[BACK]Return to harvester.pl CVS log [TXT] [DIR] Up to [LON-CAPA] / nsdl / nsdlloncapaorg

Diff for /nsdl/nsdlloncapaorg/harvester.pl between versions 1.1 and 1.9

version 1.1, 2003/07/28 14:27:05 version 1.9, 2006/05/10 16:28:56
Line 11 Line 11
 use strict;  use strict;
 use LWP::UserAgent;  use LWP::UserAgent;
 use Getopt::Std;  use Getopt::Std;
   use Digest::MD5 qw(md5_hex);
   use IO::File;
   
 use DBI;  my $basepath='/home/httpd/cgi-bin/OAI-XMLFile/XMLFile/nsdlexport/data';
 use DBD::ODBC;  
   
 require OAIcataloging_v2;  
   
 # -u flag specifies [u]pdate database; otherwise output to STDOUT  
   
 my $usage = << "EOT";  
 Usage: lon-capa.pl -u  
   
     -u (U)pdate the database  
   
     Without -u it simply prints SQL UPDATE statements to STDOUT  
 EOT  
   
 my %args;  
 getopts('u', \%args) || die $usage;  
   
 my $useDatabase = 1 if ($args{'u'});  
   
 #my $DBI_DSN='dbi:ODBC:needs2_mel_needs_3_1_dev.odbc';  
 my $DBI_DSN='dbi:ODBC:needs2_mel_needs_3_1.odbc';  
 my $DBI_USER='autocataloger';  
 my $DBI_PWD='regolatacotua';  
 my $dbh;  
   
 my $pub_month;  my $pub_month;
 my $pub_year;  my $pub_year;
Line 50  my $content_regex = 'File Not Found'; Line 28  my $content_regex = 'File Not Found';
 # Configuration  # Configuration
   
 my $debug = 0;  my $debug = 0;
 my $url = 'http://data.lite.msu.edu/cgi-bin/metadata_harvest.pl';  
   # Stats
   my %allstats=();
   my %filterstats=();
   my %knockout=();
   my %knockoutlang=();
   
 # The list of servers is from the LON-CAPA CVS repository in /loncapa/loncom/production_hosts.tab  # The list of servers is from the LON-CAPA CVS repository in /loncapa/loncom/production_hosts.tab
 my @servers = ( 'newscience.westshore.cc.mi.us', 's10.lite.msu.edu', 's12.lite.msu.edu', 'lon-capa.chem.sunysb.edu', 'schubert.tmcc.edu', 'dalton.chem.sfu.ca', 'capa2.phy.ohiou.edu', 'pollux.physics.fsu.edu', 'loncapa.physics.sc.edu', 'loncapa.math.ucf.edu', 'zappa.ags.udel.edu', 'loncapa.gwu.edu');  my @servers = (
   'newscience.westshore.edu',
   's10.lite.msu.edu',
   's12.lite.msu.edu',
   'schubert.tmcc.edu',
   'dalton.chem.sfu.ca',
   'capa2.phy.ohiou.edu',
   'pollux.physics.fsu.edu',
   'loncapa3.physics.sc.edu',
   'zappa.ags.udel.edu',
   'loncapa.gwu.edu',
   'neptune.physics.ndsu.nodak.edu',
   'capa1.uwsp.edu',
   'loncapa.Mines.EDU',
   'loncapa.chm.nau.edu',
   'library1.lon-capa.uiuc.edu',
   'lon-capa.bsu.edu',
   'psblnx03.bd.psu.edu',
   'lon-capa.acadiau.ca',
   'harvard.lon-capa.org',
   'capa1.cc.huji.ac.il',
   'lon-capa.phy.cmich.edu',
   'meitner.physics.hope.edu',
   'loncapa.vcu.edu',
   'lon-capa.ucsc.edu',
   'lon-capa.bsu.edu',
   'harvard.lon-capa.org'
   );
   
   foreach (@servers) {
       my $url='http://'.$_.'/cgi-bin/metadata_harvest.pl';
 # End Configuration  # End Configuration
   
 #my $ua = new LWP::UserAgent;  my $ua = new LWP::UserAgent;
 #$ua->timeout(600);  $ua->timeout(600);
   
