top-r-packages.R

library(rvest)
library(stringr)

# get list of gists
all_links = c()
for (i in 1:25) {
  url = paste('https://gist.github.com/erikgregorywebb?page=', i, sep ='')
  print(url)
  Sys.sleep(1)
  page = read_html(url)
  page_links = page %>% html_nodes('a') %>% html_attr('href')
  links = page_links[str_detect(page_links, 'https://gist.github.com/erikgregorywebb/')]
  all_links = c(all_links, links)
}

# extract r packages from each gist
all_packages = c()
for (i in 1:length(all_links)) {
  url = all_links[i]
  print(url)
  Sys.sleep(1)
  page = read_html(url)
  code = page %>% html_node('.highlight') %>% html_text()
  packages = str_extract_all(code, 'library\\(.*?\\)')
  all_packages = c(all_packages, packages[[1]])
}

# determine top packages used
top_packages = all_packages %>% tibble(package = `.`) %>%
  group_by(package) %>% count(sort = T) %>% 
  filter(is.na(package) == FALSE) %>%
  ungroup() %>%
  mutate(package = str_replace(package, 'library\\(', '')) %>%
  mutate(package = str_replace(package, '\\)', ''))

rank	package
1	tidyverse
2	rvest
3	dplyr
4	scales
5	stringr
6	lubridate
7	tidytext
8	ggplot2
9	jsonlite
10	scriptuRs
11	tidyr
12	igraph
13	zipcode
14	httr
15	readxl
16	wordcloud2
17	devtools
18	fuzzyjoin
19	gganimate
20	ggmap
21	ggraph
22	gifski
23	googledrive
24	googlesheets
25	pdftools