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