MovieGrabber - Completely Automated Movie Downloader

[binhex]

hi jlguerrer, do you have non ascii characters in your watched folder path by any chance?, if you could post your watched folder path here that would be good.

binhex.

[binhex]

hi rkm and al1937, i have just pushed up a beta copy of 2.1.7 for you guys to try, this should address both bugs, let me know if your still getting issues.

binhex.

[binhex]

@JO3 - Yes, I'm very familiar with different media centers. I originally started with XBMP (Xbox Media Player, prior to the XBMC project) but I have also run Sage, Myth, MediaPortal and about a dozen different Sigma/Realtek boxes. I've used their metadata-driven DB options where available, in depth.

Buy my question was simply how you go about managing the type of workflow I mentioned. I figured you might have some unique insight as you jump through considerable hoops to have your system take the extra post-processing step.

None the less, you're correct, I shouldn't have swayed off topic. The answer to my original on-topic question, it appears is that the .NFO's generated by MG are able to be used by XBMC. Thanks again for that.

hi rkm, there are 3 reasons to keep each movie in its own named folder:-

1. you can manually download and keep metadata, post images, fanart - this wouldnt be possible without folders as the images,metadata would be overwritten by the next download (dont know if this is true for xbmc, but it certainly is for media browser)

2. scrapers can more easily identify the correct movie - if you dont keep movies in their own folder then the scraper has to use the filename, which can cause mismatches esp when the filename includes audio/video, comprression, groupname etc.

3. tidiness and disk space - keeping movies in their own folder means its easy to identify associated images,metadata etc and guarantee's a clean and simple delete by deleting the movie folder. oh and it makes my OCD happy :-)

the way i would work if i were you is to add the sabnzbd+ completed folder to xbmc as a movies folder with your preferred scraper, then once sabnzbd+ has completed a download force it to run a xbmc "update library" and then a "clean library", this should scrape the movie info for you and you can then watch and delete (if option enabled in xbmc for deletion) or watch and move to the root of your movie collection.

if you want to ensure moviegrabber doesnt download an already existing movie then just add your root movie folder to the "movie downloaded" section in moviegrabber, you will of course need to go through the pain of moving each movie into its own folder (folder name must exactly match imdb name) but its a one off pain.

note moviegrabber will by default not download a movie that exists in the sabnzbd+ completed folder (assuming you leave the movie in the folder) so you do NOT need to include this path in "movies downloaded"

one other tip, i would recommend setting Enable Append Year to yes, this again helps correctly identify the movie when xbmc runs the scraper.

binhex.

[binhex]

hi rkm, i missed replying to this question

There is an option, "Enable Queuing". The tooltip is a little confusing. Is the purpose of this option simply to toggle the use of a queue instead of automatically pushing directly into automatic-download mode?

Thanks again!

ok the queue option is a way of allowing a user to decide whether they do or do not want the movie before proceeding to let sabnzbd+ download, from the FAQ:-

Q7. What is the Queued feature used for?.

A7. Due to the automatic nature of MovieGrabber it maybe necessary at times to check what nzb's are downloaded before actually sending the nzb to sabnzbd for processing, this is especially true for certain movie genres or perhaps older movies that you are not sure you always want to download. The queued feature is simply a list of movies that match the queued critera for review by the user which is NOT watched by sabnzbd. If a user decides that he does want to download the queued nzb then he can simply do this by releasing the nzb via the webgui which will then send it to the watched folder.

i personally use the queue exclusively now, with email notification i get an email telling me a movie has been queued, if it looks good then i can jump on the webui and click download. i guess it all depends how picky you are about what you download and also what sort of download quota your isp gives you.

[jlguerrer]

hi jlguerrer, do you have non ascii characters in your watched folder path by any chance?, if you could post your watched folder path here that would be good.

binhex.

I dont think i have. This is my path:

/mnt/cache/downloads/watch_nzb/movies

Thanks

[binhex]

hi jlguerrer, hmm interesting issue, ive taken a look and i cant see anything wrong with my code, the watch folder is unicode as is the name of the movie so you shouldnt be getting any encode issues, which is what the traceback is reporting. one thing i did spot is that it looks like its trying to use python 2.6 modules which of course is the wrong version of python, check this line out:-

File "/usr/lib/python2.6/genericpath.py", line 18, in exists

so i dont know if that is the issue, maybe the issue is due to you have two versions of python on the same box (im assuming you've installed python 2.7 yes?).

are you seeing the error straight after forcing a run now, or is it random as to when it crashes, or is it on a particular movie? (check history).

binhex.

