[m69bv3]

Hi everybody!

I plan to make a php front end for the library, in order to check the movie/tvshow I've seen in other way than XBMC.

This front end should acess to the XBMC library read and write, is it a problem to write in the library when XBMC is running? Should the modification taken by XBMC on the fly or after restart?

Thank's for your answers!

[azido]

Hi everybody!

I plan to make a php front end for the library, in order to check the movie/tvshow I've seen in other way than XBMC.

This front end should acess to the XBMC library read and write, is it a problem to write in the library when XBMC is running? Should the modification taken by XBMC on the fly or after restart?

Thank's for your answers!

you can access i.e. the movie database even when xbmc is running. i've checked this right now using sqlitespy changing a genre entry for a movie. the change got displayed the next time i've triggered movie information without the need of a xbmc reboot.

please keep me updated on your forthcoming, since i would really like to get such a frontend that might be a bit more comfortable than sqlitespy

[m69bv3]

Thanks for your quick answer!

I will keep you updated!

[azido]

Thanks for your quick answer!

I will keep you updated!

that would be nice & you're welcome..

[Rand Al Thor]

This sounds interesting. Can you share some of what you have planned for this frontend?

[m69bv3]

This sounds interesting. Can you share some of what you have planned for this frontend?

For movies & TvShow:

• List
  
• Search
  
• Description (Picture, plot, etc)
  
• Mark as seen/unseen
  
• Export the list for friends
  

the aim is to use with a PC web browser or my PCH

Any others ideas are welcome!

[Rand Al Thor]

Will it also be a web remote? That could be pretty cool. You would have full control over the library with access to all the movie details. This would be great to have running on a web pad or small screen. You could do all of your browsing/searching until you find what you want then play it on the "big screen." This would be a nice setup in conjunction with my nokia770 and hd projector. Cheers.

[m69bv3]

Will it also be a web remote? That could be pretty cool. You would have full control over the library with access to all the movie details. This would be great to have running on a web pad or small screen. You could do all of your browsing/searching until you find what you want then play it on the "big screen." This would be a nice setup in conjunction with my nokia770 and hd projector. Cheers.

Yep it should be possible using XBMC http API!

[azido]

For movies & TvShow:

• List
  
• Search
  
• Description (Picture, plot, etc)
  
• Mark as seen/unseen
  
• Export the list for friends
  

the aim is to use with a PC web browser or my PCH

Any others ideas are welcome!

are you willing to edit content within the frontend, or do you simply want to do sql calls and get a proper output?

[m69bv3]

are you willing to edit content within the frontend, or do you simply want to do sql calls and get a proper output?

I do not plan editing further than mark seen/unseen. But I can plan to edit information such as description or any other stuff.
Editing plot, duration, year, etc.. should be easy but adding actors or any multi-related information should be more difficult without reversing the Database Schema.

[Rand Al Thor]

billyad has a pretty cool feature in his Media Companion program. It spits out an alphabetical list of of all your movies in html. When open in a browser each movie is also a link to it's page on IMDB. Pretty basic, but cool nonetheless. You should check it out if you haven't already.

[m69bv3]

billyad has a pretty cool feature in his Media Companion program. It spits out an alphabetical list of of all your movies in html. When open in a browser each movie is also a link to it's page on IMDB. Pretty basic, but cool nonetheless. You should check it out if you haven't already.

Thank's for the tip, but I'm french I use another website than IMDB and I want a web based tool.

[Rand Al Thor]

Ahh, yeah I wasn't suggesting that this could replace or be substituted for what you were working on. I thought perhaps you could use to see how billyad did that and modify it to auto-populate the information you will be using in your web app. I am assuming he has a script of somesort that scrapes all the names and urls from the .nfo's that are then output to html. You could use a similar script that runs either on a timer or at the user request that grabs both names and links (to Imdb or what ever scraper you use in your country, just whatever is in the database), plus it could grab posters, plot synopsis, release date etc etc and populates a "web library". Having the information generated like this might be faster on the ui side than querying the database on the fly. Just a thought.

[azido]

I do not plan editing further than mark seen/unseen. But I can plan to edit information such as description or any other stuff.
Editing plot, duration, year, etc.. should be easy but adding actors or any multi-related information should be more difficult without reversing the Database Schema.

editing any field in this db shouldn't be difficult. the harder part is to "rebuild" dependencies like i.e. you edit a genre for a specific movie which has to be relinked against the genre list afterwards. that's why i asked

[m69bv3]

Ahh, yeah I wasn't suggesting that this could replace or be substituted for what you were working on. I thought perhaps you could use to see how billyad did that and modify it to auto-populate the information you will be using in your web app. I am assuming he has a script of somesort that scrapes all the names and urls from the .nfo's that are then output to html. You could use a similar script that runs either on a timer or at the user request that grabs both names and links (to Imdb or what ever scraper you use in your country, just whatever is in the database), plus it could grab posters, plot synopsis, release date etc etc and populates a "web library". Having the information generated like this might be faster on the ui side than querying the database on the fly. Just a thought.

You're right, I've ever built my own php "scraper" for movie and tvshows, but XBMC scrapers system is really effective, I'm not sure there is a more convenient way to do the stuff. The work is more on scrapers themselves (more input format, real nice search & match algorithm). For the second point, the database is really speed when queering even with more than 500 movie and 2000 Tvshows Episodes, make a parallel database is useless from my point of view. Maybe XBMC should use a real DBMS rather than SQL Lite.

editing any field in this db shouldn't be difficult. the harder part is to "rebuild" dependencies like i.e. you edit a genre for a specific movie which has to be relinked against the genre list afterwards. that's why i asked

I agree, take care about dependencies is the hard part of the work, maybe if one day the database schema is published it should be easier than just reverse engineering the actual database.


From my point of view It could be nice to cut out Database, Scrappers, UI in XBMC, just tough at XBMC like an UI with a shared database.