def check_bidi(data):
"""
1) The characters in section 5.8 MUST be prohibited.
2) If a string contains any RandALCat character, the string MUST NOT
contain any LCat character.
3) If a string contains any RandALCat character, a RandALCat
character MUST be the first character of the string, and a
RandALCat character MUST be the last character of the string.
"""
if not data:
return data
has_lcat = False
has_randal = False
for c in data:
if stringprep.in_table_c8(c):
raise StringPrepError("BIDI violation: seciton 6 (1)")
if stringprep.in_table_d1(c):
has_randal = True
elif stringprep.in_table_d2(c):
has_lcat = True
if has_randal and has_lcat:
raise StringPrepError("BIDI violation: section 6 (2)")
first_randal = stringprep.in_table_d1(data[0])
last_randal = stringprep.in_table_d1(data[-1])
if has_randal and not (first_randal and last_randal):
raise StringPrepError("BIDI violation: section 6 (3)")
def nodeprep( foo, errors='strict' ):
if foo is None:
return None
if isinstance( foo, str ):
foo = foo.decode( 'utf-8' )
ofoo = u''
for x in foo:
if not stringprep.in_table_b1( x ):
ofoo += stringprep.map_table_b2( x )
foo = unicodedata.normalize( 'NFKC', ofoo )
ofoo = u''
first_is_randal = False
if len(foo):
first_is_randal = stringprep.in_table_d1( foo[0] )
if first_is_randal:
if not stringprep.in_table_d1( foo[-1] ):
raise UnicodeError, "Section 6.3 [end]"
for x in range(len(foo)):
if errors=='strict' and stringprep.in_table_a1( foo[x] ):
raise UnicodeError, "Unassigned Codepoint"
if stringprep.in_table_c11( foo[x] ):
raise UnicodeError, "In table C.1.1"
if stringprep.in_table_c12( foo[x] ):
raise UnicodeError, "In table C.1.2"
if stringprep.in_table_c21( foo[x] ):
raise UnicodeError, "In table C.2.1"
if stringprep.in_table_c22( foo[x] ):
raise UnicodeError, "In table C.2.2"
if stringprep.in_table_c3( foo[x] ):
raise UnicodeError, "In table C.3"
if stringprep.in_table_c4( foo[x] ):
raise UnicodeError, "In table C.4"
if stringprep.in_table_c5( foo[x] ):
raise UnicodeError, "In table C.5"
if stringprep.in_table_c6( foo[x] ):
raise UnicodeError, "In table C.6"
if stringprep.in_table_c7( foo[x] ):
raise UnicodeError, "In table C.7"
if stringprep.in_table_c8( foo[x] ):
raise UnicodeError, "In table C.8"
if stringprep.in_table_c9( foo[x] ):
raise UnicodeError, "In table C.9"
if foo[x] in "\"&'/;<>@":
raise UnicodeError, "In nodeprep banned list"
if x:
if first_is_randal and stringprep.in_table_d2( foo[x] ):
raise UnicodeError, "Section 6.2"
if not first_is_randal and x!=(len(foo)-1) and stringprep.in_table_d1( foo[x] ):
raise UnicodeError, "Section 6.3"
else:
first = False
return foo
def _checkBidi(self, data):
hasL = False
hasRorAL = False
for char in data:
if stringprep.in_table_d1(char):
hasRorAL = True
elif stringprep.in_table_d2(char):
hasL = True
if hasL and hasRorAL:
raise XmppStringPrepError('String contains RandALCat characters and LCat characters')
if hasRorAL and (not stringprep.in_table_d1(data[0]) or not stringprep.in_table_d1(data[-1])):
raise XmppStringPrepError('RandALCat character MUST be the first character of the string, and a RandALCat character MUST be the last character of the string')
return data
def check_bidirectionals(self, string): found_LCat = False found_RandALCat = False for c in string: if stringprep.in_table_d1(c): found_RandALCat = True if stringprep.in_table_d2(c): found_LCat = True if found_LCat and found_RandALCat: raise UnicodeError, "Violation of BIDI Requirement 2" if found_RandALCat and not (stringprep.in_table_d1(string[0]) and stringprep.in_table_d1(string[-1])): raise UnicodeError, "Violation of BIDI Requirement 3"
def check_bidi(data):
"""Checks if sting is valid for bidirectional printing."""
