CSVFixLab
Delimiter

Detect and fix CSV delimiters

Find the actual field separator and normalize a structurally valid table to comma-separated CSV without touching the business values.

Detect and fix CSV delimiters
Detect the parsed separator, verify row width, and enable comma re-export only when the table structure is safe.
Local browser processing
Drop a CSV file hereor choose a file from your device
Auto mode confirms UTF-8 exactly. If the file is not UTF-8, choose its source encoding before downloading a repair.
Analyzing…
—

CSV report

Health score reflects checks on this page; it does not guarantee import acceptance.

—data rows
—columns
—delimiter
—encoding
Download is enabled only when the parser can safely re-serialize the file without guessing the source encoding or missing business data.

Why a .csv file may not use commas

The extension .csv does not guarantee that commas separate the fields. Regional spreadsheet settings frequently export semicolons; data pipelines may use tabs or pipes. The problem appears when the receiving importer assumes a different separator.

Delimiter problems versus commas inside data

A correct CSV can contain commas inside a field when that field is quoted. For example "Seoul, Korea" is one field, not two. Changing every semicolon or comma with Find & Replace is therefore unsafe because it ignores CSV quoting rules.

A reproducible example

Every row below has three semicolon-separated fields. A parser can prove the table shape and safely serialize the same values with commas.

Problem CSV
name;price;sku
Blue Shirt;19.99;SKU-001
Red Shirt;24.50;SKU-002
Structurally corrected CSV
name,price,sku
Blue Shirt,19.99,SKU-001
Red Shirt,24.50,SKU-002

What the checker can prove from the file

  • The detected delimiter is semicolon.
  • Each data row has the same three-field width as the header.
  • No quote parser error makes the field boundaries ambiguous.
Safe next action

Re-serialize the parsed fields with commas and standard CSV quoting when your destination expects comma-separated input.

Do not guess

Do not globally replace punctuation in the raw text when fields may contain quoted commas, semicolons, tabs, or line breaks.

Reproduce the behavior

Semicolon delimiter lab caseDownload a three-column fixture and compare the normalized output.

Read the underlying rule

CSV delimiters explained →Quoted fields and broken quotes →

What can make normalization unsafe

If a row has too many or too few fields, or a quoted field is unclosed, the parser may not know where one field ends and another begins. In that situation the delimiter is not the only problem and CSVFixLab disables the repair instead of producing a confident-looking but potentially shifted file.

Quick troubleshooting sequence

  1. Check the reported delimiter.
  2. Confirm the expected number of columns.
  3. Review any quote or column-count error before downloading.
  4. If repair is enabled, download the comma-separated copy.
  5. Spot-check fields that contain punctuation or line breaks.