Replacing ReadRuler with an R function (as the site will shut down on May 1st)

readruler_replacement.r

# Replacement for Read Ruler
# which is going to be deactivated on May 1st, 2023

# make sure to set up an app and the authentication in Pocket's interface: https://github.com/CorrelAid/pocketapi#authentication

# parameters:
# words_per_minute: avg. reading speed in words per minute
# minute_groups: vector indicating which limits to use for the reading time groups

read_ruler <- function(words_per_minute = 500, minute_groups = c(1, 2, 5, 10, 15, 20, 30, 45, 60, 75)) {

  library(pocketapi)
  
  # get queue
  data_unread <- pocket_get(state = "unread")
  
  # subset: which articles have no "min" tag?
  data_unread_new <- data_unread[!grepl(pattern="min", x = data_unread$tags, fixed = TRUE), ]
  
  # round it on the conservative side
  data_unread_new$reading_time_minutes <- ceiling(data_unread_new$word_count / words_per_minute)
  
  data_unread_new$reading_time_grouped <- cut(x = data_unread_new$reading_time_minutes,
                                              breaks = c(-Inf, minute_groups, Inf),
                                              labels = c(
                                                paste0(minute_groups, "min"),
                                                paste0(">", minute_groups[length(minute_groups)])
                                              ),
                                              right = TRUE,
                                              include.lowest = TRUE)
  
  # creating a list, separated by minute group
  data_unread_new_split <- split(data_unread_new, data_unread_new$reading_time_grouped)
  
  # loop over all minute groups and set the corresponding tag
  for (group in 1:length(data_unread_new_split)) {
    # check that we actually have articles in the group
    if(nrow(data_unread_new_split[[group]]) > 0) { 
      # then add the tag for the corresponding IDs with the corresponding estimated reading time
      pocket_tag(action_name = "tags_add",
                 item_ids = data_unread_new_split[[group]]$item_id,
                 tags = unique(data_unread_new_split[[group]]$reading_time_grouped))
    }
  }

}