ADS-mock-events.md

The ADS events are illustrated in a diagram at the bottom of this document

Event: ADS discovers that a journal article cites a Zenodo (software) record (not yet in the ADS database)

Little bit of background: the ADS reference resolving process stores data in a file of all citations where a publication in the ADS Universe has a reference string in its bibliography with one of a set of identifiers (doi:, ascl:, Github or Bitbucket URLs). In the ADS back-office, a pipeline ingests this data and tries to determine whether the object corresponding with the identifier represents software; currently this means that metadata is harvested for a DOI and it is checked whether the resource type is "software". If the pipeline determines that the DOI found corresponds with a software object, it constructs a payload that will be sent to the ADS webhook service, which will then send the following JSON data to all callback URLs that subscribed to the event relation_created:

{
  "time": "1441166640.359496",
  "id": "d969a56d-e520-405d-a24f-497ac6923781",
  "creator": "ADS",
  "payload": [
    {
      "source": {
        "identifier": {
          "id_url": "http://adsabs.harvard.edu/abs/",
          "id": "2016ApJ...818..156C",
          "id_schema": "bibcode"
        }
      },
      "relationship_type": {
        "scholix_relationship": "references",
        "original_relationship_name": "Cites",
        "original_relationship_schema": "DataCite"
      },
      "target": {
        "type": {
          "name": "software"
        },
        "identifier": {
          "id_url": "https://doi.org",
          "id": "10.5281/zenodo.11020",
          "id_schema": "DOI"
        }
      },
      "license_url": "https://creativecommons.org/publicdomain/zero/1.0/"
    }
  ],
  "source": "ADS.Discovery",
  "description": "ADS citation events",
  "event_type": "relation_created"
}

Event: The Zenodo (software) record is added to the ADS index

The metadata harvested for the Zenodo (software) record in the ADS back-office pipeline is submitted to the indexing process. As a result, a bibcode is minted for this Zenodo record. The pipeline (which is different from the one above) prepares a payload and submits this to the webhook service, which then sends out the following relation_createdto all callback URLs that subscribed to this event type:

{
  "time": "1441166640.359496",
  "id": "d969a56d-e520-405d-a24f-497ac6923781",
  "creator": "ADS",
  "payload": [
    {
      "source": {
        "identifier": {
          "id_url": "http://adsabs.harvard.edu/abs/",
          "id": "2016zen.soft123456X",
          "id_schema": "bibcode"
        }
      },
      "relationship_type": {
        "original_relationship_name": "IsIdenticalTo",
        "original_relationship_schema": "DataCite"
      },
      "target": {
        "type": {
          "name": "software"
        },
        "identifier": {
          "id_url": "https://doi.org",
          "id": "10.5281/zenodo.11020",
          "id_schema": "DOI"
        }
      },
      "license_url": "https://creativecommons.org/publicdomain/zero/1.0/"
    }
  ],
  "source": "ADS.Discovery",
  "description": "ADS citation events",
  "event_type": "relation_created"
}

Event: ADS discovers that a journal article cites a Zenodo (software) record (already in the ADS database)

This will result in the emission of a relation_added event just like the one earlier. Since an IsIdenticalTo event has been transmitted, equating the Zenodo DOI with the ADS bibcode for that record, the broker will recognize this equivalency.

Event: A bibcode changes in the ADS

Currently not supported as event.