CSVSee

You are reading the documentation for CSVSee, a tool for manipulating and visualizing data in comma-separated (CSV) files.

Motivation

This tool was originally developed to help with analyzing test results coming from Grinder and Performance Monitor. It was partly inspired by Grinder Analyzer, which serves a similar purpose in a more specific domain.

Features

• Generate graphs of just about any numerical data, particularly those having timestamps
• Match one or more CSV column names with regular expressions
• Plot the top N data sets, by average or peak value within each column
• Automatic color-coding when graphing multiple columns
• Display graphs in an interactive viewer with zooming/panning capability
• Export graphs to a .png, .svg or .pdf file
• Customizable line styles, title / axis labels, and timestamp formats
• Automatic guessing of date/time format

License

This software is open source, under the terms of the simplified BSD license.

Installation

If you want to install an official release, first download one from the downloads page, and extract it somewhere.

Then, open that directory in a terminal and run:

sudo python setup.py install

Or use pip:

sudo pip install .

One advantage of using pip is that you can uninstall later like so:

sudo pip uninstall CSVSee

If you’d rather use a copy of the latest development version, branch it using Bazaar:

bzr branch lp:csvsee

then install as before using setup.py or pip.

Python is required. The graph command also depends on matplotlib and NumPy.

Usage

The primary frontend script is csvs. Run this from the command-line to see usage notes and integrated help, or see the documentation below:

• csvs
  • csvs graph
  • csvs grep
  • csvs grinder

API

All of the supporting libraries are in a module called csvsee. You can read the autogenerated documentation here:

• csvsee
  • csvsee.dates
  • csvsee.utils
  • csvsee.graph
  • csvsee.grinder
  • Testing