I'm storing GTFS feeds into a SQL database and some times are expected to be stored above the 24:00:00 cap on time values. For example, some trains run at 12:30AM, but are listed for the previous days service, and that running time is stored as 24:30 in the GTFS specifications.
What would be the best way of getting around this? Should I just store it as a string?
Suggest to use int for that... your value could be:
Sec + Min * 60 + Hour * 3600
For the 24:30:00, you will get 88200.
When loading your value from DB, you could reverse your value by simple math equation:
Hour = int(value / 3600)
Min = int(value % 3600 / 60)
Sec = value % 3600 % 1800
I'd store two fields:
departure_time timestamp with time zone,
service_date date
Departure time would be calculated like this:
=> select '2015-07-08'::timestamptz+'24:30'::interval;
2015-07-09 00:30:00+02
This way:
you have a normal moment of time field for sorting events;
you'd not loose service date information;
you'd be able to calculate back the original GTFS data if needed.
Related
I would like some help with an SSIS problem.
I have two columns, one with a date of when demand was open and another when the demand was responded to.
My date comes in this way:
DT_ANSWERED_DATE
DT_CREATED_DATE
2021-02-04 19:48:00.000
2021-02-04 19:44:00.000
I would like to subtract DT_ANSWERED_DATE MINUS DT_CREATED_DATE
but I would like the result would be a float number:
like in this case when a subtract in excel
I get the result:
DT_ANSWERED_DATE
DT_CREATED_DATE
DT_ANSWERED_DATE minus DT_CREATED_DATE
2021-02-04 19:48:00.000
2021-02-04 19:44:00.000
0,00277777777228039
I would like to do the same thing but in a derived column at SSIS (Microsoft Visual Studio)
Thanks for the response in advance
It looks like your granularity is in minutes. This should get you the decimal number you are looking for...
DATEDIFF("mi", DT_CREATED_DATE, DT_ANSWERED_DATE) / (60 * 24)
(60 min per hour * 24 hours in a day)
Microsoft documentation... https://learn.microsoft.com/en-us/sql/integration-services/expressions/datediff-ssis-expression?view=sql-server-ver16
In your example above this results in:
4 min / (60*24) = 0.00277777777
Note:
I highly recommend using decimal vs float. Unless you really, really have a reason. 1=1 is usually not true when using a float number. It will always be true with integers or decimals.
I was wondering how to properly extract amount of hours between given 2 timestamps objects.
For instance, when the following SQL query gets executed:
select x, extract(HOUR FROM x) as result
from
(select (TIMESTAMP'2021-01-22T05:00:00' - TIMESTAMP'2021-01-01T09:00:00') as x)
The result value is 20, while I'd expect it to be 500.
It seems odd to me considering that x value indicates the expected return value.
Can anyone please explain to me what I'm doing wrong and perhaps suggest additional way of query so the desired result would return?
Thanks in advance!
I think you have to do the maths with this one as datediff in SparkSQL only supports days. This worked for me:
SELECT (unix_timestamp(to_timestamp('2021-01-22T05:00:00') ) - unix_timestamp(to_timestamp('2021-01-01T09:00:00'))) / 60 / 60 diffInHours
My results (in Synapse Notebook, not Databricks but I expect it to be the same):
The unix_timestamp function converts the timestamp to a Unix timestamp (in seconds) and then you can apply date math to it. Subtracting them gives the number of seconds between the two timestamps. Divide by 60 for the number minutes between the two dates and by 60 again for the number of hours between the two dates.
I have a table containing two time columns like this:
Time1 Time2
07:34:33 08:22:44
I want to add the time in both these columns and put the result of addition into a third column may be Time3
Any help would be appreciated..Thanks
If the value you expect as the result is 15:57:17 then you can get it by calculating for instance the number of seconds from midnight for Time1 and add that value to Time2:
select dateadd(second,datediff(second,0,time1),time2) as Time3
from your_table
I'm not sure how meaningful adding two discrete time values together is though, unless they are meant to represent duration in which case the time datatype might not be the best as it is meant for time of day data and only has a range of 00:00:00.0000000 through 23:59:59.9999999 and an addition could overflow (and hence wrap around).
If the result you want isn't 15:57:17 then you should clarify the question and add the desired output.
The engine doesn't understand addition of two time values, because it thinks you can't add two times of day. You get:
Msg 8117, Level 16, State 1, Line 8
Operand data type time is invalid for add operator.
If these are elapsed times, not times of day, you could take them apart with DATEPART, but in SQL Server 2008 you will have to use a CONVERT to put the value back together, plus have all the gymnastics to do base 60 addition.
If you have the option, it would be best to store the time values as NUMERIC with a positive scale, where the unit of measure is hours, and then break them down when finally reporting them. Something like this:
DECLARE
#r NUMERIC(7, 5);
SET #r = 8.856;
SELECT FLOOR(#r) AS Hours, FLOOR(60 * (#r - FLOOR(#r))) AS Minutes, 60 * ((60 * #r) - FLOOR(60 * #r)) AS Seconds
Returns
Hours Minutes Seconds
8 51 21.60000
There is an advantage to writing a user-defined function to do this, to eliminate the repeated 60 * #r calculations.
This is the CREATED_TIME 2012-07-17 00:00:22 and this is the Corresponding Timestamp 1342508427000. Here timestamp is 5 seconds more than the CREATED_TIME. I need to see below scenario
Currently I have a query, in which I am joining on created_time and timestamp like this-
ON (UNIX_TIMESTAMP(testingtable1.created_time) = (prod_and_ts.timestamps / 1000))
So in above case, it will not match as timestamp is 5 seconds more than created_time. But I need if the difference between either of the two is within 15 minutes then I need to match it.
So I wrote the below JOIN query- I am not sure whether this is the right way to do it or not?
ON ((UNIX_TIMESTAMP(testingtable1.created_time) - (prod_and_ts.timestamps / 1000)) / 60* 1000 <= 15)
How I can do the above case if difference between timestamps is within 15 minutes then data will get matched by the above ON clause
I'd prefer the (specifically designed for this purpose!) date and time functions instead of doing al these kinds of calculations with timestamps. You wouldn't believe how much trouble this can cause. Make sure you read and understand this and this
i am working for telecom,i need to calculate call duration of a call.the issue is we have world time 24hrs(00:00:00 - 23:59:59)for example, now the call starts at 22:31:40 and ends at 00:22:56
now i need to calculate the duration of this call. can i know the logic to calculate it
Thanks In advance
There is no DB2 built-in function to do this. However, the DB2 DATE documentation includes the source for a user-defined function which employs DAYS() and MIDNIGHT_SECONDS() to achieve the required result.
CREATE FUNCTION secondsdiff(t1 TIMESTAMP, t2 TIMESTAMP)
RETURNS INT
RETURN (
(DAYS(t1) - DAYS(t2)) * 86400 +
(MIDNIGHT_SECONDS(t1) - MIDNIGHT_SECONDS(t2))
)
#
Add 24 hours to the second time and subtract from the first. If the result is greater than 24 hours, take the 24 back out. This obviously doesn't handle calls that ARE over 24 hours, but we'll assume they arent.
... or use some DB2 function I'm unaware of.