video files, mysql

Post Reply
sustik
Newcomer
Posts: 1
Joined: Thu Dec 18, 2014 10:00 pm
Hungary

video files, mysql

Post by sustik »

Hi All,

I suspect mythtv implements a solution that I may use in another project. If I am correct I will dig in the code and find out how it is exactly done. But it would be good to get confirmation that my suspicion is correct.

I assume that mythtv uses the mysql database to store the metadata pointing to the video files (and of course other stuff). I assume the files themselves are not in mysql, instead their name (and location if several storage locations are used). Is that so?

How are the files transported from the backend to the frontend for display? What protocol is used and where should I look in the code?

I would like to implement a solution to an archive that hosts mid-to-large files. I thought I would put only the metadata in mysql and the files themselves will be accessed once their location is looked up in the database (together with some other attributes). The files should be possible to be processed in a streaming manner without downloading them just like for video files. (This is JSON data compressed.) Will I get some useful ideas browsing the mythtv code?
User avatar
Plerp
Junior
Posts: 18
Joined: Wed Dec 10, 2014 10:11 pm
Australia

Re: video files, mysql

Post by Plerp »

sustik wrote:Hi All,
I assume that mythtv uses the mysql database to store the metadata pointing to the video files (and of course other stuff). I assume the files themselves are not in mysql, instead their name (and location if several storage locations are used). Is that so?
Correct. File location settings are entered by you (unless you use defaults) and are accessible in the 'settings' option of the software. Both the frontend settings and backend have options for entry of file locations. In the back end I believe they implement something they are calling storage groups while the settings in the frontend may or may not be related. Sorry, obviously I'm not an expert. Hopefully someone else kicks in.

And yes, all metadata is stored in the mysql database (default 'mythconverg' with uid/pwd set as myth/myth).

Code is written in C++ I think and stored on GitHub (which has a nice search feature to dig into the source).

I can't answer your questions about transport protocols.
I would suspect that the significant complexity of MythTV might make the code something where 'useful ideas' are not necessarily something you would run across unless you take (quite) a bit of time getting familiar with things. However, I am just considering looking into the code myself. I'm not a contributing developer in the remotest sense. The most I've done is customize some themes for my own use.
Post Reply