openmedialibrary_platform/Darwin/lib/python3.5/encodings/utf_8_sig.py

131 lines
4 KiB
Python
Raw Normal View History

2013-10-11 17:28:32 +00:00
""" Python 'utf-8-sig' Codec
This work similar to UTF-8 with the following changes:
* On encoding/writing a UTF-8 encoded BOM will be prepended/written as the
first three bytes.
* On decoding/reading if the first three bytes are a UTF-8 encoded BOM, these
bytes will be skipped.
"""
import codecs
### Codec APIs
def encode(input, errors='strict'):
2014-09-30 16:15:32 +00:00
return (codecs.BOM_UTF8 + codecs.utf_8_encode(input, errors)[0],
len(input))
2013-10-11 17:28:32 +00:00
def decode(input, errors='strict'):
prefix = 0
if input[:3] == codecs.BOM_UTF8:
input = input[3:]
prefix = 3
(output, consumed) = codecs.utf_8_decode(input, errors, True)
return (output, consumed+prefix)
class IncrementalEncoder(codecs.IncrementalEncoder):
def __init__(self, errors='strict'):
codecs.IncrementalEncoder.__init__(self, errors)
self.first = 1
def encode(self, input, final=False):
if self.first:
self.first = 0
2014-09-30 16:15:32 +00:00
return codecs.BOM_UTF8 + \
codecs.utf_8_encode(input, self.errors)[0]
2013-10-11 17:28:32 +00:00
else:
return codecs.utf_8_encode(input, self.errors)[0]
def reset(self):
codecs.IncrementalEncoder.reset(self)
self.first = 1
def getstate(self):
return self.first
def setstate(self, state):
self.first = state
class IncrementalDecoder(codecs.BufferedIncrementalDecoder):
def __init__(self, errors='strict'):
codecs.BufferedIncrementalDecoder.__init__(self, errors)
2014-09-30 16:15:32 +00:00
self.first = 1
2013-10-11 17:28:32 +00:00
def _buffer_decode(self, input, errors, final):
if self.first:
if len(input) < 3:
if codecs.BOM_UTF8.startswith(input):
# not enough data to decide if this really is a BOM
# => try again on the next call
2014-09-30 16:15:32 +00:00
return ("", 0)
2013-10-11 17:28:32 +00:00
else:
2014-09-30 16:15:32 +00:00
self.first = 0
2013-10-11 17:28:32 +00:00
else:
2014-09-30 16:15:32 +00:00
self.first = 0
2013-10-11 17:28:32 +00:00
if input[:3] == codecs.BOM_UTF8:
2014-09-30 16:15:32 +00:00
(output, consumed) = \
codecs.utf_8_decode(input[3:], errors, final)
2013-10-11 17:28:32 +00:00
return (output, consumed+3)
return codecs.utf_8_decode(input, errors, final)
def reset(self):
codecs.BufferedIncrementalDecoder.reset(self)
2014-09-30 16:15:32 +00:00
self.first = 1
def getstate(self):
state = codecs.BufferedIncrementalDecoder.getstate(self)
# state[1] must be 0 here, as it isn't passed along to the caller
return (state[0], self.first)
def setstate(self, state):
# state[1] will be ignored by BufferedIncrementalDecoder.setstate()
codecs.BufferedIncrementalDecoder.setstate(self, state)
self.first = state[1]
2013-10-11 17:28:32 +00:00
class StreamWriter(codecs.StreamWriter):
def reset(self):
codecs.StreamWriter.reset(self)
try:
del self.encode
except AttributeError:
pass
def encode(self, input, errors='strict'):
self.encode = codecs.utf_8_encode
return encode(input, errors)
class StreamReader(codecs.StreamReader):
def reset(self):
codecs.StreamReader.reset(self)
try:
del self.decode
except AttributeError:
pass
def decode(self, input, errors='strict'):
if len(input) < 3:
if codecs.BOM_UTF8.startswith(input):
# not enough data to decide if this is a BOM
# => try again on the next call
2014-09-30 16:15:32 +00:00
return ("", 0)
2013-10-11 17:28:32 +00:00
elif input[:3] == codecs.BOM_UTF8:
self.decode = codecs.utf_8_decode
(output, consumed) = codecs.utf_8_decode(input[3:],errors)
return (output, consumed+3)
# (else) no BOM present
self.decode = codecs.utf_8_decode
return codecs.utf_8_decode(input, errors)
### encodings module API
def getregentry():
return codecs.CodecInfo(
name='utf-8-sig',
encode=encode,
decode=decode,
incrementalencoder=IncrementalEncoder,
incrementaldecoder=IncrementalDecoder,
streamreader=StreamReader,
streamwriter=StreamWriter,
)