Adding more torrent sites for rss d/l (1 Viewer)

[mb76pl]

I've noticed something...

bluenote, your details link looks like this:
h**p://isohunt.com/torrent_details/$id$/?tab=summary

in isohunt details link there is search phrase between chars "/" and "?" - for example:

h**p://isohunt.com/torrent_details/132793597/fringe+night?tab=summary

so, shouldn't the details link include this and look somethink like:

h**p://isohunt.com/torrent_details/$id$/.*?tab=summary

 

[RoChess]

I've noticed something...

so, shouldn't the details link include this and look somethink like:
h**p://isohunt.com/torrent_details/$id$/.*?tab=summary

That would only be required if the resulting HTML page doesn't contain the right information. If you manually go to that URL, it obviously does, so therefore not required.

 

[bluenote]

I figured out how to use SVN to get at the code today, so we'll see if I can interpret whats going on well enough to figure out what we need to do differently.

Nobody knowing how to use a feature thats included in one of the most popular plugins for this software is kind of disappointing though. I can only assume its not that no one knows, its that those who know, won't help and share their knowledge.

It makes me wonder if I've made the right choice in changing to this software.

Anyways, that aside, I will report back if I can figure this out well enough. I'm not a programmer though, I've only hacked together code, and not in a serious programming language for quite a few years.

 

[SpudR]

There's a bug in the TVSeries download thing - if you get the "process exception catched" error try un-highlighting the offending episode and trying again, by right clicking on it NOT highlighted and select your search...

 

[bluenote]

Hey thanks for posting.
I don't see how I can right click with the mouse without highlighting? when I mouse-over, it is automatically highlighted... is there some other way ?
Does this mean this part of the plugin cannot be used from the remote?


Ok. I got this to actually show a torrent in the list, AND, to download the torrent and launch it. (it annoyingly popped up utorrent overtop of mediaportal though).

Here it is:

Isohunt

http://isohunt.com/js/rss/$search$?iht=

<item><title>(?<name>.*?)</title><link>.*?</link><guid>http://isohunt\.com/torrent_details/(?<id>.*?)/.*?</guid><enclosure url='http://isohunt.com(?<link>.*?)' length='.*?'.*?Size: (?<size>.*?) MB.*?Seeds: (?<seeds>.*?) &amp.*?Leechers: (?<leechers>.*?) &amp.*?<pubDate>(?<date>.*?)</pubDate></item>

http://isohunt.com/torrent_details/$id$/?tab=summary

MB</b> in <b>(?<filecount>.*?)</b> file



I'm sure this can be tweaked, but I'm very pleased that it works at all

 

[Swede]

Does this really work for you? I cannot get it to work. It only keeps on searching ....

Edit: Erased everything and entered it again. It works fine!!! Great work!!!

 

[mb76pl]

gread job bluenote!

I can gladly confirm, that it works !

 

[bluenote]

Hey guys

I've also now fixed torrent portal (it just needed a couple tiny tweaks). My regex is atrocious and probably not very robust so if anybody wants to clean it up, please consider yourself invited to do so. I particularily have problems with whitespace/newlines. I've been having to avoid matching anything that contains those.

Here's updated torrent portal:

TorrentPortal

http://www.torrentportal.com/torrents-search.php?search=$search$&sort=seeders&d=desc&type=and&sizel=&sizeh=&cat=3&hidedead=on&exclude=

(?s)<tr><td class="alt\d n c"><a href="(?<link>/download/(?<id>\d*)[^"]*)">.*?(?:<td.*?/td>){2}<td.*?<a href="/detail[^>]*>(?:<[^>]*>)*(?<name>[^<]*).*?/td>(?:<td.*?/td>)<td[^>]*>(?:<[^>]*>)*(?<size>[^<]*).*?/td><td[^>]*>(?:<[^>]*>)*(?<seeds>\d*).*?/td><td[^>]*>(?:<[^>]*>)*(?<leechers>\d*).*?/td>

http://www.torrentportal.com/details/$id$/

files.*?<td>[^\d]*(?<filecount>\d*) files<br />


I'm working on:

TorrentZap

http://www.torrentzap.com/searchrss/$search$

(?s)<item><title>(?<name>.*?)</title>.*?<link>http://www.torrentzap.com/torrent/(?<id>.*?)/.*?</link>.*?<size>(?<size>.*?)</size>.*?Seeds: (?<seeds>.*?) , Leechers (?<leechers>.*?) , Downloaded.*?</description>.*?<enclosure url=\"http://dl.torrentzap.com(?<link>.*?)\" type=.*?</item>

http://www.torrentzap.com/torrent/$id$

<td>(?<filecount>.*?) file\(s\) -

But it doesn't work for me. This one probably involves the (?s) which I actually don't know if it works.



I'm also interested in the "News" area of mptvseries , but I dont know what to put in for "newzbin Search URL" so anybody that has any info on that it'd be appreciated.

 

[Brownard]

If anyone cares I got this working with EZTV's RSS feed and the Isohunt JSON API - HOWEVER I use my web server to parse the results and then output a nice string for TVSeries.

Not the cleanest solution I know (I was already running the server so there wasn't much extra needed for me) but this way I can overcome some of the problems with the current downloader - e.g. spoofing the file count for EZTV as this is not provided in the RSS feed (they are always single files) and overcoming the need for relative torrent url's (I pass the full url as a paramater to my server and then just redirect to the full URL).

Some issues with the downloader in general:

If uTorrent is already running the /MINIMIZED flag does not work and uTorrent will pop up, this can be stopped in uTorrent by unchecking Options-->Preferences-->UI Settings-->Activate Program Window, but this is very annoying when downloading normally from the web as uTorrent will remain minimised - OK for a dedicated client though.

I have a client with some tv shows on a mapped network drive, if a folder with the name of the downloading show exists on the mapped drive then TVSeries tells uTorrent to download to that folder - downloading to a mapped drive is obviously not ideal.

If anyone's interested in this let me know.

 

[bluenote]

Hey brownard

I'm envious of your setup ! I'm not crazy about having to maintain a webserver to get the functionality out of this though. (sigh)
I wonder what the chances are of having that one line of code changed. I don't suppose you can think of any other ways to (selectively) redirect?

On a side note, do you (or anyone) happen to know if there is a way to launch this functionality without a search target (and therefore search on whatever you want to enter)?
I see in the code there is provision to create a directory (if the series does not exist), so in theory, if we were able to affect the search term, nothing else would have to be done ..

thanks

Cory