/[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 100 by dpavlin, Tue Nov 22 14:45:17 2005 UTC revision 167 by dpavlin, Sat Nov 26 21:11:41 2005 UTC
# Line 5  use warnings; Line 5  use warnings;
5  use lib '/data/webpac2/lib';  use lib '/data/webpac2/lib';
6  use base qw/  use base qw/
7          Catalyst::Model          Catalyst::Model
         WebPAC::Search::Estraier  
8  /;  /;
9  use Data::Dumper;  use Data::Dumper;
10  use WebPAC::DB;  use WebPAC::DB;
11  use WebPAC::Output::TT;  use WebPAC::Output::TT;
12    use WebPAC::Search::Estraier 0.02;
13    use File::Slurp;
14    use Time::HiRes;
15    
16  =head1 NAME  =head1 NAME
17    
# Line 35  Configuration for hyperestraier in C<con Line 37  Configuration for hyperestraier in C<con
37    url: 'http://localhost:1978/node/webpac2'    url: 'http://localhost:1978/node/webpac2'
38    user: 'admin'    user: 'admin'
39    passwd: 'admin'    passwd: 'admin'
40      hits_on_page: 100
41    
42   webpac:   webpac:
43    db_path: '/data/webpac2/db'    db_path: '/data/webpac2/db'
# Line 59  sub new { Line 62  sub new {
62          my $est_cfg = $c->config->{hyperestraier};          my $est_cfg = $c->config->{hyperestraier};
63          $est_cfg->{'log'} = $log;          $est_cfg->{'log'} = $log;
64    
65            $est_cfg->{encoding} = $est_cfg->{catalyst_encoding};
66    
67          $log->debug("using config:" . Dumper($est_cfg) );          $log->debug("using config:" . Dumper($est_cfg) );
68    
69          $self->{est} = new WebPAC::Search::Estraier( %{ $est_cfg } );          $self->{est} = new WebPAC::Search::Estraier( %{ $est_cfg } );
70    
71          my $db_path = $c->config->{webpac}->{db_path};          # save config parametars in object
72          my $template_path = $c->config->{webpac}->{template_path};          foreach my $f (qw/db_path template_path hits_on_page/) {
73                    $self->{$f} = $c->config->{hyperestraier}->{$f} ||
74                            $c->config->{webpac}->{$f};
75                    $log->debug("self->{$f} = " . $self->{$f});
76            }
77            my $db_path = $self->{db_path};
78            my $template_path = $self->{template_path};
79    
80          $log->debug("using db path '$db_path', template path '$template_path'");          $log->debug("using db path '$db_path', template path '$template_path'");
81    
# Line 78  sub new { Line 89  sub new {
89                  filters => { foo => sub { shift } },                  filters => { foo => sub { shift } },
90          );          );
91    
92            # default template from config.yaml
93          $self->{template} ||= $c->config->{webpac}->{template};          $self->{template} ||= $c->config->{webpac}->{template};
94    
95          $self->{iconv} = new Text::Iconv(          $self->{iconv} = new Text::Iconv(
# Line 92  sub new { Line 104  sub new {
104                  "'"                  "'"
105          );          );
106    
107    
108          return $self;          return $self;
109    
110  }  }
111    
112    =head2 iconv_on_save
113    
114      my $out = $m->iconv_on_save( $content );
115    
116    Convert data saved to disk in Webpac encoding.
117    
118    =cut
119    
120    sub iconv_on_save {
121            my $self = shift;
122    
123            $self->{iconv_save} ||= new Text::Iconv(
124                    $self->config->{webpac}->{out_encoding},
125                    $self->config->{webpac}->{webpac_encoding},
126            );
127    
128            $self->{iconv_save}->convert( @_ );
129    }
130    
131    
132    =head2 search
133    
134      my $m->search(
135            phrase => 'query phrase',
136            add_attr => \@add_attr
137            get_attr => [ '@uri' ],
138            max => 42,
139            template => 'result_template.tt',
140      );
141    
142    All fields are standard C<WebPAC::Search::Estraier> parametars except
143    C<template> which will (if specified) return results in HTML using
144    selected template.
145    
146    =cut
147    
148  sub search {  sub search {
149          my ( $self, $query ) = @_;          my $self = shift;
150    
151            my $args = {@_};
152    
153          my $log = $self->{log};          my $log = $self->{log};
154    
155          $log->debug("search model query: -->$query<--");          $log->debug("args: " . Dumper( $args ));
156    
157          my $template_filename = $self->{template};          my $query = $args->{phrase} || $log->warn("no query phrase") && return;
158    
159          my @results = $self->{est}->search(          $log->debug("search model query: '$query'");
160                  query => $query,          if ($args->{add_attr}) {
161                  attr => [ '@uri' ],                  $log->debug(" + add_attr: " .
162                  max => 100,                          join("','", @{ $args->{add_attr} })
163          );                  );
164            }
165    
166            my $template_filename = $args->{template} || $self->{template};
167    
168            $args->{max} ||= $self->{'hits_on_page'};
169            if (! $args->{max}) {
170                    $args->{max} = 10;
171                    $log->warn("max not set when calling model. Using default of 10");
172            }
173    
174            my $times;      # store some times for benchmarking
175    
176            my $t = time();
177    
178            my @results = $self->{est}->search( %{ $args } );
179    
180            $times->{est} += time() - $t;
181    
182            my $hits = $#results + 1;
183    
184          $log->debug("loading " . ($#results + 1) . " results");          $log->debug( sprintf("search took %.2fs and returned $hits hits.", $times->{est}) );
185    
186            # just return results?
