+++ /dev/null
-""" codecs -- Python Codec Registry, API and helpers.\r
-\r
-\r
-Written by Marc-Andre Lemburg (mal@lemburg.com).\r
-\r
-(c) Copyright CNRI, All Rights Reserved. NO WARRANTY.\r
-\r
-"""#"\r
-\r
-import __builtin__, sys\r
-\r
-### Registry and builtin stateless codec functions\r
-\r
-try:\r
- from _codecs import *\r
-except ImportError, why:\r
- raise SystemError('Failed to load the builtin codecs: %s' % why)\r
-\r
-__all__ = ["register", "lookup", "open", "EncodedFile", "BOM", "BOM_BE",\r
- "BOM_LE", "BOM32_BE", "BOM32_LE", "BOM64_BE", "BOM64_LE",\r
- "BOM_UTF8", "BOM_UTF16", "BOM_UTF16_LE", "BOM_UTF16_BE",\r
- "BOM_UTF32", "BOM_UTF32_LE", "BOM_UTF32_BE",\r
- "CodecInfo", "Codec", "IncrementalEncoder", "IncrementalDecoder",\r
- "StreamReader", "StreamWriter",\r
- "StreamReaderWriter", "StreamRecoder",\r
- "getencoder", "getdecoder", "getincrementalencoder",\r
- "getincrementaldecoder", "getreader", "getwriter",\r
- "encode", "decode", "iterencode", "iterdecode",\r
- "strict_errors", "ignore_errors", "replace_errors",\r
- "xmlcharrefreplace_errors", "backslashreplace_errors",\r
- "register_error", "lookup_error"]\r
-\r
-### Constants\r
-\r
-#\r
-# Byte Order Mark (BOM = ZERO WIDTH NO-BREAK SPACE = U+FEFF)\r
-# and its possible byte string values\r
-# for UTF8/UTF16/UTF32 output and little/big endian machines\r
-#\r
-\r
-# UTF-8\r
-BOM_UTF8 = '\xef\xbb\xbf'\r
-\r
-# UTF-16, little endian\r
-BOM_LE = BOM_UTF16_LE = '\xff\xfe'\r
-\r
-# UTF-16, big endian\r
-BOM_BE = BOM_UTF16_BE = '\xfe\xff'\r
-\r
-# UTF-32, little endian\r
-BOM_UTF32_LE = '\xff\xfe\x00\x00'\r
-\r
-# UTF-32, big endian\r
-BOM_UTF32_BE = '\x00\x00\xfe\xff'\r
-\r
-if sys.byteorder == 'little':\r
-\r
- # UTF-16, native endianness\r
- BOM = BOM_UTF16 = BOM_UTF16_LE\r
-\r
- # UTF-32, native endianness\r
- BOM_UTF32 = BOM_UTF32_LE\r
-\r
-else:\r
-\r
- # UTF-16, native endianness\r
- BOM = BOM_UTF16 = BOM_UTF16_BE\r
-\r
- # UTF-32, native endianness\r
- BOM_UTF32 = BOM_UTF32_BE\r
-\r
-# Old broken names (don't use in new code)\r
-BOM32_LE = BOM_UTF16_LE\r
-BOM32_BE = BOM_UTF16_BE\r
-BOM64_LE = BOM_UTF32_LE\r
-BOM64_BE = BOM_UTF32_BE\r
-\r
-\r
-### Codec base classes (defining the API)\r
-\r
-class CodecInfo(tuple):\r
-\r
- def __new__(cls, encode, decode, streamreader=None, streamwriter=None,\r
- incrementalencoder=None, incrementaldecoder=None, name=None):\r
- self = tuple.__new__(cls, (encode, decode, streamreader, streamwriter))\r
- self.name = name\r
- self.encode = encode\r
- self.decode = decode\r
- self.incrementalencoder = incrementalencoder\r
- self.incrementaldecoder = incrementaldecoder\r
- self.streamwriter = streamwriter\r
- self.streamreader = streamreader\r
- return self\r
-\r
- def __repr__(self):\r
- return "<%s.%s object for encoding %s at 0x%x>" % (self.__class__.__module__, self.__class__.__name__, self.name, id(self))\r
-\r
-class Codec:\r
-\r
- """ Defines the interface for stateless encoders/decoders.\r
-\r
- The .encode()/.decode() methods may use different error\r
- handling schemes by providing the errors argument. These\r
- string values are predefined:\r
-\r
- 'strict' - raise a ValueError error (or a subclass)\r
- 'ignore' - ignore the character and continue with the next\r
- 'replace' - replace with a suitable replacement character;\r
- Python will use the official U+FFFD REPLACEMENT\r
- CHARACTER for the builtin Unicode codecs on\r
- decoding and '?' on encoding.\r
- 'xmlcharrefreplace' - Replace with the appropriate XML\r
- character reference (only for encoding).\r
- 'backslashreplace' - Replace with backslashed escape sequences\r
- (only for encoding).\r
-\r
- The set of allowed values can be extended via register_error.\r
-\r
- """\r
- def encode(self, input, errors='strict'):\r
-\r
- """ Encodes the object input and returns a tuple (output\r
- object, length consumed).\r
-\r
- errors defines the error handling to apply. It defaults to\r
- 'strict' handling.\r
-\r
- The method may not store state in the Codec instance. Use\r
- StreamCodec for codecs which have to keep state in order to\r
- make encoding/decoding efficient.\r
-\r
