1 |
package SWISH::PlusPlus; |
2 |
|
3 |
use 5.008004; |
4 |
use strict; |
5 |
use warnings; |
6 |
|
7 |
our $VERSION = '0.06'; |
8 |
|
9 |
use Carp; |
10 |
use File::Temp qw/ tempdir /; |
11 |
#use YAML; |
12 |
|
13 |
=head1 NAME |
14 |
|
15 |
SWISH::PlusPlus - Perl extension SWISH++ |
16 |
|
17 |
=head1 SYNOPSIS |
18 |
|
19 |
use SWISH::PlusPlus; |
20 |
blah blah blah |
21 |
|
22 |
=head1 DESCRIPTION |
23 |
|
24 |
This is perl module to use SWISH++ indexer by Paul J. Lucas. SWISH++ is |
25 |
rewrite of swish-e in C++ which is extremly fast (thank to mmap), but without |
26 |
support for properties (which this module tries to fix). |
27 |
|
28 |
Implementation of this module is crafted after L<Plucene::Simple> and it |
29 |
should be easy to replace Plucene with this module for increased |
30 |
performance. However, this module is not plug-in replacement. |
31 |
|
32 |
=head1 METHODS |
33 |
|
34 |
=head2 new |
35 |
|
36 |
Create new indexing object. |
37 |
|
38 |
my $i = SWISH::PlusPlus->new( |
39 |
index_dir => '/path/to/index', |
40 |
index => 'index++', |
41 |
search => 'search++', |
42 |
debug => 1, |
43 |
meta_in_body => 1, |
44 |
use_stopwords => 1, |
45 |
); |
46 |
|
47 |
Options to new are following: |
48 |
|
49 |
=over 5 |
50 |
|
51 |
=item C<index_dir> |
52 |
|
53 |
Path to directory in which index will be created. |
54 |
|
55 |
=item C<index> |
56 |
|
57 |
Full or partial path to SWISH++ index executable. By default, it's B<index> |
58 |
for self-compiled version. If you use Debian GNU/Linux package specify |
59 |
B<index++>. See C<Debian>. |
60 |
|
61 |
=item C<search> |
62 |
|
63 |
Full or partial path to SWISH++ search executable. By default, it's B<search>. |
64 |
|
65 |
=item C<debug> |
66 |
|
67 |
This option (off by default) will produce a lot of debugging output on |
68 |
C<STDERR> prefixed by C<##>. |
69 |
|
70 |
=item C<meta_in_body> |
71 |
|
72 |
This option (off by default) enables to search content of meta fields |
73 |
without specifing them (like they are in body of document). This will |
74 |
somewhat increate index size. |
75 |
|
76 |
=item C<use_stopwords> |
77 |
|
78 |
Use built-in SWISH++ stop words. By default, they are disabled. |
79 |
|
80 |
=back |
81 |
|
82 |
=cut |
83 |
|
84 |
sub new { |
85 |
my $class = shift; |
86 |
my $self = {@_}; |
87 |
bless($self, $class); |
88 |
|
89 |
foreach (qw(index_dir)) { |
90 |
croak "need $_" unless $self->{$_}; |
91 |
} |
92 |
|
93 |
my $index_dir = $self->{'index_dir'}; |
94 |
|
95 |
if ($index_dir !~ m#^/#) { |
96 |
chomp(my $cwd = `pwd`); |
97 |
$index_dir = "$cwd/$index_dir"; |
98 |
print STDERR "## full path to index_dir: $index_dir\n" if ($self->{'debug'}); |
99 |
$self->{'index_dir'} = $index_dir; |
100 |
} |
101 |
|
102 |
if (! -e $index_dir) { |
103 |
mkdir $index_dir || confess "can't create index ",$self->{'index'},": $!"; |
104 |
} |
105 |
|
106 |
# default executables |
107 |
$self->{'index'} ||= 'index'; |
108 |
$self->{'search'} ||= 'search'; |
109 |
|
110 |
print STDERR "## new index_dir: ",$index_dir," index: ",$self->{'index'}, " search: ",$self->{'search'},"\n" if ($self->{'debug'}); |
111 |
|
112 |
$self ? return $self : return undef; |
113 |
} |
114 |
|
115 |
|
116 |
=head2 check_bin |
117 |
|
118 |
Check if swish++ binaries specified in L<new> are available and verify |
