[outleradam]

Are you sure? I have original airdate-xx/xx/xxxx date for episodes and airdate-xxxx year for most movies. It may be viable. They must differentiate the two somehow with your epg supplier. As a work around you can manually input the show ID as mvxxxx for movies, epxxxxx for episodes and shxxxxx for everything else. I will be adding the language option soon.

[malte]

Yes, I found both columns but both don't contain reliable data.

Now I just added something like this to differentiate between movies and tvshows:

Code:

RecordedTVShowsDir=/home/lom/Videos/Aufnahmen/Serien
RecordedMoviesDir=/home/lom/Videos/Aufnahmen/Filme
RecordedDocsDir=/home/lom/Videos/Aufnahmen/Dokus
...
test "$originaldirname" = "$RecordedTVShowsDir" && IsShowMovieOrDocumentary="tvshow"
test "$originaldirname" = "$RecordedMoviesDir" && IsShowMovieOrDocumentary="movie"
test "$originaldirname" = "$RecordedDocsDir" && IsShowMovieOrDocumentary="doc"
...
if [ "$IsShowMovieOrDocumentary" = "tvshow" ]; then
...
fi

It requires special storagegroup configuration in mythtv but it works 100%.

BTW, I think I found a small break in logic while adding above mentioned modification (or maybe I just misread your code): You are reading the ProgramID and check if it is "MV", "EP" or "SH". But before searching TheTVDB you test if there is a subtitle present to check if it is a movie or a tvshow (if [ ! -z "$2" ]; then...).

[outleradam]

There's no break in the logic. mythicalLibrarian is designed to work with or without database access. Without database access there is no SH, MV or EP Identification Types, so episodes are detected by subtitles. EP Identification Types(EPIT) are handled fine without Database. Without DB access both MV and SH are handled as SH Identification Type(SHIT). A show with a subtitle is handled as an EPIT. A movie can only be handled if it is identified as a MVIT by using database entries. Since there is no way to fully identify a movie without a database year entry, MVIT is only available when the database is available.

I think you may need to rename that news show data so it has a EPIT. It should not have a MV identification type.

There is a clear difference between EPIT, MVIT and SHIT. EPIT are files which can, or should, have information in TheTVDB.com. A MVIT can or should have information in IMDB. EPIT or MVIT data should be added to TheTvDb, imdb, or similar databases, and you can add that information. A SHIT either is not a episode or contains no episode information. A SHIT will not be identified. The plan is to assign a NFO file to the SHIT once there is a clear standard on how to identify SHIT. mythicalLibrarian and XBMC will not take SHIT from any source. We've all got SHIT files in our libraries, but until XBMC supports SHIT we have to just throw away our SHIT files. I've just got to wait for XBMC to start taking SHIT and specify how it likes it before I start dishing it out.

BTW. Keep track of your modifications. I'm breaking up database access and processing. That way it can be Database=mythtv, GuideDataType=SchedulesDirect, or Database=mythtv, GuideDataType=GermanMagazine or Database=iTivio, GuideDataType=SchedulesDirect.

This seems like the cleanest way to implement it and add features without changing any working data.

[malte]

Without database access ... episodes are detected by subtitles

If I read the code correctly you detect by subtitle with and without database. But I am not deep enough into it.

I think you may need to rename that news show data so it has a EPIT. It should not have a MV identification type.

It's not only this show. And I could not find a way to script this, so renaming is not an option.

I've just got to wait for XBMC to start taking SHIT and specify how it likes it before I start dishing it out.

Ok, I start taking SHIT;-)
I enabled creating nfo files if there is no entry at tvdb.com. But you are right, it's not very nice at the moment. I have to create dummy episode numbers to get episodes into xbmc. But for me it is most important to access everything that I recorded inside XBMC without editing anything manually. I enabled the modified script this evening and have to wait how it will handle the next recordings. But it was working during my tests so it should do its job.

Your idea to break up database access sounds good. But maybe you should rename "GuideDataType=GermanMagazine" in "GuideDataType=UnreliableEPG" or something similar. I am sure there will be more epg provider with these problems.

[outleradam]

If I read the code correctly you detect by subtitle with and without database. But I am not deep enough into it.

