##// END OF EJS Templates
manifest: avoid corruption by dropping removed files with pure (issue5801)...
manifest: avoid corruption by dropping removed files with pure (issue5801) Previously, removed files would simply be marked by overwriting the first byte with NUL and dropping their entry in `self.position`. But no effort was made to ignore them when compacting the dictionary into text form. This allowed them to slip into the manifest revision, since the code seems to be trying to minimize the string operations by copying as large a chunk as possible. As part of this, compact() walks the existing text based on entries in the `positions` list, and consumed everything up to the next position entry. This typically resulted in a ValueError complaining about unsorted manifest entries. Sometimes it seems that files do get dropped in large repos- it seems to correspond to there being a new entry that would take the same slot. A much more trivial problem is that if the only changes were removals, `_compact()` didn't even run because `__delitem__` doesn't add anything to `self.extradata`. Now there's an explicit variable to flag this, both to allow `_compact()` to run, and to avoid searching the manifest in cases where there are no removals. In practice, this behavior was mostly obscured by the check in fastdelta() which takes a different path that explicitly drops removed files if there are fewer than 1000 changes. However, timeless has a repo where after rebasing tens of commits, a totally different path[1] is taken that bypasses the change count check and hits this problem. [1] https://www.mercurial-scm.org/repo/hg/file/2338bdea4474/mercurial/manifest.py#l1511

File last commit:

r42262:edbcf5b2 default
r42569:0546ead3 stable
Show More
rcutil.py
99 lines | 3.1 KiB | text/x-python | PythonLexer
Jun Wu
rcutil: move scmutil.*rcpath to rcutil (API)...
r31679 # rcutil.py - utilities about config paths, special config sections etc.
#
# Copyright Mercurial Contributors
#
# 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 os
from . import (
encoding,
pycompat,
util,
)
Jun Wu
codemod: use pycompat.iswindows...
r34646 if pycompat.iswindows:
Jun Wu
rcutil: move scmutil.*rcpath to rcutil (API)...
r31679 from . import scmwindows as scmplatform
else:
from . import scmposix as scmplatform
Yuya Nishihara
pager: use less as a fallback on Unix...
r32078 fallbackpager = scmplatform.fallbackpager
Jun Wu
rcutil: move scmutil.*rcpath to rcutil (API)...
r31679 systemrcpath = scmplatform.systemrcpath
userrcpath = scmplatform.userrcpath
Jun Wu
rcutil: extract rc directory listing logic...
r31681 def _expandrcpath(path):
'''path could be a file or a directory. return a list of file paths'''
p = util.expandpath(path)
if os.path.isdir(p):
join = os.path.join
Martin von Zweigbergk
config: read configs from directories in lexicographical order...
r42262 return sorted(join(p, f) for f, k in util.listdir(p)
if f.endswith('.rc'))
Jun Wu
rcutil: extract rc directory listing logic...
r31681 return [p]
Jun Wu
rcutil: add a method to convert environment variables to config items...
r31684 def envrcitems(env=None):
'''Return [(section, name, value, source)] config items.
The config items are extracted from environment variables specified by env,
used to override systemrc, but not userrc.
If env is not provided, encoding.environ will be used.
'''
if env is None:
env = encoding.environ
checklist = [
('EDITOR', 'ui', 'editor'),
('VISUAL', 'ui', 'editor'),
('PAGER', 'pager', 'pager'),
]
result = []
for envname, section, configname in checklist:
if envname not in env:
continue
result.append((section, configname, env[envname], '$%s' % envname))
return result
Jun Wu
rcutil: split osrcpath to return default.d paths (API)...
r31680 def defaultrcpath():
'''return rc paths in default.d'''
Jun Wu
rcutil: move scmutil.*rcpath to rcutil (API)...
r31679 path = []
defaultpath = os.path.join(util.datapath, 'default.d')
if os.path.isdir(defaultpath):
Jun Wu
rcutil: extract rc directory listing logic...
r31681 path = _expandrcpath(defaultpath)
Jun Wu
rcutil: move scmutil.*rcpath to rcutil (API)...
r31679 return path
Jun Wu
rcutil: rename rcpath to rccomponents (API)
r31682 def rccomponents():
Jun Wu
rcutil: let rccomponents return different types of configs (API)...
r31683 '''return an ordered [(type, obj)] about where to load configs.
respect $HGRCPATH. if $HGRCPATH is empty, only .hg/hgrc of current repo is
used. if $HGRCPATH is not set, the platform default will be used.
if a directory is provided, *.rc files under it will be used.
type could be either 'path' or 'items', if type is 'path', obj is a string,
and is the config file path. if type is 'items', obj is a list of (section,
name, value, source) that should fill the config directly.
'''
Jun Wu
rcutil: let environ override system configs (BC)...
r31685 envrc = ('items', envrcitems())
Jun Wu
rcutil: unindent a block...
r31693 if 'HGRCPATH' in encoding.environ:
# assume HGRCPATH is all about user configs so environments can be
# overridden.
_rccomponents = [envrc]
for p in encoding.environ['HGRCPATH'].split(pycompat.ospathsep):
if not p:
continue
_rccomponents.extend(('path', p) for p in _expandrcpath(p))
else:
Jun Wu
rcutil: extract duplicated logic to a lambda...
r31694 normpaths = lambda paths: [('path', os.path.normpath(p)) for p in paths]
_rccomponents = normpaths(defaultrcpath() + systemrcpath())
Jun Wu
rcutil: unindent a block...
r31693 _rccomponents.append(envrc)
Jun Wu
rcutil: extract duplicated logic to a lambda...
r31694 _rccomponents.extend(normpaths(userrcpath()))
Jun Wu
rcutil: rename rcpath to rccomponents (API)
r31682 return _rccomponents
Jun Wu
pager: set some environment variables if they're not set...
r31954
def defaultpagerenv():
'''return a dict of default environment variables and their values,
intended to be set before starting a pager.
'''
return {'LESS': 'FRX', 'LV': '-c'}