Python实现SQL模拟功能

博客核心内容：

---

1. Python实现insert插入功能
2. Python实现delete删除功能（remove、rename）
3. Python实现update更新功能（remove、rename）
4. Python实现：将列表中的内容对号入座的放到相应的字典当中（警报场景1）
5. Python实现：where列表中的内容进行再处理（警报场景2）
6. Python实现：用where条件去过滤单行记录
7. Python实现：如何获取单行记录中指定的字段数据（select id,name）
8. Python实现select查询功能
9. Python实现select查询功能（我自己实现的：带截图）

---

1、Python实现insert插入功能

插入之前db1.emp文件中的内容示例：

1,张德,25,18330267967,IT,2017-5-272,李四,25,19230267967,运维,2016-5-273,王五,35,15673026796,IT,2017-6-274,赵六,29,18122263467,测试,2017-5-285,董七,25,18556267967,运维6,2014-5-276,李张明,26,18122677967,测试,2017-5-217,赵四,28,18523267967,运维6,2014-5-228,董七,25,18556267967,运维6,2014-5-27

核心代码示例：

#!/usr/bin/python# -*- coding:utf-8 -*-def insert(sql):    """    function:该函数的功能是进行文本插入操作    :param sql: 用户传进来的sql语句    :return: 无    """    sql_list = sql.split()    db_name,table_name = sql_list[2].split(".")    #获取最后一行代码的ID    last_id = ""    with open("%s/%s"%(db_name,table_name),mode="r",encoding="utf-8") as fr:        content_list = fr.readlines()        last_record = content_list[-1]        last_id = last_record.split(",")[0]    #拼出最后一行记录,在操作的过程中注意insert的返回值是None    record_list = sql_list[-1].split(",")    record_list.insert(0,str(int(last_id)+1))    record_line = ",".join(record_list)    #向列表中写入记录    with open("%s/%s"%(db_name, table_name), mode="a", encoding="utf-8") as fw:        fw.write("\n"+record_line)        #记住要刷新纪录        fw.flush()if __name__ == '__main__':    while True:        sql = input("sql> ").strip()        insert(sql)

运行代码：

sql> insert into db1.emp values 张东,25,18556267367,运维6,2014-5-28sql> insert into db1.emp values 刘恩,25,18556227967,运维6,2014-4-27

插入之后db1.emp文件的内容示例：

1,张德,25,18330267967,IT,2017-5-272,李四,25,19230267967,运维,2016-5-273,王五,35,15673026796,IT,2017-6-274,赵六,29,18122263467,测试,2017-5-285,董七,25,18556267967,运维6,2014-5-276,李张明,26,18122677967,测试,2017-5-217,赵四,28,18523267967,运维6,2014-5-228,董七,25,18556267967,运维6,2014-5-279,张东,25,18556267367,运维6,2014-5-2810,刘恩,25,18556227967,运维6,2014-4-27

后续可以在拼接最后一行记录的时候进行优化，从而替换掉21/22/23三行代码.

    #拼出最后一条记录    last_record = last_id+","+sql_list[-1]

2、Python实现delete删除功能

文件修改的思路:
1、读源文件的内容,然后一行一行的向新文件中写.
2、通过os模块将源文件给删除.
3、将新文件重命名为源文件的名字.
源文件删除之前的情况：

1,张德,25,18330267967,IT,2017-5-272,李四,25,19230267967,运维,2016-5-273,王五,35,15673026796,IT,2017-6-274,赵六,29,18122263467,测试,2017-5-285,董七,25,18556267967,运维6,2014-5-276,李张明,26,18122677967,测试,2017-5-217,赵四,28,18523267967,运维6,2014-5-228,董七,25,18556267967,运维6,2014-5-27

