S3 List Status ad-infinitude

s3ListStatus.txt

...
15/05/05 21:43:26 INFO s3n.S3NativeFileSystem: listStatus s3n://kinesis-click-stream-us-east-1-bucket-4myg4if5x7au/events/2015/03/21/20 with recursive false
15/05/05 21:43:26 INFO s3n.S3NativeFileSystem: listStatus s3n://kinesis-click-stream-us-east-1-bucket-4myg4if5x7au/events/2015/03/21/21 with recursive false
15/05/05 21:43:26 INFO s3n.S3NativeFileSystem: listStatus s3n://kinesis-click-stream-us-east-1-bucket-4myg4if5x7au/events/2015/03/21/22 with recursive false
15/05/05 21:43:26 INFO s3n.S3NativeFileSystem: listStatus s3n://kinesis-click-stream-us-east-1-bucket-4myg4if5x7au/events/2015/03/21/23 with recursive false
15/05/05 21:43:26 INFO s3n.S3NativeFileSystem: listStatus s3n://kinesis-click-stream-us-east-1-bucket-4myg4if5x7au/events/2015/03/21/00/01 with recursive false
15/05/05 21:43:26 INFO s3n.S3NativeFileSystem: listStatus s3n://kinesis-click-stream-us-east-1-bucket-4myg4if5x7au/events/2015/03/21/00/02 with recursive false
15/05/05 21:43:26 INFO s3n.S3NativeFileSystem: listStatus s3n://kinesis-click-stream-us-east-1-bucket-4myg4if5x7au/events/2015/03/21/00/04 with recursive false
15/05/05 21:43:26 INFO s3n.S3NativeFileSystem: listStatus s3n://kinesis-click-stream-us-east-1-bucket-4myg4if5x7au/events/2015/03/21/00/05 with recursive false
...

S3Puller.scala

    import com.amazonaws.services.s3._, model._
    import com.amazonaws.auth.BasicAWSCredentials

    val request = new ListObjectsRequest()
    request.setBucketName(bucket)
    request.setPrefix(prefix)
    request.setMaxKeys(pageLength)
    def s3 = new AmazonS3Client(new BasicAWSCredentials(key, secret))

    val objs = s3.listObjects(request) // Note that this method returns truncated data if longer than the "pageLength" above. You might need to deal with that.
    sc.parallelize(objs.getObjectSummaries.map(_.getKey).toList)
        .flatMap { key => Source.fromInputStream(s3.getObject(bucket, key).getObjectContent: InputStream).getLines }

I have many *tar.gz files on S3. May I know if I have to user Spark Streaming to process the data? I don't get the result when I directly use the input stream provided by S3.