[SPARK-12834] Use type conversion instead of Ser/De of Pickle to transform JavaArray and JavaList

• Type: Improvement
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 1.6.1, 2.0.0
• Component/s: PySpark
• Target Version/s: 1.6.1, 2.0.0
• Assignee: Xusen Yin
• Reporter: Xusen Yin
• Created: 15/Jan/16 09:23
• Updated: 27/Jan/16 08:33
• Resolved: 26/Jan/16 06:42

Description

According to the Ser/De code in Python side:

StringIndexerModel

  def _java2py(sc, r, encoding="bytes"):
    if isinstance(r, JavaObject):
        clsName = r.getClass().getSimpleName()
        # convert RDD into JavaRDD
        if clsName != 'JavaRDD' and clsName.endswith("RDD"):
            r = r.toJavaRDD()
            clsName = 'JavaRDD'

        if clsName == 'JavaRDD':
            jrdd = sc._jvm.SerDe.javaToPython(r)
            return RDD(jrdd, sc)

        if clsName == 'DataFrame':
            return DataFrame(r, SQLContext.getOrCreate(sc))

        if clsName in _picklable_classes:
            r = sc._jvm.SerDe.dumps(r)
        elif isinstance(r, (JavaArray, JavaList)):
            try:
                r = sc._jvm.SerDe.dumps(r)
            except Py4JJavaError:
                pass  # not pickable

    if isinstance(r, (bytearray, bytes)):
        r = PickleSerializer().loads(bytes(r), encoding=encoding)
    return r

We use SerDe.dumps to serialize JavaArray and JavaList in PythonMLLibAPI, then deserialize them with PickleSerializer in Python side. However, there is no need to transform them in such an inefficient way. Instead of it, we can use type conversion to convert them, e.g. list(JavaArray) or list(JavaList). What's more, there is an issue to Ser/De Scala Array as I said in https://issues.apache.org/jira/browse/SPARK-12780

Issue Links

• links to: [Github] Pull Request #10772 (yinxusen)
• links to: [Github] Pull Request #10941 (jkbradley)