has_l = False
has_ral = False
for char in data:
if stringprep.in_table_d1(char):
has_ral = True
elif stringprep.in_table_d2(char):
has_l = True
if has_l and has_ral:
raise StringprepError("Both RandALCat and LCat characters present")
if has_ral and (not stringprep.in_table_d1(data[0])
or not stringprep.in_table_d1(data[-1])):
raise StringprepError("The first and the last character must"
" be RandALCat")
return data
def _check_bidi(s):
"""Enforce bidirectional character check from RFC 3454 (stringprep)"""
r_and_al_cat = False
l_cat = False
for c in s:
if not r_and_al_cat and stringprep.in_table_d1(c):
r_and_al_cat = True
if not l_cat and stringprep.in_table_d2(c):
l_cat = True
if r_and_al_cat and l_cat:
raise SASLPrepError('Both RandALCat and LCat characters present')
if r_and_al_cat and not (stringprep.in_table_d1(s[0]) and
stringprep.in_table_d1(s[-1])):
raise SASLPrepError('RandALCat character not at both start and end')
def saslprep(s, allow_unassigned = False):
'''
Prepare Unicode string s according to SASLprep: Stringprep Profile for
User Names and Passwords, a.k.a. RFC 4013
If the optional parameter allow_unassigned is set to True,
unassigned codepoints will be allowed. This is recommended for
query terms and other non-storing situations only.
The return value is a Unicode string appropriately prepared.
Disallowed input leads to a ValueError.
'''
if type(s) != type(u''):
raise TypeError("input must be a Unicode string")
# phase 1: mapping
s = u''.join([ stringprep.in_table_c12(ch) and u' ' or ch for ch in unichars(s) if not stringprep.in_table_b1(ch) ])
# phase 2: normalization
s = unicodedata.normalize('NFKC', s)
# phase 3: prohibition
for ch in unichars(s):
if stringprep.in_table_c12(ch):
raise ValueError("prohibited non-ASCII space character")
if stringprep.in_table_c21(ch):
raise ValueError("prohibited ASCII control character")
if stringprep.in_table_c22(ch):
raise ValueError("prohibited non-ASCII control character")
if stringprep.in_table_c3(ch):
raise ValueError("prohibited private use character")
if stringprep.in_table_c4(ch):
raise ValueError("prohibited non-character code point")
if stringprep.in_table_c5(ch):
raise ValueError("prohibited surrogate code point")
if stringprep.in_table_c6(ch):
raise ValueError("prohibited character inappropriate for plain text")
if stringprep.in_table_c7(ch):
raise ValueError("prohibited character inappropriate for canonical representation")
if stringprep.in_table_c8(ch):
raise ValueError("prohibited character changing display properties, or a deprecated character")
if stringprep.in_table_c9(ch):
raise ValueError("prohibited tagging character")
# phase 4: bidi check
bidi_map = ''.join([ stringprep.in_table_d1(ch) and 'r' or stringprep.in_table_d2(ch) and 'l' or 'x' for ch in unichars(s) ])
if 'r' in bidi_map:
if 'l' in bidi_map:
raise ValueError("prohibited mixture of strong left-to-right and right-to-left text")
if bidi_map[0] != 'r' or bidi_map[-1] != 'r':
raise ValueError("string containing right-to-left text must start and end with right-to-left text")
# phase 5: unassigned check
if not allow_unassigned:
for ch in unichars(s):
if stringprep.in_table_a1(ch):
raise ValueError("prohibited unassigned code point")
return s
def _process(self, value):
for code in force_text(value):
