python

安装 kudu-python 模块
一定要确保 pip 是最新版本!!!，并且 Cython 已经安装好，安装 kudu-python 的 1.2.0 版本。

  1. sudo pip install --upgrade pip 
  2. 或 [python -m ensurepip、python -m pip install --upgrade pip]
  1. sudo pip install -i https://pypi.douban.com/simple Cython
  2. sudo pip install -i https://pypi.tuna.tsinghua.edu.cn/simple kudu-python==1.2.0

或者

https://kudu.apache.org/docs/developing.html
http://archive.cloudera.com/kudu/redhat/6/x86_64/kudu/5/RPMS/x86_64/

  1. rpm -i http://archive.cloudera.com/kudu/redhat/6/x86_64/kudu/5/RPMS/x86_64/kudu-client-devel-1.4.0+cdh5.12.2+0-1.cdh5.12.2.p0.8.el6.x86_64.rpm
  3. rpm -i http://archive.cloudera.com/kudu/redhat/6/x86_64/kudu/5/RPMS/x86_64/kudu-client0-1.4.0+cdh5.12.2+0-1.cdh5.12.2.p0.8.el6.x86_64.rpm
  5. yum install gcc
  6. yum install gcc-c++
  7. pip install Cython kudu-python==1.2.0
  1. import kudu
  2. from kudu.client import Partitioning
  3. from datetime import datetime
  5. # Connect to Kudu master server
  6. # 连接 kudu master 服务
  7. client = kudu.connect(host='kudu.master', port=7051)
  9. # Define a schema for a new table
  10. # 为表定义一个模式
  11. builder = kudu.schema_builder()
  12. builder.add_column('key').type(kudu.int64).nullable(False).primary_key()
  13. builder.add_column('ts_val', type_=kudu.unixtime_micros, nullable=False, compression='lz4')
  14. schema = builder.build()
  16. # Define partitioning schema
  17. partitioning = Partitioning().add_hash_partitions(column_names=['key'], num_buckets=3)
  19. # Create new table
  20. client.create_table('python-example', schema, partitioning)
  22. # Open a table
  23. table = client.table('python-example')
  25. # Create a new session so that we can apply write operations
  26. session = client.new_session()
  28. # Insert a row
  29. # 往表里面插入一行数据
  30. op = table.new_insert({'key': 1, 'ts_val': datetime.utcnow()})
  31. session.apply(op)
  33. # Upsert a row
  34. op = table.new_upsert({'key': 2, 'ts_val': "2016-01-01T00:00:00.000000"})
  35. session.apply(op)
  37. # Updating a row
  38. op = table.new_update({'key': 1, 'ts_val': ("2017-01-01", "%Y-%m-%d")})
  39. session.apply(op)
  41. # Delete a row
  42. op = table.new_delete({'key': 2})
  43. session.apply(op)
  45. # Flush write operations, if failures occur, capture print them.
  46. try:
  47.     session.flush()
  48. except kudu.KuduBadStatus as e:
  49.     print(session.get_pending_errors())
  51. # Create a scanner and add a predicate
  52. # 先创建一个 scanner，然后再读取表中的数据
  53. # 表中的数据太多，你还需要添加一个所谓的 predicate 只读取规定区间内的数据
  54. scanner = table.scanner()
  55. scanner.add_predicate(table['ts_val'] == datetime(2017, 1, 1))
  57. # Open Scanner and read all tuples
  58. # Note: This doesn't scale for large scans
  59. result = scanner.open().read_all_tuples()

Java

https://www.e-learn.cn/topic/3601334