 #my $request = new HTTP::Request GET => $url;  my $request = new HTTP::Request GET => $url;
 #$request->authorization_basic('reaper', 'cat4u');  $request->authorization_basic('reaper', 'cat4u');
   
 #my $response = $ua->request( $request );  my $response = $ua->request( $request );
   
 #if ( $response->is_success ) {  if ( $response->is_success ) {
 # $content = $response->content;       print 'SUCCESS: ' . $response->message.' for '.$url."\n\n";
    $content = $response->content;
 # Delete all blank lines  # Delete all blank lines
 # $content =~ s/(?<!.)\n//g;   $content =~ s/(?<!.)\n//g;
 # Replace all ^M with spaces  # Replace all ^M with spaces
 # $content =~ s/ /\s/g;   $content =~ s/ /\s/g;
 # Push the content into an array  # Push the content into an array
 # @loncapa = split /\n/, $content;   @loncapa = split /\n/, $content;
 #} else {  } else {
 # die 'LON-CAPA request failed: ' . $response->message;       print 'LON-CAPA request failed: ' . $response->message.' for '.$url."\n\n";
 #}       next;
   }
   
 @loncapa=undef;  #@loncapa=undef;
 open (LON_FILE, 'metadata_harvest.txt') || die;  #open (LON_FILE, 'metadata_harvest.txt') || die;
   
 while (<LON_FILE>) {  #while (<LON_FILE>) {
        chomp;  #       chomp;
        push(@loncapa,$_);  #       push(@loncapa,$_);
 }  #}
   
 my %records = ();;  my %records = ();;
   
   my %stats=();
   
 foreach my $metadata (@loncapa) {  foreach my $metadata (@loncapa) {
  chomp $metadata;   chomp $metadata;
    $metadata=~s/[^\w\d\s\.\;\:\,\|\/]/ /gs;
  my @tkline = split('\|', $metadata);   my @tkline = split('\|', $metadata);
  my $title = $tkline[0];          my ($rawtype)=($tkline[3]=~/\.(\w+)$/);
  next if ( $title eq '' );          $rawtype=~tr/A-Z/a-z/;
           $allstats{$rawtype}++;
           
           my $title = $tkline[0];
    if ( $title eq '' ) { $knockout{'no_title_'.$rawtype}++; next; }
  my $author = $tkline[1];   my $author = $tkline[1];
  next if ( $author eq '' );   if ( $author eq '' ) { $knockout{'no_author_'.$rawtype}++; next; }
  my @authorname = split(' ', $author);   my @authorname = split(' ', $author);
  my $author_fname = $authorname[0];   my $author_fname = $authorname[0];
  my $author_lname = $authorname[1];   my $author_lname = $authorname[1];
Line 104  foreach my $metadata (@loncapa) { Line 127  foreach my $metadata (@loncapa) {
  }   }
  my $subject = $tkline[2];   my $subject = $tkline[2];
  next if ( ($subject eq 'Sample') || ($subject eq 'Something') );   next if ( ($subject eq 'Sample') || ($subject eq 'Something') );
  my $resourceurl = 'http://lon-capa.smete.org' . $tkline[3];   my $resourceurl = 'http://nsdl.lon-capa.org' . $tkline[3];
           my $baseid=$tkline[3];
    my ($adom,$auname)=($baseid=~/^\/res\/(\w+)\/(\w+)\//);
    $baseid=~s/\W/\_/g;
    $baseid=~s/^\_res\_//g;
    my $fileid=md5_hex($baseid);
   
  next if ( $resourceurl =~ /(.*)\/demo\/(.*)/ );   next if ( $resourceurl =~ /(.*)\/demo\/(.*)/ );
   # too many fragments out there
           next unless ($resourceurl=~/\.(html|htm|problem|assess|xhtm|xml|xhtml|gif|jpg|jpeg|png)$/i);
   
  my $keywords = $tkline[4];   my $keywords = $tkline[4];
  my $version = $tkline[5];   my $version = $tkline[5];
  my $notes = $tkline[6];   my $notes = $tkline[6];
  my $abstract = $tkline[7];   my $abstract = $tkline[7];
  next if ($abstract eq '');          $abstract=~s/ s / /gs;
  my $type = $tkline[8];          $abstract=~s/\s+/ /gs;
           my $postsubject=$subject;
           unless ($postsubject) {
              $postsubject=$keywords;
           } else {
              $postsubject.=' ('.$keywords.')';
           }
           unless ($postsubject=~/\w/) { $knockout{'nosubject_'.$rawtype}++; next; }
           unless ($abstract) { $knockout{'noabstract_'.$rawtype}++; next; }
    my $type = $rawtype;
           if ($type=~/htm/) { $type='htm'; }
   
