>
>
>
> *From:* Kris Mo
> *Sent:* Friday, June 12, 2020 2:20 AM
> *To:* Nasrulla Khan Haris
> *Cc:* dev@spark.apache.org
> *Subject:* [EXTERNAL] Re: ColumnnarBatch to InternalRow Cast exception
> with codegen enabled.
>
>
>
> Hi Nasrulla,
>
>
>
>
.
From: Kris Mo
Sent: Friday, June 12, 2020 2:20 AM
To: Nasrulla Khan Haris
Cc: dev@spark.apache.org
Subject: [EXTERNAL] Re: ColumnnarBatch to InternalRow Cast exception with
codegen enabled.
Hi Nasrulla,
Not sure what your new code is doing, but the symptom looks like you're
creating a new data
: [EXTERNAL] Re: ColumnnarBatch to InternalRow Cast exception with
codegen enabled.
Hi Nasrulla,
Not sure what your new code is doing, but the symptom looks like you're
creating a new data source that wraps around the builtin Parquet data source?
The problem here is, whole-stage codegen generated
Hi Nasrulla,
Not sure what your new code is doing, but the symptom looks like you're
creating a new data source that wraps around the builtin Parquet data
source?
The problem here is, whole-stage codegen generated code for row-based
input, but the actual input is columnar.
In other words, in
HI Spark developer,
I have a new baseRelation which Initializes ParquetFileFormat object and when
reading the data I am encountering Cast Exception below, however when I disable
codegen support with config "spark.sql.codegen.wholeStage"= false, I do not
encounter this exception.
20/06/11