File tools · CSV

Merge CSV files without uploading them.

Stack any number of CSV files into one. Columns are matched by header name, so files with the same columns in a different order still line up, and a report shows exactly what did not match.

Short version: choose the files, read the per-file report, download merged.csv. The merge is done in this tab; no file leaves your device.

How the merge works

Every file is decoded first (UTF-8, UTF-16 and Windows-1252 are detected automatically, and other encodings can be chosen by hand), then split into records with a parser that follows RFC 4180: a quoted field may contain commas, doubled quotes and line breaks, and a final line break does not create an empty record. The first record of each file is its header.

In header-name mode the output columns are the union of all headers, in the order they are first seen. A file that lacks a column gets blank cells there. If one file repeats a header, the second copy is kept as its own column named id (2) instead of overwriting the first. In position mode the first file’s header names are used and column 1 is stacked on column 1, which is what you want when the files are the same export with differently spelled headers.

Each file is sniffed for its own delimiter, so a semicolon-separated export from a European system and a comma-separated one from your own can be merged in one go. The output uses a single delimiter, line ending and encoding that you choose.

Your files stay on your device

Each file is read with your browser’s File API and merged by JavaScript running in this tab. The tool sits in a sandboxed frame whose content-security policy allows no outgoing connections, so even a bug could not send a row anywhere, and the advertising on this page lives outside that frame. Disconnect from the network after the page has loaded and the merge still works. There is no row limit built into the tool. The practical limit is your browser’s memory: a 100,000-row, 10-column file parses in well under a second in our automated test, while a file of several hundred megabytes may exhaust memory on a phone.

Worked example

File A (a.csv) has columns id,name with rows 1,Ann and 2,Bo. File B (b.csv) has name,id,city with the row Cy,3,Oslo. Matched by header name the result has the columns id,name,city, and three rows: 1,Ann, and 2,Bo, (city blank) and 3,Cy,Oslo. The report says file B added the column city and file A lacks it. A plain cat or copy of the two files would have put Cy under the id heading.

Reading the mismatch report

The table under the result lists, per file, its encoding and delimiter, how many data rows it contributed, which columns it introduced to the union and which union columns it lacks. Rows whose cell count differs from their file’s header are counted as ragged: short rows are padded with blanks and extra cells are kept in extra columns named Column N rather than dropped, because silently discarding data is worse than a surprise column.

Common mistakes

What this tool does not do

It does not remove duplicate rows, join files on a key, read Excel workbooks (save the sheet as CSV first), validate the data, or change any cell value. Use the duplicate-row remover afterwards if the files overlap, or the CSV validator if you suspect a file is malformed.

Frequently asked questions

Are my CSV files uploaded anywhere?

No. The files are read with your browser’s File API and merged by JavaScript in this tab. The tool runs in a sandboxed frame without network access, and the merged file is created locally when you press download.

What happens when the files have different columns?

In header-name mode the output has every column from every file. Cells a file does not have are left blank, and the per-file report lists which columns each file added or lacks.

Can I merge files with different delimiters or encodings?

Yes. Each file is checked for a byte order mark, then for valid UTF-8, then read as Windows-1252, and its delimiter is detected separately. You can override both under Input delimiter and Input encoding, which then apply to all selected files.

Is there a limit on rows or files?

The tool sets none. Everything is held in memory, so the limit is your device. Excel itself opens at most 1,048,576 rows per sheet, so a merged file larger than that will not open completely in Excel.

Will it remove duplicate rows?

No, merging keeps every row. If your files overlap, run the merged file through the duplicate-row remover and pick the key columns that identify a record.

Why does my merged file show garbled accents in Excel?

Excel often reads a double-clicked CSV in a legacy code page unless the file starts with a UTF-8 byte order mark. Pick “UTF-8 with byte order mark” in the output options; behaviour differs between Excel versions.

Does it keep line breaks inside quoted cells?

Yes. A cell such as an address with a line break inside quotes stays one cell, and is written back with its quotes so the next program reads it the same way.

Sources

Structure only

This page combines files as they are. It does not check that the values mean the same thing in each file or that the result suits the program you will import it into.

Advertisement
Advertisement