Mercurial > repos > rliterman > csp2

annotate CSP2/CSP2_env/env-d9b9114564458d9d-741b3de822f2aaca6c6caa4325c4afce/lib/python3.8/encodings/idna.py @ 68:5028fdace37b

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload commit 2e9511a184a1ca667c7be0c6321a36dc4e3d116d
author jpayne
date Tue, 18 Mar 2025 16:23:26 -0400
parents
children
rev   line source
jpayne@68 1 # This module implements the RFCs 3490 (IDNA) and 3491 (Nameprep)
jpayne@68 2
jpayne@68 3 import stringprep, re, codecs
jpayne@68 4 from unicodedata import ucd_3_2_0 as unicodedata
jpayne@68 5
jpayne@68 6 # IDNA section 3.1
jpayne@68 7 dots = re.compile("[\u002E\u3002\uFF0E\uFF61]")
jpayne@68 8
jpayne@68 9 # IDNA section 5
jpayne@68 10 ace_prefix = b"xn--"
jpayne@68 11 sace_prefix = "xn--"
jpayne@68 12
jpayne@68 13 # This assumes query strings, so AllowUnassigned is true
jpayne@68 14 def nameprep(label):
jpayne@68 15 # Map
jpayne@68 16 newlabel = []
jpayne@68 17 for c in label:
jpayne@68 18 if stringprep.in_table_b1(c):
jpayne@68 19 # Map to nothing
jpayne@68 20 continue
jpayne@68 21 newlabel.append(stringprep.map_table_b2(c))
jpayne@68 22 label = "".join(newlabel)
jpayne@68 23
jpayne@68 24 # Normalize
jpayne@68 25 label = unicodedata.normalize("NFKC", label)
jpayne@68 26
jpayne@68 27 # Prohibit
jpayne@68 28 for c in label:
jpayne@68 29 if stringprep.in_table_c12(c) or \
jpayne@68 30 stringprep.in_table_c22(c) or \
jpayne@68 31 stringprep.in_table_c3(c) or \
jpayne@68 32 stringprep.in_table_c4(c) or \
jpayne@68 33 stringprep.in_table_c5(c) or \
jpayne@68 34 stringprep.in_table_c6(c) or \
jpayne@68 35 stringprep.in_table_c7(c) or \
jpayne@68 36 stringprep.in_table_c8(c) or \
jpayne@68 37 stringprep.in_table_c9(c):
jpayne@68 38 raise UnicodeError("Invalid character %r" % c)
jpayne@68 39
jpayne@68 40 # Check bidi
jpayne@68 41 RandAL = [stringprep.in_table_d1(x) for x in label]
jpayne@68 42 for c in RandAL:
jpayne@68 43 if c:
jpayne@68 44 # There is a RandAL char in the string. Must perform further
jpayne@68 45 # tests:
jpayne@68 46 # 1) The characters in section 5.8 MUST be prohibited.
jpayne@68 47 # This is table C.8, which was already checked
jpayne@68 48 # 2) If a string contains any RandALCat character, the string
jpayne@68 49 # MUST NOT contain any LCat character.
jpayne@68 50 if any(stringprep.in_table_d2(x) for x in label):
jpayne@68 51 raise UnicodeError("Violation of BIDI requirement 2")
jpayne@68 52
jpayne@68 53 # 3) If a string contains any RandALCat character, a
jpayne@68 54 # RandALCat character MUST be the first character of the
jpayne@68 55 # string, and a RandALCat character MUST be the last
jpayne@68 56 # character of the string.
jpayne@68 57 if not RandAL[0] or not RandAL[-1]:
jpayne@68 58 raise UnicodeError("Violation of BIDI requirement 3")
jpayne@68 59
jpayne@68 60 return label
jpayne@68 61
jpayne@68 62 def ToASCII(label):
jpayne@68 63 try:
jpayne@68 64 # Step 1: try ASCII
jpayne@68 65 label = label.encode("ascii")
jpayne@68 66 except UnicodeError:
jpayne@68 67 pass
jpayne@68 68 else:
jpayne@68 69 # Skip to step 3: UseSTD3ASCIIRules is false, so
jpayne@68 70 # Skip to step 8.
jpayne@68 71 if 0 < len(label) < 64:
jpayne@68 72 return label
jpayne@68 73 raise UnicodeError("label empty or too long")
jpayne@68 74
jpayne@68 75 # Step 2: nameprep
jpayne@68 76 label = nameprep(label)
jpayne@68 77
jpayne@68 78 # Step 3: UseSTD3ASCIIRules is false
jpayne@68 79 # Step 4: try ASCII
jpayne@68 80 try:
jpayne@68 81 label = label.encode("ascii")
jpayne@68 82 except UnicodeError:
jpayne@68 83 pass
jpayne@68 84 else:
jpayne@68 85 # Skip to step 8.
