/[webpac2]/Webpacus/lib/Webpacus/Model/WebPAC.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 /Webpacus/lib/Webpacus/Model/WebPAC.pm

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

revision 143 by dpavlin, Fri Nov 25 01:24:31 2005 UTC revision 271 by dpavlin, Sat Dec 17 03:19:58 2005 UTC
# Line 7  use base qw/ Line 7  use base qw/
7          Catalyst::Model          Catalyst::Model
8  /;  /;
9  use Data::Dumper;  use Data::Dumper;
10  use WebPAC::DB;  use WebPAC::Store 0.08;
11  use WebPAC::Output::TT;  use WebPAC::Output::TT 0.04;
12  use WebPAC::Search::Estraier 0.02;  use WebPAC::Search::Estraier 0.05;
13  use File::Slurp;  use File::Slurp;
14    use Time::HiRes;
15    
16  =head1 NAME  =head1 NAME
17    
# Line 33  Configuration for hyperestraier in C<con Line 34  Configuration for hyperestraier in C<con
34    
35   # configuration for hyper estraier full text search engine   # configuration for hyper estraier full text search engine
36   hyperestraier:   hyperestraier:
37    url: 'http://localhost:1978/node/webpac2'    masterurl: 'http://localhost:1978/node/webpac2'
38      defaultnode: 'webpac2'
39      defaultdepth: 1
40    user: 'admin'    user: 'admin'
41    passwd: 'admin'    passwd: 'admin'
42    hits_on_page: 100    hits_on_page: 100
# Line 61  sub new { Line 64  sub new {
64          my $est_cfg = $c->config->{hyperestraier};          my $est_cfg = $c->config->{hyperestraier};
65          $est_cfg->{'log'} = $log;          $est_cfg->{'log'} = $log;
66    
67          $est_cfg->{encoding} = $est_cfg->{catalyst_encoding};          $est_cfg->{encoding} = $est_cfg->{catalyst_encoding} || $c->config->{catalyst_encoding} or $c->log->fatal("can't find catalyst_encoding");
68    
69          $log->debug("using config:" . Dumper($est_cfg) );          $log->debug("using config:" . Dumper($est_cfg) );
70    
71            if (! $est_cfg->{database}) {
72                    my $defaultnode = $est_cfg->{defaultnode} || $log->logdie("can't find defaultnode in estraier configuration");
73                    $log->info("using default node $defaultnode");
74                    $est_cfg->{database} = $defaultnode;
75            }
76    
77          $self->{est} = new WebPAC::Search::Estraier( %{ $est_cfg } );          $self->{est} = new WebPAC::Search::Estraier( %{ $est_cfg } );
78    
79          my $db_path = $c->config->{webpac}->{db_path};          # save config parametars in object
80          my $template_path = $c->config->{webpac}->{template_path};          foreach my $f (qw/db_path template_path hits_on_page webpac_encoding out_encoding defaultdepth/) {
81          $self->{template_path} = $template_path;                  $self->{$f} = $c->config->{hyperestraier}->{$f} ||
82                            $c->config->{webpac}->{$f};
83                    $log->debug("self->{$f} = " . $self->{$f});
84            }
85            my $db_path = $self->{db_path};
86            my $template_path = $self->{template_path};
87    
88          $log->debug("using db path '$db_path', template path '$template_path'");          $log->debug("using db path '$db_path', template path '$template_path'");
89    
90          $self->{db} = new WebPAC::DB(          $self->{db} = new WebPAC::Store(
91                  path => $db_path,                  path => $db_path,
92                  read_only => 1,                  read_only => 1,
93                    database => $est_cfg->{database},
94          );          );
95    
96          $self->{out} = new WebPAC::Output::TT(          $self->{out} = new WebPAC::Output::TT(
# Line 98  sub new { Line 113  sub new {
113                  "'"                  "'"
114          );          );
115    
116    
117          return $self;          return $self;
118    
119  }  }
120    
 =head2 iconv_on_save  
121    
122    my $out = $m->iconv_on_save( $content );  =head2 search
123    
124  Convert data saved to disk in Webpac encoding.    my $m->search(
125            phrase => 'query phrase',
126            add_attr => \@add_attr
127            get_attr => [ '@uri' ],
128            max => 42,
129            template => 'result_template.tt',
130            depth => 1,
131      );
132    
133    All fields are standard C<WebPAC::Search::Estraier> parametars except
134    C<template> which will (if specified) return results in HTML using
135    selected template.