代码示例：

#!/usr/bin/python# -*- coding:utf-8 -*-import osdef delete(sql):    """    function:该函数的功能是删除表中指定的行,此处我的where条件就是=号的方式    :param sql:用户传进来的SQL语句     :return: 返回值是删除的记录条数和删除是否成功    """    sql_list = sql.split()    db_name,table_name = sql_list[2].split(".")    #生成备份文件的名字    back_name = table_name+"_bak"    #获取判断字段    field,value = sql_list[4],sql_list[6]    title = "id,name,age,phone,dept,enroll_date"    del_count = 0    with open("%s/%s"%(db_name,table_name),"r",encoding="utf-8") as fr,\        open("%s/%s"%(db_name,back_name),"w",encoding="utf-8") as fw:        #只有在第一次打开文件的时候才会清空整个文件的内容        for line in fr.readlines():            user_dict = dict(zip(title.split(","),line.strip().split(",")))            if user_dict.get(field) == value:                del_count += 1            else:                fw.write(line)                fw.flush()    #在整个操作结束之后,删除源文件,并对新文件进行重命名    os.remove("%s/%s"%(db_name,table_name))    os.rename("%s/%s"%(db_name,back_name),"%s/%s"%(db_name,table_name))    return [[del_count],["delete successful"]]if __name__ == '__main__':    while True:        sql = input(">").strip()        log_list = delete(sql)        print(log_list)

运行示例：

>delete from db1.emp where id = 2[[1], ['delete successful']]>delete from db1.emp where name = 李四[[0], ['delete successful']]>delete from db1.emp where name = 赵四[[1], ['delete successful']]

删除文件之后的文件内容：

1,张德,25,18330267967,IT,2017-5-273,王五,35,15673026796,IT,2017-6-274,赵六,29,18122263467,测试,2017-5-285,董七,25,18556267967,运维6,2014-5-276,李张明,26,18122677967,测试,2017-5-218,董七,25,18556267967,运维6,2014-5-27

3、Python实现update更新功能

update更新之前文件中的内容：

1,张德,25,18330267967,IT,2017-5-272,李四,25,19230267967,运维,2016-5-273,王五,35,15673026796,IT,2017-6-274,赵六,29,18122263467,测试,2017-5-286,李张明,26,18122677967,测试,2017-5-217,赵六,28,18523267967,运维6,2014-5-228,Alex,25,18556267967,运维6,2014-5-27

代码示例：

#!/usr/bin/python# -*- coding:utf-8 -*-import osdef update(sql):    """    function:用于进行记录插入操作    :param sql: 用户传进来的sql语句    :return: 日志列表    """    sql_list = sql.split()    db_name,table_name = sql_list[1].split(".")    #创建备份文件的名字    back_name = table_name+"_bak"    print("sql_list is \033[41;1m%s\033[0m"%sql_list)    change_field = sql_list[3]    change_value = sql_list[5]    where_field = sql.partition("where")[2].split("=")[0].strip()    where_value = sql.partition("where")[2].split("=")[1].strip()    title = "id,name,age,phone,dept,enroll_date"    update_count = 0    with open("%s/%s"%(db_name,table_name),"r",encoding="utf-8") as fr, \        open("%s/%s"%(db_name,back_name),"w",encoding="utf-8") as fw:        for line in fr.readlines():            #拼出用户的相关信息            user_info = dict(zip(title.split(","),line.strip().split(",")))            if user_info[where_field] == where_value:                user_info[change_field] = change_value.strip("'")                #构建插入的记录                line_list = []                #下面这一步骤值得学习:先转化成list,在拼接成字符串                for i in title.split(","):                    line_list.append(user_info.get(i))                line_record = ",".join(line_list)                fw.write(line_record+"\n")                update_count += 1            else:                fw.write(line)        #所有操作完之后之后再刷新一下缓冲区        fw.flush()    os.remove("%s/%s" % (db_name, table_name))    os.rename("%s/%s" % (db_name, back_name), "%s/%s" % (db_name, table_name))    return [[update_count], ["update successful"]]if __name__ == '__main__':    sql = input(">").strip()    log_list = update(sql)    print(log_list)

运行代码：

>update db1.emp set name = 'sb'   where name = Alexsql_list is ['update', 'db1.emp', 'set', 'name', '=', "'sb'", 'where', 'name', '=', 'Alex'][[0], ['update successful']]

运行后文件的内容：

1,张德,25,18330267967,IT,2017-5-272,李四,25,19230267967,运维,2016-5-273,王五,35,15673026796,IT,2017-6-274,赵六,29,18122263467,测试,2017-5-286,李张明,26,18122677967,测试,2017-5-217,赵六,28,18523267967,运维6,2014-5-228,sb,25,18556267967,运维6,2014-5-27

