search

veekun / pokedex

more than you ever wanted to know about Pokémon
MIT License
1.45k stars 636 forks source link

separation in location area encounters #156

Closed zaneadix closed 8 years ago

zaneadix commented 9 years ago

I'm sure this isn't an issue at all but I'm hoping to clear something up. I've been working quite a bit on V2 of Pokeapi which is making heavy use of the data here (which is fantastic).

I've been working on wrapping my head around encounters and how they relate to location areas and have noticed a "separation" in how they're stored in the db. For example in Cerulean Cave I've noticed multiple entries for the same area.

323,147,140,1f
324,147,141,2f
325,147,142,b1f
...
472,147,24,1f
473,147,25,2f
474,147,26,b1f

After digging deeper I've noticed the areas 323-325 relate to encounters in R, B, Y, HG, and SS and the areas 472-474 relate only to encounters in FR and LG.

Is there a clear reason for this separation? Is this just due to how the data has evolved over time? Why would they not all be grouped together? Am I reading too far into this? AM I LOSING MY MIND!? HALP!

magical commented 9 years ago

When eevee ripped R/B encounters, we decided to reuse the HG/SS location areas. This turned out not to be very useful from a presentation point of view since even where areas are shared, the encounters changed too much between Gen I and II to really show a useful diff.

Therefore when i ripped FR/LG encounters i gave them their own set of location_areas, arguing that since the areas are somewhat specific to each version group anyway, we should keep them separate in the database and let the ui handle matching up identically-named areas if it really wants too.

The other devs disagreed, so subsequent rips have reused location_areas. There's an open issue to fix FR/LG: https://github.com/veekun/pokedex/issues/103.

zaneadix commented 9 years ago

Ok cool, that makes sense. Thanks for the quick response @magical :) :+1: