dirstate-v2: Add `hg debugupgraderepo` command support

This command changes changes the file formats used inside an existing
repository to what they would be in a new repository with the current config.

For example:

    hg debugupgraderepo --config format.exp-dirstate-v2=1 --run
    hg debugupgraderepo --config format.exp-dirstate-v2=0 --run

If a repository has a dirstate in v1 format, the first command would upgrade it
to dirstate-v2. Conversely, if a repository has a dirstate in v2 format, the
second command would downgrade it to v1. (Both may also run some unrelated
upgrades.)

Since `format.exp-dirstate-v2` is currently disabled by default, not specifying
it in `--config` or any configuration file would result in the second command.

Differential Revision: https://phab.mercurial-scm.org/D10769

mercurial/dirstate.py

@@ -1,1979 +1,1980
 # dirstate.py - working directory tracking for mercurial
 #
 # Copyright 2005-2007 Olivia Mackall <[e-mail]>
 #
 # 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 collections
 import contextlib
 import errno
 import os
 import stat
 
 from .i18n import _
 from .pycompat import delattr
 
 from hgdemandimport import tracing
 
 from . import (
     encoding,
     error,
     match as matchmod,
     pathutil,
     policy,
     pycompat,
     scmutil,
     sparse,
     txnutil,
     util,
 )
 
 from .interfaces import (
     dirstate as intdirstate,
     util as interfaceutil,
 )
 
 parsers = policy.importmod('parsers')
 rustmod = policy.importrust('dirstate')
 
 SUPPORTS_DIRSTATE_V2 = rustmod is not None
 
 propertycache = util.propertycache
 filecache = scmutil.filecache
 _rangemask = 0x7FFFFFFF
 
 dirstatetuple = parsers.dirstatetuple
 
 
 class repocache(filecache):
     """filecache for files in .hg/"""
 
     def join(self, obj, fname):
         return obj._opener.join(fname)
 
 
 class rootcache(filecache):
     """filecache for files in the repository root"""
 
     def join(self, obj, fname):
         return obj._join(fname)
 
 
 def _getfsnow(vfs):
     '''Get "now" timestamp on filesystem'''
     tmpfd, tmpname = vfs.mkstemp()
     try:
         return os.fstat(tmpfd)[stat.ST_MTIME]
     finally:
         os.close(tmpfd)
         vfs.unlink(tmpname)
 
 
 @interfaceutil.implementer(intdirstate.idirstate)
 class dirstate(object):
     def __init__(
         self,
         opener,
         ui,
         root,
         validate,
         sparsematchfn,
         nodeconstants,
         use_dirstate_v2,
     ):
         """Create a new dirstate object.
 
         opener is an open()-like callable that can be used to open the
         dirstate file; root is the root of the directory tracked by
         the dirstate.
         """
         self._use_dirstate_v2 = use_dirstate_v2
         self._nodeconstants = nodeconstants
         self._opener = opener
         self._validate = validate
         self._root = root
         self._sparsematchfn = sparsematchfn
         # ntpath.join(root, '') of Python 2.7.9 does not add sep if root is
         # UNC path pointing to root share (issue4557)
         self._rootdir = pathutil.normasprefix(root)
         self._dirty = False
         self._lastnormaltime = 0
         self._ui = ui
         self._filecache = {}
         self._parentwriters = 0
         self._filename = b'dirstate'
         self._pendingfilename = b'%s.pending' % self._filename
         self._plchangecallbacks = {}
         self._origpl = None
         self._updatedfiles = set()
         self._mapcls = dirstatemap
         # Access and cache cwd early, so we don't access it for the first time
         # after a working-copy update caused it to not exist (accessing it then
         # raises an exception).
         self._cwd
 
     def prefetch_parents(self):
         """make sure the parents are loaded
 
         Used to avoid a race condition.
         """
         self._pl
 
     @contextlib.contextmanager
     def parentchange(self):
         """Context manager for handling dirstate parents.
 
         If an exception occurs in the scope of the context manager,
         the incoherent dirstate won't be written when wlock is
         released.
         """
         self._parentwriters += 1
         yield
         # Typically we want the "undo" step of a context manager in a
         # finally block so it happens even when an exception
         # occurs. In this case, however, we only want to decrement
         # parentwriters if the code in the with statement exits
         # normally, so we don't have a try/finally here on purpose.
         self._parentwriters -= 1
 
     def pendingparentchange(self):
         """Returns true if the dirstate is in the middle of a set of changes
         that modify the dirstate parent.
         """
         return self._parentwriters > 0
 
     @propertycache
     def _map(self):
         """Return the dirstate contents (see documentation for dirstatemap)."""
         self._map = self._mapcls(
             self._ui,
             self._opener,
             self._root,
             self._nodeconstants,
             self._use_dirstate_v2,
         )
         return self._map
 
     @property
     def _sparsematcher(self):
         """The matcher for the sparse checkout.
 
         The working directory may not include every file from a manifest. The
         matcher obtained by this property will match a path if it is to be
         included in the working directory.
         """
         # TODO there is potential to cache this property. For now, the matcher
         # is resolved on every access. (But the called function does use a
         # cache to keep the lookup fast.)
         return self._sparsematchfn()
 
     @repocache(b'branch')
     def _branch(self):
         try:
             return self._opener.read(b"branch").strip() or b"default"
         except IOError as inst:
             if inst.errno != errno.ENOENT:
                 raise
             return b"default"
 
     @property
     def _pl(self):
         return self._map.parents()
 
     def hasdir(self, d):
         return self._map.hastrackeddir(d)
 
     @rootcache(b'.hgignore')
     def _ignore(self):
         files = self._ignorefiles()
         if not files:
             return matchmod.never()
 
         pats = [b'include:%s' % f for f in files]
         return matchmod.match(self._root, b'', [], pats, warn=self._ui.warn)
 
     @propertycache
     def _slash(self):
         return self._ui.configbool(b'ui', b'slash') and pycompat.ossep != b'/'
 
     @propertycache
     def _checklink(self):
         return util.checklink(self._root)
 
     @propertycache
     def _checkexec(self):
         return bool(util.checkexec(self._root))
 
     @propertycache
     def _checkcase(self):
         return not util.fscasesensitive(self._join(b'.hg'))
 
     def _join(self, f):
         # much faster than os.path.join()
         # it's safe because f is always a relative path
         return self._rootdir + f
 
     def flagfunc(self, buildfallback):
         if self._checklink and self._checkexec:
 
             def f(x):
                 try:
                     st = os.lstat(self._join(x))
                     if util.statislink(st):
                         return b'l'
                     if util.statisexec(st):
                         return b'x'
                 except OSError:
                     pass
                 return b''
 
             return f
 
         fallback = buildfallback()
         if self._checklink:
 
             def f(x):
                 if os.path.islink(self._join(x)):
                     return b'l'
                 if b'x' in fallback(x):
                     return b'x'
                 return b''
 
             return f
         if self._checkexec:
 
             def f(x):
                 if b'l' in fallback(x):
                     return b'l'
                 if util.isexec(self._join(x)):
                     return b'x'
                 return b''
 
             return f
         else:
             return fallback
 
     @propertycache
     def _cwd(self):
         # internal config: ui.forcecwd
         forcecwd = self._ui.config(b'ui', b'forcecwd')
         if forcecwd:
             return forcecwd
         return encoding.getcwd()
 
     def getcwd(self):
         """Return the path from which a canonical path is calculated.
 
         This path should be used to resolve file patterns or to convert
         canonical paths back to file paths for display. It shouldn't be
         used to get real file paths. Use vfs functions instead.
         """
         cwd = self._cwd
         if cwd == self._root:
             return b''
         # self._root ends with a path separator if self._root is '/' or 'C:\'
         rootsep = self._root
         if not util.endswithsep(rootsep):
             rootsep += pycompat.ossep
         if cwd.startswith(rootsep):
             return cwd[len(rootsep) :]
         else:
             # we're outside the repo. return an absolute path.
             return cwd
 
     def pathto(self, f, cwd=None):
         if cwd is None:
             cwd = self.getcwd()
         path = util.pathto(self._root, cwd, f)
         if self._slash:
             return util.pconvert(path)
         return path
 
     def __getitem__(self, key):
         """Return the current state of key (a filename) in the dirstate.
 
         States are:
           n  normal
           m  needs merging
           r  marked for removal
           a  marked for addition
           ?  not tracked
         """
         return self._map.get(key, (b"?",))[0]
 
     def __contains__(self, key):
         return key in self._map
 
     def __iter__(self):
         return iter(sorted(self._map))
 
     def items(self):
         return pycompat.iteritems(self._map)
 
     iteritems = items
 
     def parents(self):
         return [self._validate(p) for p in self._pl]
 
     def p1(self):
         return self._validate(self._pl[0])
 
     def p2(self):
         return self._validate(self._pl[1])
 
     def branch(self):
         return encoding.tolocal(self._branch)
 
     def setparents(self, p1, p2=None):
         """Set dirstate parents to p1 and p2.
 
         When moving from two parents to one, 'm' merged entries a
         adjusted to normal and previous copy records discarded and
         returned by the call.
 
         See localrepo.setparents()
         """
         if p2 is None:
             p2 = self._nodeconstants.nullid
         if self._parentwriters == 0:
             raise ValueError(
                 b"cannot set dirstate parent outside of "
                 b"dirstate.parentchange context manager"
             )
 
         self._dirty = True
         oldp2 = self._pl[1]
         if self._origpl is None:
             self._origpl = self._pl
         self._map.setparents(p1, p2)
         copies = {}
         if (
             oldp2 != self._nodeconstants.nullid
             and p2 == self._nodeconstants.nullid
         ):
             candidatefiles = self._map.non_normal_or_other_parent_paths()
 
             for f in candidatefiles:
                 s = self._map.get(f)
                 if s is None:
                     continue
 
                 # Discard 'm' markers when moving away from a merge state
                 if s[0] == b'm':
                     source = self._map.copymap.get(f)
                     if source:
                         copies[f] = source
                     self.normallookup(f)
                 # Also fix up otherparent markers
                 elif s[0] == b'n' and s[2] == -2:
                     source = self._map.copymap.get(f)
                     if source:
                         copies[f] = source
                     self.add(f)
         return copies
 
     def setbranch(self, branch):
         self.__class__._branch.set(self, encoding.fromlocal(branch))
         f = self._opener(b'branch', b'w', atomictemp=True, checkambig=True)
         try:
             f.write(self._branch + b'\n')
             f.close()
 
             # make sure filecache has the correct stat info for _branch after
             # replacing the underlying file
             ce = self._filecache[b'_branch']
             if ce:
                 ce.refresh()
         except:  # re-raises
             f.discard()
             raise
 
     def invalidate(self):
         """Causes the next access to reread the dirstate.
 
