Expand TVSeries to a sister plugin? My Films? (1 Viewer)

[emphatic]

The underlying architecture separates the movie meta-data importer from the cover art importer. While the info is both technically from the same site it is two different interfaces that the importer interacts with. And the interface is generic so it will be very easy to add new data providers. So because the meta-data and cover-art importers are seperate you could mix and match when new importers are added. So for example in the future you could pull metadata from movies-xml.com and coverart from impawards.com. Or meta-data from IMDB and coverart form movies-xml.com. Is this what you meant?

Yes. It's great to see that you take this progress in a calm matter (implementing bigger changes when things are stable). Take your time and make it right!

Emph

 

[allanp81]

I'm so looking forward to this as updating imdb info etc. on multiple clients is a major pain in the arse sometimes so to have something that will do it automatically will be amazing.

 

[piranha]

Wow. I am really impressed. So far amazing work. The only problem I can see that in my case only like 30% of the movies (mainstream too), were recognized during coverart download. The movie itself was recognized and added to the db without a problem, but no coverart available.

I think having scripting support here (similiar that is used by current IMDB implementation) would help.

 

[fforde]

Wow. I am really impressed. So far amazing work. The only problem I can see that in my case only like 30% of the movies (mainstream too), were recognized during coverart download. The movie itself was recognized and added to the db without a problem, but no coverart available.

I think having scripting support here (similiar that is used by current IMDB implementation) would help.

For the second release which will include (among other things) an IMDB scraper, I am going to try to create a generic, xml configurable scraping interface. Probably a combination of regex, and xml. on the Meedios forums they talk about a similar interface XBMC does. Haven't done much research yet, I may end up borrowing some code from the XBMC guys, but this is kind of the thing I want to do.

MeediOS :: View topic - Suggestion to use XBMC's XML scrapers for HTTP scraping

With an interface like that, it would be much easier for users to create new importers just by hacking out a few regex queries, or quickly fix a newly broken importer, which they could then post on the forums (so people dont have to wait for the next version of the plug-in).

Here it is the idea in the issue tracker: Issue 9 - moving-pictures - Google Code

 

[gamejester]

All sounds very promising fforde, can't wait to give it a try; bit sad about the mediainfo push back to version 2 but i will just have to be patient.

It was this lack of patience which led me to start to enhance the Core Movies part of MP, see this thread

https://forum.team-mediaportal.com/general-development-no-feature-request-here-48/movies-enhanced-cover-retrieval-43405/

Basically i have added a similar thing to what you where talking about abstracting the movie details search from the cover art retrieval. I have also added a movie-xml cover searcher and enhanced the IMPAwards one so it is now 100% accurate! + chucked in a few bug fixes.

The next phase for me was to finish tidying up the existing search scripts and then add mediainfo and i would be happy.....................however, perhaps that is now not the best use of my time

If you would like a hand on the coding of your new plugin drop me a line, i could look at adding in support for the various movie data sites and cover sites as I am overly familiar with them all, should be a doddle! have you thought about supporting csharpscripts, like Movies does? That way those searches are plain text readable like you want so anyone can maintain them + they are flexible as you like supporting XMLnodes, HTML scrapping, regex, etc..... Plus the implementation code is already there in the core for you to borrow

 

[daHive]

this looks extremely promising. tried the mesfilms/ant combo a while ago, but it can be a bit of a hassle. so i´m really looking forward to this thing!

 

[mattsk88]

Drop me a pm if you need help with the skinning also, (would also like to start working on a GUI for the titanium skin I'm currently working on also), Looking great keen to beta test it!

 

[gamejester]

fforde - Another idea I have been working on for movies is the concept of a 'combined source' - but again feel it could be taken a lot further in your plugin.

What is this?
Well in the work recently done on the scsript grabbers here
https://forum.team-mediaportal.com/watch-edit-videos-104/script-support-imdb-grabbing-26098/index9.html

We have come across various anomoly movies, or incomplete data, etc...

My idea here is once you have identified a moive (either automatically or manually selecting it from a list) you store 3 unique keys about it:

1. Title - not always great for searching on as funny chars and numbers are often represented differently on different sites, exmaple:
Star Wars: Episode One - The Phantom Menace
Star Wars: Episode 1 - The Phantom Menace
Star Wars: Episode I - The Phantom Menace
Star Wars: Ep 1 The Phantom Menace
etc..............

2. Year - A key factor for refining your search, but again not always consistent across sites
Superman II - IMDB - 1980
Superman II - movieXML - 1980
Superman II - IMPAwards - 1981

3. IMDB number - always unique but now all sites use it, there is a hidden search you can do on moveXML using the IMDBttNo. to retieve an exact movie.

These then never change unless you select 'wrong movie' and they are wipped. With these 3 bits of info whenever you subsiquently do a refresh movie info you a lot more useful info to seach on.

The combination of these three fields is your unique movie marker.

The user should then be able to build a 'super searcher' pulling the fiedls they want from their favorites site; ie

description - moviexml
rating - imdb
stars - imdb
actors - moviexml
etc.....

This would get the 'best guess' info back.

Then within the GUI when you see a movie has some 'poor' info in some of the fields you should be able to do a particial update and see a list from all sources. example:

I do not like the description that has been auto populated for Batman returns, I select it, hit a button and see all available descriptions from all my search sites, I then just pick the best one.

Obviously to save many people having to do this, and serve the wider community, the plugin should have an upload feature to movie-xml; so when you have a movie perfect you hit upload and it sends the data (using your account) to movie-xml.

 

[mattsk88]

Just a quick suggestion\question to do with the GUI function, Is it going to be similar to myfilms where you have the all the main views ie filmstrip list etc but choosing a film takes you to the imdb page\detail page and you launch the movie from there. Or is it going to be similar to myvideos where you need to bring up the dialogue box if you want to go to the detail screen? It would be good if selecting the movie went to the detail page unlike myvideos I reckon.

 

[emphatic]

Just a quick suggestion\question to do with the GUI function, Is it going to be similar to myfilms where you have the all the main views ie filmstrip list etc but choosing a film takes you to the imdb page\detail page and you launch the movie from there. Or is it going to be similar to myvideos where you need to bring up the dialogue box if you want to go to the detail screen? It would be good if selecting the movie went to the detail page unlike myvideos I reckon.

+1

Emph