4、Python实现：将列表中的内容对号入座的放到相应的字典当中（警报场景1）

代码示例：

#!/usr/bin/python# -*- coding:utf-8 -*-def handle_parse(sql_list):    """    function:该函数的功能是将列表中的内容对号入座的放到相应的字典当中    :param sql: 用户传进来的列表形式的数据    :return: sql_dict数据    """    print("sql_list is %s"%sql_list)    sql_dict = {        "select":[],        "from":[],        "where":[],        "limit":[]    }    tag = False    key = ""    for item in sql_list:        if tag and item in sql_dict:            tag = False        if not tag and item in sql_dict:            tag = True            key = item            continue        if tag:            sql_dict[key].append(item)    return sql_dictif __name__ == '__main__':    sql_list = input(">").strip().split()    sql_dict = handle_parse(sql_list)    print("sql_dict is %s"%sql_dict)

运行结果：

>select id,name  from db1.emp where not id >= 3 limit 2sql_list is ['select', 'id,name', 'from', 'db1.emp', 'where', 'not', 'id', '>=', '3', 'limit', '2']sql_dict is {'from': ['db1.emp'], 'where': ['not', 'id', '>=', '3'], 'limit': ['2'], 'select': ['id,name']}Process finished with exit code 0

除了select之外，insert、update、delete等操作都可以使用这种方式对号入座。

5、Python实现：where列表中的内容进行再处理（警报场景2）

代码示例：

#!/usr/bin/python# -*- coding:utf-8 -*-def handle_parse(where_list):    """    function:本函数的功能是对sql_list中的数据进行进一步的处理    :param sql_list: ['id', '>', '10', 'and', 'id', '<', '14', 'or', 'name', 'like', '李']    :return: [['id', '>', '10'], 'and', ['id', '<', '14'], 'or', ['name', 'like', '李']]    """    sens_list = ['and','or','not']    where_result = []    temp = []    tag = False    sens_word = ""    for item in where_list:        if item in sens_list:            tag = True            if temp:where_result.append(temp)            temp = []            sens_word = item        if not tag and item not in sens_list:            temp.append(item)        if tag and item not in sens_list:            tag = False            where_result.append(sens_word)            sens_word = ""            temp.append(item)    else:        where_result.append(temp)    return where_resultif __name__ == '__main__':    where_list = ['id', '>', '10', 'and', 'id', '<', '14', 'or', 'name', 'like', '李']    print(where_list)    where_list = handle_parse(where_list)    print("%s"%where_list)

运行结果：

['id', '>', '10', 'and', 'id', '<', '14', 'or', 'name', 'like', '李'][['id', '>', '10'], 'and', ['id', '<', '14'], 'or', ['name', 'like', '李']]

问题注意：temp = []不要写成list.clear（temp）

6、Python实现用where条件去过滤单行记录

#!/usr/bin/python# -*- coding:utf-8 -*-def logic_action(user_dict,where_list):    """    function:通过where列表中的条件对文本进行过滤操作    :param user_dict: 用户的记录字典形式：单行文本    :param where_list: where条件列表形式[['id', '>', '10'], 'and', ['id', '<', '14'], 'or', ['name', 'like', '李']]    :return: 返回逻辑值真或假    """    logic_result = []    for item in where_list:        if (type(item) == list):            field,label,value = item            line_value = user_dict[field]            if label == "=":                label = "=="            if value.isdigit():                value = int(value)                line_value = int(line_value)            if label != 'like':                item = str(eval("%s%s%s" %(line_value,label,value)))            else:                if value in line_value:                    item = 'True'                else:                    item = 'False'        logic_result.append(item)    logic_str = ' '.join(logic_result)    print(logic_str)    logic_res = eval("%s"%logic_str)    return logic_resif __name__ == '__main__':    user_dict = {'enroll_date': '2014-5-27', 'dept': '运维6', 'name': '张明', 'age': '25', 'id': '1', 'phone': '18556267967'}    where_list =  [['age', '=', '24'], 'and', ['id', '<', '14'], 'and', ['name', 'like', '张']]    logic_res = logic_action(user_dict, where_list)    print("logic_res is %s"%logic_res)