  my $learning_resource_type;   my $learning_resource_type;
  if ( $type eq 'problem' ) {   if ( $type eq 'problem' ) {
  $learning_resource_type = 114;   $learning_resource_type = 114;
Line 144  foreach my $metadata (@loncapa) { Line 187  foreach my $metadata (@loncapa) {
  $media_format = 0;   $media_format = 0;
  }   }
   
  my $language = $tkline[9]; # Look only for seniso   my $language = $tkline[9];
  next if ( $language ne 'seniso');  # likelihood is that the following is true (people would bother if it is not)
           if (($language=~/(seniso|notset|English)/) || (!$language)) { $language='seniso'; }
   # NSDL only does English
           if ( $language ne 'seniso') { $knockout{'lang_'.$rawtype}++; $knockoutlang{$language}++; next; } 
  my $primary_language='en-US';   my $primary_language='en-US';
  my $creation_date = $tkline[10];   my $creation_date = $tkline[10];
  my ($pub_year,$pub_month,$pub_day) = ( $creation_date =~ /^(\d{4})-(\d{2})-(\d{2})\s(\d{2}):(\d{2}):(\d{2})$/ );   my ($pub_year,$pub_month,$pub_day) = ( $creation_date =~ /^(\d{4}) (\d{2}) (\d{2})\s(\d{2}):(\d{2}):(\d{2})$/ );
  my $revision_date = $tkline[11];   my $revision_date = $tkline[11];
    my ($rev_year,$rev_month,$rev_day) = ( $revision_date =~ /^(\d{4}) (\d{2}) (\d{2})\s(\d{2}):(\d{2}):(\d{2})$/ );
  my $owner = $tkline[12];   my $owner = $tkline[12];
  my $rights_description;   my $rights_description;
  my $copyright = $tkline[13]; # public,domain,default,private (skip if private and domain)   my $copyright = $tkline[13]; # public,domain,default,private (skip if private and domain)
Line 165  foreach my $metadata (@loncapa) { Line 212  foreach my $metadata (@loncapa) {
  # Domain means restricted to a particular LON-CAPA domain   # Domain means restricted to a particular LON-CAPA domain
  # Defaults mean access open to any registered LON-CAPA user   # Defaults mean access open to any registered LON-CAPA user
  # Private means open only to author of material   # Private means open only to author of material
  next if ( $copyright eq 'private');          unless ($copyright eq 'public') { $knockout{'notpublic_'.$rawtype}++; next; }
  my $platform = "5";     # HTML Browser (not specified but construed from metadata)   my $platform = "5";     # HTML Browser (not specified but construed from metadata)
   #
 # Connect to database  # We actually do this
 if ( $useDatabase ) {  #
  $dbh= DBI->connect($DBI_DSN, $DBI_USER, $DBI_PWD, { RaiseError => 1, AutoCommit => 0 }) || die "Unable to connect to database $DBI_DSN as $DBI_USER: ($DBI::err) $DBI::errstr\n";;          $stats{$type}++;
  # Configuration information for LON-CAPA          $filterstats{$type}++;
  my $collection_id = OAIc_orgexists($dbh,'LearningOnline Network with CAPA');  #
  my $submitter_id = OAIc_personexists($dbh,'adong@smete.org');  # Create path
  my $image = 'http://www.lite.msu.edu/liteani.gif';  #
  my $cost = 1; # version.purchase_license_type_id   unless (-e $basepath.'/'.$adom) { mkdir($basepath.'/'.$adom); }
  my $collection = 'LearningOnline Network with CAPA';   unless (-e $basepath.'/'.$adom.'/'.$auname) { 
  # LON-CAPA has single authors      mkdir($basepath.'/'.$adom.'/'.$auname) || die 'Could not create '.$basepath.'/'.$adom.'/'.$auname;
  my $reg_key;   }
  if ( $object_type eq 'organization' ) {   open(XML,'>'.$basepath.'/'.$adom.'/'.$auname.'/'.$baseid.'.xml');
  if ( ! ($reg_key = OAIc_orgexists($dbh,join(' ',$author_fname,$author_lname))) ) {   print XML (<<ENDMETA);
  printf("Inserting new organization %s\n", join(' ',$author_fname, $author_lname));  <?xml version="1.0" encoding="UTF-8"?>
