Author Topic: what's the secret to a.b.etc?  (Read 144385 times)

Offline nrk666

  • Junior Indexer
  • **
  • Posts: 41
  • Helpful: +0/-0
what's the secret to a.b.etc?
« on: 2016-09-21, 06:44:05 AM »
I've been running nzedb for a couple years now and I don't seem to get much out of a.b.etc. I'm using ircscraper for pre's, have the daily dumps for backup, par2 lookups enabled, post-processing enabled, nfo lookups enabled, but I can't ever get much out of a.b.etc. With the new season, a lot of new items are turning up in a.b.etc. Is this just a regex issue? Is there something else that needs to be done to get more releases out of a.b.etc?

Filenames seem to all be of the format:

[<6digit#>]-[##/##] - "<6digit#>.r00" yEnc (##/##)

« Last Edit: 2016-09-21, 06:51:18 AM by nrk666 »

Offline nrk666

  • Junior Indexer
  • **
  • Posts: 41
  • Helpful: +0/-0
Re: what's the secret to a.b.etc?
« Reply #1 on: 2016-09-29, 10:43:48 AM »
I think my problem is categorization regex's, or the lack thereof.

I can see the releases from a.b.etc in my releases table, they are ALL at "-1" for NFO processing (not processed) by default. They seem to stay there forever. They don't show up in Misc->Other or Misc-Hashed, they don't show up in TV->Other (they are TV releases) or anywhere else. So, I tried adding a categorization regex for a.b.etc to just dump /.*/ into Misc->Other and that seemed to jar loose some post processing activity.

Am I missing some built in categorization regex's or is that something left to the user to figure out? Is there any additional guidance on what I should be looking to do here? I only found out this much by dumb luck.


Offline irishj

  • Junior Indexer
  • **
  • Posts: 46
  • Helpful: +0/-0
Re: what's the secret to a.b.etc?
« Reply #2 on: 2016-09-30, 02:34:06 PM »
Hey,

I'm having a similar issue, only noticed as the latest episodes of the big bang theory weren't available.

Latest episode shows up in a.b.etc as "347696" for 720p and "347695" for 1080p, both were missed.

Regex collections shows as ...

Ordinal 5

/^(?P<match0>\d+ - \[)\d+\/\d+\] - ".+?([-_](proof|sample|thumbs?))*(\.part\d*(\.rar)?|\.rar|\.7z)?(\d{1,3}\.rev"|\.vol\d+\+\d+\.par2"|\.[A-Za-z0-9]{2,4}"|")[-_\s]{0,3}yEnc$/ui

Ordinal 10

/^\[\s*#?scnzb@?efnet\s*\] (?P<match0>.+?) \[\d+\/(?P<match1>\d+\]) - ".+?" yEnc$/

There are 20 regex in release naming, I won't post them all here.

Offline Wally73

  • Overlord
  • ******
  • Posts: 260
  • Helpful: +28/-1
  • i'm nuts
Re: what's the secret to a.b.etc?
« Reply #3 on: 2016-10-01, 10:32:21 AM »
if they are those posts from brainisking (or what his name is) those should be reqid' on teevee (he is the only one who posts teevee ids in etc)

Offline nrk666

  • Junior Indexer
  • **
  • Posts: 41
  • Helpful: +0/-0
Re: what's the secret to a.b.etc?
« Reply #4 on: 2016-10-02, 11:40:40 AM »
Is there a special trick to getting releases in a.b.etc checked vs the a.b.teevee reqid's, or aren't the reqid's more of a "global" thing? Are just releases from a.b.teevee checked vs reqid's scraped from a.b.teevee, or are all releases checked against the reqid list?

I'll try those regex's for collections on a.b.etc on my test indexer and see what happens, thanks for those.

Offline irishj

  • Junior Indexer
  • **
  • Posts: 46
  • Helpful: +0/-0
Re: what's the secret to a.b.etc?
« Reply #5 on: 2016-10-03, 07:24:10 AM »
The posts appear to be from someone just named "brian".

Here is the info for the 720p release : https://www.binsearch.info/?b=347696&g=alt.binaries.etc&p=kingof%40pr0n.net+%28brian%29&max=250

@ nrk666 > I wouldn't bother with those regex, they are the default ones I believe so you probably already have them.  We just need to figure out how to decode / lookup those ID's and associate them with the proper release name

Offline nrk666

  • Junior Indexer
  • **
  • Posts: 41
  • Helpful: +0/-0
Re: what's the secret to a.b.etc?
« Reply #6 on: 2016-10-03, 09:25:04 AM »
Agreed. I do have everything I need to do it, I just don't understand why it isn't happening. For example, if I lookup the reqid in the predb table, I find it. But without a "categorization" regex, the releases seem to never get post proc'd where I would expect they would get matched to the reqid. 

I think I've kludged it a bit by putting in a categorization regex for a.b.etc, I'm waiting for tonight's BBT to see if its working as that's one release that sent me on this rabbit hole to begin with. I'll update with what I find tomorrow sometime, so don't sweat it for now.