- The encoder must be able to handle zero length input and\r
- return an empty object of the output object type in this\r
- situation.\r
-\r
- """\r
- raise NotImplementedError\r
-\r
- def decode(self, input, errors='strict'):\r
-\r
- """ Decodes the object input and returns a tuple (output\r
- object, length consumed).\r
-\r
- input must be an object which provides the bf_getreadbuf\r
- buffer slot. Python strings, buffer objects and memory\r
- mapped files are examples of objects providing this slot.\r
-\r
- errors defines the error handling to apply. It defaults to\r
- 'strict' handling.\r
-\r
- The method may not store state in the Codec instance. Use\r
- StreamCodec for codecs which have to keep state in order to\r
- make encoding/decoding efficient.\r
-\r
- The decoder must be able to handle zero length input and\r
- return an empty object of the output object type in this\r
- situation.\r
-\r
- """\r
- raise NotImplementedError\r
-\r
-class IncrementalEncoder(object):\r
- """\r
- An IncrementalEncoder encodes an input in multiple steps. The input can be\r
- passed piece by piece to the encode() method. The IncrementalEncoder remembers\r
- the state of the Encoding process between calls to encode().\r
- """\r
- def __init__(self, errors='strict'):\r
- """\r
- Creates an IncrementalEncoder instance.\r
-\r
- The IncrementalEncoder may use different error handling schemes by\r
- providing the errors keyword argument. See the module docstring\r
- for a list of possible values.\r
- """\r
- self.errors = errors\r
- self.buffer = ""\r
-\r
- def encode(self, input, final=False):\r
- """\r
- Encodes input and returns the resulting object.\r
- """\r
- raise NotImplementedError\r
-\r
- def reset(self):\r
- """\r
- Resets the encoder to the initial state.\r
- """\r
-\r
- def getstate(self):\r
- """\r
- Return the current state of the encoder.\r
- """\r
- return 0\r
-\r
- def setstate(self, state):\r
- """\r
- Set the current state of the encoder. state must have been\r
- returned by getstate().\r
- """\r
-\r
-class BufferedIncrementalEncoder(IncrementalEncoder):\r
- """\r
- This subclass of IncrementalEncoder can be used as the baseclass for an\r
- incremental encoder if the encoder must keep some of the output in a\r
- buffer between calls to encode().\r
- """\r
- def __init__(self, errors='strict'):\r
- IncrementalEncoder.__init__(self, errors)\r
- self.buffer = "" # unencoded input that is kept between calls to encode()\r
-\r
- def _buffer_encode(self, input, errors, final):\r
- # Overwrite this method in subclasses: It must encode input\r
- # and return an (output, length consumed) tuple\r
- raise NotImplementedError\r
-\r
- def encode(self, input, final=False):\r
- # encode input (taking the buffer into account)\r
- data = self.buffer + input\r
- (result, consumed) = self._buffer_encode(data, self.errors, final)\r
- # keep unencoded input until the next call\r
- self.buffer = data[consumed:]\r
- return result\r
-\r
- def reset(self):\r
- IncrementalEncoder.reset(self)\r
- self.buffer = ""\r
-\r
- def getstate(self):\r
- return self.buffer or 0\r
-\r
- def setstate(self, state):\r
- self.buffer = state or ""\r
-\r
-class IncrementalDecoder(object):\r
- """\r
- An IncrementalDecoder decodes an input in multiple steps. The input can be\r
- passed piece by piece to the decode() method. The IncrementalDecoder\r
- remembers the state of the decoding process between calls to decode().\r
- """\r
- def __init__(self, errors='strict'):\r
- """\r
- Creates a IncrementalDecoder instance.\r
-\r
- The IncrementalDecoder may use different error handling schemes by\r
- providing the errors keyword argument. See the module docstring\r
- for a list of possible values.\r
- """\r
- self.errors = errors\r
-\r
- def decode(self, input, final=False):\r
- """\r
- Decodes input and returns the resulting object.\r
- """\r
- raise NotImplementedError\r
-\r
- def reset(self):\r
- """\r
- Resets the decoder to the initial state.\r
- """\r
-\r
- def getstate(self):\r
- """\r
- Return the current state of the decoder.\r
-\r
- This must be a (buffered_input, additional_state_info) tuple.\r
