You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Beach Name,Measurement Timestamp,Water Temperature,Turbidity,Transducer Depth,Wave Height,Wave Period,Battery Life,Measurement Timestamp Label,Measurement ID
Calumet Beach,04/07/2018 12:30:00 AM,21.7,3.15,,0.389,3,0,04/07/2018 12:30 AM,CalumetBeach201804072430
Ohio Street Beach,06/14/2019 07:00:00 PM,16,2.18,,0.137,3,11.8,06/14/2019 7:00 PM,OhioStreetBeach201906141900
Ohio Street Beach,05/16/2018 01:00:00 PM,21.7,1.92,,0.023,1,10.3,05/16/2018 1:00 PM,OhioStreetBeach201805161300
Ohio Street Beach,05/16/2018 02:00:00 PM,21.9,1.81,,0.026,6,10.3,05/16/2018 2:00 PM,OhioStreetBeach201805161400
Ohio Street Beach,06/04/2018 09:00:00 AM,23,0.03,,0.021,8,12.4,06/04/2018 9:00 AM,OhioStreetBeach201806040900
Ohio Street Beach,06/04/2018 10:00:00 AM,17.6,3.09,,0.159,4,12.6,06/04/2018 10:00 AM,OhioStreetBeach201806041000
I know the date is CST, because this is from Chicago, but timezone is not present on the dataset. Right now I have to do a transformation (my solution is importing into a temp table and then doing something like INSERT INTO table SELECT to_timestamp(CONCAT(MeasurementTimestamp,' CST'), 'MM/dd/yyyy kk:mm:ss a z') as MeasurementTimestamp,.
Describe the solution you'd like.
I'd rather pass a format for the timestamp when importing the CSV. I am not sure if we want to do this as an extra parameter on import, or if we want to make this generic into questdb, so we can also use it at any moment with functions like to_timestamp. In that case we could have something like 'MM/dd/yyyy kk:mm:ss a !CST'), where the ! would indicate we are forcing the timezone and the date should be interpreted as being natively on that one.
The import param is probably easier, but the second option is probably more versatile.
Describe alternatives you've considered.
No response
Full Name:
Javier Ramirez
Affiliation:
QuestDB
Additional context
No response
The text was updated successfully, but these errors were encountered:
something like !CST looks ok first. but if you think about it:
it makes ! a special character - it means compatibility issues.
the same patterns is used for formatting. one has to guess how a formatted is going to act - should it force formatting in the given timezone? probably yes. should it print the timezone too? probably now. but I am less sure.
while you can specify how it should behave this specification will be specific to QuestDB and our parsers/formatters will move further from what other systems do. (or are aware of any system supporting this?)
for the reasons above I believe the simple option with forcing a specific timezone via an import config is probably wiser.
That makes sense. To the best of my knowledge, no other database allows to force a timezone when not present, so I was imagining if we could have a way of making that part of our parser adding a new token that would force the timezone. But I was probably overthinking XD
Is your feature request related to a problem?
If you have a CSV with timestamps (specially for designated, but not only) QuestDB should allow to force a timezone override, so UTC is not assumed.
For example, imagine a CSV (original at https://data.cityofchicago.org/api/views/qmqz-2xku/rows.csv) with rows like these:
I know the date is CST, because this is from Chicago, but timezone is not present on the dataset. Right now I have to do a transformation (my solution is importing into a temp table and then doing something like
INSERT INTO table SELECT to_timestamp(CONCAT(MeasurementTimestamp,' CST'), 'MM/dd/yyyy kk:mm:ss a z') as MeasurementTimestamp,
.Describe the solution you'd like.
I'd rather pass a format for the timestamp when importing the CSV. I am not sure if we want to do this as an extra parameter on import, or if we want to make this generic into questdb, so we can also use it at any moment with functions like
to_timestamp
. In that case we could have something like 'MM/dd/yyyy kk:mm:ss a !CST'), where the!
would indicate we are forcing the timezone and the date should be interpreted as being natively on that one.The import param is probably easier, but the second option is probably more versatile.
Describe alternatives you've considered.
No response
Full Name:
Javier Ramirez
Affiliation:
QuestDB
Additional context
No response
The text was updated successfully, but these errors were encountered: