[RC Diary] Pongo and Perdita (-101)

blog-post.md

[RC Diary] Pongo and Perdita (-101)

Thoughts about a week of daily blogging

I feel good about this. One of the things I've always struggled with is the lack of structure in what I am doing, so I feel like this approach is giving good results for me:

• some time in the evening to recap
• write what I plan to do the next day
• the next day get the plans and work on those if they still make sense

Feels like the exact amount of structure I needed.

The little schemer

$ sudo apt-get install mit-scheme

I am tracking my progress here: https://github.com/lazywithclass/the-little-schemer

Move single blog posts to gist

So it's easier to maintain and write, also comments! Also I think they do come up in Google searches, even though the SEO is far from optimal.

Sadly thre was no easy way to move them... will go hand by hand when I will have time and won't need brain power.

Cerebro

Calling modules

I figured out how to call modules within ClojureScript

;; a.cljs
(ns cerebro.mutations-reader
  (:require [cljs.nodejs :as node]))

(defn answer
  ""
  []
  42)

(set! (.-exports js/module) answer)

;; b.cljs
(ns cerebro.core
  (:require [cljs.nodejs :as node]
            [cerebro.mutations-reader :as reader]))

;; required to print
(node/enable-util-print!)
(println (reader/answer))

Configuring expression evaluation inside Spacemacs

It wasn't working for me at the start, but it all went together pretty ok.

cider-jack-in-clojurescript

I initially an error that looked something like

ClassNotFoundException cemerick/piggieback

So I went and google for it, looks like a library that allows to have a ClojureScript REPL on top of nREPL, which is the Clojure one AFAIK.

I've found this documentation and followed the steps, basically I put

:profiles {:dev {:dependencies [[com.cemerick/piggieback "0.2.1"]
                                [org.clojure/tools.nrepl "0.2.10"]]
                 :repl-options {:nrepl-middleware [cemerick.piggieback/wrap-cljs-repl]}}}

in my project.clj, then M-x customize-variable RET cider-cljs-lein-repl, so I could have nodejs as the default is rhino and I don't want that.

Now I can enjoy SPC-m-e-e, such a great way to code.

Setup a server to host the functional programming website

Making sure I can run try.purescript.org on my laptop

I installed cabal (part of the haskell-platform in Ubuntu), then cabal update told me some packages where missing:

• Glob -any,
• aeson -any,
• http-types >=0.8.5,
• purescript ==0.10.5,
• scotty -any

I proceeded installing all of them

$ cabal install glob aeson http-types purescript scotty

and then

$ dist/build/trypurescript/trypurescript 8081 'bower_components/purescript-*/src/**/*.purs'

correctly started the server, which I tested with

curl -X POST localhost:8081/compile --data '1 + 1'

which errored, but that means I got it working :b

Host setup

Looks like there's no need for a server, I will have to dive more into the details.

Have a look at purescript/trypurescript#64

Plans

• see if I can deploy the code onto S3 or if we need a server (assuming we goodthe PureSript way)
• lookup timsort heapsort
• continue working on Cerebro
• chapter 2 of the little schemer