- buffered_input must be a bytes object containing bytes that\r
- were passed to decode() that have not yet been converted.\r
- additional_state_info must be a non-negative integer\r
- representing the state of the decoder WITHOUT yet having\r
- processed the contents of buffered_input. In the initial state\r
- and after reset(), getstate() must return (b"", 0).\r
- """\r
- return (b"", 0)\r
-\r
- def setstate(self, state):\r
- """\r
- Set the current state of the decoder.\r
-\r
- state must have been returned by getstate(). The effect of\r
- setstate((b"", 0)) must be equivalent to reset().\r
- """\r
-\r
-class BufferedIncrementalDecoder(IncrementalDecoder):\r
- """\r
- This subclass of IncrementalDecoder can be used as the baseclass for an\r
- incremental decoder if the decoder must be able to handle incomplete byte\r
- sequences.\r
- """\r
- def __init__(self, errors='strict'):\r
- IncrementalDecoder.__init__(self, errors)\r
- self.buffer = "" # undecoded input that is kept between calls to decode()\r
-\r
- def _buffer_decode(self, input, errors, final):\r
- # Overwrite this method in subclasses: It must decode input\r
- # and return an (output, length consumed) tuple\r
- raise NotImplementedError\r
-\r
- def decode(self, input, final=False):\r
- # decode input (taking the buffer into account)\r
- data = self.buffer + input\r
- (result, consumed) = self._buffer_decode(data, self.errors, final)\r
- # keep undecoded input until the next call\r
- self.buffer = data[consumed:]\r
- return result\r
-\r
- def reset(self):\r
- IncrementalDecoder.reset(self)\r
- self.buffer = ""\r
-\r
- def getstate(self):\r
- # additional state info is always 0\r
- return (self.buffer, 0)\r
-\r
- def setstate(self, state):\r
- # ignore additional state info\r
- self.buffer = state[0]\r
-\r
-#\r
-# The StreamWriter and StreamReader class provide generic working\r
-# interfaces which can be used to implement new encoding submodules\r
-# very easily. See encodings/utf_8.py for an example on how this is\r
-# done.\r
-#\r
-\r
-class StreamWriter(Codec):\r
-\r
- def __init__(self, stream, errors='strict'):\r
-\r
- """ Creates a StreamWriter instance.\r
-\r
- stream must be a file-like object open for writing\r
- (binary) data.\r
-\r
- The StreamWriter may use different error handling\r
- schemes by providing the errors keyword argument. These\r
- parameters are predefined:\r
-\r
- 'strict' - raise a ValueError (or a subclass)\r
- 'ignore' - ignore the character and continue with the next\r
- 'replace'- replace with a suitable replacement character\r
- 'xmlcharrefreplace' - Replace with the appropriate XML\r
- character reference.\r
- 'backslashreplace' - Replace with backslashed escape\r
- sequences (only for encoding).\r
-\r
- The set of allowed parameter values can be extended via\r
- register_error.\r
- """\r
- self.stream = stream\r
- self.errors = errors\r
-\r
- def write(self, object):\r
-\r
- """ Writes the object's contents encoded to self.stream.\r
- """\r
- data, consumed = self.encode(object, self.errors)\r
- self.stream.write(data)\r
-\r
- def writelines(self, list):\r
-\r
- """ Writes the concatenated list of strings to the stream\r
- using .write().\r
- """\r
- self.write(''.join(list))\r
-\r
- def reset(self):\r
-\r
- """ Flushes and resets the codec buffers used for keeping state.\r
-\r
- Calling this method should ensure that the data on the\r
- output is put into a clean state, that allows appending\r
- of new fresh data without having to rescan the whole\r
- stream to recover state.\r
-\r
- """\r
- pass\r
-\r
- def seek(self, offset, whence=0):\r
- self.stream.seek(offset, whence)\r
- if whence == 0 and offset == 0:\r
- self.reset()\r
-\r
- def __getattr__(self, name,\r
- getattr=getattr):\r
-\r
- """ Inherit all other methods from the underlying stream.\r
- """\r
- return getattr(self.stream, name)\r
-\r
- def __enter__(self):\r
- return self\r
-\r
- def __exit__(self, type, value, tb):\r
- self.stream.close()\r
-\r
-###\r
-\r
-class StreamReader(Codec):\r
-\r
- def __init__(self, stream, errors='strict'):\r
-\r
