Clean and Validate CSV or TSV Files: find broken CSV rows before they break your import.

Data quality without guesswork

Start with a useful automatic check

A CSV import often fails before anyone knows whether the problem is the delimiter, quoting, headers or a value deep in the file. WoluTools first recognises UTF-8, UTF-8 with BOM or Windows-1252, then detects comma, semicolon, tab or pipe structure across logical records and shows the result before processing. A preview makes the chosen header row and columns visible. The baseline audit needs no schema and checks malformed structure, inconsistent field counts, empty or duplicate headers, blank rows, outside whitespace, duplicate records, control characters and text that spreadsheet software might execute.

Files are processed as streams up to 100,000 rows and 250 columns, with a 1 MB limit for each field, instead of loading the complete table into memory. A structurally ambiguous row is blocked and retained in the issue trail with its exact source row; the worker never guesses where a missing quote or delimiter belongs.

Content changes require a deliberate choice

Technical output normalisation—UTF-8, consistent line endings and RFC-style CSV quoting—does not change field meaning. Removing outer whitespace, blank rows or duplicate records does, so each action appears in the final confirmation before the job starts. Deduplication also needs a confirmed key and a choice to keep the first or last occurrence.

Optional column rules cover required values, text, integer, decimal, date, Boolean, email and URL types, minimum and maximum values, length, uniqueness, allowed values and a bounded regular expression. Dates and decimal separators stay explicit. The tool does not infer an uncertain regional date, create a missing value or transform a leading-zero identifier into a number.

Keep a verifiable trail after cleanup

The main cleaned CSV contains only the transformations you approved. A separate spreadsheet review copy prefixes recognized formula starters to reduce execution risk; no CSV mitigation is universal, so review it and keep it separate from machine imports. The ZIP adds an issue CSV, XLSX profile, reusable validation schema, README and a manifest with source hash, settings, counts and a SHA-256 inventory of every artifact. Completion and billing happen only after the ZIP passes CRC, exact-member, byte-size and hash verification.