THanks James for confirming. This helps a lot. I can move to 1.21.1. But
before that I need to try the same with the parquet also. Will check that
too and make a decision.
Regards
Prabhakar
On Tue, Aug 8, 2023 at 12:51 PM James Turton wrote:
> Okay I get the same result as you when I try with
Okay I get the same result as you when I try with Drill 1.20.3 so I
guess that there's a JSON reader bug that got fixed somewhere in between
1.20.3 and 1.21.1. Do you need to stay on Drill 1.20 or can you upgrade
to 1.21?
On 2023/08/07 08:32, Prabhakar Bhosale wrote:
hi James,
I re-tried the
hi James,
I re-tried the steps once again after looking at the output you shared with
me. But still I am not able to get the expected output on drill 1.20.1.
Then I downloaded drill 1.21.1 and tried the same steps. This time I got
the expected output.
So can you please try the same on drill 1.20.1
Hi!
I just got back from travelling. I ran a test and Drill did do what I
believe you're after.
Here's the test JSON file
|➜ ~ cat /tmp/foo/bar.json|||
|||{"id":"T06125309","cust_id":"A20","num":"VAB6169028"}|||
||
And here's my Drill session.
|apache drill> use dfs.tmp;|||
Hi James,
Any advice on the problem on schema for json file as mentioned in
my previous mail?
Regards
Prabhakar
On Mon, Jul 24, 2023 at 3:41 PM Prabhakar Bhosale
wrote:
> Hi James,
>
> I tried the same on the JSON file and it is still not working.
>
> Below is JSON file content
>
>
Hi Prabhakar
From what I recall, Drill won't consider a provided schema when
querying Parquet because Parquet files bundle their own schema. You
might need to use a SQL function like COALESCE(TRAN_AMOUNT, 1.11) and
possibly put that in a SQL view for reuse.
Regards
James
On 2023/07/11