jpayne@68 86 if 0 < len(label) < 64:
jpayne@68 87 return label
jpayne@68 88 raise UnicodeError("label empty or too long")
jpayne@68 89
jpayne@68 90 # Step 5: Check ACE prefix
jpayne@68 91 if label.startswith(sace_prefix):
jpayne@68 92 raise UnicodeError("Label starts with ACE prefix")
jpayne@68 93
jpayne@68 94 # Step 6: Encode with PUNYCODE
jpayne@68 95 label = label.encode("punycode")
jpayne@68 96
jpayne@68 97 # Step 7: Prepend ACE prefix
jpayne@68 98 label = ace_prefix + label
jpayne@68 99
jpayne@68 100 # Step 8: Check size
jpayne@68 101 if 0 < len(label) < 64:
jpayne@68 102 return label
jpayne@68 103 raise UnicodeError("label empty or too long")
jpayne@68 104
jpayne@68 105 def ToUnicode(label):
jpayne@68 106 # Step 1: Check for ASCII
jpayne@68 107 if isinstance(label, bytes):
jpayne@68 108 pure_ascii = True
jpayne@68 109 else:
jpayne@68 110 try:
jpayne@68 111 label = label.encode("ascii")
jpayne@68 112 pure_ascii = True
jpayne@68 113 except UnicodeError:
jpayne@68 114 pure_ascii = False
jpayne@68 115 if not pure_ascii:
jpayne@68 116 # Step 2: Perform nameprep
jpayne@68 117 label = nameprep(label)
jpayne@68 118 # It doesn't say this, but apparently, it should be ASCII now
jpayne@68 119 try:
jpayne@68 120 label = label.encode("ascii")
jpayne@68 121 except UnicodeError:
jpayne@68 122 raise UnicodeError("Invalid character in IDN label")
jpayne@68 123 # Step 3: Check for ACE prefix
jpayne@68 124 if not label.startswith(ace_prefix):
jpayne@68 125 return str(label, "ascii")
jpayne@68 126
jpayne@68 127 # Step 4: Remove ACE prefix
jpayne@68 128 label1 = label[len(ace_prefix):]
jpayne@68 129
jpayne@68 130 # Step 5: Decode using PUNYCODE
jpayne@68 131 result = label1.decode("punycode")
jpayne@68 132
jpayne@68 133 # Step 6: Apply ToASCII
jpayne@68 134 label2 = ToASCII(result)
jpayne@68 135
jpayne@68 136 # Step 7: Compare the result of step 6 with the one of step 3
jpayne@68 137 # label2 will already be in lower case.
jpayne@68 138 if str(label, "ascii").lower() != str(label2, "ascii"):
jpayne@68 139 raise UnicodeError("IDNA does not round-trip", label, label2)
jpayne@68 140
jpayne@68 141 # Step 8: return the result of step 5
jpayne@68 142 return result
jpayne@68 143
jpayne@68 144 ### Codec APIs
jpayne@68 145
jpayne@68 146 class Codec(codecs.Codec):
jpayne@68 147 def encode(self, input, errors='strict'):
jpayne@68 148
jpayne@68 149 if errors != 'strict':
jpayne@68 150 # IDNA is quite clear that implementations must be strict
jpayne@68 151 raise UnicodeError("unsupported error handling "+errors)
jpayne@68 152
jpayne@68 153 if not input:
jpayne@68 154 return b'', 0
jpayne@68 155
jpayne@68 156 try:
jpayne@68 157 result = input.encode('ascii')
jpayne@68 158 except UnicodeEncodeError:
jpayne@68 159 pass
jpayne@68 160 else:
jpayne@68 161 # ASCII name: fast path
jpayne@68 162 labels = result.split(b'.')
jpayne@68 163 for label in labels[:-1]:
jpayne@68 164 if not (0 < len(label) < 64):
jpayne@68 165 raise UnicodeError("label empty or too long")
jpayne@68 166 if len(labels[-1]) >= 64:
jpayne@68 167 raise UnicodeError("label too long")
jpayne@68 168 return result, len(input)
jpayne@68 169
jpayne@68 170 result = bytearray()
jpayne@68 171 labels = dots.split(input)
jpayne@68 172 if labels and not labels[-1]:
jpayne@68 173 trailing_dot = b'.'
jpayne@68 174 del labels[-1]
jpayne@68 175 else:
jpayne@68 176 trailing_dot = b''
jpayne@68 177 for label in labels:
jpayne@68 178 if result:
jpayne@68 179 # Join with U+002E
jpayne@68 180 result.extend(b'.')
jpayne@68 181 result.extend(ToASCII(label))
jpayne@68 182 return bytes(result+trailing_dot), len(input)
jpayne@68 183
jpayne@68 184 def decode(self, input, errors='strict'):
jpayne@68 185
jpayne@68 186 if errors != 'strict':
jpayne@68 187 raise UnicodeError("Unsupported error handling "+errors)
jpayne@68 188
jpayne@68 189 if not input:
jpayne@68 190 return "", 0
jpayne@68 191
jpayne@68 192 # IDNA allows decoding to operate on Unicode strings, too.
