/[Grep]/lib/Grep/Source.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 /lib/Grep/Source.pm

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

revision 85 by dpavlin, Fri Feb 23 20:47:08 2007 UTC revision 86 by dpavlin, Fri Feb 23 21:16:44 2007 UTC
# Line 194  sub scrape { Line 194  sub scrape {
194          die "no uri" unless ($uri);          die "no uri" unless ($uri);
195          die "feed is not a Grep::Model::Feed but ", ref $feed unless $feed->isa('Grep::Model::Feed');          die "feed is not a Grep::Model::Feed but ", ref $feed unless $feed->isa('Grep::Model::Feed');
196    
         sub save_html {  
                 my ( $file, $content ) = @_;  
                 if ( -w '/tmp/grep' ) {  
                         open(my $f, '>', "/tmp/grep/${file}.html") or die "can't open $file: $!";  
                         print $f $content or die "can't write to $file: $!";  
                         close $f or die "can't close $file: $!";  
                 }  
         }  
   
197          my $mech = WWW::Mechanize->new();          my $mech = WWW::Mechanize->new();
198    
199          $mech->get( $uri );          $mech->get( $uri );
200    
201          save_html( 'get', $mech->content );          $self->save( 'get.html', $mech->content );
202    
203          if ( $args->{submit_form} ) {          if ( $args->{submit_form} ) {
204                  warn "submit form on $uri\n";                  warn "submit form on $uri\n";
205                  $mech->submit_form( %{ $args->{submit_form} } ) or die "can't submit form";                  $mech->submit_form( %{ $args->{submit_form} } ) or die "can't submit form";
206                  save_html( 'submit', $mech->content );                  $self->save( 'submit.html', $mech->content );
207          }          }
208    
209          warn "parse result page\n";          warn "parse result page\n";
# Line 250  sub scrape { Line 241  sub scrape {
241                          warn "fetching page: ",$a->as_text," from $page_uri\n";                          warn "fetching page: ",$a->as_text," from $page_uri\n";
242                          if ( $mech->follow_link( url => $a->attr('href') ) ) {                          if ( $mech->follow_link( url => $a->attr('href') ) ) {
243    
244                                  save_html( "page-${nr}", $mech->content );                                  $self->save( "page-${nr}.html", $mech->content );
245    
246                                  my $page_tree = HTML::TreeBuilder->new or die "can't create page tree";                                  my $page_tree = HTML::TreeBuilder->new or die "can't create page tree";
247                                  $page_tree->parse( $mech->content ) or die "can't parse page at $page_uri";                                  $page_tree->parse( $mech->content ) or die "can't parse page at $page_uri";
# Line 288  sub scrape { Line 279  sub scrape {
279    
280  }  }
281    
282    =head2 save
283    
284      save( 'name', $content );
285    
286    Save dumps into C</tmp/grep> if writable
287    
288    =cut
289    
290    sub save {
291            my $self = shift;
292            my ( $file, $content ) = @_;
293            if ( -w '/tmp/grep' ) {
294                    open(my $f, '>', "/tmp/grep/$file") or die "can't open $file: $!";
295                    print $f $content or die "can't write to $file: $!";
296                    close $f or die "can't close $file: $!";
297                    Jifty->log->debug("saved $file ",length($content)," bytes");
298            }
299    }
300    
301  1;  1;

Legend:
Removed from v.85  
changed lines
  Added in v.86

  ViewVC Help
Powered by ViewVC 1.1.26