I am trying to use datediff on a date/time column. The date/time column is in dd/mm/yyyy hh:mm:ss format and from what I have read online Datediff works with yyyy/mm/dd hh:mm:ss format.
I have tried to convert bit I am facing problems
DATEDIFF(minute, convert(varchar(c.duration, 20)), '1970/01/01 00:00:00') as "Resolution Target"
Any help would be appreciated
Depends on database you are using ,if you are using Oracle 'yyyy-mm-dd',if it is SQL Server 'yyyy/mm/dd' can be used as datediff function.
DATEDIFF(minute, convert(varchar(c.duration), 20) , '1970/01/01 00:00:00') as "Resolution Target"
Related
Not being that great at sql, i've reached my limit.
I have a date in the yyyy/mm/dd format and i need to get all records "from a week ago"
I think i need some conversion stuff to be done cause this
d.date_begin >= DATEADD(day,-7, GETDATE())
is not working :), i'm TERRIBLE at convert and data type..
This will work if you want records from 7 days ago up to and including today's records
CAST(d.date_begin AS DATE) >= CAST(DATEADD(day,-7, GETDATE()) AS DATE)
where DATEDIFF(month,Your_date_column,getdate()) < 3
SQL server 2012 onwards, if date_begin is of datatype date
where d.date_begin >= cast(DATEADD(day,-7, GETDATE()) as date)
This will get anything in the last 7 days, regardless of time
You should store date/time values using native formats. Ok, sometimes we cannot. But you can easily convert your values to the correct format:
where cast(replace(d.date_begin, '/', '') as date) >= DATEADD(day, -7, GETDATE())
I should note that SQL Server is pretty good about conversions, so your initial code should not generate any errors -- unless you have unusual internationalization settings.
Or, actually, a better way to do this is to convert the current value to a string:
where d.date_begin >= format(dateadd(day, -7, getdate()), 'yyyy/MM/dd')
This is better because it is "sargable", meaning that SQL Server can use an index on the column if available.
SELECT * FROM tbl_name
WHERE date >= curdate() - INTERVAL DAYOFWEEK(curdate())+6 DAY
AND date < curdate() - INTERVAL DAYOFWEEK(curdate())-1 DAY
Don't manipulate d.date_begin. Making calculation on a column while comparing can give bad performance. You should manipulate getdate() to get the same format as d.date_begin. In this case it works because the format is yyyy/MM/dd - the comparison will give the same result as if both columns were date columns.
WHERE
d.date_begin >= convert(char(10),DATEADD(day,-7, GETDATE()), 111)
I am trying to convert numeric column (with 13 digits) to DATE and all i can do is this so i can see the date as a string.
CONVERT(VARCHAR(10),(DATEADD(SECOND, Start_Date/1000 ,'1/1/1970')),104)
What can i do so this will end up like DATE so i can filter it later?
The database is MS SQL.
Thanks in advance!
This expression should convert it to a datetime:
DATEADD(SECOND, Start_Date/1000 , '1970-01-01)
If you want a date, just convert this to a date:
CAST(DATEADD(SECOND, Start_Date/1000 , '1970-01-01') as DATE)
Note: I changed the date format to the ISO standard YYYY-MM-DD format.
I uploaded a table using fastload and want to convert the event_time column to a timestamp or just get the dates since I don't need the time component.
The text format is 'MM/DD/YYYY HH:MM:SS AM/PM', for example, '05/24/2013 08:12:00 AM'
I have tried
CAST(event_time AS TIMESTAMP(0))
and
CAST(SUBSTRING (event_time,1,10) AS date)
but they don't work for me. The error I get is invalid timestamp/date.
As a bonus question,
I don't think there are any outliers in the data (i.e. every row is in format as described above), but if there were, how do I account for those errors? or will Teradata automatically pass on those rows?
The CASTs don't work because this is not the default timestamp/date format.
The easiest way utilizes the TO_DATE function which is supported since TD14:
TO_DATE(event_time, 'MM/DD/YYYY HH:MM:SS AM')
Before TD14 it's:
CAST(CAST(event_time AS TIMESTAMP FORMAT 'MM/DD/YYYYbHH:MI:SSbT') AS DATE)
Regarding outliers, they will result in an "invalid date/timestamp" error. When you cast them during FastLoad to a Timestamp those rows will fail to load and will be inserted into the ET-error table. But you can't use the newer CAST-syntax in FL, must be the old Teradata-style cast:
:event_time (TIMESTAMP, FORMAT 'MM/DD/YYYYbHH:MI:SSbT')
I tried to obtain date in this format:
'05-31-2014 01:20:25 AM'
I used below code:
Sql Fiddle here, but output date strangely changed to a different date: 30-26-2011 01:30:38 AM
select format(CAST('2011-11-26 01:30:38.000' AS datetime), 'mm-dd-yyyy hh:mm:ss tt')
Sql Fiddle here
When using the FORMAT() function, mm is minutes, MM is month, so change to:
SELECT FORMAT(CAST('2011-11-26 01:30:38.000' AS DATETIME), 'MM-dd-yyyy hh:mm:ss tt')
If there's already an appropriate format available via CONVERT(), that is preferable as it performs better than the FORMAT() function.
Per MSDN convert should properly parse ISO 8601 dates with timezone using 127 as the style parameter.
The optional time zone indicator, Z, is used to make it easier to map XML datetime values that have time zone information to SQL Server datetime values that have no time zone. Z is the indicator for time zone UTC-0. Other time zones are indicated with HH:MM offset in the + or - direction. For example: 2006-12-12T23:45:12-08:00.
All of the following are valid ISO 8601 dates but return Conversion failed when converting date and/or time from character string.
select convert(datetime, N'2014-02-07T13:51:00+07:00', 127)
select convert(datetime, N'2014-02-07T13:51:00+07', 127)
select convert(datetime, N'2006-12-12T23:45:12-08:00', 127)
Anyone have a solution or workaround for this issue?
Workaround?: Use datetimeoffset:
select convert(datetimeoffset, N'2014-02-07T13:51:00+07:00', 127) --<-- This one works...
select convert(datetimeoffset, N'2014-02-07T13:51:00+07', 127)
select convert(datetimeoffset, N'2006-12-12T23:45:12-08:00') --<-- and this one works...
use datetimeoffset or datetime2 instead of datetime