jquery/no_pager/index.cgi

#!/usr/bin/perl -w

use strict;

use CGI::Simple;
use CGI::Carp qw(fatalsToBrowser warningsToBrowser);
use Search::Estraier;
use YAML::Syck;
use JSON::Syck;
use Data::Dump qw/dump/;

my $q = new CGI::Simple;
print qq{Content-type: text/html\n\r\n\r};

my $config = LoadFile('config.yml');

my $v = {
        search => '',
        hits => 0,
        page => 0,
        max_page => 0,
        time => '',
        id => time() . rand(99),
};

my $json;

#warn "config = ", dump($config);

sub json {
        return
        '<textarea id="json" style="display:none">' .
        $q->escapeHTML( JSON::Syck::Dump( $v ) ) .
        '</textarea>';
}

sub get_results {
        my ($search, $page) = @_;

        warn "get_results( $search , $page )\n";

        if (! $search || $search =~ m/^\s*$/) {
                $v->{status} = 'Enter search query';
                return('', json);
        }

        if (! $page) {
                $v->{status} = 'Error: no page number?';
                return('', json);
        }

        $search = join(" AND ", split(/\s+/, $search)) unless ($search =~ m/(?:AND|OR|\[|\])/);
        $v->{search} = $search || '';

        $v->{page} = $page;

        my $node = new Search::Estraier::Node(%{ $config->{estraier} });

        my $on_page = 30;
        my $skip = ( $page - 1 ) * $on_page;

        my $cond = new Search::Estraier::Condition;
        $cond->set_phrase( $search );
        $cond->set_max( $on_page );
        $cond->set_skip( $skip );

        my $nres = $node->search($cond, ( $config->{estraier}->{depth} || 0 ) );

        my $out;

        if (defined($nres)) {
                $v->{hits} = $nres->hits;
                $v->{time} = $nres->hint('TIME');
                $v->{max_page} = int( ($nres->hits + $on_page - 1) / $on_page );

                $v->{status} = qq{
                        Got <b>$v->{hits}</b> results for <tt>$v->{search}</tt>
                        in <em>$v->{time} s</em>
                };

                sub html_snippet {
                        my $text = shift || return;
                        my $out = '';
                        foreach my $s (split(/[\n\r]{2}/, $text)) {
                                $out .= ' ... ' if ($out);
                                my ($pre,$hit,$post) = split(/\n/,$s,3);
                                $hit =~ s/\t.*$//;
                                $out .= 
                                        $q->escapeHTML( $pre || '' ) . '<b>' .
                                        $q->escapeHTML( $hit || '' ) . '</b>' .
                                        $q->escapeHTML( $post || '');
                        }
                        return $out;
                }

                sub attr_regex {
                        my ($rdoc,$attr) = @_;
                        my $text = $rdoc->attr( $attr );
                        return unless defined($text);

                        if (my $r = $config->{estraier}->{attr_regex}->{$attr} ) {
                                my $do = '$text =~ ' . $r . ';';
                                eval $do;
                                if ($@) {
                                        warn "eval $do failed: $@\n";
                                }
                        }
                        return $text;
                }

                # for each document in results
                for my $i ( 0 ... $nres->doc_num - 1 ) {

                        my $rdoc = $nres->get_doc($i);

                        $out .= '<div class="item">' .
                                '<h1>' . $rdoc->attr('@title') . '</h1>' .
                                '<p>' . html_snippet( $rdoc->snippet ) . '</p>' .
                                '<h2>' . attr_regex( $rdoc, 'source' ) . '</h2>';
                        my $uri = attr_regex( $rdoc, '@uri' );
                        $out .=
                                qq{<a href="$uri"><tt>$uri</tt></a> } .
                                attr_regex( $rdoc, '@mdate' ) .
                                ' [' . ( $skip + $i + 1 ) . ']';
                }

        } else {
                $out .= 'error: ' . $node->status;
        }

        return ($out,json);

}

if ($q->path_info() eq '/snippet') {

        print join("\n<!-- json data -->",
                get_results(
                        $q->param('search'),
                        $q->param('page'),
                )
        );


} else {

        my ($get_results,$json);

        if ($q->param('search')) {
                ($get_results, $json) = get_results( $q->param('search'), 1 );
        } else {
                $v->{status} = 'Enter search query';
                ($get_results, $json) = ('', json);
        }

        my $f = $q->path_info;
        $f =~ s/\W+//g;
        $f ||= 'search';
        $f .= '.html';
        open(my $s, $f)  || die "$f: $!";
        while(<$s>) {
                s/<%(.+?)%>/eval "$1"/ge;
                print;
        }
        close($f);