         This is different from localrepo.invalidatedirstate() because it always
         rereads the dirstate. Use localrepo.invalidatedirstate() if you want to
         check whether the dirstate has changed before rereading it."""
 
         for a in ("_map", "_branch", "_ignore"):
             if a in self.__dict__:
                 delattr(self, a)
         self._lastnormaltime = 0
         self._dirty = False
         self._updatedfiles.clear()
         self._parentwriters = 0
         self._origpl = None
 
     def copy(self, source, dest):
         """Mark dest as a copy of source. Unmark dest if source is None."""
         if source == dest:
             return
         self._dirty = True
         if source is not None:
             self._map.copymap[dest] = source
             self._updatedfiles.add(source)
             self._updatedfiles.add(dest)
         elif self._map.copymap.pop(dest, None):
             self._updatedfiles.add(dest)
 
     def copied(self, file):
         return self._map.copymap.get(file, None)
 
     def copies(self):
         return self._map.copymap
 
     def _addpath(self, f, state, mode, size, mtime):
         oldstate = self[f]
         if state == b'a' or oldstate == b'r':
             scmutil.checkfilename(f)
             if self._map.hastrackeddir(f):
                 raise error.Abort(
                     _(b'directory %r already in dirstate') % pycompat.bytestr(f)
                 )
             # shadows
             for d in pathutil.finddirs(f):
                 if self._map.hastrackeddir(d):
                     break
                 entry = self._map.get(d)
                 if entry is not None and entry[0] != b'r':
                     raise error.Abort(
                         _(b'file %r in dirstate clashes with %r')
                         % (pycompat.bytestr(d), pycompat.bytestr(f))
                     )
         self._dirty = True
         self._updatedfiles.add(f)
         self._map.addfile(f, oldstate, state, mode, size, mtime)
 
     def normal(self, f, parentfiledata=None):
         """Mark a file normal and clean.
 
         parentfiledata: (mode, size, mtime) of the clean file
 
         parentfiledata should be computed from memory (for mode,
         size), as or close as possible from the point where we
         determined the file was clean, to limit the risk of the
         file having been changed by an external process between the
         moment where the file was determined to be clean and now."""
         if parentfiledata:
             (mode, size, mtime) = parentfiledata
         else:
             s = os.lstat(self._join(f))
             mode = s.st_mode
             size = s.st_size
             mtime = s[stat.ST_MTIME]
         self._addpath(f, b'n', mode, size & _rangemask, mtime & _rangemask)
         self._map.copymap.pop(f, None)
         if f in self._map.nonnormalset:
             self._map.nonnormalset.remove(f)
         if mtime > self._lastnormaltime:
             # Remember the most recent modification timeslot for status(),
             # to make sure we won't miss future size-preserving file content
             # modifications that happen within the same timeslot.
             self._lastnormaltime = mtime
 
     def normallookup(self, f):
         '''Mark a file normal, but possibly dirty.'''
         if self._pl[1] != self._nodeconstants.nullid:
             # if there is a merge going on and the file was either
             # in state 'm' (-1) or coming from other parent (-2) before
             # being removed, restore that state.
             entry = self._map.get(f)
             if entry is not None:
                 if entry[0] == b'r' and entry[2] in (-1, -2):
                     source = self._map.copymap.get(f)
                     if entry[2] == -1:
                         self.merge(f)
                     elif entry[2] == -2:
                         self.otherparent(f)
                     if source:
                         self.copy(source, f)
                     return
                 if entry[0] == b'm' or entry[0] == b'n' and entry[2] == -2:
                     return
         self._addpath(f, b'n', 0, -1, -1)
         self._map.copymap.pop(f, None)
 
     def otherparent(self, f):
         '''Mark as coming from the other parent, always dirty.'''
         if self._pl[1] == self._nodeconstants.nullid:
             raise error.Abort(
                 _(b"setting %r to other parent only allowed in merges") % f
             )
         if f in self and self[f] == b'n':
             # merge-like
             self._addpath(f, b'm', 0, -2, -1)
         else:
             # add-like
             self._addpath(f, b'n', 0, -2, -1)
         self._map.copymap.pop(f, None)
 
     def add(self, f):
         '''Mark a file added.'''
         self._addpath(f, b'a', 0, -1, -1)
         self._map.copymap.pop(f, None)
 
     def remove(self, f):
         '''Mark a file removed.'''
         self._dirty = True
         oldstate = self[f]
         size = 0
         if self._pl[1] != self._nodeconstants.nullid:
             entry = self._map.get(f)
             if entry is not None:
                 # backup the previous state
                 if entry[0] == b'm':  # merge
                     size = -1
                 elif entry[0] == b'n' and entry[2] == -2:  # other parent
                     size = -2
                     self._map.otherparentset.add(f)
         self._updatedfiles.add(f)
         self._map.removefile(f, oldstate, size)
         if size == 0:
             self._map.copymap.pop(f, None)
 
     def merge(self, f):
         '''Mark a file merged.'''
         if self._pl[1] == self._nodeconstants.nullid:
             return self.normallookup(f)
         return self.otherparent(f)
 
     def drop(self, f):
         '''Drop a file from the dirstate'''
         oldstate = self[f]
         if self._map.dropfile(f, oldstate):
             self._dirty = True
             self._updatedfiles.add(f)
             self._map.copymap.pop(f, None)
 
     def _discoverpath(self, path, normed, ignoremissing, exists, storemap):
         if exists is None:
             exists = os.path.lexists(os.path.join(self._root, path))
         if not exists:
             # Maybe a path component exists
             if not ignoremissing and b'/' in path:
                 d, f = path.rsplit(b'/', 1)
                 d = self._normalize(d, False, ignoremissing, None)
                 folded = d + b"/" + f
             else:
                 # No path components, preserve original case
                 folded = path
         else:
             # recursively normalize leading directory components
             # against dirstate
             if b'/' in normed:
                 d, f = normed.rsplit(b'/', 1)
                 d = self._normalize(d, False, ignoremissing, True)
                 r = self._root + b"/" + d
                 folded = d + b"/" + util.fspath(f, r)
             else:
                 folded = util.fspath(normed, self._root)
             storemap[normed] = folded
 
         return folded
 
     def _normalizefile(self, path, isknown, ignoremissing=False, exists=None):
         normed = util.normcase(path)
         folded = self._map.filefoldmap.get(normed, None)
         if folded is None:
             if isknown:
                 folded = path
             else:
                 folded = self._discoverpath(
                     path, normed, ignoremissing, exists, self._map.filefoldmap
                 )
         return folded
 
     def _normalize(self, path, isknown, ignoremissing=False, exists=None):
         normed = util.normcase(path)
         folded = self._map.filefoldmap.get(normed, None)
         if folded is None:
             folded = self._map.dirfoldmap.get(normed, None)
         if folded is None:
             if isknown:
                 folded = path
             else:
                 # store discovered result in dirfoldmap so that future
                 # normalizefile calls don't start matching directories
                 folded = self._discoverpath(
                     path, normed, ignoremissing, exists, self._map.dirfoldmap
                 )
         return folded
 
     def normalize(self, path, isknown=False, ignoremissing=False):
         """
         normalize the case of a pathname when on a casefolding filesystem
 
         isknown specifies whether the filename came from walking the
         disk, to avoid extra filesystem access.
 
         If ignoremissing is True, missing path are returned
         unchanged. Otherwise, we try harder to normalize possibly
         existing path components.
 
         The normalized case is determined based on the following precedence:
 
         - version of name already stored in the dirstate
         - version of name stored on disk
         - version provided via command arguments
         """
 
         if self._checkcase:
             return self._normalize(path, isknown, ignoremissing)
         return path
 
     def clear(self):
         self._map.clear()
         self._lastnormaltime = 0
         self._updatedfiles.clear()
         self._dirty = True
 
     def rebuild(self, parent, allfiles, changedfiles=None):
         if changedfiles is None:
             # Rebuild entire dirstate
             to_lookup = allfiles
             to_drop = []
             lastnormaltime = self._lastnormaltime
             self.clear()
             self._lastnormaltime = lastnormaltime
         elif len(changedfiles) < 10:
             # Avoid turning allfiles into a set, which can be expensive if it's
             # large.
             to_lookup = []
             to_drop = []
             for f in changedfiles:
                 if f in allfiles:
                     to_lookup.append(f)
                 else:
                     to_drop.append(f)
         else:
             changedfilesset = set(changedfiles)
             to_lookup = changedfilesset & set(allfiles)
             to_drop = changedfilesset - to_lookup
 
         if self._origpl is None:
             self._origpl = self._pl
         self._map.setparents(parent, self._nodeconstants.nullid)
 
         for f in to_lookup:
             self.normallookup(f)
         for f in to_drop:
             self.drop(f)
 
         self._dirty = True
 
     def identity(self):
         """Return identity of dirstate itself to detect changing in storage
 
         If identity of previous dirstate is equal to this, writing
         changes based on the former dirstate out can keep consistency.
         """
         return self._map.identity
 
     def write(self, tr):
         if not self._dirty:
             return
 
         filename = self._filename
         if tr:
             # 'dirstate.write()' is not only for writing in-memory
             # changes out, but also for dropping ambiguous timestamp.
             # delayed writing re-raise "ambiguous timestamp issue".
             # See also the wiki page below for detail:
             # https://www.mercurial-scm.org/wiki/DirstateTransactionPlan
 
             # emulate dropping timestamp in 'parsers.pack_dirstate'
             now = _getfsnow(self._opener)
             self._map.clearambiguoustimes(self._updatedfiles, now)
 
             # emulate that all 'dirstate.normal' results are written out
             self._lastnormaltime = 0
             self._updatedfiles.clear()
 
             # delay writing in-memory changes out
             tr.addfilegenerator(
                 b'dirstate',
                 (self._filename,),
                 self._writedirstate,
                 location=b'plain',
             )
             return
 
         st = self._opener(filename, b"w", atomictemp=True, checkambig=True)
         self._writedirstate(st)
 
     def addparentchangecallback(self, category, callback):
         """add a callback to be called when the wd parents are changed
 
         Callback will be called with the following arguments:
             dirstate, (oldp1, oldp2), (newp1, newp2)
 
         Category is a unique identifier to allow overwriting an old callback
         with a newer callback.
         """
         self._plchangecallbacks[category] = callback
 
     def _writedirstate(self, st):
         # notify callbacks about parents change
         if self._origpl is not None and self._origpl != self._pl:
             for c, callback in sorted(
                 pycompat.iteritems(self._plchangecallbacks)
             ):
                 callback(self, self._origpl, self._pl)
             self._origpl = None
         # use the modification time of the newly created temporary file as the
         # filesystem's notion of 'now'
         now = util.fstat(st)[stat.ST_MTIME] & _rangemask
 
         # enough 'delaywrite' prevents 'pack_dirstate' from dropping
         # timestamp of each entries in dirstate, because of 'now > mtime'
         delaywrite = self._ui.configint(b'debug', b'dirstate.delaywrite')
         if delaywrite > 0:
             # do we have any files to delay for?
             for f, e in pycompat.iteritems(self._map):
                 if e[0] == b'n' and e[3] == now:
                     import time  # to avoid useless import
 
                     # rather than sleep n seconds, sleep until the next
                     # multiple of n seconds
                     clock = time.time()
                     start = int(clock) - (int(clock) % delaywrite)
                     end = start + delaywrite
                     time.sleep(end - clock)
                     now = end  # trust our estimate that the end is near now
                     break
 
         self._map.write(st, now)
         self._lastnormaltime = 0
         self._dirty = False
 
     def _dirignore(self, f):
         if self._ignore(f):
             return True
         for p in pathutil.finddirs(f):
             if self._ignore(p):
                 return True
         return False
 
     def _ignorefiles(self):
         files = []
         if os.path.exists(self._join(b'.hgignore')):
             files.append(self._join(b'.hgignore'))
         for name, path in self._ui.configitems(b"ui"):
             if name == b'ignore' or name.startswith(b'ignore.'):
                 # we need to use os.path.join here rather than self._join
                 # because path is arbitrary and user-specified
                 files.append(os.path.join(self._rootdir, util.expandpath(path)))
         return files
 
     def _ignorefileandline(self, f):
         files = collections.deque(self._ignorefiles())
         visited = set()
         while files:
             i = files.popleft()
             patterns = matchmod.readpatternfile(
                 i, self._ui.warn, sourceinfo=True
             )
             for pattern, lineno, line in patterns:
                 kind, p = matchmod._patsplit(pattern, b'glob')
                 if kind == b"subinclude":
                     if p not in visited:
                         files.append(p)
                     continue
                 m = matchmod.match(
                     self._root, b'', [], [pattern], warn=self._ui.warn
                 )
                 if m(f):
                     return (i, lineno, line)
             visited.add(i)
         return (None, -1, b"")
 
     def _walkexplicit(self, match, subrepos):
         """Get stat data about the files explicitly specified by match.
 