运行结果：

False and True and Truelogic_res is FalseProcess finished with exit code

7、如何获取当行记录中指定的字段数据

代码示例：

#!/usr/bin/python# -*- coding:utf-8 -*-"""function:获取文本当中指定的字段数值"""if __name__ == '__main__':    db_name = "db1"    table_name = "emp"    title = "id,name,age,phone,dept,enroll_data"    fields_list = ["age","phone","dept"]    final_result = []    with open("%s/%s"%(db_name,table_name),"r",encoding="utf-8") as fr:        for line in fr.readlines():            #通过zip形式处理单行记录            user_info = dict(zip(title.split(","),line.strip().split(",")))            temp = []            for field in fields_list:                temp.append(user_info[field])            final_result.append(temp)    for record in final_result:        print(record)

运行结果：

['25', '18330267967', 'IT']['25', '19230267967', '运维']['35', '15673026796', 'IT']['29', '18122263467', '测试']['25', '18556267967', '运维6']['26', '18122677967', '测试']['28', '18523267967', '运维6']['25', '18556267967', '运维6']Process finished with exit code 0

8、Python实现select查询功能

代码示例：

#!/usr/bin/python# -*- coding:utf-8 -*-import osdef select_parse(sql):    """    function:该函数的功能是进行select查询操作    :param sql: 用户传进来的sql语句    :return: 返回操作日志    """    #定义一个存储结构,以字典的形式进行存储    sql_list = sql.split()    #用一个字典结构去存储相应的数据，方便处理数据    sql_dict = handle_parse(sql_list)    return sql_dictdef handle_parse(sql_list):    """    function:用字典的形式去存储数据    :param sql_list: 用户传进来的sql_list列表    :return: 返回字典形式的数据    """    sql_dict = {        "select":[],        "from":[],        "where":[],        "limit":[]    }    # 接下来就是如何将列表中的内容对号入座的放到相应的字典当中，利用到了警报    tag = False    key = ""    for item in sql_list:        if tag and item in sql_dict:            tag = False        if not tag and item in sql_dict:            tag = True            key = item            continue        if tag:            sql_dict[key].append(item)    #此时我们需要对where中的条件进行一步的清洗处理    #现在的where条件：['id', '>', '10', 'and', 'id', '<', '14', 'or', 'name', 'like', '李']    #需要处理成的形式:[['id', '>', '10'], 'and', ['id', '<', '14'], 'or', ['name', 'like', '李']]    #需要处理成的形式:['not', ['id', '>=', '3']]    if  sql_dict['where']:        sql_dict['where'] = where_parse(sql_dict['where'])    return sql_dictdef where_parse(where_list):    """    function:该函数的作用是对where中的条件进行进一步的处理    :param sql_dict: ['id', '>', '10', 'and', 'id', '<', '14', 'or', 'name', 'like', '李']    :return: ['not', ['id', '>=', '3']]    """    sens_list = ['and','or','not']    where_result = []    temp = []    opt = ""    tag = False    for item in where_list:        if item in sens_list:            tag = True            #通过if条件来限制not这种情况            if temp:                where_result.append(temp)            temp = []   #注意：此处不能用list.clear(temp)            opt = item        if not tag and item not in sens_list:            temp.append(item)        if tag and item not in sens_list:            tag = False            where_result.append(opt)            opt = ""            temp.append(item)    else:        where_result.append(temp)    return where_resultdef select(sql_dict):    """    function：进一步实现查询功能    :param sql_dict: 用户传进来的字典形式的数据    :return: 日志列表    """    #1、先通过from获取对应的库名和表名    db_name,table_name = sql_dict["from"][0].split(".")    