close(F);
}

print qq{
var $headlines = new Object();
};

my @part_arr;
my $last_part = '';
my $total = 0;

my $max_elements = 0;

while(<STDIN>) {
        chomp;

        if (!m/\t/ || m/\t$/) {
                print STDERR "SKIP '$_': no tab\n";
                next;
        }

        my ($path,$headline) = split(/\t+/,$_,2);

        if (length($headline) < $min_len) {
                print STDERR "SKIP '$_': too short\n";
                next;
        }


        # split into min_len part and rest
        my ($part,$rest) = ( substr($headline,0,$min_len), substr($headline,$min_len) );

        # escape special chars
        $part =~ s/(['\\])/\\$1/g && print STDERR "ESCAPED part '$part'\n";
        $rest =~ s/(['\\])/\\$1/g && print STDERR "ESCAPED rest '$rest'\n";
        $headline =~ s/(['\\])/\\$1/g;

        # make part lowercase
        $part = lc($part);

        $last_part = $part if (! $last_part);

        # new part?
        if ($part ne $last_part) {
                print STDERR $last_part,"\t",$#part_arr+1,"\n" if ($debug && $#part_arr > $increase_at);
                $max_elements = $#part_arr if ($#part_arr > $max_elements);
                print "${headlines}['$last_part'] = [\n ",join(",\n ",@part_arr),"];\n" if (@part_arr);
                $total += $#part_arr;
                @part_arr = ();
                $last_part = $part;
        }
        push @part_arr, "['$path','$headline']";

        # break out?
        last if ($max && $total > $max);
}

print "${headlines}['$last_part'] = [\n ",join(",\n ",@part_arr),"];\n" if (@part_arr);
print qq{

${headlines}.min_len = $min_len;
${headlines}.length = $total;

};

print STDERR "You have more than $increase_at elements, so you should\nincrease min_len to ",$min_len+1," or higher for performance benefit.\n" if ($max_elements > $increase_at);
1	dpavlin	1	#!/usr/bin/perl -w
2			#
3
4			use strict;
5			use locale;
6
7			# maximum entries
8			my $max = 0;
9			# minimum letters to search by
10	dpavlin	9	my $min_len = 3;
11	dpavlin	4	# if more than x elements, warn to increase min_len
12			my $increase_at = 500;
13	dpavlin	1
14	dpavlin	10	# name of generated index
15			my $headlines = 'headlines';
16
17	dpavlin	3	my $debug = 1;
18
19	dpavlin	1	sub print_file {
20			my $f = shift \|\| return;
21			open(F, $f) \|\| die "$f: $!";
22			while(<F>) {
23			print;
24			}
25			close(F);
26			}
27
28			print qq{
29	dpavlin	10	var $headlines = new Object();
30	dpavlin	1	};
31
32			my @part_arr;
33			my $last_part = '';
34			my $total = 0;
35
36	dpavlin	4	my $max_elements = 0;
37
38	dpavlin	1	while(<STDIN>) {
39			chomp;
40
41	dpavlin	9	if (!m/\t/ \|\| m/\t$/) {
42			print STDERR "SKIP '$_': no tab\n";
43			next;
44			}
45
46	dpavlin	7	my ($path,$headline) = split(/\t+/,$_,2);
47
48	dpavlin	9	if (length($headline) < $min_len) {
49			print STDERR "SKIP '$_': too short\n";
50			next;
51			}
52	dpavlin	1
53	dpavlin	9
54	dpavlin	1	# split into min_len part and rest
55	dpavlin	7	my ($part,$rest) = ( substr($headline,0,$min_len), substr($headline,$min_len) );
56	dpavlin	1
57	dpavlin	9	# escape special chars
58			$part =~ s/(['\\])/\\$1/g && print STDERR "ESCAPED part '$part'\n";
59			$rest =~ s/(['\\])/\\$1/g && print STDERR "ESCAPED rest '$rest'\n";
60			$headline =~ s/(['\\])/\\$1/g;
61
62	dpavlin	1	# make part lowercase
63			$part = lc($part);
64
65			$last_part = $part if (! $last_part);
66
67			# new part?
68			if ($part ne $last_part) {
69	dpavlin	7	print STDERR $last_part,"\t",$#part_arr+1,"\n" if ($debug && $#part_arr > $increase_at);
70	dpavlin	4	$max_elements = $#part_arr if ($#part_arr > $max_elements);
71	dpavlin	10	print "${headlines}['$last_part'] = [\n ",join(",\n ",@part_arr),"];\n" if (@part_arr);
72	dpavlin	1	$total += $#part_arr;
73			@part_arr = ();
74			$last_part = $part;
75			}
76	dpavlin	7	push @part_arr, "['$path','$headline']";
77	dpavlin	1
78			# break out?
79			last if ($max && $total > $max);
80			}
81
82	dpavlin	10	print "${headlines}['$last_part'] = [\n ",join(",\n ",@part_arr),"];\n" if (@part_arr);
83	dpavlin	7	print qq{
84	dpavlin	4
85	dpavlin	10	${headlines}.min_len = $min_len;
86			${headlines}.length = $total;
87	dpavlin	7
88			};
89
90			print STDERR "You have more than $increase_at elements, so you should\nincrease min_len to ",$min_len+1," or higher for performance benefit.\n" if ($max_elements > $increase_at);