First off, I just wanted to congratulate you on an amazing product. I migrated to Linux just recently after trying a few times over the last few years to do so. I could never get used to it, but this time I dove in and am not looking back. This is even better than utorrent.


Now the problem I am having is not having any clue how to get the RSS filters on the plugin work. I've never really had much experience with auto-downloading from RSS. However, I use the ElectricSheep screensaver and I am using the RSS feed to download new sheep.

The feed is here:
http://v2d6.sheepserver.net/gen/rss.xml

As you can see, the torrents are all numbers. Now, can somebody please explain exactly how the filtering works so I can have it auto-download sheep, or perhaps point me in the direction of a guide or something?

Also, to have the RSS auto-refresh the feed, I have to have it ignore TTL and set my own time...right?\

Thank you so much for your help in advance.

Hi erissiva,

Thanks - I've only added a small bit to the project, but still appreciate the compliment. George and Ivan have done awesome work on ktorrent - without their work I'd never have bothered to do my bit

The rss feed reader works using regular expressions. http://en.wikipedia.org/wiki/Regular_expressions or a great book on them is http://www.oreilly.com/catalog/regex/
You'll find them useful in a lot of other situations as well - I doubt you'll get more than basic use out of them for the rss plugin.

The plugin uses your regex to search the title, description and link of the items you send. Ideally you want to create something unique to the item you wish to download - this way when you add other feeds you won't end up downloading all of their items. For yours I'd suggest you just use the regex sheepserver (as that's in all the links on your feed).

The rss plugin only checks new items and it also checks against the history in a filter so it doesn't download the same item twice - so even if you added the feed a second time you shouldn't end up downloading things twice.

Some feeds will supply a TTL - this is the ideal amount of time between refreshes. By default the rss plugin checks according to that - if it's not set in the feed it defaults to hourly. If you ignore TTL then you can set the refresh frequency yourself. As long as the feed is active it will refresh automatically (with or without ignore ttl).

Hope this helps you out.

Cheers,

Alan.

Thanks so much for a reply. I've been taking a look at regular expressions, and I think I should be able to put together something without any trouble.


As for the auto-checking of RSS feeds, I was confused because of the way that it deals with new items. When I add the feed, it goes chronologically from top to bottom. However, new items show up at the top (where I wasn't looking) in reverse order. Shouldn't it be consistent?

Some examples:

Heroes.*.HDTV*.*(LOL|XOR)
Will match Heroes episodes in HDTV from LOL or XOR release groups


For my excludes I have:

AC3|720p|x264

Which excludes anything with AC3, 720p or x264 in the filename.[/b]

For my feeds, I mainly use:

http://tvrss.net/feed/eztv/

I know nothing about regular expressions and it took me something like 2 or 3 hours to find out I had to use '|' to search. Anyway everythings works perfectly and I'm going to watch XX0314 now

BTW, if you're using KDE it has a very nice regex editor that I found trying to get the Ktorrent RSS function working. I haven't found a way to start it directly but if you open Kwrite, CTRL-f, check regex and hit the button 'edit'. (NOTE to KTorrent developers: Perhaps there could be a button in Ktorrent that opens it?)

Hi erissiva,

You're right - it should be consistent - I never noticed the original load was chronological when I wrote it - oops.

I'd also suggest that for you split unrelated components onto separate lines. This way order becomes irrelevant.

i.e. for desperate housewives you may wish to have two lines
foxy.*ladies
(dsr(ip)?|[ph]dtv)
this will make anything with the word foxy before the word ladies and any of dsr, dsrip, pdtv or hdtv match the filter.

Also the filters break regex a little in order to provide more features. Any line which starts with ! will be treated as a not for everything else on the line
!dirty
will mean that anything with the word dirty won't match the filter. But if you put
[^(dirty)]
"dirty" would correctly get a not on the second filter, but "vdirty" would still match because the letter v is not dirty. But with the support for beginning with ! both "dirty" and "vdirty" will not match.

Hope this helps.

Cheers,

Alan.

I think I kinda understand this but it is way too difficult for a newbie like myself - but anyhow, can anyone confirm that this is the CORRECT way to do the RSS?

Lets say I'd like to autodownload Foxy Ladies as above (S03E01-E25)

Will: *Foxy*Ladies* be enough for the regex to correctly interpret the name? should it be lowercase or uppercase? The above example suggests that a lowercase will find both lower and uppercase scenarios?

What settings should the Series stuff be set to? Is Season 3-10 and Episodes 1-25 good enough or should I be more specific (IE Season 3, Episodes 1-25)?

vlad wrote:BTW, if you're using KDE it has a very nice regex editor that I found trying to get the Ktorrent RSS function working. I haven't found a way to start it directly but if you open Kwrite, CTRL-f, check regex and hit the button 'edit'. (NOTE to KTorrent developers: Perhaps there could be a button in Ktorrent that opens it?)

I found one called "KRegExpEditor" in the (K)ubuntu repositories.