Added schema rdd class

pdeyhim · Apr 15, 2014 · b8b904b · b8b904b
1 parent 67ba875
commit b8b904b
Show file tree

Hide file tree

Showing 2 changed files with 9 additions and 0 deletions.
diff --git a/python/pyspark/java_gateway.py b/python/pyspark/java_gateway.py
@@ -64,5 +64,6 @@ def run(self):
     java_import(gateway.jvm, "org.apache.spark.api.java.*")
     java_import(gateway.jvm, "org.apache.spark.api.python.*")
     java_import(gateway.jvm, "org.apache.spark.mllib.api.python.*")
+    java_import(gateway.jvm, "org.apache.spark.sql.api.java.JavaSQLContext")
     java_import(gateway.jvm, "scala.Tuple2")
     return gateway
diff --git a/python/pyspark/rdd.py b/python/pyspark/rdd.py
@@ -1387,6 +1387,14 @@ def _jrdd(self):
     def _is_pipelinable(self):
         return not (self.is_cached or self.is_checkpointed)
 
+class SchemaRDD:
+
+    def __init__(self, pyRDD):
+        self._pyRDD = pyRDD
+        self.ctx = pyRDD.ctx
+        self.sql_ctx = self.ctx._jvm.JavaSQLContext(self.ctx._jsc)
+        self._jrdd = self.ctx._pythonToJava(pyRDD._jrdd)
+
 
 def _test():
     import doctest