SparkSession — The Entry Point to Spark SQL

SparkSession is the entry point to Spark SQL. It is the very first object you have to create while developing Spark SQL applications using the fully-typed Dataset (or untyped Row-based DataFrame) data abstractions.

Note	`SparkSession` has merged SQLContext and HiveContext in one object in Spark 2.0.

You use the SparkSession.builder method to create an instance of SparkSession.

import org.apache.spark.sql.SparkSession
val spark: SparkSession = SparkSession.builder
  .appName("My Spark Application")  // optional and will be autogenerated if not specified
  .master("local[*]")               // avoid hardcoding the deployment environment
  .enableHiveSupport()              // self-explanatory, isn't it?
  .config("spark.sql.warehouse.dir", "target/spark-warehouse")
  .getOrCreate

And stop the current SparkSession using stop method.

spark.stop

You can have as many SparkSessions as you want in a single Spark application. The common use case is to keep relational entities separate per SparkSession (see catalog attribute).

scala> spark.catalog.listTables.show
+------------------+--------+-----------+---------+-----------+
|              name|database|description|tableType|isTemporary|
+------------------+--------+-----------+---------+-----------+
|my_permanent_table| default|       null|  MANAGED|      false|
|              strs|    null|       null|TEMPORARY|       true|
+------------------+--------+-----------+---------+-----------+

Internally, SparkSession requires a SparkContext and an optional SharedState (that represents the shared state across SparkSession instances).

Table 1. SparkSession’s Class and Instance Methods
Method	Description
builder	"Opens" a builder to get or create a `SparkSession` instance
version	Returns the current version of Spark.
implicits	Use `import spark.implicits._` to import the implicits conversions and create `Datasets` from (almost arbitrary) Scala objects.
emptyDataset[T]	Creates an empty `Dataset[T]`.
range	Creates a `Dataset[Long]`.
sql	Executes a SQL query (and returns a `DataFrame`).
udf	Access to user-defined functions (UDFs).
table	Creates a `DataFrame` from a table.
catalog	Access to the catalog of the entities of structured queries
read	Access to `DataFrameReader` to read a `DataFrame` from external files and storage systems.
conf	Access to the current runtime configuration.
readStream	Access to `DataStreamReader` to read streaming datasets.
streams	Access to `StreamingQueryManager` to manage structured streaming queries.
newSession	Creates a new `SparkSession`.
stop	Stops the `SparkSession`.

Tip

Use spark.sql.warehouse.dir Spark property to change the location of Hive’s hive.metastore.warehouse.dir property, i.e. the location of the Hive local/embedded metastore database (using Derby).

Refer to SharedState to learn about (the low-level details of) Spark SQL support for Apache Hive.

See also the official Hive Metastore Administration document.

Table 2. SparkSession’s (Lazily-Initialized) Attributes (in alphabetical order)
Name	Type	Description
`sessionState`	SessionState	Internally, `sessionState` clones the optional parent `SessionState` (if given when creating SparkSession) or creates a new `SessionState` using BaseSessionStateBuilder as defined by spark.sql.catalogImplementation property: in-memory (default) for `org.apache.spark.sql.internal.SessionStateBuilder` hive for `org.apache.spark.sql.hive.HiveSessionStateBuilder`
`sharedState`	SharedState

Note	baseRelationToDataFrame acts as a mechanism to plug `BaseRelation` object hierarchy in into LogicalPlan object hierarchy that `SparkSession` uses to bridge them.

Creating SparkSession Instance

Caution

FIXME

Creating SparkSession Using Builder Pattern — `builder` Method

builder(): Builder

builder creates a new Builder that you use to build a fully-configured SparkSession using a fluent API.

import org.apache.spark.sql.SparkSession
val builder = SparkSession.builder

Tip	Read about Fluent interface design pattern in Wikipedia, the free encyclopedia.

Accessing Version of Spark — `version` Method

version: String

version returns the version of Apache Spark in use.

Internally, version uses spark.SPARK_VERSION value that is the version property in spark-version-info.properties properties file on CLASSPATH.

Implicit Conversions — `implicits` object

The implicits object is a helper class with the Scala implicit methods (aka conversions) to convert Scala objects to Datasets, DataFrames and Columns. It also defines Encoders for Scala’s "primitive" types, e.g. Int, Double, String, and their products and collections.

Note	Import the implicits by `import spark.implicits._`. `val spark = SparkSession.builder.getOrCreate() import spark.implicits._`

implicits object offers support for creating Dataset from RDD of any type (for which an encoder exists in scope), or case classes or tuples, and Seq.

implicits object also offers conversions from Scala’s Symbol or $ to Column.

It also offers conversions from RDD or Seq of Product types (e.g. case classes or tuples) to DataFrame. It has direct conversions from RDD of Int, Long and String to DataFrame with a single column name _1.

Note	It is only possible to call `toDF` methods on `RDD` objects of `Int`, `Long`, and `String` "primitive" types.

Creating Empty Dataset — `emptyDataset` method

emptyDataset[T: Encoder]: Dataset[T]

emptyDataset creates an empty Dataset (assuming that future records being of type T).

scala> val strings = spark.emptyDataset[String]
strings: org.apache.spark.sql.Dataset[String] = [value: string]

scala> strings.printSchema
root
 |-- value: string (nullable = true)

emptyDataset creates a LocalRelation logical query plan.

Creating Dataset from Local Collections and RDDs — `createDataset` methods

createDataset[T : Encoder](data: Seq[T]): Dataset[T]
createDataset[T : Encoder](data: RDD[T]): Dataset[T]

createDataset is an experimental API to create a Dataset from a local Scala collection, i.e. Seq[T], Java’s List[T], or a distributed RDD[T].

scala> val one = spark.createDataset(Seq(1))
one: org.apache.spark.sql.Dataset[Int] = [value: int]

scala> one.show
+-----+
|value|
+-----+
|    1|
+-----+

createDataset creates a LocalRelation logical query plan (for the input data collection) or LogicalRDD (for the input RDD[T]).

Tip	You’d be better off using Scala implicits and `toDS` method instead (that does this conversion automatically for you). `val spark: SparkSession = ... import spark.implicits._ scala> val one = Seq(1).toDS one: org.apache.spark.sql.Dataset[Int] = [value: int]`

Internally, createDataset first looks up the implicit expression encoder in scope to access the AttributeReferences (of the schema).

Note	Only unresolved expression encoders are currently supported.

The expression encoder is then used to map elements (of the input Seq[T]) into a collection of InternalRows. With the references and rows, createDataset returns a Dataset with a LocalRelation logical query plan.

Creating Dataset With Single Long Column — `range` methods

range(end: Long): Dataset[java.lang.Long]
range(start: Long, end: Long): Dataset[java.lang.Long]
range(start: Long, end: Long, step: Long): Dataset[java.lang.Long]
range(start: Long, end: Long, step: Long, numPartitions: Int): Dataset[java.lang.Long]

range family of methods create a Dataset of Long numbers.

scala> spark.range(start = 0, end = 4, step = 2, numPartitions = 5).show
+---+
| id|
+---+
|  0|
|  2|
+---+

Note	The three first variants (that do not specify `numPartitions` explicitly) use SparkContext.defaultParallelism for the number of partitions `numPartitions`.

Internally, range creates a new Dataset[Long] with Range logical plan and Encoders.LONG encoder.

Creating Empty DataFrame — `emptyDataFrame` method

emptyDataFrame: DataFrame

emptyDataFrame creates an empty DataFrame (with no rows and columns).

It calls createDataFrame with an empty RDD[Row] and an empty schema StructType(Nil).

Creating DataFrames from RDDs with Explicit Schema — `createDataFrame` method

createDataFrame(rowRDD: RDD[Row], schema: StructType): DataFrame

createDataFrame creates a DataFrame using RDD[Row] and the input schema. It is assumed that the rows in rowRDD all match the schema.

Executing SQL Queries (aka SQL Mode) — `sql` Method

sql(sqlText: String): DataFrame

sql executes the sqlText SQL statement and creates a DataFrame.

Note	`sql` is imported in spark-shell so you can execute SQL statements as if `sql` were a part of the environment. `scala> spark.version res0: String = 2.2.0-SNAPSHOT scala> :imports 1) import spark.implicits._ (72 terms, 43 are implicit) 2) import spark.sql (1 terms)`

scala> sql("SHOW TABLES")
res0: org.apache.spark.sql.DataFrame = [tableName: string, isTemporary: boolean]

scala> sql("DROP TABLE IF EXISTS testData")
res1: org.apache.spark.sql.DataFrame = []

// Let's create a table to SHOW it
spark.range(10).write.option("path", "/tmp/test").saveAsTable("testData")

scala> sql("SHOW TABLES").show
+---------+-----------+
|tableName|isTemporary|
+---------+-----------+
| testdata|      false|
+---------+-----------+

Internally, sql requests the current ParserInterface to execute a SQL query that gives a LogicalPlan.

Note	`sql` uses `SessionState` to access the current `ParserInterface`.

sql then creates a DataFrame using the current SparkSession (itself) and the LogicalPlan.

