The date column that I have is in varchar2 and I want to convert those values in YYYY-MM-DD
DATE
7/26/2013
7/29/2013
8/1/2013
8/4/2013
7/28/2013
7/31/2013
8/3/2013
7/30/2013
8/5/2013
7/25/2013
8/2/2013
8/6/2013
7/27/2013
You should never store dates in a VARCHAR column
So in order to display it differently now, you need to first convert the string to a date and then back to a string
If you are certain that all "dates" do have the right (and same) format, then the following should work:
select to_char(to_date(date, 'mm/dd/yyyy'), 'yyyy-mm-dd')
from the_table;
But I wouldn't be surprised if that gives you an error because one or more rows have a date which is formatted differently. Something which could not have happened had you defined the column to be of type DATE right from the beginning.
You should really, really consider changing that column to be a real DATE column.
Btw: DATE is a horrible name for such a column. It is a reserved word and it simply doesn't tell you anything about what is actually stored in that column (a "start date", an "end date", a "birth date", a "due date", ...)
You have to convert your string to date and than convert to char with expected format
select TO_CHAR(TO_DATE(datecol, 'MM/DD/YYYY'),'YYYY-MM-DD')
from your_table
Sql Fiddle Demo
you can use REGEXP_REPLACE function to manipulate strings or to_date and to_char to convert string to date and than to string back with different format.
select to_char(TO_DATE("Date", 'MM/DD/YYYY'),'yyyy-mm-dd')
from Table1
SQL FIDDLE
Related
Based on the example mentioned here in the Snowflake documentation, why are the date and timestamp values returning different values just by changing the ORDER BY clause? Also, I am trying to convert a string to a date format which is not returning correct results in Snowflake while this works fine in other SQL based Engines. Need help from experts on this.
This query
SELECT '20200710', TO_DATE('20200710');
is returning the following output
20200710 | 1970-08-22
Also tried:
SELECT TO_DATE('20200710', 'YYYY-MM-DD');
and got the error:
Can't parse '20200710' as date with format 'YYYY-MM-DD'
To convert to a date data type, you would use:
SELECT TO_DATE('20200710', 'YYYYMMDD')
I would recommend just keeping the date data type. But if you want a string in the format YYYY-MM-DD:
SELECT TO_CHAR(TO_DATE('20200710', 'YYYYMMDD'), 'YYYY-MM-DD')
I have a BIRT report where user will be entering the dates in dd-mm-yyyy format however I need to convert dd-mm-yyyy to MON-YYYY format.
I have tried to use VARCHAR_FORMAT(FIELDNAME,'MON-YYYY') however it doesn't work.
select …….
where VARCHAR_FORMAT(fieldname,'MON-YYYY') = '2017-05-15';
User would end the date as
15/05/2017
The value present in the database for this field is 2017-05-15 07:30:00.0
update
Apparently the column is not a string but a datetime which means the conversion is only
to_date(fieldname, 'MON-YYYY')
But if the column is used in a Where clause it shouldn’t be converted at all.
——
Use to_date and to_char to first convert your string to a date and then back to a string with the right format
to_char(to_date(fieldname, 'DD-MM-YYYY'), 'MON-YYYY')
select *
from table (values
timestamp('2017-05-15-07.30.00')
) t(fieldname)
where
fieldname between to_date('15/05/2017', 'DD/MM/YYYY') and to_date('15/05/2017', 'DD/MM/YYYY') + 1 day
--date(fieldname) = to_date('15/05/2017', 'DD/MM/YYYY')
;
You may run it as is.
Both cases work, but to use the 2-nd one efficiently, you must create an index by the date(fieldname) expression (since db2 10.5) or add generated always column to the table with the same expression and index on it.
In my DB, there is a date field in the format yyyymmdd.
I have to get all the dates in the format dd-mm-yyyy for that particlar date.
ex:
Date
20170130
20170228
20170325
for the above dates, I need the output in the below format with the dates and day of the particular dates
date day
30-01-2017 tuesday
28-02-2017 tuesday
25-03-2017 saturday
If the column is a string, then it can hold invalid date values such as February 31, one way to avoid this is by a small function such as this:
create or replace
function my_to_date( p_str in varchar2 ) return date
is
begin
return to_date( p_str );
exception
when others then
return null;
end;
\\
select to_char(my_to_date('20170231'),'DD-MM-YYYY Day')
from dual
\\
Demo
Try below:
Select to_char(yrdate, 'dd-mm-yyyy'), to_char(yrdate, 'D') from yrtable
It sounds like your dates aren't actually DATE fields but some kind of CHAR field? The best option would be to convert to DATE and then convert back to CHAR:
SELECT TO_CHAR(TO_DATE(mydate, 'YYYYMMDD'), 'DD-MM-YYYY Day')
FROM mytable;
This uses the YYYYMMDD mask to convert your string into a date, then uses the mask DD-MM-YYYY Day to convert it back into a string. Use day if you want the day name in lowercase (as in your OP).
#user2778168 answer will give you the results you want. But why?
Your database does not have dates stored in yyyymmdd format or any other date format for at mater, unless it's defined with a character type definition. Oracle stores all dates in a single internal structure, and with only slight variations timestamps are the same. The format used only tells Oracle how to display the value or to convert a string to a date. Unless a specific format is specified Oracle uses the NLS_DATE_FORMAT for this determination. See here and scan down to "Datetime Format Models" for format specifications.
To see this run the following:
select value
from nls_session_parameters
where parameter = 'NLS_DATE_FORMAT';
Select yrdate default_format
, to_char(yrdate, 'dd-mm-yyyy') specified_format
, dump(yrdate) unformated
from yrtable;
alter session set nls_date_format = 'Month dd,yyyy';
Rerun the above queries.
It seems you hold date column(date1) in character format. Suppose you have a table named days:
SQL> desc days
date1 varchar2(10)
then,
we should convert it into date, and then into char format, with aliases in quotation marks to provide lowercase aliases as you wanted.
perhaps your database's date language is non-english like mine(turkish), then you need to convert it to english.
lastly, it'a appropriate to order the results according to converted date, seen from your output. So we can use the following SQL :
select to_char(to_date(date1,'yyyymmdd'),'dd-mm-yyyy') "date",
to_char(to_date(date1,'yyyymmdd'),'day','nls_date_language=english') "day"
from days
order by to_date(date1,'yyyymmdd');
D e m o
I have a character field that stamps in the order of MMDDYYHHMMSS (note: not a date but character field). I am wanting to kick this out to a date field in my SQL into this format dd.mm.yyyy. hh24:mi.
My problem is that the sql kicks it out to YYYY-MM-DD field without the time. This section of the sql looks like this:
TO_DATE(SUBSTR(MOPACTIVITY.MOPID,3,2)||'.'||SUBSTR(MOPACTIVITY.MOPID,1,2)
||'.'||'20'||SUBSTR(MOPACTIVITY.MOPID,5,2)||'.'||SUBSTR(MOPACTIVITY.MOPID,7,2)
||':'||SUBSTR(MOPACTIVITY.MOPID,9,2)||':'||SUBSTR(MOPACTIVITY.MOPID,11,2)
, 'dd.mm.yyyy. hh24:mi:ss') "XXX",
Any thoughs on how to get the time to convert too?
No need for such a complicated expression:
to_date(MOPID, 'MMDDYYHH24MISS')
will convert the column to a real DATE column assuming the time part is in 24 hour format (00-23, not 00-12). And this will also fail if you don't really have valid dates in the varchar column.
this out to a date field in my SQL into this format
A DATE column does not have "a format"!
The format is only applied when you display it.
In case you mean you want to convert the varchar stored in your column into another varchar that has a different date formatting, the easiest is probably to simply convert the above expression back to a varchar:
to_char(to_date(MOPID, 'MMDDYYHH24MISS'), 'dd.mm.yyyy. hh24:mi')
Before applying something like that, allow me one comment:
Store dates in DATE columns, never ever store them in a VARCHAR column.
If you had done that from the beginning, all you would have to do know is to simply apply a single to_char() to your DATE column to get the display format you want.
I have Following dummy table with data:
ACID srno date(mm/dd/yyyy) name
3 1 04/12/2010 mahesh
3 2 04/12/2010 mahendra
Now if I try with Following SQL Transact:
select srno from dummy
where name = 'mahesh'
and date= convert(datetime,'12/04/2010',101) –- I have date in dd/MM/yyyy Format
and ACID=3
It’s Not returning the srno of the table. That means Date is not execute convert statement as above
What’s the reason?
Try using style 103 instead of 101.
select srno from dummy
where name = 'mahesh'
and date= convert(datetime,'12/04/2010',103) –- I have date in dd/MM/yyyy Format
and ACID=3
If you convert 12/04/2010 using format 101, you get date "December 4, 2010", which is not in your database. Use format 103 to convert a date in format dd/mm/yyyy to DateTime.
The database stores dates using the DateTime type which is format-agnostic. It does have a default format for string conversions, which seems to be mm/dd/yyyy (101) on your database.
However, when you convert a string to add it to your table, you want to specify the format of your input string, in your example dd/mm/yyyy (103).
Take a look at the MSDN article for CAST and CONVERT which details all format styles that you can use with dates.
To be honest, if you want to specify a DATE LITERAL in SQL Server, please stick with the simplest YYYYMMDD format, e.g.
and dummy.date = '20100412'
It is robust and works for all regional, user language and dateformat settings. This assumes the other side of the comparison is already a date column. Even if you had to CAST it, using this format you don't need to specify a format
and dummy.date = cast('20100412' as datetime)