187            return @results unless ($args->{'template'});
188    
189            #
190            # construct HTML results
191            #
192    
193          my @html_results;          my @html_results;
194    
# Line 119  sub search { Line 196  sub search {
196    
197                  my $mfn = $1 if ( $results[$i]->{'@uri'} =~ m#/(\d+)$#);                  my $mfn = $1 if ( $results[$i]->{'@uri'} =~ m#/(\d+)$#);
198    
199                  # $log->debug("load_ds( $mfn )");                  #$log->debug("load_ds( $mfn )");
200    
201                    $t = time();
202    
203                    my $ds = $self->{db}->load_ds( $mfn ) || $log->error("can't load_ds( $mfn )") && next;
204    
205                    $times->{db} += time() - $t;
206    
207                    #$log->debug( "ds = " . Dumper( \@html_results ) );
208    
209                    $t = time();
210    
                 my $ds = $self->{db}->load_ds( $mfn ) || next;  
           
211                  my $html = $self->{out}->apply(                  my $html = $self->{out}->apply(
212                          template => $template_filename,                          template => $template_filename,
213                          data => $ds,                          data => $ds,
214                  );                  );
215    
216                    $times->{out} += time() - $t;
217    
218                    $t = time();
219    
220                  $html = $self->{iconv}->convert( $html ) || $log->error("can't convert: $html");                  $html = $self->{iconv}->convert( $html ) || $log->error("can't convert: $html");
221    
222                    $times->{iconv} += time() - $t;
223    
224                  push @html_results, $html;                  push @html_results, $html;
225    
226          }          }
227    
228            #$log->debug( '@html_results = ' . Dumper( \@html_results ) );
229    
230            $log->debug( sprintf(
231                    "time spent: db = %.2f, out = %.2f, iconv = %.2f",
232                    $times->{db}, $times->{out}, $times->{iconv},
233            ) );
234    
235          return \@html_results;          return \@html_results;
236  }  }
237    
238    =head2 record
239    
240      my $html = $m->record(
241            mfn => 42,
242            template => 'foo.tt',
243      );
244    
245    This will load one record, convert it to html using C<template> and return
246    it.
247    
248    =cut
249    
250    sub record {
251            my $self = shift;
252    
253            my $args = {@_};
254            my $log = $self->{log};
255            $log->debug("args: " . Dumper( $args ));
256    
257            foreach my $f (qw/mfn template/) {
258                    $log->die("need $f") unless ($args->{$f});
259            }
260    
261            my $mfn = $args->{mfn};
262    
263            my $ds = $self->{db}->load_ds( $mfn ) || $log->error("can't load_ds( $mfn )") && next;
264    
265            my $html = $self->{out}->apply(
266                    template => $args->{template},
267                    data => $ds,
268            );
269    
270            $html = $self->{iconv}->convert( $html ) || $log->error("can't convert: $html");
271    
272            return $html;
273    }
274    
275    =head2 save_html
276    
277              $m->save_html( '/full/path/to/file', $content );
278    
279    It will use C<iconv_on_save> to convert content encoding back to
280    Webpac codepage, recode JavaScript Unicode entities (%u1234),
281    strip extra newlines at beginning and end, and save to
282    C</full/path/to/file.new> and if that succeeds, just rename
283    it over original file which should be atomic on filesystem level.
284    
285    =cut
286    
287    sub save_html {
288            my ($self, $path, $content) = @_;
289    
290            $content = $self->iconv_on_save( $content ) || die "no content?";
291    
292            sub _conv_js {
293                    my $t = shift || return;
294                    return $self->{iconv}->convert(chr(hex($t)));
295            }
296            $content =~ s/%u([a-fA-F0-9]{4})/_conv_js($1)/gex;
297            $content =~ s/^[\n\r]+//s;
298            $content =~ s/[\n\r]+$/\n/s;
299    
300            write_file($path . '.new', $content) || die "can't save ${path}.new $!";
301            rename $path . '.new', $path || die "can't rename to $path: $!";
302    }
303    
304    =head2 load_html
305    
306      my $html = $m->load_html('/full/path/to/file');
307    
308    This will convert file from Webpac encoding to Catalyst and
309    convert that data to escaped HTML (for sending into
310    C<< <textarea/> >> tags in html.
311    
312    =cut
313    
314    sub load_html {
315            my ($self, $path) = @_;
316    
317            die "no path?" unless ($path);
318    
319            my $content = read_file($path) || die "can't read $path: $!";
320            #$content = $q->escapeHTML($iconv_utf8->convert($content));
321            $content = $self->{iconv}->convert($content);
322    
323            return $content;
324    }
325    
326  =head1 AUTHOR  =head1 AUTHOR
327    
Legend:
Removed from v.100  
changed lines
  Added in v.167
  ViewVC Help
Powered by ViewVC 1.1.26