示例#1
0
def getrandstr(rng, charset, count):
    """return string containing *count* number of chars/bytes, whose elements are drawn from specified charset, using specified rng"""
    # NOTE: tests determined this is 4x faster than rng.sample(),
    # which is why that's not being used here.

    # check alphabet & count
    if count < 0:
        raise ValueError("count must be >= 0")
    letters = len(charset)
    if letters == 0:
        raise ValueError("alphabet must not be empty")
    if letters == 1:
        return charset * count

    # get random value, and write out to buffer
    def helper():
        # XXX: break into chunks for large number of letters?
        value = rng.randrange(0, letters**count)
        i = 0
        while i < count:
            yield charset[value % letters]
            value //= letters
            i += 1

    if isinstance(charset, unicode):
        return join_unicode(helper())
    else:
        return join_byte_elems(helper())
示例#2
0
def getrandstr(rng, charset, count):
    """return string containing *count* number of chars/bytes, whose elements are drawn from specified charset, using specified rng"""
    # NOTE: tests determined this is 4x faster than rng.sample(),
    # which is why that's not being used here.

    # check alphabet & count
    if count < 0:
        raise ValueError("count must be >= 0")
    letters = len(charset)
    if letters == 0:
        raise ValueError("alphabet must not be empty")
    if letters == 1:
        return charset * count

    # get random value, and write out to buffer
    def helper():
        # XXX: break into chunks for large number of letters?
        value = rng.randrange(0, letters**count)
        i = 0
        while i < count:
            yield charset[value % letters]
            value //= letters
            i += 1

    if isinstance(charset, unicode):
        return join_unicode(helper())
    else:
        return join_byte_elems(helper())
示例#3
0
def saslprep(source, param="value"):
    """Normalizes unicode strings using SASLPrep stringprep profile.

    The SASLPrep profile is defined in :rfc:`4013`.
    It provides a uniform scheme for normalizing unicode usernames
    and passwords before performing byte-value sensitive operations
    such as hashing. Among other things, it normalizes diacritic
    representations, removes non-printing characters, and forbids
    invalid characters such as ``\\n``. Properly internationalized
    applications should run user passwords through this function
    before hashing.

    :arg source:
        unicode string to normalize & validate

    :param param:
        Optional noun identifying source parameter in error messages
        (Defaults to the string ``"value"``). This is mainly useful to make the caller's error
        messages make more sense contextually.

    :raises ValueError:
        if any characters forbidden by the SASLPrep profile are encountered.

    :raises TypeError:
        if input is not :class:`!unicode`

    :returns:
        normalized unicode string

    .. note::

        This function is not available under Jython,
        as the Jython stdlib is missing the :mod:`!stringprep` module
        (`Jython issue 1758320 <http://bugs.jython.org/issue1758320>`_).

    .. versionadded:: 1.6
    """
    # saslprep - http://tools.ietf.org/html/rfc4013
    # stringprep - http://tools.ietf.org/html/rfc3454
    #              http://docs.python.org/library/stringprep.html

    # validate type
    # XXX: support bytes (e.g. run through want_unicode)?
    #      might be easier to just integrate this into cryptcontext.
    if not isinstance(source, unicode):
        raise TypeError("input must be unicode string, not %s" %
                        (type(source), ))

    # mapping stage
    #   - map non-ascii spaces to U+0020 (stringprep C.1.2)
    #   - strip 'commonly mapped to nothing' chars (stringprep B.1)
    in_table_c12 = stringprep.in_table_c12
    in_table_b1 = stringprep.in_table_b1
    data = join_unicode(_USPACE if in_table_c12(c) else c for c in source
                        if not in_table_b1(c))

    # normalize to KC form
    data = unicodedata.normalize('NFKC', data)
    if not data:
        return _UEMPTY

    # check for invalid bi-directional strings.
    # stringprep requires the following:
    #   - chars in C.8 must be prohibited.
    #   - if any R/AL chars in string:
    #       - no L chars allowed in string
    #       - first and last must be R/AL chars
    # this checks if start/end are R/AL chars. if so, prohibited loop
    # will forbid all L chars. if not, prohibited loop will forbid all
    # R/AL chars instead. in both cases, prohibited loop takes care of C.8.
    is_ral_char = stringprep.in_table_d1
    if is_ral_char(data[0]):
        if not is_ral_char(data[-1]):
            raise ValueError("malformed bidi sequence in " + param)
        # forbid L chars within R/AL sequence.
        is_forbidden_bidi_char = stringprep.in_table_d2
    else:
        # forbid R/AL chars if start not setup correctly; L chars allowed.
        is_forbidden_bidi_char = is_ral_char

    # check for prohibited output - stringprep tables A.1, B.1, C.1.2, C.2 - C.9
    in_table_a1 = stringprep.in_table_a1
    in_table_c21_c22 = stringprep.in_table_c21_c22
    in_table_c3 = stringprep.in_table_c3
    in_table_c4 = stringprep.in_table_c4
    in_table_c5 = stringprep.in_table_c5
    in_table_c6 = stringprep.in_table_c6
    in_table_c7 = stringprep.in_table_c7
    in_table_c8 = stringprep.in_table_c8
    in_table_c9 = stringprep.in_table_c9
    for c in data:
        # check for chars mapping stage should have removed
        assert not in_table_b1(c), "failed to strip B.1 in mapping stage"
        assert not in_table_c12(c), "failed to replace C.1.2 in mapping stage"

