csvkit is a toolkit for inspecting, dissecting, and manipulating delimited files (CSVs, pipe-delimited, tab-separated, etc.) in furtherance of data preparation and loading. This image lets you run csvkit without having to install Python, pip, and csvkit itself.
(provided that you have Docker available...)
To run it:
docker run --rm -it filefrog/csvkit [options]
This docker image is built such that any commands you pass it are
interpreted as csv* sub-commands. I.e. look means "run the
csvlook command". You can of course just call it with a full
and correct command to run, but I'm lazy.
By default, the container operates in the /csv working
directory; you'll most likely want to mount this so that your data
files are visible to csvkit processes. This works:
docker run --rm -it \
--mount "type=bind,source=$PWD,destination=/csv,ro=true \
filefrog/csvkit [options]
It's good practice to make the mountpoint readonly if you aren't expecting to munge the CSV files in-place.
In some sense, docker run ... filefrog/csvkit look can be
thought of as an alias for the actual csvlook command, except
you don't have to install the software, or its dependencies.
This repo also comes with a bin/ directory that contains a
shortcut shell script that you can symlink to all of the csv*
utility names to make the illusion completely seemless.
cp bin/csvkit ~/bin
for x in clean cut format grep join json \
look py sort sql stack stat; do
ln -sf ~/bin/csvkit ~/bin/csv$x
done