/[SWISH-PlusPlus]/trunk/PlusPlus.pm
This is repository of my old source code which isn't updated any more. Go to git.rot13.org for current projects!
ViewVC logotype

Diff of /trunk/PlusPlus.pm

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 3 by dpavlin, Fri Dec 3 15:23:23 2004 UTC revision 11 by dpavlin, Sun Dec 5 13:30:57 2004 UTC
# Line 4  use 5.008004; Line 4  use 5.008004;
4  use strict;  use strict;
5  use warnings;  use warnings;
6    
7  our $VERSION = '0.02';  our $VERSION = '0.05';
8    
9  use Carp;  use Carp;
10    use File::Temp qw/ tempdir /;
11    #use YAML;
12    
13  =head1 NAME  =head1 NAME
14    
# Line 29  performance. However, this module is not Line 31  performance. However, this module is not
31    
32  =head1 METHODS  =head1 METHODS
33    
34  =head2 open  =head2 new
35    
36  Create new indexing object.  Create new indexing object.
37    
38    my $i = SWISH::PlusPlus->open(    my $i = SWISH::PlusPlus->new(
39          index_dir => '/path/to/index',          index_dir => '/path/to/index',
40          index => 'index++',          index => 'index++',
41          search => 'search++',          search => 'search++',
42            debug => 1,
43            meta_in_body => 1,
44            use_stopwords => 1,
45    );    );
46    
47  Options to open are following:  Options to new are following:
48    
49  =over 5  =over 5
50    
# Line 57  B<index++>. See C<Debian>. Line 62  B<index++>. See C<Debian>.
62    
63  Full or partial path to SWISH++ search executable. By default, it's B<search>.  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  =back
81    
82  =cut  =cut
83    
84  sub open {  sub new {
85          my $class = shift;          my $class = shift;
86          my $self = {@_};          my $self = {@_};
87          bless($self, $class);          bless($self, $class);
# Line 78  sub open { Line 98  sub open {
98          $self->{'index'} ||= 'index';          $self->{'index'} ||= 'index';
99          $self->{'search'} ||= 'search';          $self->{'search'} ||= 'search';
100    
101            print STDERR "## new index_dir: ",$self->{'index_dir'}," index: ",$self->{'index'}, " search: ",$self->{'search'},"\n" if ($self->{'debug'});
102    
103          $self ? return $self : return undef;          $self ? return $self : return undef;
104  }  }
105    
106    
107  =head2 check_bin  =head2 check_bin
108    
109  Check if swish++ binaries specified in L<open> are available and verify  Check if swish++ binaries specified in L<new> are available and verify
110  version signature.  version signature.
111    
112    if ($i->check_bin) {    if ($i->check_bin) {
# Line 121  sub check_bin { Line 143  sub check_bin {
143    
144  }  }
145    
146    =head2 index_document
147    
148    Quick way to add simple data to index.
149    
150      $i->index_document($key, $data);
151      $i->index_document( 42 => 'meaning of life' );
152    
153    =cut
154    
155    sub index_document {
156            my $self = shift;
157    
158            my %doc = @_;
159    
160            foreach my $id (keys %doc) {
161                    $self->_create_doc(
162                            path => $id,
163                            body => $doc{$id},
164                    );
165            }
166    
167            return 1;
168    }
169    
170    =head2 add
171    
172    Add document with metadata to index.
173    
174      $i->add(
175            path => 'path/to/document',
176            title => 'this is result title',
177            meta => {
178                    description => 'this is description meta tag',
179                    date => '2004-11-04',
180                    author => 'Dobrica Pavlinusic',
181            }
182            body => 'this is text without meta data',
183      );
184    
185    This is thin wrapper round L<_create_doc>.
186    
187    =cut
188    
189    sub add {
190            my $self = shift;
191    
192            $self->_create_doc(@_);
193    
194            return 1;
