Re: [stella] ROM image classifications

[Glenn Saunders <cybpunks2@xxxxxxxxxxxxx>]

The hardest part about defining the actual XML schema is the relationships.

The top level of a game isn't really the ROM image, it's the title 
itself.  The title is independent of any ROM image.  In most cases you are 
going to have at least two different ROM images for that title, the NTSC 
and PAL versions.  From there you might also have revisions, like the 
copyright notice version of Asteroids.  Then you have hacks and pirates and 
proto betas and so on.  And for Starpath multiload games you have multiple 
files that group together into one entity also.

When designing a database, you usually want a numeric primary key rather 
than a string, so that if you decide to rename the title, you don't have to 
rename all children of that title.  It also helps when you have two games 
that have the same name but are in fact different titles.

So I would suggest an initial grouping of the 2600 catalog based on 
original titles and their different versions, and probably using guids as a 
primary key because it's easier for anybody to add to the db without 
worrying about using a duplicate integer id.

AA has already done most of the actual schema work.  You just have to 
convert the SQL table structure down into an XML schema and add the 
necessary goodies that emulators need.


----------------------------------------------------------------------------------------------
Archives (includes files) at http://www.biglist.com/lists/stella/archives/
Unsub & more at http://www.biglist.com/lists/stella/