Page 2 of 3

Re: RSS remember episode

Posted: Wed Apr 23, 2014 7:41 pm
by sledgehammer_999
I don't know the RSS innards but always thought that files downloaded from RSS rules are marked as read. Maybe this is a bug...
(If Dayman sees this, he may comment better)

Re: RSS remember episode

Posted: Wed Apr 23, 2014 8:21 pm
by ciaobaby
If you delete from the download history how is qBT supposed to 'know' it HAS been downloaded?

Re: RSS remember episode

Posted: Wed Apr 23, 2014 8:37 pm
by Neipas09
[quote="ciaobaby"]
If you delete from the download history how is qBT supposed to 'know' it HAS been downloaded?
[/quote]

Simple... By marking the file as READ in the RSS Feed.
RSS Downloader looks at the RSS Feeds.

IF Feed = Unread
THEN Downloader searches for matching rules
IF rule matches UNREAD feed
THEN torrent is added to download queue

This is easily fixed by adding another step in the process where
IF Downloader ADDS torrent to Queue
THEN mark file in FEED as READ

Re: RSS remember episode

Posted: Thu Apr 24, 2014 8:55 am
by ciaobaby
Simple... By marking the file as READ in the RSS Feed.
Your client does NOT have access to write to the RSS Feed, so that is impossible.
IF Downloader ADDS torrent to Queue
THEN mark file in FEED as READ
It does that already, but you keep interfering and removing the information on what it has 'read'.

Re: RSS remember episode

Posted: Thu Apr 24, 2014 9:01 am
by sledgehammer_999
@ciaobaby

The client should mark as read any item on the feed that has matched a rule, regardless if the user deleted the download later.

Re: RSS remember episode

Posted: Thu Apr 24, 2014 12:08 pm
by ciaobaby
Sure,  but it can ONLY mark it read in the local history of the RSS feed, if the user removes that historical information and the feed is updated from source, that information is no longer available, so everything in that feed is effectively new and WILL be downloaded again.

Re: RSS remember episode

Posted: Thu Apr 24, 2014 2:09 pm
by Neipas09
So there's no way for the RSS Downloader to run the same command that the MARK AS READ button does when you click it for ALL of your RSS Feeds?

Re: RSS remember episode

Posted: Thu Apr 24, 2014 3:07 pm
by sledgehammer_999
[quote="Neipas09"]
So there's no way for the RSS Downloader to run the same command that the MARK AS READ button does when you click it for ALL of your RSS Feeds?
[/quote]

There is way to run it(from code). And probably that it doesn't do it already is a bug. I already have said that 2 times.
I'll PM Dayman for this since he knows better the RSS code for qbittorrent.

Re: RSS remember episode

Posted: Thu Apr 24, 2014 4:10 pm
by Neipas09
[quote="sledgehammer_999"]
[quote="Neipas09"]
So there's no way for the RSS Downloader to run the same command that the MARK AS READ button does when you click it for ALL of your RSS Feeds?
[/quote]

There is way to run it(from code). And probably that it doesn't do it already is a bug. I already have said that 2 times.
I'll PM Dayman for this since he knows better the RSS code for qbittorrent.
[/quote]

Oh i apologize then. I didn't realize this has already been addressed.
Just trying to help make this amazing app even better! :)

Re: RSS remember episode

Posted: Sat Apr 26, 2014 5:07 pm
by Dayman
We use "id" field in rss stream to identify unique articles. If this field is not available we fall back to using "news_link" field as "id" substitue, if "news_link" is not supplied we use title and finally ignore current article if we can't find/generate idenitifer.

So, what happens here is that rss publisher doesn't supply "id" or does supply it, but the new article has different "id", thus we handle it as such.
If there was no "id" in feed, then we use "news_link", which was probably different from the previous article and again we handle it as new article.

Re: RSS remember episode

Posted: Sat Apr 26, 2014 7:14 pm
by sledgehammer_999
[quote="Dayman"]
We use "id" field in rss stream to identify unique articles. If this field is not available we fall back to using "news_link" field as "id" substitue, if "news_link" is not supplied we use title and finally ignore current article if we can't find/generate idenitifer.