195    }
196    =head2 search
197    
198    Search your index.
199    
200      my @results = $i->search("swhish query");
201    
202    Returns array with result IDs.
203    
204    =cut
205    
206    sub search {
207            my $self = shift;
208    
209            my $query = shift || return;
210    
211            $self->_close_index;
212    
213            my @results;
214    
215            # escape double quotes in query for shell
216            $query =~ s/"/\\"/g;
217    
218            my $open_cmd = $self->{'search'}." -i ".$self->{'index_dir'}.'/index "'.$query.'" |';
219            print STDERR "## search $open_cmd\n" if ($self->{'debug'});
220    
221            open(SEARCH, $open_cmd) || confess "can't start $open_cmd: $!";
222            while(<SEARCH>) {
223                    next if (/^#/);
224                    chomp;
225                    print STDERR "## $_\n" if ($self->{'debug'});
226                    my ($rank,$path,$size,$title) = split(/ /,$_,4);
227                    push @results, {
228                            rank => $rank,
229                            path => $path,
230                            size => $size,
231                            title => $title,
232                    }
233            }
234    
235            close(SEARCH) || confess "can't close search";
236    
237            #print STDERR "## results: ",Dump(@results),"\n" if ($self->{'debug'});
238    
239            return @results;
240    }
241    
242    =head1 PRIVATE METHODS
243    
244    Private methods implement internals for creating temporary file needed for
245    swish++. You should have no need to call them directly, and they are here
246    just to have documentation.
247    
248    =head2 _init_indexer
249    
250    Create temporary directory in which files for indexing will be created and
251    start index process.
252    
253      my $i->_init_indexer || die "can't start indexer";
254    
255    It will also create empty file C<_stopwords_> to disable stop words.
256    
257    =cut
258    
259    sub _init_indexer {
260            my $self = shift;
261    
262            $self->{'tmp_dir'} = tempdir( CLEANUP => 1 ) || confess "can't create temporary directory: $!";
263    
264            chdir $self->{'tmp_dir'} || confess "can't chdir to ".$self->{'tmp_dir'}.": $!";
265    
266            my $opt = "-v 4";
267    
268            unless ($self->{'use_stopwrods'}) {
269                    open(STOP, '>', "_stopwords_") || carp "can't create empty stopword file, skipping\n";
270                    print STOP "  ";
271                    close(STOP);
272                    $opt .= " -s _stopwords_";
273            }
274    
275            my $open_cmd = '| '.$self->{'index'}.' '.$opt.' -e "html:*" -i '.$self->{'index_dir'}.'/index -';
276    
277    
278            open($self->{'index_fh'}, $open_cmd) || confess "can't start index with $open_cmd: $!";
279    
280    
281            return $self->{'index_fh'};
282    }
283    
284    =head2 _create_doc
285    
286    Create temporary file and pass it's name to swish++
287    
288      $i->_create_doc(
289            path => 'path/to/store/in/index',
290            title => 'this is title in results',
291            body => 'data to story in body tag',
292            meta => {
293                    'meta name' => 'data for this meta',
294                    'another' => 'again more data',
295            }
296      );
297    
298    To delete document, just omit body and meta data.
299    
300    =cut
301    
302    sub _create_doc {
303            my $self = shift;
304    
305            my $arg = {@_};
306    
307            # open indexer if needed
308            $self->{'index_fh'} ||= $self->_init_indexer;
309    
310            my $path = $self->{'tmp_dir'} || confess "no tmp_dir?";
311    
312            open(TMP, '>', $arg->{'path'}) || die "can't create temp file ".$arg->{'path'}.": $!";
313    
314            print TMP '<html><head>';
315    
316            $arg->{'body'} ||= '';
317    
318            if ($arg->{'meta'}) {
319                    foreach my $name (keys %{$arg->{'meta'}}) {
320                            my $content = $arg->{'meta'}->{$name};
321                            print TMP qq{<meta name="$name" content="$content">};
322                            $arg->{'body'} .= " $content" if ($self->{'meta_in_body'});
323                    }
324            }
325    
326            if (defined($arg->{'title'})) {
327                    print TMP '<title>' . ($arg->{'title'} || '') . '</title>';
328                    $arg->{'body'} .= " ".$arg->{'title'} if ($self->{'meta_in_body'});
329            }
330    
331            print TMP '</head><body>' . $arg->{'body'} . '</body></html>';
332            
333            close(TMP) || confess "can't close tmp file ".$arg->{'path'}.": $!";
334    
335            print { $self->{'index_fh'} } $arg->{'path'}."\n";
