python DataFrame merge 函数 合并两个DataFrame（与sql联系）

pd.merge(df1,df2,on="class_id",how="right")

创建两个dataframe

temp_dict1 = [{"id":1,"name":"cyw","age":20,"class_id":1},
             {"id":2,"name":"xiaogang","age":18,"class_id":2},
            {"id":2,"name":"老王","age":18,"class_id":3},
            {"id":2,"name":"张三","age":18,"class_id":2},
            {"id":2,"name":"李四","age":18,"class_id":3},
            {"id":2,"name":"赵子龙","age":18,"class_id":4}]

temp_dict2 = [{"class_id":1,"class":"一班"},
             {"class_id":2,"class":"二班"},
              {"class_id":3,"class":"三班"},
              {"class_id":4,"class":"四班"},
              {"class_id":5,"class":"五班"}
            ]
classes = pd.DataFrame(temp_dict2)
students = pd.DataFrame(temp_dict1)

内连接（取两个表中交集）

pd.merge(df1,df2,on="class_id") 

相当与在sql中

select * from students inner join classes on students.class_id = classes.class_id;

左连接（获取左表所有记录，即使右表没有对应匹配的记录,没有匹配到的数据将用NAN替代）

pd.merge(df1,df2,on="class_id",how="left")
#等价于pd.merge(df1,df2,left_on="class_id",right_on="class_id",how="left")

相当sql中

select * from students left join classes on students.class_id = classes.class_id;

 

 右连接（与 LEFT JOIN 相反，用于获取右表所有记录，即使左表没有对应匹配的记录，没有匹配到的数据将用NAN替代） 

pd.merge(df1,df2,on="class_id",how="right")
#等价于pd.merge(df1,df2,left_on="class_id",right_on="class_id",how="right")

相当sql中

select * from students right join classes on students.class_id = classes.class_id;