R/compliance.R

In academictwitteR: Access the Twitter Academic Research Product Track V2 API Endpoint

post_query <- function(url, params, bearer_token = get_bearer()) {
  bearer_token <- check_bearer(bearer_token)
  r <- httr::POST(url, httr::add_headers(Authorization = bearer_token), body = params, encode = "json")
  status_code <- httr::status_code(r)
  if (!status_code %in% c(200)) {
    stop(paste("Something went wrong. Status code:", httr::status_code(r)))
  }
  r
}

.gen_tweet_ids_file <- function(x, force_ids) {
  if (length(x) == 1 & !force_ids) {
    if (!file.exists(x)) {
      stop("File ", x, "does not exist.", call. = FALSE)
    }
    file <- x
  } else {
    file <- tempfile(fileext = ".txt")
    writeLines(as.character(x), file)
  }
  return(file)
}

#' Create Compliance Job
#'
#' This function creates a new compliance job and upload the Tweet IDs or user IDs. By default, the parameter `x` with the length of one is assumed to be a text file containing either Tweet IDs or user IDs. This default behavior can be bypassed using `force_ids` For example, if you want to check for just a single Tweet ID. 
#' 
#' @param x either a character vector of Tweet IDs or user IDs; or a plain text file that each line contains a Tweet ID or user ID.
#' @param type the type of the job, whether "tweets" or "users".
#' @param force_ids logical, make sure `x` is treated as a character vector of Tweet IDs or user IDs.
#' @inheritParams get_all_tweets
#' 
#' @return the job ID (invisibly)
#' @export
#' 
#' @examples
#' \dontrun{
#' create_compliance_job(x = "tweetids.txt", type = "tweets")
#' }
create_compliance_job <- function(x,
                                  type = "tweets",
                                  bearer_token = get_bearer(),
                                  force_ids = FALSE,
                                  verbose = TRUE) {
  if (!type %in% c("tweets", "users")) {
    stop("Unknown `type` parameter: ", type, ". It must be \"tweet\" or \"users\".")
  }
  file <- .gen_tweet_ids_file(x = x, force_ids = force_ids)
  r <- post_query(url = "https://api.twitter.com/2/compliance/jobs",
                  bearer_token,
                  params = list("type" = type))
  rcontent <- jsonlite::fromJSON(httr::content(r, "text"))
  .vcat(verbose, "Job ID: ", rcontent$data$id, "\n")
  r <- httr::PUT(rcontent$data$upload_url,
                 body = httr::upload_file(file))
  if(r$status_code == 200){
    .vcat(verbose, "Upload Completed", "\n")
  } else {
    stop(paste("Something went wrong. Status code:", r$status_code, "\n"))
  }
  invisible(rcontent$data$id)
}

#' List Compliance Jobs
#' 
#' This function lists all compliance jobs.
#' @inheritParams create_compliance_job
#' 
#' @return a data frame 
#' @export
#' 
#' @examples
#' \dontrun{
#' list_compliance_jobs()
#' }
list_compliance_jobs <-function(type = "tweets", 
                                bearer_token = get_bearer()){
  if (!type %in% c("tweets", "users")) {
    stop("Unknown `type` parameter: ", type, ". It must be \"tweet\" or \"users\".")
  }
  res <- make_query("https://api.twitter.com/2/compliance/jobs",
                    params = list("type" = type), 
                    bearer_token = bearer_token)
  return(res$data)
}

#' Get Compliance Result
#' 
#' This function retrieves the information for a single compliance job.
#' 
#' @param id string, the job id
#' @inheritParams create_compliance_job
#' 
#' @return a data frame 
#' @export
#' 
#' @examples
#' \dontrun{
#' get_compliance_result("1460077048991555585")
#' }
get_compliance_result <- function(id, 
                                  bearer_token = get_bearer(), verbose = TRUE){
  r <- make_query(paste0("https://api.twitter.com/2/compliance/jobs/",id),
                  params = list(), 
                  bearer_token = bearer_token)
  status <- r$data$status
  if(status == "in_progress"){
    stop("Compliance check is still in progress.\n")
  } else if (status == "complete"){ 
    # Download if ready
    .vcat(verbose, "Downloading...\n")
    dl <- httr::GET(r$data$download_url)
    filename <- tempfile(pattern = id, fileext = ".json")
    writeLines(httr::content(dl, "text", encoding = "UTF-8"), filename)
    res <- jsonlite::stream_in(file(filename), verbose = FALSE)
    unlink(res)
    return(res)
  } else if (status == "expired"){
    stop("Upload expired, please retry.\n")
  } else {
    stop(paste("Something went wrong. Status:", status, "\n"))
  }
}