Skip to main content
Version: 15.0.1

Standalone Validator

This feature allows you to validate one of your organized datasets stored locally or on Pennsieve. This can save you time by allowing you to avoid stepping through Organize dataset in Upload Dataset or Prepare Dataset Step-by-Step UIs when validating your dataset is your only goal.

Background

All SPARC datasets must conform to the SPARC Dataset Structure format described here. It is recommended that, when possible, you resolve any validation errors before submitting your dataset to the SPARC Curation Team for review.

How to

To access this interface click on the Advanced Features at the bottom of the homepage, then click on the Validate dataset card.

  • First choose which validation workflow you will perform:
    • Pennsieve validation: For validating a dataset stored on Pennsieve.
    • Local validation: For validating a dataset stored on your computer or a local drive.
  • Pennsieve validation workflow:
    • Select which dataset you would like to validate from the dropdown menu.
    • Click Confirm to import the dataset.
    • Once imported, click the Validate Dataset button to begin the validation process.
    • Once validation is complete you will be presented with a table of results with two columns.
      • Path: The Path column will contain the path to the file, folder, or field that was validated.
      • Message: The Message column will contain the message informing you what was determined to be invalid about the file, folder, or field specified in the Path column.
    • To fix your errors, simply alter your dataset to correct the given error messages then re-run validation for confirmation.
  • Local validation workflow:
    • Select the location of the local dataset you would like to validate.
    • Click Confirm to finalize your selection.
    • Click Validate Dataset to begin the validation process.
    • Once validation is complete you will be presented with a table of results with two columns.
      • Path: The Path column will contain the path to the file, folder, or field that was validated.
      • Message: The Message column will contain the message informing you what was determined to be invalid about the file, folder, or field specified in the Path column.
    • To fix your errors, simply alter your dataset to correct the given error messages then re-run validation for confirmation.

Was this page helpful?