Tip	spark-sql is the main SQL environment in Spark to work with pure SQL statements (where you do not have to use Scala to execute them). `spark-sql> show databases; default Time taken: 0.028 seconds, Fetched 1 row(s)`

Accessing UDF Registration Interface — `udf` Attribute

udf: UDFRegistration

udf attribute gives access to UDFRegistration that allows registering user-defined functions for SQL-based queries.

val spark: SparkSession = ...
spark.udf.register("myUpper", (s: String) => s.toUpperCase)

val strs = ('a' to 'c').map(_.toString).toDS
strs.registerTempTable("strs")

scala> sql("SELECT *, myUpper(value) UPPER FROM strs").show
+-----+-----+
|value|UPPER|
+-----+-----+
|    a|    A|
|    b|    B|
|    c|    C|
+-----+-----+

Internally, it is simply an alias for SessionState.udfRegistration.

Creating DataFrame for Table — `table` method

table(tableName: String): DataFrame

table creates a DataFrame from records in the tableName table (if exists).

val df = spark.table("mytable")

Accessing Metastore — `catalog` Attribute

catalog: Catalog

catalog attribute is a (lazy) interface to the current metastore, i.e. data catalog (of relational entities like databases, tables, functions, table columns, and temporary views).

Tip	All methods in `Catalog` return `Datasets`.

scala> spark.catalog.listTables.show
+------------------+--------+-----------+---------+-----------+
|              name|database|description|tableType|isTemporary|
+------------------+--------+-----------+---------+-----------+
|my_permanent_table| default|       null|  MANAGED|      false|
|              strs|    null|       null|TEMPORARY|       true|
+------------------+--------+-----------+---------+-----------+

Internally, catalog creates a CatalogImpl (that uses the current SparkSession).

Accessing DataFrameReader — `read` method

read: DataFrameReader

read method returns a DataFrameReader that is used to read data from external storage systems and load it into a DataFrame.

val spark: SparkSession = // create instance
val dfReader: DataFrameReader = spark.read

Runtime Configuration — `conf` attribute

conf: RuntimeConfig

conf returns the current runtime configuration (as RuntimeConfig) that wraps SQLConf.

Caution

FIXME

`readStream` method

readStream: DataStreamReader

readStream returns a new DataStreamReader.

`streams` Attribute

streams: StreamingQueryManager

streams attribute gives access to StreamingQueryManager (through SessionState).

val spark: SparkSession = ...
spark.streams.active.foreach(println)

`streamingQueryManager` Attribute

streamingQueryManager is…

`listenerManager` Attribute

listenerManager is…

`ExecutionListenerManager`

ExecutionListenerManager is…

`functionRegistry` Attribute

functionRegistry is…

`experimentalMethods` Attribute

experimental: ExperimentalMethods

experimentalMethods is an extension point with ExperimentalMethods that is a per-session collection of extra strategies and Rule[LogicalPlan]s.

Note	`experimental` is used in SparkPlanner and SparkOptimizer. Hive and Structured Streaming use it for their own extra strategies and optimization rules.

`newSession` method

newSession(): SparkSession

newSession creates (starts) a new SparkSession (with the current SparkContext and SharedState).

scala> println(sc.version)
2.0.0-SNAPSHOT

scala> val newSession = spark.newSession
newSession: org.apache.spark.sql.SparkSession = org.apache.spark.sql.SparkSession@122f58a

Stopping SparkSession — `stop` Method

stop(): Unit

stop stops the SparkSession, i.e. stops the underlying SparkContext.

Create DataFrame from BaseRelation — `baseRelationToDataFrame` Method

baseRelationToDataFrame(baseRelation: BaseRelation): DataFrame

Internally, baseRelationToDataFrame creates a DataFrame from the input BaseRelation wrapped inside LogicalRelation.

Note	LogicalRelation is an logical plan adapter for `BaseRelation` (so `BaseRelation` can be part of a logical plan).

Note

baseRelationToDataFrame is used when:

DataFrameReader loads data from a data source that supports multiple paths
DataFrameReader loads data from an external table using JDBC
TextInputCSVDataSource creates a base Dataset (of Strings)
TextInputJsonDataSource creates a base Dataset (of Strings)

Building SessionState — `instantiateSessionState` Internal Method

instantiateSessionState(className: String, sparkSession: SparkSession): SessionState

instantiateSessionState finds the className that is then used to create and immediatelly build a BaseSessionStateBuilder.

instantiateSessionState reports a IllegalArgumentException while constructing a SessionState:

Error while instantiating '[className]'

Note	`instantiateSessionState` is used exclusively when `SparkSession` is requested for `SessionState` (and one is not available yet).

SparkSession — The Entry Point to Spark SQL