pullq: Pull request queue visualization

This is a tool inspired by review gator with a number of departures:

Github-only support
A departure in the presentation style
Decoupling of stats gathering and presentation
Additional dynamic filters (authors, repositories, labels, status)

Quickstart

A pre-built javascript application exist in the repo, so to test this program you can take these two steps:

Generate a data-file for your repositories
Look at the output

Generating the data-file

You will need to install Leiningen, the Clojure build tool. It will take care of gathering other dependencies for you (putting them in your ~/.m2 folder).

To generate the data file, you will need a config file, usually stored in pullq.conf, you will also need to provide a github token, either in the GITHUB_TOKEN environment variable, or by provide the -t command line flag.

# user repo minimum-oks
org awesome-repo 2
org some-repo 1

# other repositories
my-other-org repo3 1

You can then generate the data with:

lein run

Looking at the output

(cd build  && python -m http.server)

You can now browse to http://localhost:8000, have fun!

Generating the Javascript application

You will need leiningen. In the repository, run: lein cljsbuild once min

Development Mode

Start Cider from Emacs:

Put this in your Emacs config file:

(setq cider-cljs-lein-repl
	"(do (require 'figwheel-sidecar.repl-api)
         (figwheel-sidecar.repl-api/start-figwheel!)
         (figwheel-sidecar.repl-api/cljs-repl))")

Navigate to a clojurescript file and start a figwheel REPL with cider-jack-in-clojurescript or (C-c M-J)

Run application:

lein clean
lein figwheel dev

Figwheel will automatically push cljs changes to the browser. Wait a bit, then browse to http://localhost:3449.

Production Build

To compile clojurescript to javascript:

lein clean
lein cljsbuild once min

Caveats

This is extracted from internal tooling at Exoscale and might be a bit rough around the edges.

chrisglass / pullq