max-mapper

richfitz

Installation on OSX is currently a PITA if you don't have brew (I don't so that my system stays more similar to users of things, and because I'm a masochist apparently). See this issue as this will be fixed in jq 1.5. So, create an executable shell script sheep in the jq directory containing

#!/bin/sh
echo "baa"

and run YACC=./sheep ./configure - make should run fine after that.

There is a library target for jq, which is great because now we only have to wrap a library, rather than write one. Further, there is a set of Python bindings that we can use as a starting point. The repo is here.

Ruby bindings here

Stebalien

#!/bin/bash
set -e
[[ "$(git symbolic-ref --short HEAD)" == "master" ]] || exit 0

msg() {
    echo "> $@"
}

dir="$(pwd)"
tmp="$(mktemp -d)"

davharris

#Based on https://github.com/robotlolita/fuck-you
# and https://en.wikipedia.org/wiki/Transformation_of_text

screw_you = function(x){
  chars   = strsplit(paste0(c(" ", rev(letters), rev(LETTERS), 0, 9:1, "&_?!\".';'"), collapse = ""), "")[[1]]
  flipped = strsplit(
    " zʎxʍʌnʇsɹbdouɯlʞɾıɥɓɟǝpɔqɐZ⅄XMΛ∩⊥SᴚΌԀONW⅂⋊ſIH⅁ℲƎᗡƆ𐐒∀068ㄥ9ގㄣƐᘔ⇂⅋‾¿¡„,˙'؛", 
    "")[[1]]
  
  split = strsplit(as.character(substitute(x)), "")[[1]]

ateucher

## convert back and forth between -180/180 and 0/360 systems

to_lon_360 <- function(lon_180) {
  (lon_180 + 360) %% 360
}

to_lon_180 <- function(lon_360) {
  ((lon_360 + 180) %% 360) - 180
}

gadenbuie

getCiteCounts <- function(citekeys, bibtex){
  ## This function takes a list of citation keys, for ex. from lit review notes
  ## and a master bibtex file with citekeys and DOIs
  ## Returns a vector of number of times each article was cited from CrossRef
  
  
  # Check for and load/install required packages
  # --> stringr, rcrossref
  if (!'stringr' %in% installed.packages()) install.packages('stringr')
  require(stringr, quietly=TRUE)

kbroman

options(show.signif.stars=FALSE, scipen=10, htmlhelp=TRUE)
options(width=115, repos="http://cran.rstudio.com",
        CRAN = "http://cran.rstudio.com",
        browserNLdisabled = TRUE,
        deparse.max.lines = 2)
if (interactive()) {
  suppressMessages(require(devtools))
}
options(error= function() cat("    \\\   \n     \\\

pkoppstein

# This gist defines a jq filter for combining two JSON objects by resolving conflicts
# in a way that is appropriate for mapping relational tables to objects. 
# "combine" as defined here is both commutative and associative.
# Example:
#           { "id": 123, "son": "Son1", "surname": "S"}
# | combine({ "id": 123, "son": "Son2", "surname": "S"})
# produces: { "id": 123, "son": ["Son1",Son2], "surname": "S"}


# Combine two entities in an array-oriented fashion.

tpoisot

{
   "c_key": "ConsumerKey", 
   "c_sec": "ConsumerSecret",
   "t_key": "TokenKey",
   "t_sec": "TokenSecret"
}

noamross

throttle = new.env(parent = emptyenv())
throttle$recent = data.frame(domain =  character(), last_visit = character())

#' A throttled version of GET
#' 
#' This uses \code{httr::GET} to fetch a web page, but throttles based on domains.
#' 
#' \code{slowGET} keeps a list of domains recently accessed by itself in a 
#' separate environment.  If a domain has been accessed since \code{pause}
#' seconds ago, it will delay execution until that time has passed