• 1
  • 10
  • 11
  • 12(current)
  • 13
  • 14
  • 20
New DB Structure for all libraries + DBs [update 2015 06 13 / Released SQL 1.3]
(2014-10-20, 08:22)phate89 Wrote: But I don't think it's a good thing to do since kodi doesn't update everytime the info but only once.
For example I'm Italian and if I scrape the whole series when I add the first episode I will get the info there are at the moment I scrape the first time. They might be incomplete (old/missing artwork, less episode votes missing translated episode/plot). Probably it's not a big deal for someone but for example for foreign users that wants localized plot and tiles (and maybe provide it theirself) it is..
This is a general issue/limitation, not related to missing episodes. Either way, fetching extra info is not necessary for the feature.

(2014-10-20, 08:22)phate89 Wrote: I think it's possible to get it with an addon.
There are 2 options. The first is a plugin that shows a list of tv shows you have in the db. When you select a tv show it will get data from thetvdb, compare them with files available in the database and shows missing. This is strictly to see what files are missing and quite easy and faster.
The second one is trickier and probably requires helix (Gotham can't add sources from code).
Everytime after an update (or manually) the addon runs and search for missing files. If files are missing it creates dummy files in addon_data folder and it scrape that folder so you will have them directly in the database. The problem with this approach is that you run twice the scan and it's harder to build but it's a lot more powerful (for example you could add a tag to dummy episodes and filter them easily).
I might try to write something like this if I found some time but it's not easy
As discussed to death in the other thread: it just such an ugly hack that will at best only work in some circumstance (thetvdb is not the only scraper you know. You'll have to reimplement the whole scraper system. Use may have overwritten scanner config, in which case the whole scanning approach will break down. To name a few). It' not good enough. And easier to implement in core anyway.
Reply
Sorry for the silence lately but my life has been a mess!

Will be back soon on this and will produce the physical schema to advance.
If you think I'm useful please use the +/- button to raise my reputation
Reply
(2014-10-26, 12:29)takoi Wrote: is is a general issue/limitation, not related to missing episodes. Either way, fetching extra info is not necessary for the feature.
It is. If you want to show them you need the info of the episode. It's not only a general issue. It's a minor issue that becomes major when you scrape everything the first time.


(2014-10-26, 12:29)takoi Wrote: As discussed to death in the other thread: it just such an ugly hack that will at best only work in some circumstance (thetvdb is not the only scraper you know. You'll have to reimplement the whole scraper system. Use may have overwritten scanner config, in which case the whole scanning approach will break down. To name a few). It' not good enough. And easier to implement in core anyway.
That's not true. You don't need to reimplement anything.
Excluding that thetvdb is used by the majority of users if you want you can implement as many as you want easily and set it as an option of the addon. You don't need to implement anything. It's enough to let the user create a video source with the addon_data and chose the scraper they want with the settings they want.
There are only 3 major limits:
You can't display them with a different color in list (for example greyed out), you can't toggle them with an option (but you can create playlists to list them and add options to remove all shortcuts instantly) and you can't use more than one scraper.
Implementing it in the core is not that easy at all. You need to change the whole episode handling and you need a way to update the episode infos after the first scrape (that will slow down the update).
And again that's not a database problem. The current is not a limit for this..
Reply
Hope things are better M!!!
Reply
Still quite messed up but trying to get back to normality!

I promise I will work on this as soon as I can!

M
If you think I'm useful please use the +/- button to raise my reputation
Reply
Team,
I've upgraded the folder with the physical model.

Now I will produce the SQL for the tale creation and then the views.
If you think I'm useful please use the +/- button to raise my reputation
Reply
You're back! Hope life is treating you well.
Reply
Life is getting better and now I'm trying to generate the correct SQL for SQLite and MySQL

so that we can advance on this. Maybe for Kodi 15 Smile
If you think I'm useful please use the +/- button to raise my reputation
Reply
thumbs up! i'm watching this effort because at some points games are gonna need a first-class library
Reply
Any updates on this? Maybe this parts of this could be a project or subprojects for Google Summer of Code?

http://forum.kodi.tv/forumdisplay.php?fid=220
Reply
It looks like at least one or two GSOC projects could be indirectly dependent on this

http://forum.kodi.tv/forumdisplay.php?fid=220

Also a few that would directly benefit from the work being done here by this project
Reply
I've started generating the SQL and the real DB and found several errors in the relationships.

using SQLServer Express and it's tools I've created the DB and is available for everyone to see. Also there is the SQL.

From that one I generated the MySQL

In my OneDrive you can find the SQL for SQLServer, the DB. Hoping the team will import this to enhance the library management Smile
If you think I'm useful please use the +/- button to raise my reputation
Reply
here the folder with everything: http://1drv.ms/1IO2Jn3
If you think I'm useful please use the +/- button to raise my reputation
Reply
(2015-05-06, 18:12)m.savazzi Wrote: I've started generating the SQL and the real DB and found several errors in the relationships.

using SQLServer Express and it's tools I've created the DB and is available for everyone to see. Also there is the SQL.

From that one I generated the MySQL

In my OneDrive you can find the SQL for SQLServer, the DB. Hoping the team will import this to enhance the library management Smile

I looked at that only yesterday and I wanted to reply this evening to let you know about that relationships Big Grin
Here's the logical view where I fixed relationship errors:
https://dl.dropboxusercontent.com/u/2219...gical3.pdf
Check if we found the same errors or if you missed something
Reply
I also fixed the person - asset relationship because the asset is not connected directly to the person, it's connected to the person's role. A person could be a director, producer and actor of the same movie simultaneously etc
Reply
  • 1
  • 10
  • 11
  • 12(current)
  • 13
  • 14
  • 20

Logout Mark Read Team Forum Stats Members Help
New DB Structure for all libraries + DBs [update 2015 06 13 / Released SQL 1.3]3