l4/pkg/python/contrib/Lib/encodings/utf_8_sig.py

   1 """ Python 'utf-8-sig' Codec
   2 This work similar to UTF-8 with the following changes:
   3
   4 * On encoding/writing a UTF-8 encoded BOM will be prepended/written as the
   5   first three bytes.
   6
   7 * On decoding/reading if the first three bytes are a UTF-8 encoded BOM, these
   8   bytes will be skipped.
   9 """
  10 import codecs
  11
  12 ### Codec APIs
  13
  14 def encode(input, errors='strict'):
  15     return (codecs.BOM_UTF8 + codecs.utf_8_encode(input, errors)[0], len(input))
  16
  17 def decode(input, errors='strict'):
  18     prefix = 0
  19     if input[:3] == codecs.BOM_UTF8:
  20         input = input[3:]
  21         prefix = 3
  22     (output, consumed) = codecs.utf_8_decode(input, errors, True)
  23     return (output, consumed+prefix)
  24
  25 class IncrementalEncoder(codecs.IncrementalEncoder):
  26     def __init__(self, errors='strict'):
  27         codecs.IncrementalEncoder.__init__(self, errors)
  28         self.first = True
  29
  30     def encode(self, input, final=False):
  31         if self.first:
  32             self.first = False
  33             return codecs.BOM_UTF8 + codecs.utf_8_encode(input, self.errors)[0]
  34         else:
  35             return codecs.utf_8_encode(input, self.errors)[0]
  36
  37     def reset(self):
  38         codecs.IncrementalEncoder.reset(self)
  39         self.first = True
  40
  41 class IncrementalDecoder(codecs.BufferedIncrementalDecoder):
  42     def __init__(self, errors='strict'):
  43         codecs.BufferedIncrementalDecoder.__init__(self, errors)
  44         self.first = True
  45
  46     def _buffer_decode(self, input, errors, final):
  47         if self.first:
  48             if len(input) < 3:
  49                 if codecs.BOM_UTF8.startswith(input):
  50                     # not enough data to decide if this really is a BOM
  51                     # => try again on the next call
  52                     return (u"", 0)
  53                 else:
  54                     self.first = None
  55             else:
  56                 self.first = None
  57                 if input[:3] == codecs.BOM_UTF8:
  58                     (output, consumed) = codecs.utf_8_decode(input[3:], errors, final)
  59                     return (output, consumed+3)
  60         return codecs.utf_8_decode(input, errors, final)
  61
  62     def reset(self):
  63         codecs.BufferedIncrementalDecoder.reset(self)
  64         self.first = True
  65
  66 class StreamWriter(codecs.StreamWriter):
  67     def reset(self):
  68         codecs.StreamWriter.reset(self)
  69         try:
  70             del self.encode
  71         except AttributeError:
  72             pass
  73
  74     def encode(self, input, errors='strict'):
  75         self.encode = codecs.utf_8_encode
  76         return encode(input, errors)
  77
  78 class StreamReader(codecs.StreamReader):
  79     def reset(self):
  80         codecs.StreamReader.reset(self)
  81         try:
  82             del self.decode
  83         except AttributeError:
  84             pass
  85
  86     def decode(self, input, errors='strict'):
  87         if len(input) < 3:
  88             if codecs.BOM_UTF8.startswith(input):
  89                 # not enough data to decide if this is a BOM
  90                 # => try again on the next call
  91                 return (u"", 0)
  92         elif input[:3] == codecs.BOM_UTF8:
  93             self.decode = codecs.utf_8_decode
  94             (output, consumed) = codecs.utf_8_decode(input[3:],errors)
  95             return (output, consumed+3)
  96         # (else) no BOM present
  97         self.decode = codecs.utf_8_decode
  98         return codecs.utf_8_decode(input, errors)
  99
 100 ### encodings module API
 101
 102 def getregentry():
 103     return codecs.CodecInfo(
 104         name='utf-8-sig',
 105         encode=encode,
 106         decode=decode,
 107         incrementalencoder=IncrementalEncoder,
 108         incrementaldecoder=IncrementalDecoder,
 109         streamreader=StreamReader,
 110         streamwriter=StreamWriter,
 111     )