Error "Incompatible Jackson version" when using with PySpark 2.2.0 about clickhouse-java HOT 4 CLOSED

Sorry.
I just solved this problem removing
rm $SPARK_PATH/jars/clickhouse-jdbc-0.1-SNAPSHOT-jar-with-dependencies.jar

and using
cp clickhouse-jdbc-0.1-SNAPSHOT.jar $SPARK_PATH/jars/

and putting these jars to $SPARK_PATH/jars/
https://mvnrepository.com/artifact/org.apache.httpcomponents/httpclient/4.5.2
https://mvnrepository.com/artifact/org.apache.httpcomponents/httpcore/4.4.4

from clickhouse-java.

The problem begin to arise exactly when I put jar to spark jars dir

cp clickhouse-jdbc-0.1-SNAPSHOT-jar-with-dependencies.jar $SPARK_PATH/jars/

After this, even when I just read file in pyspark (commenting all clickhouse code) it gives me

Caused by: com.fasterxml.jackson.databind.JsonMappingException: Incompatible Jackson version: 2.7.3

So it crushes pyspark.
When I remove clickhouse-jdbc-0.1-SNAPSHOT-jar-with-dependencies.jar from $SPARK_PATH/jars/ my program begins to load DataFrame, but of course it cannot dump it to clickhouse.

from clickhouse-java.

Hello.
The problem is that current driver implementation has dependencies. In fact jackson isn't used for actual JDBC interface but for some specific methods, so it doesn't really matter which jackson version is provided.
Ideally drivers must be free of dependencies, but is not the case now and it needs a lot of effort.
You've found out the correct workaround.
You may need to add some other dependencies the same way if Spark doesn't provide ones and if you need the features provided by these dependencies: lz4, guava, joda-time.

from clickhouse-java.

Jackson was replaced by Gson(for better performance), and the latter is an optional dependency(because there's a simple JSON parser in jdbc driver).

from clickhouse-java.