I have JSON data as below: I need to convert that date or mongo_date
into utc timestamp, to analyse the data in hive as per timeline example per year, per month, per week using map reduce
{
"_id" : ObjectId("51ac77050e9edcdad271ce2d"),
"company" : null,
"date" : "19760224",
"mongo_date" : ISODate("1976-02-24T00:00:00Z")
Hive understands this format: 'yyyy-MM-dd HH:mm:ss.SSS'
.
Use unix_timestamp()
to convert to seconds passed from 1970-01-01, then use from_unixtime()
to convert to proper format:
select from_unixtime(UNIX_TIMESTAMP("2017-01-01T05:01:10Z", "yyyy-MM-dd'T'HH:mm:ss'Z'"),"yyyy-MM-dd HH:mm:ss");
Result:
2017-01-01 05:01:10
Update. This method is to remove Z
and replace T
with space using regexp_replace
and convert to timestamp if necessary, without using unix_timestamp()
, this will preserve milliseconds:
select timestamp(regexp_replace("2019-05-17T17:03:09.775Z", '^(.+?)T(.+?)Z$','$1 $2'));
Result:
2019-05-17 17:03:09.775