Skip to content

Dataset publishing prevented or slowed by errors related to possible validation of metadata entered in Geospatial Bounding Box fields #9421

@jggautier

Description

@jggautier

What steps does it take to reproduce the issue?

  • When does this issue occur?
    When using a web browser to try to create or edit a dataset version that includes "invalid" data in the Bounding Box fields in the Geospatial metadata block.

    And when trying to create a dataset version using the Native API's "Create a dataset in a Dataverse collection" and "Import a Dataset into a Dataverse Collection" endpoints while the JSON file that's passed in the API call includes "invalid" data in Geospatial Bounding Box fields.

  • Which page(s) does it occurs on?
    The dataset create and edit pages.

  • What happens?
    When using the API endpoints, the dataset version can't be created or updated. The endpoints return two error messages, one error message when there are non-numerical values in any of the Geospatial Bounding Box fields, and another error message when the number in the North Latitude field is less than the number in the South Latitude field.

    When using a web browser, the dataset can be saved, eventually. After clicking Save Changes I get the error "Error – The metadata could not be updated. If you believe this is an error, please contact Demo Dataverse Support for assistance." but I can see that the metadata was updated.

    The Publish button doesn't work until I refresh the page, which removes the error message. Then I can publish the dataset.

    And the dataset URL includes "&version=DRAFT" while the page shows the Info banner that reads that a draft version wasn't found. This might be related to an already documented bug.

  • To whom does it occur (all users, curators, superusers)?
    Superusers, curators, depositors

  • What did you expect to happen?
    Has validation been added to the Geospatial Bounding Box fields, maybe as part of the Geospatial Search added in v5.13 (Geospatial search #8239)? If it has, the error message in the GUI should make clear what's preventing the dataset version from being saved. The API endpoints return more specific error messages, but I'm not sure if they are clear to others.

Which version of Dataverse are you using?
v5.13

Metadata

Metadata

Assignees

No one assigned

    Type

    No type

    Projects

    Status

    No status

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions