>>/1361/
Yeah, Cloudflare 403s are a pain. I am not sure about Tor and Cloudflare, that may just be a step too far in complexity, but the 'normal' way to get through a 403 Cloudflare block is to visit the site in your browser (which does the clever problem solving) and then use Hydrus Companion to copy the solved cookies over to Hydrus.
https://gitgud.io/prkc/hydrus-companion
However, this may not work with Tor. Tor probably sets its own User-Agent and I guess is using different IP addresses regularly, and those are the two things that need to be the same as your browser for the Hydrus Companion trick to work. Maybe if you did the work in the Tor browser, but then copying cookies across may be even more difficult.
That said, some sites are just strict when it comes to Cloudflare rules, so if your VPN or whatever network solution is frowned upon by Cloudflare this week, then you'll just have trouble with that site. My own answer is mostly just to navigate away from those sites and stick to what I know works for me. I'm already overloaded with stuff to process, so no point spending 50% effort trying to fetch 5% more stuff that I missed.
For updating, no worries, the normal method is to extract the release on top of your old install folder. The extract straight from the zip file has no 'database' files, so it extracts safely over what you already have without overwriting anything user-created. Please check the help here:
https://hydrusnetwork.github.io/hydrus/getting_started_installing.html#updating
I personally make a backup, extract the fresh release on top, and boot. My update code will investigate your database, notice it is an older version, and apply several rules to update it in place. You will keep the same session and everything as when you last closed the client. The idea of the backup at that time is that if the update fails for some reason (rare, but it happens to a handful of users a few times a year), or more likely I messed something up, I can roll back immediately to where I was, no danger.
If you update using the normal routine and keep your old database, it will remember all its old known urls and be able to skip repeated downloads.
Let me know if you have any trouble with it!