fr = open("%s/%s"%(db_name,table_name),"r",encoding="utf-8")    #2、我们去解析where语句中的内容,(这是重点)    where_list = where_action(fr,sql_dict["where"])    #3、获取指定个数的行文本    limit_list = limit_action(where_list,sql_dict.get("limit"))    #4、获取指定字段个数的信息    final_result = search_action(limit_list,sql_dict.get("select"))    return final_resultdef search_action(limit_list,select_list):    """    function:本函数的功能是获取指定字段个数的数据    :param limit_list: list列表    :param select_list: [id,name]    :return: 返回一个指定字段的集合    """    #final_result用来存放最终的结果集合    final_result = []    title = "id,name,age,phone,dept,enroll_data"    #先考虑*的情况    if select_list[0] == "*":        final_result = limit_list    else:        fields_list = select_list[0].split(",")        for line in limit_list:            user_info = dict(zip(title.split(","),line.split(",")))            #用一个临时列表去存放中间数据            r_l = []            for field in fields_list:                r_l.append(user_info[field])            final_result.append(r_l)    return final_resultdef limit_action(where_list,limit_list):    """    function:获取指定个数的行文本：    :param where_list: ['2,李四,25,19230267967,运维,2016-5-27\n', '6,李张明,26,18122677967,测试,2017-5-21\n']    :param limit_num: 数值    :return:     """    limit_res = []    if  limit_list:        where_list  = where_list[0:int(limit_list[0])]    return where_listdef where_action(fr,where_list):    """    function：通过where中的条件对文件中的内容进行过滤    :param fr: 文件的句柄    :param where_list: where列表中的内容    :return: 返回一个列表的内容    """    #res_list用于存放结果集    res_list = []    #先考虑是否含有where条件    if  where_list:        title = "id,name,age,phone,dept,enroll_date"        for line in fr.readlines():            user_dict = dict(zip(title.split(","),line.strip().split(",")))            #用logic_action            logic_res = logic_action(user_dict,where_list)            if logic_res:                res_list.append(line)    #在考虑无where条件    else:        res_list = fr.readlines()    return res_listdef logic_action(user_dict,where_list):    """    function:通过where列表中的条件对文本进行过滤操作    :param user_dict: 用户的记录字典形式：单行文本    :param where_list: where条件列表形式[['id', '>', '10'], 'and', ['id', '<', '14'], 'or', ['name', 'like', '李']]    :return: 返回逻辑值真或假    """    logic_result = []    for item in where_list:        if (type(item) == list):            field,label,value = item            line_value = user_dict[field]            if label == "=":                label = "=="            if value.isdigit():                value = int(value)                line_value = int(line_value)            if label != 'like':                item = str(eval("%s%s%s" %(line_value,label,value)))            else:                if value in line_value:                    item = 'True'                else:                    item = 'False'        #如果不是列表类型,就是敏感词汇，此时直接加入到logic_result当中        logic_result.append(item)    logic_str = ' '.join(logic_result)    logic_res = eval("%s"%logic_str)    return logic_resif __name__ == '__main__':    while True:        sql = input(">").strip()        #此时的sql语句已经被解析成字典形式        sql_dict = select_parse(sql)        final_result = select(sql_dict)        for line in final_result:            print(line)

