From: Rick W. <wes...@pu...> - 2010-01-12 14:11:04
|
brett lentz wrote: > On Mon, Jan 11, 2010 at 3:44 PM, Rick Westerman <wes...@pu...> wrote: > >> Ah, I think I just answered my own question on why there has to be a >> per-game Tiles.xml when the data is being generated from a global >> Tiles.xml file. It looks like -- for whatever reason -- there does >> need to be a one-to-one correspondence between the tiles in >> TileSets.xml and Tiles.xml. Thus since TileSets is limited on a per- >> game basis then so does Tiles.xml. >> >> -- Rick >> > > If memory serves, the reason we decided to do it this way is because > there are a few tiles out there that use duplicated tile numbers. > > So, we need to be able to do things like, "Tile #88 in game X is image > #1088" while "Tile #88 in game Y is image #2088". > > > ---Brett. > Definitely need that ability. And thus the need for a per-game TileSets which has the 'pic' option (along with others such as quantity and permissible upgrades.) It is just irritating that there is this firm one-to-one correspondence between the Tiles.xml file (which gets generated from a global file) and TileSets.xml. A missing or extra tile in one or the other other causes the program to complain & crash ... and not always coherently. On the other hand things are the way they are. I've certainly written my share of programs that have strange behaviors. Often because of retroactive modifications and growth. Since I hate it when people make suggestions on what I could do better without actually doing the work -- and since I am not offering to make the program any tighter (at least at this point) -- I'll just shut up on the topic and work with what exists. I am almost done with 1876-30. It has been like pulling teeth but maybe I will have a good document at the end of this process. -- Rick Westerman wes...@pu... |