Python codecs 模块，html() 实例源码

def get_special_case_params():
    #  The windows and unix specific tests should be tested on both unix and Windows to detect crashes.
    special_cases = [
        [u"noexist", u"noexist"],
        [u"tests/ascii/ex1", u"noexist"],
        [u"noexist", u"tests/ascii/ex1"],
        [u"tests/ascii/ex1", u"tests/ascii/ex1", "--outfile", "/dev/null"],
        [u"tests/ascii/ex1", u"tests/ascii/ex2"],
        [u"tests/utf_8/ex3", u"tests/utf_8/ex4"],
        [u"tests/utf_8/ex3", u"tests/utf_8/ex4", u"--oldfile-encoding", u"\"utf-8\"", u"--newfile-encoding", u"\"utf-8\""],
        [u"tests/utf_8/ex3", u"tests/utf_8/ex4", u"--oldfile-encoding", u"\"utf-8\"", u"--newfile-encoding", u"\"utf-8\"", u"--output-encoding", u"\"utf-8\""],
        [u"tests/ascii/ex5", u"tests/ascii/ex6"],
        [u"tests/ascii/ex7", u"tests/ascii/ex8"],
        [u"tests/ascii/a.json", u"tests/ascii/b.json"],
        [u"tests/ascii/a.json", u"tests/ascii/b.json", u"--push-delimiters", u"\"{\"", u"\"[\"", u"--pop-delimiters", u"\"}\"", u"\"]\"", u"--include-delimiters"],
        [u"tests/utf_8/fancy1", u"tests/utf_8/fancy2", u"--delimiters", u"???", u"--include-delimiters", u"--parameters-encoding", u"\"utf-8\"", u"--output-encoding", u"\"utf-8\"", u"--newfile-encoding", u"\"utf-8\"", u"--oldfile-encoding", u"\"utf-8\""],
        [u"tests/utf_8/fancy1", u"tests/utf_8/fancy2", u"--delimiters", u"\"\\u65e5\\u672c\\u56fd\"", u"--include-delimiters", u"--parameters-encoding", u"\"utf-8\"", u"--output-encoding", u"\"utf-8\"", u"--newfile-encoding", u"\"utf-8\"", u"--oldfile-encoding", u"\"utf-8\""],
        [u"tests/utf_8/this-is-encoded-in-utf-8", u"tests/utf_16/this-is-encoded-in-utf-16", u"--output-encoding", u"\"utf-8\"", u"--newfile-encoding", u"\"utf-16\"", u"--oldfile-encoding", u"\"utf-8\"", u"--enable-mark"],
        [u"tests/ascii/a.html", u"tests/ascii/b.html", u"-m", u"html"]
    ]
    return special_cases[random.randint(0, len(special_cases)-1)]

def error_handler(error):
  """Error handler for surrogateescape decoding.

  Should be used with an ASCII-compatible encoding (e.g., 'latin-1' or 'utf-8').
  Replaces any invalid byte sequences with surrogate code points.

  As specified in
  https://docs.python.org/2/library/codecs.html#codecs.register_error.
  """
  # We can't use this with UnicodeEncodeError; the UTF-8 encoder doesn't raise
  # an error for surrogates. Instead, use encode.
  if not isinstance(error, UnicodeDecodeError):
    raise error

  result = []
  for i in range(error.start, error.end):
    byte = ord(error.object[i])
    if byte < 128:
      raise error
    result.append(unichr(0xdc00 + byte))

  return ''.join(result), error.end

def from_path(path, manager=None, allow_nested=False, citation_clearing=True, encoding='utf-8', **kwargs):
    """Loads a BEL graph from a file resource. This function is a thin wrapper around :func:`from_lines`.

