| 1 |
#!/usr/bin/perl -w |
| 2 |
|
| 3 |
# This script will fatch list of articles on which you have access |
| 4 |
# (using IP authorisation) from ScienceDirect |
| 5 |
# |
| 6 |
# This version requires CSV dumps from ScienceDirect for Holdings data |
| 7 |
# and categories, but can output much more data about each record |
| 8 |
|
| 9 |
use LWP::UserAgent; |
| 10 |
use HTML::TreeBuilder; |
| 11 |
require Text::CSV; |
| 12 |
use Text::Unaccent; |
| 13 |
use strict; |
| 14 |
|
| 15 |
my $debug=1; |
| 16 |
|
| 17 |
my $file; |
| 18 |
|
| 19 |
# uncomment following line if you want to use file instead of http connection |
| 20 |
#$file="list.html"; |
| 21 |
|
| 22 |
# configure ScienceDirect CVS files location |
| 23 |
my $csv_dir="/data/isis_data/sciencedirect"; |
| 24 |
my $j_holdings="sd_JournalHoldingsRpt.txt"; |
| 25 |
my $j_category="sd_Journal_Category.txt"; |
| 26 |
|
| 27 |
# URL to list of subscribed journals |
| 28 |
|
| 29 |
my $url = 'http://www.sciencedirect.com/science?_ob=JournalListURL&_type=subscribed&_stype=title&subjColl=all&_auth=y&_update=y&_frameSeg=M&_title=all&_acct=C000050661&_version=1&_urlVersion=0&_userid=1034703&md5=6d4b6e263318a1d7d2a3b523d861f920'; |
| 30 |
|
| 31 |
my $html_codepage="iso-8859-1"; |
| 32 |
|
| 33 |
my $csv = Text::CSV->new(); |
| 34 |
my $journal; |
| 35 |
my $c_wo_h = 0; # category without holding record |
| 36 |
my $c_nr = 0; # number of categories assigned |
| 37 |
|
| 38 |
my $j_basic = 0; |
| 39 |
my $j_detailed = 0; |
| 40 |
|
| 41 |
print STDERR "unrolling $j_holdings\n"; |
| 42 |
|
| 43 |
sub nuc { |
| 44 |
# normalizing UC |
| 45 |
my $s=shift @_ || return ""; |
| 46 |
$s=unac_string($html_codepage,$s); |
| 47 |
$s=~s/[^\w]/ /g; |
| 48 |
$s=~s/ +/ /g; |
| 49 |
return uc($s); |
| 50 |
} |
| 51 |
|
| 52 |
open(H,"$csv_dir/$j_holdings") || die "can't open $csv_dir/$j_holdings: $!"; |
| 53 |
my $line = <H>; # skip header line |
| 54 |
while(<H>) { |
| 55 |
chomp; |
| 56 |
$csv->parse($_) || warn "can't parse '$_': ".$csv->error_input; |
| 57 |
my @data = $csv->fields; |
| 58 |
my $key = nuc($data[0]); |
| 59 |
push @data,""; # for categories later... |
| 60 |
$journal->{$key} = \@data; |
| 61 |
} |
| 62 |
close(H); |
| 63 |
|
| 64 |
print STDERR "unrolling $j_category\n"; |
| 65 |
|
| 66 |
open(C,"$csv_dir/$j_category") || die "can't open $csv_dir/$j_category: $!"; |
| 67 |
$line = <C>; # skip header line |
| 68 |
while(<C>) { |
| 69 |
chomp; |
| 70 |
$csv->parse($_) || warn "can't parse '$_': ".$csv->error_input; |
| 71 |
my @data = $csv->fields; |
| 72 |
my $key = nuc($data[1]); |
| 73 |
if (! $journal->{$key}) { |
| 74 |
$c_wo_h++; |
| 75 |
next; |
| 76 |
} |
| 77 |
|
| 78 |
foreach my $i (4, 6, 8, 10) { |
| 79 |
push @{$journal->{$key}},$data[$i] || ""; |
| 80 |
if ($data[$i]) { |
| 81 |
$c_nr++; |
| 82 |
} |
| 83 |
} |
| 84 |
} |
| 85 |
close(C); |
| 86 |
|
| 87 |
print STDERR "$c_nr categories assigned, $c_wo_h categories with holdings\n"; |
| 88 |
|
| 89 |
$debug++ if (lc($ARGV[0]) eq "-d"); |
| 90 |
|
| 91 |
|
| 92 |
my $res; |
| 93 |
if (! $file) { |
| 94 |
my $ua = new LWP::UserAgent; |
| 95 |
$ua->agent("Mjesec educational harvester -- contact mglavica\@ffzg.hr 0.0"); |
| 96 |
$ua->timeout(60); |
| 97 |
#$ua->env_proxy(); |
| 98 |
#$ua->proxy(['http', 'ftp'], 'http://proxy.carnet.hr:8001/'); |
| 99 |
|
| 100 |
print STDERR "getting '$url'...\n" if ($debug); |
| 101 |
my $req = HTTP::Request->new(GET => $url); |
| 102 |
|
| 103 |
$res = $ua->request($req); |
| 104 |
} elsif (! -e $file) { |
| 105 |
die "can't find feed file '$file'"; |
| 106 |
} |
| 107 |
|
| 108 |
if ($file || $res->is_success) { |
| 109 |
print STDERR "parsing html...\n" if ($debug); |
| 110 |
my $tree = HTML::TreeBuilder->new; |
| 111 |
if ($file) { |
| 112 |
$tree->parse_file("list.html"); |
| 113 |
} else { |
| 114 |
$tree->parse($res->content); |
| 115 |
} |
| 116 |
|
| 117 |
foreach my $tr ($tree->look_down('_tag', 'tr')) { |
| 118 |
my $link; |
| 119 |
foreach my $link ($tr->look_down('_tag','a')) { |
| 120 |
if ($link->attr('href') =~ m{/science\?_ob=JournalURL}) { |
| 121 |
my $j=nuc($link->as_text); |
| 122 |
if ($journal->{$j}) { |
| 123 |
my $i=0; |
| 124 |
foreach my $line (@{$journal->{$j}}) { |
| 125 |
print $i++,": $line\n"; |
| 126 |
} |
| 127 |
$j_detailed++; |
| 128 |
} else { |
| 129 |
print "0: ",$link->as_text."\n"; |
| 130 |
print "7: http://www.sciencedirect.com",$link->attr('href')."\n"; |
| 131 |
$j_basic++; |
| 132 |
print STDERR "can't find details for $j\n" if ($debug); |
| 133 |
} |
| 134 |
|
| 135 |
print "\n"; |
| 136 |
} |
| 137 |
} |
| 138 |
} |
| 139 |
|
| 140 |
$tree->delete; # clear memory! |
| 141 |
|
| 142 |
} else { |
| 143 |
warn "can't fetch web page from '$url'"; |
| 144 |
} |
| 145 |
|
| 146 |
print STDERR "Processed ",($j_basic+$j_detailed)," journals, $j_basic with basic data and $j_detailed detailed\n"; |
| 147 |
|
Parent Directory
| 
Revision Log