| 1 |
#!/usr/bin/perl |
| 2 |
# -*- Mode: Perl -*- |
| 3 |
# $Basename: Wais.pm $ |
| 4 |
# $Revision: 1.4 $ |
| 5 |
# Author : Ulrich Pfeifer |
| 6 |
# Created On : Mon Sep 16 11:08:04 1996 |
| 7 |
# Last Modified By: Ulrich Pfeifer |
| 8 |
# Last Modified On: Wed Nov 5 17:34:02 1997 |
| 9 |
# Language : CPerl |
| 10 |
# Update Count : 156 |
| 11 |
# Status : Unknown, Use with caution! |
| 12 |
# |
| 13 |
# (C) Copyright 1997, Ulrich Pfeifer, all rights reserved. |
| 14 |
# |
| 15 |
# |
| 16 |
|
| 17 |
package WAIT::Wais; |
| 18 |
|
| 19 |
require WAIT::Query::Wais; |
| 20 |
require WAIT::Database; |
| 21 |
use Fcntl; |
| 22 |
use strict; |
| 23 |
use vars qw(%DB %TB); |
| 24 |
|
| 25 |
my %FORMATTER; |
| 26 |
|
| 27 |
BEGIN { # check for available formatters |
| 28 |
%FORMATTER = qw(text WAIT::Format::Base); |
| 29 |
for my $inc (@INC) { |
| 30 |
if (-d "$inc/WAIT/Format") { |
| 31 |
for my $format ( <$inc/WAIT/Format/*.pm>) { |
| 32 |
my ($name) = ($format =~ /(\w+)\.pm$/); |
| 33 |
my $module = "WAIT::Format::$name"; |
| 34 |
$name = lc $name; |
| 35 |
$FORMATTER{$name} = $module; |
| 36 |
} |
| 37 |
} |
| 38 |
} |
| 39 |
} |
| 40 |
|
| 41 |
|
| 42 |
sub _database { |
| 43 |
my $path = shift; |
| 44 |
my ($dir, $dn, $tn) = ($path =~ m:(.*)/([^/]+)/([^/]+)$:); |
| 45 |
|
| 46 |
return $DB{"$dir/$dn"} if exists $DB{"$dir/$dn"}; |
| 47 |
$DB{"$dir/$dn"} = WAIT::Database->open(name => $dn, directory => $dir, |
| 48 |
mode => O_RDONLY); |
| 49 |
return $DB{"$dir/$dn"}; |
| 50 |
} |
| 51 |
|
| 52 |
sub _table { |
| 53 |
my $path = shift; |
| 54 |
|
| 55 |
return $TB{$path} if exists $TB{$path}; |
| 56 |
my $db = _database($path); |
| 57 |
my ($dir, $dn, $tn) = ($path =~ m:(.*)/([^/]+)/([^/]+)$:); |
| 58 |
$TB{$path} = $db->table(name => $tn); |
| 59 |
$TB{$path}; |
| 60 |
} |
| 61 |
|
| 62 |
sub Search { |
| 63 |
my (@requests) = @_; |
| 64 |
my $request; |
| 65 |
my $result = new WAIT::Wais::Result; |
| 66 |
for $request (@requests) { |
| 67 |
my $query = $request->{'query'}; |
| 68 |
my $database = $request->{'database'}; |
| 69 |
my $tag = $request->{'tag'} || $request->{'database'}; |
| 70 |
my ($dir, $dn, $tn) = ($database =~ m:(.*)/([^/]+)/([^/]+)$:); |
| 71 |
my $tb = _table($database); |
| 72 |
unless (defined $tb) { |
| 73 |
$result->add(Tag => $tag, Error => 'Could not open database'); |
| 74 |
return $result; |
| 75 |
} |
| 76 |
my $wquery; |
| 77 |
eval {$wquery = WAIT::Query::Wais::query($tb, $query)}; |
| 78 |
if ($@ ne '') { |
| 79 |
$result->add(Tag => $tag, Error => $@); |
| 80 |
return $result; |
| 81 |
} |
| 82 |
my %po = $wquery->execute(); |
| 83 |
$result->add(Tag => $tag, Database => $database, |
| 84 |
Table => $tb, Postings => \%po) |
| 85 |
} |
| 86 |
$result; |
| 87 |
} |
| 88 |
|
| 89 |
sub Retrieve { |
| 90 |
my %parm = @_; |
| 91 |
my $result = new WAIT::Wais::Result; |
| 92 |
