Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Bengi010]

About a week ago I went through my movie collection and added or corrected the Genre tag and used Keywords to further define any extra genre descriptions.  So for example The Bourne Identity genre was action and keywords were something like mystery, crime, spy, thriller or something to that effect. 

The point was that I had a very limited and specific set of key words I was using to organize my entire collection, about a dozen total.  I realize now that I should have created a custom field to avoid what just happened, but I had no way to know you were about to unveil this new metadata scraping tool.

 From what I read it was supposed to leave untouched any fields that had information already entered but it did not.  Now there are literally hundreds of keywords in my library and it is as good as useless.  Not to mention other data that has been lost or changed.  For example the movie 11-11-11 was changed to a Turkish film called 11'e 10 kala and the data for that film entered. 

Also the cover art of many films have been changed, not for the better and I can't get the art I painstakingly downloaded and organized back. 

In a minute I'm going to restore my library from a backup prior to all this awfulness, but how do I prevent it from happening again?

I would like to use the metadata scraping function, it seems like a great tool, but only if I can pick and chose what fields it overwrites and it leaves my hard work well alone.

[Bengi010]

Two more related questions.

I have 3 other computers in the house running J River, most likely they all have had their libraries compromised.  Is there a way to easily import the library data and structure and cover from the host pc to the others? 
In the past when I have tried using a library backup from the host pc on a client pc it broke all the links because obviously they are different when the files are on a local hdd from when they have to be accessed through the network.

Second, I've noticed that the cover art used for a "Series" is sometimes different from the art used for the actual files.  (It seems to be saved in the Username/appdata/roaming/jriver/media center 17/cover art/series) folder in the case of TV Shows, for movies I can't find it anywhere.) 
Whereas the cover art I want to use is located in the folder where the actual movie file is (M:\My Videos\Babel for example) and is labeled Folder.jpg. 
The only way I've found to remove the erroneous cover art is to open the folder I just mentioned and manually delete them. 
Using the functions "remove cover art" or "quick find in file/cover art directory" has no effect.

This second issue is relevant in my case to all my movies because of the view structure I've created.  In order to group all movies from a series like Harry Potter or James Bond under one heading I give them all the same Series tag and have the library sort by Series instead of Name.  So the cover art shown in the library is the art for the Series, which is different from the art for the File, and very difficult to change.

[MrHaugen]

Is there a way to easily import the library data and structure and cover from the host pc to the others? 

Start library server on one of the computers and set the others ac library clients. The data will sync to the others. This is in my opinion, the only sane choice when you have several computers.

[Bengi010]

It looks like a few other people have had the same issues with the new movie & tv info import.  It's new so I'm sure there are some kinks to work out.  As to MrHaugen's suggestion about using the library server, I am giving it a try and have a question.  It seems like the client pc can only see whichever library the server has loaded, is this true?  I've looked everywhere in the Forums and Wiki and can find no mention of this. 

[Matt]

The new system only gets metadata for files without metadata.

You can read the definition of this here:
http://yabb.jriver.com/interact/index.php?topic=69316.msg466618#msg466618

The most basic piece of metadata is 'Description', so if that's empty, the automatic lookup may engage.

[Bengi010]

Well that explains it then.  I don't use the description field, and never bothered with actors or directors either.  I'm interested in sorting by genre date and a small specific set of keywords and that's all I use other than the obvious, name sub-type, season, episode.

My biggest gripe is that it writes over my keywords.  I'm only using about a dozen and I love the drop down click to select function.  However as soon as auto import kicks in I have hundreds of irrelevant keywords to sort through and I hate it.  Can I use the auto import function, but select only certain fields I want filled?  I'd love to have it fill in a description, actors and director, but leave everything else alone.

[MrHaugen]

In my opinion, the engine should not overwrite any tag fields if it's populated. I can understand someones frustration if this is not the case.

[Hendrik]

In my opinion, the engine should not overwrite any tag fields if it's populated. I can understand someones frustration if this is not the case.

That would suck for me.
Carnac populates the "Name" field with some random garbage extracted from the filename, if the Metadata Engine wouldn't override that, it would be useless to me.

If anything, it should be configurable what fields to write, and which don't.

[MrHaugen]

Name might be one of the exceptions, yes. Other than that, I think most people would want to keep their tags. If they don't, it should be easy to make import clear the tags with the new tag on import tool, right?

I'm all for full control of overwrite options as well as full management of library fields, sources and data. I'm just not sure if it will ever be developed.

[Bengi010]

I agree with MrHaugen, I think it would be great if the auto tagging were completely customizable.  As I said earlier if I could just use it to fill in certain fields like description and actors without changing the rest of my stuff that would be great.  Just not writing over fields with data already in them is a good quick fix, but often there is garbage data that I don't want or use in random fields, so what would be best for me is to have a list of all the fields that could be auto-populated with a check box next to each so I can pick and choose which to use.

BTW what is Carnac?

[JimH]

http://wiki.jriver.com/index.php/Carnac