Here is the syntax for I could figure out for timezone.
select to_timestamp('2015-03-30 20:49:59.0 UTC', '-MM-dd HH:mm:ss.s z') as
Original, to_char(to_timestamp('2015-03-30 20:49:59.0 UTC', '-MM-dd
HH:mm:ss.s z'), 'z') as New_TZ from sys.version;
+++
| Original
Currently in Drill there isn't support for 'timestamp with time zone'
data type, all the timestamp/date information is stored in UTC and no
timezone information is maintained, so AFAIK there isn't really a way to
convert dates/timestamp to a specific timezone. However if your input
data contain
How big is the data set? Is it just a single file, or multiple? And how
long is it taking?
It does not have to read the entire file to produce output. I suspect it's
taking a long time to set up the query.
On Mon, Mar 30, 2015 at 3:39 PM, Ganesha Muthuraman
wrote:
> I see that for a large data
I see that for a large data set ( in my case a large CSV ), even a LIMIT 5 type
of SQL take a very long time. Is it that Drill has to read through the entire
file to produce any output even if I just want, say the first 10 lines of the
file?
Is there a way to let Drill stream the results back a
I have a column that’s UTC, how can I convert this column to a specific
timezone?
> select to_timestamp(columns[13], '-MM-dd HH:mm:ss') as `datetime_utc`
> from `weather/2012` limit 10;
+--+
| datetime_utc |
+--+
| 2012-02-01 05:54:00.0 |
| 2012-02-01 06:54:00.0 |
| 2
Thanks - I knew that - I was trying to understand better what was going on
under the hood. Kristine's link got me what I needed.
On Mon, Mar 30, 2015 at 1:48 PM, Sudhakar Thota wrote:
> Vince,
>
> Can you try doing this again and query one more time.
>
> alter system set `store.json.all_text_mod
Vince,
Can you try doing this again and query one more time.
alter system set `store.json.all_text_mode` = true
I used your file on sandbox and queried. It is working.
0: jdbc:drill:> select * from sys.options where name like '%all_text%';
+++++--
Great doc Kristine, thanks!
On Mon, Mar 30, 2015 at 11:06 AM, Kristine Hahn wrote:
> When you set all_text_mode to true, Drill reads all data from the JSON
> files as VARCHAR. After reading the data, use a SELECT statement in Drill
> to cast data as follows:
>
> * Cast JSON numeric values to [SQ
When you set all_text_mode to true, Drill reads all data from the JSON
files as VARCHAR. After reading the data, use a SELECT statement in Drill
to cast data as follows:
* Cast JSON numeric values to [SQL types](/docs/data-types), such as
BIGINT, DECIMAL, FLOAT, and INTEGER.
* Cast JSON strings to
My guess is that setting store.json.all_text_mode = True effectively tells
Drill not to skip inferrering the type of the columns, which perhaps
avoids any complications due to a change in schema that confuses things (in
this case from Float8 to null)? But I am not confident I understand what's
rea
10 matches
Mail list logo