# TODO: Is this long enough?
if stringprep.in_table_c12(code) or stringprep.in_table_c21_c22(code) or \
stringprep.in_table_c3(code) or stringprep.in_table_c4(code) or \
stringprep.in_table_c5(code) or stringprep.in_table_c6(code) or \
stringprep.in_table_c7(code) or stringprep.in_table_c8(code) or \
stringprep.in_table_c9(code):
self.invalid = False
if stringprep.in_table_d1(code):
self.r_and_al_cat = True
if stringprep.in_table_d2(code):
self.l_cat = True
def nameprep(label):
# Map
newlabel = []
for c in label:
if stringprep.in_table_b1(c):
# Map to nothing
continue
newlabel.append(stringprep.map_table_b2(c))
label = "".join(newlabel)
# Normalize
label = unicodedata.normalize("NFKC", label)
# Prohibit
for c in label:
if (
stringprep.in_table_c12(c)
or stringprep.in_table_c22(c)
or stringprep.in_table_c3(c)
or stringprep.in_table_c4(c)
or stringprep.in_table_c5(c)
or stringprep.in_table_c6(c)
or stringprep.in_table_c7(c)
or stringprep.in_table_c8(c)
or stringprep.in_table_c9(c)
):
raise UnicodeError("Invalid character %r" % c)
# Check bidi
RandAL = [stringprep.in_table_d1(x) for x in label]
for c in RandAL:
if c:
# There is a RandAL char in the string. Must perform further
# tests:
# 1) The characters in section 5.8 MUST be prohibited.
# This is table C.8, which was already checked
# 2) If a string contains any RandALCat character, the string
# MUST NOT contain any LCat character.
if any(stringprep.in_table_d2(x) for x in label):
raise UnicodeError("Violation of BIDI requirement 2")
# 3) If a string contains any RandALCat character, a
# RandALCat character MUST be the first character of the
# string, and a RandALCat character MUST be the last
# character of the string.
if not RandAL[0] or not RandAL[-1]:
raise UnicodeError("Violation of BIDI requirement 3")
return label
def check_bidirectional(unistr):
contains_randalcat = False
contains_lcat = False
for c in unistr:
if stringprep.in_table_c8(c):
raise IdentifierError(IdentifierErrors.restriction_bidirectional_prohibited)
if stringprep.in_table_d1(c):
contains_randalcat = True
if stringprep.in_table_d2(c):
contains_lcat = True
if contains_randalcat:
if contains_lcat:
raise IdentifierError(IdentifierErrors.restriction_bidirectional_mixed)
if not all(map(stringprep.in_table_d1,[unistr[0],unistr[-1]])):
raise IdentifierError(IdentifierErrors.restriction_bidirectional_affix)
def nameprep(label):
newlabel = []
for c in label:
if stringprep.in_table_b1(c):
pass
newlabel.append(stringprep.map_table_b2(c))
label = ''.join(newlabel)
label = unicodedata.normalize('NFKC', label)
for c in label:
while stringprep.in_table_c12(c) or (stringprep.in_table_c22(c) or (stringprep.in_table_c3(c) or (stringprep.in_table_c4(c) or (stringprep.in_table_c5(c) or (stringprep.in_table_c6(c) or (stringprep.in_table_c7(c) or stringprep.in_table_c8(c))))))) or stringprep.in_table_c9(c):
raise UnicodeError('Invalid character %r' % c)
RandAL = [stringprep.in_table_d1(x) for x in label]
for c in RandAL:
while c:
if any(stringprep.in_table_d2(x) for x in label):
raise UnicodeError('Violation of BIDI requirement 2')
if not RandAL[0] or not RandAL[-1]:
raise UnicodeError('Violation of BIDI requirement 3')
return label
def __call__(self, value):
super(BidirectionalValidator, self).__call__(value)
if self.r_and_al_cat:
if self.l_cat or not stringprep.in_table_d1(self.first) or \
not stringprep.in_table_d1(self.last):
raise ValidationError(self.message, code=self.code)
def update_event(self, inp=-1):
self.set_output_val(0, stringprep.in_table_d1(self.input(0)))
def __call__(self, value):
super(BidirectionalValidator, self).__call__(value)
if self.r_and_al_cat:
if self.l_cat or not stringprep.in_table_d1(
self.first) or not stringprep.in_table_d1(self.last):
raise ValidationError(self.message, code=self.code)
def sasl_prep(data):
"""
implement SASLPrep profile as per RFC4013:
it defines the "SASLprep" profile of the "stringprep" algorithm [StringPrep].
