python摸爬滚打之day022----模块(序列化操作)

1、pickle

　　可以将我们python中的任意数据类型转化成bytes并写入到文件中.  同样也可以把文件中写好的bytes转换回我们python的数据.

　　pickle可以直接序列化对象.

 class Fruit:
     def __init__(self,name,color):
         self.name = name
         self.color = color

     def eat(self):
         return "haochi"

 p1 = Fruit("apple1","red1")
 p2 = Fruit("apple2","red4")
 p3 = Fruit("apple3","red2")
 p4 = Fruit("apple4","red6")
 lst = [p1,p2,p3,p4]

 pickle.dump(lst,open("fruit_info","wb"))

 lst = pickle.load(open("fruit_info","rb"))
 for i in lst:
     print(i.name)

pickle序列化对象

2、shelve

　　shelve提供python的持久化操作. 什么叫持久化操作呢? 说白话,就是把数据写到硬盘上. 在操作shelve的时候非常的像操作一个字典.  这个东西到后期. 就像redis差不多.

 # shelve  数据的持久化操作, 像操作字典一样去操作shelve  -----------------------------------------------------------------
 d = shelve.open("shelve_file")
 d["wlh"] = "王力宏"
 d["ljj"] = {"name":"林俊杰","id":"singer"}
 d["zj"] = {"name":"张杰","wife":{"name":"谢娜","gender":"女","worke":"大本营"}}
 d.close()

 d = shelve.open("shelve_file")
 d.pop("zj")     # 能删除,但是再查询时会报错.
 d.close()

 d = shelve.open("shelve_file",writeback=True)     # 如果是嵌套字典的话, 必须得加上writeback回写参数, 不然没法修改
 d["ljj"]["id"] = "歌手"
 print(d["ljj"]["id"])
 d.close()

shelve

3、configparser

　　 该模块适用于配置文件的格式与windows ini文件类似，可以包含一个或多个节(section), 每个节可以有多个参数(键=值)

 import configparser
 config = configparser.ConfigParser()
 config["DEFAULT"] = {                  # DEFAULT 是默认配置, 下面的一些子配置都会带上DEFAULT 里面的配置.
     "默认设置1":"...1",
     "默认设置2":"....2",
     "默认设置3":".....3",
 }
 config["配置11"] = {
     "设置11":"",
     "设置12":"",
     "设置13":"",
 }
 config["配置22"] = {
     "设置21":"",
     "设置22":"",
     "设置23":"",
 }
 config["配置33"] = {
     "设置31":"",
     "设置32":"",
     "设置33":"",
 }

 with open("peizhi.ini","w",encoding="utf8") as f:
     config.write(f)         # 这里的write()方法是configparser自己重新写的,不是调用的文件操作的write()

 # 查
 config = configparser.ConfigParser()
 config.read("peizhi.ini",encoding="utf8")
 print(config.sections())     # ['配置11', '配置22', '配置33']
 print(config["配置33"]["设置33"])       #

 # 添加一个章节      先读取文件, 再修改, 最后写入文件
 config = configparser.ConfigParser()
 config.read("peizhi.ini",encoding="utf8")
 config.add_section("配置44")
 config["配置44"] = {
     "设置41": "",
     "设置42": "",
     "设置43": "",
 }
 config.write(open("peizhi.ini","w",encoding="utf8"))

 # 修改章节
 config = configparser.ConfigParser()
 config.read("peizhi.ini",encoding="utf8")
 config.set("配置44","设置43","")
 config.write(open("peizhi.ini","w",encoding="utf8"))

 # 删除章节或  章节元素
 config = configparser.ConfigParser()
 config.read("peizhi.ini",encoding="utf8")
 config.remove_section("配置44")         # 删除"配置44"章节
 config.remove_option("配置44","设置43")    # 删除"配置44"里的"设置43"元素
 config.write(open("peizhi.ini","w",encoding="utf8"))

 # 对config的所有修改操作都是先创建config对象,然后用config读取内容, 修改完之后再将config写进文件.

configparser

4、json

　　json是所有语言都通用的模块

　   4.1  json和pickle的比较?

