[SPARK-23327][SQL] Update the description and tests of three external…

… API or functions

## What changes were proposed in this pull request?
Update the description and tests of three external API or functions `createFunction `, `length` and `repartitionByRange `

## How was this patch tested?
N/A

Author: gatorsmile <[email protected]>

Closes apache#20495 from gatorsmile/updateFunc.

(cherry picked from commit c36fecc)
Signed-off-by: gatorsmile <[email protected]>

R/pkg/R/functions.R

@@ -1026,7 +1026,9 @@ setMethod("last_day",
           })
 
 #' @details
-#' \code{length}: Computes the length of a given string or binary column.
+#' \code{length}: Computes the character length of a string data or number of bytes
+#' of a binary data. The length of string data includes the trailing spaces.
+#' The length of binary data includes binary zeros.
 #'
 #' @rdname column_string_functions
 #' @aliases length length,Column-method

python/pyspark/sql/functions.py

@@ -1705,10 +1705,12 @@ def unhex(col):
 @ignore_unicode_prefix
 @since(1.5)
 def length(col):
-    """Calculates the length of a string or binary expression.
+    """Computes the character length of string data or number of bytes of binary data.
+    The length of character data includes the trailing spaces. The length of binary data
+    includes binary zeros.
 
-    >>> spark.createDataFrame([('ABC',)], ['a']).select(length('a').alias('length')).collect()
-    [Row(length=3)]
+    >>> spark.createDataFrame([('ABC ',)], ['a']).select(length('a').alias('length')).collect()
+    [Row(length=4)]
     """
     sc = SparkContext._active_spark_context
     return Column(sc._jvm.functions.length(_to_java_column(col)))

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/catalog/SessionCatalog.scala

@@ -988,8 +988,11 @@ class SessionCatalog(
   // -------------------------------------------------------
 
   /**
-   * Create a metastore function in the database specified in `funcDefinition`.
+   * Create a function in the database specified in `funcDefinition`.
    * If no such database is specified, create it in the current database.
+   *
+   * @param ignoreIfExists: When true, ignore if the function with the specified name exists
+   *                        in the specified database.
    */
   def createFunction(funcDefinition: CatalogFunction, ignoreIfExists: Boolean): Unit = {
     val db = formatDatabaseName(funcDefinition.identifier.database.getOrElse(getCurrentDatabase))
@@ -1061,7 +1064,7 @@ class SessionCatalog(
   }
 
   /**
-   * Check if the specified function exists.
+   * Check if the function with the specified name exists
    */
   def functionExists(name: FunctionIdentifier): Boolean = {
     val db = formatDatabaseName(name.database.getOrElse(getCurrentDatabase))

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/stringExpressions.scala

@@ -1653,19 +1653,19 @@ case class Left(str: Expression, len: Expression, child: Expression) extends Run
  * A function that returns the char length of the given string expression or
  * number of bytes of the given binary expression.
  */
-// scalastyle:off line.size.limit
 @ExpressionDescription(
-  usage = "_FUNC_(expr) - Returns the character length of `expr` or number of bytes in binary data.",
+  usage = "_FUNC_(expr) - Returns the character length of string data or number of bytes of " +
+    "binary data. The length of string data includes the trailing spaces. The length of binary " +
+    "data includes binary zeros.",
   examples = """
     Examples:
-      > SELECT _FUNC_('Spark SQL');
-       9
-      > SELECT CHAR_LENGTH('Spark SQL');
-       9
-      > SELECT CHARACTER_LENGTH('Spark SQL');
-       9
+      > SELECT _FUNC_('Spark SQL ');
+       10
+      > SELECT CHAR_LENGTH('Spark SQL ');
+       10
+      > SELECT CHARACTER_LENGTH('Spark SQL ');
+       10
   """)
-// scalastyle:on line.size.limit
 case class Length(child: Expression) extends UnaryExpression with ImplicitCastInputTypes {
   override def dataType: DataType = IntegerType
   override def inputTypes: Seq[AbstractDataType] = Seq(TypeCollection(StringType, BinaryType))
@@ -1687,7 +1687,7 @@ case class Length(child: Expression) extends UnaryExpression with ImplicitCastIn
  * A function that returns the bit length of the given string or binary expression.
  */
 @ExpressionDescription(
-  usage = "_FUNC_(expr) - Returns the bit length of `expr` or number of bits in binary data.",
+  usage = "_FUNC_(expr) - Returns the bit length of string data or number of bits of binary data.",
   examples = """
     Examples:
       > SELECT _FUNC_('Spark SQL');
@@ -1716,7 +1716,8 @@ case class BitLength(child: Expression) extends UnaryExpression with ImplicitCas
  * A function that returns the byte length of the given string or binary expression.
  */
 @ExpressionDescription(
-  usage = "_FUNC_(expr) - Returns the byte length of `expr` or number of bytes in binary data.",
+  usage = "_FUNC_(expr) - Returns the byte length of string data or number of bytes of binary " +
+    "data.",
   examples = """
     Examples:
       > SELECT _FUNC_('Spark SQL');

sql/core/src/main/scala/org/apache/spark/sql/Dataset.scala

@@ -2825,6 +2825,7 @@ class Dataset[T] private[sql](
    *
    * At least one partition-by expression must be specified.
    * When no explicit sort order is specified, "ascending nulls first" is assumed.
+   * Note, the rows are not sorted in each partition of the resulting Dataset.
    *
    * @group typedrel
    * @since 2.3.0
@@ -2848,6 +2849,7 @@ class Dataset[T] private[sql](
    *
    * At least one partition-by expression must be specified.
    * When no explicit sort order is specified, "ascending nulls first" is assumed.
+   * Note, the rows are not sorted in each partition of the resulting Dataset.
    *
    * @group typedrel
    * @since 2.3.0

