File: password_obfuscation.py

package info (click to toggle)
python-irodsclient 3.0.0-1
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid, trixie
  • size: 1,352 kB
  • sloc: python: 16,650; xml: 525; sh: 104; awk: 5; sql: 3; makefile: 3
file content (412 lines) | stat: -rw-r--r-- 10,229 bytes parent folder | download
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
import hashlib
import sys
import os
import time
import random
import string
from irods import MAX_PASSWORD_LENGTH

seq_list = [
    0xD768B678,
    0xEDFDAF56,
    0x2420231B,
    0x987098D8,
    0xC1BDFEEE,
    0xF572341F,
    0x478DEF3A,
    0xA830D343,
    0x774DFA2A,
    0x6720731E,
    0x346FA320,
    0x6FFDF43A,
    0x7723A320,
    0xDF67D02E,
    0x86AD240A,
    0xE76D342E,
]

# Don't forget to drink your Ovaltine
wheel = [
    "0",
    "1",
    "2",
    "3",
    "4",
    "5",
    "6",
    "7",
    "8",
    "9",
    "A",
    "B",
    "C",
    "D",
    "E",
    "F",
    "G",
    "H",
    "I",
    "J",
    "K",
    "L",
    "M",
    "N",
    "O",
    "P",
    "Q",
    "R",
    "S",
    "T",
    "U",
    "V",
    "W",
    "X",
    "Y",
    "Z",
    "a",
    "b",
    "c",
    "d",
    "e",
    "f",
    "g",
    "h",
    "i",
    "j",
    "k",
    "l",
    "m",
    "n",
    "o",
    "p",
    "q",
    "r",
    "s",
    "t",
    "u",
    "v",
    "w",
    "x",
    "y",
    "z",
    "!",
    '"',
    "#",
    "$",
    "%",
    "&",
    "'",
    "(",
    ")",
    "*",
    "+",
    ",",
    "-",
    ".",
    "/",
]

default_password_key = "a9_3fker"
default_scramble_prefix = ".E_"
v2_prefix = "A.ObfV2"


def str_to_int(input_string):
    """Convert `input_string` into an integer.

    Parameters
    ----------
    input_string : str
        Value to convert.

    Returns
    -------
    int
        Converted value.
    """
    return sum((ord(char) for char in input_string))


# Decode a password from a .irodsA file
def decode(s, uid=None):
    # This value lets us know which seq value to use
    # Referred to as "rval" in the C code
    seq_index = ord(s[6]) - ord("e")
    seq = seq_list[seq_index]

    # How much we bitshift seq by when we use it
    # Referred to as "addin_i" in the C code
    # Since we're skipping five bytes that are normally read,
    # we start at 15
    bitshift = 15

    # The uid is used as a salt.
    if uid is None:
        try:
            uid = os.getuid()
        except AttributeError:
            # Spoof UID for Non-POSIX
            uid = str_to_int(os.getlogin())

    # The first byte is a dot, the next five are literally irrelevant
    # garbage, and we already used the seventh one. The string to decode
    # starts at byte eight.
    encoded_string = s[7:]

    decoded_string = ""

    for c in encoded_string:
        if ord(c) == 0:
            break
        # How far this character is from the target character in wheel
        # Referred to as "add_in" in the C code
        offset = ((seq >> bitshift) & 0x1F) + (uid & 0xF5F)

        bitshift += 3
        if bitshift > 28:
            bitshift = 0

        # The character is only encoded if it's one of the ones in wheel
        if c in wheel:
            # index of the target character in wheel
            wheel_index = (len(wheel) + wheel.index(c) - offset) % len(wheel)
            decoded_string += wheel[wheel_index]
        else:
            decoded_string += c

    return decoded_string


# encode passwords to store in the .irodsA file
def encode(s, uid=None, mtime=None):
    # mtime & 65535 needs to be within 20 seconds of the
    # .irodsA file's mtime & 65535
    if mtime is None:
        mtime = int(time.time())

    # How much we bitshift seq by when we use it
    # Referred to as "addin_i" in the C code
    # We can't skip the first five bytes this time,
    # so we start at 0
    bitshift = 0

    # The uid is used as a salt.
    if uid is None:
        try:
            uid = os.getuid()
        except AttributeError:
            # Spoof UID for Non-POSIX
            uid = str_to_int(os.getlogin())

    # This value lets us know which seq value to use
    # Referred to as "rval" in the C code
    # The C code is very specific about this being mtime & 15,
    # but it's never checked. Let's use zero.
    seq_index = 0
    seq = seq_list[seq_index]

    to_encode = ""

    # The C code DOES really care about this value matching
    # the seq_index, though
    to_encode += chr(ord("S") - ((seq_index & 0x7) * 2))

    # And this is also a song and dance to
    # convince the C code we are legitimate
    to_encode += chr(((mtime >> 4) & 0xF) + ord("a"))
    to_encode += chr((mtime & 0xF) + ord("a"))
    to_encode += chr(((mtime >> 12) & 0xF) + ord("a"))
    to_encode += chr(((mtime >> 8) & 0xF) + ord("a"))

    # We also want to actually encode the passed string
    to_encode += s

    # Yeah, the string starts with a dot. Whatever.
    encoded_string = "."

    for c in to_encode:
        if ord(c) == 0:
            break

        # How far this character is from the target character in wheel
        # Referred to as "add_in" in the C code
        offset = ((seq >> bitshift) & 0x1F) + (uid & 0xF5F)

        bitshift += 3
        if bitshift > 28:
            bitshift = 0

        # The character is only encoded if it's one of the ones in wheel
        if c in wheel:
            # index of the target character in wheel
            wheel_index = (wheel.index(c) + offset) % len(wheel)
            encoded_string += wheel[wheel_index]
        else:
            encoded_string += c

