dirstate: avoid a race with multiple commits in the same process
(issue2264, issue2516)
The race happens when two commits in a row change the same file
without changing its size, *if* those two commits happen in the same
second in the same process while holding the same repo lock. For
example:
commit 1:
M a
M b
commit 2: # same process, same second, same repo lock
M b # modify b without changing its size
M c
This first manifested in transplant, which is the most common way to
do multiple commits in the same process. But it can manifest in any
script or extension that does multiple commits under the same repo
lock. (Thus, the test script tests both transplant and a custom script.)
The problem was that dirstate.status() failed to notice the change to
b when localrepo is about to do the second commit, meaning that change
gets left in the working directory. In the context of transplant, that
means either a crash ("RuntimeError: nothing committed after
transplant") or a silently inaccurate transplant, depending on whether
any other files were modified by the second transplanted changeset.
The fix is to make status() work a little harder when we have
previously marked files as clean (state 'normal') in the same process.
Specifically, dirstate.normal() adds files to self._lastnormal, and
other state-changing methods remove them. Then dirstate.status() puts
any files in self._lastnormal into state 'lookup', which will make
localrepository.status() read file contents to see if it has really
changed. So we pay a small performance penalty for the second (and
subsequent) commits in the same process, without affecting the common
case. Anything that does lots of status updates and checks in the
same process could suffer a performance hit.
Incidentally, there is a simpler fix: call dirstate.normallookup() on
every file updated by commit() at the end of the commit. The trouble
with that solution is that it imposes a performance penalty on the
common case: it means the next status-dependent hg command after every
"hg commit" will be a little bit slower. The patch here is more
complex, but only affects performance for the uncommon case.
# config.py - configuration parsing for Mercurial
#
# Copyright 2009 Matt Mackall <mpm@selenic.com> and others
#
# This software may be used and distributed according to the terms of the
# GNU General Public License version 2 or any later version.
from i18n import _
import error, util
import re, os
class sortdict(dict):
'a simple sorted dictionary'
def __init__(self, data=None):
self._list = []
if data:
self.update(data)
def copy(self):
return sortdict(self)
def __setitem__(self, key, val):
if key in self:
self._list.remove(key)
self._list.append(key)
dict.__setitem__(self, key, val)
def __iter__(self):
return self._list.__iter__()
def update(self, src):
for k in src:
self[k] = src[k]
def items(self):
return [(k, self[k]) for k in self._list]
def __delitem__(self, key):
dict.__delitem__(self, key)
self._list.remove(key)
class config(object):
def __init__(self, data=None):
self._data = {}
self._source = {}
if data:
for k in data._data:
self._data[k] = data[k].copy()
self._source = data._source.copy()
def copy(self):
return config(self)
def __contains__(self, section):
return section in self._data
def __getitem__(self, section):
return self._data.get(section, {})
def __iter__(self):
for d in self.sections():
yield d
def update(self, src):
for s in src:
if s not in self:
self._data[s] = sortdict()
self._data[s].update(src._data[s])
self._source.update(src._source)
def get(self, section, item, default=None):
return self._data.get(section, {}).get(item, default)
def source(self, section, item):
return self._source.get((section, item), "")
def sections(self):
return sorted(self._data.keys())
def items(self, section):
return self._data.get(section, {}).items()
def set(self, section, item, value, source=""):
if section not in self:
self._data[section] = sortdict()
self._data[section][item] = value
self._source[(section, item)] = source
def parse(self, src, data, sections=None, remap=None, include=None):
sectionre = re.compile(r'\[([^\[]+)\]')
itemre = re.compile(r'([^=\s][^=]*?)\s*=\s*(.*\S|)')
contre = re.compile(r'\s+(\S|\S.*\S)\s*$')
emptyre = re.compile(r'(;|#|\s*$)')
unsetre = re.compile(r'%unset\s+(\S+)')
includere = re.compile(r'%include\s+(\S|\S.*\S)\s*$')
section = ""
item = None
line = 0
cont = False
for l in data.splitlines(True):
line += 1
if cont:
m = contre.match(l)
if m:
if sections and section not in sections:
continue
v = self.get(section, item) + "\n" + m.group(1)
self.set(section, item, v, "%s:%d" % (src, line))
continue
item = None
cont = False
m = includere.match(l)
if m:
inc = util.expandpath(m.group(1))
base = os.path.dirname(src)
inc = os.path.normpath(os.path.join(base, inc))
if include:
try:
include(inc, remap=remap, sections=sections)
except IOError, inst:
raise error.ParseError(_("cannot include %s (%s)")
% (inc, inst.strerror),
"%s:%s" % (src, line))
continue
if emptyre.match(l):
continue
m = sectionre.match(l)
if m:
section = m.group(1)
if remap:
section = remap.get(section, section)
if section not in self:
self._data[section] = sortdict()
continue
m = itemre.match(l)
if m:
item = m.group(1)
cont = True
if sections and section not in sections:
continue
self.set(section, item, m.group(2), "%s:%d" % (src, line))
continue
m = unsetre.match(l)
if m:
name = m.group(1)
if sections and section not in sections:
continue
if self.get(section, name) is not None:
del self._data[section][name]
continue
raise error.ParseError(l.rstrip(), ("%s:%s" % (src, line)))
def read(self, path, fp=None, sections=None, remap=None):
if not fp:
fp = util.posixfile(path)
self.parse(path, fp.read(), sections, remap, self.read)