Which is by no means a guarantee that the data did or did not change. I deal with government data all the time, and their sites just list the ZIP/Excel/CSV/PDF files. You actually have to fetch the files in order to check if they changed (or their content changed).

My approach is

• Read persistent file with ZIP/CSV file checksums
• Read site and parse links
• For each link with a file I want/know
  • Fetch file into memory
  • Calculate SHA256
  • Compare to previous SHA256
  • same and next
  • save file
  • store SHA256
  • log/mail/other action(s)