How can I remove milliseconds from Timestamp in Oracle?

Question

I want to export data from 'My_Customer' table in .csv format.
I am extracting a time field but I am getting the output along with Milliseconds.
Eg:
The output I am getting,
25-10-2017 12:53:12:00076
I want to remove/truncate the milliseconds.

Expected Output-
25-10-2017 12:53:12.

I have tried to_char() but it isn't giving proper output. (Instead of giving the exact time available in that field, it is rounding off).

Solution 1

Convert your timestamp to a character string:

SQL> SELECT TO_CHAR (SYSTIMESTAMP, 'YYYY-MON-DD HH24:MI:SS') AS my_date
  FROM DUAL;

MY_DATE
-----------------------------
2017-OCT-13 00:38:26

Solution 2

You can either cast it to a timestamp with no fractional seconds (this will round to the nearest second):

CAST( your_timestamp AS TIMESTAMP(0) )

Or to a DATE data type (this will truncate to the nearest second):

CAST( your_timestamp AS DATE )

Or you can convert it to a formatted string and specify the format model you want to use (this will truncate to the nearest second):

TO_CHAR( your_timestamp, 'YYYY-MM-DD HH24:MI:SS' )

Like this:

SQL Fiddle

Oracle 11g R2 Schema Setup:

CREATE TABLE your_table ( your_timestamp ) AS
  SELECT TIMESTAMP '2017-10-25 12:53:12.10076' FROM DUAL;

Query 1:

SELECT CAST( your_timestamp AS TIMESTAMP(0) ) AS "Timestamp",
       CAST( your_timestamp AS DATE ) AS "Date",
       TO_CHAR( your_timestamp, 'DD-MM-YYYY HH24:MI:SS' ) AS "String"
FROM   your_table

Results:

|             Timestamp |                 Date |              String |
|-----------------------|----------------------|---------------------|
| 2017-10-25 12:53:12.0 | 2017-10-25T12:53:12Z | 25-10-2017 12:53:12 |

How the TIMESTAMP and DATE are formatted will depend on your NLS_TIMESTAMP_FORMAT and NLS_DATE_FORMAT session parameters but you can directly control the formatting of TO_CHAR when you specify a format model.

Comments

[Cyrille MODIANO]

Why don't you share the result of what you tried and what didn't work? TO_CHAR() should work

[XING]

Such questions was already answered very well and iterating same thing doesnot make sense. Feel free to mark it duplicate.

[XING]

Such questions was already answered very well and iterating same thing doesnot make sense. Feel free to mark it duplicate.