How do I delete one of my two duplicate rows of data in Postgres?

问题

I’m using Postgres 9.5. I have the below query that is designed to find identical rows of data (but unique IDs) in my table.

select e.name, 
       e.day, 
       e.distance, 
       e.created_at, 
       e2.created_at 
from events e, 
     events e2 
where e.name = e2.name 
  and e.distance = e2.distance 
  and e.day = e2.day 
  and e.web_crawler_id = e2.web_crawler_id 
  and e.id <> e2.id 
  and e.web_crawler_id = 1 
order by e.day desc;

I ultimately want to delete one of the duplicate rows — so perhaps deleting the row with the greatest “created_at” date. But I’m unsure how to write a query to only return one of the two identical rows. How do I do that?

回答1:

There are many ways but without changing your sql much you can just do greater than instead of <> for id:

select e.name, e.day, e.distance, e.created_at, e2.created_at 
from events e, events e2
where e.name = e2.name
and e.distance = e2.distance
and e.day = e2.day 
and e.web_crawler_id = e2.web_crawler_id 
and e.id > e2.id 
and e.web_crawler_id = 1 

回答2:

You can use LIMIT:

select e.name, e.day, e.distance, e.created_at, e2.created_at from events e, events e2 where e.name = e2.name and e.distance = e2.distance and e.day = e2.day and e.web_crawler_id = e2.web_crawler_id and e.id <> e2.id and e.web_crawler_id = 1 order by e.day desc LIMIT 1;

来源：https://stackoverflow.com/questions/40431252/how-do-i-delete-one-of-my-two-duplicate-rows-of-data-in-postgres