- """ Creates a StreamReader instance.\r
-\r
- stream must be a file-like object open for reading\r
- (binary) data.\r
-\r
- The StreamReader may use different error handling\r
- schemes by providing the errors keyword argument. These\r
- parameters are predefined:\r
-\r
- 'strict' - raise a ValueError (or a subclass)\r
- 'ignore' - ignore the character and continue with the next\r
- 'replace'- replace with a suitable replacement character;\r
-\r
- The set of allowed parameter values can be extended via\r
- register_error.\r
- """\r
- self.stream = stream\r
- self.errors = errors\r
- self.bytebuffer = ""\r
- # For str->str decoding this will stay a str\r
- # For str->unicode decoding the first read will promote it to unicode\r
- self.charbuffer = ""\r
- self.linebuffer = None\r
-\r
- def decode(self, input, errors='strict'):\r
- raise NotImplementedError\r
-\r
- def read(self, size=-1, chars=-1, firstline=False):\r
-\r
- """ Decodes data from the stream self.stream and returns the\r
- resulting object.\r
-\r
- chars indicates the number of characters to read from the\r
- stream. read() will never return more than chars\r
- characters, but it might return less, if there are not enough\r
- characters available.\r
-\r
- size indicates the approximate maximum number of bytes to\r
- read from the stream for decoding purposes. The decoder\r
- can modify this setting as appropriate. The default value\r
- -1 indicates to read and decode as much as possible. size\r
- is intended to prevent having to decode huge files in one\r
- step.\r
-\r
- If firstline is true, and a UnicodeDecodeError happens\r
- after the first line terminator in the input only the first line\r
- will be returned, the rest of the input will be kept until the\r
- next call to read().\r
-\r
- The method should use a greedy read strategy meaning that\r
- it should read as much data as is allowed within the\r
- definition of the encoding and the given size, e.g. if\r
- optional encoding endings or state markers are available\r
- on the stream, these should be read too.\r
- """\r
- # If we have lines cached, first merge them back into characters\r
- if self.linebuffer:\r
- self.charbuffer = "".join(self.linebuffer)\r
- self.linebuffer = None\r
-\r
- # read until we get the required number of characters (if available)\r
- while True:\r
- # can the request be satisfied from the character buffer?\r
- if chars >= 0:\r
- if len(self.charbuffer) >= chars:\r
- break\r
- elif size >= 0:\r
- if len(self.charbuffer) >= size:\r
- break\r
- # we need more data\r
- if size < 0:\r
- newdata = self.stream.read()\r
- else:\r
- newdata = self.stream.read(size)\r
- # decode bytes (those remaining from the last call included)\r
- data = self.bytebuffer + newdata\r
- try:\r
- newchars, decodedbytes = self.decode(data, self.errors)\r
- except UnicodeDecodeError, exc:\r
- if firstline:\r
- newchars, decodedbytes = self.decode(data[:exc.start], self.errors)\r
- lines = newchars.splitlines(True)\r
- if len(lines)<=1:\r
- raise\r
- else:\r
- raise\r
- # keep undecoded bytes until the next call\r
- self.bytebuffer = data[decodedbytes:]\r
- # put new characters in the character buffer\r
- self.charbuffer += newchars\r
- # there was no data available\r
- if not newdata:\r
- break\r
- if chars < 0:\r
- # Return everything we've got\r
- result = self.charbuffer\r
- self.charbuffer = ""\r
- else:\r
- # Return the first chars characters\r
- result = self.charbuffer[:chars]\r
- self.charbuffer = self.charbuffer[chars:]\r
- return result\r
-\r
- def readline(self, size=None, keepends=True):\r
-\r
- """ Read one line from the input stream and return the\r
- decoded data.\r
-\r
- size, if given, is passed as size argument to the\r
- read() method.\r
-\r
- """\r
- # If we have lines cached from an earlier read, return\r
- # them unconditionally\r
- if self.linebuffer:\r
- line = self.linebuffer[0]\r
- del self.linebuffer[0]\r
- if len(self.linebuffer) == 1:\r
- # revert to charbuffer mode; we might need more data\r
- # next time\r
- self.charbuffer = self.linebuffer[0]\r
- self.linebuffer = None\r
- if not keepends:\r
