Utility: Batch Metadata Importer 3.x

[lash]

Yes that fixed the probelm with the directory listing, thanks stuckless!

[lash]

I am having some pretty basic trouble. I tried reading through this thread but its quite long. So I am sorry if there is a duplicate here.

1) Is there a way to see what data the program is pulling from the media files for the query?

2) Is there a designate setting to somehow tell the program a file is a TV or movie file? Or is it just done via the regex matching?

3) With the default settings, running a scan produces something I am not sure if it is intended or not. When no match is found (which is what is happening for everything right now), the program kicks out the display title to include all the directories.

eg Display Title: TV/4th and Long/Season 1/4th And Long - Episode 01

Is this the way its supposed to work, or is there an error somewhere in my install?

5) I need some clarity in the regular expression capability of the app, if possible?

For example:

${MediaTitle}\Season ${SeasonNumber}\(.*) - Episode ${EpisodeNumber} - (*${EpisodeTitle})

This syntax I think should work for the following examples:

4th and Long\Season 1\4th And Long - Episode 01
4th and Long\Season 1\4th And Long - Episode 06 - Great sacrifices great rewards

However, it doesn't seem to work. I need to pull the top directory for the title, as some of the filenames include unicode characters for the series title but the top level directory is in standard western encoding.

Does the app support directories in the regex expression and is the zero or more character: *, supported? Is there some special syntax the program requires, such as escaping out of special characters (eg "\").

Thank you!

[stuckless]

Originally Posted by lash

Is there a way to see what data the program is pulling from the media files for the query?

Not sure I totally understand... but i think the web ui may show that... or for sure it's available in the mailcheck log (all searches / responses are logged)

Is there a designate setting to somehow tell the program a file is a TV or movie file? Or is it just done via the regex matching?

If you are using the web ui, you can force the type, or if it's a sage recording the type is TV, or it parses the filenames (basically looking for S##). The filename parsing is handled by the tv filename scrapers in the xbmc/scrapers dir.

With the default settings, running a scan produces something I am not sure if it is intended or not. When no match is found (which is what is happening for everything right now), the program kicks out the display title to include all the directories.

eg Display Title: TV/4th and Long/Season 1/4th And Long - Episode 01

Is this the way its supposed to work, or is there an error somewhere in my install?

I think that is because the the title, in sage, will include that, if a title is not explicitly set. ie, bmt is not adding that as the title. If bmt skips a file, then it remains untouched by bmt.

I need some clarity in the regular expression capability of the app, if possible?

For example:

${MediaTitle}\Season ${SeasonNumber}\(.*) - Episode ${EpisodeNumber} - (*${EpisodeTitle})

This syntax I think should work for the following examples:

4th and Long\Season 1\4th And Long - Episode 01
4th and Long\Season 1\4th And Long - Episode 06 - Great sacrifices great rewards

However, it doesn't seem to work. I need to pull the top directory for the title, as some of the filenames include unicode characters for the series title but the top level directory is in standard western encoding.

Does the app support directories in the regex expression and is the zero or more character: *, supported? Is there some special syntax the program requires, such as escaping out of special characters (eg "\").

The regular expression engine is the Java regular express engine. I have no idea how that compares to other regex engines. I don't quite understand your regex pattern, wich includes ${}. WHere are you setting this regex pattern? I would seem to me what you need to do is create a new filename scraper for your filenames (in the xbmc scrapers dir). Filename scrapers are xbmc xml files that include regex statements. Take a look at the existing scrapers... it might help. Not sure, but if you add new scrapers, you may have to restart the sagetv server... i'm sure if i've coded it to recheck that folder after startup. And one final note, you'll notice that the regex's don't use the standard windows backslash (\) for dir separator. BMT uses a URI syntax, which is portable for all OSes, and therefore the directory separator is a forward slash (/). (although for good measure you could have your regex include both [\\/].

good luck.

[MeInMaui]

Originally Posted by can3gxw

Mike,

Is that a newer version than the SageMC_6_3_9b_12.zip?

I'm on build 17. I'll try to get it out tonight.

Aloha,
Mike

[ckentman]

Hi,

So I am new to posting on the forums, and first off have to give my props to everyone who have made sage what it is today... AWESOME!!!

Here is my problem, and I apologize if it has been discussed, I have searched for the resolution already with no luck.

