mercurial/pure/base85.py
author Arseniy Alekseyev <aalekseyev@janestreet.com>
Fri, 26 Apr 2024 19:10:35 +0100
changeset 51626 865efc020c33
parent 48875 6000f5b25c9b
permissions -rw-r--r--
dirstate: remove the python-side whitelist of allowed matchers This whitelist is too permissive because it allows matchers that contain disallowed ones deep inside, for example through `intersectionmatcher`. It is also too restrictive because it doesn't pass through some of the matchers we support, such as `patternmatcher`. It's also unnecessary because unsupported matchers raise `FallbackError` and we fall back anyway. Making this change makes more of the tests use rust code path, and therefore subtly change behavior. For example, rust status in largefiles repos seems to have strange behavior.
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
7701
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
     1
# base85.py: pure python base85 codec
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
     2
#
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
     3
# Copyright (C) 2009 Brendan Cully <brendan@kublai.com>
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
     4
#
8225
46293a0c7e9f updated license to be explicit about GPL version 2
Martin Geisler <mg@lazybytes.net>
parents: 7881
diff changeset
     5
# This software may be used and distributed according to the terms of the
10263
25e572394f5c Update license to GPLv2+
Matt Mackall <mpm@selenic.com>
parents: 9029
diff changeset
     6
# GNU General Public License version 2 or any later version.
7701
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
     7
27334
9007f697e8ef base85: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 16598
diff changeset
     8
7701
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
     9
import struct
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    10
35944
01b4d88ccb24 py3: use pycompat.bytestr to convert _b85chars to bytes
Pulkit Goyal <7895pulkit@gmail.com>
parents: 27334
diff changeset
    11
from .. import pycompat
01b4d88ccb24 py3: use pycompat.bytestr to convert _b85chars to bytes
Pulkit Goyal <7895pulkit@gmail.com>
parents: 27334
diff changeset
    12
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 36191
diff changeset
    13
_b85chars = pycompat.bytestr(
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
    14
    b"0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdef"
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
    15
    b"ghijklmnopqrstuvwxyz!#$%&()*+-;<=>?@^_`{|}~"
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 36191
diff changeset
    16
)
7835
2505e9f84153 Optimization of pure.base85.b85encode
Mads Kiilerich <mads@kiilerich.com>
parents: 7701
diff changeset
    17
_b85chars2 = [(a + b) for a in _b85chars for b in _b85chars]
7701
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    18
_b85dec = {}
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    19
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 36191
diff changeset
    20
7701
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    21
def _mkb85dec():
8632
9e055cfdd620 replace "i in range(len(xs))" with "i, x in enumerate(xs)"
Martin Geisler <mg@lazybytes.net>
parents: 8225
diff changeset
    22
    for i, c in enumerate(_b85chars):
9e055cfdd620 replace "i in range(len(xs))" with "i, x in enumerate(xs)"
Martin Geisler <mg@lazybytes.net>
parents: 8225
diff changeset
    23
        _b85dec[c] = i
7701
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    24
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 36191
diff changeset
    25
7701
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    26
def b85encode(text, pad=False):
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    27
    """encode text in base85 format"""
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    28
    l = len(text)
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    29
    r = l % 4
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    30
    if r:
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
    31
        text += b'\0' * (4 - r)
7701
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    32
    longs = len(text) >> 2
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
    33
    words = struct.unpack(b'>%dL' % longs, text)
7701
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    34
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
    35
    out = b''.join(
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 36191
diff changeset
    36
        _b85chars[(word // 52200625) % 85]
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 36191
diff changeset
    37
        + _b85chars2[(word // 7225) % 7225]
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 36191
diff changeset
    38
        + _b85chars2[word % 7225]
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 36191
diff changeset
    39
        for word in words
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 36191
diff changeset
    40
    )
7701
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    41
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    42
    if pad:
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    43
        return out
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    44
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    45
    # Trim padding
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    46
    olen = l % 4
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    47
    if olen:
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    48
        olen += 1
9029
0001e49f1c11 compat: use // for integer division
Alejandro Santos <alejolp@alejolp.com>
parents: 8632
diff changeset
    49
    olen += l // 4 * 5
7701
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    50
    return out[:olen]
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    51
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 36191
diff changeset
    52
7701
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    53
def b85decode(text):
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    54
    """decode base85-encoded text"""
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    55
    if not _b85dec:
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    56
        _mkb85dec()
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    57
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    58
    l = len(text)
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    59
    out = []
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    60
    for i in range(0, len(text), 5):
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 36191
diff changeset
    61
        chunk = text[i : i + 5]
36191
80301c90a2dc py3: converts bytes to pycompat.bytestr to get bytechrs while enumerating
Pulkit Goyal <7895pulkit@gmail.com>
parents: 35944
diff changeset
    62
        chunk = pycompat.bytestr(chunk)
7701
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    63
        acc = 0
8632
9e055cfdd620 replace "i in range(len(xs))" with "i, x in enumerate(xs)"
Martin Geisler <mg@lazybytes.net>
parents: 8225
diff changeset
    64
        for j, c in enumerate(chunk):
7701
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    65
            try:
8632
9e055cfdd620 replace "i in range(len(xs))" with "i, x in enumerate(xs)"
Martin Geisler <mg@lazybytes.net>
parents: 8225
diff changeset
    66
                acc = acc * 85 + _b85dec[c]
7701
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    67
            except KeyError:
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 36191
diff changeset
    68
                raise ValueError(
43666
4394687b298b pure: use string for exception in the pure version of base85
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 43077
diff changeset
    69
                    'bad base85 character at position %d' % (i + j)
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 36191
diff changeset
    70
                )
7701
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    71
        if acc > 4294967295:
43667
4cd911040ba5 pure: use string for another exception in the pure version of base85
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 43666
diff changeset
    72
            raise ValueError('Base85 overflow in hunk starting at byte %d' % i)
7701
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    73
        out.append(acc)
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    74
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    75
    # Pad final chunk if necessary
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    76
    cl = l % 5
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    77
    if cl:
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    78
        acc *= 85 ** (5 - cl)
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    79
        if cl > 1:
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 36191
diff changeset
    80
            acc += 0xFFFFFF >> (cl - 2) * 8
7701
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    81
        out[-1] = acc
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    82
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
    83
    out = struct.pack(b'>%dL' % (len(out)), *out)
7701
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    84
    if cl:
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 36191
diff changeset
    85
        out = out[: -(5 - cl)]
7701
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    86
4bdead043d8d Pure python base85 fallback
Brendan Cully <brendan@kublai.com>
parents:
diff changeset
    87
    return out