         Return a triple (results, dirsfound, dirsnotfound).
         - results is a mapping from filename to stat result. It also contains
           listings mapping subrepos and .hg to None.
         - dirsfound is a list of files found to be directories.
         - dirsnotfound is a list of files that the dirstate thinks are
           directories and that were not found."""
 
         def badtype(mode):
             kind = _(b'unknown')
             if stat.S_ISCHR(mode):
                 kind = _(b'character device')
             elif stat.S_ISBLK(mode):
                 kind = _(b'block device')
             elif stat.S_ISFIFO(mode):
                 kind = _(b'fifo')
             elif stat.S_ISSOCK(mode):
                 kind = _(b'socket')
             elif stat.S_ISDIR(mode):
                 kind = _(b'directory')
             return _(b'unsupported file type (type is %s)') % kind
 
         badfn = match.bad
         dmap = self._map
         lstat = os.lstat
         getkind = stat.S_IFMT
         dirkind = stat.S_IFDIR
         regkind = stat.S_IFREG
         lnkkind = stat.S_IFLNK
         join = self._join
         dirsfound = []
         foundadd = dirsfound.append
         dirsnotfound = []
         notfoundadd = dirsnotfound.append
 
         if not match.isexact() and self._checkcase:
             normalize = self._normalize
         else:
             normalize = None
 
         files = sorted(match.files())
         subrepos.sort()
         i, j = 0, 0
         while i < len(files) and j < len(subrepos):
             subpath = subrepos[j] + b"/"
             if files[i] < subpath:
                 i += 1
                 continue
             while i < len(files) and files[i].startswith(subpath):
                 del files[i]
             j += 1
 
         if not files or b'' in files:
             files = [b'']
             # constructing the foldmap is expensive, so don't do it for the
             # common case where files is ['']
             normalize = None
         results = dict.fromkeys(subrepos)
         results[b'.hg'] = None
 
         for ff in files:
             if normalize:
                 nf = normalize(ff, False, True)
             else:
                 nf = ff
             if nf in results:
                 continue
 
             try:
                 st = lstat(join(nf))
                 kind = getkind(st.st_mode)
                 if kind == dirkind:
                     if nf in dmap:
                         # file replaced by dir on disk but still in dirstate
                         results[nf] = None
                     foundadd((nf, ff))
                 elif kind == regkind or kind == lnkkind:
                     results[nf] = st
                 else:
                     badfn(ff, badtype(kind))
                     if nf in dmap:
                         results[nf] = None
             except OSError as inst:  # nf not found on disk - it is dirstate only
                 if nf in dmap:  # does it exactly match a missing file?
                     results[nf] = None
                 else:  # does it match a missing directory?
                     if self._map.hasdir(nf):
                         notfoundadd(nf)
                     else:
                         badfn(ff, encoding.strtolocal(inst.strerror))
 
         # match.files() may contain explicitly-specified paths that shouldn't
         # be taken; drop them from the list of files found. dirsfound/notfound
         # aren't filtered here because they will be tested later.
         if match.anypats():
             for f in list(results):
                 if f == b'.hg' or f in subrepos:
                     # keep sentinel to disable further out-of-repo walks
                     continue
                 if not match(f):
                     del results[f]
 
         # Case insensitive filesystems cannot rely on lstat() failing to detect
         # a case-only rename.  Prune the stat object for any file that does not
         # match the case in the filesystem, if there are multiple files that
         # normalize to the same path.
         if match.isexact() and self._checkcase:
             normed = {}
 
             for f, st in pycompat.iteritems(results):
                 if st is None:
                     continue
 
                 nc = util.normcase(f)
                 paths = normed.get(nc)
 
                 if paths is None:
                     paths = set()
                     normed[nc] = paths
 
                 paths.add(f)
 
             for norm, paths in pycompat.iteritems(normed):
                 if len(paths) > 1:
                     for path in paths:
                         folded = self._discoverpath(
                             path, norm, True, None, self._map.dirfoldmap
                         )
                         if path != folded:
                             results[path] = None
 
         return results, dirsfound, dirsnotfound
 
     def walk(self, match, subrepos, unknown, ignored, full=True):
         """
         Walk recursively through the directory tree, finding all files
         matched by match.
 
         If full is False, maybe skip some known-clean files.
 
         Return a dict mapping filename to stat-like object (either
         mercurial.osutil.stat instance or return value of os.stat()).
 
         """
         # full is a flag that extensions that hook into walk can use -- this
         # implementation doesn't use it at all. This satisfies the contract
         # because we only guarantee a "maybe".
 
         if ignored:
             ignore = util.never
             dirignore = util.never
         elif unknown:
             ignore = self._ignore
             dirignore = self._dirignore
         else:
             # if not unknown and not ignored, drop dir recursion and step 2
             ignore = util.always
             dirignore = util.always
 
         matchfn = match.matchfn
         matchalways = match.always()
         matchtdir = match.traversedir
         dmap = self._map
         listdir = util.listdir
         lstat = os.lstat
         dirkind = stat.S_IFDIR
         regkind = stat.S_IFREG
         lnkkind = stat.S_IFLNK
         join = self._join
 
         exact = skipstep3 = False
         if match.isexact():  # match.exact
             exact = True
             dirignore = util.always  # skip step 2
         elif match.prefix():  # match.match, no patterns
             skipstep3 = True
 
         if not exact and self._checkcase:
             normalize = self._normalize
             normalizefile = self._normalizefile
             skipstep3 = False
         else:
             normalize = self._normalize
             normalizefile = None
 
         # step 1: find all explicit files
         results, work, dirsnotfound = self._walkexplicit(match, subrepos)
         if matchtdir:
             for d in work:
                 matchtdir(d[0])
             for d in dirsnotfound:
                 matchtdir(d)
 
         skipstep3 = skipstep3 and not (work or dirsnotfound)
         work = [d for d in work if not dirignore(d[0])]
 
         # step 2: visit subdirectories
         def traverse(work, alreadynormed):
             wadd = work.append
             while work:
                 tracing.counter('dirstate.walk work', len(work))
                 nd = work.pop()
                 visitentries = match.visitchildrenset(nd)
                 if not visitentries:
                     continue
                 if visitentries == b'this' or visitentries == b'all':
                     visitentries = None
                 skip = None
                 if nd != b'':
                     skip = b'.hg'
                 try:
                     with tracing.log('dirstate.walk.traverse listdir %s', nd):
                         entries = listdir(join(nd), stat=True, skip=skip)
                 except OSError as inst:
                     if inst.errno in (errno.EACCES, errno.ENOENT):
                         match.bad(
                             self.pathto(nd), encoding.strtolocal(inst.strerror)
                         )
                         continue
                     raise
                 for f, kind, st in entries:
                     # Some matchers may return files in the visitentries set,
                     # instead of 'this', if the matcher explicitly mentions them
                     # and is not an exactmatcher. This is acceptable; we do not
                     # make any hard assumptions about file-or-directory below
                     # based on the presence of `f` in visitentries. If
                     # visitchildrenset returned a set, we can always skip the
                     # entries *not* in the set it provided regardless of whether
                     # they're actually a file or a directory.
                     if visitentries and f not in visitentries:
                         continue
                     if normalizefile:
                         # even though f might be a directory, we're only
                         # interested in comparing it to files currently in the
                         # dmap -- therefore normalizefile is enough
                         nf = normalizefile(
                             nd and (nd + b"/" + f) or f, True, True
                         )
                     else:
                         nf = nd and (nd + b"/" + f) or f
                     if nf not in results:
                         if kind == dirkind:
                             if not ignore(nf):
                                 if matchtdir:
                                     matchtdir(nf)
                                 wadd(nf)
                             if nf in dmap and (matchalways or matchfn(nf)):
                                 results[nf] = None
                         elif kind == regkind or kind == lnkkind:
                             if nf in dmap:
                                 if matchalways or matchfn(nf):
                                     results[nf] = st
                             elif (matchalways or matchfn(nf)) and not ignore(
                                 nf
                             ):
                                 # unknown file -- normalize if necessary
                                 if not alreadynormed:
                                     nf = normalize(nf, False, True)
                                 results[nf] = st
                         elif nf in dmap and (matchalways or matchfn(nf)):
                             results[nf] = None
 
         for nd, d in work:
             # alreadynormed means that processwork doesn't have to do any
             # expensive directory normalization
             alreadynormed = not normalize or nd == d
             traverse([d], alreadynormed)
 
         for s in subrepos:
             del results[s]
         del results[b'.hg']
 