136    
137  =cut  =cut
138    
139  sub iconv_on_save {  sub search {
140          my $self = shift;          my $self = shift;
141    
142          $self->{iconv_save} ||= new Text::Iconv(          my $args = {@_};
                 $self->config->{webpac}->{out_encoding},  
                 $self->config->{webpac}->{webpac_encoding},  
         );  
143    
144          $self->{iconv_save}->convert( @_ );          my $log = $self->{log};
 }  
145    
146            $log->debug("args: " . Dumper( $args ));
147    
148  =head2 search          my $query = $args->{phrase} || $log->warn("no query phrase") && return;
149    
150    my $m->search( 'query phrase', 'result_template.tt', \@add_attr );          $log->debug("search model query: '$query'");
151            if ($args->{add_attr}) {
152                    $log->debug(" + add_attr: " .
153                            join("','", @{ $args->{add_attr} })
154                    );
155            }
156    
157  =cut          my $template_filename = $args->{template} || $self->{template};
158    
159  sub search {          $args->{max} ||= $self->{'hits_on_page'};
160          my ( $self, $query, $template, $add_attr ) = @_;          if (! $args->{max}) {
161                    $args->{max} = 10;
162                    $log->warn("max not set when calling model. Using default of 10");
163            }
164    
165          my $log = $self->{log};          my $times;      # store some times for benchmarking
166    
167          $log->debug("search model query: '$query', add_attr: '" . join("','", @{$add_attr}) . "'");          my $t = time();
168    
169          my $template_filename = $template || $self->{template};          # transfer depth of search
170            if (! $args->{depth}) {
171                    my $default = $self->{defaultdepth} || $log->logdie("can't find defaultdepth in estraier configuration");
172                    $args->{depth} = $default;
173                    $log->warn("using default search depth $default");
174            }
175    
176          my @results = $self->{est}->search(          my @results = $self->{est}->search( %{ $args } );
177                  phrase => $query,  
178                  get_attr => [ '@uri' ],          $times->{est} += time() - $t;
179                  max => $self->{est}->{hits_on_page} || 30,  
180                  add_attr => $add_attr,          my $hits = $#results + 1;
181          );  
182            $log->debug( sprintf("search took %.2fs and returned $hits hits.", $times->{est}) );
183    
184          $log->debug("loading " . ($#results + 1) . " results");          # just return results?
185            return @results unless ($args->{'template'});
186    
187            #
188            # construct HTML results
189            #
190    
191          my @html_results;          my @html_results;
192    
193          for my $i ( 0 .. $#results ) {          for my $i ( 0 .. $#results ) {
194    
195                  my $mfn = $1 if ( $results[$i]->{'@uri'} =~ m#/(\d+)$#);                  my ($database, $prefix, $id);
196                    if ( $results[$i]->{'@uri'} =~ m!/([^/]+)/([^/]+)#(\d+)$!) {
197                            ($database, $prefix,$id) = ($1,$2,$3);
198                    } else {
199                            $log->warn("can't decode prefix#id from " .  $results[$i]->{'@uri'});
200                            next;
201                    }
202    
203                    #$log->debug("load_ds( id => $id, prefix => '$prefix' )");
204    
205                    $t = time();
206    
207                    my $ds = $self->{db}->load_ds( database => $database, prefix => $prefix, id => $id );
208                    if (! $ds) {
209                            $log->error("can't load_ds( ${database}/${prefix}#${id} )");
210                            next;
211                    }
212    
213                  #$log->debug("load_ds( $mfn )");                  $times->{db} += time() - $t;
214    
                 my $ds = $self->{db}->load_ds( $mfn ) || $log->error("can't load_ds( $mfn )") && next;  
           
215                  #$log->debug( "ds = " . Dumper( \@html_results ) );                  #$log->debug( "ds = " . Dumper( \@html_results ) );
216    
217                    $t = time();
218    
219                  my $html = $self->{out}->apply(                  my $html = $self->{out}->apply(
220                          template => $template_filename,                          template => $template_filename,
221                          data => $ds,                          data => $ds,
222                            record_uri => "${database}/${prefix}/${id}",
223                  );                  );
224    
225                    $times->{out} += time() - $t;
226    
227                    $t = time();
228    
229                  $html = $self->{iconv}->convert( $html ) || $log->error("can't convert: $html");                  $html = $self->{iconv}->convert( $html ) || $log->error("can't convert: $html");
230    
231                    $times->{iconv} += time() - $t;
232    
233                  push @html_results, $html;                  push @html_results, $html;
234    
235          }          }
236    
237          #$log->debug( '@html_results = ' . Dumper( \@html_results ) );          #$log->debug( '@html_results = ' . Dumper( \@html_results ) );
238    
239            $log->debug( sprintf(
240                    "time spent: db = %.2f, out = %.2f, iconv = %.2f",
241                    $times->{db}, $times->{out}, $times->{iconv},
242            ) );
243    
244          return \@html_results;          return \@html_results;
245  }  }
246    
247    =head2 record
248    
249      my $html = $m->record(
250            mfn => 42,
251            template => 'foo.tt',
252      );
253    
254    This will load one record, convert it to html using C<template> and return
255    it.
