Execute Script File should expose more control parameters

[allwefantasy]

With Byzer, you can run a local byzer script file with following command:

/opt/apps/SPARK3/spark-current/bin/spark-submit --class streaming.core.StreamingApp \
--master yarn \
--deploy-mode client \
--driver-memory 40g \
--conf spark.executor.cores=3 \
--conf spark.serializer=org.apache.spark.serializer.KryoSerializer \
--conf spark.executor.instances=2 \
--conf spark.mlsql.path.schemas=oss \
/opt/byzer/byzer-lang/main/byzer-lang-3.3.0-2.12-2.4.0-SNAPSHOT.jar  \
-streaming.platform spark \
-streaming.name byzer-engine \
-streaming.thrift false \
-streaming.rest false \
-streaming.spark.service false \
-streaming.enableHiveSupport true \
-streaming.platform_hooks tech.mlsql.runtime.SparkSubmitMLSQLScriptRuntimeLifecycle \
-streaming.mlsql.script.path file:///home/byzer/byzer-project/customer_identity.byzer \
-streaming.mlsql.script.owner owner \
-streaming.mlsql.script.jobName scheduler

however ,you may try to control more behaviors when executing the script.

[allwefantasy]

For example, if you want to control the default home prefix , you can try add following config item:

-streaming.mlsql.script.query.defaultPathPrefix  /home/winubuntu/byzer-home

The query parameter in Rest can be passed to the system by key with prefix streaming.mlsql.script.query..