         # step 3: visit remaining files from dmap
         if not skipstep3 and not exact:
             # If a dmap file is not in results yet, it was either
             # a) not matching matchfn b) ignored, c) missing, or d) under a
             # symlink directory.
             if not results and matchalways:
                 visit = [f for f in dmap]
             else:
                 visit = [f for f in dmap if f not in results and matchfn(f)]
             visit.sort()
 
             if unknown:
                 # unknown == True means we walked all dirs under the roots
                 # that wasn't ignored, and everything that matched was stat'ed
                 # and is already in results.
                 # The rest must thus be ignored or under a symlink.
                 audit_path = pathutil.pathauditor(self._root, cached=True)
 
                 for nf in iter(visit):
                     # If a stat for the same file was already added with a
                     # different case, don't add one for this, since that would
                     # make it appear as if the file exists under both names
                     # on disk.
                     if (
                         normalizefile
                         and normalizefile(nf, True, True) in results
                     ):
                         results[nf] = None
                     # Report ignored items in the dmap as long as they are not
                     # under a symlink directory.
                     elif audit_path.check(nf):
                         try:
                             results[nf] = lstat(join(nf))
                             # file was just ignored, no links, and exists
                         except OSError:
                             # file doesn't exist
                             results[nf] = None
                     else:
                         # It's either missing or under a symlink directory
                         # which we in this case report as missing
                         results[nf] = None
             else:
                 # We may not have walked the full directory tree above,
                 # so stat and check everything we missed.
                 iv = iter(visit)
                 for st in util.statfiles([join(i) for i in visit]):
                     results[next(iv)] = st
         return results
 
     def _rust_status(self, matcher, list_clean, list_ignored, list_unknown):
         # Force Rayon (Rust parallelism library) to respect the number of
         # workers. This is a temporary workaround until Rust code knows
         # how to read the config file.
         numcpus = self._ui.configint(b"worker", b"numcpus")
         if numcpus is not None:
             encoding.environ.setdefault(b'RAYON_NUM_THREADS', b'%d' % numcpus)
 
         workers_enabled = self._ui.configbool(b"worker", b"enabled", True)
         if not workers_enabled:
             encoding.environ[b"RAYON_NUM_THREADS"] = b"1"
 
         (
             lookup,
             modified,
             added,
             removed,
             deleted,
             clean,
             ignored,
             unknown,
             warnings,
             bad,
             traversed,
         ) = rustmod.status(
             self._map._rustmap,
             matcher,
             self._rootdir,
             self._ignorefiles(),
             self._checkexec,
             self._lastnormaltime,
             bool(list_clean),
             bool(list_ignored),
             bool(list_unknown),
             bool(matcher.traversedir),
         )
 
         if matcher.traversedir:
             for dir in traversed:
                 matcher.traversedir(dir)
 
         if self._ui.warn:
             for item in warnings:
                 if isinstance(item, tuple):
                     file_path, syntax = item
                     msg = _(b"%s: ignoring invalid syntax '%s'\n") % (
                         file_path,
                         syntax,
                     )
                     self._ui.warn(msg)
                 else:
                     msg = _(b"skipping unreadable pattern file '%s': %s\n")
                     self._ui.warn(
                         msg
                         % (
                             pathutil.canonpath(
                                 self._rootdir, self._rootdir, item
                             ),
                             b"No such file or directory",
                         )
                     )
 
         for (fn, message) in bad:
             matcher.bad(fn, encoding.strtolocal(message))
 
         status = scmutil.status(
             modified=modified,
             added=added,
             removed=removed,
             deleted=deleted,
             unknown=unknown,
             ignored=ignored,
             clean=clean,
         )
         return (lookup, status)
 
     def status(self, match, subrepos, ignored, clean, unknown):
         """Determine the status of the working copy relative to the
         dirstate and return a pair of (unsure, status), where status is of type
         scmutil.status and:
 
           unsure:
             files that might have been modified since the dirstate was
             written, but need to be read to be sure (size is the same
             but mtime differs)
           status.modified:
             files that have definitely been modified since the dirstate
             was written (different size or mode)
           status.clean:
             files that have definitely not been modified since the
             dirstate was written
         """
         listignored, listclean, listunknown = ignored, clean, unknown
         lookup, modified, added, unknown, ignored = [], [], [], [], []
         removed, deleted, clean = [], [], []
 
         dmap = self._map
         dmap.preload()
 
         use_rust = True
 
         allowed_matchers = (
             matchmod.alwaysmatcher,
             matchmod.exactmatcher,
             matchmod.includematcher,
         )
 
         if rustmod is None:
             use_rust = False
         elif self._checkcase:
             # Case-insensitive filesystems are not handled yet
             use_rust = False
         elif subrepos:
             use_rust = False
         elif sparse.enabled:
             use_rust = False
         elif not isinstance(match, allowed_matchers):
             # Some matchers have yet to be implemented
             use_rust = False
 
         if use_rust:
             try:
                 return self._rust_status(
                     match, listclean, listignored, listunknown
                 )
             except rustmod.FallbackError:
                 pass
 
         def noop(f):
             pass
 
         dcontains = dmap.__contains__
         dget = dmap.__getitem__
         ladd = lookup.append  # aka "unsure"
         madd = modified.append
         aadd = added.append
         uadd = unknown.append if listunknown else noop
         iadd = ignored.append if listignored else noop
         radd = removed.append
         dadd = deleted.append
         cadd = clean.append if listclean else noop
         mexact = match.exact
         dirignore = self._dirignore
         checkexec = self._checkexec
         copymap = self._map.copymap
         lastnormaltime = self._lastnormaltime
 
         # We need to do full walks when either
         # - we're listing all clean files, or
         # - match.traversedir does something, because match.traversedir should
         #   be called for every dir in the working dir
         full = listclean or match.traversedir is not None
         for fn, st in pycompat.iteritems(
             self.walk(match, subrepos, listunknown, listignored, full=full)
         ):
             if not dcontains(fn):
                 if (listignored or mexact(fn)) and dirignore(fn):
                     if listignored:
                         iadd(fn)
                 else:
                     uadd(fn)
                 continue
 
             # This is equivalent to 'state, mode, size, time = dmap[fn]' but not
             # written like that for performance reasons. dmap[fn] is not a
             # Python tuple in compiled builds. The CPython UNPACK_SEQUENCE
             # opcode has fast paths when the value to be unpacked is a tuple or
             # a list, but falls back to creating a full-fledged iterator in
             # general. That is much slower than simply accessing and storing the
             # tuple members one by one.
             t = dget(fn)
             state = t[0]
             mode = t[1]
             size = t[2]
             time = t[3]
 
             if not st and state in b"nma":
                 dadd(fn)
             elif state == b'n':
                 if (
                     size >= 0
                     and (
                         (size != st.st_size and size != st.st_size & _rangemask)
                         or ((mode ^ st.st_mode) & 0o100 and checkexec)
                     )
                     or size == -2  # other parent
                     or fn in copymap
                 ):
                     if stat.S_ISLNK(st.st_mode) and size != st.st_size:
                         # issue6456: Size returned may be longer due to
                         # encryption on EXT-4 fscrypt, undecided.
                         ladd(fn)
                     else:
                         madd(fn)
                 elif (
                     time != st[stat.ST_MTIME]
                     and time != st[stat.ST_MTIME] & _rangemask
                 ):
                     ladd(fn)
                 elif st[stat.ST_MTIME] == lastnormaltime:
                     # fn may have just been marked as normal and it may have
                     # changed in the same second without changing its size.
                     # This can happen if we quickly do multiple commits.
                     # Force lookup, so we don't miss such a racy file change.
                     ladd(fn)
                 elif listclean:
                     cadd(fn)
             elif state == b'm':
                 madd(fn)
             elif state == b'a':
                 aadd(fn)
             elif state == b'r':
                 radd(fn)
         status = scmutil.status(
             modified, added, removed, deleted, unknown, ignored, clean
         )
         return (lookup, status)
 
     def matches(self, match):
         """
         return files in the dirstate (in whatever state) filtered by match
         """
         dmap = self._map
         if rustmod is not None:
             dmap = self._map._rustmap
 
         if match.always():
             return dmap.keys()
         files = match.files()
         if match.isexact():
             # fast path -- filter the other way around, since typically files is
             # much smaller than dmap
             return [f for f in files if f in dmap]
         if match.prefix() and all(fn in dmap for fn in files):
             # fast path -- all the values are known to be files, so just return
             # that
             return list(files)
         return [f for f in dmap if match(f)]
 
     def _actualfilename(self, tr):
         if tr:
             return self._pendingfilename
         else:
             return self._filename
 
     def savebackup(self, tr, backupname):
         '''Save current dirstate into backup file'''
         filename = self._actualfilename(tr)
         assert backupname != filename
 
         # use '_writedirstate' instead of 'write' to write changes certainly,
         # because the latter omits writing out if transaction is running.
         # output file will be used to create backup of dirstate at this point.
         if self._dirty or not self._opener.exists(filename):
             self._writedirstate(
                 self._opener(filename, b"w", atomictemp=True, checkambig=True)
             )
 
         if tr:
             # ensure that subsequent tr.writepending returns True for
             # changes written out above, even if dirstate is never
             # changed after this
             tr.addfilegenerator(
                 b'dirstate',
                 (self._filename,),
                 self._writedirstate,
                 location=b'plain',
             )
 
             # ensure that pending file written above is unlinked at
             # failure, even if tr.writepending isn't invoked until the
             # end of this transaction
             tr.registertmp(filename, location=b'plain')
 
         self._opener.tryunlink(backupname)
         # hardlink backup is okay because _writedirstate is always called
         # with an "atomictemp=True" file.
         util.copyfile(
             self._opener.join(filename),
             self._opener.join(backupname),
             hardlink=True,
         )
 
     def restorebackup(self, tr, backupname):
         '''Restore dirstate by backup file'''
         # this "invalidate()" prevents "wlock.release()" from writing
         # changes of dirstate out after restoring from backup file
         self.invalidate()
         filename = self._actualfilename(tr)
         o = self._opener
         if util.samefile(o.join(backupname), o.join(filename)):
             o.unlink(backupname)
         else:
             o.rename(backupname, filename, checkambig=True)
 
     def clearbackup(self, tr, backupname):
         '''Clear backup file'''
         self._opener.unlink(backupname)
 
 
 class dirstatemap(object):
     """Map encapsulating the dirstate's contents.
 
     The dirstate contains the following state:
 
     - `identity` is the identity of the dirstate file, which can be used to
       detect when changes have occurred to the dirstate file.
 
     - `parents` is a pair containing the parents of the working copy. The
       parents are updated by calling `setparents`.
 
     - the state map maps filenames to tuples of (state, mode, size, mtime),
       where state is a single character representing 'normal', 'added',
       'removed', or 'merged'. It is read by treating the dirstate as a
       dict.  File state is updated by calling the `addfile`, `removefile` and
       `dropfile` methods.
 
     - `copymap` maps destination filenames to their source filename.
 
     The dirstate also provides the following views onto the state:
 
     - `nonnormalset` is a set of the filenames that have state other
       than 'normal', or are normal but have an mtime of -1 ('normallookup').
 
     - `otherparentset` is a set of the filenames that are marked as coming
       from the second parent when the dirstate is currently being merged.
 
     - `filefoldmap` is a dict mapping normalized filenames to the denormalized
       form that they appear as in the dirstate.
 
