Skip to content

InvenioRDM records export formats

Introduced in v12.0.0b1

InvenioRDM provides various record serialisation formats to allow users to easily export bibliographic records in a machine-readable way or transfer them to different systems. These formats adhere to widely used metadata standards such as DataCite, Dublin Core, BibTeX, MARCXML, DCAT-AP or GeoJSON, which is commonly used for geospatial metadata.

By providing a range of export formats, Invenio-app-rdm enables users to share and exchange metadata records with other systems in a format that is compatible with their respective standards. This makes it easier to ensure that metadata records are accurate, complete, and consistent across different systems.

A bibliographic record can be exported to a different format in a couple of ways: 1. in the UI, using the Export dropdown menu in the record landing page (screenshot with the menu open) 2. via REST APIs, using the Accept header:

GET /api/records/ HTTP/1.1
Accept: <export-format-mime-type>;charset=UTF-8

Find below the Accept header MIME type for each export format.


The JSON export format in InvenioRDM exports metadata records in a structured JSON format, which includes all the metadata fields associated with the record, as well as any custom fields. This makes it possible to export metadata records from Invenio-app-rdm in a machine-readable format that can be used for a variety of purposes, such as data analysis, visualization, or integration with other software systems.

To export a metadata record in JSON format in InvenioRDM, users can simply select the "JSON" option from the export menu, which will generate a JSON file containing the record's metadata in the appropriate format. The JSON file can then be downloaded and used as needed.

To export a record via REST API to JSON format, use the application/json MIME type:

GET /api/records/ HTTP/1.1
Accept: application/json;charset=UTF-8


The DataCite export format is a JSON and XML serialization of the DataCite internal data model. You can read more about the InvenioRDM data model .

The official documentation for the DataCite metadata schema can be found on their website.

To export a record to DataCite XML format, use the application/vnd.datacite.datacite+xml MIME type:

GET /api/records/ HTTP/1.1
Accept: application/vnd.datacite.datacite+xml;charset=UTF-8

To export a record to DataCite JSON format, use the application/vnd.datacite.datacite+json MIME type:

GET /api/records/ HTTP/1.1
Accept: application/application/vnd.datacite.datacite+json;charset=UTF-8


DCAT-AP is a specification for describing public sector datasets and data services across Europe. It provides a common metadata vocabulary for describing datasets, catalogs, and services. The goal is to make it easier for users to find, access, and use data. DCAT-AP is based on the Data Catalog Vocabulary (DCAT) specification, which is maintained by the W3C.

The DCAT-AP data model is based on the Dublin Core metadata elements and includes additional properties for describing datasets and data services. It defines a core set of metadata elements that must be provided for each dataset, as well as optional and recommended elements. Some of the core elements include the title, description, publisher, and keyword.

Here is a link to the DCAT-AP data model (format) documentation:

To export a record to DCAT-AP format, use theapplication/dcat+xml MIME type:

GET /api/records/ HTTP/1.1
Accept: application/dcat+xml;charset=UTF-8


CSL (Citation Style Language) is a widely used XML-based format for defining citation styles used in academic writing. InvenioRDM provides CSL export format as a way to allow users to easily export metadata of their records in a citation format that can be used with various reference management software. This feature allows users to easily generate bibliographies and citations for their research work.

To export a record to CSL format, use the application/vnd.citationstyles.csl+json MIME type:

GET /api/records/ HTTP/1.1
Accept: application/vnd.citationstyles.csl+json;charset=UTF-8


MARCXML is an XML representation used to represent bibliographic records in the MARC 21 (MAchine Readable Cataloging) format, which is widely used in libraries for the description of bibliographic items. MARCXML includes a standard set of data elements, control fields, and subfields, each with its own tag and indicator values, that allow for the representation of bibliographic metadata.

Here's a link to the official documentation of the MARC 21 data model:

To export a record to MARCXML format, use the application/marcxml+xml MIME type:

GET /api/records/ HTTP/1.1
Accept: application/marcxml+xml;charset=UTF-8


BibTeX is a bibliographic file format commonly used with LaTeX documents. It allows users to easily organize and manage bibliographic references in their documents. The BibTeX format uses a plain text file with a .bib extension and consists of entries, each describing a reference.

Each BibTeX entry starts with a @ symbol, followed by the type of reference (e.g., @article, @book, @inproceedings, etc.) and a unique identifier for the reference. The fields for each entry include the author(s), title, year, publisher, journal, and any other relevant information about the reference.

BibTeX is widely used in the academic community and is supported by many reference management software programs.

Here's a link to the BibTeX documentation:

To export a record to BibTex format, use the application/x-bibtex MIME type:

GET /api/records/ HTTP/1.1
Accept: application/x-bibtex;charset=UTF-8

Dublin Core

Dublin Core is a metadata standard used to describe digital resources such as web pages, images, and videos. It consists of a set of 15 metadata elements that provide basic descriptive information about a resource, including its title, creator, subject, description, publisher, and date. The elements can be used in a variety of combinations and can be extended to meet specific needs.

Dublin Core is widely used across many domains and is especially popular in the library and cultural heritage communities. It is a simple and flexible standard that allows for easy exchange and sharing of metadata between systems.

Here's a link to the Dublin Core Metadata Element Set specification:

Example search request:

GET /api/records/ HTTP/1.1
Accept: application/x-dc+xml;charset=UTF-8


GeoJSON is a format for encoding geographic data structures using JSON. It supports various types of geometries such as points, lines, and polygons, and allows for additional properties to be associated with those geometries. GeoJSON is often used to exchange data between web services and to display geospatial data on web maps.

Here is a link to the official GeoJSON specification:

To export a record to GeoJSON format, use the application/vnd.geojson+json MIME type:

GET /api/records/ HTTP/1.1
Accept: application/vnd.geojson+json;charset=UTF-8