python happybase 操作 hbase

启动 hbase

bin/start-hbase.sh

启动 thrift 服务

bin/hbase-daemon.sh start thrift

连接 hbase 数据库

import happybase

connection = happybase.Connection(host='qq3', port=9090)

happybase.Connection(host=’localhost’, port=9090, timeout=None, autoconnect=True, table_prefix=None, table_prefix_separator=b’_’, compat=’0.98’, transport=’buffered’, protocol=’binary’)

host：主机名
port：端口
timeout：超时时间
autoconnect：连接是否直接打开
table_prefix：用于构造表名的前缀
table_prefix_separator：用于table_prefix的分隔符
compat：兼容模式
transport：运输模式
protocol：协议

获取所有表名

table_name_list = connection.tables()
print(table_name_list)

[b'ConfigurationManagementGraph', b'actor', b'janusgraph', b'new_actor', b'person']

创建新表：

connection = happybase.Connection(host='qq3', port=9090)
families = {
    'info': dict(),
    'score': dict()
}
connection.create_table('students', families)

table_name_list = connection.tables()
print(table_name_list)

[b'ConfigurationManagementGraph', b'actor', b'janusgraph', b'new_actor', b'person', b'students']

禁用表

connection.disable_table(name)

启用表

connection.enable_table(name)

删除表

connection.delete_table(name,disable=False)

disable:是否先禁用：为True时可直接删除已启用的表
否则应该先禁用再删除

判断表是否被启用 返回一个bool值

connection.is_table_enabled(name)

获取表的所有列簇信息

connection = happybase.Connection(host='qq3', port=9090)
students_table = connection.table('students')
info = students_table.families()
print(info)

{b’info’: {‘name’: b’info:’, ‘max_versions’: 3, ‘compression’: b’NONE’, ‘in_memory’: False, ‘bloom_filter_type’: b’NONE’, ‘bloom_filter_vector_size’: 0, ‘bloom_filter_nb_hashes’: 0, ‘block_cache_enabled’: False, ‘time_to_live’: 2147483647}, b’score’: {‘name’: b’score:’, ‘max_versions’: 3, ‘compression’: b’NONE’, ‘in_memory’: False, ‘bloom_filter_type’: b’NONE’, ‘bloom_filter_vector_size’: 0, ‘bloom_filter_nb_hashes’: 0, ‘block_cache_enabled’: False, ‘time_to_live’: 2147483647}}

插入/修改数据

put(self, row, data, timestamp=None, wal=True)

向表中插入两条数据

connection = happybase.Connection(host='qq3', port=9090)
students_table = connection.table('students')
data_1 = {"info:name":'lucy',"info:age":"16","info:gender":"M","score:Math":"87","score:Chinese":"78","score:English":"80"}
data_2 = {"info:name":'jack',"info:age":"15","info:gender":"F","score:Math":"90","score:Chinese":"75","score:English":"78"}
students_table.put('2019030101',data_1)
students_table.put('2019030102',data_2)

查询单条数据 返回结果为字典

row(self, row, columns=None, timestamp=None, include_timestamp=False)

connection = happybase.Connection(host='qq3', port=9090)
students_table = connection.table('students')
data_1 = students_table.row('2019030101')
print(data_1)

{b'info:age': b'16', b'info:gender': b'M', b'info:name': b'lucy', b'score:Chinese': b'78', b'score:English': b'80', b'score:Math': b'87'}

查询多条数据 返回结果为列表

rows(self, rows, columns=None, timestamp=None,include_timestamp=False)

connection = happybase.Connection(host='qq3', port=9090)
students_table = connection.table('students')
data = students_table.rows(rows=['2019030101','2019030102'],columns=['info:name','score:English'])
print(data)