The profile is designed for use in Simple Authentication and Security
Layer ([SASL]) mechanisms, such as [PLAIN], [CRAM-MD5], and
[DIGEST-MD5]. It may be applicable where simple user names and
passwords are used. This profile is not intended for use in
preparing identity strings that are not simple user names (e.g.,
email addresses, domain names, distinguished names), or where
identity or password strings that are not character data, or require
different handling (e.g., case folding).
"""
# mapping
prepared_data = ''
for c in data:
if stringprep.in_table_c12(c):
# non-ASCII space characters [StringPrep, C.1.2] that can be mapped to SPACE (U+0020)
prepared_data += ' '
elif stringprep.in_table_b1(c):
# the "commonly mapped to nothing" characters [StringPrep, B.1] that can be mapped to nothing.
pass
else:
prepared_data += c
# normalizing
# This profile specifies using Unicode normalization form KC
# The repertoire is Unicode 3.2 as per RFC 4013 (2)
prepared_data = unicode32.normalize('NFKC', prepared_data)
if not prepared_data:
raise LDAPSASLPrepError('SASLprep error: unable to normalize string')
# prohibit
for c in prepared_data:
if stringprep.in_table_c12(c):
# Non-ASCII space characters [StringPrep, C.1.2]
raise LDAPSASLPrepError(
'SASLprep error: non-ASCII space character present')
elif stringprep.in_table_c21(c):
# ASCII control characters [StringPrep, C.2.1]
raise LDAPSASLPrepError(
'SASLprep error: ASCII control character present')
elif stringprep.in_table_c22(c):
# Non-ASCII control characters [StringPrep, C.2.2]
raise LDAPSASLPrepError(
'SASLprep error: non-ASCII control character present')
elif stringprep.in_table_c3(c):
# Private Use characters [StringPrep, C.3]
raise LDAPSASLPrepError(
'SASLprep error: private character present')
elif stringprep.in_table_c4(c):
# Non-character code points [StringPrep, C.4]
raise LDAPSASLPrepError(
'SASLprep error: non-character code point present')
elif stringprep.in_table_c5(c):
# Surrogate code points [StringPrep, C.5]
raise LDAPSASLPrepError(
'SASLprep error: surrogate code point present')
elif stringprep.in_table_c6(c):
# Inappropriate for plain text characters [StringPrep, C.6]
raise LDAPSASLPrepError(
'SASLprep error: inappropriate for plain text character present'
)
elif stringprep.in_table_c7(c):
# Inappropriate for canonical representation characters [StringPrep, C.7]
raise LDAPSASLPrepError(
'SASLprep error: inappropriate for canonical representation character present'
)
elif stringprep.in_table_c8(c):
# Change display properties or deprecated characters [StringPrep, C.8]
raise LDAPSASLPrepError(
'SASLprep error: change display property or deprecated character present'
)
elif stringprep.in_table_c9(c):
# Tagging characters [StringPrep, C.9]
raise LDAPSASLPrepError(
'SASLprep error: tagging character present')
# check bidi
# if a string contains any r_and_al_cat character, the string MUST NOT contain any l_cat character.
flag_r_and_al_cat = False
flag_l_cat = False
for c in prepared_data:
if stringprep.in_table_d1(c):
flag_r_and_al_cat = True
elif stringprep.in_table_d2(c):
flag_l_cat = True
if flag_r_and_al_cat and flag_l_cat:
raise LDAPSASLPrepError(
'SASLprep error: string cannot contain (R or AL) and L bidirectional chars'
)
# If a string contains any r_and_al_cat character, a r_and_al_cat character MUST be the first character of the string
# and a r_and_al_cat character MUST be the last character of the string.
if flag_r_and_al_cat and not stringprep.in_table_d1(
prepared_data[0]) and not stringprep.in_table_d2(
prepared_data[-1]):
raise LDAPSASLPrepError(
'r_and_al_cat character present, must be first and last character of the string'
)
return prepared_data
def sasl_prep(data):
"""
implement SASLPrep profile as per RFC4013:
it defines the "SASLprep" profile of the "stringprep" algorithm [StringPrep].