运行结果示例：

>select * from db1.emp where id > 56,李张明,26,18122677967,测试,2017-5-217,赵四,28,18523267967,运维6,2014-5-228,董七,25,18556267967,运维6,2014-5-27>select id,name from db1.emp where id > 5['6', '李张明']['7', '赵四']['8', '董七']

9、Python实现select查询功能

主要功能模块：

代码示例：

#!/usr/bin/python# -*- coding:utf-8 -*-import osdef select_parse(sql):    """    function:解析成字典形式    :param sql:     :return:     """    sql_dict = {        "select":[],        "from":[],        "where":[],        "limit":[],    }    sql_list = sql.split()    sens_list = ["select","from","where","limit"]    tag = False    key_word = ""    for item in sql_list:        if tag and item in sens_list:            tag = False            key_word = ""        if not tag and item in sens_list:            tag = True            key_word = item            continue        if tag:            sql_dict[key_word].append(item)    return sql_dictdef select(sql_dict):    """    function:本函数的功能是对实现select解析功能    :param sql_dict: {'select': ['*'], 'where': ['id', '>', '5'], 'from': ['db1.emp'], 'limit': ['2']}    :return: 返回最终的结果集合    """    #1、获取数据库名和表名    db_name,table_name = sql_dict["from"][0].split(".")    #2、获取整个文件的内容    content_list =[]    with open("%s/%s"%(db_name,table_name),"r",encoding="utf-8") as fr:        content_list = fr.readlines()    #3、对where中的条件进行进一步的处理: [['id', '>', '10'], 'and', ['id', '<', '14'], 'or', ['name', 'like', '李']]    sql_dict["where"] = where_parse(sql_dict["where"])    #4、通过where条件对content_list表中的内容进行第一次过滤    where_result = handle_where(content_list,sql_dict["where"])    #5、通过limit条件对where_result中的条件进行进一步的过滤    limit_result = limit_parse(where_result,sql_dict["limit"])    #6、输出select函数中指定的字段信息    final_result = select_handle(limit_result,sql_dict['select'])    return final_resultdef select_handle(limit_result,select_list):    """    function:获取指定的字段信息    :param limit_result: limit中的列表    :param select_list: 指定字段或者*    :return: 返回一个列表    """    #先考虑*的情况    final_result = []    title = "id,name,age,phone,dept,enroll_date"    if select_list[0] == "*":        return limit_result    else:        for item in limit_result:            temp = []            user_info = dict(zip(title.split(","),item.split(",")))            for item in select_list[0].split(","):                temp.append(user_info[item])            final_result.append(temp)        return final_resultdef limit_parse(where_result,limit_list):    """    function:通过limit过滤功能对where中的条件进行进一步的过滤    :param where_result: where过滤后产生的结果    :param limit_list:  limit列表中的条件    :return: 返回limit过滤之后的列表    """    limit_result = []    if limit_list:        limit_list = where_result[0:int(limit_list[0])]        return limit_list    else:        #如果列表中的内容为空,则进行进一步的处理        return where_resultdef handle_where(content_list,where_list):    """    function:通过where列表中的内容对content_list中的行文本进行进一步的过滤    :param content_list: 列表内容    :param where_list: where条件列表    :return: 返回一个过滤之后的列表    """    where_result = []    title = "id,name,age,phone,dept,enroll_date"    if where_list:        for line in content_list:            #以字典的形式去封装相应的信息            user_info = dict(zip(title.split(","),line.split(",")))            logic_result = logic_action(user_info,where_list)            if logic_result:                where_result.append(line)        return where_result    #如果where列表中内容为空,则直接返回整个列表内容    else:        where_result = content_list        return where_resultdef logic_action(user_dict,where_list):    """    function:通过where列表中的条件对文本进行过滤操作    :param user_dict: 用户的记录字典形式：单行文本    :param where_list: where条件列表形式[['id', '>', '10'], 'and', ['id', '<', '14'], 'or', ['name', 'like', '李']]    :return: 返回逻辑值真或假    """    logic_result = []    for item in where_list:        if (type(item) == list):            field,label,value = item            line_value = user_dict[field]            if label == "=":                label = "=="            if value.isdigit():                value = int(value)                line_value = int(line_value)            if label != 'like':                item = str(eval("%s%s%s" %(line_value,label,value)))            else:                if value in line_value:                    item = 'True'                else:                    item = 'False'        logic_result.append(item)    logic_str = ' '.join(logic_result)    logic_res = eval("%s"%logic_str)    return logic_resdef where_parse(where_list):    """    function:对where中的条件进行进一步的处理    :param where_list:     :return:     """    where_result = []    sens_list = ["and","or","not"]    key_word = ""    temp = []    if where_list:        tag = False        for item in where_list:            if item in sens_list:                tag = True                key_word = item                if temp:where_result.append(temp)                temp = []            if not tag and  item not in sens_list:                temp.append(item)            if tag and item not in sens_list:                tag = False                where_result.append(key_word)                key_word = ""                temp.append(item)        else:            where_result.append(temp)    return where_resultif __name__ == '__main__':    while True:        sql = input(">").strip()        print("\033[50;1m%s\033[0m"%sql)        sql_dict = select_parse(sql)        final_result = select(sql_dict)        for line in final_result:            print(line)

运行结果示例：

>select * from db1.emp limit 3select * from db1.emp limit 31,张德,25,18330267967,IT,2017-5-272,李四,25,19230267967,运维,2016-5-273,王五,35,15673026796,IT,2017-6-27>select * from db1.emp where not id >= 3select * from db1.emp where not id >= 31,张德,25,18330267967,IT,2017-5-272,李四,25,19230267967,运维,2016-5-27>select * from db1.emp where id > 2 and id < 10 or name like 李select * from db1.emp where id > 2 and id < 10 or name like 李2,李四,25,19230267967,运维,2016-5-273,王五,35,15673026796,IT,2017-6-274,赵六,29,18122263467,测试,2017-5-285,董七,25,18556267967,运维6,2014-5-276,李张明,26,18122677967,测试,2017-5-217,赵四,28,18523267967,运维6,2014-5-228,董七,25,18556267967,运维6,2014-5-27>select id,phone from db1.emp where name like 李select id,phone from db1.emp where name like 李['2', '19230267967']['6', '18122677967']