     - `dirfoldmap` is a dict mapping normalized directory names to the
       denormalized form that they appear as in the dirstate.
     """
 
     def __init__(self, ui, opener, root, nodeconstants, use_dirstate_v2):
         self._ui = ui
         self._opener = opener
         self._root = root
         self._filename = b'dirstate'
         self._nodelen = 20
         self._nodeconstants = nodeconstants
         assert (
             not use_dirstate_v2
         ), "should have detected unsupported requirement"
 
         self._parents = None
         self._dirtyparents = False
 
         # for consistent view between _pl() and _read() invocations
         self._pendingmode = None
 
     @propertycache
     def _map(self):
         self._map = {}
         self.read()
         return self._map
 
     @propertycache
     def copymap(self):
         self.copymap = {}
         self._map
         return self.copymap
 
     def clear(self):
         self._map.clear()
         self.copymap.clear()
         self.setparents(self._nodeconstants.nullid, self._nodeconstants.nullid)
         util.clearcachedproperty(self, b"_dirs")
         util.clearcachedproperty(self, b"_alldirs")
         util.clearcachedproperty(self, b"filefoldmap")
         util.clearcachedproperty(self, b"dirfoldmap")
         util.clearcachedproperty(self, b"nonnormalset")
         util.clearcachedproperty(self, b"otherparentset")
 
     def items(self):
         return pycompat.iteritems(self._map)
 
     # forward for python2,3 compat
     iteritems = items
 
     def __len__(self):
         return len(self._map)
 
     def __iter__(self):
         return iter(self._map)
 
     def get(self, key, default=None):
         return self._map.get(key, default)
 
     def __contains__(self, key):
         return key in self._map
 
     def __getitem__(self, key):
         return self._map[key]
 
     def keys(self):
         return self._map.keys()
 
     def preload(self):
         """Loads the underlying data, if it's not already loaded"""
         self._map
 
     def addfile(self, f, oldstate, state, mode, size, mtime):
         """Add a tracked file to the dirstate."""
         if oldstate in b"?r" and "_dirs" in self.__dict__:
             self._dirs.addpath(f)
         if oldstate == b"?" and "_alldirs" in self.__dict__:
             self._alldirs.addpath(f)
         self._map[f] = dirstatetuple(state, mode, size, mtime)
         if state != b'n' or mtime == -1:
             self.nonnormalset.add(f)
         if size == -2:
             self.otherparentset.add(f)
 
     def removefile(self, f, oldstate, size):
         """
         Mark a file as removed in the dirstate.
 
         The `size` parameter is used to store sentinel values that indicate
         the file's previous state.  In the future, we should refactor this
         to be more explicit about what that state is.
         """
         if oldstate not in b"?r" and "_dirs" in self.__dict__:
             self._dirs.delpath(f)
         if oldstate == b"?" and "_alldirs" in self.__dict__:
             self._alldirs.addpath(f)
         if "filefoldmap" in self.__dict__:
             normed = util.normcase(f)
             self.filefoldmap.pop(normed, None)
         self._map[f] = dirstatetuple(b'r', 0, size, 0)
         self.nonnormalset.add(f)
 
     def dropfile(self, f, oldstate):
         """
         Remove a file from the dirstate.  Returns True if the file was
         previously recorded.
         """
         exists = self._map.pop(f, None) is not None
         if exists:
             if oldstate != b"r" and "_dirs" in self.__dict__:
                 self._dirs.delpath(f)
             if "_alldirs" in self.__dict__:
                 self._alldirs.delpath(f)
         if "filefoldmap" in self.__dict__:
             normed = util.normcase(f)
             self.filefoldmap.pop(normed, None)
         self.nonnormalset.discard(f)
         return exists
 
     def clearambiguoustimes(self, files, now):
         for f in files:
             e = self.get(f)
             if e is not None and e[0] == b'n' and e[3] == now:
                 self._map[f] = dirstatetuple(e[0], e[1], e[2], -1)
                 self.nonnormalset.add(f)
 
     def nonnormalentries(self):
         '''Compute the nonnormal dirstate entries from the dmap'''
         try:
             return parsers.nonnormalotherparententries(self._map)
         except AttributeError:
             nonnorm = set()
             otherparent = set()
             for fname, e in pycompat.iteritems(self._map):
                 if e[0] != b'n' or e[3] == -1:
                     nonnorm.add(fname)
                 if e[0] == b'n' and e[2] == -2:
                     otherparent.add(fname)
             return nonnorm, otherparent
 
     @propertycache
     def filefoldmap(self):
         """Returns a dictionary mapping normalized case paths to their
         non-normalized versions.
         """
         try:
             makefilefoldmap = parsers.make_file_foldmap
         except AttributeError:
             pass
         else:
             return makefilefoldmap(
                 self._map, util.normcasespec, util.normcasefallback
             )
 
         f = {}
         normcase = util.normcase
         for name, s in pycompat.iteritems(self._map):
             if s[0] != b'r':
                 f[normcase(name)] = name
         f[b'.'] = b'.'  # prevents useless util.fspath() invocation
         return f
 
     def hastrackeddir(self, d):
         """
         Returns True if the dirstate contains a tracked (not removed) file
         in this directory.
         """
         return d in self._dirs
 
     def hasdir(self, d):
         """
         Returns True if the dirstate contains a file (tracked or removed)
         in this directory.
         """
         return d in self._alldirs
 
     @propertycache
     def _dirs(self):
         return pathutil.dirs(self._map, b'r')
 
     @propertycache
     def _alldirs(self):
         return pathutil.dirs(self._map)
 
     def _opendirstatefile(self):
         fp, mode = txnutil.trypending(self._root, self._opener, self._filename)
         if self._pendingmode is not None and self._pendingmode != mode:
             fp.close()
             raise error.Abort(
                 _(b'working directory state may be changed parallelly')
             )
         self._pendingmode = mode
         return fp
 
     def parents(self):
         if not self._parents:
             try:
                 fp = self._opendirstatefile()
                 st = fp.read(2 * self._nodelen)
                 fp.close()
             except IOError as err:
                 if err.errno != errno.ENOENT:
                     raise
                 # File doesn't exist, so the current state is empty
                 st = b''
 
             l = len(st)
             if l == self._nodelen * 2:
                 self._parents = (
                     st[: self._nodelen],
                     st[self._nodelen : 2 * self._nodelen],
                 )
             elif l == 0:
                 self._parents = (
                     self._nodeconstants.nullid,
                     self._nodeconstants.nullid,
                 )
             else:
                 raise error.Abort(
                     _(b'working directory state appears damaged!')
                 )
 
         return self._parents
 
     def setparents(self, p1, p2):
         self._parents = (p1, p2)
         self._dirtyparents = True
 
     def read(self):
         # ignore HG_PENDING because identity is used only for writing
         self.identity = util.filestat.frompath(
             self._opener.join(self._filename)
         )
 
         try:
             fp = self._opendirstatefile()
             try:
                 st = fp.read()
             finally:
                 fp.close()
         except IOError as err:
             if err.errno != errno.ENOENT:
                 raise
             return
         if not st:
             return
 
         if util.safehasattr(parsers, b'dict_new_presized'):
             # Make an estimate of the number of files in the dirstate based on
             # its size. This trades wasting some memory for avoiding costly
             # resizes. Each entry have a prefix of 17 bytes followed by one or
             # two path names. Studies on various large-scale real-world repositories
             # found 54 bytes a reasonable upper limit for the average path names.
             # Copy entries are ignored for the sake of this estimate.
             self._map = parsers.dict_new_presized(len(st) // 71)
 
         # Python's garbage collector triggers a GC each time a certain number
         # of container objects (the number being defined by
         # gc.get_threshold()) are allocated. parse_dirstate creates a tuple
         # for each file in the dirstate. The C version then immediately marks
         # them as not to be tracked by the collector. However, this has no
         # effect on when GCs are triggered, only on what objects the GC looks
         # into. This means that O(number of files) GCs are unavoidable.
         # Depending on when in the process's lifetime the dirstate is parsed,
         # this can get very expensive. As a workaround, disable GC while
         # parsing the dirstate.
         #
         # (we cannot decorate the function directly since it is in a C module)
         parse_dirstate = util.nogc(parsers.parse_dirstate)
         p = parse_dirstate(self._map, self.copymap, st)
         if not self._dirtyparents:
             self.setparents(*p)
 
         # Avoid excess attribute lookups by fast pathing certain checks
         self.__contains__ = self._map.__contains__
         self.__getitem__ = self._map.__getitem__
         self.get = self._map.get
 
     def write(self, st, now):
         st.write(
             parsers.pack_dirstate(self._map, self.copymap, self.parents(), now)
         )
         st.close()
         self._dirtyparents = False
         self.nonnormalset, self.otherparentset = self.nonnormalentries()
 
     @propertycache
     def nonnormalset(self):
         nonnorm, otherparents = self.nonnormalentries()
         self.otherparentset = otherparents
         return nonnorm
 
     @propertycache
     def otherparentset(self):
         nonnorm, otherparents = self.nonnormalentries()
         self.nonnormalset = nonnorm
         return otherparents
 
     def non_normal_or_other_parent_paths(self):
         return self.nonnormalset.union(self.otherparentset)
 
     @propertycache
     def identity(self):
         self._map
         return self.identity
 
     @propertycache
     def dirfoldmap(self):
         f = {}
         normcase = util.normcase
         for name in self._dirs:
             f[normcase(name)] = name
         return f
 
 
 if rustmod is not None:
 
     class dirstatemap(object):
         def __init__(self, ui, opener, root, nodeconstants, use_dirstate_v2):
             self._use_dirstate_v2 = use_dirstate_v2
             self._nodeconstants = nodeconstants
             self._ui = ui
             self._opener = opener
             self._root = root
             self._filename = b'dirstate'
             self._nodelen = 20  # Also update Rust code when changing this!
             self._parents = None
             self._dirtyparents = False
 
             # for consistent view between _pl() and _read() invocations
             self._pendingmode = None
 
+            self._use_dirstate_tree = self._ui.configbool(
+                b"experimental",
+                b"dirstate-tree.in-memory",
+                False,
+            )
+
         def addfile(self, *args, **kwargs):
             return self._rustmap.addfile(*args, **kwargs)
 
         def removefile(self, *args, **kwargs):
             return self._rustmap.removefile(*args, **kwargs)
 
         def dropfile(self, *args, **kwargs):
             return self._rustmap.dropfile(*args, **kwargs)
 
         def clearambiguoustimes(self, *args, **kwargs):
             return self._rustmap.clearambiguoustimes(*args, **kwargs)
 
         def nonnormalentries(self):
             return self._rustmap.nonnormalentries()
 
         def get(self, *args, **kwargs):
             return self._rustmap.get(*args, **kwargs)
 
         @property
         def copymap(self):
             return self._rustmap.copymap()
 
         def preload(self):
             self._rustmap
 
         def clear(self):
             self._rustmap.clear()
             self.setparents(
                 self._nodeconstants.nullid, self._nodeconstants.nullid
             )
             util.clearcachedproperty(self, b"_dirs")
             util.clearcachedproperty(self, b"_alldirs")
             util.clearcachedproperty(self, b"dirfoldmap")
 
         def items(self):
             return self._rustmap.items()
 
         def keys(self):
             return iter(self._rustmap)
 
         def __contains__(self, key):
             return key in self._rustmap
 
         def __getitem__(self, item):
             return self._rustmap[item]
 
         def __len__(self):
             return len(self._rustmap)
 
         def __iter__(self):
             return iter(self._rustmap)
 