- line = line.splitlines(False)[0]\r
- return line\r
-\r
- readsize = size or 72\r
- line = ""\r
- # If size is given, we call read() only once\r
- while True:\r
- data = self.read(readsize, firstline=True)\r
- if data:\r
- # If we're at a "\r" read one extra character (which might\r
- # be a "\n") to get a proper line ending. If the stream is\r
- # temporarily exhausted we return the wrong line ending.\r
- if data.endswith("\r"):\r
- data += self.read(size=1, chars=1)\r
-\r
- line += data\r
- lines = line.splitlines(True)\r
- if lines:\r
- if len(lines) > 1:\r
- # More than one line result; the first line is a full line\r
- # to return\r
- line = lines[0]\r
- del lines[0]\r
- if len(lines) > 1:\r
- # cache the remaining lines\r
- lines[-1] += self.charbuffer\r
- self.linebuffer = lines\r
- self.charbuffer = None\r
- else:\r
- # only one remaining line, put it back into charbuffer\r
- self.charbuffer = lines[0] + self.charbuffer\r
- if not keepends:\r
- line = line.splitlines(False)[0]\r
- break\r
- line0withend = lines[0]\r
- line0withoutend = lines[0].splitlines(False)[0]\r
- if line0withend != line0withoutend: # We really have a line end\r
- # Put the rest back together and keep it until the next call\r
- self.charbuffer = "".join(lines[1:]) + self.charbuffer\r
- if keepends:\r
- line = line0withend\r
- else:\r
- line = line0withoutend\r
- break\r
- # we didn't get anything or this was our only try\r
- if not data or size is not None:\r
- if line and not keepends:\r
- line = line.splitlines(False)[0]\r
- break\r
- if readsize<8000:\r
- readsize *= 2\r
- return line\r
-\r
- def readlines(self, sizehint=None, keepends=True):\r
-\r
- """ Read all lines available on the input stream\r
- and return them as list of lines.\r
-\r
- Line breaks are implemented using the codec's decoder\r
- method and are included in the list entries.\r
-\r
- sizehint, if given, is ignored since there is no efficient\r
- way to finding the true end-of-line.\r
-\r
- """\r
- data = self.read()\r
- return data.splitlines(keepends)\r
-\r
- def reset(self):\r
-\r
- """ Resets the codec buffers used for keeping state.\r
-\r
- Note that no stream repositioning should take place.\r
- This method is primarily intended to be able to recover\r
- from decoding errors.\r
-\r
- """\r
- self.bytebuffer = ""\r
- self.charbuffer = u""\r
- self.linebuffer = None\r
-\r
- def seek(self, offset, whence=0):\r
- """ Set the input stream's current position.\r
-\r
- Resets the codec buffers used for keeping state.\r
- """\r
- self.stream.seek(offset, whence)\r
- self.reset()\r
-\r
- def next(self):\r
-\r
- """ Return the next decoded line from the input stream."""\r
- line = self.readline()\r
- if line:\r
- return line\r
- raise StopIteration\r
-\r
- def __iter__(self):\r
- return self\r
-\r
- def __getattr__(self, name,\r
- getattr=getattr):\r
-\r
- """ Inherit all other methods from the underlying stream.\r
- """\r
- return getattr(self.stream, name)\r
-\r
- def __enter__(self):\r
- return self\r
-\r
- def __exit__(self, type, value, tb):\r
- self.stream.close()\r
-\r
-###\r
-\r
-class StreamReaderWriter:\r
-\r
- """ StreamReaderWriter instances allow wrapping streams which\r
- work in both read and write modes.\r
-\r
- The design is such that one can use the factory functions\r
- returned by the codec.lookup() function to construct the\r
- instance.\r
-\r
- """\r
- # Optional attributes set by the file wrappers below\r
- encoding = 'unknown'\r
-\r
- def __init__(self, stream, Reader, Writer, errors='strict'):\r
-\r
- """ Creates a StreamReaderWriter instance.\r
-\r
- stream must be a Stream-like object.\r
-\r
- Reader, Writer must be factory functions or classes\r
- providing the StreamReader, StreamWriter interface resp.\r
-\r
- Error handling is done in the same way as defined for the\r
- StreamWriter/Readers.\r
-\r
- """\r
- self.stream = stream\r
- self.reader = Reader(stream, errors)\r
- self.writer = Writer(stream, errors)\r
- self.errors = errors\r
-\r
- def read(self, size=-1):\r
-\r
- return self.reader.read(size)\r
-\r
- def readline(self, size=None):\r
-\r
