If the process that parses the URL could report an error (e.g., "not a URL" or "not an academia URL"), that would be great.

I put this in a MacOS service that I'll post and this would be helpful for when things go wrong.

I'm not 100% on this as I'm not quite sure how the academia file naming works, but wouldn't it be possible to check for the existence of the download file before doing all the work getting the URL from academia.edu?

Getting this error

.rbenv/versions/2.6.4/lib/ruby/2.6.0/uri/rfc3986_parser.rb:21:in split': URI must be ascii only`

on this URL:

https://www.academia.edu/1963397/Santuario_costiero_di_Ardea_le_terrecotte_del_tempio_arcaico_in_Tetti_di_Terracotta_La_decorazione_architettonica_fittile_tra_Etruria_e_Lazio_in_età_arcaica_Atti_delle_giornate_di_studio_Sapienza_Università_di_Roma_25_marzo_e_15_ottobre_2010_Officina_di_Etruscologia_5_Roma_2011_251_258

You'll see a couple of à in there.

My Nokogiri knowledge is zero, so I can't fix it. :-(