Free web tool
Normalize CSV headers to snake_case, camelCase, kebab-case, or safe identifier forms.
Turn headers such as Product Name and Unit Price (JPY) into consistent names such as product_name or productName. Only the header row changes; the data rows do not.
Spaces, parentheses, full-width punctuation, and mixed naming styles can make headers awkward in APIs, scripts, and databases. The tool first applies Unicode NFKC normalization, separates words, then rebuilds the name in the selected style.
A lowercase letter or digit followed by an uppercase letter creates a word boundary before tokenization. So productName can be understood as “product name” and rebuilt as product_name, productName, or another selected form.
If a header becomes empty after normalization, it falls back to column. If multiple headers normalize to the same name, later ones get suffixes such as _2 and _3 so the output header stays unique.
In the current implementation, safe_identifier also joins tokens with underscores, so it looks close to snake_case. It does not promise compatibility with every database or language; reserved words and destination-specific naming rules still belong to the destination system.
Cell values, row order, and column order stay untouched. That makes this useful as a preparation step when the data itself is already correct but the field names need to be machine-friendly.