Pokedex_PrintListing has a comment noting how it depends on wCurSpecies == wNamedObjectIndexBuffer; an assert would be more convenient, but is not possible since WRAM label addresses are not defined yet here.
For some reason I thought BattleAnim_Ret was never used, due to how the
battle anim loop would be exited early from. It seems I was wrong.
Removing commands before anim_ret without it being part of the `enum`
seems to break things.
I mean, technically, being able to read beyond a data table's size is
bad practice, but there's so many cases of this that it's not worth
mentioning.
For most of these cases, at least, it's safe to assume the index won't
ever get high enough to surpass such a table's length, because it's
either hardcoded or decided by the game without the use of
uncontrollable data. As such, they don't expose any actual bugs.