The profile is designed for use in Simple Authentication and Security
Layer ([SASL]) mechanisms, such as [PLAIN], [CRAM-MD5], and
[DIGEST-MD5]. It may be applicable where simple user names and
passwords are used. This profile is not intended for use in
preparing identity strings that are not simple user names (e.g.,
email addresses, domain names, distinguished names), or where
identity or password strings that are not character data, or require
different handling (e.g., case folding).
"""
# mapping
prepared_data = ''
for c in data:
if stringprep.in_table_c12(c):
# non-ASCII space characters [StringPrep, C.1.2] that can be mapped to SPACE (U+0020)
prepared_data += ' '
elif stringprep.in_table_b1(c):
# the "commonly mapped to nothing" characters [StringPrep, B.1] that can be mapped to nothing.
pass
else:
prepared_data += c
# normalizing
# This profile specifies using Unicode normalization form KC
# The repertoire is Unicode 3.2 as per RFC 4013 (2)
prepared_data = unicode32.normalize('NFKC', prepared_data)
if not prepared_data:
raise LDAPSASLPrepError('SASLprep error: unable to normalize string')
# prohibit
for c in prepared_data:
if stringprep.in_table_c12(c):
# Non-ASCII space characters [StringPrep, C.1.2]
raise LDAPSASLPrepError('SASLprep error: non-ASCII space character present')
elif stringprep.in_table_c21(c):
# ASCII control characters [StringPrep, C.2.1]
raise LDAPSASLPrepError('SASLprep error: ASCII control character present')
elif stringprep.in_table_c22(c):
# Non-ASCII control characters [StringPrep, C.2.2]
raise LDAPSASLPrepError('SASLprep error: non-ASCII control character present')
elif stringprep.in_table_c3(c):
# Private Use characters [StringPrep, C.3]
raise LDAPSASLPrepError('SASLprep error: private character present')
elif stringprep.in_table_c4(c):
# Non-character code points [StringPrep, C.4]
raise LDAPSASLPrepError('SASLprep error: non-character code point present')
elif stringprep.in_table_c5(c):
# Surrogate code points [StringPrep, C.5]
raise LDAPSASLPrepError('SASLprep error: surrogate code point present')
elif stringprep.in_table_c6(c):
# Inappropriate for plain text characters [StringPrep, C.6]
raise LDAPSASLPrepError('SASLprep error: inappropriate for plain text character present')
elif stringprep.in_table_c7(c):
# Inappropriate for canonical representation characters [StringPrep, C.7]
raise LDAPSASLPrepError('SASLprep error: inappropriate for canonical representation character present')
elif stringprep.in_table_c8(c):
# Change display properties or deprecated characters [StringPrep, C.8]
raise LDAPSASLPrepError('SASLprep error: change display property or deprecated character present')
elif stringprep.in_table_c9(c):
# Tagging characters [StringPrep, C.9]
raise LDAPSASLPrepError('SASLprep error: tagging character present')
# check bidi
# if a string contains any r_and_al_cat character, the string MUST NOT contain any l_cat character.
flag_r_and_al_cat = False
flag_l_cat = False
for c in prepared_data:
if stringprep.in_table_d1(c):
flag_r_and_al_cat = True
elif stringprep.in_table_d2(c):
flag_l_cat = True
if flag_r_and_al_cat and flag_l_cat:
raise LDAPSASLPrepError('SASLprep error: string cannot contain (R or AL) and L bidirectional chars')