In a nutshell, automatic updates and command line are not working. My mailcheck.log file shows the following error:

2009-10-28 12:00:33,406 ERROR [LibraryImporter] org.jdna.sage.MetadataUpdaterPlugin (MetadataUpdaterPlugin.java:201) - Failed while performing Automatic Metadata/Fanart lookup
java.lang.NoSuchMethodError: sagex.phoenix.Phoenix.getSageTVRootDir()Ljava/io/File;
at org.jdna.configuration.ConfigurationManager.load(ConfigurationManager.java:46)
at org.jdna.configuration.ConfigurationManager.<init>(ConfigurationManager.java:40)
at org.jdna.configuration.ConfigurationManager.getInstance(ConfigurationManager.java:32)
at org.jdna.media.util.AutomaticUpdateMetadataVisitor.fetchMetaData(AutomaticUpdateMetadataVisitor.java:83)
at org.jdna.media.util.AutomaticUpdateMetadataVisitor.fetchMetaData(AutomaticUpdateMetadataVisitor.java:69)
at org.jdna.media.util.AutomaticUpdateMetadataVisitor.visit(AutomaticUpdateMetadataVisitor.java:49)
at org.jdna.sage.MetadataUpdaterPlugin.extractMetadata(MetadataUpdaterPlugin.java:176)
at sage.media.format.g.a(Unknown Source)
at sage.media.format.g.a(Unknown Source)
at sage.b1.a(Unknown Source)
at sage.ce.a(Unknown Source)
at sage.ce.a(Unknown Source)
at sage.ce.a(Unknown Source)
at sage.ce.k1(Unknown Source)
at sage.ce.access$400(Unknown Source)
at sage.ce$1.run(Unknown Source)
at sage.cs$a.run(Unknown Source)

Any help is greatly appreciated. I believe I have installed correctly... 6.3.9a of SageMC, 3.1 of BMT, and Jetty... All are functioning correctly, and manual updates work, but not command line (same error) and automatic.

Thanks in advance...

[stuckless]

Sounds like a jar version mismatch.... (most likely caused if you had BMT installed and then installed SageMC)

given that you are running SageMC 6.3.9a, then you'll need to install BMT AFTER you install sage mc. So, just downloading and extracting the BMT 3.1 package again should fix your problem.

[ckentman]

Thank you Stuckless... that did the trick.

[lash]

Thank you again stuckless! I was working on the TV Title Mask under Sage Persistence. Now I see that is an output sting isn't it. No wonder it wasn't working .

Thanks again, I will get back to work on it.

[lash]

Another new user question, but this time a quick one. I am noticing the parental ratings of movies via IMDB are pretty inconsistent. With MyMovies its much better. So my question is about MyMovies.

What field and how does BMT match a filename to? For example if I have a movie file "<movie_title>.<ext>", but in MyMovies its listed as "<movie_title> - Director's Cut". Will that still match? Or do I have to manualy remove the " - Director's Cut"?

There are two name fields in MyMovies: Local title and Original Title. Which of these fields has to match the filename?

Thank you!

[stuckless]

The support for mymovies and dvdprofiler should be much better, and it's on my list... unfortunately, my list never gets smaller, no matter how many things i knock off

i'm not sure which title field is used in the mymovies (at the momement), but how bmt searches for titles, is the same way it searches for titles online. ie, the mymovies/dvdprofiler implementation basically indexes all the titles in your xml, and then bmt will use that index when searching. So, the closer the name of the file matches the title stored in the xml, then the better your chances of getting a hit. (ideally, bmt, should just use the filename and do a lookup using that, but due to how the "search" is implemented, the actual filename is never passed into the search provider, but i'm working on that... someday)

[lash]

Thanks again! I will test it out to figure out which field, no problem.

<ignore>Is there something special I need to do to read the mymovies XML? I realize that the descriptions are coming from IMDB not MyMovies. I noticed that there is a setting called "Metadata Provider(s)" under the "Metadata Configuration" and there is no mention of MyMovies there. It says:

tvdb
imdb-2
themoviedb.org
imdb.xml

Should Mymovies be there somewhere? If so what is the proper name to call the provider?

Thanks yet again!</ignore>

Edit: I figured it out by looking at it, so ignore that question .

[sflamm]

The support for mymovies and dvdprofiler should be much better, and it's on my list... unfortunately, my list never gets smaller, no matter how many things i knock off

Consider yourself a victim of your own success!

[lash]

Is everyone new to SageTV as annoying as me? Apologies for yet another question.

I've moved from the setup stage to the UI stage and I am seeing a strange issue. If I run a scan via the web front end or do a media refresh in SageMC, the media shows up with weird titles like:

"Movies/2001 Movies/2010 The Year We Make Contact (1984)/2010: The Year We Make Contact (MGM/UA)"

However, if I do an individual metadata tagging via the BMT SageMC interface, eg Edit Video Info->Metadata\Fanart, the file title becomes a much more normal:

"2010: The Year We Make Contact (MGM/UA)"

I did not mess with the movie regex but I must've messed this up somewhere but I can't figure out where. Is there a setting that effects the web interface or media refresh but not the individual one?