336    }
337    
338    =head2 _close_index
339    
340    Close index after indexing.
341    
342      $i->_close_index;
343    
344    You have to close index before searching.
345    
346    =cut
347    
348    sub _close_index {
349            my $self = shift;
350    
351            return unless ($self->{'index_fh'});
352    
353            print STDERR "## close index\n" if ($self->{'debug'});
354    
355            close($self->{'index_fh'});
356            undef $self->{'index_fh'};
357    }
358    
359  1;  1;
360  __END__  __END__
361    
# Line 135  None by default. Line 370  None by default.
370  Debian version of swish++ is often old (version 5 at moment of this writing  Debian version of swish++ is often old (version 5 at moment of this writing
371  while version 6 is available in source code), so this module by default  while version 6 is available in source code), so this module by default
372  uses executable names B<index> and B<search> for self-compiled version  uses executable names B<index> and B<search> for self-compiled version
373  instead of one from Debian package. See L<open> how to specify Debian  instead of one from Debian package. See L<new> how to specify Debian
374  default binaries B<index++> and B<search++>.  default binaries B<index++> and B<search++>.
375    
376  =head1 SEE ALSO  =head2 SWISH++
377    
378  Mention other useful documentation such as the documentation of  Aside from very good rewrite in C++, SWISH++ is fatster because it has
379  related modules or operating system documentation (such as man pages  claver heuristics about which data in input files are words to index and
380  in UNIX), or any relevant external documentation such as RFCs or  which are not. It's based on English language and might be best choice if
381  standards.  you plan to install large amount of long text documents.
382    
383    However, if you plan to index all data from structured storage (e.g. RDBMS)
384    you might want B<all> words from data to end up in index as opposed to just
385    those which look like English words. This is especially important if you
386    don't plan to index English texts with this module.
387    
388    With distribution build versions of SWISH++ you might have problems with
389    disepearing words. To overcome this problem, you will have to compile and
390    configure SWISH++ yourself (because language characteristics are
391    compilation-time option).
392    
393    Compilation of SWISH++ is easy process well described on project's web
394    pages. To see my very relaxed sample configuration take a look at C<swish++>
395    directory included in distribution.
396    
397    =head2 SWISH++ config
398    
399    C<config.h> located in C<swish++> directory of this distribution is relaxed
400    SWISH++ configuration that will index all words passed to it. This
401    configuration is needed for B<date test> because default configuration
402    doesn't recognize 2004-12-05 as date. Have in mind that your index size
403    might explode.
404    
405  If you have a mailing list set up for your module, mention it here.  =head1 SEE ALSO
406    
407  If you have a web site set up for your module, mention it here.  C<swish++> web site L<http://homepage.mac.com/pauljlucas/software/swish/>
408    
409  =head1 AUTHOR  =head1 AUTHOR
410    
411  Dobrica Pavlinusic, E<lt>dpavlin@E<gt>  Dobrica Pavlinusic, E<lt>dpavlin@rot13.orgE<gt>
412    
413  =head1 COPYRIGHT AND LICENSE  =head1 COPYRIGHT AND LICENSE
414    
Legend:
Removed from v.3  
changed lines
  Added in v.11
  ViewVC Help
Powered by ViewVC 1.1.26