标签：调用 cti datalist demo1 高效 反序 ble 比较 clear
python3 pickle持久化的储存数据。
python程序运行中得到了一些字符串，列表，字典等数据，想要长久的保存下来，方便以后使用，而不是简单的放入内存中关机断电就丢失数据。python模块大全中pickle模块就排上用场了， 他可以将对象转换为一种可以传输或存储的格式。
pickle完全用python来实现的，cpickle用C来实现的，cpickle的速度要比pickle快好多倍，电脑中如果有cpickle的话建议使用cpickle。
1、一个字典a，用dumple（）存储到本地文件，所存数据的格式就是字典，而普通的file.write（）写入文件的是字符串。读取时，load（）返回的是一个字典，file.read（）返回的是一个字符串。
1 import pickle 2 3 4 a = {" name ": "Tom", "age": "40"} 5 with open(‘text.txt‘, ‘wb‘) as file: 6 pickle.dump(a, file) 7 8 with open(‘text.txt‘, ‘rb‘) as file2: 9 b = pickle.load(file2) 10 11 print(type(b)) 12 print(b)
执行结果：
/usr/bin/python3.5 /home/rxf/python3_1000/1000/python3_server/python_pickle/example.py <class ‘dict‘> {‘age‘: ‘40‘, ‘ name ‘: ‘Tom‘}
2、一个列表info，用 pickle.dumps()方法将info序列化为string形式，而不是存入文件中。用pickle.loads()方法从string（文件名称data1）读出序列化前的对象。
1 import pickle 2 import pprint 3 4 info = [1, 2, 3, ‘abc‘, ‘ilovepython‘] 5 print(‘原始数据：‘) 6 pprint.pprint(info) 7 8 data1 = pickle.dumps(info) 9 data2 = pickle.loads(data1) 10 11 print("序列化：%r" % data1) 12 print("反序列化： %r" % data2)
执行结果：
/usr/bin/python3.5 /home/rxf/python3_1000/1000/python3_server/python_pickle/demo1.py 原始数据： [1, 2, 3, ‘abc‘, ‘ilovepython‘] 序列化：b‘\x80\x03]q\x00(K\x01K\x02K\x03X\x03\x00\x00\x00abcq\x01X\x0b\x00\x00\x00ilovepythonq\x02e.‘ 反序列化： [1, 2, 3, ‘abc‘, ‘ilovepython‘] Process finished with exit code 0
3、pickle模块主要函数实例
1 # pickle模块主要函数的应用举例 2 import pickle 3 import pprint 4 5 dataList = [[8, 1, ‘python‘], 6 [8, 1, ‘python‘], 7 [8, 0, ‘python‘], 8 [8, 1, ‘C++‘], 9 [8, 1, ‘C++‘]] 10 dataDic = {0: [1, 2, 3, 4], 11 1: (‘a‘, ‘b‘), 12 2: {‘c‘: ‘yes‘, ‘d‘: ‘no‘}} 13 print("原始数据dataList：") 14 pprint.pprint(dataList) 15 print(‘\n‘) 16 print("原始数据dataDic：") 17 pprint.pprint(dataDic) 18 19 # 使用dump()将数据序列化到文件中 20 fw = open(‘dataFile.txt‘, ‘wb‘) 21 # Pickle the list using the highest protocol available. 22 pickle.dump(dataList, fw) 23 # Pickle dictionary using protocol 0. 24 pickle.dump(dataDic, fw) 25 fw.close() 26 27 # 使用load()将数据从文件中序列化读出 28 fr = open(‘dataFile.txt‘, ‘rb‘) 29 data1 = pickle.load(fr) 30 print(‘\n‘+"反序列化1：%r" % data1) 31 data2 = pickle.load(fr) 32 print("反序列化2：%r" % data2 + ‘\n‘) 33 fr.close() 34 35 # 使用dumps()和loads()举例 36 p = pickle.dumps(dataList) 37 print(pickle.loads(p)) 38 p = pickle.dumps(dataDic) 39 print(pickle.loads(p))
执行结果：
