check empty df

tests/spark_etl/integration/test_spark_connector.py

@@ -97,7 +97,11 @@ def getSparkDF(inpath, outpath, spark, schemafile):
     output_df = regex_df
 
     # Let's now write to the output dir
-    cluster_utils.write_spark_df(output_df, outpath)
+    df_empty = (len(output_df.head(1))==0)
+    print("Is the dataframe empty?  %s" % df_empty)
+    if not df_empty:
+        print("Saving dataframe")
+        cluster_utils.write_spark_df(output_df,outpath)
     return output_df