[SPARK-5579][SQL][DataFrame] Support for project/filter using SQL expressions

```scala
df.selectExpr("abs(colA)", "colB")
df.filter("age > 21")
```

Author: Reynold Xin <rxin@databricks.com>

Closes #4348 from rxin/SPARK-5579 and squashes the following commits:

2baeef2 [Reynold Xin] Fix Python.
b416372 [Reynold Xin] [SPARK-5579][SQL][DataFrame] Support for project/filter using SQL expressions.

1 files changed, 10 insertions, 0 deletions

diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/SqlParser.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/SqlParser.scala
index 5c006e9d4c..a9bd079c70 100755
--- a/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/SqlParser.scala
+++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/SqlParser.scala
@@ -36,6 +36,16 @@ import org.apache.spark.sql.types._
  * for a SQL like language should checkout the HiveQL support in the sql/hive sub-project.
  */
 class SqlParser extends AbstractSparkSQLParser {
+
+  def parseExpression(input: String): Expression = {
+    // Initialize the Keywords.
+    lexical.initialize(reservedWords)
+    phrase(expression)(new lexical.Scanner(input)) match {
+      case Success(plan, _) => plan
+      case failureOrError => sys.error(failureOrError.toString)
+    }
+  }
+
   // Keyword is a convention with AbstractSparkSQLParser, which will scan all of the `Keyword`
   // properties via reflection the class in runtime for constructing the SqlLexical object
   protected val ABS = Keyword("ABS")