Thanks yet again!

[lash]

Yet another 2 questions .

1) How do I name files for TV episodes that fall into the "specials" field on the tvdb? What is the season for these?

2) What is the best way to name double episodes?

e.g. I am using a differnt naming scheme but using the default one, would:

"Avatar - S01E19E20 - The Siege Of The North (Parts 1 and 2)"

work?

Thanks again!

[ranger]

OK Here's the scoop. I have around 1200 TV shows recorded and stored on hard drive. I am evaluating SAGE as an alternative to BTV and XBMC, kind of hoping I can find something that comes close to an all in one package. TV show metadata from sources like TVDB is important to me and so I am looking at BMT, BUT I don't want to have to rename all my files to test this and I don't have a working knowledge of REGEX so I am going to play the totally newbie.

How can I get BMT to recognize my existing TV show filenames? The format is SeriesName\SeriesName - SxE - EpisodeName.ext

Sorry if this is answered someplace else in the 102 pages of posts, but I got bleary-eyed looking for it.

[stuckless]

Originally Posted by lash

I've moved from the setup stage to the UI stage and I am seeing a strange issue. If I run a scan via the web front end or do a media refresh in SageMC, the media shows up with weird titles like:

"Movies/2001 Movies/2010 The Year We Make Contact (1984)/2010: The Year We Make Contact (MGM/UA)"

However, if I do an individual metadata tagging via the BMT SageMC interface, eg Edit Video Info->Metadata\Fanart, the file title becomes a much more normal:

"2010: The Year We Make Contact (MGM/UA)"

I did not mess with the movie regex but I must've messed this up somewhere but I can't figure out where. Is there a setting that effects the web interface or media refresh but not the individual one?

Are these DVDs? I sometimes see this with DVDs.

Originally Posted by lash

1) How do I name files for TV episodes that fall into the "specials" field on the tvdb? What is the season for these?

2) What is the best way to name double episodes?

"Avatar - S01E19E20 - The Siege Of The North (Parts 1 and 2)"

work?

That should work, although, it basically will find S01E19 and use the metadata for that. ie, the E20 is just dropped. BMT has no concept of a single TV show that has 2 episodes.

Originally Posted by ranger

OK Here's the scoop. I have around 1200 TV shows recorded and stored on hard drive. I am evaluating SAGE as an alternative to BTV and XBMC, kind of hoping I can find something that comes close to an all in one package. TV show metadata from sources like TVDB is important to me and so I am looking at BMT, BUT I don't want to have to rename all my files to test this and I don't have a working knowledge of REGEX so I am going to play the totally newbie.

How can I get BMT to recognize my existing TV show filenames? The format is SeriesName\SeriesName - SxE - EpisodeName.ext

That format should work.

[ranger]

Originally Posted by stuckless

That format should work.

Thanks for the response, but if it should work, then I need help figuring out why it isn't working.
Using the web interface for BMT and performing a scan, the metadata isn't populated. The show title, episode title, season and episode numbers do not populate correctly and so no metadata is found.
An example of how the Display Title populates is "Hogan's Heroes/Hogan's Heroes - 1x12 - The Scientist"
Any ideas?

[stuckless]

Could you post a complete filename... I'll check to see if there is a problem in the scanner. I typically name files as S01E01, but I was pretty sure that it supported SxE as well. If not, then maybe I can quickly update the regex to allow for it.

[ranger]

Originally Posted by stuckless

Could you post a complete filename... I'll check to see if there is a problem in the scanner. I typically name files as S01E01, but I was pretty sure that it supported SxE as well. If not, then maybe I can quickly update the regex to allow for it.

Thanks. I really appreciate the help. All TV Series are in their own folder in the format "Folder\Filename". For example
Stargate Universe\Stargate Universe - 1x6 - Water.avi
The Office (US)\The Office (US) - 6x3 - The Promotion.avi

[stuckless]

Originally Posted by ranger

Thanks. I really appreciate the help. All TV Series are in their own folder in the format "Folder\Filename". For example
Stargate Universe\Stargate Universe - 1x6 - Water.avi
The Office (US)\The Office (US) - 6x3 - The Promotion.avi

It appears that bmt does not like the SxE format... but that's easy to extend using the tvfilename scrapers. I can't do it tonight, but I'll try to get a new title scraper out tomorrow night that will hand the SxE format.