GitHub - DMPRoadmap/orion

Orion (Roadmap ROR Integration) is a lightweight service designed to integrate the Research Organization Registry (ROR) into systems like Roadmap. Built in Go, Orion exposes a simple HTTP API for linking email domains to research organizations.

Built with efficiency and minimal resource consumption in mind, Orion avoids using a traditional database, aligning with principles of sustainable and lightweight infrastructure. Instead, it leverages a structured local filesystem-based storage to serve organization records and domain-to-ROR mappings.

This design enables fast, efficient lookups with minimal dependencies and overhead. This is ideal for lightweight deployments or containerized environments.

Getting Started

Requirements

Go
Docker (optional)
Input data to generate ROR records and domain mappings (download .json file from https://zenodo.org/records/18419061 and rename it to ror-data.json)

Generate Storage Data

Run the following to build the /storage structure used by the API:

docker run -it --rm   -v $HOME/tmp/storage:/storage   -p 8080:8080 -v $(pwd):/app   -w /app   orion   bash
go run json.go

This will create:

/storage/orgs/ → individual JSON files for each ROR ID

/storage/domains/ → hashed domain .txt files mapping to ROR IDs

Start the API Server

Locally:

go run api.go

Or in Docker:

docker build -t orion .

docker run -it --rm \
  -v $HOME/tmp/storage:/storage \
  -v $(pwd):/app \
  -w /app \
  -p 8080:8080 \
  orion bash

# then inside the container:
go run api.go

Testing the API

Once the server is running (either locally or inside Docker), you can test the endpoints using curl or tools like Postman.

Test 1: Search by Domain

This returns the list of ROR IDs associated with a domain.

Sample Request:

curl -X POST http://localhost:8080/submit \
  -H "Content-Type: application/json" \
  -d '{
        "cmd": "search_by_domain",
        "value": "dcc.ac.uk"
      }'

Sample Response

{
  "ids": ["01k9d6864"],
  "orgs": [
    {
    ...

    "id": "https://ror.org/01k9d6864",
    "links": [
      {
        "type": "website",
        "value": "http://www.dcc.ac.uk/"
      },
      {
        "type": "wikipedia",
        "value": "https://en.wikipedia.org/wiki/Digital_Curation_Centre"
      }
    ],

    ...

    "names": [
      {
        "lang": null,
        "types": [
          "acronym"
        ],
        "value": "DCC"
      },
      {
        "lang": "en",
        "types": [
          "ror_display",
          "label"
        ],
        "value": "Digital Curation Centre"
      }
    ],

    ...
  }
  ]

Test 2: Search by ROR ID

This returns the full JSON record of a single organization.

Sample Request:

curl -X POST http://localhost:8080/submit \
  -H "Content-Type: application/json" \
  -d '{
        "cmd": "search_by_ror_id",
        "value": "01k9d6864"
      }'

Sample Response:

[
  {
    ...

    "id": "https://ror.org/01k9d6864",
    "links": [
      {
        "type": "website",
        "value": "http://www.dcc.ac.uk/"
      },
      {
        "type": "wikipedia",
        "value": "https://en.wikipedia.org/wiki/Digital_Curation_Centre"
      }
    ],

    ...

    "names": [
      {
        "lang": null,
        "types": [
          "acronym"
        ],
        "value": "DCC"
      },
      {
        "lang": "en",
        "types": [
          "ror_display",
          "label"
        ],
        "value": "Digital Curation Centre"
      }
    ],

    ...
  }
]

Note: You can also send an array of ROR IDs to get multiple org records in one call.

Request format:

{
    "cmd": "search_by_ror_id",
    "value": ["rorID1", "rorID2"]
}

Response format:

[
  {
      ...
    "id": "https://ror.org/rorID1",
    "name": "First Organisation",
      ...
  },
  {
      ...
    "id": "https://ror.org/rorID2",
    "name": "Second Organisation",
      ...
  }
]

License

SPDX-License-Identifier: AGPL-3.0-or-later