Getting Started with APIs

If you are a researcher or curator who wants to automate parts of your workflow, this section should help you get started. The Introduction section lists resources for other groups who may be interested in Dataverse Software APIs such as developers of integrations and support teams.

Servers You Can Test With

Rather than using a production Dataverse installation, API users are welcome to use https://demo.dataverse.org for testing. You can email support@dataverse.org if you have any trouble with this server.

If you would rather have full control over your own test server, deployments to AWS, Docker, and more are covered in the Developer Guide and the Installation Guide.

Getting an API Token

Many Dataverse Software APIs require an API token.

Once you have identified a server to test with, create an account, click on your name, and get your API token. For more details, see the API Tokens and Authentication section.

curl Examples and Environment Variables

The examples in this guide use curl for the following reasons:

  • curl commands are succinct.

  • curl commands can be copied and pasted into a terminal.

  • This guide is programming language agnostic. It doesn’t prefer any particular programming language.

You’ll find curl examples that look like this:

export SERVER_URL=https://demo.dataverse.org
export QUERY=data

curl $SERVER_URL/api/search?q=$QUERY

What’s going on above is the declaration of “environment variables” that are substituted into a curl command. You should run the “export” commands but change the value for the server URL or the query (or whatever options the command supports). Then you should be able to copy and paste the curl command and it should “just work”, substituting the variables like this:

curl https://demo.dataverse.org/api/search?q=data

If you ever want to check an environment variable, you can “echo” it like this:

echo $SERVER_URL

With curl version 7.56.0 and higher, it is recommended to use –form-string with outer quote rather than -F flag without outer quote.

For example, curl command parameter below might cause error such as warning: garbage at end of field specification: ,"categories":["Data"]}.

-F jsonData={\"description\":\"My description.\",\"categories\":[\"Data\"]}

Instead, use –form-string with outer quote. See https://github.com/curl/curl/issues/2022

--form-string 'jsonData={"description":"My description.","categories":["Data"]}'

If you don’t like curl, don’t have curl, or want to use a different programming language, you are encouraged to check out the Python, Javascript, R, and Java options in the Client Libraries section.

Depositing Data

Creating a Dataverse Collection

See Create a Dataverse Collection.

Creating a Dataset

See Create a Dataset in a Dataverse Collection.

Uploading Files

See Add a File to a Dataset. In addition, when a Dataverse installation is configured to use S3 storage with direct upload enabled, there is API support to send a file directly to S3. This facilitates an efficient method to upload big files, but is more complex. The procedure is described in the Direct DataFile Upload/Replace API section of the Developer Guide.

Publishing a Dataverse Collection

See Publish a Dataverse Collection.

Publishing a Dataset

See Publish a Dataset.

Finding and Downloading Data

Finding Datasets

A quick example search for the word “data” is https://demo.dataverse.org/api/search?q=data

See the Search API section for details.

Finding Recently Published Dataverse Collections, Datasets, and Files

See Date Range Search Example.

It’s also possible to find recently published datasets via OAI-PMH.

Downloading Files

The Data Access API section explains how to download files.

To download all the files in a dataset, see Downloading All Files in a Dataset.

In order to download individual files, you must know their database IDs which you can get from the dataverse_json metadata at the dataset level. See Export Metadata of a Dataset in Various Formats.

Downloading Metadata

Dataset metadata is available in a variety of formats listed at Supported Metadata Export Formats.

See Export Metadata of a Dataset in Various Formats.

Listing the Contents of a Dataverse Collection

See Show Contents of a Dataverse Collection.

Managing Permissions

Granting Permission

See Assign a New Role on a Dataverse Collection.

Revoking Permission

See Delete Role Assignment from a Dataverse Collection.

Listing Permissions (Role Assignments)

See List Role Assignments in a Dataverse Collection.

Getting the OpenAPI Document

You can access our OpenAPI document using the /openapi endpoint. The default format is YAML if no parameter is provided, but you can also obtain the JSON version by either passing format=json as a query parameter or by sending Accept:application/json (case-sensitive) as a header.

Note

See curl Examples and Environment Variables if you are unfamiliar with the use of export below.

export SERVER_URL=https://demo.dataverse.org
export FORMAT=json

curl "$SERVER_URL/openapi?format=$FORMAT"

The fully expanded example above (without environment variables) looks like this:

curl "https://demo.dataverse.org/openapi?format=json"

We are aware that our OpenAPI document is not perfect. You can find more information about validating the document under OpenAPI in the Developer Guide.

Beyond “Getting Started” Tasks

In addition to the tasks listed above, your Dataverse installation supports many other operations via API.

See Lists of Dataverse APIs and Types of Dataverse Software API Users to get oriented.

If you’re looking for some inspiration for how you can use the Dataverse Software APIs, there are open source projects that integrate with the Dataverse Software listed in the Apps section.

Getting Help

See Getting Help.