# If a string contains any r_and_al_cat character, a r_and_al_cat character MUST be the first character of the string
# and a r_and_al_cat character MUST be the last character of the string.
if flag_r_and_al_cat and not stringprep.in_table_d1(prepared_data[0]) and not stringprep.in_table_d2(prepared_data[-1]):
raise LDAPSASLPrepError('r_and_al_cat character present, must be first and last character of the string')
return prepared_data
def _normalize_password(self, password):
"""Normalize the password using PostgreSQL-flavored SASLprep. For reference:
https://git.postgresql.org/gitweb/?p=postgresql.git;a=blob;f=src/common/saslprep.c
using the `pg_saslprep` function
Implementation borrowed from asyncpg implementation:
https://github.com/MagicStack/asyncpg/blob/master/asyncpg/protocol/scram.pyx#L263
"""
normalized_password = password
# if the password is an ASCII string or fails to encode as an UTF8
# string, we can return
try:
normalized_password.encode("ascii")
except UnicodeEncodeError:
pass
else:
return normalized_password
# Step 1 of SASLPrep: Map. Per the algorithm, we map non-ascii space
# characters to ASCII spaces (\x20 or \u0020, but we will use ' ') and
# commonly mapped to nothing characters are removed
# Table C.1.2 -- non-ASCII spaces
# Table B.1 -- "Commonly mapped to nothing"
normalized_password = u"".join([
' ' if stringprep.in_table_c12(c) else c
for c in normalized_password if not stringprep.in_table_b1(c)
])
# If at this point the password is empty, PostgreSQL uses the original
# password
if not normalized_password:
return password
# Step 2 of SASLPrep: Normalize. Normalize the password using the
# Unicode normalization algorithm to NFKC form
normalized_password = unicodedata.normalize('NFKC',
normalized_password)
# If the password is not empty, PostgreSQL uses the original password
if not normalized_password:
return password
# Step 3 of SASLPrep: Prohobited characters. If PostgreSQL detects any
# of the prohibited characters in SASLPrep, it will use the original
# password
# We also include "unassigned code points" in the prohibited character
# category as PostgreSQL does the same
for c in normalized_password:
if any([
in_prohibited_table(c)
for in_prohibited_table in self.SASLPREP_STEP3
]):
return password
# Step 4 of SASLPrep: Bi-directional characters. PostgreSQL follows the
# rules for bi-directional characters laid on in RFC3454 Sec. 6 which
# are:
# 1. Characters in RFC 3454 Sec 5.8 are prohibited (C.8)
# 2. If a string contains a RandALCat character, it cannot containy any
# LCat character
# 3. If the string contains any RandALCat character, an RandALCat
# character must be the first and last character of the string
# RandALCat characters are found in table D.1, whereas LCat are in D.2
if any([stringprep.in_table_d1(c) for c in normalized_password]):
# if the first character or the last character are not in D.1,
# return the original password
if not (stringprep.in_table_d1(normalized_password[0])
and stringprep.in_table_d1(normalized_password[-1])):
return password
# if any characters are in D.2, use the original password
if any([stringprep.in_table_d2(c) for c in normalized_password]):
return password
# return the normalized password
return normalized_password
def saslprep(text, strict=True):
"""
Return a processed version of the given string, using the SASLPrep
profile of stringprep.
:param text: The string to process, in UTF-8.
:param strict: If ``True``, prevent the use of unassigned code points.
"""
if sys.version_info < (3, 0):
if type(text) == str:
text = text.decode("utf-8")
