Some of my articles ended up in there.
I don’t know how it happend, but I am interested to know because it could be a good feature that I didn’t how to use.
The Archive feed is for articles that have no assigned feed. Articles can end up in here if you star an article but later unsubscribe from its feed. If you use the Share Anything feature, articles will also be stored here.
You can manually archive an article from the Select… menu. If you do this with a more recent article you might find it re-appears in the feed again because it’s still in the site’s RSS feed, but the archived copy will remain forever.
Archived articles are not purged. You can manually delete them, but they won’t be removed during the regular clean-up TT-RSS does.
I’m really thankful for your answer!
I’ve been trying to find how to archive articles for a whole week, but couldn’t find the button for it. Now, that you’ve pointed where it is I can enjoy the feature properly.
On top of that, because you mentioned starred articles without feeds, I’d like to ask you more thing: is it what “orphaned articles” means, when running update.sh?
Also: my instante seems to avoid purging starred or published articles, too. Is it a misconfiguration I’ve made or that’s it works?
Short answer (as best as I can tell): No.
Long answer: I don’t think so. A quick look through the code suggests orphaned articles are where parts of the article exist in one database table but not another. TT-RSS stores information about articles across multiple tables, that check during the update makes sure that information isn’t left in one table if it’s been removed the other.
Starred articles are never removed. They are starred, which tells TT-RSS you want to keep them.
Published articles (as far as I know, I’ve never looked specifically) are removed when the article would normally be purged (whatever the setting for that article’s feed is, or the global setting). Unless, of course, you’ve starred that article; then it is never removed.
pretty much this, it just so happened that articles are stored in two tables, one points to another, articles without references from the latter table are cleaned up from the former periodically.
this was originally implemented this way because i thought that storing articles for multiple users subscribed to one feed once was worth the added bullshit, however this became pointless after per-user plugins became a thing.