[SPARK-16128][SQL] Allow setting length of characters to be truncated…

… to, in Dataset.show function.

## What changes were proposed in this pull request?

Allowing truncate to a specific number of character is convenient at times, especially while operating from the REPL. Sometimes those last few characters make all the difference, and showing everything brings in whole lot of noise.

## How was this patch tested?
Existing tests. + 1 new test in DataFrameSuite.

For SparkR and pyspark, existing tests and manual testing.

Author: Prashant Sharma <[email protected]>
Author: Prashant Sharma <[email protected]>

Closes apache#13839 from ScrapCodes/add_truncateTo_DF.show.

.gitignore

@@ -77,3 +77,4 @@ spark-warehouse/
 # For R session data
 .RData
 .RHistory
+.Rhistory

R/pkg/R/DataFrame.R

@@ -176,8 +176,8 @@ setMethod("isLocal",
 #' @param x A SparkDataFrame
 #' @param numRows The number of rows to print. Defaults to 20.
 #' @param truncate Whether truncate long strings. If true, strings more than 20 characters will be
-#' truncated and all cells will be aligned right
-#'
+#'    truncated. However, if set greater than zero, truncates strings longer than `truncate`
+#'    characters and all cells will be aligned right.
 #' @family SparkDataFrame functions
 #' @rdname showDF
 #' @name showDF
@@ -193,7 +193,12 @@ setMethod("isLocal",
 setMethod("showDF",
           signature(x = "SparkDataFrame"),
           function(x, numRows = 20, truncate = TRUE) {
-            s <- callJMethod(x@sdf, "showString", numToInt(numRows), truncate)
+            if (is.logical(truncate) && truncate) {
+              s <- callJMethod(x@sdf, "showString", numToInt(numRows), numToInt(20))
+            } else {
+              truncate2 <- as.numeric(truncate)
+              s <- callJMethod(x@sdf, "showString", numToInt(numRows), numToInt(truncate2))
+            }
             cat(s)
           })
 

R/pkg/inst/tests/testthat/test_sparkSQL.R

@@ -1582,7 +1582,15 @@ test_that("showDF()", {
                     "|  30|   Andy|\n",
                     "|  19| Justin|\n",
                     "+----+-------+\n", sep = "")
+  expected2 <- paste("+---+----+\n",
+                     "|age|name|\n",
+                     "+---+----+\n",
+                     "|nul| Mic|\n",
+                     "| 30| And|\n",
+                     "| 19| Jus|\n",
+                     "+---+----+\n", sep = "")
   expect_output(showDF(df), expected)
+  expect_output(showDF(df, truncate = 3), expected2)
 })
 
 test_that("isLocal()", {

python/pyspark/sql/dataframe.py

@@ -271,7 +271,9 @@ def show(self, n=20, truncate=True):
         """Prints the first ``n`` rows to the console.
 
         :param n: Number of rows to show.
-        :param truncate: Whether truncate long strings and align cells right.
+        :param truncate: If set to True, truncate strings longer than 20 chars by default.
+            If set to a number greater than one, truncates long strings to length ``truncate``
+            and align cells right.
 
         >>> df
         DataFrame[age: int, name: string]
@@ -282,8 +284,18 @@ def show(self, n=20, truncate=True):
         |  2|Alice|
         |  5|  Bob|
         +---+-----+
-        """
-        print(self._jdf.showString(n, truncate))
+        >>> df.show(truncate=3)
+        +---+----+
+        |age|name|
+        +---+----+
+        |  2| Ali|
+        |  5| Bob|
+        +---+----+
+        """
+        if isinstance(truncate, bool) and truncate:
+            print(self._jdf.showString(n, 20))
+        else:
+            print(self._jdf.showString(n, int(truncate)))
 
     def __repr__(self):
         return "DataFrame[%s]" % (", ".join("%s: %s" % c for c in self.dtypes))

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

@@ -232,16 +232,18 @@ class Dataset[T] private[sql](
    * Compose the string representing rows for output
    *
    * @param _numRows Number of rows to show
-   * @param truncate Whether truncate long strings and align cells right
+   * @param truncate If set to more than 0, truncates strings to `truncate` characters and
+   *                   all cells will be aligned right.
    */
-  private[sql] def showString(_numRows: Int, truncate: Boolean = true): String = {
+  private[sql] def showString(_numRows: Int, truncate: Int = 20): String = {
     val numRows = _numRows.max(0)
     val takeResult = toDF().take(numRows + 1)
     val hasMoreData = takeResult.length > numRows
     val data = takeResult.take(numRows)
 
     // For array values, replace Seq and Array with square brackets
-    // For cells that are beyond 20 characters, replace it with the first 17 and "..."
+    // For cells that are beyond `truncate` characters, replace it with the
+    // first `truncate-3` and "..."
     val rows: Seq[Seq[String]] = schema.fieldNames.toSeq +: data.map { row =>
       row.toSeq.map { cell =>
         val str = cell match {
@@ -251,7 +253,13 @@ class Dataset[T] private[sql](
           case seq: Seq[_] => seq.mkString("[", ", ", "]")
           case _ => cell.toString
         }
-        if (truncate && str.length > 20) str.substring(0, 17) + "..." else str
+        if (truncate > 0 && str.length > truncate) {
+          // do not show ellipses for strings shorter than 4 characters.
+          if (truncate < 4) str.substring(0, truncate)
+          else str.substring(0, truncate - 3) + "..."
+        } else {
+          str
+        }
       }: Seq[String]
     }
 
