Looks like Atomic parsley describes those tags. I will have to download MetaX. Couldn't find a link for iDentify (not exactly a good Google search term). If you have the link, won't you pass it on.

It seems Media Reader is getting to have a gazillion options - not sure that it continues to be user-friendly. Might need some wizards. I was going to shelve the project but it just won't die ...

Hi,

yes, I am sorry I am being late on it. Work has been a bit crazy. It should go out this weekend as most of it is already done. And I will continue to provide fixes, so feel free to send bugs to me.

Looking forward to seeing the last drop for this. Waiting on rebuilding my library after an xbmc wipe and reinstall so I can get everything set up properly.

Thanks for the input. Let me add a couple of comments (and yes, the doc is an ever-expanding challenge):

Second, since some popular skins really prefer it if you have all your movies and supporting files in the same directory, optionally create this directory, move the media there, then create the supporting files inside the new directory. Otherwise in a setup like mine where are the media files are in the same directory, things get insanely cluttered after the supporting files are added.

This is pretty much how I work, too. I dump a new file into a directory '_Incoming' or such, then run the 'Process New' and then 'Sanitize Movies'. This will prompt you fror a target directory which must be on the same drive and redistribute the files, i.e. move every file (including the accompanying trailers, nfos, artwork,etc.) into its own directory. The anme of the directory is the TITLE of the move. So make sure you don't have any clashes or it won't move a particular file. This can generally be remedied afterwards.

I will add add the IMDB lookup for the drop this weekend. It's a no-brainer. Is your whole title simply the IMDB ID or can it be gleaned from the name? How about a regex with a capturing group for the IMDB ID? Do you use the full ID (including the 'tt' or just the number)?

I will have to look at the sample code for reading the tags. I am not big on buying stuff from iTunes just to test this. Do they have samples they give away like Amazon?

Anyhow, so I think one of your problems is already addressed, the second one will be this weekend with the thumbnailing (It is using ffmpeg for the thumbnail creation).

So check again Sunday night or subscribe to the announcement thread in the forum.

I am also still looking at speeding up the 'process new' I/O with Java's NIO but am not sure how much that will do. It's really only a fraction of the socket time anyhow.

Also, please note in the setup docs that one should use the CLI version of mediaInfo as opposed to the gui (otherwise you have to clsoe the gui on each file to continue).

Links to the required programs in your setup docs would also be very helpful indeed.

I had a nice long post typed out with various suggestions and improvements and my arguments behind them, as well as compliments for your work. But I bumped the back button on the keyboard and lost everything I typed. So here is a quick summary. Forgive me if I sound a little terse. I retyped my opening 3 times, then lost everything when I was two sentances from finishing. I was gonna say to hell with it, but some of these suggestions are really needed.

Many people include the imdb code in their file names for movies. When this imdb code is there, Media reader wont find any entires for the files. So, look for the imdb code in the file name, and use it for lookups. Also, include an imdb code token for renaming movies. For example, my file names are like "Reservoir Dogs [1992] [tt0105236].mp4". Heck, some people use only the imdb code for their file names.

Second, since some popular skins really prefer it if you have all your movies and supporting files in the same directory, optionally create this directory, move the media there, then create the supporting files inside the new directory. Otherwise in a setup like mine where are the media files are in the same directory, things get insanely cluttered after the supporting files are added.

Third, media files complatible with iTunes based devices can contain metaData (stores in a similar manner to ID3 tags in mp3 files) containing all the same information you scrape the imdb. Some people go to great pains to get all this data in their files. Using it when it is present would save bandwidth, and likely speed up conversions of entire libraries. There are cross platform open source projects that can read and write this data, the older, program AtomicParsley is more popular, and a bit more stable, but hasnt been actively developed in years. The newer mp4v2 project is still lacking in some features the last I checked, but can be faster, and is under pretty active development (last I checked).

If you look at the regex used in the perl Module, Video::Filename, you will see a very complex series of regexs that get a lot of data from damn near any video filename. I use this module myself in a program I wrote called iDentify ( http://www.macupdate.com/info.php/id/29847/identify ), which does the same thing for iTunes libraries as Media Reader does for XBMC libraries. Its not flashy (just a perl script with a crappy gui generated by command line programs), and requires nothing from the user other than having the necessary info in the file name (some naming conventions are more reliable than others, but I have only encountered a few non-sensical ones that throws it off.) However, it gets all the data from the imdb/tvdb, and even makes a few good guesses at imp award posters with about an 80% success rate for finding posters. You can even include your own posters if you like. However, that is neither here nor there, since the point of this post was not to pimp my program but to suggest features to improve yours.

Since all my files contain the imdb codes in the file names, I would have to manually update each of my 2000+ files in Media Reader to get info already contained in the metadata, I'm afriad Im going to have to revisit Media Reader at a later time when either you can do lookups directly from pre-suppled imdb codes found in the file name, or Media reader uses metadata already supplied inside the media files.