Change Date Format from an array in SQL SELECT Statement - sql

I have a column updated_at that returns an array
["2019-01-05T17:28:32.506-05:00","2019-06-15T13:22:02.625-04:00"]
But I want the output date format like this 2019-01-03.
How can I accomplish this in sql databricks?
Thanks!


Try unnest and cast that as a date:
with ts_array as
(select array['2019-01-05T17:28:32.506-05:00','2019-06-15T13:22:02.625-04:00'] as tsa)
select unnest(tsa)::date from ts_array ;


You can use "date_trunc" SQL function to get the output in date format.
date_trunc(fmt, ts) - Returns timestamp ts truncated to the unit specified by the format model fmt. fmt should be one of [“YEAR”, “YYYY”, “YY”, “MON”, “MONTH”, “MM”, “DAY”, “DD”, “HOUR”, “MINUTE”, “SECOND”, “WEEK”, “QUARTER”]
Examples:
> SELECT date_trunc('YEAR', '2015-03-05T09:32:05.359');
2015-01-01 00:00:00
> SELECT date_trunc('MM', '2015-03-05T09:32:05.359');
2015-03-01 00:00:00
> SELECT date_trunc('DD', '2015-03-05T09:32:05.359');
2015-03-05 00:00:00
> SELECT date_trunc('HOUR', '2015-03-05T09:32:05.359');
2015-03-05 09:00:00
Reference: Databricks - SQL Functions.
Hope this helps.

Related

BigQuery doesn't convert some string values to date

My table has its dates in string format. I'm trying to transform them to date, but BQ is not accepting some values and I can't find the reason.
What I've tried:
PARSE_TIMESTAMP('%Y%m%d', CLIENTE.dtnasc)
PARSE_TIMESTAMP('%Y%m%d', cast( CLIENTE.dtnasc as string))
CAST(CLIENTE.dtnasc AS DATE)
Error example (it's not only this value):
Failed to parse input string "1991-02-11 00:00:00"
How the date is without transfomartion:
2000-01-01 00:00:00 (string)

Try this:
SELECT PARSE_TIMESTAMP('%Y-%m-%d %H:%M:%S', '1991-02-11 00:00:00')
SELECT PARSE_TIMESTAMP('%Y-%m-%d %H:%M:%S', CLIENTE.dtnasc)
FROM mytable

Consider below option (the simplest I can think of, having that your string column matches already format of timestamp - 2000-01-01 00:00:00 (string))
SELECT DATE(TIMESTAMP(CLIENTE.dtnasc))

convert date unix to timestamp

I want to convert type date unix to timestamp in Informix.
My column date1 contains values as 1598915961, 1598911249, 1598911255...
expected output: 2020-02-13 15:00:00
How should I do it, please?

In Informix, you can use dbinfo() and 'utc_to_datetime':
select dbinfo('utc_to_datetime', myepoch)

The idea is that you can add the seconds to the date '1970-01-01'. I don't have Informix on-hand, but the syntax is something like this:
select datetime('1970-01-01') + interval date1 second

Date format with UTC

I have to select the value 2019-03-25 from a date column of a table but in the following format:
2019-03-25T00:00:00.000+02:00
Hon can I get it?
Oracle 10g
Thanks!

The date datatype does not store milliseconds and timezone information, so I undertand your question as how to format a date to the target forma, with fixed values for milliseconds and timezone.
If so, you can use to_char() like so:
to_char(mycol, 'yyyy-mm-dd"T"hh24:mi:ss".000+2:00"')

You can CAST your DATE to a TIMESTAMP and then use FROM_TZ to set the time zone and then format it to your requirements using TO_CHAR:
SELECT TO_CHAR(
FROM_TZ( CAST( your_date AS TIMESTAMP ), '+02:00' ),
'YYYY-MM-DD"T"HH24:MI:SS.FF3TZH:TZM'
) AS formatted_date
FROM your_table;
Which, for your sample data:
CREATE TABLE your_table ( your_date ) AS
SELECT DATE '2019-03-25' FROM DUAL
Outputs:
| FORMATTED_DATE |
| :---------------------------- |
| 2019-03-25T00:00:00.000+02:00 |
db<>fiddle here

We can not store timezone information in db that's why we can simply use below format,
select to_char(sysdate,'YYYY-MM-DD')||'T'||to_char(sysdate,'HH24:MI:SS') from dual

AWS Athena (Presto) - how to format Timestamp to Date Format?

I have a column in Athena with Timestamp Data Type and format is: 2019-08-28 00:00:00.000
How to format it to Date format using SQL to be:
DD-MON-YYYY
Thanks.

WITH test AS (
SELECT '2019-08-28 00:00:00.000' AS str
)
SELECT format_datetime(cast(str AS timestamp), 'dd-MM-YYYY')
FROM test
Result:
_col0
1 28-08-2019

PostgreSQL: truncate hour/min/second from a timestamp

I am using the following query to change all date to the Monday of the corresponding week:
select date_trunc('week', join_date) as join_wk from my_table
This query converts 2017-08-23 11:30:02 to 2017-08-21 00:00:00
I am wondering if it is possible to remove the hour/min/secondfrom the output 2017-08-21 00:00:00? i.e. make the output in the format of 2017-08-21

date_trunc returns a timestamp. You could cast it to a date to lose the time part of it:
SELECT DATE_TRUNC('week', join_date)::DATE AS join_wk FROM my_table
-- Here ----------------------------^