hdgarrood

This gist previously contained a draft version of my post announcing the malicious code in the purescript npm installer.

You can now find the published post on my blog: https://harry.garrood.me/blog/malicious-code-in-purescript-npm-installer

hdgarrood

'use strict';

const {dirname, isAbsolute, join, resolve} = require('path');
const {existsSync} = require('fs');
const {PassThrough} = require('stream');

const inspectWithKind = require('inspect-with-kind');
const npmCliDir = require('npm-cli-dir');
const optional = require('optional');
const resolveFromNpm = require('resolve-from-npm');

hdgarrood

Haskell projects with lots of dependencies on Travis CI

If your Haskell project has lots of dependencies, you can find yourself reaching the timeout while compiling them - the free service has a timeout of 50 minutes. How do we address this?

First, a quick reminder of build statuses in Travis CI:

All builds finish with one of four possible statuses: succeeded, failed, errored, or (manually) cancelled. The difference between a failed and errored

hdgarrood

{-# LANGUAGE OverloadedStrings #-}

import Control.Monad (forM_)
import Data.Text.Lazy (Text)
import qualified Data.Text.Lazy as T
import qualified Data.Text.Lazy.Encoding as T
import qualified Data.ByteString as BS
import qualified Data.ByteString.Lazy as BSL
import qualified Blaze.ByteString.Builder as B

hdgarrood

module DataTypes where

data Foo = Foo

mkFoo = Foo

data Bar = Bar

data Baz = Baz1 | Baz2

hdgarrood

Why there can be no EuclideanRing b => EuclideanRing (a -> b) instance

One of the defining features of a EuclideanRing is that you can divide any pair of elements, as long as the divisor is nonzero. Specifically, if you have a euclidean ring R, with x, y in R, and y /= zero, we need to have x = (x / y) * y + (x `mod` y).

We do have a Ring b => Ring (a -> b) instance, where multiplication is defined as follows:

hdgarrood

var master = [15.985, 15.774, 16.664, 16.111, 15.889, 15.603, 15.598, 15.690, 15.602, 16.159];

var reordered = [15.831, 16.182, 16.519, 15.383, 15.936, 15.888, 15.645, 15.775, 15.749, 16.076];

function mean(arr) {
  return arr.reduce(function(x,y) { return x + y; }) / arr.length;
}

function variance(arr) {
  var m = mean(arr);

hdgarrood

from collections import namedtuple
import numpy as np

# This is an implementation of the version of the simplex algorithm we are
# expected to execute by hand in LPMS. There is just one important difference:
# because python and numpy use zero-based indexing, this does too. So for
# example you need to subtract 1 from each element of your basic and nonbasic
# sets.

LPState = namedtuple("LPState", ["basic", "nonbasic"])

hdgarrood

a9d20f4 Add test for generic deriving with awkward records (Harry Garrood, 54 seconds ago)

diff --git a/examples/passing/RecordLabels.purs b/examples/passing/RecordLabels.purs
new file mode 100644
index 0000000..b58a4bd
--- /dev/null
+++ b/examples/passing/RecordLabels.purs
@@ -0,0 +1,5 @@
+module Main where
+

hdgarrood

Dependencies

First you'll need a docker install: https://docs.docker.com/installation/

psci

After installing Docker, we'll run a PureScript psci session by running the 0.6.1.1 image. The -it flags allocate a pseudo tty and make it interactive. We're pulling the image from the default docker hub, so the images are here and the code that builds them is on GitHub

docker run -it biscarch/purescript:0.6.1.1