Take only the first record of each condition

Question

Good afternoon!

I have the following table:

How to get only the first record of each id_uni_sching? That is, I need to get the id_sching 534,382 and 524,075 (First time of each id_uni_sching) ?

Note: I am using Postgres

Answer 1

You can add using the array_agg function and access the index of the item you want, if you sort by some column with timestamp is perfect.

select id_uni_sching,
      (array_agg(id_sching order by sta_sching))[1] as primeiro_id_sching,
      (array_agg(dta_sching order by sta_sching))[1] as primeiro_dta_sching,
      (array_agg(sta_sching order by sta_sching))[1] as primeiro_sta_sching
from seguinte_tabela
group by id_uni_sching

You can do the opposite by sta_sching desc, Okay, I’m assuming that sta_sching is a time column, if you don’t have to go a little deeper by creating a subquery for row_number(). In this case you can assign a number to each Row, and sort by this number.

Answer 2

A solution using DISTINCT ON:

SELECT DISTINCT ON (id_uni_sching) id_uni_sching, id_sching, dta_sching, sta_sching
FROM tabela1
ORDER BY id_uni_sching, dta_sching, sta_sching;