[Bstrdsmkr]

I'm doing something similar in 1Channel. I've been experimenting with different ways to do it, here's the simplest I've found so far: https://github.com/bstrdsmkr/1Channel/bl...layback.py

You can actually see some commented out code where I've messed with using different threads and etc. I used a property of the class object instead of writing to a file because some users were having issues with permissions and etc. As a bonus, it turned out to be noticeably more efficient, about a 3-5% cpu drop on my 3.0ghz dual core. (I think the file writes were getting queued and backing up).

Catching the onSeek() event should let you handle users trying to manually fast forward too far. Thinking about it more, I think you could probably do it by percentage and get close enough. getTotalTime() gives you the number of seconds for a given file size, so from that you can get a good idea of how many seconds you've downloaded. Then you could play with it and give yourself a little margin of error to make sure you don't seek too far.

[Popeye]

Thank you!! Excellent information and help.

[wireman99]

Popeye,

Just a quick update on the "crazy-names, no rar errors".

After several hours of digging around I've come to several conclusions.

1): Doesn't matter the method used to send the nzb to SAB. Whether sent by Pneumatic or having SAB pickup the nzb from a watch folder. The results are the same. Several header names get changed to the short name format (the.name.rar) and the repair Par volumes are always at the top of the queue.

2): Changing the number of connections set in SAB to connect to the Usenet server changes the number of header names changed in the queue. The more connections the more names that are changed.

3): Lastly. After much frustration I switched from using the Newznab plugin, configured to use nzb.su, to the Nzbmatriz plugin. Doing this allowed all nzb's that were failing to work. I suspect it's the way nzb.su is constructing the nzb file.

When I get additional time I'll compare the way the two different nzb's are constructed to see what the difference might be. Not sure if this will help you in dealing with this issue as it's entirely outside of Pneumatic.

Also really like the new speed and eta features you've added. Take care and keep up the good work.

[Popeye]

Popeye,

Just a quick update on the "crazy-names, no rar errors".

After several hours of digging around I've come to several conclusions.

1): Doesn't matter the method used to send the nzb to SAB. Whether sent by Pneumatic or having SAB pickup the nzb from a watch folder. The results are the same. Several header names get changed to the short name format (the.name.rar) and the repair Par volumes are always at the top of the queue.

2): Changing the number of connections set in SAB to connect to the Usenet server changes the number of header names changed in the queue. The more connections the more names that are changed.

3): Lastly. After much frustration I switched from using the Newznab plugin, configured to use nzb.su, to the Nzbmatriz plugin. Doing this allowed all nzb's that were failing to work. I suspect it's the way nzb.su is constructing the nzb file.

When I get additional time I'll compare the way the two different nzb's are constructed to see what the difference might be. Not sure if this will help you in dealing with this issue as it's entirely outside of Pneumatic.

Also really like the new speed and eta features you've added. Take care and keep up the good work.

Thanks for the investigation information I have made the same conclusions and also looked through the SAB source code.
1, is correct. It all depends on the nzb.
2, also correct, the name is resolved once the article is read. Standard is to use the same subject as attachment or use "the.name.rar" in thevsubject , this is what SAB looks for.
3, you are most probaly right, the index sites uses different code to produce the meta information.
I know what to do and will have the code cooking for a few more days until ready.

btw, glad you liked the eta. More in this area soon...

[Popeye]

Code:

*
               , , , , ,
               # # # # #
              _#_#_#_#_#_
             {_` ` ` ` `_}
            _{_._._._._._}_
           {_  H A P P Y  _}
          _{_._._._._._._._}_
         {_ B I R T H D A Y _}
     .---{_._._._._._._._._._}---.
    (   `"""""""""""""""""""""`   )
jgs  `~~~~~~~~~~~~~~~~~~~~~~~~~~~`

[atmorell]

Congratulation, project has come very far in one year

[Popeye]

Congratulation, project has come very far in one year

Thank you!

Just realized, the lines of code committed is inversely proportional to the length of coping grinded...

[grope]

Excellent work popeye, appreciate all the hard work you and the others are putting in

Does this work with nzbmatrix? I'm considering buying a subscription to that site but only if it works with this plugin

[obmib]

Hey grope,

take a look at this: https://github.com/pzydog/xbmc-nzbmatrix/downloads
Someone posted that plugin a few pages before

[Popeye]

Hi guys, I've made a simple tool to automatically create .Strm files from .Nzb files, here's the first release. Hopefully there's not too many bugs! Let me know if you have any problems.
....
....
....

Cool! I'm sorry I havent had the time to actually try it though. Would you mind sharing the source? Win only or?

[Popeye]

I have updated the first post with links to the excellent extensions you guys have made. Sorry if i have missed someone. PM me if you want me to change the info or stuff..

[tpunder]

This project is excellent, it has completely changed the way I think about my TV storage. I've set up a branch of Sickbeard that spits out STRM files with all of the goodness that comes with Sickbeard itself. Currently testing it now on a couple of my shows. Going to see how it does for a few days and make sure nothing breaks. Shrinking my collection down to a few gigs from 10s of terabytes is making me giddy!

[Popeye]

This project is excellent, it has completely changed the way I think about my TV storage. I've set up a branch of Sickbeard that spits out STRM files with all of the goodness that comes with Sickbeard itself. Currently testing it now on a couple of my shows. Going to see how it does for a few days and make sure nothing breaks. Shrinking my collection down to a few gigs from 10s of terabytes is making me giddy!

Great you like it! Did a rough calculation 36000 items with cover, strm and nfo would use 2GB...
Special branch? Can you explain a bit more? Link to source?

To all my users,
Just finished a working proof of concept code for the "crazy name" issue. I have quite some cleaning to do since it required a total rewrite of some of the core code. Hopefully everything will work even better now..stay tuned for a few days more

[tpunder]

Still making some tweaks, I've come up with a couple ideas to clean up the way I implemented it. I'll post a link to the source once it's in better shape. Basically I added a new search setting that outputs .strm files instead of sending links to sab and does all of the usual post processing on them to move them into your library directories. The way I did it the first time through duplicated a lot of code and I want to make it a bit less redundant.

[tpunder]

Great you like it! Did a rough calculation 36000 items with cover, strm and nfo would use 2GB...
Special branch? Can you explain a bit more? Link to source?

I think this is good to go. Here's a link to the specific branch on github. Strm file support for Sickbeard

In Search settings, choose XBMC STRM. If the screen looks screwy, you might have to clear your browser cache.

On the Post Processing Tab, in the Post-Processing section:

TV Download Dir: Configure this to where you want the strm files to be created temporarily before post-processing. Do not use your library directories.
Keep Original Files: uncheck
Scan and Process: check

The other options set as desired.

One thing to note, your api key for the various nzb sites is written to the .strm files. If you ever change that key, the files will break. Unless you save the nzb locally? Not sure how that part works of Pneumatic works.

To test, I nuked one of my season folders for a TV show and had it rescrape. Worked well for me!