[al1937]

Hi Binhex

I still got this error when unzipping over 2.1.6

09/03/2012 10:10:37 Error In MyMovies Creation Function
Traceback (most recent call last):
File "MovieGrabber.py", line 1948, in MetaData
AttributeError: class GlobalVars has no attribute 'last_run'
09/03/2012 10:10:37 Error In NZBs RSS Function
Traceback (most recent call last):
File "MovieGrabber.py", line 3645, in NZBsIndex
File "MovieGrabber.py", line 2085, in MetaData
File "MovieGrabber.py", line 1948, in MetaData
AttributeError: class GlobalVars has no attribute 'last_run'
09/03/2012 10:11:32 Error In MyMovies Creation Function
Traceback (most recent call last):
File "MovieGrabber.py", line 1948, in MetaData
AttributeError: class GlobalVars has no attribute 'last_run'
09/03/2012 10:11:32 Error In NZBMatrix RSS Function
Traceback (most recent call last):
File "MovieGrabber.py", line 3200, in NZBMatrixIndex
File "MovieGrabber.py", line 2085, in MetaData
File "MovieGrabber.py", line 1948, in MetaData
AttributeError: class GlobalVars has no attribute 'last_run'

But I tried a fresh copy of 2.1.7 and it seems to be working fine. Thanks a lot.

Cheers al1937

[RKM]

Hi Binhex,
I've never had any issues with metadata, etc -- without using folders. To give you a simple example, this is one of the options in XBMC that works just fine:
Pulp Fiction.mkv (movie)
Pulp Fiction.nfo (metadata)
Pulp Fiction.srt (subtitles, if not incorporated into the MKV)
Pulp Fiction.tbn (DVD box thumbnail)
Pulp Fiction-fanart.jpg (fanart background)

I have about 1,200 movies organized this way and it's always felt very organized to me. Again, I'm definitely *not* saying my way is the right way, simply saying that I think it's a matter of preference, and there are a lot of people that don't use folders.

The point about the foldername being more accurate so that the scraper has more success makes sense, but the thing I don't get, if you can name a folder, couldn't you rename a movie filename without much more work? If so, the file can be named as accurately as the folder.

All this is really just to summarize with ... I hope that MovieGrabber supports filenames to remove future movies from the download queue. If it only supports folders, I'm going to need to figure out some workaround to get 1,200 movies into MG.

About the "Enable Queuing" option, thank you for the very good description. This is what I had assumed it did, until I read the tooltip. The tooltip completely confused me and made me worried that it was doing something else. You may want to replace the tooltip with the explanation you just typed, it made a lot more sense to me!

Thanks again for the detailed description on all of those poitns, it was very helpful. I'll head over to test 2.1.7 now!

[RKM]

Regarding 2.1.7-
NZBmatrix & NZBs.org, both seem to work fine. I did get a small error somewhere in the UI, but I'm not sure where. I think it could have been on restarting.

Here are the logs for that:

09/03/2012 14:57:26 Error In Webui Function
Traceback (most recent call last):
File "MovieGrabber.py", line 5491, in <module>
File "argparse.pyc", line 1691, in parse_args
File "argparse.pyc", line 2347, in error
File "argparse.pyc", line 2335, in exit
SystemExit: 2

On a separate note, I have "Enable Queuing" set to "Yes" in the Web UI (and here's the .INI line: enable_queuing = yes). I had it set to "Yes" before starting. Unfortunately, MG still automatically moved every movie into SAB's directory for download, for every movie that matched my criteria. Am I doing something wrong, is there something else I need to set along with this flag?

[binhex]

hi rkm, yes it looks like xbmc works differently to media browser (media center plugin) in that it prefixes the metadata/images/etc with the movie name to prevent the filename clash

The point about the foldername being more accurate so that the scraper has more success makes sense, but the thing I don't get, if you can name a folder, couldn't you rename a movie filename without much more work? If so, the file can be named as accurately as the folder.

this is very true, the only downside for this is that you loose the release group name by doing this, which if for instance you want to try and trace missing subtitles, or you want to check that the version you have down is the best encoded version by looking at the release group nfo then you cant, but this is only a minor downside.

ok i will add it to my list to take a look at allowing filename matches as well as folder matches.

i will update the tooltip with a more accurate description :-), also on your question about the queue, once enabled you will need to choose how you want to identify movies for queuing, your two choices are by genre or by release year, you can define both and then if either match it will be queued, the settings are in configuration/imdb/queued definitions.

for instance i have queue date set to 2000, so any movie OLDER than 2000 will be queued, or if there is a genre your not overly keen on, say romance, then tick this option to queue movies that match romance. you can also force every movie to be queued by setting the queue date to sometime in the future :-). i dont overly like the tooltip on this section either, might rework it.

thanks for the traceback, ive spotted the error and corrected it, it was indeed triggered by a restart, im going to upload again to beta folder, if you can download in around an hour and try a restart and see if the problem goes away that would be good.

binhex.

[Dixao]

I'm having a problem. I click "Run Now" after configuring everything and Movegrabber does nothing... no updates on the logs, nothing. Then after letting it sit by itself for a few hours I'll come back and will randomly have movies in queue or successfully downloaded. Is this normal? It feels like it's broke. I click Run Now and wait hours and hours for any possible results. How is the program suppose to react after clicking "Run Now" from Home?

