Import already QUEUED as paused/merge

[mysteryman]

==reason
Recently, I decided to start using yahoo pipes as referenced in a few posts here for multiple rss feeds. I did this, because whenever I use just one rss feed, there are problems with missing things, servers going down, etc. At some point, for one reason or another, the rss stops working for a few days before I really notice it's broke.

For the past couple weeks this multiple rss has worked excellently, with one exception. Because there are multiple rss feeds, sometimes there are duplicate lines. I have tried everything I can think of on yahoo's side. The unique operator does not help, because when you move from one index provider to another the subject in rss may change from "[7980]-[FULL]-[Something]-[ Something.S02E06.HDTV.XviD-2HD ]-[01/18] - "Something.s02e06.hdtv.xvid-2hd.nfo" yEnc " to "[7980]-[FULL]-[Something]-[ Something.S02E06.HDTV.XviD-2HD ]-[00/18] - "Something.s02e06.hdtv.xvid-2hd.sfv" yEnc " (notice file name listed) If there are any other ways to do this type of filtering in pipes I would be interested, however, I believe the following would be better suited to the task.

==important part
1. I believe it would be EXTREMELY handy if you could extend the current altbinz "import already downloaded as paused" feature to pause in-queue items as well (ones not downloaded yet). That would solve the problem 98% .... which brings me to the second half.

2. It would be even better if you could merge the two together. Let's say the first rss was incomplete, file 10 has parts 1-10, but needs parts 11-20 (easy numbers). Currently, the easiest way to fix this... a person would have to notice it, then requeue the entire collection while unpausing the broken file. It would be great if altbinz could figure out that file 10 was incomplete when it was downloaded... and either skip the auto pause (causing some wasted bandwidth but still working perfectly), or copy the missing parts into the still in queue item.

[davidq666]

==reason
...
For the past couple weeks this multiple rss has worked excellently, with one exception. Because there are multiple rss feeds, sometimes there are duplicate lines. I have tried everything I can think of on yahoo's side.
...

==important part
1. I believe it would be EXTREMELY handy if you could extend the current altbinz "import already downloaded as paused" feature to pause in-queue items as well (ones not downloaded yet). That would solve the problem 98% .... which brings me to the second half.

2. It would be even better if you could merge the two together. Let's say the first rss was incomplete, file 10 has parts 1-10, but needs parts 11-20 (easy numbers). Currently, the easiest way to fix this... a person would have to notice it, then requeue the entire collection while unpausing the broken file. It would be great if altbinz could figure out that file 10 was incomplete when it was downloaded... and either skip the auto pause (causing some wasted bandwidth but still working perfectly), or copy the missing parts into the still in queue item.

to the reason part:

Avoiding dupes can only be done by adapting the rss sources so that there is no or very little overlapping. Depending on the amount of Somethings u wanna filter out u should choose between using search or group rss-feeds. If there aren't to many filters to be filled one might consider using only search-feeds. Each of ur filters would then corespond with one idividual rss-feed. If the number of filters is very high this might be to tedious or breaking the restrictions of the indexers. then u could use binsearch to watch certain groups and let the alt.binz filters do the rest. the only way u might get dupes this way might be by crossposts. Using Group- and Search-eeds simultaneaously is likely to cause dupes except if u can exclude the already watched groups from the searches.

the important part:

1.) i don't get it. how should an option to "import already downloaded as paused" be extended not yet downloaded files?

2.) I am not sure i get what u mean. do u mean one corupt file or multiple files missing? why reimport instead of repair? might the "merge collections with same name" option in settings->nzb help?

[mysteryman]

I think you misunderstood me. I have far too many rss filters to do separate searches for each. Also, this does not give me the redundancy that I wanted to begin with. I have a feed for both binsearch and nzbindex running in pipes. This way, either site goes down, and I still get my rss. The reason I get duplicates is because the 'title' is (seemingly) randomly chosen by each index. One time it could show the sfv first, other times a rar file, etc. So, often (when they don't match) the one from both binsearch AND nzbindex show up. which causes altbinz to download numerous .1 files. (wasting bw)

as for the actual request...

1. let's say you queue superawesome.rar - superawesome.r25 ... and while they are downloading (not finished yet) you queue superawesome .r25 again. Currently, you end up with superawesome.r25 AND superawesome.r25.1 ... Altbinz does not pause r25 because it's not been downloaded yet (even though it will be in a couple minutes). I would like an option to pause the second .r25 when there is already a r25 in queue.

2. I was referring to one file, with one or more missing parts/articles. Now that I've re-explained the reason, the original probably makes more sense. I already have the same item being queued 2 times, all by itself. Also, as for doing it manually... doing a par recovery on a large set can take 20+ minutes, at 20mbit a single file takes seconds. Redownloading (when possible) is always faster for me.

Sometimes a rss entry will have an incomplete file due to bad timing. A few minutes later, altbinz will download the remaining files (when a new entry shows up on the rss, or in my case due to multiple rss feeds). The problem is, the two lines both have an entry for (let's say) r10. The first line has half the articles, and the second line has the other half of the articles. Together, we should be able to build a complete file. What ACTUALLY happens, is that we get one incomplete file, and one paused (depending on timing). This problem may only happen when the two rss lines come in during separate refreshes.
Summary (features only)

• import already queued as paused
• ignore 'pause already downloaded' option in the case of corrupt downloads. This would waste a small amount of bw, but would not bother me in the least.
• if queued, and not yet downloaded, merge the two r10 lines together (filtering duplicate article numbers). This would allow the first download to be successful, even if it was originally incomplete.

[Hecks]

Way to overcomplicate the use of RSS .  You're really expecting too much of it, or aren't using the right feeds.  Polling both Binsearch and NZBIndex in the unllikely event that one will go down is overkill, especially seeing as the tiny bit of extra effort needed to search manually if this actually happened is far outweighed by the kind of daily nuisance with having multiple feeds of the type you describe where posts aren't properly consolidated.  There are a gazillion better feeds out there for this kind of thing.

[davidq666]

...
Polling both Binsearch and NZBIndex in the unllikely event that one will go down is overkill, especially seeing as the tiny bit of extra effort needed to search manually if this actually happened is far outweighed by the kind of daily nuisance with having multiple feeds of the type you describe where posts aren't properly consolidated.
...

i agree

the word overkill was the first thing coming to my mind