In my work I spend a lot of time sharing data with others and I've often been frustrated when I had to download a data export from an anlytics machine R job only to upload it to a GitHub gist for sharing. I used the Vim gist plugin for that, but I wished I could just let the R job do the sharing for me.
Here's the package that does this - and right from an R data.table object.
You can install this from GitHub using devtools by:
library(devtools)
devtools::install_github('adjust/csv-gists-r');Consistent with the Vim gist plugin, this package looks for your GitHub token
under ~/.gist-vim. It expects the format:
token 45cdef12323ab
NOTE: Make sure the GitHub token that you use has permissions to create gists. You can edit these permissions from your Authentication Settings on GitHub. You can manage your tokens from https://github.com/settings/tokens.
The package has only one function called gist.csv. It takes a data.table (or
data.frame)
library(csvgists)
resp <- gist.csv(data.frame(a=1:10, b=letters[1:10]), description='my gist')
# To update an existing gist - give its id:
resp <- gist.csv(data.frame(a=1:11, b=letters[1:11]), id=resp$id)These commands produce a gist with the data.frame as a CSV format and print
the HTML URL for you ready to share with others.
NOTE: the default visibility of the gists is private, but you can set this to
public too. See ?gist.csv for usage details.
Some Excel setups require ; as a CSV separator and other custom settings. R takes care of that by
having a standard write.csv2 function. csvgists allows you to use a custom csv.writer function
by passing a gist.csv argument.
gist <- gist.csv(DT, csv.writer=write.csv2)Feature requests as GitHub issues or pull requests are welcome.