Intelligent RSS download

[Aohzan]

Type: Feature Request

First, @jesec, thanks for your work

Summary

When auto download from RSS feeds, don’t take multiple time, the same episode number.

Idea of implementation

From regex rules, detect episode number like s01e01 or 1x1

[jesec]

Flood won’t requeue the item if an item with the same URL has already been queued.

Won’t there be other differences (like resolution, publisher, etc) in title even if the episode number is the same? It seems unlikely to me that there will be feed items that are exactly same in title.

Current RegEx based matches should be sufficient for all those cases. Admittedly you might have to use a more sophisticated expression. Testers like regex101.com can help here.

[Aohzan]

A single feed can have multiple release from the same episode number (example: title.s01e01.720p.hdtv and title.s01e01.1080.bdrip...).
It's in this case that Flood will download every release of I just put title.* in the search pattern.

Currently, to avoid this, my search pattern have a fixed resolution, fixed source, fixed team. But if the specific resolution set isn't in the feed, I can miss a episode

[Aohzan]

uTorrent have the feature :

[jesec]

It is expected that rules need to have precise patterns. Regular expressions support OR. You may specify multiple resolution strings to match for example.

[Aohzan]

Yes I know, but it will download all the resolution as soon as they are in the feed. In my case, I only want the first release available for the episode

[jesec]

Um... I think in most cases, there are clear preferences in certain properties like resolution or publisher. It is a rare case that the time of availability go above these preferences.

Even if there is such need, most people probably won't be concerned about the duplicates. Disk and bandwidth are cheap these days (and it won't be particularly burdensome to remove duplicates).

Some new record keeping functions would have to be implemented for this feature.

[Aohzan]

Ok I understand the complexity to implement this feature, not usefull for everyone. Thanks