splitcsv

SplitCSV

Split a CSV file into one or more parts. The header is written to every output file; data rows are distributed round-robin.

Features

• Round-robin distribution — rows are spread evenly across output files
• Header in every file — each part is a valid, self-contained CSV
• One or more outputs — single output to copy/convert (e.g. change delimiter), or many to split
• Configurable delimiter — optional -delimiter for semicolon- or tab-separated files
• Split into N parts — use -n N -o <dir> to create part1.csv … partN.csv in a directory
• Streaming — processes row-by-row; constant memory even for very large files
• Standard library only — no external dependencies
• Single binary — easy to install and script

Install

go install github.com/dmitryravilov/splitcsv/cmd/splitcsv@latest

Or build from source:

git clone https://github.com/dmitryravilov/splitcsv.git
cd splitcsv
go build -o splitcsv ./cmd/splitcsv

Usage

Explicit output files:

splitcsv [-delimiter c] <input.csv> <out1.csv> [out2.csv ...]

Split into N parts (outputs go to a directory):

splitcsv -n N -o <output_dir> [-delimiter c] <input.csv>

Examples

# Split into three parts (explicit filenames)
splitcsv data.csv part1.csv part2.csv part3.csv

# Split into five parts in a directory
splitcsv -n 5 -o ./chunks data.csv

# Single output (copy with header)
splitcsv data.csv single.csv

# Semicolon-delimited input and output
splitcsv -delimiter ';' european.csv out1.csv out2.csv

Requirements

Go 1.22 or later.

License

MIT