/[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 142 by dpavlin, Fri Nov 25 00:23:33 2005 UTC revision 222 by dpavlin, Mon Dec 5 19:15:01 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 0.03;
11  use WebPAC::Output::TT;  use WebPAC::Output::TT 0.02;
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
43    
44   webpac:   webpac:
45    db_path: '/data/webpac2/db'    db_path: '/data/webpac2/db'
# Line 64  sub new { Line 68  sub new {
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          );          );
# Line 97  sub new { Line 112  sub new {
112                  "'"                  "'"
113          );          );
114    
115    
116          return $self;          return $self;
117    
118  }  }
119    
 =head2 iconv_on_save  
120    
121    my $out = $m->iconv_on_save( $content );  =head2 search
122    
123  Convert data saved to disk in Webpac encoding.    my $m->search(
124            phrase => 'query phrase',
125            add_attr => \@add_attr
126            get_attr => [ '@uri' ],
127            max => 42,
128            template => 'result_template.tt',
129            depth => 1,
130      );
131    
132    All fields are standard C<WebPAC::Search::Estraier> parametars except
133    C<template> which will (if specified) return results in HTML using
134    selected template.
135    
136  =cut  =cut
137    
138  sub iconv_on_save {  sub search {
139          my $self = shift;          my $self = shift;
140    
141          $self->{iconv_save} ||= new Text::Iconv(          my $args = {@_};
                 $self->config->{webpac}->{out_encoding},  
                 $self->config->{webpac}->{webpac_encoding},  
         );  
142    
143          $self->{iconv_save}->convert( @_ );          my $log = $self->{log};
 }  
144    
145            $log->debug("args: " . Dumper( $args ));
146    
147  =head2 search          my $query = $args->{phrase} || $log->warn("no query phrase") && return;
148    
149    my $m->search( 'query phrase', 'result_template.tt', \@add_attr );          $log->debug("search model query: '$query'");
150            if ($args->{add_attr}) {
151                    $log->debug(" + add_attr: " .
152                            join("','", @{ $args->{add_attr} })
153                    );
154            }
155    
156  =cut          my $template_filename = $args->{template} || $self->{template};
157    
158  sub search {          $args->{max} ||= $self->{'hits_on_page'};
159          my ( $self, $query, $template, $add_attr ) = @_;          if (! $args->{max}) {
160                    $args->{max} = 10;
161                    $log->warn("max not set when calling model. Using default of 10");
162            }
163    
164          my $log = $self->{log};          my $times;      # store some times for benchmarking
165    
166          $log->debug("search model query: '$query', add_attr: '" . join("','", @{$add_attr}) . "'");          my $t = time();
167    
168          my $template_filename = $template || $self->{template};          # transfer depth of search
169            if (! $args->{depth}) {
170                    my $default = $self->{defaultdepth} || $log->logdie("can't find defaultdepth in estraier configuration");
171                    $args->{depth} = $default;
172                    $log->warn("using default search depth $default");
173            }
174    
175          my @results = $self->{est}->search(          my @results = $self->{est}->search( %{ $args } );
176                  phrase => $query,  
177                  get_attr => [ '@uri' ],          $times->{est} += time() - $t;
178                  max => 100,  
179                  add_attr => $add_attr,          my $hits = $#results + 1;
180          );  
181            $log->debug( sprintf("search took %.2fs and returned $hits hits.", $times->{est}) );
182    
183          $log->debug("loading " . ($#results + 1) . " results");          # just return results?
184            return @results unless ($args->{'template'});
185    
186            #
187            # construct HTML results
188            #
189    
190          my @html_results;          my @html_results;
191    
# Line 153  sub search { Line 195  sub search {
195    
196                  #$log->debug("load_ds( $mfn )");                  #$log->debug("load_ds( $mfn )");
197    
198                    $t = time();
199    
200                  my $ds = $self->{db}->load_ds( $mfn ) || $log->error("can't load_ds( $mfn )") && next;                  my $ds = $self->{db}->load_ds( $mfn ) || $log->error("can't load_ds( $mfn )") && next;
201            
202                    $times->{db} += time() - $t;
203    
204                  #$log->debug( "ds = " . Dumper( \@html_results ) );                  #$log->debug( "ds = " . Dumper( \@html_results ) );
205    
206                    $t = time();
207    
208                  my $html = $self->{out}->apply(                  my $html = $self->{out}->apply(
209                          template => $template_filename,                          template => $template_filename,
210                          data => $ds,                          data => $ds,
211                  );                  );
212    
213                    $times->{out} += time() - $t;
214    
215                    $t = time();
216    
217                  $html = $self->{iconv}->convert( $html ) || $log->error("can't convert: $html");                  $html = $self->{iconv}->convert( $html ) || $log->error("can't convert: $html");
218    
219                    $times->{iconv} += time() - $t;
220    
221                  push @html_results, $html;                  push @html_results, $html;
222    
223          }          }
224    
225          #$log->debug( '@html_results = ' . Dumper( \@html_results ) );          #$log->debug( '@html_results = ' . Dumper( \@html_results ) );
226    
227            $log->debug( sprintf(
228                    "time spent: db = %.2f, out = %.2f, iconv = %.2f",
229                    $times->{db}, $times->{out}, $times->{iconv},
230            ) );
231    
232          return \@html_results;          return \@html_results;
233  }  }
234    
235    =head2 record
236    
237      my $html = $m->record(
238            mfn => 42,
239            template => 'foo.tt',
240      );
241    
242    This will load one record, convert it to html using C<template> and return
243    it.