}
1	dpavlin	2	#!/usr/bin/perl -w
2
3			use strict;
4
5			use CGI::Simple;
6	dpavlin	5	use CGI::Carp qw(fatalsToBrowser warningsToBrowser);
7	dpavlin	2	use Search::Estraier;
8	dpavlin	5	use YAML::Syck;
9	dpavlin	15	use JSON::Syck;
10	dpavlin	5	use Data::Dump qw/dump/;
11	dpavlin	2
12			my $q = new CGI::Simple;
13			print qq{Content-type: text/html\n\r\n\r};
14
15	dpavlin	5	my $config = LoadFile('config.yml');
16
17	dpavlin	15	my $v = {
18			search => '',
19			hits => 0,
20			page => 0,
21			max_page => 0,
22			time => '',
23			id => time() . rand(99),
24			};
25
26			my $json;
27
28	dpavlin	5	#warn "config = ", dump($config);
29
30	dpavlin	17	sub json {
31			return
32			'<textarea id="json" style="display:none">' .
33			$q->escapeHTML( JSON::Syck::Dump( $v ) ) .
34			'</textarea>';
35			}
36
37	dpavlin	15	sub get_results {
38			my ($search, $page) = @_;
39	dpavlin	2
40	dpavlin	17	warn "get_results( $search , $page )\n";
41
42			if (! $search \|\| $search =~ m/^\s*$/) {
43	dpavlin	15	$v->{status} = 'Enter search query';
44	dpavlin	17	return('', json);
45	dpavlin	15	}
46	dpavlin	2
47	dpavlin	15	if (! $page) {
48			$v->{status} = 'Error: no page number?';
49	dpavlin	17	return('', json);
50	dpavlin	15	}
51	dpavlin	2
52	dpavlin	15	$search = join(" AND ", split(/\s+/, $search)) unless ($search =~ m/(?:AND\|OR\|\[\|\])/);
53			$v->{search} = $search \|\| '';
54	dpavlin	2
55	dpavlin	15	$v->{page} = $page;
56	dpavlin	2
57	dpavlin	15	my $node = new Search::Estraier::Node(%{ $config->{estraier} });
58	dpavlin	2
59	dpavlin	15	my $on_page = 30;
60			my $skip = ( $page - 1 ) * $on_page;
61	dpavlin	2
62	dpavlin	15	my $cond = new Search::Estraier::Condition;
63			$cond->set_phrase( $search );
64			$cond->set_max( $on_page );
65			$cond->set_skip( $skip );
66	dpavlin	2
67	dpavlin	15	my $nres = $node->search($cond, ( $config->{estraier}->{depth} \|\| 0 ) );
68	dpavlin	2
69	dpavlin	15	my $out;
70	dpavlin	2
71	dpavlin	15	if (defined($nres)) {
72			$v->{hits} = $nres->hits;
73			$v->{time} = $nres->hint('TIME');
74			$v->{max_page} = int( ($nres->hits + $on_page - 1) / $on_page );
75
76			$v->{status} = qq{
77			Got <b>$v->{hits}</b> results for <tt>$v->{search}</tt>
78			in <em>$v->{time} s</em>
79			};
80
81			sub html_snippet {
82			my $text = shift \|\| return;
83			my $out = '';
84			foreach my $s (split(/[\n\r]{2}/, $text)) {
85			$out .= ' ... ' if ($out);
86			my ($pre,$hit,$post) = split(/\n/,$s,3);
87			$hit =~ s/\t.*$//;
88			$out .=
89			$q->escapeHTML( $pre \|\| '' ) . '<b>' .
90			$q->escapeHTML( $hit \|\| '' ) . '</b>' .
91			$q->escapeHTML( $post \|\| '');
92			}
93			return $out;
94	dpavlin	2	}
95
96	dpavlin	15	sub attr_regex {
97			my ($rdoc,$attr) = @_;
98			my $text = $rdoc->attr( $attr );
99			return unless defined($text);
100	dpavlin	2
101	dpavlin	15	if (my $r = $config->{estraier}->{attr_regex}->{$attr} ) {
102			my $do = '$text =~ ' . $r . ';';
103			eval $do;
104			if ($@) {
105			warn "eval $do failed: $@\n";
106			}
107	dpavlin	14	}
108	dpavlin	15	return $text;
109	dpavlin	14	}
110
111	dpavlin	15	# for each document in results
112			for my $i ( 0 ... $nres->doc_num - 1 ) {
113	dpavlin	2
114	dpavlin	15	my $rdoc = $nres->get_doc($i);
115	dpavlin	2
116	dpavlin	15	$out .= '<div class="item">' .
117			'<h1>' . $rdoc->attr('@title') . '</h1>' .
118			'<p>' . html_snippet( $rdoc->snippet ) . '</p>' .
119			'<h2>' . attr_regex( $rdoc, 'source' ) . '</h2>';
120			my $uri = attr_regex( $rdoc, '@uri' );
121			$out .=
122			qq{<a href="$uri"><tt>$uri</tt></a> } .
123			attr_regex( $rdoc, '@mdate' ) .
124			' [' . ( $skip + $i + 1 ) . ']';
125			}
126	dpavlin	2
127	dpavlin	15	} else {
128			$out .= 'error: ' . $node->status;
129	dpavlin	2	}
130	dpavlin	15
131	dpavlin	17	return ($out,json);
132	dpavlin	15
133	dpavlin	2	}
134
135	dpavlin	15	if ($q->path_info() eq '/snippet') {
136	dpavlin	2
137	dpavlin	15	print join("\n<!-- json data -->",
138			get_results(
139			$q->param('search'),
140			$q->param('page'),
141			)
142			);
143	dpavlin	2
144
145			} else {
146
147	dpavlin	17	my ($get_results,$json);
148	dpavlin	2
149	dpavlin	17	if ($q->param('search')) {
150			($get_results, $json) = get_results( $q->param('search'), 1 );
151			} else {
152			$v->{status} = 'Enter search query';
153			($get_results, $json) = ('', json);
154			}
155
156	dpavlin	2	my $f = $q->path_info;
157			$f =~ s/\W+//g;
158			$f \|\|= 'search';
159			$f .= '.html';
160			open(my $s, $f) \|\| die "$f: $!";
161			while(<$s>) {
162	dpavlin	4	s/<%(.+?)%>/eval "$1"/ge;
163	dpavlin	2	print;
164			}
165			close($f);
166
167			}