 dic = {1:"青铜",2:"白银",3:"黄金",4:"铂金",5:"钻石"}

 ret = json.dumps(dic,ensure_ascii=False)    # 用 ensure_ascii=False 来处理非中文显示
 print(ret,type(ret))       # json结果是 str 类型
 # ret = pickle.dumps(dic)
 # print(ret,type(ret))      # pickle结果是 bytes 类型

　　4.2  json写进文件

 dic = {'王力宏':{"ji":"新加坡","gender":"男","album":"盖世英雄"},
        "林俊杰":"爱笑的眼睛",
        "张靓颖":{"gender":"女","album":"如果这就是爱情"},
        "张杰":"云中的angle"
        }
 with open("singer_info.json","w",encoding="utf8") as f:
     json.dump(dic,f,ensure_ascii=False,indent=4)          # indent=4 即在json文件中改成好看点的格式
 # 即下面格式:
 # {
 #     "王力宏": {
 #         "ji": "新加坡",
 #         "gender": "男",
 #         "album": "盖世英雄"
 #     },
 #     "林俊杰": "爱笑的眼睛",
 #     "张靓颖": {
 #         "gender": "女",
 #         "album": "如果这就是爱情"
 #     },
 #     "张杰": "云中的angle"
 # }

json数据写进文件

　　4.3  json也能处理对象, 也能将对象写进文件(通过间接方式将对象转成字典的形式,然后序列化字典写进文件)

 class Fruit:
     def __init__(self,name,color):
         self.name = name
         self.color = color

 def trans1(obj):
     return {
         "name": obj.name,
         "color": obj.color,
     }
 # 用json序列化对象 ???
 ap = Fruit("苹果","红色")

 # 方法一: __dict__  对obj的各种封装信息(也是字典)进行序列化
 # ret = json.dumps(ap.__dict__, ensure_ascii=False)
 # print(ret)

 # 方法二: 对函数返回的结果进行序列化
 ret = json.dumps(ap,default=trans1,ensure_ascii=False)   # 把obj对象作为参数传进trans里,将返回的字典结果进行序列化
 print(ret)

json序列化对象

 class Fruit:
     def __init__(self,name,color):
         self.name = name
         self.color = color

 def trans2(dic):
     print(type(dic))
     return Fruit(dic["name"],dic["color"])      # 返回自己构造的Fruit对象

 # 用json反序列化
 s = '{"name": "苹果", "color": "红色"}'
 ap = json.loads(s,object_hook=trans2)     # 先将s进行反序列化, 然后将反序列化结果作为参数传进trans2 ,在函数里自己构造对象
 print(ap.color)

json反序列化对象

　　4.4  如何将大量数据写进文件?

　　 　　不用 dump. 改用dumps和loads. 对每一行分别进行处理.

　　　　用dumps将数据逐个序列化, 并写进文件;  用loads逐行将字符串反序列化, 并输出(不能用load, 逐行读出来的字符串是不符合json语法规范, 没法load, 会直接报错).

 # 用json写进大量数据
 dic1 = {"name":"乱世佳人","show_time":1975,"score":8.5,"time":"180min"}
 dic2 = {"name":"闻香识女人","show_time":1998,"score":8.8,"time":"160min"}
 dic3 = {"name":"飞越疯人院","show_time":1973,"score":8.7,"time":"140min"}
 dic4 = {"name":"霸王别姬","show_time":2001,"score":9.3,"time":"150min"}
 lst = [dic1,dic2,dic3,dic4]
 # with open("movie.json","w",encoding="utf8") as f:
 #     for i in lst:
 #         s = json.dumps(i,ensure_ascii=False)
 #         f.write(s+"\n")

 with open("movie.json","r",encoding="utf8") as f:
     for line in f:
         line = line.strip()
         if not line:
             continue
         cont = json.loads(line)      # 只是一个反序列化操作, 把字符串式的line转成字典形式.
         # cont = json.load(line)        # 错误写法, 这里line只是读取了json文件的一行，并没有全部读取，
                                         # line里面所存的字符串是不符合json语法的，没法load, 所以读取出错.
         print(cont)

json写入大量数据