Editing ORCID (section)

= OAI-PMH (Experimental) =
'''[[User:Libjlrs]] 2016-09-29 Warning: this is experimental. I hope others will read, understand, question and improve on this before it becomes some form of best practice!'''

Some metadata profiles support attributes that are useful to propagate ORCIDs e.g. RIOXX (see the [[#Exposing_the_ORCID_in_RIOXX]] section above, and the 'rioxxterms:author' section on http://www.rioxx.net/profiles/v2-0-final/ ).

A lot of harvesters use the oai_dc metadata profile when harvesting records. The DC schema doesn't allow attributes on the dc:creator element, so ORCIDs cannot be added in a sensible way to this profile.

The details below create an additional metadata prefix that adds an attribute to the dc:creator field. The original <code>metadataPrefix=oai_dc</code> output remains unchanged. Using <code>metadataPrefix=oai_dc_orcid</code> will add the orcid attribute where available.

Currently there is not an XSD that describes this change, so a harvester that validates the feed may stop when it encounters an ORCID.

This example is based on the <code>EPrints::Plugin::Export::OAI_DC</code> module from EPrints 3.3.10. You may need to tweak it to make it work for you!

File: <code>~/archives/ARCHIVEID/cfg/plugins/EPrints/Plugin/Export/OAI_DC_ORCID.pm</code>

<source lang="perl">
package EPrints::Plugin::Export::OAI_DC_ORCID;

use EPrints::Plugin::Export::OAI_DC;

@ISA = ( "EPrints::Plugin::Export::OAI_DC", "EPrints::Plugin::Export::DC" );

use strict;

sub new
{
        my( $class, %opts ) = @_;

        my $self = $class->SUPER::new( %opts );

        $self->{name} = "Dublin Core - OAI + ORCID Schema";
        $self->{accept} = [ 'dataobj/eprint' ];
        $self->{visible} = "";
        $self->{suffix} = ".xml";
        $self->{mimetype} = "text/xml";

        $self->{metadataPrefix} = "oai_dc_orcid";
        $self->{xmlns} = "http://www.openarchives.org/OAI/2.0/oai_dc/";
        $self->{schemaLocation} = "http://www.openarchives.org/OAI/2.0/oai_dc.xsd";

        return $self;
}

sub xml_dataobj
{
        my( $plugin, $dataobj ) = @_;

        my $data = $plugin->convert_dataobj( $dataobj );

        my $dc = $plugin->{session}->make_element(
                "oai_dc:dc",
                "xmlns:oai_dc" => "http://www.openarchives.org/OAI/2.0/oai_dc/",
                "xmlns:dc" => "http://purl.org/dc/elements/1.1/",
                "xmlns:xsi" => "http://www.w3.org/2001/XMLSchema-instance",
                "xsi:schemaLocation" =>
        "http://www.openarchives.org/OAI/2.0/oai_dc/ http://www.openarchives.org/OAI/2.0/oai_dc.xsd" );

        # turn the list of pairs into XML blocks (indented by 8) and add them
        # them to the DC element.
        foreach( @{$data} )
        {
                $dc->appendChild(  $plugin->{session}->render_data_element( 8, "dc:".$_->[0], $_->[1], %{$_->[2]} ) );
                # produces <key>value</key>
        }

        return $dc;
}

sub convert_dataobj
{
        my( $plugin, $eprint ) = @_;

        my $dataset = $eprint->{dataset};

        my @dcdata = ();

        push @dcdata, $plugin->simple_value( $eprint, title => "title" );

        if( $eprint->exists_and_set( "creators" ) )
        {
                my $creators = $eprint->get_value( "creators" );
                foreach my $creator ( @$creators )
                {
                        next if !defined $creator;

                        # Add ORCID if it exists
                        if( EPrints::Utils::is_set( $creator->{orcid} ) ){
                                push @dcdata, [ "creator", EPrints::Utils::make_name_string( $creator->{name} ), { 'orcid' => $creator->{orcid} } ];
                        } else {
                                push @dcdata, [ "creator", EPrints::Utils::make_name_string( $creator->{name} ) ];
                        }
                }
        }

        if( $eprint->exists_and_set( "subjects" ) )
        {
                my $subjectid;
                foreach $subjectid ( @{$eprint->get_value( "subjects" )} )
                {
                        my $subject = EPrints::DataObj::Subject->new( $plugin->{session}, $subjectid );
                        # avoid problems with bad subjects
                                next unless( defined $subject );
                        push @dcdata, [ "subject", EPrints::Utils::tree_to_utf8( $subject->render_description() ) ];
                }
        }

        push @dcdata, $plugin->simple_value( $eprint, abstract => "description" );
        push @dcdata, $plugin->simple_value( $eprint, publisher => "publisher" );

        if( $eprint->exists_and_set( "editors_name" ) )
        {
                my $editors = $eprint->get_value( "editors_name" );
                if( defined $editors )
                {
                        foreach my $editor ( @{$editors} )
                        {
                                push @dcdata, [ "contributor", EPrints::Utils::make_name_string( $editor ) ];
                        }
                }
        }
        ## Date for discovery. For a month/day we don't have, assume 01.
        if( $eprint->exists_and_set( "date" ) )
        {
                my $date = $eprint->get_value( "date" );
                if( defined $date )
                {
                        $date =~ s/(-0+)+$//;
                        push @dcdata, [ "date", $date ];
                }
        }

        if( $eprint->exists_and_set( "type" ) )
        {
                push @dcdata, [ "type", EPrints::Utils::tree_to_utf8( $eprint->render_value( "type" ) ) ];
        }

        my $ref = "NonPeerReviewed";
        if( $eprint->exists_and_set( "refereed" ) && $eprint->get_value( "refereed" ) eq "TRUE" )
        {
                $ref = "PeerReviewed";
        }
        push @dcdata, [ "type", $ref ];


        my @documents = $eprint->get_all_documents();
        my $mimetypes = $plugin->{session}->get_repository->get_conf( "oai", "mime_types" );
        foreach( @documents )
        {
                my $format = $mimetypes->{$_->get_value("format")};
                $format = $_->get_value("format") unless defined $format;
                #$format = "application/octet-stream" unless defined $format;
                push @dcdata, [ "format", $format ];
                push @dcdata, [ "identifier", $_->get_url() ];
        }

        # Most commonly a DOI or journal link
        push @dcdata, $plugin->simple_value( $eprint, official_url => "relation" );

        # The citation for this eprint
        push @dcdata, [ "identifier",
                EPrints::Utils::tree_to_utf8( $eprint->render_citation() ) ];

        # The URL of the abstract page
        if( $eprint->is_set( "eprintid" ) )
        {
                push @dcdata, [ "relation", $eprint->get_url() ];
        }
        # dc.language not handled yet.
        # dc.source not handled yet.
        # dc.coverage not handled yet.
        # dc.rights not handled yet.

        return \@dcdata;
}

1;
</source>