There are functions in R that work on URL, such as download.file or even
read.table (plus many others, including some from dedicated packages such as
RCurl). One may thus think that (s)he can just go on a Github or Dropbox
webpage, copy the URL of the file their are interested in and paste it directly
into R. It’s almost that, except for a little tweak. Indeed the URL that you
would copy this way is the URL to the file display on the webpage. What you need
instead is the URL of the actual file. And this can be accessed by ending the URL
with ?raw=1 (or, equivalently, ?raw=t, ?raw=T, ?raw=true, ?raw=True,
?raw=TRUE, etc…). By default, if you copy-paste from Dropbox, your URL will
end by ?dl=0. What you have to do here is just replace this ?dl=0 by ?raw=1
in your URL. Copied from Github, your URL will not end with ?dl=0 or ?raw=1,
then just add ?raw=1 to the end of your URL. Here is an example on how to
download and laod an rda file from GitHub:
download.file("https://github.com/.../data.rda?raw=1", "data.rda")
load("data.rda")
file.remove("data.rda")Below is an example on how to read a table directly from Dropbox:
data <- read.table("https://www.dropbox.com/.../data.txt?raw=1")