You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Inspired by #335 i decided to look into whether we had any other messed up order columns in pokemon_moves. Answer: yes. We have several cases where a pokemon learns multiple moves at the same level and those rows have a mix of NULL and non-NULL order. Also the order doesn't start at 1? There's even one pokemon (Muk) where both moves have a NULL order.
selectp.identifieras pokemon, vg.identifieras version_group, tmp.*from (select pokemon_id, version_group_id, level, array_agg("order"order by"order") as"order"from pokemon_moves where level !=0group by pokemon_id, version_group_id, level havingcount(*) !=1and not every("order"is not null)) as tmp join pokemon p ontmp.pokemon_id=p.idjoin version_groups vg ontmp.version_group_id=vg.idorder by version_group_id, pokemon_id, level;
Inspired by #335 i decided to look into whether we had any other messed up order columns in pokemon_moves. Answer: yes. We have several cases where a pokemon learns multiple moves at the same level and those rows have a mix of
NULL
and non-NULL
order. Also the order doesn't start at 1? There's even one pokemon (Muk) where both moves have aNULL
order.Results:
https://gist.github.com/magical/589445759d7a7ab159fd2be74f228811
The text was updated successfully, but these errors were encountered: