The DSpace CSV Metadata Quality Checker is a collection of sanity checks and automated fixes for a number of common issues in metadata files.

Select a CSV file to process (or try test.csv)

Skip fields(s)

Optionally indicate fields to skip during analysis. Separate multiple fields with a comma, for example: dcterms.issued,dcterms.subject.

Validate field(s) against AGROVOC

Optionally indicate fields to validate against AGROVOC. Separate multiple fields with a comma, for example: dcterms.subject,cg.coverage.country. Note: this can take an extra minute or more depending on your data. If you have a problem please try again and it will generally be faster the second time.

Enable "unsafe" fixes

This will remove newlines, perform normalization of Unicode characters, and attempt to fix mojibake character encoding issues. Read more about these unsafe fixes.

Enable experimental checks

Attempt to validate whether the value of an item's dc.language.iso or dcterms.language field matches the actual language of text used in its title, abstract, and citation. Read more about these experimental checks.