/usr/bin/python3.5 /home/rxf/python3_1000/1000/python3_server/python_pickle/demo2.py 原始数据dataList： [[8, 1, ‘python‘], [8, 1, ‘python‘], [8, 0, ‘python‘], [8, 1, ‘C++‘], [8, 1, ‘C++‘]] 原始数据dataDic： {0: [1, 2, 3, 4], 1: (‘a‘, ‘b‘), 2: {‘c‘: ‘yes‘, ‘d‘: ‘no‘}} 反序列化1：[[8, 1, ‘python‘], [8, 1, ‘python‘], [8, 0, ‘python‘], [8, 1, ‘C++‘], [8, 1, ‘C++‘]] 反序列化2：{0: [1, 2, 3, 4], 1: (‘a‘, ‘b‘), 2: {‘d‘: ‘no‘, ‘c‘: ‘yes‘}} [[8, 1, ‘python‘], [8, 1, ‘python‘], [8, 0, ‘python‘], [8, 1, ‘C++‘], [8, 1, ‘C++‘]] {0: [1, 2, 3, 4], 1: (‘a‘, ‘b‘), 2: {‘d‘: ‘no‘, ‘c‘: ‘yes‘}} Process finished with exit code 0
4、要注意的是，在load（file）时，要让python能够找到类的定义，否则会报错：
1 import pickle 2 3 4 class Person: 5 def __init__(self, name, age): 6 self.name = name 7 self.age = age 8 9 def show(self): 10 print(self.name+"_"+str(self.age)) 11 12 aa = Person("Battier", 6) 13 aa.show() 14 15 f = open(‘./demo3.txt‘, ‘wb‘) 16 pickle.dump(aa, f, 0) 17 f.close() 18 19 # del Person 20 f = open(‘./demo3.txt‘, ‘rb‘) 21 bb = pickle.load(f) 22 23 f.close() 24 bb.show()
如果不注释掉del Person的话，那么会报错：（意思就是当前的模块找不到类了）
/usr/bin/python3.5 /home/rxf/python3_1000/1000/python3_server/python_pickle/demo3.py Battier_6 Traceback (most recent call last): File "/home/rxf/python3_1000/1000/python3_server/python_pickle/demo3.py", line 21, in <module> bb = pickle.load(f) AttributeError: Can‘t get attribute ‘Person‘ on <module ‘__main__‘ from ‘/home/rxf/python3_1000/1000/python3_server/python_pickle/demo3.py‘> Process finished with exit code 1
5、清空pickler的“备忘”，使用Pickler实例在序列化对象的时候，它会“记住”已经被序列化的对象引用，所以对同一对象多次调用dump(obj)，pickler不会“傻呼呼”的去多次序列化。
1 import pickle 2 import io 3 4 5 class Person: 6 def __init__(self, name, age): 7 self.name = name 8 self.age = age 9 10 def show(self): 11 print(self.name + "_"+str(self.age)) 12 13 aa = Person("Battier", 6) 14 aa.show() 15 16 17 fle = io.BytesIO() 18 pick = pickle.Pickler(fle) 19 pick.dump(aa) 20 val1 = fle.getvalue() 21 print(len(val1)) 22 23 pick.clear_memo() 24 pick.dump(aa) 25 val2 = fle.getvalue() 26 print(len(val2)) 27 fle.close()
上面代码运行结果：
/usr/bin/python3.5 /home/rxf/python3_1000/1000/python3_server/python_pickle/demo4.py Battier_6 69 138 Process finished with exit code 0
再注释掉pick.clear_memo()后，运行结果如下：
/usr/bin/python3.5 /home/rxf/python3_1000/1000/python3_server/python_pickle/demo4.py Battier_6 69 74 Process finished with exit code 0
主要是因为，python的pickle如果不clear_memo，则不会多次去序列化对象。
标签：调用 cti datalist demo1 高效 反序 ble 比较 clear
原文地址：http://www.cnblogs.com/ranxf/p/7800179.html
