New versions of this module and support for Omeka S version 3.0 and above are available on GitLab, which seems to respect users and privacy better than the previous repository.
OAI-PMH Repository is a module for Omeka S that implements an Open Archives Initiative Protocol for Metadata Harvesting (OAI-PMH) repository for Omeka S, allowing Omeka S items, item sets, and media to be harvested by OAI-PMH harvesters. The module implements version 2.0 of the protocol.
It provides a way to map non-Dublin Core properties to Dublin Core properties in order to expose all values of all items, in particular records for author managed with foaf. Of course, the mapping is far from perfect, because it is not simple to map two ontologies that don't describe the same content.
Anyway, there is a specific format, simple_xml, similar to the default oai_dc,
that returns resource as a flat xml, included all values of all vocabularies
that are used.
See general end user documentation for installing a module.
This module requires the module Common, that should be installed first.
- From the zip
Download the last release OaiPmhRepository.zip from the list of releases, and
uncompress it in the modules directory.
- From the source and for development:
If the module was installed from the source, rename the name of the folder of
the module to OaiPmhRepository.
There is a global oai server, with all the data of the Omeka S instance, and a server for each site, if wanted, to manage the case where there are multiple institutional entities that share the same Omeka S server. Nevertheless, all options are defined globally in order to keep consistency between servers, in particulars for the identifiers of the sets and the items.
Name for this OAI-PMH repository. This value is sent as part of the response to an Identify request, and it is how the repository will be identified by well-behaved harvesters.
Default: The name of the Omeka S installation.
The oai-identifier specification requires repositories to specify a namespace identifier. This will be used to form globally unique IDs for the exposed metadata items. This value is required to be a domain name you have registered. Using other values will generate invalid identifiers.
Default: If it can, the module will try to automatically detect the domain of
the server hosting the site, and use that as the default namespace identifier.
If a name can’t be detected (for example, if the site is accessed through the
localhost domain), the default will be default.must.change (as you might
think, this value is intended to be changed, not used as-is). The module will
function with this, or any other string, as the namespace identifier, but this
breaks the assumption that each identifier is globally unique. Best practice is
to set this value to the domain name the Omeka server is published at, possibly
with a prefix like "oai."
Select the formats that will be available. Only oai_dc is required.
Whether the repository should expose direct URLs to all the files associated with an item as part of its returned metadata. This gives harvesters the ability to directly access the media described by the metadata.
Default: true
Whether the module should expose empty oai sets. If enabled, only collections or sites that actually contain at least one public item will be included in the ListSets output. If disabled, all public oai sets are included in ListSets output.
Default: true
The global repository contains all the resources of Omeka S, in one place. If
enabled, it can exposes different types of oai sets: the sites, or the item
sets. The url is simply /oai.
TODO: Expose the resource classes as oai sets or subsets. Expose sets via created via queries.
The site repositories simulate multiple oai servers, with the site pools of items and the attached item sets as oai sets.
There is only one format currently, but new ones can be added, for example when the resource have a unique standard identifier.
This option allows to expose all metadata in the standard oai_dc and use refined
terms for Dublin Core elements, for example "dcterms:abstract" will be merged
with "dc:description". For other merges, the event oaipmhrepository.values.pre
can be used. For a quick hack, the file data/mapping/generic_dcterms.php can
be updated.
This option allows to expose any metadata as another one. The default mapping maps Bibliographic Ontology (bibo) and Friend of a Friend (foaf) into Dublin Core. Of course is not perfect, but you can adapt it to your real data. See the default For foaf, a spreadsheet can help to improve the mapping.
Mets can use any format to describe the item and the files. The formats that are
commonly used are dc and mods (currently unavailable).
The OAI-PMH pages can be displayed and browsed with a themable responsive human
interface based on Bootstrap. To theme it, create a folder asset/xsl in your
theme and copy the file oai-pmh-repository.xsl inside it, then edit it. The
css file can be copied and updated as well.
