Nothing Special   »   [go: up one dir, main page]

Skip to content

Names

Names are a specific type of vocabulary, which represent a records' creator or contributor. For example, the deposit form offers search as you type suggestions and auto-fills the corresponding information.

You can find information behind the design and usage of the vocabulary in RFC 0054.

Data model

A Name record contains:

  • The name of the author split in family name and given name, or a single name attribute with the full name formatted as <family name>, <given name>. Note that if family name and given name are present they will overwrite name.
  • A list of identifiers, composed by their identifier value and scheme. The scheme can potentially be autocompleted if it is known by the idutils library (e.g. ORCID).
  • A list of affiliations, which can be represented by its name or, if it belongs to the Affiliations vocabulary, by its id.
- family_name: Carberry
  given_name: Josiah
  id: 0000-0002-1825-0097
  identifiers:
    - identifier: https://orcid.org/0000-0002-1825-0097
      scheme: orcid
  affiliations:
    - name: Wesleyan University
- name: Haak, Laurel L
  id: 0000-0001-5109-3700
  identifiers:
    - identifier: https://orcid.org/0000-0001-5109-3700
  affiliations:
    - id: 04fa4r544

How to import and update your name records

InvenioRDM ships with an example set of names and ORCID identifiers. To disable these from being loaded, create a blank file called app_data/vocabularies/names_orcid.yaml.

The Names vocabulary uses the new DataStreams API for processing vocabularies. You can find more information about this new API in RFC 0053.

Use of the vocabularies command

Name records will not be managed by the usual invenio rdm-records fixtures command, but instead a set of invenio vocabularies ... commands.

There are several ways to import names records. The most straight forward is by a DataStream definition file, where you will define how entries from a data source will be read, if they need any transformation, and finally where they should be written to.

For a simple import you can read entries from a YAML file with raw metadata objects, skip transformations, and use a service API to write and persist the entries to the database. Here is an example of this definition file, lets call it vocabularies-future.yaml:

names:
  readers:
    - type: yaml
      args:
        origin: "./app_data/names.yaml"
  writers:
    - type: names-service
      args:
        update: false

Finally, to run an import using this vocabularies-future.yaml file you can call the vocabularies import command:

invenio vocabularies import \
  --vocabulary names \
  --filepath ./vocabularies-future.yaml

In addition, you can also update vocabulary records in case you updated the source data file using the vocabularies update command:

invenio vocabularies update \
  --vocabulary names \
  --filepath ./vocabularies-future.yaml

Creating a names.yaml file

The Names vocabulary has been implemented with the ORCID public dataset as a possible source to import entries from. This means that the functionality to read entries from this format is already available. For example, you can use the vocabularies convert command to convert this dataset into a YAML file with the appropriate names metadata format:

invenio vocabularies convert \
  --vocabulary names \
  --origin /path/to/ORCID_2021_10_summaries.tar.gz \
  --target names.yaml

Alternatively, you can simply import it directly:

Long and blocking operation

Note that the import process is done synchronously and the ORCID dataset is very large. Therefore, this operation can take a long time.

invenio vocabularies import \
  --vocabulary names \
  --origin /path/to/ORCID_2021_10_summaries.tar.gz