my $tb = _table($parm{database}); |
| 93 |
|
| 94 |
unless (defined $tb) { |
| 95 |
$result->add(Tag => 'document', Error => 'Could not open database'); |
| 96 |
return $result; |
| 97 |
} |
| 98 |
|
| 99 |
my $did = ref($parm{docid})?$parm{docid}->did:$parm{docid}; |
| 100 |
|
| 101 |
my %rec = $tb->fetch($did); |
| 102 |
|
| 103 |
# another CPAN hack |
| 104 |
if ($rec{docid} =~ m(^data/)) { |
| 105 |
$rec{docid} = $tb->dir . '/' . $rec{docid}; |
| 106 |
} |
| 107 |
|
| 108 |
my $text = $tb->fetch_extern($rec{docid}); |
| 109 |
|
| 110 |
my @txt; |
| 111 |
$tb->open; |
| 112 |
if ($parm{query}) { |
| 113 |
@txt = WAIT::Query::Wais::query($tb,$parm{query})->hilight($text); |
| 114 |
} else { |
| 115 |
@txt = $tb->layout->tag($text); |
| 116 |
} |
| 117 |
|
| 118 |
if ($parm{lines}) { |
| 119 |
@txt = filter($parm{lines}, @txt); |
| 120 |
} |
| 121 |
|
| 122 |
my $type = lc $parm{type}; |
| 123 |
|
| 124 |
my $module = (exists $FORMATTER{$type})?$FORMATTER{$type}:$FORMATTER{text}; |
| 125 |
my $path = $module; |
| 126 |
$path =~ s(::)(/)g; |
| 127 |
|
| 128 |
require "$path.pm"; |
| 129 |
my $format = new $module; |
| 130 |
$text = $format->as_string(\@txt, sub {$tb->fetch($did)}); |
| 131 |
$result->add(Tag => 'document', Text => $text); |
| 132 |
} |
| 133 |
|
| 134 |
sub filter { |
| 135 |
my $filter = shift; |
| 136 |
my @result; |
| 137 |
my @context; |
| 138 |
my $lines = 0; |
| 139 |
my $clines = 0; |
| 140 |
my $elipsis = 0; |
| 141 |
|
| 142 |
while (@_) { |
| 143 |
my %tag = %{shift @_}; |
| 144 |
my $txt = shift @_; |
| 145 |
|
| 146 |
for (split /(\n)/, $txt) { |
| 147 |
if ($_ eq "\n") { |
| 148 |
if (exists $tag{_qt}) { |
| 149 |
#die "Weird!"; |
| 150 |
push @result, {_i=>1}, "[WEIRD]"; |
| 151 |
} elsif ($lines) { |
| 152 |
push @result, {}, $_; |
| 153 |
$lines--; |
| 154 |
} else { |
| 155 |
push @context, {}, $_; |
| 156 |
$clines++; |
| 157 |
} |
| 158 |
} else { |
| 159 |
if (exists $tag{_qt}) { |
| 160 |
push @result, {_i=>1}, "\n[ $elipsis lines ]\n" if $elipsis; |
| 161 |
push @result, @context, {%tag}, $_; |
| 162 |
delete $tag{_qt}; |
| 163 |
@context = (); $clines = 0; $elipsis=0; |
| 164 |
$lines = $filter+1; |
| 165 |
} elsif ($lines) { |
| 166 |
push @result, \%tag, $_; |
| 167 |
} else { |
| 168 |
push @context, \%tag, $_; |
| 169 |
} |
| 170 |
} |
| 171 |
if ($clines>$filter) { |
| 172 |
my (%tag, $txt); |
| 173 |
while ($clines>$filter) { |
| 174 |
%tag = %{shift @context}; |
| 175 |
$txt = shift @context; |
| 176 |
if ($txt =~ /\n/) { |
| 177 |
$clines--; |
| 178 |
$elipsis++; |
| 179 |
} |
| 180 |
} |
| 181 |
} |
| 182 |
} |
| 183 |
} |
| 184 |
@result; |
| 185 |
} |
| 186 |
|
| 187 |
package WAIT::Wais::Result; |
| 188 |
|
| 189 |
sub new { |
| 190 |
my $type = shift; |
| 191 |
my %par = @_; |
| 192 |
my $self = {'header' => [], 'diagnostics' => [], 'text' => ''}; |
| 193 |
|
| 194 |
bless $self, $type; |