You got me thinking just now about that. Punk'd and Tosh.0 do not have subtitles. I should revise that to "or EPIT"

It's not only this show. And I could not find a way to script this, so renaming is not an option.

Have you tried to talk to the guide data provider? My girlfriend is taking a very basic introduction to computers course. She was asking me questions about it when the following caught my eye and made me think about your guide data

To be valuable, information must be accurate, organized, timely, accessable, useful and cost-effective to produce.

It seems that the guide data that you are working with is only timely and cost-effective. Have you tried checking with SchedulesDirect to see if they support your area? They may be able to help you out.

Ok, I start taking SHIT;-)
I enabled creating nfo files if there is no entry at tvdb.com. But you are right, it's not very nice at the moment. I have to create dummy episode numbers to get episodes into xbmc. But for me it is most important to access everything that I recorded inside XBMC without editing anything manually. I enabled the modified script this evening and have to wait how it will handle the next recordings. But it was working during my tests so it should do its job.

Keep in mind that a Library is only a library because it's information is categorically filed. Without a organized filing system, it is only a repository, and no better then searching through the files on a hard disk. You can add a folder.jpg and all of the files in that folder will show with that jpg. I'm really hoping that the XBMC database abstraction brings about good things with the NFO recognition. I would like to place files in a folder by the show name, but toss an NFO file in with them and name them by the time they were recorded. So that way you'd see episodes/Teenage Mutant Ninja Turtles/18JAN09-07:30:00.mpg. and it would show a picture of the box art, description, actors, studio and what not for the show.

Your idea to break up database access sounds good. But maybe you should rename "GuideDataType=GermanMagazine" in "GuideDataType=UnreliableEPG" or something similar. I am sure there will be more epg provider with these problems.

I would like to make this program as versitile as possible, but consistency is very important with guide data. We can't use unreliable epg. So if anyone has any concerns or questions about guide data, please bring them up.

[watson540]

http://pastebin.org/85238

only one movie in there "dickie roberts" and let me say..not even worth the 1 gig of hd space for even 30 seconds

most recent recordings are on the bottom

[outleradam]

ok, good, the programid comes with a mv tag

Just need to get a show now. Hopefully they will come with sh tags.

I think this will be fully possible if I can get a sh.

imagination movers:
mc2x programid: EP010806230002
thetvdb Zap2itid: EP01080623

dora
mc2xprogramid:EP003849750174
thetvdb Zap2itid: SH384975

I'm seeing patterns. I think it could increase episode recognition if I were able to remove leading EP/SH + leading 0's from ALL data received from different sources.

Removing the last 4 digits from mc2x programid would be the way to go for it.

Any idea about shows? Do you have any news shows recorded or one of the 3-4 daily spongebob's on NIK without data? I also need a movie of some sort.

[outleradam]

found it.

Code:

[LIST=1]
[*]"1054","2010-01-30 12:00:00","2010-01-30 12:30:00","SpongeBob SquarePants","","A square yellow sponge lives in the underwater city of Bikini Bottom.","Animated","myth","0","0","0","9999","1","Deleted","56","179391763","SH003077660000","2010-01-30 23:26:34","597198848","0","1","0000-00-00","0","0","0","0","1","-75","1054_20100130120000.mpg","2010-01-30 12:00:00","2010-01-30 12:30:00","Default","Default","0","0","0","Default"
[*]"1054","2010-01-30 12:30:00","2010-01-30 13:01:00","SpongeBob SquarePants","","A square yellow sponge lives in the underwater city of Bikini Bottom.","Animated","myth","0","0","0","9999","1","Deleted","56","179391763","SH003077660000","2010-01-30 23:26:29","607963136","0","1","0000-00-00","0","0","0","0","1","-75","1054_20100130123000.mpg","2010-01-30 12:30:00","2010-01-30 13:00:00","Default","Default","1","0","0","Default"
[*]"1006","2010-01-30 14:00:00","2010-01-30 14:31:00","Lost: The Final Season: Beginning of the End","","Remembering the biggest moments of the series.","Other","myth","0","0","0","9999","1","Deleted","55","44501252","SH012158000000","2010-01-30 23:25:08","619137024","0","0","0000-00-00","0","0","0","0","1","99","1006_20100130140000.mpg","2010-01-30 14:00:00","2010-01-30 14:30:00","Default","Default","0","0","0","Default"

