CSV to TSV converter

Parsing the CSV before re-writing it

The parser reads quotes properly: a quoted field may contain commas, a line break inside quotes stays inside the cell, and two quotes in a row collapse to one. That is what makes the conversion safe for tables exported from spreadsheets, where free-text columns often contain commas.

The delimiter is detected heuristically from the first part of the file rather than assumed. If the head contains more semicolons than commas, semicolons are treated as the separator; tabs win over commas too. A file that mixes styles can therefore be read with the wrong separator, so check the row count reported after conversion against your own count.

• Quoted fields keep embedded commas and line breaks
• Doubled quotes inside a quoted field collapse to one quote
• Delimiter is guessed from the file head — comma, semicolon or tab
• A UTF-8 byte-order mark at the start is dropped

Blank lines disappear, empty cells survive

A line with no content at all is skipped, so trailing blank lines do not become empty rows. A row with separators but no text between them is kept, because those empty cells carry the column structure.

No header option here

Unlike CSV to JSON, this page has no "first row is a header" switch. The first row is treated as text and written out as the first row, exactly as it appeared. Nothing is renamed or typed.

How the tab output is written

Each cell is quoted only when it would otherwise break a tab-separated row: if it contains a tab, a quote character, or a line break. Ordinary commas and semicolons are written as plain text, so most rows come out unquoted, which is what strict TSV readers expect.

The result starts with a byte-order mark so spreadsheet software reads it as UTF-8. Accented Latin characters, Cyrillic and CJK text are preserved; the BOM can look odd when you open the file in a plain text editor.

• Quoted only if the cell contains a tab, quote or newline
• Commas and semicolons become ordinary cell text
• Rows are joined with CRLF line endings
• Output keeps the input name with a .tsv suffix

Cells that contain tabs

If a CSV field itself contains a tab, that cell is wrapped in quotes so the column structure survives. Some strict TSV tools do not understand quoted cells, so check the consumer before importing data with embedded tabs.

When this is the right conversion

Use it when a comma is ambiguous in your data — long free-text columns, addresses, prices with thousands separators — and you want a format that pastes cleanly into an editor or feeds a command-line tool without quoting rules.

It is also the quick fix when an importer splits rows on commas and your file already quotes them correctly: re-delimiting removes the ambiguity without editing the file by hand.

• Pasting rows into editors and terminal tools
• Database importers where commas appear inside text
• Preparing a table for a tool that expects plain TSV
• Nothing restructured: same rows, same columns, same text

If you need records instead of a table

To get keyed objects rather than a re-delimited table, use CSV to JSON, which offers header and indentation controls. This page deliberately leaves the shape alone.

Limits and what is not guaranteed

One file may be up to 50 MiB and you may convert up to 10 files in a batch. Everything runs in this browser tab, so a table containing customer data never leaves the machine — no upload, no account.

There is no CSV validation promise here: the parser is tolerant, and a malformed file is converted as it was read rather than rejected. The reported row count is the practical sanity check.

• Up to 10 files, 50 MiB each
• Row count is reported after conversion
• No type conversion: every cell remains text
• Heuristic delimiter detection can misread mixed-style files

Round trip is not byte-identical

Converting the TSV back to CSV re-quotes only what a comma would break, so quoting style, line endings and the byte-order mark differ from the original file. Cell content is preserved; formatting is not.

Questions