srvanderplas

// Browser fingerprinting is a technique to "mark" anonymous users using JS
// (or other things).  To build an "identity" of sorts the browser is queried
// for a list of its plugins, the screen size and several other things, then
// hashes them.  The idea is that these bits of information produce an unique
// "fingerprint" of sorts; the more elaborate the list of data points is, the
// more unique this fingerprint becomes.  And you wouldn't even need to set a
// cookie to recognize this user when she visits again.
//
// For more information on this topic consult
// [Ars Technica](http://arstechnica.com/tech-policy/news/2010/05/how-your-web-browser-rats-you-out-online.ars)

frankfarach

#!/bin/bash

##############################################################################
# Name: templateR.sh 
# Author: Frank Farach
# Description: Script to clone templateR repo from  GitHub account, 
# with options to name the local working directory and RStudio
# project. This depends on the existence of a repository called templateR
# in the GitHub account of user named in the global variable USER below.
# See: http://github.com/frankfarach/templateR

ramnathv

This is a proof-of-concept of a parser to convert a markdown file to a ctv file, that is required for task views. The basic idea is to use a combination of YAML and markdown sections in the md file, parse it and convert it into a payload, and render it using a template as specified by the ctv package.

Structure of Markdown File

---
name: Working with data on the web
maintainer: Scott Chamberlain, Karthik Ram, Christopher Gandrud
email: scott at ropensci.org
version: 2013-09-17

millermedeiros

Setup Mac OS X

I've done the same process every couple years since 2013 (Mountain Lion, Mavericks, High Sierra, Catalina) and I updated the Gist each time I've done it.

I kinda regret for not using something like Boxen (or anything similar) to automate the process, but TBH I only actually needed to these steps once every couple years...

hadley

# A tutorial navigates through a series of states, either manually, with
# nxt(), prv(), jmp() etc, or automatically, by using addTaskCallback() and
# checking that preconditions are met.
#   
# Each state has:
#  * a message
#  * a next state (a function name)  
#    (eventually may have multiple next named states)
#  * an auto test 
#

ramnathv

to_json = function(df, orient = "columns", json = T){
  dl = as.list(df)
  dl = switch(orient, 
    columns = dl,
    records = do.call('zip_vectors_', dl),
    values = do.call('zip_vectors_', setNames(dl, NULL))
  )
  if (json){
    dl = rjson::toJSON(dl)
  }

bearloga

library(shiny)
library(ggplot2)

# Download DataTables JS file and TableTools full package from http://datatables.net/download/
# Change the paths appropriately:
addResourcePath('datatables','/Users/yourusername/Downloads/DataTables-1.9.4/media')
addResourcePath('tabletools','/Users/yourusername/Downloads/TableTools-2.1.5/media')

runApp(list(
	ui = basicPage(

leeper

library('devtools')
library('utils')
library('httr')

update_github <-
function(ask = TRUE, ...){
  installed <- installed.packages()
  oldVersion <- installed[,'Version']
  urls <- sapply(names(oldVersion), function(x){
    d <- packageDescription(x)

DanHerbert

OBSOLETE

This entire guide is based on an old version of Homebrew/Node and no longer applies. It was only ever intended to fix a specific error message which has since been fixed. I've kept it here for historical purposes, but it should no longer be used. Homebrew maintainers have fixed things and the options mentioned don't exist and won't work.

I still believe it is better to manually install npm separately since having a generic package manager maintain another package manager is a bad idea, but the instructions below don't explain how to do that.

Fixing npm On Mac OS X for Homebrew Users

Installing node through Homebrew can cause problems with npm for globally installed packages. To fix it quickly, use the solution below. An explanation is also included at the end of this document.

hadley

iclass <- function(x) {
  c(
    if (is.matrix(x)) "matrix",
    if (is.array(x) && !is.matrix(x)) "array",
    if (is.double(x)) "double",
    if (is.integer(x)) "integer",
    mode(x)
  )
}