An alias (redirect 301) for backward compatibility with Omeka Classic, that used
/oai-pmh-repository/request, or any other old OAI-PMH repository, so the
harvesters will continue to harvest your metadata.
Number of individual items that can be returned in a response at once. Larger values will increase memory usage but reduce the number of database queries and HTTP requests. Smaller values will reduce memory usage but increase the number of DB queries and requests.
Default: 50 (recommended in most cases)
Amount of time in minutes a resumptionToken is valid for. The specification suggests a number in the tens of minutes. This boils down to the length of time a harvester has to request the next part of an incomplete list request.
Default: 10 (minutes)
The module ships with several default formats. Other modules can alter these or add their own.
Dublin Core (prefix oai_dc)
The Dublin Core is required by the OAI-PMH specification for all repositories. Omeka S metadata fields are mapped one-to-one with fields for this output format.
Dublin Core Terms (prefix oai_dcterms)
The OAI-PMH standard does not manage the complete set of Dublin Core, but it is
largely used. So this format is similar to oai_dc, but with all 55 terms.
NOTE: The namespace and the schema don’t exist.
CDWA Lite (prefix cdwalite)
The mapping between Omeka’s metadata and CDWA Lite metadata is more complicated, and certain fields may not be populated correctly. The chief advantage of using CDWA Lite output is that file URLs can be output in a controlled format, unlike Dublin Core. Harvesters may therefore be able to harvest or link to files in addition to metadata.
MODS (prefix mods)
This output crosswalks the Dublin Core metadata to MODS using the mapping recommended by the Library of Congress.
METS (prefix mets)
The Metadata Encoding and Transmission Standard format exposes files to harvesters alongside Dublin Core metadata.
RDF (prefix rdf)
This format exposes metadata as RDF/XML. Unlike many of the other formats, RDF allows the repository to expose metadata from different standards all in the same output.
NOTE: The rdf output is currently not implemented, but it is available through the standard api of Omeka S as json.
Omeka XML (prefix omeka-xml)
This output format uses an Omeka-specific XML output that includes all metadata elements without requiring crosswalking or subsetting, but is not well-supported by harvesters or other tools and because the RRCHNM itself removed the schema from the last site.
NOTE: Because of its limited support by harvesters, the format is not implemented in Omeka S. Nevertheless, you can use the format "simple xml" that provides all metadata too.
This format exposes metadata as a simple flat XML. Unlike most of the other standard formats, this format expose all metadata from different vocabularies.
Other formats can be added or replace an existing one via a key in the config
['oaipmhrepository']['metadata_formats'].
The format EAD, designed to describe archives, is available through the module
EAD.
The output can be customized via the filter oaipmhrepository.values.pre, that
is triggered before processing and that contains all the values of the current
resource, so it is possible to remove, to update or to append some values, or to
convert some properties from other vocabularies into the standard formats.
The old event oaipmhrepository.values called for each value is deprecated
since version 3.3.5.2 and was removed in version 3.3.6.
- Normalize simple_xml (or omeka xml v5/v6?).
- Improve data output for media html in simple xml, in particular for html.
- Include namespace and metadata of modules when needed for simple_xml (module Mapping, etc.)
- Native rdf output (from json serialization? From easyrdf?).
- Improve performance for big bases and items with many medias. With a cache? In a table? As a xml file?
Use it at your own risk.
It’s always recommended to backup your files and your databases and to check your archives regularly so you can roll back if needed.
See online issues on the module issues page on GitLab.
This module is published under GNU/GPL.
This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 3 of the License, or (at your option) any later version.
This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
The human interface is published under the CeCILL-B BSD-like licence. See its header for other licenses notes. It is based on Bootstrap.
See commits for full list of contributors.
- Copyright Yu-Hsun Lin, 2009
- Copyright John Flatness, 2009-2016
- Copyright Julian Maurice for BibLibre, 2016-2017
- Copyright Daniel Berthereau, 2014-2023 (see Daniel-KM)
This Omeka S module is based on the rewrite of the OAI-PMH Repository plugin for Omeka by BibLibre and provide the same features as the original plugin and many more.