         # forward for python2,3 compat
         iteritems = items
 
         def _opendirstatefile(self):
             fp, mode = txnutil.trypending(
                 self._root, self._opener, self._filename
             )
             if self._pendingmode is not None and self._pendingmode != mode:
                 fp.close()
                 raise error.Abort(
                     _(b'working directory state may be changed parallelly')
                 )
             self._pendingmode = mode
             return fp
 
         def setparents(self, p1, p2):
             self._parents = (p1, p2)
             self._dirtyparents = True
 
         def parents(self):
             if not self._parents:
                 if self._use_dirstate_v2:
                     offset = len(rustmod.V2_FORMAT_MARKER)
                 else:
                     offset = 0
                 read_len = offset + self._nodelen * 2
                 try:
                     fp = self._opendirstatefile()
                     st = fp.read(read_len)
                     fp.close()
                 except IOError as err:
                     if err.errno != errno.ENOENT:
                         raise
                     # File doesn't exist, so the current state is empty
                     st = b''
 
                 l = len(st)
                 if l == read_len:
                     st = st[offset:]
                     self._parents = (
                         st[: self._nodelen],
                         st[self._nodelen : 2 * self._nodelen],
                     )
                 elif l == 0:
                     self._parents = (
                         self._nodeconstants.nullid,
                         self._nodeconstants.nullid,
                     )
                 else:
                     raise error.Abort(
                         _(b'working directory state appears damaged!')
                     )
 
             return self._parents
 
         @propertycache
         def _rustmap(self):
             """
             Fills the Dirstatemap when called.
             """
             # ignore HG_PENDING because identity is used only for writing
             self.identity = util.filestat.frompath(
                 self._opener.join(self._filename)
             )
 
             try:
                 fp = self._opendirstatefile()
                 try:
                     st = fp.read()
                 finally:
                     fp.close()
             except IOError as err:
                 if err.errno != errno.ENOENT:
                     raise
                 st = b''
 
-            use_dirstate_tree = self._ui.configbool(
-                b"experimental",
-                b"dirstate-tree.in-memory",
-                False,
-            )
             self._rustmap, parents = rustmod.DirstateMap.new(
-                use_dirstate_tree, self._use_dirstate_v2, st
+                self._use_dirstate_tree, self._use_dirstate_v2, st
             )
 
             if parents and not self._dirtyparents:
                 self.setparents(*parents)
 
             self.__contains__ = self._rustmap.__contains__
             self.__getitem__ = self._rustmap.__getitem__
             self.get = self._rustmap.get
             return self._rustmap
 
         def write(self, st, now):
             parents = self.parents()
             packed = self._rustmap.write(
                 self._use_dirstate_v2, parents[0], parents[1], now
             )
             st.write(packed)
             st.close()
             self._dirtyparents = False
 
         @propertycache
         def filefoldmap(self):
             """Returns a dictionary mapping normalized case paths to their
             non-normalized versions.
             """
             return self._rustmap.filefoldmapasdict()
 
         def hastrackeddir(self, d):
             self._dirs  # Trigger Python's propertycache
             return self._rustmap.hastrackeddir(d)
 
         def hasdir(self, d):
             self._dirs  # Trigger Python's propertycache
             return self._rustmap.hasdir(d)
 
         @propertycache
         def _dirs(self):
             return self._rustmap.getdirs()
 
         @propertycache
         def _alldirs(self):
             return self._rustmap.getalldirs()
 
         @propertycache
         def identity(self):
             self._rustmap
             return self.identity
 
         @property
         def nonnormalset(self):
             nonnorm = self._rustmap.non_normal_entries()
             return nonnorm
 
         @propertycache
         def otherparentset(self):
             otherparents = self._rustmap.other_parent_entries()
             return otherparents
 
         def non_normal_or_other_parent_paths(self):
             return self._rustmap.non_normal_or_other_parent_paths()
 
         @propertycache
         def dirfoldmap(self):
             f = {}
             normcase = util.normcase
             for name in self._dirs:
                 f[normcase(name)] = name
             return f

mercurial/upgrade_utils/actions.py

@@ -1,1008 +1,1049
 # upgrade.py - functions for in place upgrade of Mercurial repository
 #
 # Copyright (c) 2016-present, Gregory Szorc
 #
 # 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
 
 from ..i18n import _
 from .. import (
     error,
     localrepo,
     pycompat,
     requirements,
     revlog,
     util,
 )
 
 from ..utils import compression
 
 if pycompat.TYPE_CHECKING:
     from typing import (
         List,
         Type,
     )
 
 
 # list of requirements that request a clone of all revlog if added/removed
 RECLONES_REQUIREMENTS = {
     requirements.GENERALDELTA_REQUIREMENT,
     requirements.SPARSEREVLOG_REQUIREMENT,
     requirements.REVLOGV2_REQUIREMENT,
     requirements.CHANGELOGV2_REQUIREMENT,
 }
 
 
 def preservedrequirements(repo):
     return set()
 
 
 FORMAT_VARIANT = b'deficiency'
 OPTIMISATION = b'optimization'
 
 
 class improvement(object):
     """Represents an improvement that can be made as part of an upgrade."""
 
     ### The following attributes should be defined for each subclass:
 
     # Either ``FORMAT_VARIANT`` or ``OPTIMISATION``.
     # A format variant is where we change the storage format. Not all format
     # variant changes are an obvious problem.
     # An optimization is an action (sometimes optional) that
     # can be taken to further improve the state of the repository.
     type = None
 
     # machine-readable string uniquely identifying this improvement. it will be
     # mapped to an action later in the upgrade process.
     name = None
 
     # message intended for humans explaining the improvement in more detail,
     # including the implications of it ``FORMAT_VARIANT`` types, should be
     # worded
     # in the present tense.
     description = None
 
     # message intended for humans explaining what an upgrade addressing this
     # issue will do. should be worded in the future tense.
     upgrademessage = None
 
     # value of current Mercurial default for new repository
     default = None
 
     # Message intended for humans which will be shown post an upgrade
     # operation when the improvement will be added
     postupgrademessage = None
 
     # Message intended for humans which will be shown post an upgrade
     # operation in which this improvement was removed
     postdowngrademessage = None
 
-    # By default for now, we assume every improvement touches all the things
+    # By default we assume that every improvement touches requirements and all revlogs
 
     # Whether this improvement touches filelogs
     touches_filelogs = True
 
     # Whether this improvement touches manifests
     touches_manifests = True
 
     # Whether this improvement touches changelog
     touches_changelog = True
 
     # Whether this improvement changes repository requirements
     touches_requirements = True
 
+    # Whether this improvement touches the dirstate
+    touches_dirstate = False
+
 
 allformatvariant = []  # type: List[Type['formatvariant']]
 
 
 def registerformatvariant(cls):
     allformatvariant.append(cls)
     return cls
 
 
 class formatvariant(improvement):
     """an improvement subclass dedicated to repository format"""
 
     type = FORMAT_VARIANT
 
     @staticmethod
     def fromrepo(repo):
         """current value of the variant in the repository"""
         raise NotImplementedError()
 
     @staticmethod
     def fromconfig(repo):
         """current value of the variant in the configuration"""
         raise NotImplementedError()
 
 
 class requirementformatvariant(formatvariant):
     """formatvariant based on a 'requirement' name.
 
     Many format variant are controlled by a 'requirement'. We define a small
     subclass to factor the code.
     """
 
     # the requirement that control this format variant
     _requirement = None
 
     @staticmethod
     def _newreporequirements(ui):
         return localrepo.newreporequirements(
             ui, localrepo.defaultcreateopts(ui)
         )
 
     @classmethod
     def fromrepo(cls, repo):
         assert cls._requirement is not None
         return cls._requirement in repo.requirements
 
     @classmethod
     def fromconfig(cls, repo):
         assert cls._requirement is not None
         return cls._requirement in cls._newreporequirements(repo.ui)
 
 
 @registerformatvariant
 class fncache(requirementformatvariant):
     name = b'fncache'
 
     _requirement = requirements.FNCACHE_REQUIREMENT
 
     default = True
 
     description = _(
         b'long and reserved filenames may not work correctly; '
         b'repository performance is sub-optimal'
     )
 
     upgrademessage = _(
         b'repository will be more resilient to storing '
         b'certain paths and performance of certain '
         b'operations should be improved'
     )
 
 
 @registerformatvariant
+class dirstatev2(requirementformatvariant):
+    name = b'dirstate-v2'
+    _requirement = requirements.DIRSTATE_V2_REQUIREMENT
+
+    default = False
+
+    description = _(
+        b'version 1 of the dirstate file format requires '
+        b'reading and parsing it all at once.'
+    )
+
+    upgrademessage = _(b'"hg status" will be faster')
+
+    touches_filelogs = False
+    touches_manifests = False
+    touches_changelog = False
+    touches_requirements = True
+    touches_dirstate = True
+
+
+@registerformatvariant
 class dotencode(requirementformatvariant):
     name = b'dotencode'
 
     _requirement = requirements.DOTENCODE_REQUIREMENT
 
     default = True
 
     description = _(
         b'storage of filenames beginning with a period or '
         b'space may not work correctly'
     )
 
     upgrademessage = _(
         b'repository will be better able to store files '
         b'beginning with a space or period'
     )
 
 
 @registerformatvariant
 class generaldelta(requirementformatvariant):
     name = b'generaldelta'
 
     _requirement = requirements.GENERALDELTA_REQUIREMENT
 
     default = True
 
     description = _(
         b'deltas within internal storage are unable to '
         b'choose optimal revisions; repository is larger and '
         b'slower than it could be; interaction with other '
         b'repositories may require extra network and CPU '
         b'resources, making "hg push" and "hg pull" slower'
     )
 
     upgrademessage = _(
         b'repository storage will be able to create '
         b'optimal deltas; new repository data will be '
         b'smaller and read times should decrease; '
         b'interacting with other repositories using this '
         b'storage model should require less network and '
         b'CPU resources, making "hg push" and "hg pull" '
         b'faster'
     )
 
 
 @registerformatvariant
 class sharesafe(requirementformatvariant):
     name = b'share-safe'
     _requirement = requirements.SHARESAFE_REQUIREMENT
 
     default = False
 
     description = _(
         b'old shared repositories do not share source repository '
         b'requirements and config. This leads to various problems '
         b'when the source repository format is upgraded or some new '
         b'extensions are enabled.'
     )
 
     upgrademessage = _(
         b'Upgrades a repository to share-safe format so that future '
         b'shares of this repository share its requirements and configs.'
     )
 
     postdowngrademessage = _(
         b'repository downgraded to not use share safe mode, '
         b'existing shares will not work and needs to'
         b' be reshared.'
     )
 
     postupgrademessage = _(
         b'repository upgraded to share safe mode, existing'
         b' shares will still work in old non-safe mode. '
         b'Re-share existing shares to use them in safe mode'
         b' New shares will be created in safe mode.'
     )
 
     # upgrade only needs to change the requirements
     touches_filelogs = False
     touches_manifests = False
     touches_changelog = False
     touches_requirements = True
 
 
 @registerformatvariant
 class sparserevlog(requirementformatvariant):
     name = b'sparserevlog'
 
     _requirement = requirements.SPARSEREVLOG_REQUIREMENT
 
     default = True
 
     description = _(
         b'in order to limit disk reading and memory usage on older '
         b'version, the span of a delta chain from its root to its '
         b'end is limited, whatever the relevant data in this span. '
         b'This can severly limit Mercurial ability to build good '
         b'chain of delta resulting is much more storage space being '
         b'taken and limit reusability of on disk delta during '
         b'exchange.'
     )
 
     upgrademessage = _(
         b'Revlog supports delta chain with more unused data '
         b'between payload. These gaps will be skipped at read '
         b'time. This allows for better delta chains, making a '
         b'better compression and faster exchange with server.'
     )
 
 
 @registerformatvariant
 class persistentnodemap(requirementformatvariant):
     name = b'persistent-nodemap'
 
     _requirement = requirements.NODEMAP_REQUIREMENT
 
     default = False
 
     description = _(
         b'persist the node -> rev mapping on disk to speedup lookup'
     )
 
     upgrademessage = _(b'Speedup revision lookup by node id.')
 
