I have such setup, but with some problems:

• it always does full crawl (doesn't do deduplication)
• I am unable to control crawl order (don't have enough knowledge)

But need more motivation than I have...

Also I want some Android OS opensource client for that.

My OSes are Debian, Gentoo, software on them Docker, Apache, Postgres, certbot and some more, the clients are Firefox and Dino.

I saw that messages have different identifiers on different servers. What should I write in my text to make the hyperlink to work on all servers?

https://xn--e1aghfa.xn--e1aqbccjfc.xn--p1acf/post/15
https://lemmy.ml/post/49064481

15 ≠ 49064481

The site https://lemmy.ml/ is unreachable from my computer because of government politics. I want to refer to the copy of message, which is downloaded into my instance. But it have a new ID in my DB.

I think that my setup is still working partially

see earlier https://lemmy.ml/post/48608351

My browser says (in F12 console of firefox):

SyntaxError client.js:2:1170023
    hw https://лемми.опенсорс.рус/static/0799f19b/js/client.js:2

the file starts with

/*! For license information please see client.js.LICENSE.txt */
(()=>{var e,t,n,r,o={45850:(e,t,n)=>{"use strict";n.d(t,

and DeepSeek says that e,t,n,r,o should be separated by spaces, i.e. e, t, n, r, o

The lemmy.ml instance adds that image without problems.