I noticed that some Feeds are not displayed at all and for some feeds only one single article is displayed in RSSOwl though the feed (RSS, Atom) itself is ok.
www.mindfactory.de/xml/rss/mindstar_artikel.xml and maybe all mindfactory RSS feeds
Hi, I've noticed the same for a while now. I can't make out a pattern which kinds of feeds work and which don't.
Some popular blog feeds that don't work:
Bildblog.de, RSS and Atom both don't work
RSS doesn't work
Atom seems to work
Gestern Nacht im Taxi
seems to be Atom, does not work
Hope you can find the problem and fix it!
Thank you for the good work on RSS Owl!
Actually, I'm not sure that I know which kind of feed these are. How can I tell whether a feed is RSS or Atom?
To test if it is RSS or Atom, try searching for "Atom" in source code (you get it with Ctrl+U in most browsers, or just download it with wget/curl/whatever.
Some of those RSS feeds (e.g. lawblog.de, netzpolitik.org, not-safe-for-work.de, logbuch-netzpolitik.de, wrint.de, …) don't work with flattr elements (the problematic code is in the "<atom:link…" line).
Thanks for the tip. I see
which seems to be Atom, and
which says RSS, but still mentions Atom (as namespace?).
Are the flattr elements the reason that some feeds only show one article in RSS Owl? If so, could you just ignore flattr somehow?
Right, this should be the reason for this bug. I already tried to hunt it down but failed to find the related code. See https://sourceforge.net/p/rssowl/discussion/296910/thread/f2b993f4/?limit=25#4b08 for that.
Thank you for the link to the previous discussion.
So somehow the feed source gets interpreted while loading or just after loading, and something in the <atom:link ... makes RSSOwl stop loading (or stop displaying?) the feed items. Can you figure out which part of that line is responsible? In the other thread you wrote that <atom10:link is OK -- so it looks as if the <atom:link alone is enough to break the feed. Maybe RSSOwl tries to interpret (or follow?) the URL in a <atom:link but not in a <atom10:link ?
It seems only Benjamin Pasero knows the code well enough to find the bug and fix it, and he seems too busy to come around to do that.
Could a temporary work-around be to filter or delete all <atom:link rel="payment" ... lines while parsing?
Yes, could be. But I don't know where and how to do this.
I found a workaround for myself: using Proxomitron as a proxy in RSSOwl, I can remove all lines that start with "<atom" if they contain "payment". That seems to do the trick, I have the full feeds now.
I still hope that there will be a fix for RSSOwl in the (near?) future... If I were a programmer I'd try and help, but I'm just a user...
Proxomitron is not a great solution for linux users, however, I played around with the Mindfactory feeds and it turns out that RSSOwl just discards n-1 news items because they share the same link and date.
Setting up Proxomitron rules to delete these lines fixes the feeds:
(It also removes the title link, which may interfere with list view. In newspaper mode, the other links are still present, as they use the RSS base link)
So that seems to be a fail of both parties. MF because they don't set up proper, unique links to product pages, and RSSOwl because it treats different news items as the same as soon as link and date match; title and description are fully ignored.
This filter rule also fixes the previous problem of being unable to copy/move news around into a news bin, and also enables proper execution (every time!) of other rules, like marking something read because of matching words in the title.
Log in to post a comment.
Sign up for the SourceForge newsletter:
You seem to have CSS turned off.
Please don't fill out this field.