Github user dongjoon-hyun commented on a diff in the pull request: https://github.com/apache/spark/pull/14004#discussion_r69942945 --- Diff: sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/regexpExpressions.scala --- @@ -198,6 +203,66 @@ case class StringSplit(str: Expression, pattern: Expression) override def prettyName: String = "split" } +/** + * Splits a string into arrays of sentences, where each sentence is an array of words. + * The 'lang' and 'country' arguments are optional, and if omitted, the default locale is used. + */ +@ExpressionDescription( + usage = "_FUNC_(str, lang, country) - Splits str into an array of array of words.", + extended = "> SELECT _FUNC_('Hi there! Good morning.');\n [['Hi','there'], ['Good','morning']]") +case class Sentences( + str: Expression, + language: Expression = Literal(""), + country: Expression = Literal("")) + extends Expression with ImplicitCastInputTypes with CodegenFallback { + + def this(str: Expression) = this(str, Literal(""), Literal("")) + def this(str: Expression, language: Expression) = this(str, language, Literal("")) + + override def nullable: Boolean = true + override def dataType: DataType = + ArrayType(ArrayType(StringType, containsNull = false), containsNull = false) + override def inputTypes: Seq[AbstractDataType] = Seq(StringType, StringType, StringType) + override def children: Seq[Expression] = str :: language :: country :: Nil + + override def eval(input: InternalRow): Any = { + val string = str.eval(input) + if (string == null) { + null + } else { + val locale = try { + new Locale(language.eval(input).asInstanceOf[UTF8String].toString, + country.eval(input).asInstanceOf[UTF8String].toString) + } catch { + case _: NullPointerException | _: ClassCastException => Locale.getDefault --- End diff -- Oh, it was ambiguous. `new Locale("abc", "xyz")` is created, it's not null. The following works but uses JVM Default locale. `BreakIterator.getSentenceInstance` does not return `null` with wrong locale instance. ``` scala> val x = java.text.BreakIterator.getSentenceInstance(new java.util.Locale("xxx", "bbb")) x: java.text.BreakIterator = [checksum=0xcba403eb] ```
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. --- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org