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.
# error.py - Mercurial exceptions
#
# Copyright 2005-2008 Matt Mackall <mpm@selenic.com>
#
# This software may be used and distributed according to the terms of the
# GNU General Public License version 2 or any later version.
"""Mercurial exceptions.
This allows us to catch exceptions at higher levels without forcing
imports.
"""
# Do not import anything here, please
class RevlogError(Exception):
pass
class LookupError(RevlogError, KeyError):
def __init__(self, name, index, message):
self.name = name
if isinstance(name, str) and len(name) == 20:
from node import short
name = short(name)
RevlogError.__init__(self, '%s@%s: %s' % (index, name, message))
def __str__(self):
return RevlogError.__str__(self)
class CommandError(Exception):
"""Exception raised on errors in parsing the command line."""
class Abort(Exception):
"""Raised if a command needs to print an error and exit."""
def __init__(self, *args, **kw):
Exception.__init__(self, *args)
self.hint = kw.get('hint')
class ConfigError(Abort):
'Exception raised when parsing config files'
class ParseError(Exception):
'Exception raised when parsing config files (msg[, pos])'
class RepoError(Exception):
pass
class RepoLookupError(RepoError):
pass
class CapabilityError(RepoError):
pass
class RequirementError(RepoError):
"""Exception raised if .hg/requires has an unknown entry."""
pass
class LockError(IOError):
def __init__(self, errno, strerror, filename, desc):
IOError.__init__(self, errno, strerror, filename)
self.desc = desc
class LockHeld(LockError):
def __init__(self, errno, filename, desc, locker):
LockError.__init__(self, errno, 'Lock held', filename, desc)
self.locker = locker
class LockUnavailable(LockError):
pass
class ResponseError(Exception):
"""Raised to print an error with part of output and exit."""
class UnknownCommand(Exception):
"""Exception raised if command is not in the command table."""
class AmbiguousCommand(Exception):
"""Exception raised if command shortcut matches more than one command."""
# derived from KeyboardInterrupt to simplify some breakout code
class SignalInterrupt(KeyboardInterrupt):
"""Exception raised on SIGTERM and SIGHUP."""
class SignatureError(Exception):
pass