>>> s='梯'
# python3.x 字符串 str 對(duì)應(yīng) unicode
>>> s,type(s)
('梯', <class 'str'>)
>>> s_gbk=s.encode('gbk')
>>> s_utf8=s.encode('utf-8')
# 1個(gè)漢字，gbk編碼為2個(gè)字節(jié)， utf-8 編碼為3個(gè)字節(jié)
# 2位16進(jìn)制 占1個(gè)字節(jié)
>>> s_gbk,s_utf8
(b'\xcc\xdd', b'\xe6\xa2\xaf')
>>> len(s_gbk),len(s_utf8)
(2, 3)

# 寫文件時(shí) encoding 表示文件存儲(chǔ)到磁盤的編碼格式
>>> open('gbk.txt','w',encoding='gbk').write('梯')
1
>>> open('utf8.txt','w',encoding='utf-8').write('梯')
1
# 'rb' 二進(jìn)制模式，讀取文件內(nèi)容為bytes類型，為文件實(shí)際存儲(chǔ)的字節(jié)串
# gbk編碼，1個(gè)中文占2個(gè)字節(jié)
>>> open('gbk.txt','rb').read()
b'\xcc\xdd'
# utf-8編碼，1個(gè)中文占3個(gè)字節(jié)
>>> open('utf8.txt','rb').read()
b'\xe6\xa2\xaf'

# 讀文件時(shí) encoding 表示解釋器讀取文件字節(jié)串解碼使用的編碼格式
# 需與文件存儲(chǔ)時(shí)編碼格式相同
>>> open('gbk.txt','r',encoding='gbk').read()
'梯'
>>> open('utf8.txt','r',encoding='utf-8').read()
'梯'
>>> dgbk=open('gbk.txt','rb').read();dgbk
b'\xcc\xdd'
>>> dutf8=open('utf8.txt','rb').read();dutf8
b'\xe6\xa2\xaf'
# 二進(jìn)制模式，讀取文件內(nèi)容后進(jìn)行手動(dòng)解碼
# python3.x 解釋器默認(rèn)編碼為utf-8
>>> dgbk.decode('gbk');dutf8.decode()
'梯'
'梯'

import locale
locale.getpreferredencoding()

>>> import locale
>>> locale.getpreferredencoding()
'cp936'
>>> locale.getpreferredencoding(False)
'cp936'

>>> '梯'.encode()
b'\xe6\xa2\xaf'
>>> '梯'.encode('utf-8')
b'\xe6\xa2\xaf'
>>> '梯'.encode('gbk')
b'\xcc\xdd'
>>> '梯'.encode('cp936')
b'\xcc\xdd'
# utf-16 帶小端字節(jié)序
>>> '梯'.encode('utf-16')
b'\xff\xfe\xafh'
# utf-16 忽略小端字節(jié)序(little BOM)
>>> '梯'.encode('utf-16-le')
b'\xafh'

>>> fpath='temp.txt'
# open()函數(shù)encoding未傳時(shí)默認(rèn)編碼取 locale.getpreferredencoding(False)
# 'r+' 文本模式打開(kāi)文件
>>> f=open(fpath,'r+')
# f.encoding 獲取文件對(duì)象的編碼方式
>>> f.encoding
'cp936'
# 文件指針會(huì)停留在文件結(jié)尾
>>> f.write('梯')
1
# 文件指針移動(dòng)到開(kāi)頭
>>> f.seek(0)
0
>>> f.read()
'梯'
>>> f.close()
# 'rb' 二進(jìn)制模式直接讀取文件的字節(jié)串
# 與 cp936 編碼一致
>>> open(fpath,'rb').read()
b'\xcc\xdd'
# 通過(guò) 相同的編碼 cp936或gbk 讀取文件內(nèi)容
>>> open(fpath,'r',encoding='cp936').read()
'梯'
>>> open(fpath,'r',encoding='gbk').read()
'梯'
>>> open(fpath,'r',encoding='utf-8').read()
Traceback (most recent call last):
  File "<pyshell#24>", line 1, in <module>
    open(fpath,'r',encoding='utf-8').read()
  File "D:\python3\lib\codecs.py", line 322, in decode
    (result, consumed) = self._buffer_decode(data, self.errors, final)
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xcc in position 0: invalid continuation byte