119 |
version signature. |
120 |
|
121 |
if ($i->check_bin) { |
122 |
print "swish++ binaries found\n"; |
123 |
}; |
124 |
|
125 |
It will also setup property |
126 |
|
127 |
$i->{'version'} |
128 |
|
129 |
which you can examine to see version. |
130 |
|
131 |
=cut |
132 |
|
133 |
sub check_bin { |
134 |
my $self = shift; |
135 |
|
136 |
my $i = `$self->{'index'} -V 2>&1` || confess "can't find '",$self->{'index'},"' binary"; |
137 |
my $s = `$self->{'search'} -V 2>&1` || confess "can't find '",$self->{'search'},"' binary"; |
138 |
|
139 |
chomp $i; |
140 |
chomp $s; |
141 |
|
142 |
confess $self->{'index'}," binary is not SWISH++" unless ($i =~ m/^SWISH\+\+/); |
143 |
confess $self->{'search'}," binary is not SWISH++" unless ($s =~ m/^SWISH\+\+/); |
144 |
|
145 |
if ($i eq $s) { |
146 |
$self->{'version'} = $i; |
147 |
return 1; |
148 |
} else { |
149 |
carp "version difference: index is $i while search is $s"; |
150 |
return; |
151 |
} |
152 |
|
153 |
} |
154 |
|
155 |
=head2 index_document |
156 |
|
157 |
Quick way to add simple data to index. |
158 |
|
159 |
$i->index_document($key, $data); |
160 |
$i->index_document( 42 => 'meaning of life' ); |
161 |
|
162 |
=cut |
163 |
|
164 |
sub index_document { |
165 |
my $self = shift; |
166 |
|
167 |
my %doc = @_; |
168 |
|
169 |
foreach my $id (keys %doc) { |
170 |
$self->_create_doc( |
171 |
path => $id, |
172 |
body => $doc{$id}, |
173 |
); |
174 |
} |
175 |
|
176 |
return 1; |
177 |
} |
178 |
|
179 |
=head2 add |
180 |
|
181 |
Add document with metadata to index. |
182 |
|
183 |
$i->add( |
184 |
path => 'path/to/document', |
185 |
title => 'this is result title', |
186 |
meta => { |
187 |
description => 'this is description meta tag', |
188 |
date => '2004-11-04', |
189 |
author => 'Dobrica Pavlinusic', |
190 |
} |
191 |
body => 'this is text without meta data', |
192 |
); |
193 |
|
194 |
This is thin wrapper round L<_create_doc>. |
195 |
|
196 |
=cut |
197 |
|
198 |
sub add { |
199 |
my $self = shift; |
200 |
|
201 |
$self->_create_doc(@_); |
202 |
|
203 |
return 1; |
204 |
} |
205 |
=head2 search |
206 |
|
207 |
Search your index. |
208 |
|
209 |
my @results = $i->search("swhish query"); |
210 |
|
211 |
Returns array with result IDs. |
212 |
|
213 |
=cut |
214 |
|
215 |
sub search { |
216 |
my $self = shift; |
217 |
|
218 |
my $query = shift || return; |
219 |
|
220 |
$self->_close_index; |
221 |
|
222 |
my @results; |
223 |
|
224 |
# escape double quotes in query for shell |
225 |
$query =~ s/"/\\"/g; |
226 |
|
227 |
my $open_cmd = $self->{'search'}." -i ".$self->{'index_dir'}.'/index "'.$query.'" |'; |
228 |
print STDERR "## search $open_cmd\n" if ($self->{'debug'}); |
229 |
|
230 |
open(SEARCH, $open_cmd) || confess "can't start $open_cmd: $!"; |
231 |
while(<SEARCH>) { |
232 |
next if (/^#/); |
233 |
chomp; |
234 |
print STDERR "## $_\n" if ($self->{'debug'}); |
235 |
my ($rank,$path,$size,$title) = split(/ /,$_,4); |
236 |
push @results, { |
237 |
rank => $rank, |
238 |
path => $path, |
239 |
size => $size, |
240 |
title => $title, |
241 |
} |
242 |
} |
243 |
|
244 |
close(SEARCH) || confess "can't close search"; |
245 |
|
246 |
#print STDERR "## results: ",Dump(@results),"\n" if ($self->{'debug'}); |
247 |
|
248 |
return @results; |
249 |
} |
250 |
|
251 |
=head2 finish_update |
252 |
|
253 |
