So I have a table in Snowflake that I uploaded and made the mistake of uploading some data with the incorrect datetime. The first few columns of the table are different json results, that correspond to the date they were fetched.
Example:
name date 0 [{},{},{},{},{}] 8/20/2019 1 [{}] 12/22/2019 2 [{},{},{},{}] 11/15/2019 3 [{},{},{}] 1/10/2019 4 [{},{},{},{}] 12/1/2019
The INSERT was pretty straight forward, I staged the file and inserted it into an already created table.
I want to go in and change the date by two days, and will make a note to change how I am generating the date and data, so I don’t have to do this manually anymore.
Is is possible to correct the column for a certain set of ids instead of a date range like I did below?
ALTER TABLE json_date ALTER COLUMN date Where date >"12-01-2019" and date < "12-30-2019" dateadd(day,2,date);
Advertisement
Answer
If this is simply a column, then you would use update
:
update json_date set date = dateadd(day, 2, date) where date > '2019-12-01';