scrape-brewtoad.r

library(rvest)
library(dplyr)
library(readr)

# How many pages do you want to harvest (currently over 10,000 available)
npages <- 5
for (i in 1:npages) {
  # If first time through the loop, initialize list to hold results
  if (i==1) d <- list()
  # Report progress
  message(paste0("Harvesting page ", i, "/", npages))
  # The URL for the table
  url <- paste0("https://www.brewtoad.com/recipes?page=", i, "&sort=created_at&view_as_table=true")
  # harvest the data
  d[[i]] <- url %>%
    read_html %>%
    html_table %>%
    as.data.frame %>%
    tbl_df
}
rm(npages, i, url)

# bind_rows on the list to convert to a data_frame
d <- d %>%
  bind_rows %>%
  # Remove the rating column - sparse, most are zero
  select(-Rating) %>% 
  # remove the percent sign and turn abv into a numeric value
  mutate(ABV=as.numeric(gsub("%", "", ABV))) %>% 
  # remove dupes
  distinct
# Write to file
d %>% write_csv("brewtoad.csv")