@@ -273,7 +281,7 @@ class Dataset[T] private[sql](
 
     // column names
     rows.head.zipWithIndex.map { case (cell, i) =>
-      if (truncate) {
+      if (truncate > 0) {
         StringUtils.leftPad(cell, colWidths(i))
       } else {
         StringUtils.rightPad(cell, colWidths(i))
@@ -285,7 +293,7 @@ class Dataset[T] private[sql](
     // data
     rows.tail.map {
       _.zipWithIndex.map { case (cell, i) =>
-        if (truncate) {
+        if (truncate > 0) {
           StringUtils.leftPad(cell.toString, colWidths(i))
         } else {
           StringUtils.rightPad(cell.toString, colWidths(i))
@@ -523,7 +531,32 @@ class Dataset[T] private[sql](
    * @since 1.6.0
    */
   // scalastyle:off println
-  def show(numRows: Int, truncate: Boolean): Unit = println(showString(numRows, truncate))
+  def show(numRows: Int, truncate: Boolean): Unit = if (truncate) {
+    println(showString(numRows, truncate = 20))
+  } else {
+    println(showString(numRows, truncate = 0))
+  }
+  // scalastyle:on println
+
+  /**
+   * Displays the Dataset in a tabular form. For example:
+   * {{{
+   *   year  month AVG('Adj Close) MAX('Adj Close)
+   *   1980  12    0.503218        0.595103
+   *   1981  01    0.523289        0.570307
+   *   1982  02    0.436504        0.475256
+   *   1983  03    0.410516        0.442194
+   *   1984  04    0.450090        0.483521
+   * }}}
+ *
+   * @param numRows Number of rows to show
+   * @param truncate If set to more than 0, truncates strings to `truncate` characters and
+   *                    all cells will be aligned right.
+   * @group action
+   * @since 1.6.0
+   */
+  // scalastyle:off println
+  def show(numRows: Int, truncate: Int): Unit = println(showString(numRows, truncate))
   // scalastyle:on println
 
   /**

sql/core/src/test/scala/org/apache/spark/sql/DataFrameSuite.scala

@@ -723,7 +723,7 @@ class DataFrameSuite extends QueryTest with SharedSQLContext {
     testData.select($"*").show(1000)
   }
 
-  test("showString: truncate = [true, false]") {
+  test("showString: truncate = [0, 20]") {
     val longString = Array.fill(21)("1").mkString
     val df = sparkContext.parallelize(Seq("1", longString)).toDF()
     val expectedAnswerForFalse = """+---------------------+
@@ -733,15 +733,36 @@ class DataFrameSuite extends QueryTest with SharedSQLContext {
                                    ||111111111111111111111|
                                    |+---------------------+
                                    |""".stripMargin
-    assert(df.showString(10, false) === expectedAnswerForFalse)
+    assert(df.showString(10, truncate = 0) === expectedAnswerForFalse)
     val expectedAnswerForTrue = """+--------------------+
                                   ||               value|
                                   |+--------------------+
                                   ||                   1|
                                   ||11111111111111111...|
                                   |+--------------------+
                                   |""".stripMargin
-    assert(df.showString(10, true) === expectedAnswerForTrue)
+    assert(df.showString(10, truncate = 20) === expectedAnswerForTrue)
+  }
+
+  test("showString: truncate = [3, 17]") {
+    val longString = Array.fill(21)("1").mkString
+    val df = sparkContext.parallelize(Seq("1", longString)).toDF()
+    val expectedAnswerForFalse = """+-----+
+                                   ||value|
+                                   |+-----+
+                                   ||    1|
+                                   ||  111|
+                                   |+-----+
+                                   |""".stripMargin
+    assert(df.showString(10, truncate = 3) === expectedAnswerForFalse)
+    val expectedAnswerForTrue = """+-----------------+
+                                  ||            value|
+                                  |+-----------------+
+                                  ||                1|
+                                  ||11111111111111...|
+                                  |+-----------------+
+                                  |""".stripMargin
+    assert(df.showString(10, truncate = 17) === expectedAnswerForTrue)
   }
 
   test("showString(negative)") {

sql/core/src/test/scala/org/apache/spark/sql/DatasetSuite.scala

@@ -730,7 +730,7 @@ class DatasetSuite extends QueryTest with SharedSQLContext {
 
   private def checkShowString[T](ds: Dataset[T], expected: String): Unit = {
     val numRows = expected.split("\n").length - 4
-    val actual = ds.showString(numRows, truncate = true)
+    val actual = ds.showString(numRows, truncate = 20)
 
     if (expected != actual) {
       fail(