Unlimited archive size display


I want to know exact workings for displaying large archives:

1) I don't see all articles in large feed, with Unlimited archive size. It doesn't help to secondary click and choose See all items from "" in popup windows.

2) After feed updates, I think it multiplies several articles so now I have instead around 2000 articles (confirmed by visiting related site) nearly 10.000 of them. And I can't even see and delete all duplicates (see 1). Will there be a function to eliminate duplicates (same titles)?

3) Should I delete unwanted articles (e.g. read but don't want to have them in the archive)? Is this the way it's supposed to work? Shouldn't FeedReader compare timestamps of articles and UPDATE (not INSERT) them into database (I have such impression for updating feature)?

Otherwise great speedup and elimination of very annoying blocking of application when large feeds get updated but still some misteries as this plus missing of flags etc. Great effort, though!

1) - At the moment

1) - At the moment Feedreader has undocumented "feature" that it displays only first 1000 articles in every feed. If your feed has more than 1000 then only first 1000 is displayed. I know it's annoying and we must think something out. This limitation is because things get quite slow if for example user has 10 feeds (every feed has 1000 articles) in folder and he/she opens folder. Then 10000 articles are queried and this makes things slow. Also - it does not have point from user interface logic to display 10000 articles in a time. It's just not usable/browseable any more.

2) and 3) why are you getting duplicates? Can you show me the feed what acts this badly.

Flags are already there... :)

Answer is in other topic

Aha, this - Marking the feeds - is actually an answer to 2 and 3. Reported to feed producers to generate guid sub tags in item tags of articles. More is explained here:

http://blogs.law.harvard.edu/tech/rss#ltguidgtSubelementOfLtitemgt

Answer to 1 (showing ALL articles in large feed) is still a mistery - older articles are found by searching but not by browsing.