This method will close index. |
254 |
|
255 |
$i->finish_update; |
256 |
|
257 |
It will be called on DESTROY when $i goes out of scope. |
258 |
|
259 |
=cut |
260 |
|
261 |
sub finish_update { |
262 |
my $self = shift; |
263 |
|
264 |
$self->_close_index; |
265 |
} |
266 |
|
267 |
sub DESTROY { |
268 |
my $self = shift; |
269 |
$self->finish_update; |
270 |
} |
271 |
|
272 |
=head1 PRIVATE METHODS |
273 |
|
274 |
Private methods implement internals for creating temporary file needed for |
275 |
swish++. You should have no need to call them directly, and they are here |
276 |
just to have documentation. |
277 |
|
278 |
=head2 _init_indexer |
279 |
|
280 |
Create temporary directory in which files for indexing will be created and |
281 |
start index process. |
282 |
|
283 |
my $i->_init_indexer || die "can't start indexer"; |
284 |
|
285 |
It will also create empty file C<_stopwords_> to disable stop words. |
286 |
|
287 |
=cut |
288 |
|
289 |
sub _init_indexer { |
290 |
my $self = shift; |
291 |
|
292 |
$self->{'tmp_dir'} = tempdir( CLEANUP => 1 ) || confess "can't create temporary directory: $!"; |
293 |
|
294 |
chdir $self->{'tmp_dir'} || confess "can't chdir to ".$self->{'tmp_dir'}.": $!"; |
295 |
|
296 |
my $opt = "-v " . ($self->{'debug'} || '0'); |
297 |
|
298 |
unless ($self->{'use_stopwrods'}) { |
299 |
open(STOP, '>', "_stopwords_") || carp "can't create empty stopword file, skipping\n"; |
300 |
print STOP " "; |
301 |
close(STOP); |
302 |
$opt .= " -s _stopwords_"; |
303 |
} |
304 |
|
305 |
my $open_cmd = '| '.$self->{'index'}.' '.$opt.' -e "html:*" -i '.$self->{'index_dir'}.'/index -'; |
306 |
|
307 |
print STDERR "## open index $open_cmd\n" if ($self->{'index'}); |
308 |
|
309 |
open($self->{'index_fh'}, $open_cmd) || confess "can't start index with $open_cmd: $!"; |
310 |
|
311 |
|
312 |
return $self->{'index_fh'}; |
313 |
} |
314 |
|
315 |
=head2 _create_doc |
316 |
|
317 |
Create temporary file and pass it's name to swish++ |
318 |
|
319 |
$i->_create_doc( |
320 |
path => 'path/to/store/in/index', |
321 |
title => 'this is title in results', |
322 |
body => 'data to story in body tag', |
323 |
meta => { |
324 |
'meta name' => 'data for this meta', |
325 |
'another' => 'again more data', |
326 |
} |
327 |
); |
328 |
|
329 |
To delete document, just omit body and meta data. |
330 |
|
331 |
=cut |
332 |
|
333 |
sub _create_doc { |
334 |
my $self = shift; |
335 |
|
336 |
my $arg = {@_}; |
337 |
|
338 |
# open indexer if needed |
339 |
$self->{'index_fh'} ||= $self->_init_indexer; |
340 |
|
341 |
my $path = $self->{'tmp_dir'} || confess "no tmp_dir?"; |
342 |
|
343 |
open(TMP, '>', $arg->{'path'}) || die "can't create temp file ".$arg->{'path'}.": $!"; |
344 |
|
345 |
print TMP '<html><head>'; |
346 |
|
347 |
$arg->{'body'} ||= ''; |
348 |
|
349 |
if ($arg->{'meta'}) { |
350 |
foreach my $name (keys %{$arg->{'meta'}}) { |
351 |
my $content = $arg->{'meta'}->{$name}; |
352 |
print TMP qq{<meta name="$name" content="$content">}; |
353 |
$arg->{'body'} .= " $content" if ($self->{'meta_in_body'}); |
354 |
} |
355 |
} |
356 |
|
357 |
if (defined($arg->{'title'})) { |
358 |
print TMP '<title>' . ($arg->{'title'} || '') . '</title>'; |
359 |
$arg->{'body'} .= " ".$arg->{'title'} if ($self->{'meta_in_body'}); |
360 |
} |
361 |
|
362 |
print TMP '</head><body>' . $arg->{'body'} . '</body></html>'; |