Thanks for the help.

Config :

Code: Select all

[folders]
movies_downloaded_dir = J:\Movies
movies_to_replace_dir = 
watch_dir = C:\Users\****\Downloads
nzb_dir = C:\Users\****\AppData\Local\sabnzbd\logs
completed_dir = J:\New Downloads
metadata_dir = 
cherrypylog_dir = C:\Users\****\Desktop\moviegrabber\logs
warninglog_dir = C:\Users\****\Desktop\moviegrabber\logs
errorlog_dir = C:\Users\******\Desktop\moviegrabber\logs

[switches]
enable_downloaded = yes
enable_replace = no
enable_favorites = no
enable_preferred = no
enable_queuing = no
enable_metadata = no
enable_release_group = no
enable_email_notify = no
enable_append_year = no
enable_posters = yes

[imdb]
good_genre = action,adventure,animation,biography,comedy,crime,documentary,drama,family,fantasy,film-Noir,game-show,history,horror,music,musical,mystery,news,reality-tv,romance,sci-fi,short,sport,talk-show,thriller,war,western
good_rating = 7.0
good_votes = 0
good_date = 1900
preferred_genre = 
preferred_rating = 0.0
queue_genre = 
queue_date = 1900
bad_genre = 
bad_title = 
fav_dir = 
fav_writer = 
fav_actor = 
fav_char = 
fav_title = 

[email_settings]
email_server = 
email_server_port = 587
email_server_ssl = 
email_username = 
email_password = 
email_from = 
email_to = 

[website_settings]
index_site = nzbmatrix
nzbmatrix_username = Dixao
nzbmatrix_key = *****************************
nzbmatrix_cat = all formats
nzbmatrix_lang = english
nzbmatrix_search = +brrip
nzbmatrix_minsize = 0
nzbmatrix_maxsize = 0
nzbmatrix_badreport = 
nzbmatrix_release_group = 
nzbs_id = 
nzbs_hash = 
nzbs_cat = all formats
nzbs_search = 
nzbs_minsize = 0
nzbs_maxsize = 0
nzbs_badreport = 
nzbs_release_group = 
newznab_hostname = 
newznab_portnumber = 
newznab_key = 
newznab_cat = all formats
newznab_search = 
newznab_minsize = 0
newznab_maxsize = 0
newznab_min_completion = 1
newznab_badreport = 
newznab_release_group = 

[misc]
color_scheme = darkblue
schedule_hour = 0
schedule_minute = 3
max_items_shown = 100
history_sort_order = post_date_desc
queued_sort_order = post_date_desc
moviegrabber_username = 
moviegrabber_password = 
moviegrabber_ssl = no
launch_browser = yes
test_mode = no
last_run = 10/03/2012 07:46:27
version = 2.1.7

[dragonfyre13]

Just an FYI - I put up an unraid package here: http://lime-technology.com/forum/index. ... c=18680.15

If anyone uses unraid, ideally that should help. It pulls down the package from sourceforge, autoconfigures host and port, deploys to a directory you specify, and flips it from requiring 2.7 to 2.6. It depends on the python argparse library (by the way, seems to be the only reason this needs 2.7 from my tests), and will prompt you if not installed. Runs on current beta release of unraid (v5) as well as stable (v4.7). It's been updated to version 2.1.7 as well.

[jlguerrer]

hi jlguerrer, hmm interesting issue, ive taken a look and i cant see anything wrong with my code, the watch folder is unicode as is the name of the movie so you shouldnt be getting any encode issues, which is what the traceback is reporting. one thing i did spot is that it looks like its trying to use python 2.6 modules which of course is the wrong version of python, check this line out:-

File "/usr/lib/python2.6/genericpath.py", line 18, in exists

so i dont know if that is the issue, maybe the issue is due to you have two versions of python on the same box (im assuming you've installed python 2.7 yes?).

are you seeing the error straight after forcing a run now, or is it random as to when it crashes, or is it on a particular movie? (check history).

binhex.

Hi binhex,

I have both python 2.6 and 2.7, there was something wrong in my python setup it’s now solved and I don’t have de ascii error anymore.
I have been trying to setup the downloaded movies feature but I doesn’t seem to work as in my current set up I use dots instead of spaces “The.movie.name.(YYYY)”. I believe it’s a common way to store the movies especially in Linux systems. Could it be possible to include an option to replace spaces by dots in the IMDB title before checking in the downloaded movies folder?

Thanks

[binhex]

hi jlguerrer, yep this shouldnt be a problem, i will add it to my list.

Could it be possible to include an option to replace spaces by dots in the IMDB title before checking in the downloaded movies folder?

[binhex]

2.1.7 has been released, see first post in this thread for changelog.

Upgrading

1. shutdown moviegrabber
2. unzip to same folder overwriting all files/folders
3. start moviegrabber

binhex.