python连接HBase

环境

启动hbase的Thrift RPC

./hbase-daemon.sh start thrift

生成python的Thrift模块

cd hbase-1.2.1/hbase-thrift/src/main/resources/org/apache/hadoop/hbase/thrift
thrift --gen py Hbase.thrift

#生成gen-py文件
.
├── gen-py
│   ├── hbase
│   │   ├── constants.py
│   │   ├── Hbase.py
│   │   ├── Hbase-remote
│   │   ├── __init__.py
│   │   └── ttypes.py
│   └── __init__.py
└── Hbase.thrift
#把gen-py/hbase复制到项目下

.
├── hbase
│   ├── constants.py
│   ├── Hbase.py
│   ├── Hbase.pyc
│   ├── Hbase-remote
│   ├── __init__.py
│   ├── __init__.pyc
│   ├── ttypes.py
│   └── ttypes.pyc
└── hbase_client.py

# # -*- coding: utf-8 -*-

from thrift.transport import TSocket
from thrift.transport import TTransport
from thrift.protocol import TBinaryProtocol

from hbase import Hbase
from hbase.ttypes import ColumnDescriptor, Mutation

class HbaseClient(object):
def __init__(self, host='localhost', port=9090):
transport = TTransport.TBufferedTransport(TSocket.TSocket(host, port))
protocol = TBinaryProtocol.TBinaryProtocol(transport)
self.client = Hbase.Client(protocol)
transport.open()

def get_tables(self):
"""
获取所有表
"""
return self.client.getTableNames()

def create_table(self, table, *columns):
"""
创建表
"""
self.client.createTable(table, map(lambda column: ColumnDescriptor(column), columns))

def put(self, table, row, columns, attributes=None):
"""
添加记录
@:param columns {"k:1":"11"}
"""
self.client.mutateRow(table, row, map(lambda (k,v): Mutation(column=k, value=v), columns.items()), attributes)

def scan(self, table, start_row="", columns=None, attributes=None):
"""
获取记录
"""

scanner = self.client.scannerOpen(table, start_row, columns, attributes)
while True:
r = self.client.scannerGet(scanner)
if not r:
break
yield dict(map(lambda (k, v): (k, v.value),r[0].columns.items()))

if __name__ == "__main__":
client = HbaseClient("192.168.19.74", 9090)
client.create_table("student", "name", "coruse")
print(client.get_tables())
client.put("student", "1", {"name:":"zhangsan", "coruse:art": "88", "coruse:math": "12"})
client.put("student", "2", {"name:":"lisi", "coruse:art": "90", "coruse:math": "100"})
client.put("student", "3", {"name:":"lisi2"})
for v in client.scan("student", columns=["name"]):
print(v)
for v in client.scan("student"):
print(v)