# Mapping:
#
# - non-ASCII space characters [StringPrep, C.1.2] that can be
# mapped to SPACE (U+0020), and
#
# - the 'commonly mapped to nothing' characters [StringPrep, B.1]
# that can be mapped to nothing.
buffer = ""
for char in text:
if stringprep.in_table_c12(char):
buffer += " "
elif not stringprep.in_table_b1(char):
buffer += char
# Normalization using form KC
text = unicodedata.normalize("NFKC", buffer)
# Check for bidirectional string
buffer = ""
first_is_randal = False
if text:
first_is_randal = stringprep.in_table_d1(text[0])
if first_is_randal and not stringprep.in_table_d1(text[-1]):
raise SASLPrepFailure("Section 6.3 [end]")
# Check for prohibited characters
for x in range(len(text)):
if strict and stringprep.in_table_a1(text[x]):
raise SASLPrepFailure("Unassigned Codepoint")
if stringprep.in_table_c12(text[x]):
raise SASLPrepFailure("In table C.1.2")
if stringprep.in_table_c21(text[x]):
raise SASLPrepFailure("In table C.2.1")
if stringprep.in_table_c22(text[x]):
raise SASLPrepFailure("In table C.2.2")
if stringprep.in_table_c3(text[x]):
raise SASLPrepFailure("In table C.3")
if stringprep.in_table_c4(text[x]):
raise SASLPrepFailure("In table C.4")
if stringprep.in_table_c5(text[x]):
raise SASLPrepFailure("In table C.5")
if stringprep.in_table_c6(text[x]):
raise SASLPrepFailure("In table C.6")
if stringprep.in_table_c7(text[x]):
raise SASLPrepFailure("In table C.7")
if stringprep.in_table_c8(text[x]):
raise SASLPrepFailure("In table C.8")
if stringprep.in_table_c9(text[x]):
raise SASLPrepFailure("In table C.9")
if x:
if first_is_randal and stringprep.in_table_d2(text[x]):
raise SASLPrepFailure("Section 6.2")
if not first_is_randal and x != len(text) - 1 and stringprep.in_table_d1(text[x]):
raise SASLPrepFailure("Section 6.3")
return text
def saslprep(text, strict=True):
"""
Return a processed version of the given string, using the SASLPrep
profile of stringprep.
:param text: The string to process, in UTF-8.
:param strict: If ``True``, prevent the use of unassigned code points.
"""
if sys.version_info < (3, 0):
if type(text) == str:
text = text.decode('utf-8')
# Mapping:
#
# - non-ASCII space characters [StringPrep, C.1.2] that can be
# mapped to SPACE (U+0020), and
#
# - the 'commonly mapped to nothing' characters [StringPrep, B.1]
# that can be mapped to nothing.
buffer = ''
for char in text:
if stringprep.in_table_c12(char):
buffer += ' '
elif not stringprep.in_table_b1(char):
buffer += char
# Normalization using form KC
text = unicodedata.normalize('NFKC', buffer)
# Check for bidirectional string
buffer = ''
first_is_randal = False
if text:
first_is_randal = stringprep.in_table_d1(text[0])
if first_is_randal and not stringprep.in_table_d1(text[-1]):
raise UnicodeError('Section 6.3 [end]')
# Check for prohibited characters
for x in range(len(text)):
if strict and stringprep.in_table_a1(text[x]):
raise UnicodeError('Unassigned Codepoint')
if stringprep.in_table_c12(text[x]):
raise UnicodeError('In table C.1.2')
if stringprep.in_table_c21(text[x]):
raise UnicodeError('In table C.2.1')
if stringprep.in_table_c22(text[x]):
raise UnicodeError('In table C.2.2')
if stringprep.in_table_c3(text[x]):
raise UnicodeError('In table C.3')
if stringprep.in_table_c4(text[x]):
raise UnicodeError('In table C.4')
if stringprep.in_table_c5(text[x]):
raise UnicodeError('In table C.5')
if stringprep.in_table_c6(text[x]):
raise UnicodeError('In table C.6')
if stringprep.in_table_c7(text[x]):
raise UnicodeError('In table C.7')
if stringprep.in_table_c8(text[x]):
raise UnicodeError('In table C.8')
if stringprep.in_table_c9(text[x]):
raise UnicodeError('In table C.9')
if x:
if first_is_randal and stringprep.in_table_d2(text[x]):
raise UnicodeError('Section 6.2')
if not first_is_randal and \
x != len(text) - 1 and \
stringprep.in_table_d1(text[x]):
raise UnicodeError('Section 6.3')
return text