Dedupe-csv

​ A lightweight module designed to be run from the command line. By default, it takes a csv file, removes duplicate lines, and exports the unique lines to a new CSV. Also included, are optional parameters to check for duplicates across a subset of columns and to switch between keeping the first and last duplicates. ​ dedupe-csv is based on pandas' drop_duplicates function and has the same functionality. ​

Quick start

​ To use dedupe-csv open the terminal on the folder where the file is located. ​

Options

​ The dedupe-csv command accepts the following options: ​

• file: Specifies the input CSV file. Replace "data.csv" with the path to your CSV file.
• column: (Optional) if omitted it will look for identical rows. If specified, the header name in the CSV file will be used to identify duplicates. Replace "header name/s" with the header/s name you want to use. You can use multiple headers.
• keep: (Optional) Specifies the strategy for keeping duplicates. The default value is "first".
• Available options are:
  • "first": Keep the first occurrence of each duplicate.
  • "last": Keep the last occurrence of each duplicate.​

dedupe-csv file="data.csv" column="header name" keep="first" ​

Install

​ Using npm: ​

$ npm install -g dedupe-csv

​ Using yarn: ​

$ yarn global add dedupe-csv

​

Usage

Remove duplicate lines

This will remove any entries that are identical across all headers and keep the first entry found

$ dedupe-csv file="data.csv"

data.csv

brand,style,rating
Yum,cup,1
Yum,cup,1

Will output the following file:

data_deduped.csv

brand,style,rating
Yum,cup,1

Remove duplicates by column name

This will remove any entries that have the same duplicated column

$ dedupe-csv file="data.csv" column="brand"

data-column.csv

brand,style,rating
Yum,cup,1
Yum,pack,2
Foo,cup,1
Foo,pack,2

Will output the following file:

data-column_deduped.csv

brand,style,rating
Yum,cup,1
Foo,cup,1

Remove duplicates using multiple headers

This will remove any entries that match the same duplicated columns. To use it, add columns between quotes and separated by commas.

$ dedupe-csv file="data-multiple.csv" column="brand,style"

data-multiple.csv

brand,style,rating
Yum,cup,1
Yum,cup,2
Foo,cup,1
Foo,pack,3
Foo,pack,1

Will output the following file:

data-multiple_dedupe.csv

brand,style,rating
Yum,cup,1
Foo,cup,1
Foo,pack,3

Remove duplicates by header and keep last

This will remove any duplicated entries that match by column name and keep the last found ​ To use add the option keep="last"

$ dedupe-csv file="data.csv" column="brand" keep="last"

data-last.csv

brand,style,rating
Yum,cup,1
Yum,pack,2
Foo,cup,1
Foo,pack,2

Will output the following file:

data-last_deduped.csv

brand,style,rating
Yum,pack,2
Foo,pack,2