feeds.fun
feeds.fun copied to clipboard
Tiendil
News reader with tags
Feed normalization can remove duplicated entries, more entries will have tags, etc. There are at least two types of aggregated feeds: 1. Complex Reddit feeds - Example: https://www.reddit.com/r/linux+Gentoo+Kalilinux+archlinux+debian+Ubuntu+voidlinux+openSUSE+linuxquestions+linux4noobs+pop_os/.rss - It...
It may reduce the size of the table by removing duplicates. - According to gh-172 - Before implementation, repeat analysis from gh-172
I planned that the `l_entries` table would be constant, but there are already a few places where the records are changed.
Currently, `EntriesProcessor` starts processing N entries at once in parallel and then waits for all of them to finish before starting the next batch. This could slow down the processing...
Currently `InfiniteTasks` loop has constant sleep `delay_between_runs` seconds. It is ok for now, but in the future, it could affect the speed of processing spikes of new news. `InfiniteTasks` should...
Otherwise, error messages will eat up the Sentry limits too fast.
Sometimes, tag processors skip entries. For example, OpenAI processors skip a lot of entries because no API key is found for them. Initially, such entries were reprocessed periodically, but this...
For example, if we filter out read articles and mark one unread news as read, it takes several seconds to update the news list.
Since it is free for the moment, I wonder if this is possible. Thank you for your wonderful work!