jpayne@68 193 if not isinstance(input, bytes):
jpayne@68 194 # XXX obviously wrong, see #3232
jpayne@68 195 input = bytes(input)
jpayne@68 196
jpayne@68 197 if ace_prefix not in input:
jpayne@68 198 # Fast path
jpayne@68 199 try:
jpayne@68 200 return input.decode('ascii'), len(input)
jpayne@68 201 except UnicodeDecodeError:
jpayne@68 202 pass
jpayne@68 203
jpayne@68 204 labels = input.split(b".")
jpayne@68 205
jpayne@68 206 if labels and len(labels[-1]) == 0:
jpayne@68 207 trailing_dot = '.'
jpayne@68 208 del labels[-1]
jpayne@68 209 else:
jpayne@68 210 trailing_dot = ''
jpayne@68 211
jpayne@68 212 result = []
jpayne@68 213 for label in labels:
jpayne@68 214 result.append(ToUnicode(label))
jpayne@68 215
jpayne@68 216 return ".".join(result)+trailing_dot, len(input)
jpayne@68 217
jpayne@68 218 class IncrementalEncoder(codecs.BufferedIncrementalEncoder):
jpayne@68 219 def _buffer_encode(self, input, errors, final):
jpayne@68 220 if errors != 'strict':
jpayne@68 221 # IDNA is quite clear that implementations must be strict
jpayne@68 222 raise UnicodeError("unsupported error handling "+errors)
jpayne@68 223
jpayne@68 224 if not input:
jpayne@68 225 return (b'', 0)
jpayne@68 226
jpayne@68 227 labels = dots.split(input)
jpayne@68 228 trailing_dot = b''
jpayne@68 229 if labels:
jpayne@68 230 if not labels[-1]:
jpayne@68 231 trailing_dot = b'.'
jpayne@68 232 del labels[-1]
jpayne@68 233 elif not final:
jpayne@68 234 # Keep potentially unfinished label until the next call
jpayne@68 235 del labels[-1]
jpayne@68 236 if labels:
jpayne@68 237 trailing_dot = b'.'
jpayne@68 238
jpayne@68 239 result = bytearray()
jpayne@68 240 size = 0
jpayne@68 241 for label in labels:
jpayne@68 242 if size:
jpayne@68 243 # Join with U+002E
jpayne@68 244 result.extend(b'.')
jpayne@68 245 size += 1
jpayne@68 246 result.extend(ToASCII(label))
jpayne@68 247 size += len(label)
jpayne@68 248
jpayne@68 249 result += trailing_dot
jpayne@68 250 size += len(trailing_dot)
jpayne@68 251 return (bytes(result), size)
jpayne@68 252
jpayne@68 253 class IncrementalDecoder(codecs.BufferedIncrementalDecoder):
jpayne@68 254 def _buffer_decode(self, input, errors, final):
jpayne@68 255 if errors != 'strict':
jpayne@68 256 raise UnicodeError("Unsupported error handling "+errors)
jpayne@68 257
jpayne@68 258 if not input:
jpayne@68 259 return ("", 0)
jpayne@68 260
jpayne@68 261 # IDNA allows decoding to operate on Unicode strings, too.
jpayne@68 262 if isinstance(input, str):
jpayne@68 263 labels = dots.split(input)
jpayne@68 264 else:
jpayne@68 265 # Must be ASCII string
jpayne@68 266 input = str(input, "ascii")
jpayne@68 267 labels = input.split(".")
jpayne@68 268
jpayne@68 269 trailing_dot = ''
jpayne@68 270 if labels:
jpayne@68 271 if not labels[-1]:
jpayne@68 272 trailing_dot = '.'
jpayne@68 273 del labels[-1]
jpayne@68 274 elif not final:
jpayne@68 275 # Keep potentially unfinished label until the next call
jpayne@68 276 del labels[-1]
jpayne@68 277 if labels:
jpayne@68 278 trailing_dot = '.'
jpayne@68 279
jpayne@68 280 result = []
jpayne@68 281 size = 0
jpayne@68 282 for label in labels:
jpayne@68 283 result.append(ToUnicode(label))
jpayne@68 284 if size:
jpayne@68 285 size += 1
jpayne@68 286 size += len(label)
jpayne@68 287
jpayne@68 288 result = ".".join(result) + trailing_dot
jpayne@68 289 size += len(trailing_dot)
jpayne@68 290 return (result, size)
jpayne@68 291
jpayne@68 292 class StreamWriter(Codec,codecs.StreamWriter):
jpayne@68 293 pass
jpayne@68 294
jpayne@68 295 class StreamReader(Codec,codecs.StreamReader):
jpayne@68 296 pass
jpayne@68 297
jpayne@68 298 ### encodings module API
jpayne@68 299
jpayne@68 300 def getregentry():
jpayne@68 301 return codecs.CodecInfo(
jpayne@68 302 name='idna',
jpayne@68 303 encode=Codec().encode,
jpayne@68 304 decode=Codec().decode,
jpayne@68 305 incrementalencoder=IncrementalEncoder,
jpayne@68 306 incrementaldecoder=IncrementalDecoder,
jpayne@68 307 streamwriter=StreamWriter,
jpayne@68 308 streamreader=StreamReader,
jpayne@68 309 )