python在读写文件之前需要创建文件对象-Python对象序列化写入文件对象

1.创建Python文件对象的读写模式(r,w模式)与创建Java输入输出流；

FileInputStream inputStream=new FileInputStream(new File("E:\workspace\tmpfile\farrago.txt"));

FileOutputStream outputStream=new FileOutputStream(new File("E:\workspace\tmpfile\outagainb.txt"));

2.序列化Python对象，序列化成字符串如toJson和序列化成二进制；

3.向文件对象中写入序列化对象。

PS：有的序列化包，有方便的方法直接将序列化和写入过程合二为一。numpy的save和load感觉就是序列化而已，那么其必要性不大了？

import umsgpack

import pickle

import numpy as np

"""

1.创建Python文件对象的读写模式(r,w模式)与创建Java输入输出流；

FileInputStream inputStream=new FileInputStream(new File("E:\workspace\tmpfile\farrago.txt"));

FileOutputStream outputStream=new FileOutputStream(new File("E:\workspace\tmpfile\outagainb.txt"));

2.序列化Python对象，序列化成字符串如toJson和序列化成二进制；

3.向文件对象中写入序列化对象。

PS：有的序列化包，有方便的方法直接将序列化和写入过程合二为一。numpy的save和load感觉就是序列化而已，那么其必要性不大了？

"""

with open('test0.bin', 'wb') as f:

rs = umsgpack.packb({u"compact": True, u"schema": 0})

f.write(rs)

with open('test0.bin', 'rb') as f:

print(umsgpack.unpackb(f.read()))

print("_+"*20)

with open('test.bin', 'wb') as f:

print(umsgpack.pack({u"compact": True, u"schema": 0}, f))

print(umsgpack.pack([1,2,3], f))

with open('test.bin', 'rb') as f:

print(umsgpack.unpack(f))

print(umsgpack.unpack(f))

print("_+"*20)

with open('test2.bin', 'wb') as f:

rs = pickle.dumps({u"compact": True, u"schema": 0})

f.write(rs)

print(pickle.loads(rs))

with open('test2.bin', 'rb') as f:

print(pickle.load(f))

print("_+"*20)

with open('test3.bin', 'wb') as f:

pickle.dump({u"compact": True, u"schema": 0},f)

with open('test3.bin', 'rb') as f:

print(pickle.load(f))

print("_+="*20)

with open('test4.bin', 'wb') as f:

"""有没有觉得numpy的save与load就是个二进制序列化协议"""

np.save(f, 'abc')

np.save(f, np.arange(10))

with open('test4.bin', 'rb') as f:

print(np.load(f))

print(np.load(f))

print("888" * 20)