shinyburndown

Generates "burndown" type charts about issues in your GitHub repo. Available as both a Shiny web app and a command-line tool.

Live demo: https://colindiesh.shinyapps.io/github_burndown/

Screenshots

Simple graph of 'open issues over time'. Note that it includes open PRs as 'open issues'

This is my favorite chart where color reveals how long an issue has been open for, and then dots below the 'main curve' show old issues being closed. This can help show how both old and new issues are being solved. Note that some vertical stripes in this plot are basically closing stale issues

Installation

Install dependencies

install.packages(c(
  "tidyverse",
  "lubridate",
  "gh",
  "glue",
  "cli",
  "fs",
  "shiny",
  "plotly"
))

Or install as a package (from the repo directory):

# Install devtools if needed
install.packages("devtools")
# Install shinyburndown and all dependencies
devtools::install()

Usage

Shiny App

Run the interactive web app:

shiny::runApp()

Or from the command line:

Rscript -e "shiny::runApp()"

Command Line

Rscript burndown_cli.R <org/repo> [options]

Examples:

# Basic usage
Rscript burndown_cli.R GMOD/jbrowse-components
# Force refresh cached data
Rscript burndown_cli.R GMOD/jbrowse-components --refresh
# Custom output directory
Rscript burndown_cli.R GMOD/jbrowse-components --output-dir=plots
# Show summary only, skip plot generation
Rscript burndown_cli.R GMOD/jbrowse-components --no-plots
# Use colorblind-friendly viridis palette
Rscript burndown_cli.R GMOD/jbrowse-components --viridis

Options:

Option	Description
`--refresh`	Force refresh of cached data
`--output-dir=DIR`	Directory for output plots (default: `output`)
`--no-plots`	Skip generating plots, just show summary
`--viridis`	Use colorblind-friendly viridis color palette
`--width=N`	Plot width in inches (default: 8)
`--height=N`	Plot height in inches (default: 4)

GitHub Token

For repos with many issues, you may hit GitHub API rate limits. Set a token:

export GITHUB_TOKEN=your_token_here

Or create a .Renviron file:

GITHUB_TOKEN=your_token_here

Generate a token at: https://github.com/settings/tokens

Caching

The CLI caches fetched data in cache/<org>_<repo>.rds. Subsequent runs load from cache unless --refresh is passed.

Output

The CLI generates 7 plots in both PDF and PNG formats:

months_vs_timespan - Months to complete vs time span covered
issues_timeline - Issues completed over time
completion_time - Close date vs months taken
close_date - Close date vs issue number
completion_histogram - Distribution of time to complete
open_issues - Total issues open over time
closed_per_day - Issues closed per day distribution

Deploy to shinyapps.io

library(rsconnect)
deployApp()