    # insert the seq_index (which is NOT encoded):
    encoded_string = chr(seq_index + ord("e")).join(
        [encoded_string[:6], encoded_string[6:]]
    )

    # aaaaand, append a null character. because we want to print
    # a null character to the file. because that's a good idea.
    encoded_string += chr(0)

    return encoded_string


# Hash some stuff to create ANOTHER encoder ring.
def get_encoder_ring(key=default_password_key):

    md5_hasher = hashlib.md5()
    # key (called keyBuf in the C) is padded
    # or truncated to 100 characters
    md5_hasher.update(key.ljust(100, chr(0))[:100].encode("ascii"))
    first = md5_hasher.digest()

    md5_hasher = hashlib.md5()
    md5_hasher.update(first)
    second = md5_hasher.digest()

    md5_hasher = hashlib.md5()
    md5_hasher.update(first + second)
    third = md5_hasher.digest()

    return first + second + third + third


# unscramble passwords stored in the database
def unscramble(
    s,
    key=default_password_key,
    scramble_prefix=default_scramble_prefix,
    block_chaining=False,
):
    if key is None:
        key = default_password_key

    if not s.startswith(scramble_prefix):
        # not scrambled. or if it is, not
        # in a way we can unscramble
        return s

    # the prefix is nonsense, strip it off
    to_unscramble = s[len(scramble_prefix) :]

    # get our encoder ring (called cpKey in the C code)
    encoder_ring = get_encoder_ring(key)
    encoder_ring_index = 0

    # for block chaining
    chain = 0

    unscrambled_string = ""
    for c in to_unscramble:
        if c in wheel:
            # the index of the target character in wheel
            wheel_index = (
                wheel.index(c) - encoder_ring[encoder_ring_index % 61] - chain
            ) % len(wheel)
            unscrambled_string += wheel[wheel_index]
            if block_chaining:
                chain = ord(c) & 0xFF
        else:
            unscrambled_string += c
        encoder_ring_index += 1

    return unscrambled_string


# scramble passwords to store in the database
def scramble(
    s,
    key=default_password_key,
    scramble_prefix=default_scramble_prefix,
    block_chaining=False,
):
    if key is None:
        key = default_password_key

    to_scramble = s

    # get our encoder ring (called cpKey in the C code)
    encoder_ring = get_encoder_ring(key)
    encoder_ring_index = 0

    # for block chaining
    chain = 0

    scrambled_string = ""
    for c in to_scramble:
        if c in wheel:
            # the index of the target character in wheel
            wheel_index = (
                wheel.index(c) + encoder_ring[encoder_ring_index % 61] + chain
            ) % len(wheel)
            scrambled_string += wheel[wheel_index]
            if block_chaining:
                chain = ord(scrambled_string[-1]) & 0xFF
        else:
            scrambled_string += c
        encoder_ring_index += 1

    return scramble_prefix + scrambled_string


# port of https://github.com/irods/irods/blob/4-2-stable/lib/core/src/obf.cpp#L1113
def scramble_v2(s, first_key, second_key):

    to_scramble = (
        random.SystemRandom().choice(string.printable) + v2_prefix[1:10] + s[:150]
    )

    key = first_key[:90] + second_key[:100]

    # Must pad the key with null bytes to generate the same hash as the server code
    # https://github.com/irods/irods/blob/4.2.1/lib/core/src/obf.cpp#L1303-L1314
    if sys.version_info > (2, 7, 7):
        key = "{:\x00<100}".format(key)
    else:
        # https://bugs.python.org/issue12546
        key += "\x00" * max(100 - len(key), 0)

    md5_hasher = hashlib.md5()
    md5_hasher.update(key.encode("utf-8"))
    hashed_key = md5_hasher.hexdigest()

    return scramble(
        to_scramble, key=hashed_key, scramble_prefix="", block_chaining=True
    )


def obfuscate_new_password_with_key(new_password, obfuscation_key):
    lcopy = MAX_PASSWORD_LENGTH - 10 - len(new_password)
    new_password = new_password[:MAX_PASSWORD_LENGTH]
    if lcopy > 15:
        # https://github.com/irods/irods/blob/4.2.1/plugins/database/src/db_plugin.cpp#L1094-L1095
        padding = "1gCBizHWbwIYyWLoysGzTe6SyzqFKMniZX05faZHWAwQKXf6Fs"
        new_password += padding[:lcopy]
    return scramble(new_password, obfuscation_key, scramble_prefix="")


# port of https://github.com/irods/irods_client_icommands/blob/4.2.1/src/iadmin.cpp#L878-L930
def obfuscate_new_password(new, old, signature):
    pwd_len = len(new)
    new = new[:MAX_PASSWORD_LENGTH]
    lcopy = MAX_PASSWORD_LENGTH - 10 - pwd_len

    if lcopy > 15:
        # https://github.com/irods/irods/blob/4.2.1/plugins/database/src/db_plugin.cpp#L1094-L1095
        padding = "1gCBizHWbwIYyWLoysGzTe6SyzqFKMniZX05faZHWAwQKXf6Fs"
        new = new + padding[:lcopy]

    return scramble_v2(new, old, signature)


def create_temp_password(temp_hash, source_password):
    password = (temp_hash + source_password).ljust(100, chr(0))
    password_md5 = hashlib.md5(password.encode("utf-8"))

    # Return hexdigest
    return password_md5.hexdigest()