reprex: help me help you!

reprex R package: description and philosophy
http://jennybc.github.io/reprex/
part of rOpenSci Community Call 2017 March 7
https://github.com/ropensci/commcalls/issues/14

Transcript

1.  
   Jennifer Bryan 

   RStudio, University of British Columbia
   @JennyBryan @jennybc
   reprex:
   the package, the point
   

   View Slide

2. How to ask questions so they get answered!
   Possibly by yourself!
   2017 March 07
   w/ James (JD) Long, Jenny Bryan, Scott Chamberlain
   an rOpenSci Community call:
   https://github.com/ropensci/commcalls/issues/14
   Collaborative notes:
   http://bit.ly/commcallv13
   

   View Slide

3. View Slide

4. http://stat545.com
   300-500 GitHub issue comments
   by me Sept-Nov each year
   

   View Slide

5. conversations about code are more productive
   when they contain code
   code that actually runs
   code that I don’t have to run
   code that I can easily run
   

   View Slide

6. https://github.com/jennybc/reprex
   http://jennybc.github.io/reprex/
   

   View Slide

7. install.packages("reprex")
   ## or
   devtools::install_github("jennybc/reprex")
   

   View Slide

8. View Slide

9. code that actually runs
   code is run in separate, fresh R session
   ergo, it must be self-contained!
   must load all necessary packages
   must create all necessary objects
   

   View Slide

10. not self-contained — forgot to load necessary package
    not self-contained — forgot to define template
    YAAAASSSSSS
    

    View Slide

11. code that I don’t have to run
    did I mention: I touch 300 - 500 issues/course?
    I can often get the point w/o running code IFF
    I can see the output
    SHOW ME SOME OUTPUT, not just the code
    

    View Slide

12. View Slide

13. code that I can easily run
    do not copy/paste your R transcript
    do not take a screenshot of your R session
    must I edit out the prompts?
    must I comment the output?
    HELP ME HELP YOU
    

    View Slide

14. > library(purrr)
    > x <- list(one = list(always = "a", sometimes = "b"),
    + two = list(always = "b", sometimes = NULL))
    > x %>% map_chr("always")
    one two
    "a" "b"
    > x %>% map_chr("sometimes")
    Error: Result 2 is not a length 1 atomic vector
    
    

    View Slide

15. library(purrr)
    x <- list(one = list(always = "a", sometimes = "b"),
    two = list(always = "b", sometimes = NULL))
    x %>% map_chr("always")
    #> one two
    #> "a" "b"
    x %>% map_chr("sometimes")
    #> Error: Result 2 is not a length 1 atomic vector
    
    

    View Slide

16. live demo
    
    

    View Slide

17. basic copy + reprex()
    copy + add-in
    select or focus + add-in
    venue = “so”, venue= “R”
    si = TRUE
    input, outfile
    include a figure
    reprex_invert()
    reprex_clean()
    reprex_rescue()
    stuff I hope to show
    

    View Slide

18. Why make a reprex?
    

    View Slide

19. 1. get help
    2. bug report
    

    View Slide

20. https://github.com/jennybc/reprex#what-is-a-reprex
    

    View Slide

21. writing a good reprex can seem like a hassle
    but if asker can’t be bothered,
    why should the answerer?
    practice this often and
    it will make you a better programmer
    it will increase your knowledge of the language
    

    View Slide

22. writing a reprex helps you ambush your problem
    from behind
    “demonstrate my problem in a small example”
    sounds so much more doable than
    “figure out how my code is borked”
    > x% of the time I solve my own problem,
    but the reprex mentality is enabling
    

    View Slide

23. View Slide