 
 @registerformatvariant
 class copiessdc(requirementformatvariant):
     name = b'copies-sdc'
 
     _requirement = requirements.COPIESSDC_REQUIREMENT
 
     default = False
 
     description = _(b'Stores copies information alongside changesets.')
 
     upgrademessage = _(
         b'Allows to use more efficient algorithm to deal with ' b'copy tracing.'
     )
 
 
 @registerformatvariant
 class revlogv2(requirementformatvariant):
     name = b'revlog-v2'
     _requirement = requirements.REVLOGV2_REQUIREMENT
     default = False
     description = _(b'Version 2 of the revlog.')
     upgrademessage = _(b'very experimental')
 
 
 @registerformatvariant
 class changelogv2(requirementformatvariant):
     name = b'changelog-v2'
     _requirement = requirements.CHANGELOGV2_REQUIREMENT
     default = False
     description = _(b'An iteration of the revlog focussed on changelog needs.')
     upgrademessage = _(b'quite experimental')
 
 
 @registerformatvariant
 class removecldeltachain(formatvariant):
     name = b'plain-cl-delta'
 
     default = True
 
     description = _(
         b'changelog storage is using deltas instead of '
         b'raw entries; changelog reading and any '
         b'operation relying on changelog data are slower '
         b'than they could be'
     )
 
     upgrademessage = _(
         b'changelog storage will be reformated to '
         b'store raw entries; changelog reading will be '
         b'faster; changelog size may be reduced'
     )
 
     @staticmethod
     def fromrepo(repo):
         # Mercurial 4.0 changed changelogs to not use delta chains. Search for
         # changelogs with deltas.
         cl = repo.changelog
         chainbase = cl.chainbase
         return all(rev == chainbase(rev) for rev in cl)
 
     @staticmethod
     def fromconfig(repo):
         return True
 
 
 _has_zstd = (
     b'zstd' in util.compengines
     and util.compengines[b'zstd'].available()
     and util.compengines[b'zstd'].revlogheader()
 )
 
 
 @registerformatvariant
 class compressionengine(formatvariant):
     name = b'compression'
 
     if _has_zstd:
         default = b'zstd'
     else:
         default = b'zlib'
 
     description = _(
         b'Compresion algorithm used to compress data. '
         b'Some engine are faster than other'
     )
 
     upgrademessage = _(
         b'revlog content will be recompressed with the new algorithm.'
     )
 
     @classmethod
     def fromrepo(cls, repo):
         # we allow multiple compression engine requirement to co-exist because
         # strickly speaking, revlog seems to support mixed compression style.
         #
         # The compression used for new entries will be "the last one"
         compression = b'zlib'
         for req in repo.requirements:
             prefix = req.startswith
             if prefix(b'revlog-compression-') or prefix(b'exp-compression-'):
                 compression = req.split(b'-', 2)[2]
         return compression
 
     @classmethod
     def fromconfig(cls, repo):
         compengines = repo.ui.configlist(b'format', b'revlog-compression')
         # return the first valid value as the selection code would do
         for comp in compengines:
             if comp in util.compengines:
                 e = util.compengines[comp]
                 if e.available() and e.revlogheader():
                     return comp
 
         # no valide compression found lets display it all for clarity
         return b','.join(compengines)
 
 
 @registerformatvariant
 class compressionlevel(formatvariant):
     name = b'compression-level'
     default = b'default'
 
     description = _(b'compression level')
 
     upgrademessage = _(b'revlog content will be recompressed')
 
     @classmethod
     def fromrepo(cls, repo):
         comp = compressionengine.fromrepo(repo)
         level = None
         if comp == b'zlib':
             level = repo.ui.configint(b'storage', b'revlog.zlib.level')
         elif comp == b'zstd':
             level = repo.ui.configint(b'storage', b'revlog.zstd.level')
         if level is None:
             return b'default'
         return bytes(level)
 
     @classmethod
     def fromconfig(cls, repo):
         comp = compressionengine.fromconfig(repo)
         level = None
         if comp == b'zlib':
             level = repo.ui.configint(b'storage', b'revlog.zlib.level')
         elif comp == b'zstd':
             level = repo.ui.configint(b'storage', b'revlog.zstd.level')
         if level is None:
             return b'default'
         return bytes(level)
 
 
 def find_format_upgrades(repo):
     """returns a list of format upgrades which can be perform on the repo"""
     upgrades = []
 
     # We could detect lack of revlogv1 and store here, but they were added
     # in 0.9.2 and we don't support upgrading repos without these
     # requirements, so let's not bother.
 
     for fv in allformatvariant:
         if not fv.fromrepo(repo):
             upgrades.append(fv)
 
     return upgrades
 
 
 def find_format_downgrades(repo):
     """returns a list of format downgrades which will be performed on the repo
     because of disabled config option for them"""
 
     downgrades = []
 
     for fv in allformatvariant:
         if fv.name == b'compression':
             # If there is a compression change between repository
             # and config, destination repository compression will change
             # and current compression will be removed.
             if fv.fromrepo(repo) != fv.fromconfig(repo):
                 downgrades.append(fv)
             continue
         # format variant exist in repo but does not exist in new repository
         # config
         if fv.fromrepo(repo) and not fv.fromconfig(repo):
             downgrades.append(fv)
 
     return downgrades
 
 
 ALL_OPTIMISATIONS = []
 
 
 def register_optimization(obj):
     ALL_OPTIMISATIONS.append(obj)
     return obj
 
 
 class optimization(improvement):
     """an improvement subclass dedicated to optimizations"""
 
     type = OPTIMISATION
 
 
 @register_optimization
 class redeltaparents(optimization):
     name = b're-delta-parent'
 
     type = OPTIMISATION
 
     description = _(
         b'deltas within internal storage will be recalculated to '
         b'choose an optimal base revision where this was not '
         b'already done; the size of the repository may shrink and '
         b'various operations may become faster; the first time '
         b'this optimization is performed could slow down upgrade '
         b'execution considerably; subsequent invocations should '
         b'not run noticeably slower'
     )
 
     upgrademessage = _(
         b'deltas within internal storage will choose a new '
         b'base revision if needed'
     )
 
 
 @register_optimization
 class redeltamultibase(optimization):
     name = b're-delta-multibase'
 
     type = OPTIMISATION
 
     description = _(
         b'deltas within internal storage will be recalculated '
         b'against multiple base revision and the smallest '
         b'difference will be used; the size of the repository may '
         b'shrink significantly when there are many merges; this '
         b'optimization will slow down execution in proportion to '
         b'the number of merges in the repository and the amount '
         b'of files in the repository; this slow down should not '
         b'be significant unless there are tens of thousands of '
         b'files and thousands of merges'
     )
 
     upgrademessage = _(
         b'deltas within internal storage will choose an '
         b'optimal delta by computing deltas against multiple '
         b'parents; may slow down execution time '
         b'significantly'
     )
 
 
 @register_optimization
 class redeltaall(optimization):
     name = b're-delta-all'
 
     type = OPTIMISATION
 
     description = _(
         b'deltas within internal storage will always be '
         b'recalculated without reusing prior deltas; this will '
         b'likely make execution run several times slower; this '
         b'optimization is typically not needed'
     )
 
     upgrademessage = _(
         b'deltas within internal storage will be fully '
         b'recomputed; this will likely drastically slow down '
         b'execution time'
     )
 
 
 @register_optimization
 class redeltafulladd(optimization):
     name = b're-delta-fulladd'
 
     type = OPTIMISATION
 
     description = _(
         b'every revision will be re-added as if it was new '
         b'content. It will go through the full storage '
         b'mechanism giving extensions a chance to process it '
         b'(eg. lfs). This is similar to "re-delta-all" but even '
         b'slower since more logic is involved.'
     )
 
     upgrademessage = _(
         b'each revision will be added as new content to the '
         b'internal storage; this will likely drastically slow '
         b'down execution time, but some extensions might need '
         b'it'
     )
 
 
 def findoptimizations(repo):
     """Determine optimisation that could be used during upgrade"""
     # These are unconditionally added. There is logic later that figures out
     # which ones to apply.
     return list(ALL_OPTIMISATIONS)
 
 
 def determine_upgrade_actions(
     repo, format_upgrades, optimizations, sourcereqs, destreqs
 ):
     """Determine upgrade actions that will be performed.
 
     Given a list of improvements as returned by ``find_format_upgrades`` and
     ``findoptimizations``, determine the list of upgrade actions that
     will be performed.
 
     The role of this function is to filter improvements if needed, apply
     recommended optimizations from the improvements list that make sense,
     etc.
 
