Inspect the edit history of a bunch of wikipedia pages to measure the proportion of the page contributed by each editor

wikipedia-page-edit-proportions.R

library(tidyverse)
the_pages <- c(
  "https://en.wikipedia.org/wiki/Angela_McGowan",
  "https://en.wikipedia.org/wiki/Caroline_Bird_(archaeologist)",
  "https://en.wikipedia.org/wiki/Jo_McDonald",
  "https://en.wikipedia.org/wiki/Laurajane_Smith",
  "https://en.wikipedia.org/wiki/Louise_Zarmati",
  "https://en.wikipedia.org/wiki/Marcia-Anne_Dobres",
  "https://en.wikipedia.org/wiki/Sarah_Colley",
  "https://en.wikipedia.org/wiki/Sharon_Sullivan"
)

the_page_history <- 
  str_replace(the_pages,
              'https://en.wikipedia.org/wiki/',
              'https://en.wikipedia.org/w/index.php?title=') %>% 
  str_c(., '&action=history')



library(rvest)
# get a summary of edit history for each page
edit_byte_change_per_editor <- 
  map(the_page_history,
      ~read_html(.x) %>% 
        html_nodes('.mw-diff-bytes , bdi') %>% 
        html_text() %>% 
        matrix(., ncol = 2, byrow = T) %>% 
        data.frame() %>% 
        set_names(nm = c("editor", "edit_size")) %>% 
        mutate_all(as.character) %>% 
        mutate(edit_size = parse_number(edit_size))
  )

# put the page names onto this list
names(edit_byte_change_per_editor) <- # thanks LYW for the short version!
  str_extract(the_pages, "(?<=wiki/).*")

# plot them all
bind_rows(edit_byte_change_per_editor, 
          .id = 'page') %>% 
  group_by(page, editor) %>% 
  summarise(total_edit_size_bytes = sum(edit_size)) %>% 
  mutate(percent_total_article = total_edit_size_bytes / sum(total_edit_size_bytes)) %>% 
  ggplot(aes(reorder(editor, 
                     percent_total_article),
             percent_total_article)) +
  geom_col() +
  theme_minimal(base_size = 10) +
  coord_flip() +
  facet_wrap( ~ page, 
              scales = "free") +
  xlab("Editor") +
  ylab("Proportion of bytes added to the page")

# plot by the total edit size, thanks to LYW for this!
bind_rows(edit_byte_change_per_editor, 
          .id = 'page') %>% 
  group_by(page) %>% 
  summarise(total_edit_size_bytes = sum(edit_size)) %>%
  ggplot(aes(reorder(page, total_edit_size_bytes), 
             total_edit_size_bytes)) + 
  geom_col() +
  theme_minimal(base_size = 10) +
  coord_flip() +
  xlab("Archaeologist Page") +
  ylab("Total Edit Size Bytes")