sql/core/src/main/scala/org/apache/spark/sql/execution/command/functions.scala

@@ -40,24 +40,28 @@ import org.apache.spark.sql.types.{StringType, StructField, StructType}
  *    CREATE [OR REPLACE] FUNCTION [IF NOT EXISTS] [databaseName.]functionName
  *    AS className [USING JAR\FILE 'uri' [, JAR|FILE 'uri']]
  * }}}
+ *
+ * @param ignoreIfExists: When true, ignore if the function with the specified name exists
+ *                        in the specified database.
+ * @param replace: When true, alter the function with the specified name
  */
 case class CreateFunctionCommand(
     databaseName: Option[String],
     functionName: String,
     className: String,
     resources: Seq[FunctionResource],
     isTemp: Boolean,
-    ifNotExists: Boolean,
+    ignoreIfExists: Boolean,
     replace: Boolean)
   extends RunnableCommand {
 
-  if (ifNotExists && replace) {
+  if (ignoreIfExists && replace) {
     throw new AnalysisException("CREATE FUNCTION with both IF NOT EXISTS and REPLACE" +
       " is not allowed.")
   }
 
   // Disallow to define a temporary function with `IF NOT EXISTS`
-  if (ifNotExists && isTemp) {
+  if (ignoreIfExists && isTemp) {
     throw new AnalysisException(
       "It is not allowed to define a TEMPORARY function with IF NOT EXISTS.")
   }
@@ -79,12 +83,12 @@ case class CreateFunctionCommand(
       // Handles `CREATE OR REPLACE FUNCTION AS ... USING ...`
       if (replace && catalog.functionExists(func.identifier)) {
         // alter the function in the metastore
-        catalog.alterFunction(CatalogFunction(func.identifier, className, resources))
+        catalog.alterFunction(func)
       } else {
         // For a permanent, we will store the metadata into underlying external catalog.
         // This function will be loaded into the FunctionRegistry when a query uses it.
         // We do not load it into FunctionRegistry right now.
-        catalog.createFunction(CatalogFunction(func.identifier, className, resources), ifNotExists)
+        catalog.createFunction(func, ignoreIfExists)
       }
     }
     Seq.empty[Row]

sql/core/src/main/scala/org/apache/spark/sql/functions.scala

@@ -2267,7 +2267,9 @@ object functions {
   }
 
   /**
-   * Computes the length of a given string or binary column.
+   * Computes the character length of a given string or number of bytes of a binary string.
+   * The length of character strings include the trailing spaces. The length of binary strings
+   * includes binary zeros.
    *
    * @group string_funcs
    * @since 1.5.0

sql/core/src/test/scala/org/apache/spark/sql/execution/command/DDLParserSuite.scala

@@ -236,39 +236,39 @@ class DDLParserSuite extends PlanTest with SharedSQLContext {
       Seq(
         FunctionResource(FunctionResourceType.fromString("jar"), "/path/to/jar1"),
         FunctionResource(FunctionResourceType.fromString("jar"), "/path/to/jar2")),
-      isTemp = true, ifNotExists = false, replace = false)
+      isTemp = true, ignoreIfExists = false, replace = false)
     val expected2 = CreateFunctionCommand(
       Some("hello"),
       "world",
       "com.matthewrathbone.example.SimpleUDFExample",
       Seq(
         FunctionResource(FunctionResourceType.fromString("archive"), "/path/to/archive"),
         FunctionResource(FunctionResourceType.fromString("file"), "/path/to/file")),
-      isTemp = false, ifNotExists = false, replace = false)
+      isTemp = false, ignoreIfExists = false, replace = false)
     val expected3 = CreateFunctionCommand(
       None,
       "helloworld3",
       "com.matthewrathbone.example.SimpleUDFExample",
       Seq(
         FunctionResource(FunctionResourceType.fromString("jar"), "/path/to/jar1"),
         FunctionResource(FunctionResourceType.fromString("jar"), "/path/to/jar2")),
-      isTemp = true, ifNotExists = false, replace = true)
+      isTemp = true, ignoreIfExists = false, replace = true)
     val expected4 = CreateFunctionCommand(
       Some("hello"),
       "world1",
       "com.matthewrathbone.example.SimpleUDFExample",
       Seq(
         FunctionResource(FunctionResourceType.fromString("archive"), "/path/to/archive"),
         FunctionResource(FunctionResourceType.fromString("file"), "/path/to/file")),
-      isTemp = false, ifNotExists = false, replace = true)
+      isTemp = false, ignoreIfExists = false, replace = true)
     val expected5 = CreateFunctionCommand(
       Some("hello"),
       "world2",
       "com.matthewrathbone.example.SimpleUDFExample",
       Seq(
         FunctionResource(FunctionResourceType.fromString("archive"), "/path/to/archive"),
         FunctionResource(FunctionResourceType.fromString("file"), "/path/to/file")),
-      isTemp = false, ifNotExists = true, replace = false)
+      isTemp = false, ignoreIfExists = true, replace = false)
     comparePlans(parsed1, expected1)
     comparePlans(parsed2, expected2)
     comparePlans(parsed3, expected3)