244    
245    =cut
246    
247    sub record {
248            my $self = shift;
249    
250            my $args = {@_};
251            my $log = $self->{log};
252            $log->debug("args: " . Dumper( $args ));
253    
254            foreach my $f (qw/mfn template/) {
255                    $log->die("need $f") unless ($args->{$f});
256            }
257    
258            my $mfn = $args->{mfn};
259    
260            my $ds = $self->{db}->load_ds( $mfn ) || $log->error("can't load_ds( $mfn )") && next;
261    
262            my $html = $self->{out}->apply(
263                    template => $args->{template},
264                    data => $ds,
265            );
266    
267            $html = $self->{iconv}->convert( $html ) || $log->error("can't convert: $html");
268    
269            return $html;
270    }
271    
272  =head2 save_html  =head2 save_html
273    
274    $m->save_html( '/full/path/to/file', $content );    $m->save_html( '/full/path/to/file', $content );
275    
276  It will use C<iconv_on_save> to convert content encoding back to  It will use C<Text::Iconv> to convert content encoding back to
277  Webpac codepage, recode JavaScript Unicode entities (%u1234),  Webpac codepage, recode JavaScript Unicode entities (%u1234),
278  strip extra newlines at beginning and end, and save to  strip extra newlines at beginning and end, and save to
279  C</full/path/to/file.new> and if that succeeds, just rename  C</full/path/to/file.new> and if that succeeds, just rename
# Line 188  it over original file which should be at Line 284  it over original file which should be at
284  sub save_html {  sub save_html {
285          my ($self, $path, $content) = @_;          my ($self, $path, $content) = @_;
286    
         $content = $self->iconv_on_save( $content ) || die "no content?";  
   
287          sub _conv_js {          sub _conv_js {
288                  my $t = shift || return;                  my $t = shift || return;
289                  return $self->{iconv}->convert(chr(hex($t)));                  return $self->{iconv}->convert(chr(hex($t)));
# Line 198  sub save_html { Line 292  sub save_html {
292          $content =~ s/^[\n\r]+//s;          $content =~ s/^[\n\r]+//s;
293          $content =~ s/[\n\r]+$/\n/s;          $content =~ s/[\n\r]+$/\n/s;
294    
295          write_file($path . '.new', $content) || die "can't save ${path}.new $!";          my ($from, $to) = (
296                    $self->{out_encoding},
297                    $self->{webpac_encoding},
298            );
299    
300            $self->{log}->debug("using iconv to convert from $from to $to encoding");
301    
302            my $iconv_on_save = new Text::Iconv($from, $to)
303                    || $self->{log}->fatal("can't create iconv for saving");
304    
305            $content = $iconv_on_save->convert( $content ) || die "no content?";
306    
307            write_file($path . '.new', {binmode => ':raw' }, $content) || die "can't save ${path}.new $!";
308          rename $path . '.new', $path || die "can't rename to $path: $!";          rename $path . '.new', $path || die "can't rename to $path: $!";
309  }  }
310    
# Line 217  sub load_html { Line 323  sub load_html {
323    
324          die "no path?" unless ($path);          die "no path?" unless ($path);
325    
326          my $content = read_file($path) || die "can't read $path: $!";          my $content = read_file($path, {binmode => ':raw' }) || die "can't read $path: $!";
327          #$content = $q->escapeHTML($iconv_utf8->convert($content));          #$content = $q->escapeHTML($iconv_utf8->convert($content));
328          $content = $self->{iconv}->convert($content);          $content = $self->{iconv}->convert($content);
329    
Legend:
Removed from v.142  
changed lines
  Added in v.222
  ViewVC Help
Powered by ViewVC 1.1.26