github() doesn't retrieve the code correctly from an ipynb file on github.

Example here

Solution

• github() should detect files ending in .ipynb and parse accordingly

It could be useful to automatically include the source URL in rawr() output, such that

rawr("https://www.kaggle.com/vrtjso/mercari-eda-more-info-than-you-can-imagine", 
  to_file=T, 
  source = T)

Would return

# From: https://www.kaggle.com/vrtjso/mercari-eda-more-info-than-you-can-imagine

library(data.table) # Loading data
library(ggplot2) # Data visualization
library(treemapify) # Treemap visualization
etc

Note the extra lines at the top of the output

github() currently attempts to parse everything as though it were a file (with readLines(), but should be able to recognise README.md and parse accordingly.

Stack Overflow allows users to link to a specific answer. stackoverflow() should recognise when a url is referring to a specific answer and return only code for that answer.

Lower priority: the same can be applied to urls for comments.

rawr() retrieved just a couple of lines (not necessarily from the beginning) from https://github.com/mjaniec2013/ExecutionTime/blob/master/ExecutionTime.R

Code used:

library(rawr)

code <- rawr("https://github.com/mjaniec2013/ExecutionTime/blob/master/ExecutionTime.R")

cat(code)

rawr expedites the transfer of R code from web to interpreter which is great for beginners and more experienced programmers alike.

A list of compatible urls, organised by topic, as well as a way of browsing/searching these urls, would be an excellent addition.

The urls would be 'best in class' tutorials and demonstrations, thus improving the speed with which R users can learn a new topic, and relieve any complications arising from those sources being in unstandardised formats (i.e. rawr takes care of differences across websites).

If required, checksums could be used to ensure the extracted code is unaltered from the time the url was added to the list. I have not found a way to validate externally read files (e.g. S3 or Cloud Storage) yet, but that could also be useful.

r2c() and/or rtc() seem sensible aliases for convenient use of rawr_to_clip()

CRAN_collisions("r2c")
$packages
character(0)

$functions
[1] package_names  function_names
<0 rows> (or 0-length row.names)

CRAN_collisions("rtc")
$packages
character(0)

$functions
[1] package_names  function_names
<0 rows> (or 0-length row.names)