/[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 9 by dpavlin, Sun Dec 5 00:59:50 2004 UTC revision 13 by dpavlin, Sun Dec 5 14:35:54 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 = '0.03';  our $VERSION = '0.06';
8    
9  use Carp;  use Carp;
10  use File::Temp qw/ tempdir /;  use File::Temp qw/ tempdir /;
# Line 31  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++',
# Line 44  Create new indexing object. Line 44  Create new indexing object.
44          use_stopwords => 1,          use_stopwords => 1,
45    );    );
46    
47  Options to open are following:  Options to new are following:
48    
49  =over 5  =over 5
50    
# Line 81  Use built-in SWISH++ stop words. By defa Line 81  Use built-in SWISH++ stop words. By defa
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 90  sub open { Line 90  sub open {
90                  croak "need $_" unless $self->{$_};                  croak "need $_" unless $self->{$_};
91          }          }
92    
93          if (! -e $self->{'index_dir'}) {          my $index_dir = $self->{'index_dir'};
94                  mkdir $self->{'index_dir'} || confess "can't create index ",$self->{'index'},": $!";  
95            if ($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          # default executables
107          $self->{'index'} ||= 'index';          $self->{'index'} ||= 'index';
108          $self->{'search'} ||= 'search';          $self->{'search'} ||= 'search';
109    
110          print STDERR "## open index_dir: ",$self->{'index_dir'}," index: ",$self->{'index'}, " search: ",$self->{'search'},"\n" if ($self->{'debug'});          print STDERR "## new index_dir: ",$index_dir," index: ",$self->{'index'}, " search: ",$self->{'search'},"\n" if ($self->{'debug'});
111    
112          $self ? return $self : return undef;          $self ? return $self : return undef;
113  }  }
# Line 106  sub open { Line 115  sub open {
115    
116  =head2 check_bin  =head2 check_bin
117    
118  Check if swish++ binaries specified in L<open> are available and verify  Check if swish++ binaries specified in L<new> are available and verify
119  version signature.  version signature.
120    
121    if ($i->check_bin) {    if ($i->check_bin) {
# Line 218  sub search { Line 227  sub search {
227          my $open_cmd = $self->{'search'}." -i ".$self->{'index_dir'}.'/index "'.$query.'" |';          my $open_cmd = $self->{'search'}." -i ".$self->{'index_dir'}.'/index "'.$query.'" |';
228          print STDERR "## search $open_cmd\n" if ($self->{'debug'});          print STDERR "## search $open_cmd\n" if ($self->{'debug'});
229    
230          CORE::open(SEARCH, $open_cmd) || confess "can't start $open_cmd: $!";          open(SEARCH, $open_cmd) || confess "can't start $open_cmd: $!";
231          while(<SEARCH>) {          while(<SEARCH>) {
232                  next if (/^#/);                  next if (/^#/);
233                  chomp;                  chomp;
# Line 239  sub search { Line 248  sub search {
248          return @results;          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  =head1 PRIVATE METHODS
273    
274  Private methods implement internals for creating temporary file needed for  Private methods implement internals for creating temporary file needed for
# Line 263  sub _init_indexer { Line 293  sub _init_indexer {
293    
294          chdir $self->{'tmp_dir'} || confess "can't chdir to ".$self->{'tmp_dir'}.": $!";          chdir $self->{'tmp_dir'} || confess "can't chdir to ".$self->{'tmp_dir'}.": $!";
295    
296          my $opt = "-v 4";          my $opt = "-v " . ($self->{'debug'} || '0');
297    
298          unless ($self->{'use_stopwrods'}) {          unless ($self->{'use_stopwrods'}) {
299                  CORE::open(STOP, '>', "_stopwords_") || carp "can't create empty stopword file, skipping\n";                  open(STOP, '>', "_stopwords_") || carp "can't create empty stopword file, skipping\n";
300                  print STOP "  ";                  print STOP "  ";
301                  close(STOP);                  close(STOP);
302                  $opt .= " -s _stopwords_";                  $opt .= " -s _stopwords_";
# Line 274  sub _init_indexer { Line 304  sub _init_indexer {
304    
305          my $open_cmd = '| '.$self->{'index'}.' '.$opt.' -e "html:*" -i '.$self->{'index_dir'}.'/index -';          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          CORE::open($self->{'index_fh'}, $open_cmd) || confess "can't start index with $open_cmd: $!";          open($self->{'index_fh'}, $open_cmd) || confess "can't start index with $open_cmd: $!";
310    
311    
312          return $self->{'index_fh'};          return $self->{'index_fh'};
# Line 309  sub _create_doc { Line 340  sub _create_doc {
340    
341          my $path = $self->{'tmp_dir'} || confess "no tmp_dir?";          my $path = $self->{'tmp_dir'} || confess "no tmp_dir?";
342    
343          CORE::open(TMP, '>', $arg->{'path'}) || die "can't create temp file ".$arg->{'path'}.": $!";          open(TMP, '>', $arg->{'path'}) || die "can't create temp file ".$arg->{'path'}.": $!";
344    
345          print TMP '<html><head>';          print TMP '<html><head>';
346    
347          $arg->{'body'} ||= '';          $arg->{'body'} ||= '';
348    
349          if ($arg->{'meta'}) {          if ($arg->{'meta'}) {
350                  confess "not yet implemented";                  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($arg->{'title'})) {          if (defined($arg->{'title'})) {
# Line 366  None by default. Line 401  None by default.
401  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
402  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
403  uses executable names B<index> and B<search> for self-compiled version  uses executable names B<index> and B<search> for self-compiled version
404  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
405  default binaries B<index++> and B<search++>.  default binaries B<index++> and B<search++>.
406    
407  =head2 SWISH++  =head2 SWISH++
# Line 390  Compilation of SWISH++ is easy process w Line 425  Compilation of SWISH++ is easy process w
425  pages. To see my very relaxed sample configuration take a look at C<swish++>  pages. To see my very relaxed sample configuration take a look at C<swish++>
426  directory included in distribution.  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  =head1 SEE ALSO
437    
438  C<swish++> web site L<http://homepage.mac.com/pauljlucas/software/swish/>  C<swish++> web site L<http://homepage.mac.com/pauljlucas/software/swish/>
Legend:
Removed from v.9  
changed lines
  Added in v.13
  ViewVC Help
Powered by ViewVC 1.1.26