# python2.x內(nèi)置open()函數(shù)沒(méi)有encoding入?yún)?，無(wú)法指定編碼方法
>>> fpath='temp.txt'
>>> f=open(fpath,'w')
>>> f.write('a梯')
>>> f.close()
>>> open(fpath,'r').read()
'a\xcc\xdd'
# 'rb' 二進(jìn)制模式直接讀取文件的字節(jié)串
# 與 cp936 編碼一致
>>> open(fpath,'rb').read()
'a\xcc\xdd'
>>> len('a梯')
3
# python2.x通過(guò) codecs.open(encoding)指定編碼
>>> import codecs
>>> f=codecs.open(fpath,'w','utf-8')
# 寫入漢字時(shí)需要通過(guò)u'漢字'創(chuàng)建Unicode字符串
# 因?yàn)閜ython2.x解釋器默認(rèn)編碼為ASCII碼，
# '梯'字符串str用操作系統(tǒng)cp936自動(dòng)編碼為 '\xcc\xdd'
# 解釋器需要對(duì)其解碼為Unicode，解碼時(shí) \xcc 不在ASCII范圍內(nèi)，會(huì)報(bào)錯(cuò)
>>> f.write(u'梯')
>>> f.close()
# 'rb' 二進(jìn)制模式直接讀取文件的字節(jié)串
# 與 utf-8 編碼一致
>>> codecs.open(fpath,'rb').read()
'\xe6\xa2\xaf'

>>> fpath='utf8bom.txt'
# encoding='utf-8-sig', 寫文件時(shí)，會(huì)在文件開(kāi)頭加入 字節(jié)順序標(biāo)記
>>> open(fpath,'w',encoding='utf-8-sig').write('梯')
1
>>> open(fpath,'rb').read()
b'\xef\xbb\xbf\xe6\xa2\xaf'
>>> open(fpath,'r',encoding='utf-8').read()
'\ufeff梯'
# encoding='utf-8-sig', 讀文件時(shí)，忽略字節(jié)順序標(biāo)記
>>> open(fpath,'r',encoding='utf-8-sig').read()
'梯'

>>> import sys
# sys.byteorder 當(dāng)前系統(tǒng)的字節(jié)序， 'little'（小端）或 'big'（大端）
>>> sys.byteorder
'little'
# 寫文件時(shí)，encoding='utf-16' ，默認(rèn)添加當(dāng)前系統(tǒng)字節(jié)序(little)
>>> open(fpath,'w',encoding='utf-16').write('梯')
1
# 二進(jìn)制模式，讀取實(shí)際存放字節(jié)串，utf-16小端字節(jié)序?yàn)?\xff\xfe 
>>> open(fpath,'rb').read()
b'\xff\xfe\xafh'
>>> open(fpath,'r',encoding='utf-16').read()
'梯'
# encoding='utf-16-le'，讀文件時(shí)，忽略字節(jié)順序標(biāo)記
>>> open(fpath,'r',encoding='utf-16-le').read()
'\ufeff梯'

# encoding='utf-16-le'，寫文件時(shí)，不添加BOM
>>> open(fpath,'w',encoding='utf-16-le').write('梯')
1
>>> open(fpath,'rb').read()
b'\xafh'
>>> open(fpath,'r',encoding='utf-16-le').read()
'梯'
# encoding='utf-16-le'，寫文件后，不能用 utf-16 讀取
>>> open(fpath,'r',encoding='utf-16').read()
Traceback (most recent call last):
  File "<pyshell#102>", line 1, in <module>
    open(fpath,'r',encoding='utf-16').read()
  File "D:\python3\lib\codecs.py", line 322, in decode
    (result, consumed) = self._buffer_decode(data, self.errors, final)
  File "D:\python3\lib\encodings\utf_16.py", line 67, in _buffer_decode
    raise UnicodeError("UTF-16 stream does not start with BOM")
UnicodeError: UTF-16 stream does not start with BOM