        # check for forbidden chars
        if in_table_a1(c):
            raise ValueError("unassigned code points forbidden in " + param)
        if in_table_c21_c22(c):
            raise ValueError("control characters forbidden in " + param)
        if in_table_c3(c):
            raise ValueError("private use characters forbidden in " + param)
        if in_table_c4(c):
            raise ValueError("non-char code points forbidden in " + param)
        if in_table_c5(c):
            raise ValueError("surrogate codes forbidden in " + param)
        if in_table_c6(c):
            raise ValueError("non-plaintext chars forbidden in " + param)
        if in_table_c7(c):
            # XXX: should these have been caught by normalize?
            # if so, should change this to an assert
            raise ValueError("non-canonical chars forbidden in " + param)
        if in_table_c8(c):
            raise ValueError("display-modifying / deprecated chars "
                             "forbidden in" + param)
        if in_table_c9(c):
            raise ValueError("tagged characters forbidden in " + param)

        # do bidi constraint check chosen by bidi init, above
        if is_forbidden_bidi_char(c):
            raise ValueError("forbidden bidi character in " + param)

    return data
示例#4
0
def saslprep(source, param="value"):
    """Normalizes unicode strings using SASLPrep stringprep profile.

    The SASLPrep profile is defined in :rfc:`4013`.
    It provides a uniform scheme for normalizing unicode usernames
    and passwords before performing byte-value sensitive operations
    such as hashing. Among other things, it normalizes diacritic
    representations, removes non-printing characters, and forbids
    invalid characters such as ``\\n``. Properly internationalized
    applications should run user passwords through this function
    before hashing.

    :arg source:
        unicode string to normalize & validate

    :param param:
        Optional noun identifying source parameter in error messages
        (Defaults to the string ``"value"``). This is mainly useful to make the caller's error
        messages make more sense contextually.

    :raises ValueError:
        if any characters forbidden by the SASLPrep profile are encountered.

    :raises TypeError:
        if input is not :class:`!unicode`

    :returns:
        normalized unicode string

    .. note::

        This function is not available under Jython,
        as the Jython stdlib is missing the :mod:`!stringprep` module
        (`Jython issue 1758320 <http://bugs.jython.org/issue1758320>`_).

    .. versionadded:: 1.6
    """
    # saslprep - http://tools.ietf.org/html/rfc4013
    # stringprep - http://tools.ietf.org/html/rfc3454
    #              http://docs.python.org/library/stringprep.html

    # validate type
    # XXX: support bytes (e.g. run through want_unicode)?
    #      might be easier to just integrate this into cryptcontext.
    if not isinstance(source, unicode):
        raise TypeError("input must be unicode string, not %s" %
                        (type(source),))

    # mapping stage
    #   - map non-ascii spaces to U+0020 (stringprep C.1.2)
    #   - strip 'commonly mapped to nothing' chars (stringprep B.1)
    in_table_c12 = stringprep.in_table_c12
    in_table_b1 = stringprep.in_table_b1
    data = join_unicode(
        _USPACE if in_table_c12(c) else c
        for c in source
        if not in_table_b1(c)
        )

    # normalize to KC form
    data = unicodedata.normalize('NFKC', data)
    if not data:
        return _UEMPTY

    # check for invalid bi-directional strings.
    # stringprep requires the following:
    #   - chars in C.8 must be prohibited.
    #   - if any R/AL chars in string:
    #       - no L chars allowed in string
    #       - first and last must be R/AL chars
    # this checks if start/end are R/AL chars. if so, prohibited loop
    # will forbid all L chars. if not, prohibited loop will forbid all
    # R/AL chars instead. in both cases, prohibited loop takes care of C.8.
    is_ral_char = stringprep.in_table_d1
    if is_ral_char(data[0]):
        if not is_ral_char(data[-1]):
            raise ValueError("malformed bidi sequence in " + param)
        # forbid L chars within R/AL sequence.
        is_forbidden_bidi_char = stringprep.in_table_d2
    else:
        # forbid R/AL chars if start not setup correctly; L chars allowed.
        is_forbidden_bidi_char = is_ral_char

    # check for prohibited output - stringprep tables A.1, B.1, C.1.2, C.2 - C.9
    in_table_a1 = stringprep.in_table_a1
    in_table_c21_c22 = stringprep.in_table_c21_c22
    in_table_c3 = stringprep.in_table_c3
    in_table_c4 = stringprep.in_table_c4
    in_table_c5 = stringprep.in_table_c5
    in_table_c6 = stringprep.in_table_c6
    in_table_c7 = stringprep.in_table_c7
    in_table_c8 = stringprep.in_table_c8
    in_table_c9 = stringprep.in_table_c9
    for c in data:
        # check for chars mapping stage should have removed
        assert not in_table_b1(c), "failed to strip B.1 in mapping stage"
        assert not in_table_c12(c), "failed to replace C.1.2 in mapping stage"

        # check for forbidden chars
        if in_table_a1(c):
            raise ValueError("unassigned code points forbidden in " + param)
        if in_table_c21_c22(c):
            raise ValueError("control characters forbidden in " + param)
        if in_table_c3(c):
            raise ValueError("private use characters forbidden in " + param)
        if in_table_c4(c):
            raise ValueError("non-char code points forbidden in " + param)
        if in_table_c5(c):
            raise ValueError("surrogate codes forbidden in " + param)
        if in_table_c6(c):
            raise ValueError("non-plaintext chars forbidden in " + param)
        if in_table_c7(c):
            # XXX: should these have been caught by normalize?
            # if so, should change this to an assert
            raise ValueError("non-canonical chars forbidden in " + param)
        if in_table_c8(c):
            raise ValueError("display-modifying / deprecated chars "
                             "forbidden in" + param)
        if in_table_c9(c):
            raise ValueError("tagged characters forbidden in " + param)

        # do bidi constraint check chosen by bidi init, above
        if is_forbidden_bidi_char(c):
            raise ValueError("forbidden bidi character in " + param)

    return data