So, what happens here is that rss publisher doesn't supply "id" or does supply it, but the new article has different "id", thus we handle it as such.
If there was no "id" in feed, then we use "news_link", which was probably different from the previous article and again we handle it as new article.
[/quote]

In this case shouldn't the item which has different id/news_link download anyway, regardless if the user deletes it from the transfer list?
@OP can you confirm what is/isn't happening in your case? Does it re-download only if you delete that item first?

Re: RSS remember episode

Posted: Sat Apr 26, 2014 9:34 pm
by Dayman
[quote="sledgehammer_999"]
In this case shouldn't the item which has different id/news_link download anyway, regardless if the user deletes it from the transfer list?
[/quote]

Transfer list is completely separated from RSS downloader. There's no way it can influence the articles.

Re: RSS remember episode

Posted: Tue Jun 03, 2014 3:55 pm
by maruseru
[quote="sledgehammer_999"]
[quote="Dayman"]
We use "id" field in rss stream to identify unique articles. If this field is not available we fall back to using "news_link" field as "id" substitue, if "news_link" is not supplied we use title and finally ignore current article if we can't find/generate idenitifer.

So, what happens here is that rss publisher doesn't supply "id" or does supply it, but the new article has different "id", thus we handle it as such.
If there was no "id" in feed, then we use "news_link", which was probably different from the previous article and again we handle it as new article.
[/quote]

In this case shouldn't the item which has different id/news_link download anyway, regardless if the user deletes it from the transfer list?
@OP can you confirm what is/isn't happening in your case? Does it re-download only if you delete that item first?
[/quote]
sorry for the late answer @sledgehammer_999, I didn't know this thread was active again, no it doesn't re-download files that have been downloaded and deleted before. however if another article containing the same episode appears it will download it. even if it's the same file.

I have no idea how it would react if there are two identical ids.

Re: RSS remember episode

Posted: Tue Jun 17, 2014 3:13 pm
by hyperion1is
I found this thread and I will post here if is not offtopic (not to open a new thread).

After switching from uTorrent to qBittorrent, just out of curiosity Iooked into this RSS feature also. Silly me, I was in need for it for a long time, but I thought that will not work for me, using a private tracker. But it works.
Anyway. Now I'm using RSS feature on qBt. So far so good. My only worry is about this duplicate episode issue. It didn't happened to me, so far, but I handled only 50 items in rss feed until now.
My idea is, until new RSS features are implemented, us (users) to make the most of what is already available. To exchange ideas about the filter parameters (if there aren't any privacy concern. I don't think it should be. Or is it? . This issue can't be handled with parameters alone?

I don't know in which degree this topic is up to date. Rss (now)  marks as read a downloaded item.
1. Tracker that I use, gives the option to check which tv show you want to get, and generates a custom RSS url. (so I don't need a filter for a specific tv show).
2. I use only one tracker.
3. My storage capacity is limited so I only download 480p versions, although they are not defined as such in feed.

My filter parameters are:
- must contain: "hdtv" - to avoid webrip
- must not contain: 720 1080 .S??. (I have used | operator but its displayed with space in my filter). 720 and 1080 is to avoid HD files. ".S??." is to prevent to download an entire season.
- "use regular expressions" is unchecked.

How I said, so far so good. Rss downloader downloaded only what I wanted.
@maruseru: what are the issues you confront with? 2 sources? like "fqm" and "dtv" or whatever? If so, in the example adding "fqm" in "must not contain" doesn't solve it for you?
I hope I'm not offtopic.

Re: RSS remember episode

Posted: Tue Jun 17, 2014 4:00 pm
by ciaobaby
Probably it won't be a problem on a 'private' trackers where the normally is only one 'uploader' for a particular feed.

It's only public trackers where the feed is aggregated from several uploaders or sources that duplicates will occur in a feed.