Offline nrk666

  • Junior Indexer
  • **
  • Posts: 41
  • Helpful: +0/-0
Re: what's the secret to a.b.etc?
« Reply #7 on: 2016-10-03, 02:32:59 PM »
I can't grok why I need a categorization regex before the release is de-obfuscated. I would think that once a release is matched to a reqid that categorization would occur AFTER the reqid match. The reqid is the information that changes "release #####" to a sortable release name.

Offline nrk666

  • Junior Indexer
  • **
  • Posts: 41
  • Helpful: +0/-0
Re: what's the secret to a.b.etc?
« Reply #8 on: 2016-10-05, 06:25:09 AM »
I might have found a problem, whether its "the" problem or not, I don't know yet.

My PreDB has two entries for the same requestID. One is a porn release from 2014, the other is a show from Monday. Is it possible my release didn't get matched because it found two entries in the predb table? (The unfixed reqid release is still in the releases table, it hasn't been assigned either reqid entry).

Should I just purge out reqid's from the predb table older than say, 2016? (I don't really care about old releases, this is just a personal indexer.)


Offline Darius

  • Enforcer
  • *****
  • Posts: 120
  • Helpful: +7/-1
Re: what's the secret to a.b.etc?
« Reply #9 on: 2016-10-05, 12:30:07 PM »
Same reqid from 2 different groups? It happens and is unrelated to your problem, as reqids are searched by group posted.

Offline nrk666

  • Junior Indexer
  • **
  • Posts: 41
  • Helpful: +0/-0
Re: what's the secret to a.b.etc?
« Reply #10 on: 2016-10-07, 01:02:49 PM »
Ok, that makes sense, but that also might be why I'm failing to match the reqid to the release.

From what Wally suggested, the reqid might be getting scraped from irc from the a.b.teevee prebots and if the reqid's are group specific in the DB, then thats probably why they aren't being applied to releases in a.b.etc.

Is there a way to fix that?

Offline nrk666

  • Junior Indexer
  • **
  • Posts: 41
  • Helpful: +0/-0
Re: what's the secret to a.b.etc?
« Reply #11 on: 2016-11-04, 08:04:34 AM »
So, I've still been trying to figure this out.

From a few random support posts I've read, it seems like the WEB based RequestID lookups automatically started referencing a.b.etc to the a.b.teevee RequestIDs, but if you don't do web based RequestID lookups anymore (the default nzedb reqid lookup is dead) and just use the IRCScraped PreDB, the LOCAL lookups don't query a.b.teevee RequestID's for a.b.etc.

Is this accurate? If so, how can we restore this functionality?

I just use IRCScraping and want releases in a.b.etc to query the local PreDB DB and use the a.b.teevee RequestID's.


Offline Bart39

  • Overlord
  • ******
  • Posts: 171
  • Helpful: +19/-0
Re: what's the secret to a.b.etc?
« Reply #12 on: 2016-11-05, 07:20:25 AM »
no, thats not accurate - the alternate group matching applies to both web and local reqid lookups

also i don't seem to have the same problem you are reporting - all my (valid) request id's in a.b.etc are matched
« Last Edit: 2016-11-05, 07:54:10 AM by Bart39 »

Offline nrk666

  • Junior Indexer
  • **
  • Posts: 41
  • Helpful: +0/-0
Re: what's the secret to a.b.etc?
« Reply #13 on: 2016-11-05, 01:00:05 PM »
I cannot track it down. I've re-installed from bare OS a few times now and a.b.etc won't match to a.b.teevee RequestID's from local PreDB. In my Site config, do I just turn off "Lookup RequestID's" even though the URL field is blank? Or does that globally disable RequestID matching, even with local PreDB? Right now it is set to Lookup RequestID's with the URL field blank. Is that the way it should be?

Specifically, I'm talking about a show that airs on Thursday on CBS and gets posted to a.b.etc almost immediately following its completion around 8:30pm EDT.

My default newznab install has no problems, this is just the last problem I'm having with nzedb. I want to use nzedb because the categorization configurations are so much better than newznab.

Edit: seems its on Thursdays now. Who knew. Timeshifting ftw.
« Last Edit: 2016-11-06, 09:51:58 AM by nrk666 »

Offline nrk666

  • Junior Indexer
  • **
  • Posts: 41
  • Helpful: +0/-0
Re: what's the secret to a.b.etc?
« Reply #14 on: 2016-11-11, 01:03:00 PM »
Did a test last night:

Version: latest STABLE git release from 11/08 built from scratch running on freshly installed Ubuntu 16.04, MariaDB 10.1, Tmux/TPG, PHP7 from Ondrej's PPA, nginx. Followed the wiki Ubuntu install docs on the right panel.

RequestID Lookups DISABLED in Admin-EditSite with the URL entry blanked out. (DOES THIS NEED TO BE BLANK AND ENABLED TO QUERY PREDB?)

TMUX settings:
PostProcess Additional : All
Fix Release Names: Yes
Decrypt Hash Based Release Names: All

IRCScraper properly scraped the correct PreDB announcement (ReqID 358836 from a.b.teevee), visible in PreDB browser.

Release created for 358836 from a.b.etc and was categorized into Misc->Other and was never renamed from PreDB.

I'll try to provide any additional information possible, just tell me what to look for.