Re: How to troubleshoot 'Could not find hash cache for joinId' which is failing always for some users and never for others

We had a similar problem while using explicit JOIN in the query (we didn't have sub-queries). We worked it around by switching to sort merge join from hash join (to get it run, though it'll be bit slower than hash join as per documentation) Try adding the hint */*+ USE_SORT_MERGE_JOIN*/ *in the SE

Re: Right way to use phoenix client - 4.3.1 jar

Hi Arun, Below mail archive might answer your question as well... http://mail-archives.apache.org/mod_mbox/incubator-phoenix-user/201409.mbox/%3CCAAF1Jdh8XNyF-E4_u36rT8OVgK5Ss=dvxroe0nnrwmdp-gn...@mail.gmail.com%3E Thanks, Kathir On Wed, Jun 3, 2015 at 2:37 PM, Nick Dimiduk wrote: > Hi Arun

Re: Phoenix Array Type Null Element

ix Array Type Null Element >> > >> > >> > >> > Note that for an array of primitive types (double, float, BIGINT, >> integer, >> > smallint, tinyint), we don't support inserting null as an array element. >> > >> > On Thursday,

Re: Phoenix Array Type Null Element

th the patch that I had attached. > > > > But for the BIGINT case we need to see the reason. > > > > Regards > > Ram > > > > *From:* Kathiresan S [mailto:kathiresanselva...@gmail.com] > *Sent:* Thursday, May 7, 2015 7:24 PM > > *To:* user@phoe

Re: UDF - Access Tuple Column Values By Name

the Tuple argument > in the evaluate method as it might go away in the future > (PHOENIX-1887). > > Thanks, > James > > On Wed, May 6, 2015 at 6:14 AM, Kathiresan S > wrote: > > Hi Jaime, > > > > In my case, the number of arguments passed to the custom funct

Re: Phoenix Array Type Null Element

On Thu, May 7, 2015 at 2:47 AM, Vasudevan, Ramkrishna S < ramkrishna.s.vasude...@intel.com> wrote: > Hi Kathir > > > > Ideally nulls should work with all the ARRAY types. Do you find any > issues in the behavior of these ARRAY data types when you use ‘nulls’? > >

Re: UDF - Access Tuple Column Values By Name

d, May 6, 2015 at 8:43 AM, Jaime Solano wrote: > Hi Kathir, > > I don't know if that's posible, but seems like the better approach is to > just define your function to take two arguments. > > What do you think? > > Best, > -Jaime > On May 6, 2015 7:50 AM, &q

UDF - Access Tuple Column Values By Name

Hi, In UDF, in the implemented evaluate(tuple,ptr) method, i could access the expression object of the columns passed to the UDF, by invoking *getChildren.get(index)*. and then by invoking expression.evaluate(tuple,ptr), i can get the value copied to the ptr object. But, how do i access other elem

Re: Phoenix Array Type Null Element

What are the other Array base data types (like VARCHAR), that could take null values? Do the below data types allow NULL as well? BIGINT ARRAY, DOUBLE ARRAY, DATE ARRAY, VARBINARY ARRAY Thanks, Kathir On Wed, May 6, 2015 at 6:35 AM, Kathiresan S wrote: > Thanks Ram. > > > Than

Re: Phoenix Array Type Null Element

th a patch ASAP. > > > > Regards > > Ram > > > > *From:* Kathiresan S [mailto:kathiresanselva...@gmail.com > ] > *Sent:* Tuesday, May 5, 2015 10:23 PM > *To:* user@phoenix.apache.org > *Subject:* Phoenix Array Type Null Element > > > > Hi, &

Re: Phoenix Array Type Null Element

Also, could you please let me know the array data base types that allow null (in case VARCHAR ARRAY will not allow NULLs)? Thanks, Kathir On Tue, May 5, 2015 at 12:53 PM, Kathiresan S wrote: > Hi, > > Is it possible to insert null elements in an array type column? > > CREATE TAB

Phoenix Array Type Null Element

Hi, Is it possible to insert null elements in an array type column? CREATE TABLE ARRAYTEST124 (ID VARCHAR, NAME VARCHAR ARRAY CONSTRAINT PK PRIMARY KEY(ID)) UPSERT INTO ARRAYTEST124 (ID, NAME) VALUES('123',ARRAY['ABC','XYZ',null]) UPSERT INTO ARRAYTEST124 (ID, NAME) VALUES('123',ARRAY['ABC',null

Re: UDF in SELECT

er side. > That phoenix automatically do it for you. > > Thanks, > Rajeshbabu. > > On Sun, May 3, 2015 at 1:02 AM, Kathiresan S > wrote: > >> Hi, >> >> I've tried the new UDF feature of Phoenix. >> >> Below are the version details >>

UDF in SELECT

Hi, I've tried the new UDF feature of Phoenix. Below are the version details Phoenix : master (tag : v4.4.0-HBase-0.98-rc0) Hbase : hbase-0.98.12-hadoop2 I created a UDF and tested with a simple SELECT statement. Looks like, its executed on the client side. Also, in one of the JIRAs

Re: Phoenix Array Flatten Function

Apr 23, 2015 at 5:44 AM, Kathiresan S < > kathiresanselva...@gmail.com> wrote: > >> Hi, >> >> Do we have something like flatten function in Apache drill, for array >> data types in phoenix? >> >> https://cwiki.apache.org/confluence/display/DRILL/FLATTEN+Function >> >> Thanks, >> Kathir >> > >

Phoenix Array Flatten Function

Hi, Do we have something like flatten function in Apache drill, for array data types in phoenix? https://cwiki.apache.org/confluence/display/DRILL/FLATTEN+Function Thanks, Kathir