Import Movies Catalog

[JavaWiz]

Any way that I can set my "Location" and offline ID field to the ANT Medialabel field? You seem to be generating a number/id based on the db name.

This is where I'm putting the disk id for offline movies.

Otherwise its all looking really good.

The location field is a unique key in the db. So... how about this:

• The location and offlineID fields default with the generated number. This is basically 'AMC_' + your catalog name + MovieNumber.
  
• If the MediaLabel field exists
  • AND it's value points to an existing file, the location field is updated with this value AND the offlineFlag is set to false
    
  • If value does not point to a file, the offlineID field is set to the MediaLabel value, and the location field is left with the generated value above.
    

So in your case, the OfflineID would contain the MediaLabel field, and the Location field would contain the unique generated value.

[psycik]

yeah i've actually done that manually - and I also went and updated the details using your lookup details since it was better than what I'd got from dvdempire.

I guess as long as i can specify that for offline files that the media label is shown then the internal location field can be whateverit wants...

Is a imagesource 1 a file based image?? or something else entirely?

So that's probably a good solution what you've come up with. I think in future I'll just enter into MovieWiz and manually go and get the image. I was just handy to get the images with ant the first time.

[mymymy]

is it possible to partially populate the data, ie the location and title, and then have the rest of the information looked up manually when the xml file is imported? This way I could get a much better hit ratio on the titles when uploading movies in bulk. Or alternatively is there a way to re-lookup all movies in the db in one go?

[JavaWiz]

Is a imagesource 1 a file based image?? or something else entirely?

Yes, don't have the code in front of me, but I believe that is true. The codes get translated in the Manage tab under the thumbnail. (ie Local Source, Yahoo, Imdb,...)

[JavaWiz]

is it possible to partially populate the data, ie the location and title, and then have the rest of the information looked up manually when the xml file is imported? This way I could get a much better hit ratio on the titles when uploading movies in bulk. Or alternatively is there a way to re-lookup all movies in the db in one go?

Good ideas, both of them. Currently these features do not exist, but it shouldn't be too hard to implement one or both of them.

Look for changes in a future release.

[mymymy]

wow, quick response.

A third way, perhaps the best solution, might be to make the title resolver into an interface that would take the file location, and return the title and allow pluggable resolvers to be implemented. This would give the greatest flexibility, but would probably be more work on your end. It would also be beyond most, but would have the advantage that once a title resolver implementation had been written many others could use it. Just a thought...

[JavaWiz]

wow, quick response.

A third way, perhaps the best solution, might be to make the title resolver into an interface that would take the file location, and return the title and allow pluggable resolvers to be implemented. This would give the greatest flexibility, but would probably be more work on your end. It would also be beyond most, but would have the advantage that once a title resolver implementation had been written many others could use it. Just a thought...

Actually, it already works similar to this.

There is a routine that resolves a filename into a potential show/episode title. Then there is an abstract interface (I think it's called WizScraper), that takes show/episode as input, and returns information fields. The class that implements the interface will retrieve info based on it's source (ie IMDB or TheTvDb, TvGuide,...).

[mymymy]

cool, so a simple change to make the resolving of the filename to a show/episode/movie title an interface and a configuration option to allow me to specify the concrete class that will do the resolution and jobs a good 'un .

excellent work on the plugin by the way.

is there any option to make pluggable filters for the movies? Currently you can have comedy/action/romance etc, but having this be pluggable so that other filters could be added would be a great feature I can see filters like 'movies from the 1980s', 'movies under 90 mins', 'oscar winners' (not sure if this info is available from IMDB) etc

Is the source code available?

[JavaWiz]

cool, so a simple change to make the resolving of the filename to a show/episode/movie title an interface and a configuration option to allow me to specify the concrete class that will do the resolution and jobs a good 'un .

The current WizTools support thread can be found here and the Wiki entry here. The version on the wiki is the public release, version in that thread has not been released yet, but I hope to within the next week or so.

The config to MovieWiz is not plugable, although I've been considering adding a scripting feature, so that users could write code to scrape their own sites. That might provide you with the features you are looking for.

is there any option to make pluggable filters for the movies? Currently you can have comedy/action/romance etc, but having this be pluggable so that other filters could be added would be a great feature I can see filters like 'movies from the 1980s', 'movies under 90 mins', 'oscar winners' (not sure if this info is available from IMDB) etc

Well, you can setup specific categories to do this, and manually add entries into each category. The config program allows you to create your own categories.

Is the source code available?

I have not released the source code for my specific plugins or the wiztools, but the wiztools provides skeleton code for creating a plugin, an example plugin, a tutorial and API help file. If folks have had specific questions, I have supplied snippets to illustrate how I've done things.

[mymymy]

looks good. thanks.