| 195 |
} |
| 196 |
|
| 197 |
sub _header { |
| 198 |
my ($database, $did, $score) = @_; |
| 199 |
my $types; |
| 200 |
my $tb = WAIT::Wais::_table($database); |
| 201 |
my %rec = $tb->fetch($did); |
| 202 |
my $lines = $rec{'lines'} || 0; |
| 203 |
my $length = $rec{'size'} || 0; |
| 204 |
unless ($length) { |
| 205 |
($length) = ($rec{docid} =~ /(\d+)$/) |
| 206 |
} |
| 207 |
unless ($rec{docid} =~ m(^/)) { |
| 208 |
$rec{docid} = $tb->dir . '/' . $rec{docid}; |
| 209 |
} |
| 210 |
my $headline = $rec{headline} || ''; |
| 211 |
if (exists $rec{types}) { |
| 212 |
$types = [split ',', $rec{types}] |
| 213 |
} else { |
| 214 |
$types = [keys %FORMATTER]; |
| 215 |
} |
| 216 |
|
| 217 |
[$score, $lines, $length, $headline, $types, |
| 218 |
WAIT::Wais::Docid->new('wait',$database, $did)]; |
| 219 |
} |
| 220 |
|
| 221 |
sub add { |
| 222 |
my $self = shift; |
| 223 |
my %parm = @_; |
| 224 |
my $tag = $parm{Tag}; |
| 225 |
my $docid; |
| 226 |
|
| 227 |
if ($parm{Postings}) { |
| 228 |
my @result; |
| 229 |
my @left = @{$self->{'header'}}; |
| 230 |
my @right; |
| 231 |
for (keys %{$parm{Postings}}) { |
| 232 |
push @right, _header($parm{Database}, $_, $parm{Postings}->{$_}) |
| 233 |
} |
| 234 |
while (($#left >= $[) or ($#right >= $[)) { |
| 235 |
if ($#left < $[) { |
| 236 |
for (@right) { |
| 237 |
push @result, [$tag, @{$_}]; |
| 238 |
} |
| 239 |
last; |
| 240 |
} |
| 241 |
if ($#right < $[) { |
| 242 |
push @result, @left; |
| 243 |
last; |
| 244 |
} |
| 245 |
if ($left[0]->[1] > $right[0]->[0]) { |
| 246 |
push @result, shift @left; |
| 247 |
} else { |
| 248 |
push @result, [$tag, @{shift @right}]; |
| 249 |
} |
| 250 |
} |
| 251 |
$self->{'header'} = \@result; |
| 252 |
} |
| 253 |
if ($parm{Errors}) { |
| 254 |
my %diag = %{$parm{Errors}}; |
| 255 |
for (keys %diag) { |
| 256 |
push(@{$self->{'diagnostics'}}, [$tag, $_, $diag{$_}]); |
| 257 |
} |
| 258 |
} |
| 259 |
if ($parm{Text}) { |
| 260 |
$self->{'text'} .= $parm{Text}; |
| 261 |
} |
| 262 |
|
| 263 |
$self; |
| 264 |
} |
| 265 |
|
| 266 |
|
| 267 |
sub diagnostics { |
| 268 |
my $self = shift; |
| 269 |
|
| 270 |
@{$self->{'diagnostics'}}; |
| 271 |
} |
| 272 |
|
| 273 |
sub header { |
| 274 |
my $self = shift; |
| 275 |
|
| 276 |
@{$self->{'header'}}; |
| 277 |
} |
| 278 |
|
| 279 |
sub text { |
| 280 |
my $self = shift; |
| 281 |
|
| 282 |
$self->{'text'}; |
| 283 |
} |
| 284 |
|
| 285 |
package WAIT::Wais::Docid; |
| 286 |
|
| 287 |
sub new { |
| 288 |
my $type = shift; |
| 289 |
my ($server, $database, $dodid) = @_; |
| 290 |
my $self = join ';', $server, $database, $dodid; |
| 291 |
bless \$self, $type; |
| 292 |
} |
| 293 |
|
| 294 |
sub did { |
| 295 |
($_[0]->split)[2]; |
| 296 |
} |
| 297 |
|
| 298 |
sub split { |
| 299 |
my $self = shift; |
| 300 |
|
| 301 |
split /;/, $$self; |
| 302 |
} |
| 303 |
|
| 304 |
1; |
Parent Directory
| 
Revision Log