Get Quotes From Tweet ID

[FiveVoltLogic]

Hey all,

I've been collecting data using the package for subsequent SNA and it's been mostly great. I've used get_all_tweets to make my original collection, and then run get_retweeted_by on the results that had retweets.

I would now like to collect information from the tweets that quote my original collection. I've seen that there is an argument is_quote in get_all_tweets but I'm not sure this will help. Really, I need something similar to get_retweeted_by that gets quotes instead (user ID, text body).

I have seen that there is an endpoint for this and I have had a go at changing the endpoint for the get_retweeted_by to point to the quote_tweets endpoint and recompiled. It gets some results, though I'm not sure how to test this properly as the numbers it returns are not consistent with the numbers on the twitter website. - I have been through and checked that it only contains relevant fields available for this endpoint.

I've included the code below:

#' Get users who have quoted a tweet
#' 
#' This function fetches users who quoted a tweet
#'
#' @param x string containing one tweet id or a vector of tweet ids
#' @inheritParams get_all_tweets
#' @return a data frame
#' @export
#'
#' @examples
#' \dontrun{
#' tweets <- c("1392887366507970561","1409931481552543749")
#' get_quoted_by(tweets, bearer_token = get_bearer())
#' }
get_quoted_by <- function(x, bearer_token = get_bearer(), data_path = NULL, verbose = TRUE) {
  url <- "https://api.twitter.com/2/tweets/"
  endpoint <- "/quote_tweets"
  ## Building parameters for get_tweets()
  params <- list(
    "tweet.fields" = "attachments,author_id,context_annotations,conversation_id,created_at,entities,geo,id,in_reply_to_user_id,lang,public_metrics,possibly_sensitive,referenced_tweets,source,text,withheld",
    "user.fields" = "created_at,description,entities,id,location,name,pinned_tweet_id,profile_image_url,protected,public_metrics,url,username,verified,withheld"
  )  
  ## loop through x
  new_df <- data.frame()
  for(i in seq_along(x)){
    .vcat(verbose, paste0("Processing ", x[i], "\n"))
    requrl <- paste0(url, x[i], endpoint)
    next_token <- ""
    while(!is.null(next_token)) {
      if(next_token != ""){
        params[["pagination_token"]] <- next_token
      }
      dat <- make_query(url = requrl, params = params, bearer_token = bearer_token, verbose = verbose)
      next_token <- dat$meta$next_token #this is NULL if there are no pages left
      if (!is.null(dat$data)) {
        new_rows <- dat$data
        new_rows$from_id <- x[i]
        new_df <- dplyr::bind_rows(new_df, new_rows) # add new rows
        .vcat(verbose, "Total data points: ",nrow(new_df), "\n")
      }
      Sys.sleep(1)
      if (is.null(next_token)) {
        .vcat(verbose, "This is the last page for ",  x[i], ": finishing collection. \n")
        params[["pagination_token"]] <- NULL
        break
      }
    }
  }
  return(new_df)
}

I'm not completely familiar with how twitter reports its numbers, nor am I particularly familiar with how to test what I've just done.

I have a couple of questions then regarding this:

• Is this the best way to go about collecting quote tweets, or is there another way that I've missed?
• How do I go about writing a test for this to ensure the numbers being returned are consistent with the numbers reported on the website? (does that even matter?)

Thanks!

Will