jsonschema_dataset.json

{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "https://datalad.org/catalog.dataset.schema.json",
  "title": "dataset",
  "description": "A dataset in a DataLad Catalog",
  "type": "object",
  "properties": {
    "type": {
      "description": "The type of node",
      "title": "Type",
      "type": "string",
      "pattern": "dataset"
    },
    "dataset_id": {
      "description": "The dataset ID",
      "title": "Dataset ID",
      "type": "string"
    },
    "dataset_version": {
      "description": "The dataset VERSION",
      "title": "Dataset version",
      "type": "string"
    },
    "name":  {
      "description": "The long name of the dataset",
      "title": "Name",
      "type": "string"
    },
    "short_name":  {
      "description": "The short name of the dataset",
      "title": "Short name",
      "type": "string"
    },
    "alias": {
      "description": "An alias of the dataset, used for shortened URL access within the catalog",
      "title": "Alias",
      "type": "string"
    },
    "description": {
      "description": "A 1-2 paragraph description of the dataset",
      "title": "Description",
      "type": ["array", "string"]
    },
    "doi": {
      "description": "The dataset's digital object identifier",
      "title": "DOI",
      "type": "string"
    },
    "url": {
      "description": "The location of the datalad dataset's annex",
      "title": "DataLad URL",
      "type": ["array", "string"],
      "items": {
        "type": "string"
      }
    },
    "download_url": {
      "description": "The URL at which the complete dataset can be downloaded directly, typically via HTTP protocol and without further access restrictions",
      "title": "Download URL",
      "type": ["array", "string"],
      "items": {
        "type": "string"
      }
    },
    "homepage_url": {
      "description": "A unique URL with content that describes the current dataset and which may provide further data access",
      "title": "Homepage URL",
      "type": ["array", "string"],
      "items": {
        "type": "string"
      }
    },
    "license": {
      "description": "The license under which the dataset is made available",
      "type": "object",
      "properties": {
        "name": {
          "description": "The license name",
          "title": "Name",
          "type": "string"
        },
        "url": {
          "description": "A URL where a description of the license can be viewed",
          "title": "URL",
          "type": "string"
        }
      },
      "additionalProperties": false,
      "dependentSchemas": {
        "name": {
          "required": ["name"]
        },
        "url": {
          "required": ["name"]
        }
      }
    },
    "authors": { "$ref": "https://datalad.org/catalog.authors.schema.json"},
    "access_request_contact" : {
      "description": "The contact details of the person/office from which to request access to the dataset content",
      "title": "Access Request Contact",
      "type": "object",
      "properties": {
        "givenName": {
          "description": "The author's given name, such as 'John'",
          "title": "Given name",
          "type": "string"
        },
        "familyName": {
          "description": "The author's last/family name, such as 'Doe'",
          "title": "Family name",
          "type": "string"
        },
        "name": {
          "description": "A concatenation of the honorificSuffix, givenName and familyName properties, such as 'Dr. John Doe'",
          "title": "Name",
          "type": "string"
        },
        "email": {
          "description": "Email address of the author, such as johndoe@example.com",
          "title": "Email",
          "type": "string",
          "format": "email"
        },
        "honorificSuffix": {
          "description": "Title of the author, such as 'Dr.'",
          "title": "Honorific suffix",
          "type": "string"
        },
        "identifiers": {
          "description": "Identifiers for the person",
          "title": "Identifiers",
          "type": "array",
          "items": {
            "type": "object",
            "properties": {
              "type": {
                "description": "Type of identifier, such as 'ORCID'",
                "title": "Type",
                "type": "string"
              },
              "identifier": {
                "description": "The actual identifier, such as the ORCID ID '0000-1212-5566'",
                "title": "Identifier",
                "type": "string"
              }
            }
          }
        }
      },
      "required": ["givenName", "familyName", "email"]
    },
    "access_request_url": { 
      "type": "string",
      "description": "The url of the online location from which to request access to the dataset content",
      "title": "Access Request URL",
      "format": "uri"
    },
    "keywords": {
      "description": "Tags or keywords describing the dataset",
      "title": "Keywords",
      "type": "array",
      "items": {
        "type": "string"
      },
      "uniqueItems": true
    },
    "funding": {
      "description": "Sources of funding for the dataset",
      "title": "Funding",
