##// END OF EJS Templates
tests: avoid grafting the same change over and over...
tests: avoid grafting the same change over and over The test case added in a1381eea7c7d (graft: do not use `.remove` on a smart set (regression), 2014-04-28) added a test case that grafted the same change (renaming 'a' to 'b') three times over. It had description "graft works on complex revset", but AFACT, all that it cared about was that some ancestor of the working copy was in the set of revisions to graft. So this patch changes the test to do that instead. (I plan to later make it so that grafting these renames on top of each won't create the empty commits they currently create.) Differential Revision: https://phab.mercurial-scm.org/D7804

File last commit:

r44082:4cd91104 stable
r44496:f3ad014b default
Show More
base85.py
88 lines | 2.0 KiB | text/x-python | PythonLexer
# base85.py: pure python base85 codec
#
# Copyright (C) 2009 Brendan Cully <brendan@kublai.com>
#
# This software may be used and distributed according to the terms of the
# GNU General Public License version 2 or any later version.
from __future__ import absolute_import
import struct
from .. import pycompat
_b85chars = pycompat.bytestr(
b"0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdef"
b"ghijklmnopqrstuvwxyz!#$%&()*+-;<=>?@^_`{|}~"
)
_b85chars2 = [(a + b) for a in _b85chars for b in _b85chars]
_b85dec = {}
def _mkb85dec():
for i, c in enumerate(_b85chars):
_b85dec[c] = i
def b85encode(text, pad=False):
"""encode text in base85 format"""
l = len(text)
r = l % 4
if r:
text += b'\0' * (4 - r)
longs = len(text) >> 2
words = struct.unpack(b'>%dL' % longs, text)
out = b''.join(
_b85chars[(word // 52200625) % 85]
+ _b85chars2[(word // 7225) % 7225]
+ _b85chars2[word % 7225]
for word in words
)
if pad:
return out
# Trim padding
olen = l % 4
if olen:
olen += 1
olen += l // 4 * 5
return out[:olen]
def b85decode(text):
"""decode base85-encoded text"""
if not _b85dec:
_mkb85dec()
l = len(text)
out = []
for i in range(0, len(text), 5):
chunk = text[i : i + 5]
chunk = pycompat.bytestr(chunk)
acc = 0
for j, c in enumerate(chunk):
try:
acc = acc * 85 + _b85dec[c]
except KeyError:
raise ValueError(
'bad base85 character at position %d' % (i + j)
)
if acc > 4294967295:
raise ValueError('Base85 overflow in hunk starting at byte %d' % i)
out.append(acc)
# Pad final chunk if necessary
cl = l % 5
if cl:
acc *= 85 ** (5 - cl)
if cl > 1:
acc += 0xFFFFFF >> (cl - 2) * 8
out[-1] = acc
out = struct.pack(b'>%dL' % (len(out)), *out)
if cl:
out = out[: -(5 - cl)]
return out