[(b'2019030101', {b'info:name': b'lucy', b'score:English': b'80'}), (b'2019030102', {b'info:name': b'jack', b'score:English': b'78'})]

scan 查看数据 scan生成的是一个迭代器，可用next()取数据，也可遍历取数据

scan(self, row_start=None, row_stop=None, row_prefix=None, columns=None, filter=None, timestamp=None, include_timestamp=False, batch_size=1000, scan_batching=None, limit=None, sorted_columns=False, reverse=False)

row_start：起始行，默认None，即第一行，可传入行号指定从哪一行开始
row_stop：结束行，默认None，即最后一行，可传入行号指定到哪一行结束(不获取此行数据)
row_prefix：行号前缀，默认为None，即不指定前缀扫描，可传入前缀来扫描符合此前缀的行
columns：列，默认为None，即获取所有列，可传入一个list或tuple来指定获取列
filter：过滤字符串
timestamp：时间戳。默认为None，即返回最大的那个时间戳的数据。可传入一个时间戳来获取小于此时间戳的最大时间戳的版本数据
include_timestamp：是否返回时间戳数据，默认为False
batch_size：用于检索结果的批量大小
scan_batching：服务端扫描批处理
limit：数量
sorted_columns：是否返回排序的列(根据行名称排序)
reverse：是否执行反向扫描

connection = happybase.Connection(host='qq3', port=9090)
students_table = connection.table('students')
res = students_table.scan()

print(next(res))
for i in res:
    print(i)

b'2019030101', {b'info:age': b'16', b'info:gender': b'M', b'info:name': b'lucy', b'score:Chinese': b'78', b'score:English': b'80', b'score:Math': b'87'})
(b'2019030102', {b'info:age': b'15', b'info:gender': b'F', b'info:name': b'jack', b'score:Chinese': b'75', b'score:English': b'78', b'score:Math': b'90'})

批量插入数据

batch(self, timestamp=None, batch_size=None, transaction=False, wal=True)

connection = happybase.Connection(host='qq3', port=9090)
students_table = connection.table('students')
bat = students_table.batch()
bat.put('2019030103', {"info:name":'lili',"info:age":"15","info:gender":"F","score:Math":"82","score:Chinese":"73","score:English":"81"})
bat.put('2019030104', {"info:name":'rose',"info:age":"18","info:gender":"F","score:Math":"77","score:Chinese":"80","score:English":"88"})
bat.put('2019030105', {"info:name":'tom',"info:age":"16","info:gender":"M","score:Math":"87","score:Chinese":"73","score:English":"92"})
bat.send()
res = students_table.scan()

for i in res:
    print(i)

(b'2019030101', {b'info:age': b'16', b'info:gender': b'M', b'info:name': b'lucy', b'score:Chinese': b'78', b'score:English': b'80', b'score:Math': b'87'})
(b'2019030102', {b'info:age': b'15', b'info:gender': b'F', b'info:name': b'jack', b'score:Chinese': b'75', b'score:English': b'78', b'score:Math': b'90'})
(b'2019030103', {b'info:age': b'15', b'info:gender': b'F', b'info:name': b'lili', b'score:Chinese': b'73', b'score:English': b'81', b'score:Math': b'82'})
(b'2019030104', {b'info:age': b'18', b'info:gender': b'F', b'info:name': b'rose', b'score:Chinese': b'80', b'score:English': b'88', b'score:Math': b'77'})
(b'2019030105', {b'info:age': b'16', b'info:gender': b'M', b'info:name': b'tom', b'score:Chinese': b'73', b'score:English': b'92', b'score:Math': b'87'})

删除数据 可以整条删除 或者 删除某些column

delete(self, row, columns=None, timestamp=None, wal=True)

connection = happybase.Connection(host='qq3', port=9090)
students_table = connection.table('students')
students_table.delete('2019030101',['info:name','score:English'])
data_1 = students_table.row('2019030101')
print(data_1)
students_table.delete('2019030101')
data_1 = students_table.row('2019030101')
print(data_1)

{b'info:age': b'16', b'info:gender': b'M', b'score:Chinese': b'78', b'score:Math': b'87'}
{}

使用with来管理batch 写入或者删除

connection = happybase.Connection(host='qq3', port=9090)
students_table = connection.table('students')
with students_table.batch() as bat:
    bat.delete('2019030102')
    bat.delete('2019030103')
    bat.delete('2019030104')
    bat.put('2019030106',{"info:name":'tims',"info:age":"16","info:gender":"M","score:Math":"87","score:Chinese":"73","score:English":"92"})

res = students_table.scan()
for i in res:
    print(i)

(b'2019030105', {b'info:age': b'16', b'info:gender': b'M', b'info:name': b'tom', b'score:Chinese': b'73', b'score:English': b'92', b'score:Math': b'87'})
(b'2019030106', {b'info:age': b'16', b'info:gender': b'M', b'info:name': b'tims', b'score:Chinese': b'73', b'score:English': b'92', b'score:Math': b'87'})