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "name": {
            "description": "Name of the funding source, such as a funder or grant scheme",
            "title": "Name",
            "type": "string"
          },
          "identifier": {
            "description": "Identifier of the fund, such as a grant number",
            "title": "Identifier",
            "type": "string"
          },
          "description": {
            "description": "Free form description of grant or funding",
            "title": "Description",
            "type": "string"
          }
        },
        "required": []
      },
      "uniqueItems": true
    },
    "publications": {
      "description": "Publications related to the dataset",
      "title": "Publications",
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "type": {
            "description": "Type of publication, such as a scholarly article, book, blog post",
            "title": "Type",
            "type": "string"
          },
          "title": {
            "description": "Title of the publication",
            "title": "Title",
            "type": "string"
          },
          "doi": {
            "description": "The publication's digital object identifier",
            "title": "DOI",
            "type": "string"
          },
          "datePublished": {
            "description": "The publication date year",
            "title": "Date published",
            "type": ["number", "string"]
          },
          "publicationOutlet": {
            "description": "The publication outlet / venue, such as the journal, publisher name, or news outlet",
            "title": "Publication outlet",
            "type": "string"
          },
          "authors": { "$ref": "https://datalad.org/catalog.authors.schema.json"}
        },
        "required": ["title", "doi", "authors"]
      },
      "uniqueItems": true
    },
    "subdatasets": {
      "description": "Subdatasets of the current dataset",
      "title": "Subdatasets",
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "dataset_id": {
            "description": "The subdataset ID",
            "title": "Dataset ID",
            "type": "string"
          },
          "dataset_version": {
            "description": "The subdataset VERSION",
            "title": "Dataset version",
            "type": "string"
          },
          "dataset_path": {
            "description": "The subdataset PATH relative to its parent",
            "title": "Dataset path",
            "type": "string"
          }
        },
        "required": ["dataset_id", "dataset_version", "dataset_path"]
      },
      "uniqueItems": true
    },
    "metadata_sources": {"$ref": "https://datalad.org/catalog.metadata_sources.schema.json"},
    "additional_display": {
      "description": "Additional items to display in tabs on dataset page",
      "title": "Additional display",
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "name": {
            "description": "Name of the section that will appear as the tab name",
            "title": "Name",
            "type": "string"
          },
          "content": {
            "description": "The content that will appear in the tab when opened, specified as key-value pairs",
            "title": "Content",
            "type": "object"
          },
          "icon": {
            "description": "An icon from the Font Awesome Free collection (v6.1.1), e.g. 'fas fa-database', that will be displayed in the tab heading",
            "title": "icon",
            "type": "string"
          }
        },
        "required": ["name", "content"]
      },
      "uniqueItems": true
    },
    "top_display": {
      "description": "Additional items to display at the top of the dataset page (along with keywords, description, etc)",
      "title": "Top display",
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "name": {
            "description": "Name of the item that will appear as part of the top display",
            "title": "Name",
            "type": "string"
          },
          "value": {
            "description": "Value of the item that will appear as part of the top display",
            "title": "Value",
            "type": ["number", "string"]
          }
        },
        "required": ["name", "value"]
      },
      "uniqueItems": true,
      "maxItems": 5
    },
    "notebooks": {
      "description": "Jupyter notebooks that will be linked to from the 'Explore with Binder' button and will typically be used to demonstrate the use of the dataset",
      "title": "Jupyter notebooks",
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "git_repo_url": {
            "description": "The URL to a publicly accessible git repository containing the notebook",
            "title": "Repo URL",
            "type": "string"
          },
          "notebook_path": {
            "description": "The path to the notebook, relative to the repository root, including the filename and extension",
            "title": "Notebook path",
            "type": "string"
          }
        },
        "required": ["git_repo_url", "notebook_path"]
      },
      "uniqueItems": true
    }
  },
  "required": [ "type", "dataset_id", "dataset_version", "metadata_sources"]
}