Replace magic lengths with constants in PySpark.

Write the length of the accumulators section up-front rather
than terminating it with a negative length.  I find this
easier to read.

1 files changed, 6 insertions, 0 deletions

diff --git a/python/pyspark/serializers.py b/python/pyspark/serializers.py
index 54fed1c9c7..fbc280fd37 100644
--- a/python/pyspark/serializers.py
+++ b/python/pyspark/serializers.py
@@ -19,6 +19,12 @@ import struct
 import cPickle
 
 
+class SpecialLengths(object):
+    END_OF_DATA_SECTION = -1
+    PYTHON_EXCEPTION_THROWN = -2
+    TIMING_DATA = -3
+
+
 class Batch(object):
     """
     Used to store multiple RDD entries as a single Java object.