/[webpac2]/trunk/bin/isi-download-results.pl
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/bin/isi-download-results.pl

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

revision 1292 by dpavlin, Sat Sep 19 20:35:43 2009 UTC revision 1332 by dpavlin, Sun Apr 18 11:04:49 2010 UTC
# Line 3  Line 3 
3  use warnings;  use warnings;
4  use strict;  use strict;
5    
6    use WWW::Mechanize;
7    use Data::Dump qw(dump);
8    use File::Path;
9    use Text::Unaccent;
10    
11  # Advanced search syntax:  # Advanced search syntax:
12  # http://images.isiknowledge.com/WOK46/help/WOS/h_advanced_examples.html  # http://images.isiknowledge.com/WOK46/help/WOS/h_advanced_examples.html
13    
14  our $q = 'AD=Croatia';  our $q = 'AD=Croatia';
15  my $range_size = 500;  my $range_size = 500;
16  my $overlap    = 10; # between previous and this range  my $overlap    = 3; # between previous and this range
17    my $skip_results = 0;
18  $q = 'TS=psychology AND AD=Croatia' if @ARGV; # FIXME debug  my $cites_by_year = 0;
19    
20  my $max_cites = 5000; # ISI limit to get cites  my $max_cites = 5000; # ISI limit to get cites
 our @ranges;  
21    
22  use WWW::Mechanize;  if ( 0 ) {
23  use Data::Dump qw(dump);          $q = 'TS=psychology AND AD=Croatia';
24  use File::Path;          $range_size = 50;
25            $overlap    = 0;
26            $max_cites  = 50;
27    }
28    
29    $q = unac_string( 'utf-8', join(' ', @ARGV) ) if @ARGV;
30    
31  our $mech = WWW::Mechanize->new(  our $mech = WWW::Mechanize->new(
32          autocheck => 1,          autocheck => 1,
# Line 25  our $mech = WWW::Mechanize->new( Line 34  our $mech = WWW::Mechanize->new(
34  );  );
35    
36  our $step = 0;  our $step = 0;
37    our @ranges;
38    
39  my $dir = '/tmp/isi/';  my $dir = '/tmp/isi/';
40  rmtree $dir if -e $dir;  #rmtree $dir if -e $dir;
41  mkdir $dir;  mkdir $dir unless -d $dir;
42    
43  sub save_mech {  sub save_mech {
44          my $path = shift;          my $path = shift;
# Line 164  sub years { Line 174  sub years {
174    
175          my @y = sort keys %$years;          my @y = sort keys %$years;
176    
177            @ranges = ();
178    
179            if ( $cites_by_year ) {
180                    push @ranges, [ $_ ] foreach @y;
181                    warn "# cites_by_year ranges ", dump @ranges;
182                    return;
183            }
184    
185          my $y = shift @y;          my $y = shift @y;
186          my $size = $years->{$y};          my $size = $years->{$y};
187    
         @ranges = ();  
188          my $cites_range;          my $cites_range;
189          $cites_range = [$y] if $y;          $cites_range = [$y] if $y;
190    
# Line 196  sub years { Line 213  sub years {
213    
214  search;  search;
215  years;  years;
216  get_results;  get_results unless $skip_results;
217    
218    
219  citations;  citations;
# Line 206  do { Line 223  do {
223          if ( @ranges ) {          if ( @ranges ) {
224                  $part .= $ranges[0]->[0] . '.';                  $part .= $ranges[0]->[0] . '.';
225                  search;                  search;
226                    citations;
227          }          }
228          $part .= 'citing';          $part .= 'citing';
229          get_results $part;          get_results $part;
Legend:
Removed from v.1292  
changed lines
  Added in v.1332
  ViewVC Help
Powered by ViewVC 1.1.26