256    
257    =cut
258    
259    sub record {
260            my $self = shift;
261    
262            my $args = {@_};
263            my $log = $self->{log};
264            $log->debug("args: " . Dumper( $args ));
265    
266            foreach my $f (qw/record_uri template/) {
267                    $log->fatal("need $f") unless ($args->{$f});
268            }
269    
270            my ($database, $prefix, $id);
271    
272            if ($args->{record_uri} =~ m#^([^/]+)/([^/]+)/([^/]+)$#) {
273                    ($database, $prefix, $id) = ($1,$2,$3);
274            } else {
275                    $log->error("can't parse $args->{record_uri} into prefix, database and uri");
276                    return;
277            }
278    
279            my $ds = $self->{db}->load_ds( id => $id, prefix => $prefix, database => $database );
280            if (! $ds) {
281                    $log->error("can't load_ds( $database/$prefix/$id )");
282                    return;
283            }
284    
285            my $html = $self->{out}->apply(
286                    template => $args->{template},
287                    data => $ds,
288                    record_uri => $args->{record_uri},
289            );
290    
291            $html = $self->{iconv}->convert( $html ) || $log->error("can't convert: $html");
292    
293            return $html;
294    }
295    
296  =head2 save_html  =head2 save_html
297    
298    $m->save_html( '/full/path/to/file', $content );    $m->save_html( '/full/path/to/file', $content );
299    
300  It will use C<iconv_on_save> to convert content encoding back to  It will use C<Text::Iconv> to convert content encoding back to
301  Webpac codepage, recode JavaScript Unicode entities (%u1234),  Webpac codepage, recode JavaScript Unicode entities (%u1234),
302  strip extra newlines at beginning and end, and save to  strip extra newlines at beginning and end, and save to
303  C</full/path/to/file.new> and if that succeeds, just rename  C</full/path/to/file.new> and if that succeeds, just rename
# Line 189  it over original file which should be at Line 308  it over original file which should be at
308  sub save_html {  sub save_html {
309          my ($self, $path, $content) = @_;          my ($self, $path, $content) = @_;
310    
         $content = $self->iconv_on_save( $content ) || die "no content?";  
   
311          sub _conv_js {          sub _conv_js {
312                  my $t = shift || return;                  my $t = shift || return;
313                  return $self->{iconv}->convert(chr(hex($t)));                  return $self->{iconv}->convert(chr(hex($t)));
# Line 199  sub save_html { Line 316  sub save_html {
316          $content =~ s/^[\n\r]+//s;          $content =~ s/^[\n\r]+//s;
317          $content =~ s/[\n\r]+$/\n/s;          $content =~ s/[\n\r]+$/\n/s;
318    
319          write_file($path . '.new', $content) || die "can't save ${path}.new $!";          my ($from, $to) = (
320                    $self->{out_encoding},
321                    $self->{webpac_encoding},
322            );
323    
324            $self->{log}->debug("using iconv to convert from $from to $to encoding");
325    
326            my $iconv_on_save = new Text::Iconv($from, $to)
327                    || $self->{log}->fatal("can't create iconv for saving");
328    
329            $content = $iconv_on_save->convert( $content ) || die "no content?";
330    
331            write_file($path . '.new', {binmode => ':raw' }, $content) || die "can't save ${path}.new $!";
332          rename $path . '.new', $path || die "can't rename to $path: $!";          rename $path . '.new', $path || die "can't rename to $path: $!";
333  }  }
334    
# Line 218  sub load_html { Line 347  sub load_html {
347    
348          die "no path?" unless ($path);          die "no path?" unless ($path);
349    
350          my $content = read_file($path) || die "can't read $path: $!";          my $content = read_file($path, {binmode => ':raw' }) || die "can't read $path: $!";
351          #$content = $q->escapeHTML($iconv_utf8->convert($content));          #$content = $q->escapeHTML($iconv_utf8->convert($content));
352          $content = $self->{iconv}->convert($content);          $content = $self->{iconv}->convert($content);
353    
Legend:
Removed from v.143  
changed lines
  Added in v.271
  ViewVC Help
Powered by ViewVC 1.1.26