- return self.reader.readline(size)\r
-\r
- def readlines(self, sizehint=None):\r
-\r
- return self.reader.readlines(sizehint)\r
-\r
- def next(self):\r
-\r
- """ Return the next decoded line from the input stream."""\r
- return self.reader.next()\r
-\r
- def __iter__(self):\r
- return self\r
-\r
- def write(self, data):\r
-\r
- return self.writer.write(data)\r
-\r
- def writelines(self, list):\r
-\r
- return self.writer.writelines(list)\r
-\r
- def reset(self):\r
-\r
- self.reader.reset()\r
- self.writer.reset()\r
-\r
- def seek(self, offset, whence=0):\r
- self.stream.seek(offset, whence)\r
- self.reader.reset()\r
- if whence == 0 and offset == 0:\r
- self.writer.reset()\r
-\r
- def __getattr__(self, name,\r
- getattr=getattr):\r
-\r
- """ Inherit all other methods from the underlying stream.\r
- """\r
- return getattr(self.stream, name)\r
-\r
- # these are needed to make "with codecs.open(...)" work properly\r
-\r
- def __enter__(self):\r
- return self\r
-\r
- def __exit__(self, type, value, tb):\r
- self.stream.close()\r
-\r
-###\r
-\r
-class StreamRecoder:\r
-\r
- """ StreamRecoder instances provide a frontend - backend\r
- view of encoding data.\r
-\r
- They use the complete set of APIs returned by the\r
- codecs.lookup() function to implement their task.\r
-\r
- Data written to the stream is first decoded into an\r
- intermediate format (which is dependent on the given codec\r
- combination) and then written to the stream using an instance\r
- of the provided Writer class.\r
-\r
- In the other direction, data is read from the stream using a\r
- Reader instance and then return encoded data to the caller.\r
-\r
- """\r
- # Optional attributes set by the file wrappers below\r
- data_encoding = 'unknown'\r
- file_encoding = 'unknown'\r
-\r
- def __init__(self, stream, encode, decode, Reader, Writer,\r
- errors='strict'):\r
-\r
- """ Creates a StreamRecoder instance which implements a two-way\r
- conversion: encode and decode work on the frontend (the\r
- input to .read() and output of .write()) while\r
- Reader and Writer work on the backend (reading and\r
- writing to the stream).\r
-\r
- You can use these objects to do transparent direct\r
- recodings from e.g. latin-1 to utf-8 and back.\r
-\r
- stream must be a file-like object.\r
-\r
- encode, decode must adhere to the Codec interface, Reader,\r
- Writer must be factory functions or classes providing the\r
- StreamReader, StreamWriter interface resp.\r
-\r
- encode and decode are needed for the frontend translation,\r
- Reader and Writer for the backend translation. Unicode is\r
- used as intermediate encoding.\r
-\r
- Error handling is done in the same way as defined for the\r
- StreamWriter/Readers.\r
-\r
- """\r
- self.stream = stream\r
- self.encode = encode\r
- self.decode = decode\r
- self.reader = Reader(stream, errors)\r
- self.writer = Writer(stream, errors)\r
- self.errors = errors\r
-\r
- def read(self, size=-1):\r
-\r
- data = self.reader.read(size)\r
- data, bytesencoded = self.encode(data, self.errors)\r
- return data\r
-\r
- def readline(self, size=None):\r
-\r
- if size is None:\r
- data = self.reader.readline()\r
- else:\r
- data = self.reader.readline(size)\r
- data, bytesencoded = self.encode(data, self.errors)\r
- return data\r
-\r
- def readlines(self, sizehint=None):\r
-\r
- data = self.reader.read()\r
- data, bytesencoded = self.encode(data, self.errors)\r
- return data.splitlines(1)\r
-\r
- def next(self):\r
-\r
- """ Return the next decoded line from the input stream."""\r
- data = self.reader.next()\r
- data, bytesencoded = self.encode(data, self.errors)\r
- return data\r
-\r
- def __iter__(self):\r
- return self\r
-\r
- def write(self, data):\r
-\r
- data, bytesdecoded = self.decode(data, self.errors)\r
- return self.writer.write(data)\r
-\r
- def writelines(self, list):\r
-\r
- data = ''.join(list)\r
- data, bytesdecoded = self.decode(data, self.errors)\r
- return self.writer.write(data)\r
-\r
- def reset(self):\r
-\r
- self.reader.reset()\r
- self.writer.reset()\r
-\r
- def __getattr__(self, name,\r
- getattr=getattr):\r
-\r
- """ Inherit all other methods from the underlying stream.\r
- """\r
- return getattr(self.stream, name)\r
-\r
- def __enter__(self):\r
- return self\r
-\r
- def __exit__(self, type, value, tb):\r
- self.stream.close()\r
-\r
-### Shortcuts\r
-\r
-def open(filename, mode='rb', encoding=None, errors='strict', buffering=1):\r
-\r
- """ Open an encoded file using the given mode and return\r
- a wrapped version providing transparent encoding/decoding.\r
-\r
- Note: The wrapped version will only accept the object format\r
- defined by the codecs, i.e. Unicode objects for most builtin\r
- codecs. Output is also codec dependent and will usually be\r
- Unicode as well.\r
-\r
- Files are always opened in binary mode, even if no binary mode\r
- was specified. This is done to avoid data loss due to encodings\r
- using 8-bit values. The default file mode is 'rb' meaning to\r
- open the file in binary read mode.\r
-\r
- encoding specifies the encoding which is to be used for the\r
- file.\r
-\r
- errors may be given to define the error handling. It defaults\r
- to 'strict' which causes ValueErrors to be raised in case an\r
- encoding error occurs.\r
-\r
- buffering has the same meaning as for the builtin open() API.\r
- It defaults to line buffered.\r
-\r
- The returned wrapped file object provides an extra attribute\r
- .encoding which allows querying the used encoding. This\r
- attribute is only available if an encoding was specified as\r
- parameter.\r
-\r
- """\r
- if encoding is not None:\r
- if 'U' in mode:\r
- # No automatic conversion of '\n' is done on reading and writing\r
- mode = mode.strip().replace('U', '')\r
- if mode[:1] not in set('rwa'):\r
- mode = 'r' + mode\r
- if 'b' not in mode:\r
- # Force opening of the file in binary mode\r
- mode = mode + 'b'\r
- file = __builtin__.open(filename, mode, buffering)\r
- if encoding is None:\r
- return file\r
- info = lookup(encoding)\r
- srw = StreamReaderWriter(file, info.streamreader, info.streamwriter, errors)\r
- # Add attributes to simplify introspection\r
- srw.encoding = encoding\r
- return srw\r
-\r
-def EncodedFile(file, data_encoding, file_encoding=None, errors='strict'):\r
-\r
- """ Return a wrapped version of file which provides transparent\r
- encoding translation.\r
-\r
- Strings written to the wrapped file are interpreted according\r
- to the given data_encoding and then written to the original\r
- file as string using file_encoding. The intermediate encoding\r
- will usually be Unicode but depends on the specified codecs.\r
-\r
- Strings are read from the file using file_encoding and then\r
- passed back to the caller as string using data_encoding.\r
-\r
- If file_encoding is not given, it defaults to data_encoding.\r
-\r
- errors may be given to define the error handling. It defaults\r
- to 'strict' which causes ValueErrors to be raised in case an\r
- encoding error occurs.\r
-\r
- The returned wrapped file object provides two extra attributes\r
- .data_encoding and .file_encoding which reflect the given\r
- parameters of the same name. The attributes can be used for\r
- introspection by Python programs.\r
-\r
- """\r
- if file_encoding is None:\r
- file_encoding = data_encoding\r
- data_info = lookup(data_encoding)\r
- file_info = lookup(file_encoding)\r
- sr = StreamRecoder(file, data_info.encode, data_info.decode,\r
- file_info.streamreader, file_info.streamwriter, errors)\r
- # Add attributes to simplify introspection\r
- sr.data_encoding = data_encoding\r
- sr.file_encoding = file_encoding\r
- return sr\r
-\r
-### Helpers for codec lookup\r
-\r
-def getencoder(encoding):\r
-\r
- """ Lookup up the codec for the given encoding and return\r
- its encoder function.\r
-\r
- Raises a LookupError in case the encoding cannot be found.\r
-\r
- """\r
- return lookup(encoding).encode\r
-\r
-def getdecoder(encoding):\r
-\r
- """ Lookup up the codec for the given encoding and return\r
- its decoder function.\r
-\r
- Raises a LookupError in case the encoding cannot be found.\r
-\r
- """\r
- return lookup(encoding).decode\r
-\r
-def getincrementalencoder(encoding):\r
-\r
- """ Lookup up the codec for the given encoding and return\r
- its IncrementalEncoder class or factory function.\r
-\r
- Raises a LookupError in case the encoding cannot be found\r
- or the codecs doesn't provide an incremental encoder.\r
-\r
- """\r
- encoder = lookup(encoding).incrementalencoder\r
- if encoder is None:\r
- raise LookupError(encoding)\r
- return encoder\r
-\r
-def getincrementaldecoder(encoding):\r
-\r
- """ Lookup up the codec for the given encoding and return\r
- its IncrementalDecoder class or factory function.\r
-\r
- Raises a LookupError in case the encoding cannot be found\r
- or the codecs doesn't provide an incremental decoder.\r
-\r
- """\r
- decoder = lookup(encoding).incrementaldecoder\r
- if decoder is None:\r
- raise LookupError(encoding)\r
- return decoder\r
-\r
-def getreader(encoding):\r
-\r
- """ Lookup up the codec for the given encoding and return\r
- its StreamReader class or factory function.\r
-\r
- Raises a LookupError in case the encoding cannot be found.\r
-\r
- """\r
- return lookup(encoding).streamreader\r
-\r
-def getwriter(encoding):\r
-\r
- """ Lookup up the codec for the given encoding and return\r
- its StreamWriter class or factory function.\r
-\r
- Raises a LookupError in case the encoding cannot be found.\r
-\r
- """\r
- return lookup(encoding).streamwriter\r
-\r
-def iterencode(iterator, encoding, errors='strict', **kwargs):\r
- """\r
- Encoding iterator.\r
-\r
- Encodes the input strings from the iterator using a IncrementalEncoder.\r
-\r
- errors and kwargs are passed through to the IncrementalEncoder\r
- constructor.\r
- """\r
- encoder = getincrementalencoder(encoding)(errors, **kwargs)\r
- for input in iterator:\r
- output = encoder.encode(input)\r
- if output:\r
- yield output\r
- output = encoder.encode("", True)\r
- if output:\r
- yield output\r
-\r
-def iterdecode(iterator, encoding, errors='strict', **kwargs):\r
- """\r
- Decoding iterator.\r
-\r
- Decodes the input strings from the iterator using a IncrementalDecoder.\r
-\r
- errors and kwargs are passed through to the IncrementalDecoder\r
- constructor.\r
- """\r
- decoder = getincrementaldecoder(encoding)(errors, **kwargs)\r
- for input in iterator:\r
- output = decoder.decode(input)\r
- if output:\r
- yield output\r
- output = decoder.decode("", True)\r
- if output:\r
- yield output\r
-\r
-### Helpers for charmap-based codecs\r
-\r
-def make_identity_dict(rng):\r
-\r
- """ make_identity_dict(rng) -> dict\r
-\r
- Return a dictionary where elements of the rng sequence are\r
- mapped to themselves.\r
-\r
- """\r
- res = {}\r
- for i in rng:\r
- res[i]=i\r
- return res\r
-\r
-def make_encoding_map(decoding_map):\r
-\r
- """ Creates an encoding map from a decoding map.\r
-\r
- If a target mapping in the decoding map occurs multiple\r
- times, then that target is mapped to None (undefined mapping),\r
- causing an exception when encountered by the charmap codec\r
- during translation.\r
-\r
- One example where this happens is cp875.py which decodes\r
- multiple character to \\u001a.\r
-\r
- """\r
- m = {}\r
- for k,v in decoding_map.items():\r
- if not v in m:\r
- m[v] = k\r
- else:\r
- m[v] = None\r
- return m\r
-\r
-### error handlers\r
-\r
-try:\r
- strict_errors = lookup_error("strict")\r
- ignore_errors = lookup_error("ignore")\r
- replace_errors = lookup_error("replace")\r
- xmlcharrefreplace_errors = lookup_error("xmlcharrefreplace")\r
- backslashreplace_errors = lookup_error("backslashreplace")\r
-except LookupError:\r
- # In --disable-unicode builds, these error handler are missing\r
- strict_errors = None\r
- ignore_errors = None\r
- replace_errors = None\r
- xmlcharrefreplace_errors = None\r
- backslashreplace_errors = None\r
-\r
-# Tell modulefinder that using codecs probably needs the encodings\r
-# package\r
-_false = 0\r
-if _false:\r
- import encodings\r
-\r
-### Tests\r
-\r
-if __name__ == '__main__':\r
-\r
- # Make stdout translate Latin-1 output into UTF-8 output\r
- sys.stdout = EncodedFile(sys.stdout, 'latin-1', 'utf-8')\r
-\r
- # Have stdin translate Latin-1 input into UTF-8 input\r
- sys.stdin = EncodedFile(sys.stdin, 'utf-8', 'latin-1')\r