Maxim Gekk created SPARK-25931:
----------------------------------
Summary: Benchmarking creation of Jackson parser
Key: SPARK-25931
URL: https://issues.apache.org/jira/browse/SPARK-25931
Project: Spark
Issue Type: Test
Components: SQL
Affects Versions: 2.4.0
Reporter: Maxim Gekk
Existing JSON benchmarks perlineParsing and perlineParsingOfWideColumn don't
invoke Jackson parser at all due to an optimization for empty schema introduced
SPARK-24959. Need to add new benchmark which should forcibly create Jackson
parser for short and wide columns. For example:
{code:scala}
spark.read
.schema(schema)
.json(path)
.filter((_: Row) => true)
.count()
{code}
The *.filter((_: Row) => true)* prevents projection pushdown to JSON datasource
and forces fully parsing of JSON content.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
