read csv file - convert string to datetime

Community Support
1

Hi all!

 

In my csv file date is stored as 20130315185540686 (meaning 2013.03.15T18:55:40.686).

 

Is there a possibility to convert that string into datetime “on the run” during the reading process? Or perhaps split it into two separate date and time columns?

 

I am using following string to read csv file:

.Q.fs[{:tst/orderlog/ upsert .Q.en[:tst] flip symbolsystypmomentidactionpricevolumeid_dealprice_dealownaccount!(“SCCJJIFIJFIS”;“,”)0:x}]`:3.csv

 

 

Thank you.

2

update date:“D”$8#'date,time:“T”$-9#'date from (date:10#enlist “20130315185540686”)

so embed it like

.Q.fs[{:tst/orderlog/ upsert .Q.en[:tst] update “D”$8#'date,time:“T”$-9#'date from flipsymbolsystypmomentidactionpricevolumeid_dealprice_dealownaccount!(“SCCJJIFIJFIS”;“,”)0:x}]`:3.csv

I dont know which of your col names

symbolsystypmomentidactionpricevolumeid_dealprice_dealownaccount

refers to date[time] though… so I just called it date; change it as necessary.

3

btw, for this use type id “*” to read the date[time] column from the csv

4

Thank you,

 

seems I can make it work after some effort

 

5

Or you can do:

update “Z”$((8#‘date),’“T”,/:(-9#'date)) from (date:10#enlist “20130315185540686”)

 

.Q.fs[{:tst/orderlog/ upsert .Q.en[:tst] update “Z”$((8#‘date),’“T”,/:(-9#'date)) from flipsymbolsystypmomentidactionpricevolumeid_dealprice_dealownaccount!(“SCCJJIFIJFIS”;“,”)0:x}]`:3.csv

 

6

I have made it almost same

 

update (“D”$8#'moment)+(“T”$-9#'moment)

Is it a good approach?

7

update sum(“DT”;8 9)0:date from 
is a bit faster (no each)  and shorter

Cheers,

  Attila

8

I think this is fine, as long as the type of the result is what you need (well you result will end up to be time stamp type)…

 

But you can easily recast it anyway… So you’re good!! :)