    :param str path: A file path
    :param manager: database connection string to cache, pre-built :class:`Manager`, or None to use default cache
    :type manager: None or str or pybel.manager.Manager
    :param bool allow_nested: if true, turn off nested statement failures
    :param bool citation_clearing: Should :code:`SET Citation` statements clear evidence and all annotations?
                                Delegated to :class:`pybel.parser.ControlParser`
    :param str encoding: the encoding to use when reading this file. Is passed to :code:`codecs.open`.
                     See the python `docs <https://docs.python.org/3/library/codecs.html#standard-encodings>`_ for a
                     list of standard encodings. For example, files starting with a UTF-8 BOM should use
                     :code:`utf_8_sig`
    :param dict kwargs: keyword arguments to :func:`pybel.io.line_utils.parse_lines`
    :rtype: BELGraph
    """
    log.info('Loading from path: %s', path)
    with codecs.open(os.path.expanduser(path), encoding=encoding) as file:
        return from_lines(
            lines=file,
            manager=manager,
            allow_nested=allow_nested,
            citation_clearing=citation_clearing,
            **kwargs
        )

def run(self, edit, encoding, file_name, need_codecs):
        self.view.set_name('ConvertToUTF8 Instructions')
        self.view.set_scratch(True)
        self.view.settings().set("word_wrap", True)
        msg = 'File: {0}\nEncoding: {1}\nError: '.format(file_name, encoding)
        if need_codecs:
            msg = msg + 'Codecs missing\n\n'
            branch = self.get_branch(sublime.platform(), sublime.arch())
            if branch:
                ver = '33' if ST3 else '26'
                msg = msg + 'Please install Codecs{0} plugin (https://github.com/seanliang/Codecs{0}/tree/{1}).\n'.format(ver, branch)
            else:
                import platform
                msg = msg + 'Please send the following information to sunlxy (at) yahoo.com:\n====== Debug Information ======\nVersion: {0}-{1}\nPlatform: {2}\nPath: {3}\nEncoding: {4}\n'.format(
                    sublime.version(), sublime.arch(), platform.platform(), sys.path, encoding
                )
        else:
            msg = msg + 'Unsupported encoding, see http://docs.python.org/library/codecs.html#standard-encodings\n\nPlease try other tools such as iconv.\n'

        self.view.insert(edit, 0, msg)
        self.view.set_read_only(True)
        self.view.window().focus_view(self.view)

def encode(self, input, errors='strict'):
        assert errors == 'strict'
        #return codecs.encode(input, self.base_encoding, self.name), len(input)

        # The above line could totally be all we needed, relying on the error
        # handling to replace the unencodable Unicode characters with our extended
        # byte sequences.
        #
        # However, there seems to be a design bug in Python (probably intentional):
        # the error handler for encoding is supposed to return a **Unicode** character,
        # that then needs to be encodable itself...  Ugh.
        #
        # So we implement what codecs.encode() should have been doing: which is expect
        # error handler to return bytes() to be added to the output.
        #
        # This seems to have been fixed in Python 3.3.  We should try using that and
        # use fallback only if that failed.
        # https://docs.python.org/3.3/library/codecs.html#codecs.register_error

        length = len(input)
        out = b''
        while input:
            try:
                part = codecs.encode(input, self.base_encoding)
                out += part
                input = '' # All converted
            except UnicodeEncodeError as e:
                # Convert the correct part
                out += codecs.encode(input[:e.start], self.base_encoding)
                replacement, pos = self.error(e)
                out += replacement
                input = input[pos:]
        return out, length

def pocketsphinx(self):
        # NOTE: copy()
        # Return a shallow copy of x.
        # source: https://docs.python.org/3/library/copy.html
        """Get pocketsphinx speech to text settings."""
        return self._data.get('pocketsphinx', {}).copy()

def pocketsphinx(self):
        # NOTE: copy()
        # Return a shallow copy of x.
        # source: https://docs.python.org/3/library/copy.html
        """Get pocketsphinx speech to text settings."""
        return self._data.get('pocketsphinx', {}).copy()