Ted Blackman created SPARK-8116:
-----------------------------------
Summary: sc.range() doesn't match python range()
Key: SPARK-8116
URL: https://issues.apache.org/jira/browse/SPARK-8116
Project: Spark
Issue Type: Bug
Components: PySpark
Affects Versions: 1.4.0, 1.4.1
Reporter: Ted Blackman
Priority: Minor
Python's built-in range() and xrange() functions can take 1, 2, or 3 arguments.
Ranges with just 1 argument are probably used the most frequently, e.g.:
for i in range(len(myList)): ...
However, in pyspark, the SparkContext range() method throws an error when
called with a single argument, due to the way its arguments get passed into
python's range function.
There's no good reason that I can think of not to support the same syntax as
the built-in function. To fix this, we can set the default of the sc.range()
method's `stop` argument to None, and then inside the method, if it is None,
replace `stop` with `start` and set `start` to 0, which is what the c
implementation of range() does:
https://github.com/python/cpython/blob/master/Objects/rangeobject.c#L87
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
