Add a snakifyColumns method that uses the snakify method defined in t…

…he Lyft web development framework

src/main/scala/com/github/mrpowers/spark/daria/sql/FunctionsAsColumnExt.scala

@@ -1,7 +1,7 @@
 package com.github.mrpowers.spark.daria.sql
 
 import org.apache.spark.sql.Column
-import org.apache.spark.sql.{functions=>F}
+import org.apache.spark.sql.{functions => F}
 
 object FunctionsAsColumnExt {
 
@@ -14,7 +14,12 @@ object FunctionsAsColumnExt {
 
     def lower(): Column = t(F.lower)
 
-    def regexp_replace(pattern: String, replacement: String): Column = F.regexp_replace(col, pattern,replacement)
+    def regexp_replace(pattern: String, replacement: String): Column =
+      F.regexp_replace(
+        col,
+        pattern,
+        replacement
+      )
 
   }
 

src/main/scala/com/github/mrpowers/spark/daria/sql/transformations.scala

@@ -91,6 +91,36 @@ object transformations {
     }
   }
 
+  /**
+   * snakifies all the columns of a DataFrame
+   *
+   * import com.github.mrpowers.spark.daria.sql.transformations._
+   *
+   * {{{
+   * val sourceDf = Seq(
+   *   ("funny", "joke")
+   * ).toDF("ThIs", "BiH")
+   *
+   * val actualDf = sourceDf.transform(snakeCaseColumns)
+   *
+   * actualDf.show()
+   *
+   * +-----+----+
+   * |th_is|bi_h|
+   * +-----+----+
+   * |funny|joke|
+   * +-----+----+
+   * }}}
+   */
+  def snakifyColumns()(df: DataFrame): DataFrame = {
+    df.columns.foldLeft(df) { (memoDF, colName) =>
+      memoDF.withColumnRenamed(
+        colName,
+        com.github.mrpowers.spark.daria.utils.StringHelpers.snakify(colName)
+      )
+    }
+  }
+
   /**
    * Convert camel case columns to snake case
    * Example: SomeColumn -> some_column

src/main/scala/com/github/mrpowers/spark/daria/utils/StringHelpers.scala

@@ -37,6 +37,29 @@ object StringHelpers {
       .toLowerCase
   }
 
+  /**
+   * Copied from the Lyft Framework https://github.com/lift/framework/blob/master/core/util/src/main/scala/net/liftweb/util/StringHelpers.scala
+   *
+   * Turn a string of format "FooBar" into snake case "foo_bar"
+   *
+   * Note: snakify is not reversible, ie. in general the following will _not_ be true:
+   *
+   * s == camelify(snakify(s))
+   *
+   * @return the underscored string
+   */
+  def snakify(name: String) =
+    name
+      .replaceAll(
+        "([A-Z]+)([A-Z][a-z])",
+        "$1_$2"
+      )
+      .replaceAll(
+        "([a-z\\d])([A-Z])",
+        "$1_$2"
+      )
+      .toLowerCase
+
   def camelCaseToSnakeCase(str: String): String = {
     str
       .replaceAll(

src/test/scala/com/github/mrpowers/spark/daria/sql/TransformationsTest.scala

@@ -124,6 +124,33 @@ object TransformationsTest extends TestSuite with DataFrameComparer with ColumnC
 
     }
 
+    'snakifyColumns - {
+
+      val sourceDF = spark.createDF(
+        List(("funny", "joke")),
+        List(
+          ("ThIs", StringType, true),
+          ("BiH", StringType, true)
+        )
+      )
+
+      val actualDF = sourceDF.transform(transformations.snakifyColumns())
+
+      val expectedDF = spark.createDF(
+        List(("funny", "joke")),
+        List(
+          ("th_is", StringType, true),
+          ("bi_h", StringType, true)
+        )
+      )
+
+      assertSmallDataFrameEquality(
+        actualDF,
+        expectedDF
+      )
+
+    }
+
     'camelCaseToSnakeCaseColumns - {
       "convert camel case columns to snake case" - {
 

src/test/scala/com/github/mrpowers/spark/daria/utils/ArrayHelpersSpec.scala → src/test/scala/com/github/mrpowers/spark/daria/utils/ArrayHelpersTest.scala

@@ -2,7 +2,7 @@ package com.github.mrpowers.spark.daria.utils
 
 import utest._
 
-class ArrayHelpersSpec extends TestSuite {
+object ArrayHelpersTest extends TestSuite {
 
   val tests = Tests {
 

src/test/scala/com/github/mrpowers/spark/daria/utils/StringHelpersTest.scala

@@ -0,0 +1,34 @@
+package com.github.mrpowers.spark.daria.utils
+
+import utest._
+
+object StringHelpersTest extends TestSuite {
+
+  val tests = Tests {
+
+    'escapeForSqlRegexp - {
+      assert(StringHelpers.escapeForSqlRegexp("D/E") == Some("D\\/E"))
+      assert(StringHelpers.escapeForSqlRegexp("(E/F)") == Some("\\(E\\/F\\)"))
+      assert(StringHelpers.escapeForSqlRegexp("") == Some(""))
+      assert(StringHelpers.escapeForSqlRegexp("E|G") == Some("E\\|G"))
+      assert(StringHelpers.escapeForSqlRegexp("E;;G") == Some("E;;G"))
+      assert(StringHelpers.escapeForSqlRegexp("^AB-C") == Some("^AB\\-C"))
+      assert(StringHelpers.escapeForSqlRegexp("^AB+C") == Some("^AB\\+C"))
+      assert(StringHelpers.escapeForSqlRegexp(null) == None)
+    }
+
+    'toSnakeCase - {
+      assert(StringHelpers.toSnakeCase("A b C") == "a_b_c")
+    }
+
+    'snakify - {
+      assert(StringHelpers.snakify("SomeColumn") == "some_column")
+    }
+
+    'camelCaseToSnakeCase - {
+      assert(StringHelpers.camelCaseToSnakeCase("thisIsCool") == "this_is_cool")
+    }
+
+  }
+
+}