Currently users need to download the data via download_data() to a directory they specify, and then process the data with process_data().
For users who may not care about if the downloaded .txt files exist someplace locally they can get to, couldn't we essentially download the .txt files to a temp directory on the user's machine, then run process_data() on the files in the temp directory, and do it all in one shot? Perhaps this could be a new function simply named get_data() ?