363 |
|
364 |
close(TMP) || confess "can't close tmp file ".$arg->{'path'}.": $!"; |
365 |
|
366 |
print { $self->{'index_fh'} } $arg->{'path'}."\n"; |
367 |
} |
368 |
|
369 |
=head2 _close_index |
370 |
|
371 |
Close index after indexing. |
372 |
|
373 |
$i->_close_index; |
374 |
|
375 |
You have to close index before searching. |
376 |
|
377 |
=cut |
378 |
|
379 |
sub _close_index { |
380 |
my $self = shift; |
381 |
|
382 |
return unless ($self->{'index_fh'}); |
383 |
|
384 |
print STDERR "## close index\n" if ($self->{'debug'}); |
385 |
|
386 |
close($self->{'index_fh'}); |
387 |
undef $self->{'index_fh'}; |
388 |
} |
389 |
|
390 |
1; |
391 |
__END__ |
392 |
|
393 |
=head2 EXPORT |
394 |
|
395 |
None by default. |
396 |
|
397 |
=head1 RELATED |
398 |
|
399 |
=head2 Debian |
400 |
|
401 |
Debian version of swish++ is often old (version 5 at moment of this writing |
402 |
while version 6 is available in source code), so this module by default |
403 |
uses executable names B<index> and B<search> for self-compiled version |
404 |
instead of one from Debian package. See L<new> how to specify Debian |
405 |
default binaries B<index++> and B<search++>. |
406 |
|
407 |
=head2 SWISH++ |
408 |
|
409 |
Aside from very good rewrite in C++, SWISH++ is fatster because it has |
410 |
claver heuristics about which data in input files are words to index and |
411 |
which are not. It's based on English language and might be best choice if |
412 |
you plan to install large amount of long text documents. |
413 |
|
414 |
However, if you plan to index all data from structured storage (e.g. RDBMS) |
415 |
you might want B<all> words from data to end up in index as opposed to just |
416 |
those which look like English words. This is especially important if you |
417 |
don't plan to index English texts with this module. |
418 |
|
419 |
With distribution build versions of SWISH++ you might have problems with |
420 |
disepearing words. To overcome this problem, you will have to compile and |
421 |
configure SWISH++ yourself (because language characteristics are |
422 |
compilation-time option). |
423 |
|
424 |
Compilation of SWISH++ is easy process well described on project's web |
425 |
pages. To see my very relaxed sample configuration take a look at C<swish++> |
426 |
directory included in distribution. |
427 |
|
428 |
=head2 SWISH++ config |
429 |
|
430 |
C<config.h> located in C<swish++> directory of this distribution is relaxed |
431 |
SWISH++ configuration that will index all words passed to it. This |
432 |
configuration is needed for B<date test> because default configuration |
433 |
doesn't recognize 2004-12-05 as date. Have in mind that your index size |
434 |
might explode. |
435 |
|
436 |
=head1 SEE ALSO |
437 |
|
438 |
C<swish++> web site L<http://homepage.mac.com/pauljlucas/software/swish/> |
439 |
|
440 |
=head1 AUTHOR |
441 |
|
442 |
Dobrica Pavlinusic, E<lt>dpavlin@rot13.orgE<gt> |
443 |
|
444 |
=head1 COPYRIGHT AND LICENSE |
445 |
|
446 |
Copyright (C) 2004 by Dobrica Pavlinusic |
447 |
|
448 |
This library is free software; you can redistribute it and/or modify |
449 |
it under the same terms as Perl itself, either Perl version 5.8.4 or, |
450 |
at your option, any later version of Perl 5 you may have available. |
451 |
|
452 |
|
453 |
=cut |