csvpivot

CSV Pivot

Pivot tables for CSV files in the terminal.

Tested on Python 3.6 and 2.7.

Installing

pip install csvpivot

Usage

Say you have a CSV file such as:

name,country,gender,salary
Oliver,UK,M,10000
Jack,UK,M,21000
Emily,UK,F,32000
Harry,UK,M,43000
Adam,France,M,54000
Paul,France,M,65000
Louise,France,F,76000
Alice,France,F,87000
Emma,Germany,F,98000

We could then find the average salary in each country:

$ csvpivot test.csv --rows country --values 'mean(salary)'

country,mean(salary)
France,70500
Germany,98000
UK,26500

It would be useful to find out the maximum and minimum values too though:

$ csvpivot test.csv --rows country --values 'mean(salary)' 'min(salary)' 'max(salary)'

country,mean(salary),min(salary),max(salary)
France,70500,54000,87000
Germany,98000,98000,98000
UK,26500,10000,43000

As well as mean, min, and max, CSV Pivot also supports median, sum, stddev, count, countuniq, concat, and concatuniq. All require numerical values apart from the last two. If numbers contain commas they are interpreted as thousands separators and removed.

Columns are also supported. So we could break down our data by gender:

$ csvpivot test.csv --rows country --values 'mean(salary)' --columns gender

country,mean(salary):F,mean(salary):M
France,81500,59500
Germany,98000,
UK,32000,24666.666666666668