  my $success = OAIc_insert_org($dbh,$collection_id,$submitter_id,'',join(' ',$author_fname,$author_lname),'','','','','','','','');  
  $reg_key = OAIc_orgexists($dbh,join(' ',$author_fname,$author_lname));  <oaidc:dc xmlns="http://purl.org/dc/elements/1.1/" 
  }            xmlns:oaidc="http://www.openarchives.org/OAI/2.0/oai_dc/" 
  } else {            xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" 
  if ( ! ($reg_key = OAIc_personexists_name($dbh,join(' ',$author_fname,$author_lname))) ) {            xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/ 
  printf("Inserting new person(author) %s\n", join(' ',$author_fname, $author_lname));                                http://www.openarchives.org/OAI/2.0/oai_dc.xsd"
  my $success = OAIc_insert_person($dbh,$collection_id,$submitter_id,$author_lname,$author_fname,'','');  >
  $reg_key = OAIc_personexists_name($dbh,join(' ',$author_fname,$author_lname));      <title>$title</title>
  }      <creator>$author</creator>
  }      <identifier>$resourceurl</identifier>
  my $updated;      <subject>$postsubject</subject>
  my $inserted;      <language>$primary_language</language>
  if ( my $general_key = OAIc_loexists($dbh,$title) ) {      <description>$abstract</description>
  # Do nothing      <date>$rev_year-$rev_month-$rev_day</date>
  $updated = $updated + 1;  </oaidc:dc>
  } else {  ENDMETA
  printf("Inserting new record for %s\n",$title);        close (XML);
  my $success = OAIc_insert_lo($dbh, $title, $primary_language, $abstract, $image, $pub_month, $pub_year, $keywords, $submitter_id, $reg_key, $collection_id, $collection_id, $media_format, $platform, , '', $resourceurl, '', 1, $reg_key, $collection_id, $collection_id, '', '', '', $learning_resource_type, $rights_description, $cost);  
  $inserted = $inserted + 1;  
  }  
 }  }
   foreach my $thistype (sort keys %stats) {
 if (! $useDatabase ) { # Print information if no database updates requested     print "\n$thistype: $stats{$thistype}";
  printf("Title: %s\n", $title);  
  printf("Author First Name: %s\n", $author_fname);  
  printf("Author Last Name: %s\n", $author_lname);  
  printf("Subject: %s\n", $subject);  
  printf("URL: %s\n", $resourceurl);  
  printf("Keywords: %s\n", $keywords);  
  printf("Version: %s\n", $version);  
  printf("Notes: %s\n", $notes);  
  printf("Abstract: %s\n", $abstract);  
  printf("Learning Resource Type: %d\n", $learning_resource_type);  
  printf("Media Format: %d\n", $media_format);  
  printf("Primary Language: %s\n", $primary_language);  
  printf("Creation Date: %s\n", $creation_date);  
  printf("Revision Date: %s\n", $revision_date);  
  printf("Copyright: %s\n", $copyright);  
  printf("Publication Year: %4d\tPublication Month: %02d\n", $pub_year, $pub_month);  
 }  
   
 if ( $useDatabase ) {  
  $dbh->commit;  
  $dbh->disconnect;  
 }  }
   print "\n----\n";
   }
   print "\nDone.\n";
   foreach my $thistype (sort keys %allstats) {
      print "\n$thistype: $allstats{$thistype} ($filterstats{$thistype}) title: $knockout{'no_title_'.$thistype} author: $knockout{'no_author_'.$thistype} lang: $knockout{'lang_'.$thistype} priv: $knockout{'private_'.$thistype} domain: $knockout{'domain_'.$thistype} custom: $knockout{'custom_'.$thistype}";
   }
   print "\n----\n";
   foreach my $thislang (sort keys %knockoutlang) {
   print "\n>$thislang<: $knockoutlang{$thislang}";
 }  }
   print "\n";
Removed from v.1.1  
changed lines
  Added in v.1.9

FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>