Hi Dobes,
Looking at the BSON spec [1], it seems that BSON works like JSON: a null isn't
a null of some type, it is just null. So, if Drill sees, say, 10 nulls, and has
to create a vector to store them, it doesn't know which type to use. In fact,
if BSON worked any other way, there could not be
Charles,
Yeah, makes sense to me, although I guess sorting would have to be applied as
well in that case. Does it push down sorting to the storage plugin?
On 2/25/2020 8:21:42 PM, Charles Givre wrote:
Hey Paul, Dobes,
I was thinking about this, and it would seem that a LIMIT pushdown would be
Hi Paul,
A simple filter I tried was: WHERE createdAt > TIMESTAMP "2020-02-25"
This wasn't pushed down.
I think I recall doing another query where it did send a filter to MongoDB so I
was curious what I could expect to be applied at the mongodb level and what
would not.
Would drill be able to
Hi Paul,
It seems to me that the type of the columns in a JSON file is "JSON" - e.g.
map, array, number, string, null, or boolean. In mongodb it is "BSON", which
adds dates, integers, and a few other things.
Lacking further guidance from the user, I would expect drill to handle all JSON
& BSO
Hi Prabhakar,
One more thought if you can't upgrade your app server to Java 8, and if
back-porting Drill to Java 7 is not practical. As it turns out, all versions of
Java are compatible with your network. So, perhaps you can use a network
connection to bridge the two.
If your queries are of mo
Hey Paul, Dobes,
I was thinking about this, and it would seem that a LIMIT pushdown would be an
obvious candidate for most plugins as well. I don't think that is in the Mongo
plugin, but that would be a good addition.
--C
> On Feb 25, 2020, at 11:19 PM, Paul Rogers wrote:
>
> d
Hi Dobes,
Your use case is exactly the one we hope Drill can serve: integrate data from
multiple sources. We may have to work on Drill a bit to get it there, however.
A quick check of Mongo shows that it does implement filter push down. Check out
the class MongoPushDownFilterForScan. The detail
Hi Dobes,
You've run into the classic drawback of runtime schema inference: if Drill
never sees a column value in its first sample, then it has no way to "predict
the future" and guess what type will eventually show up. So, Drill guesses
"nullable INT" which turns out to almost always be wrong.
Hi,
I am trying to understand drill's performance how we can best use it for our
project. We use mongo as our primary "live" database and I am looking at
syncing data to Amazon S3 and using Drill to run reports off of that.
I was hoping that I could have Drill connect directly to mongo for som
Hi,
I was experimenting with the mongo storage system and I found that when I query
a field that doesn't usually have any value, I get this error "You tried to
write a Float8 type when you are using a ValueWriter of type
NullableIntWriterImpl."
Based on a bit of googling I found that this mea
Thanks Paul, that helps a lot.
Regards
Prabhakar
On Tue, Feb 25, 2020 at 1:30 PM Paul Rogers
wrote:
> Hi Prabhakar,
>
> As it turns out, Drill is built for Java 8-13, but we've not built for
> Java 7 in quite some time. (Java 7 reached end of life several years back.)
>
> That said, you can try
Hi Prabhakar,
As it turns out, Drill is built for Java 8-13, but we've not built for Java 7
in quite some time. (Java 7 reached end of life several years back.)
That said, you can try to clone the project sources and do a build.
Unfortunately, the JDBC driver tends to use quite a bit of Drill's
12 matches
Mail list logo
