[outleradam]

For my development purposes, I would like to know if the developers have yet made a decision on implementation of non-scraped but categorically filed shows into the video library

I have sucessfully scraped a movie from MythTV's database into XBMC's library using nothing but the mythconverg database for pre-xbmc recognition. My program, soon to be called mythicalLibrarian has the potential to categorically file every single recording from mythtv into a library.

I would like to know if there is plans to make use of libraries which are not only Movies and Series. Should I be aiming for recordings named by category like: Talk, News, Shopping, Children, Music, Documentary, Animals, Special, Sitcom, drama, Reality... etc?

ABC's World News With Diane Sawyer=News=2009-12-30 17:30:00.mpg

Or is non-series, non-movie, non-planned for XBMC?

[jmarshall]

The initial plan is to support:

1. Libraries for episodic and movie information on a per-share basis (i.e. you can call the "root" of the library whatever you like, and can have as many of them as you want)
2. Each of the libraries in 1 are browsable either at the filesystem level or at a consolidated level.
3. Metadata information will be loosely connected with file locations - i.e. when you move a file it won't lose the information associated with that file. This also allows online information to be obtained via scrapers for things like plugins or upnp and the like, where URLs may change. Obviously these would still be only browsable by the filesystem (as the URL locations can change.)

Cheers,
Jonathan

[outleradam]

What I'm really wondering about is simply the file name format. What information do you need from files in order to successfully add content to the library in an XBMC recognizable format? I'm working with you on proper formatting so that I don't go off creating a unusable library. Here's an example of the information I have at my fingertips currently for every recording. There is more information available, however this is what I've determined to be pertinant information for cataloguing shows.

Code:

RESOVED SHOW NAME:Casper's Haunted Christmas-
RECSTART:2009-12-20 08:00:00
ProgramID:MV000979360000
CHANNEL ID:1035
CATEGORY:Children
FILESIZE: 2535010'kB'

Channel ID could be resolved to a friendly name like "Nickelodeon" or whatever.

Assuming this was not a movie, then I would still be able to get all information, however, currently I've got my program set up to fail if the show is not a series or a movie.

So from my understanding, the official XBMC format for movies is
Movies/Casper's Haunted Christmas/Casper's Haunted Christmas (2000).mpg
The official XBMC format for tv shows is
Shows/Sesame Street/Sesame Street.S40E08 (Stinky's Annual Birthday Flower).mpg

Please don't laugh at me because I record these shows for my 4 year old's entertainment.

I would like to know what XBMC says about other TV recordings Tell me how to format when a show is not recognized as a series or episode.

[jmarshall]

For movies, as long as you can give us the title and year we'll be able to look it up. If that's through an nfo file or through the filename that's fine. In terms of the filename, it's <moviename> (year).ext, yes.

[outleradam]

The information I've given you is through a script I'm creating called mythicalLibrarian. Those are just some random information I pulled out of my debug log. mythicalLibrarian deposits files on a network attached storage, while allowing mythtv to still serve the file when it is running. However, with mythicalLibrarian you do not need a MythTV server on the network at all times and mythTV will remove it's files when it needs space.

Maybe I have not been speciffic enough... For non-movie non-episode... like in the first post.... what shoud be the naming convention for non-movies and non-episodes? I have already stated the formats for movies and episodes.

I want you to tell me the file naming convention for ALL OTHER RECORDINGS not series and not movies

[jmarshall]

In that case we have no metadata information whatsoever, so will be showing the filename only.

[outleradam]

Without catagorization, a library is simply a repository. A library is not a library without catagorization. There has to be references which are useful for cataloguing purposes. I listed the data which can be obtained from every recording.

Channel name can be referenced and data about the television network can be extrapolated from the internet.
Show name can be referenced and data about the television show can be extrapolated from the internet
The programID can be referenced to Zap2it (my EPG supplier) for full information about the show as displayed on the program guide.
Category can be referenced as a grouping method and icons can be assigned to everything in a speciffic category like sports or news.

This is all VERY useful information for cataloguing. I can place whatever metadata will be required in the file name so that it can be catagorized.

I would like to know what the plans are, if any, for naming convention for library filing of non-movies non-episodes, because I am equipped to supply any information required.

[jmarshall]

I wouldn't rule out XBMC reading additional information from the filename. It would be nice to unify the way that we extract tvshow and movie title/year from filenames as it is. However, it's fair to say that this is always unreliable, and other methods (eg via an accompanying XML file) are preferred.

Cheers,
Jonathan

[outleradam]

Having a separate file requires manual cleanup if the file is deleted. That is not the way to go. If tagging is required, I can tag the filename.
showname-channel:nick-time:1500-genre:children.mpg

If metatags can be added to the end of the file, then that would be a great way of doing it. I can write additional data tags into the target file if it can be read by getting a tail in xbmc.

[jayspam]

The initial plan is to support:

1. Libraries for episodic and movie information on a per-share basis (i.e. you can call the "root" of the library whatever you like, and can have as many of them as you want)
2. Each of the libraries in 1 are browsable either at the filesystem level or at a consolidated level.
3. Metadata information will be loosely connected with file locations - i.e. when you move a file it won't lose the information associated with that file. This also allows online information to be obtained via scrapers for things like plugins or upnp and the like, where URLs may change. Obviously these would still be only browsable by the filesystem (as the URL locations can change.)

Cheers,
Jonathan

Hopefully you don't perceive this as being pushy, but do you have an approximate ETA or priority level of when user-designated categories will be implemented? I have a few documentaries that are currently without a home so to speak (as they are not quite a movie, not quite a TV Show).

[spiff]

exactly whenitsdone.

[jayspam]

exactly whenitsdone.

lol, exactly the response I expected. Thanks anyway.