在大数据时代下，和数据打打交道是家常便饭。那么常用储存数据的一种方式：数据库，用起来那也是相当的得心应手，今天就用python连接各类常见数据库！

1.sqlite

非常轻量的关系型数据库，不需要安装服务端，解压即用

import sqlite3
import pandas as pd

conn = sqlite3.connect('test.db')
df = pd.read_sql_query('select * from company', con=conn)

print(df)

2.mysql

最受欢迎、使用最多的关系型数据库之一

import pandas as pd
from sqlalchemy import create_engine

mysql_url = 'mysql+pymysql://root:root@localhost:3306/test?charset=utf8'
engine = create_engine(mysql_url)
data = pd.read_sql_table('user1', engine)

print(data)

3.postgresql

对象关系型数据库，开源人士自行研发的数据库

import pandas as pd
from sqlalchemy import create_engine

postgres_url = 'postgres://postgres:root@localhost:5432/db1'
engine = create_engine(postgres_url)
data = pd.read_sql_table('company', engine)

print(data)

4.mongodb

文档型数据库，将数据已一个个json文档格式进行存储

from pymongo import MongoClient
import pandas as pd

client = MongoClient('localhost', 27017)
collection = client['test']['user']
data = collection.find()
df = pd.DataFrame(list(data))

print(df)

5.redis

缓存数据库，快，并且支持五大数据类型

import asyncio
from aredis import StrictRedis


async def example():
    client = StrictRedis(host='127.0.0.1', port=6379, db=0)
    await client.flushdb()
    uid = await client.get('id')
    uname = await client.get('name')
    uage = await client.get('age')
    print(f'user:id={uid},name={uname},age={uage}')


loop = asyncio.get_event_loop()
loop.run_until_complete(example())
loop.close()

6.hive

数据仓库工具，可以把HDFS上的结构化数据映射成表，以供查询

import pandas as pd
from pyhive import hive

conn = hive.Connection(
    host='localhost',
    port=10000,
    database='test',
    username='hive',
    password=''
)
data = pd.read_sql_query('select * from test', con=conn)

print(data)

7.clickhouse

列示存储数据库，可用于构建在hive之上的一个在线查询分析，为了弥补hive查询速度慢的劣势，我们公司已经在用了

from clickhouse_driver import Client

client = Client(host='127.0.0.1', port='9000', user='ck_user', password='ck_pass')
sql = 'select * from db_name.tb_name limit 0, 10'
ans = client.execute(sql)

8.habse

也是列示存储数据库，通过rowkey，类似于主键来查询数据，能够因对超大数据的挑战

import happybase

connection = happybase.Connection('localhost', autoconnect=False)
connection.open()

print(connection.tables())
connection.close()