jasonkarns

$ grep -RIho -E '\$npm_package_[_0-9a-z]*' . 2>/dev/null | sort -u
$npm_package_
$npm_package_actions
$npm_package_bin_brew_publish
$npm_package_bin_node_build
$npm_package_bin_packfiles
$npm_package_bin_validate_analytics_data
$npm_package_browser
$npm_package_config_
$npm_package_config_app

jasonkarns

[ "$NODENV_COMMAND" = npx ] || return 0

remove_from_path() {
  local path_to_remove="$1"
  local path_before
  local result=":${PATH//\~/$HOME}:"
  while [ "$path_before" != "$result" ]; do
    path_before="$result"
    result="${result//:$path_to_remove:/:}"
  done

jasonkarns

Why choose nodenv?

Because nodenv was forked from rbenv, many of the reasons to choose nodenv over others are similar to why one would choose rbenv over rvm. (Nodenv also inherits a much longer battle-tested history than others.)

Auto switching

One of the biggest reasons to choose nodenv over the others is that nodenv can automatically select the appropriate node version for a given project. Neither nvm nor n have auto-switching capabilities out of the box. They both require some form of 'use' command to activate a given version; which lasts for the lifetime of the shell (or until another version is activated).

jasonkarns

I'm opening this issue because:

• npm is crashing.
• npm is producing an incorrect install.
• npm is doing something I don't understand.
• Other (see below for feature requests):

supporting information:

• npm -v prints: 5.0.1

jasonkarns

module.exports = function findMaxValidDate(min, max) {
  var next = Math.floor((min+max)/2);

  if(min == next) return min;

  if(new Date(next) == "Invalid Date") {
    return findMaxValidDate(min, next);
  } else {
    return findMaxValidDate(next, max);
  }

jasonkarns

require 'axe/configuration'

module Axe
  class FindsPage
    WEBDRIVER_NAMES = [ :page, :browser, :driver, :webdriver ]

    class << self
      alias :in :new
    end

jasonkarns

require 'singleton'

class Configuration
  include Singleton

  def init_with(attributes)
    attributes.map.each do |k,v|
      self.class.instance.instance_variable_set(:"@#{k}", v)
    end
  end

jasonkarns

# named captures shouldn't *have* to be compared as a hash;
# MatchData still returns the captures as a plain indexed array,
# so matching against the indexed captures should still be allowed
it "matches indexed captures against named captures" do
  expect(Regexp.new("(?<num>123)")).to match("a123a").with_captures("123")
end

# doesn't make sense to show the expected captures if the match itself fails; they're irrelevant at that point
it "shows plain match failure message if match itself fails" do
  expect {

jasonkarns

def matches?(page)
  @audit = @a11y_check.call Page.new page

  ::RSpec::Expectations::FailureAggregator.new(@audit.invocation, nil).aggregate do
    @audit.results.violations.each {|v| ::RSpec::Expectations.fail_with v.failure_message }
  end

  @audit.passed?
end

jasonkarns

module CustomMatchers
  class Match < RSpec::Matchers::BuiltIn::Match
    def matches?(actual)
      # first ensure the regex matched
      return false unless result = super
      # only continue if specifying captures
      return result unless expected_captures = @captures

      actual_captures = to_hash result