[Kde-games-devel] Re : Re : What to do for default theme for Kapman?

[Mauricio Piacentini]

Matthew Woehlke wrote:
> We also need a way to identify what maze an svg belongs to (I was 
> thinking about generating a checksum or something, or I guess you could 
> just add some sort of identifier in the maze xml), because I don't want 
> mazes to be tied to themes :-).

Well, Kapman is not using themes (yet). But you guys should probably use 
the same approach used by the rest of kdegames: a .desktop file that 
contains meta-information about the theme, and specifies the .svg file 
(or files) to be used. This .desktop file is localizable, and it can 
contain extra information (like the fact that the theme is a full maze 
graphic one or not), so less logic and detection of files and paths are 
needed at the game level. Without proper support for loading themes (and 
the companion .desktop files) we can not consider Kapman for the main 
module at this time, as it would not be fully localizable.

You wrote that you have a theme description in-progress, can you share 
this? Maybe we can help, as almost all of the other games have faced 
this issue in the past year. And the solution might be different from 
game to game: KGoldrunner for example shares the runner graphics, while 
other games use a single .svg for all elements, including backgrounds.

You mention that you already wrote most of the code for tile levels, is 
this right? I re-read some of the messages, and what I feel from this 
discussion is that for one reason or another some of your goals or 
priorities at the moment might not match the ones from the original 
developers. And both you and them are very polite, and do not want 
confrontation, while disagreeing respectfully about some points :)
I might be wrong, but if this is the case, then of course contributing 
is more difficult. At the same time, their vision might simply be 
different, and they might be looking for simpler solutions at this time, 
and do not want to get distracted from what they consider is essential 
to complete the game. A lot of maybe's and might's... Anyway, if this is 
the case, it might be better to let them complete their initial vision, 
as a rough first cut. When this is done, it is easier to start accepting 
external input, I think. Maybe not optimal, but easier :) Notice that I 
am discouraging you guys from working together, but the original 
developers should probably steer the process for a while longer in this 
final sprint, imo.

Regards,
Mauricio Piacentini