[/LIST]
MC2XML does come with MV/SH/EP identifiers on their programid Now to strip down the data to a recognizable format. I will work on this sometime this week. today is not a good day though. I'm working on my media center and very little sleep.

notes:
1.strip down all ep+leading 0's including schedulesdirect
2.MC2XML handled the same as normal data except programid is zap2itseriesid
3.MC2XML will need trailing 4 digits removed from programid before zap2itid is applied seriesid is not applicable
4 movieairdate should only be cleared for all except MVIT

[outleradam]

TVDB has unreliable data for zap2it ID, sometimes it's a mv, sometimes it's an SH. So, we're going to remove the MV/SH/EP identifier.

TVDBZap2itid=`echo $ProgramID| tr -d MVSHEP | sed 's/0*//' `

To unify MC2XML, ScheduelsDirect/Zap2it and Tribune, program id can be translated into a proprietary Zap2it id like this:

Zap2itid=`echo $ProgramID| tr -d MVSHEP | sed 's/0*//' | sed 's/.\{4\}$//' `

This should leave dora like this
myth Zap2itid: 384975
TVDBzap2itid:384975

[outleradam]

mc2xml support Added to svn: http://mythicallibrarian.googlecode.com/...brarian.sh

With mc2xml support came a fix for random MV/SH data at TheTvDb.com. So, this is even better. Upgrade, and make sure you delete ALL working files/database leave only: *.tracking, doover.sh and showtranslations.

[outleradam]

I'm working on a configuration script. This script would basically take care of all the user settings from a GUI. This will address many configuration issues

[barney_1]

That's awesome. Will the configuration data be stored separately from the main script? I'd love to do upgrades without having to reconfigure.

[outleradam]

Currently no, but I am working on defaults so that it will first check a file, then use it's own pre-programmed data. It will of course allow the user to select if they want to use a different value. Defaults/preprogrammed will show up in the menus though

[outleradam]

Added a few features on request tonight

First, SYMLINK=Disabled means move the file and delete the database entries along with associated picture files. Thanks to barney_1

Seccond, custom user jobs can be run on a completed or failed job.
Completed means the file was moved/symlinked.Failed jobs means generic guide data , the tvdb was down, the show was not recognized, or permission errors.

Here is a list of some of the variables which can be used in mythicalLibrarian SucessfulJob or FailedJob as set in the user settings

$ShowName = input show name
$NewShowName = sucessfully resolved show name
$exx = Episode Number or "Movie" in case of a movie
$sxx = Season number or blank in case of movie
$epn = Subtitle
$ChanID = ChannelID
$ProgramID= Zap2it listing
$plot = plot
$stars = stars
$ShowStartTime = begin recording time
$ShowCategory = category like children or sports
$OriginalAirDate = original air date
$XMLTVGrabber = your guide data type
$MovieAirDate = the original year the movie aired
$mythicalLibrarianProgramIDCheck = SH MV or EP
$ProgramIDType= Generic episode with no data, Movie, or Series With Episode Data
$Zap2itSeriesID= Zap2it ID with SH, MV or EP stripped
$seriesid = TheTVDB series ID
$TvDbTime = current tvdb time
$LastUpdated = last updated time (for Episodes only others wil be blank)
$absolouteEpisodeNumber = tvdb order (for Episodes only)
$ShowFileName = name of the file (if sucessful)
$originalext = file's "EXT"ension
$originaldirname = dirname of the file to be moved
$FileBaseName = name of the file to be moved without ext
$MyUserName = name of user running mythicalLibrarian

[stratoscape]

Tried to follow this thread to its beginnings but got bored trying. I'm thinking all the hoopla is still about recognition which really is not that big of a deal to me so far, sure it would be really cool to one day see "reality" show get properly classified such as "Survivor", or "Who do you think you are" but at the end of the day I am usually just happy that MythicalLibrarian transcoded it at all anyways.

I have been puzzled as to why the tvdb.com always gets it right but mythical still dumps it to unclassified show or something.

whatever, I still am using and digging this with no plans to upgrade from April 4th SVN unknown (cant remember)

regards

Strato