CSViser ADVISER

CSV validator — RFC 4180

Upload your CSV and get a full validation report: errors, warnings, and a column-by-column profile. Every issue includes a plain-English explanation and a one-click fix.

Upload your CSV to validate

CSV, XLSX, or JSON — any size

Free account — takes 10 seconds to sign up

Validation runs automatically after upload — no extra steps needed.

What does the validator check?

CSViser validates against RFC 4180, the closest thing to a formal CSV standard. The checks fall into three categories:

  • Errors — the file cannot be parsed reliably: unclosed quotes, inconsistent column counts per row, invalid byte sequences (encoding errors).
  • Warnings — the file parses but may cause problems downstream: non-standard delimiter (semicolon, tab), quoted fields with unescaped quotes, rows with trailing commas.
  • Info — observations about structure: detected delimiter, line-ending style (CRLF / LF), whether a header row is present, column count and fill rate.

Most common CSV errors explained

Unclosed quote

A field begins with a double-quote but the closing quote is missing or malformed. Usually caused by a value that contains a double-quote but was not properly escaped (RFC 4180 requires "" inside a quoted field). Example: Alice,"hello world — the quote opens but never closes.

Inconsistent number of fields

Some rows have more or fewer columns than the header. Common causes: a value with an unescaped comma, a row with a missing field, or a trailing comma in the header row. Parsers like pandas raise ParserError or silently misalign columns.

Invalid UTF-8 / encoding error

The file contains bytes that are not valid UTF-8. Typically from Windows-1251 or ISO-8859-1 exports. Fix with the encoding converter.

Mixed line endings (CRLF + LF)

The file mixes Windows-style line endings (\r\n) and Unix-style (\n). RFC 4180 specifies CRLF; mixing the two confuses parsers that read line-by-line. Common when files are concatenated from different sources.

What RFC 4180 doesn't cover

RFC 4180 defines syntax, not semantics. It says nothing about whether a date field is in the right format, whether an email is valid, or whether a number column contains only numbers. For that level of validation, use the Convert data tab in CSViser — it checks and normalises dates, emails, phone numbers, and numbers column by column.

Frequently asked questions

My file opens fine in Excel but fails validation — why?

Excel applies its own error-recovery heuristics when opening CSV files. It silently handles unclosed quotes and inconsistent column counts by guessing. A strict RFC 4180 parser used by a database, API, or data pipeline has no such tolerance — it errors on the first malformed row.

Can I fix the errors without downloading and re-uploading?

Yes. After validation, the Fix issues tab appears with one-click fixes for all detected problems: encoding conversion, BOM removal, line ending normalisation, blank row removal, and whitespace trimming. After fixing, the file is re-validated automatically.