A command-line tool and Python library that draws basic graphs in the terminal.
Graph types supported:
- Bar Graphs
- Color charts
- Multi-variable
- Stacked charts
- Histograms
- Horizontal or Vertical
- Calendar heatmaps
- Emoji!
$ termgraph data/ex1.dat
# Reading data from data/ex1.dat
2007: ▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 183.32
2008: ▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 231.23
2009: ▇ 16.43
2010: ▇▇▇▇ 50.21
2011: ▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 508.97
2012: ▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 212.05
2014: ▏ 1.00
from termgraph import Data, Args, BarChart
# Create data
data = Data([[10], [25], [50], [40]], ["Q1", "Q2", "Q3", "Q4"])
# Configure chart options
args = Args(
title="Quarterly Sales",
width=50,
format="{:.0f}",
suffix="K"
)
# Create and display chart
chart = BarChart(data, args)
chart.draw()Output:
# Quarterly Sales
Q1: ▇▇▇▇▇▇▇▇▇▇ 10K
Q2: ▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 25K
Q3: ▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 50K
Q4: ▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 40K
Use --scale to compare charts across multiple runs without changing the
number of units represented by each tick:
printf 'A 6\nB 12\n' | termgraph --scale 4 --format '{:g}'A: ▇▇ 6
B: ▇▇▇ 12
Each tick represents up to four units. Positive values round up to a whole
tick and zero draws no bar. The scale must be positive and finite, and the
data must be finite and non-negative. --scale overrides --width without
clipping bars. It supports horizontal, stacked, and vertical bar charts;
stacked segments round independently. It cannot be combined with
--different-scale, --histogram, or --calendar.
The same option is available in Python: BarChart(data, Args(scale=4)).draw().
Omitting it keeps the existing automatic scaling.
An example using emoji as custom tick:
termgraph data/ex1.dat --custom-tick "🏃" --width 20 --title "Running Data"
# Running Data
2007: 🏃🏃🏃🏃🏃🏃🏃 183.32
2008: 🏃🏃🏃🏃🏃🏃🏃🏃🏃 231.23
2009: 16.43
2010: 🏃 50.21
2011: 🏃🏃🏃🏃🏃🏃🏃🏃🏃🏃🏃🏃🏃🏃🏃🏃🏃🏃🏃🏃 508.97
2012: 🏃🏃🏃🏃🏃🏃🏃🏃 212.05
2014: 1.00
Note: Color charts use ANSI escape codes, so may not be able to copy/paste from terminal into other uses.
$ termgraph data/ex4.dat --color {cyan/yellow} --space-betweentermgraph data/ex7.dat --color {green,magenta} --stacked
Termgraph accepts counts before text labels, so you can pipe the output of
sort | uniq -c directly into it:
printf '%s\n' 'update settings' 'create' 'update settings' | sort | uniq -c | termgraphNo extra flag or column-swapping command is needed. For example, 2 update settings
is read as the label update settings with the value 2.
With whitespace-separated input, numeric words inside a label are also read as
values. For labels such as HTTP 404, use a delimiter to keep the entire label in
one field, for example 2,HTTP 404. A label that is entirely numeric is ambiguous
in count-first input; use the usual label-first order for those rows, such as
404,2. See file and standard-input parsing
for details.
Calendar Heatmap, expects first column to be date in yyyy-mm-dd
$ termgraph --calendar --start-dt 2017-07-01 data/cal.dat
-
Create a data file with labels and numeric values, separated by commas or whitespace. Text labels can come before or after the values. If every field is numeric, the first field is treated as the label. See the
sort | uniq -cexample for count-first input. -
termgraph [datafile]
-
Help: termgraph -h
usage: termgraph [-h] [options] [filename]
draw basic graphs on terminal
positional arguments:
filename data file name (comma or space separated). Defaults to stdin.
options:
-h, --help show this help message and exit
--title TITLE Title of graph
--width WIDTH width of graph in characters default:50
--scale SCALE fixed units per tick for non-negative bar charts; overrides --width
--format FORMAT format specifier to use.
--suffix SUFFIX string to add as a suffix to all data points.
--no-labels Do not print the label column
--no-values Do not print the values at end
--space-between Print a new line after every field
--color [COLOR ...] Graph bar color( s )
--vertical Vertical graph
--stacked Stacked bar graph
--histogram Histogram
--bins BINS Bins of Histogram
--different-scale Categories have different scales.
--calendar Calendar Heatmap chart
--start-dt START_DT Start date for Calendar chart
--custom-tick CUSTOM_TICK
Custom tick mark, emoji approved
--delim DELIM Custom delimiter, default , or space
--verbose Verbose output, helpful for debugging
--label-before Display the values before the bars
--no-readable Disable the readable numbers
--percentage Display the number in percentage
--version Display version and exit
All chart types are available as classes:
from termgraph import (
Data, Args,
BarChart, StackedChart, VerticalChart, HistogramChart
)
# Basic setup
data = Data([[10], [20]], ["A", "B"])
args = Args(title="My Chart")
# Choose your chart type
chart = BarChart(data, args) # Horizontal bars
# chart = StackedChart(data, args) # Stacked bars
# chart = VerticalChart(data, args) # Vertical bars
# chart = HistogramChart(data, args) # Histogram
chart.draw()📚 Complete Python API Documentation
For comprehensive examples, detailed API reference, and advanced usage patterns, see the complete documentation:
- Getting Started Guide - Examples and best practices
- Data Class API - Data preparation and validation
- Chart Classes API - All chart types with examples
- Args Configuration - Complete configuration options
Quick Args options:
title: Chart titlewidth: Width in characters (default: 50)scale: Fixed units per tick for non-negative bar charts (default: None)format: Number format string (default: "{:<5.2f}")suffix: Add suffix to all valuesno_labels: Don't show labelsno_values: Don't show valuescolors: List of color names
I wanted a quick way to visualize data stored in a simple text file. I initially created some scripts in R that generated graphs but this was a two step process of creating the graph and then opening the generated graph.
After seeing command-line sparklines I figured I could do the same thing using block characters for bar charts.
All contributions are welcome! For detailed information about the project structure, development workflow, and contribution guidelines, please see CONTRIBUTING.md.
Quick Start:
- 🐛 Bug reports and 🚀 feature requests: Use GitHub Issues
- 🔧 Code contributions: See our development workflow
- 📚 Documentation: Help improve our guides and examples
Code Quality: We use ruff for linting and formatting, mypy for type checking, and maintain comprehensive test coverage.
Thanks to all the contributors!
MIT License, see LICENSE.txt