     Returns a list of action names.
     """
     newactions = []
 
     for d in format_upgrades:
         name = d._requirement
 
         # If the action is a requirement that doesn't show up in the
         # destination requirements, prune the action.
         if name is not None and name not in destreqs:
             continue
 
         newactions.append(d)
 
     newactions.extend(o for o in sorted(optimizations) if o not in newactions)
 
     # FUTURE consider adding some optimizations here for certain transitions.
     # e.g. adding generaldelta could schedule parent redeltas.
 
     return newactions
 
 
 class UpgradeOperation(object):
     """represent the work to be done during an upgrade"""
 
     def __init__(
         self,
         ui,
         new_requirements,
         current_requirements,
         upgrade_actions,
         removed_actions,
         revlogs_to_process,
         backup_store,
     ):
         self.ui = ui
         self.new_requirements = new_requirements
         self.current_requirements = current_requirements
         # list of upgrade actions the operation will perform
         self.upgrade_actions = upgrade_actions
-        self._upgrade_actions_names = set([a.name for a in upgrade_actions])
         self.removed_actions = removed_actions
         self.revlogs_to_process = revlogs_to_process
         # requirements which will be added by the operation
         self._added_requirements = (
             self.new_requirements - self.current_requirements
         )
         # requirements which will be removed by the operation
         self._removed_requirements = (
             self.current_requirements - self.new_requirements
         )
         # requirements which will be preserved by the operation
         self._preserved_requirements = (
             self.current_requirements & self.new_requirements
         )
         # optimizations which are not used and it's recommended that they
         # should use them
         all_optimizations = findoptimizations(None)
         self.unused_optimizations = [
             i for i in all_optimizations if i not in self.upgrade_actions
         ]
 
         # delta reuse mode of this upgrade operation
+        upgrade_actions_names = self.upgrade_actions_names
         self.delta_reuse_mode = revlog.revlog.DELTAREUSEALWAYS
-        if b're-delta-all' in self._upgrade_actions_names:
+        if b're-delta-all' in upgrade_actions_names:
             self.delta_reuse_mode = revlog.revlog.DELTAREUSENEVER
-        elif b're-delta-parent' in self._upgrade_actions_names:
+        elif b're-delta-parent' in upgrade_actions_names:
             self.delta_reuse_mode = revlog.revlog.DELTAREUSESAMEREVS
-        elif b're-delta-multibase' in self._upgrade_actions_names:
+        elif b're-delta-multibase' in upgrade_actions_names:
             self.delta_reuse_mode = revlog.revlog.DELTAREUSESAMEREVS
-        elif b're-delta-fulladd' in self._upgrade_actions_names:
+        elif b're-delta-fulladd' in upgrade_actions_names:
             self.delta_reuse_mode = revlog.revlog.DELTAREUSEFULLADD
 
         # should this operation force re-delta of both parents
         self.force_re_delta_both_parents = (
-            b're-delta-multibase' in self._upgrade_actions_names
+            b're-delta-multibase' in upgrade_actions_names
         )
 
         # should this operation create a backup of the store
         self.backup_store = backup_store
 
-        # whether the operation touches different revlogs at all or not
-        self.touches_filelogs = self._touches_filelogs()
-        self.touches_manifests = self._touches_manifests()
-        self.touches_changelog = self._touches_changelog()
-        # whether the operation touches requirements file or not
-        self.touches_requirements = self._touches_requirements()
-        self.touches_store = (
-            self.touches_filelogs
-            or self.touches_manifests
-            or self.touches_changelog
-        )
+    @property
+    def upgrade_actions_names(self):
+        return set([a.name for a in self.upgrade_actions])
+
+    @property
+    def requirements_only(self):
         # does the operation only touches repository requirement
-        self.requirements_only = (
-            self.touches_requirements and not self.touches_store
+        return (
+            self.touches_requirements
+            and not self.touches_filelogs
+            and not self.touches_manifests
+            and not self.touches_changelog
+            and not self.touches_dirstate
         )
 
-    def _touches_filelogs(self):
+    @property
+    def touches_filelogs(self):
         for a in self.upgrade_actions:
             # in optimisations, we re-process the revlogs again
             if a.type == OPTIMISATION:
                 return True
             elif a.touches_filelogs:
                 return True
         for a in self.removed_actions:
             if a.touches_filelogs:
                 return True
         return False
 
-    def _touches_manifests(self):
+    @property
+    def touches_manifests(self):
         for a in self.upgrade_actions:
             # in optimisations, we re-process the revlogs again
             if a.type == OPTIMISATION:
                 return True
             elif a.touches_manifests:
                 return True
         for a in self.removed_actions:
             if a.touches_manifests:
                 return True
         return False
 
-    def _touches_changelog(self):
+    @property
+    def touches_changelog(self):
         for a in self.upgrade_actions:
             # in optimisations, we re-process the revlogs again
             if a.type == OPTIMISATION:
                 return True
             elif a.touches_changelog:
                 return True
         for a in self.removed_actions:
             if a.touches_changelog:
                 return True
         return False
 
-    def _touches_requirements(self):
+    @property
+    def touches_requirements(self):
         for a in self.upgrade_actions:
             # optimisations are used to re-process revlogs and does not result
             # in a requirement being added or removed
             if a.type == OPTIMISATION:
                 pass
             elif a.touches_requirements:
                 return True
         for a in self.removed_actions:
             if a.touches_requirements:
                 return True
 
+    @property
+    def touches_dirstate(self):
+        for a in self.upgrade_actions:
+            # revlog optimisations do not affect the dirstate
+            if a.type == OPTIMISATION:
+                pass
+            elif a.touches_dirstate:
+                return True
+        for a in self.removed_actions:
+            if a.touches_dirstate:
+                return True
+
         return False
 
     def _write_labeled(self, l, label):
         """
         Utility function to aid writing of a list under one label
         """
         first = True
         for r in sorted(l):
             if not first:
                 self.ui.write(b', ')
             self.ui.write(r, label=label)
             first = False
 
     def print_requirements(self):
         self.ui.write(_(b'requirements\n'))
         self.ui.write(_(b'   preserved: '))
         self._write_labeled(
             self._preserved_requirements, "upgrade-repo.requirement.preserved"
         )
         self.ui.write((b'\n'))
         if self._removed_requirements:
             self.ui.write(_(b'   removed: '))
             self._write_labeled(
                 self._removed_requirements, "upgrade-repo.requirement.removed"
             )
             self.ui.write((b'\n'))
         if self._added_requirements:
             self.ui.write(_(b'   added: '))
             self._write_labeled(
                 self._added_requirements, "upgrade-repo.requirement.added"
             )
             self.ui.write((b'\n'))
         self.ui.write(b'\n')
 
     def print_optimisations(self):
         optimisations = [
             a for a in self.upgrade_actions if a.type == OPTIMISATION
         ]
         optimisations.sort(key=lambda a: a.name)
         if optimisations:
             self.ui.write(_(b'optimisations: '))
             self._write_labeled(
                 [a.name for a in optimisations],
                 "upgrade-repo.optimisation.performed",
             )
             self.ui.write(b'\n\n')
 
     def print_upgrade_actions(self):
         for a in self.upgrade_actions:
             self.ui.status(b'%s\n   %s\n\n' % (a.name, a.upgrademessage))
 
     def print_affected_revlogs(self):
         if not self.revlogs_to_process:
             self.ui.write((b'no revlogs to process\n'))
         else:
             self.ui.write((b'processed revlogs:\n'))
             for r in sorted(self.revlogs_to_process):
                 self.ui.write((b'  - %s\n' % r))
         self.ui.write((b'\n'))
 
     def print_unused_optimizations(self):
         for i in self.unused_optimizations:
             self.ui.status(_(b'%s\n   %s\n\n') % (i.name, i.description))
 
     def has_upgrade_action(self, name):
         """Check whether the upgrade operation will perform this action"""
         return name in self._upgrade_actions_names
 
     def print_post_op_messages(self):
         """print post upgrade operation warning messages"""
         for a in self.upgrade_actions:
             if a.postupgrademessage is not None:
                 self.ui.warn(b'%s\n' % a.postupgrademessage)
         for a in self.removed_actions:
             if a.postdowngrademessage is not None:
                 self.ui.warn(b'%s\n' % a.postdowngrademessage)
 
 
 ###  Code checking if a repository can got through the upgrade process at all. #
 
 
 def requiredsourcerequirements(repo):
     """Obtain requirements required to be present to upgrade a repo.
 
     An upgrade will not be allowed if the repository doesn't have the
     requirements returned by this function.
     """
     return {
         # Introduced in Mercurial 0.9.2.
         requirements.STORE_REQUIREMENT,
     }
 
 
 def blocksourcerequirements(repo):
     """Obtain requirements that will prevent an upgrade from occurring.
 
     An upgrade cannot be performed if the source repository contains a
     requirements in the returned set.
     """
     return {
         # The upgrade code does not yet support these experimental features.
         # This is an artificial limitation.
         requirements.TREEMANIFEST_REQUIREMENT,
         # This was a precursor to generaldelta and was never enabled by default.
         # It should (hopefully) not exist in the wild.
         b'parentdelta',
         # Upgrade should operate on the actual store, not the shared link.
         requirements.SHARED_REQUIREMENT,
     }
 
 
 def check_revlog_version(reqs):
     """Check that the requirements contain at least one Revlog version"""
     all_revlogs = {
         requirements.REVLOGV1_REQUIREMENT,
         requirements.REVLOGV2_REQUIREMENT,
     }
     if not all_revlogs.intersection(reqs):
         msg = _(b'cannot upgrade repository; missing a revlog version')
         raise error.Abort(msg)
 
 
 def check_source_requirements(repo):
     """Ensure that no existing requirements prevent the repository upgrade"""
 
     check_revlog_version(repo.requirements)
     required = requiredsourcerequirements(repo)
     missingreqs = required - repo.requirements
     if missingreqs:
         msg = _(b'cannot upgrade repository; requirement missing: %s')
         missingreqs = b', '.join(sorted(missingreqs))
         raise error.Abort(msg % missingreqs)
 
     blocking = blocksourcerequirements(repo)
     blockingreqs = blocking & repo.requirements
     if blockingreqs:
         m = _(b'cannot upgrade repository; unsupported source requirement: %s')
         blockingreqs = b', '.join(sorted(blockingreqs))
         raise error.Abort(m % blockingreqs)
 
 
 ### Verify the validity of the planned requirement changes ####################
 
 
 def supportremovedrequirements(repo):
     """Obtain requirements that can be removed during an upgrade.
 
     If an upgrade were to create a repository that dropped a requirement,
     the dropped requirement must appear in the returned set for the upgrade
     to be allowed.
     """
     supported = {
         requirements.SPARSEREVLOG_REQUIREMENT,
         requirements.COPIESSDC_REQUIREMENT,
         requirements.NODEMAP_REQUIREMENT,
         requirements.SHARESAFE_REQUIREMENT,
         requirements.REVLOGV2_REQUIREMENT,
         requirements.CHANGELOGV2_REQUIREMENT,
         requirements.REVLOGV1_REQUIREMENT,
+        requirements.DIRSTATE_V2_REQUIREMENT,
     }
     for name in compression.compengines:
         engine = compression.compengines[name]
         if engine.available() and engine.revlogheader():
             supported.add(b'exp-compression-%s' % name)
             if engine.name() == b'zstd':
                 supported.add(b'revlog-compression-zstd')
     return supported
 
 
 def supporteddestrequirements(repo):
     """Obtain requirements that upgrade supports in the destination.
 
     If the result of the upgrade would create requirements not in this set,
     the upgrade is disallowed.
 
     Extensions should monkeypatch this to add their custom requirements.
     """
     supported = {
         requirements.DOTENCODE_REQUIREMENT,
         requirements.FNCACHE_REQUIREMENT,
         requirements.GENERALDELTA_REQUIREMENT,
         requirements.REVLOGV1_REQUIREMENT,  # allowed in case of downgrade
         requirements.STORE_REQUIREMENT,
         requirements.SPARSEREVLOG_REQUIREMENT,
         requirements.COPIESSDC_REQUIREMENT,
         requirements.NODEMAP_REQUIREMENT,
         requirements.SHARESAFE_REQUIREMENT,
         requirements.REVLOGV2_REQUIREMENT,
         requirements.CHANGELOGV2_REQUIREMENT,
         requirements.DIRSTATE_V2_REQUIREMENT,
     }
     for name in compression.compengines:
         engine = compression.compengines[name]
         if engine.available() and engine.revlogheader():
             supported.add(b'exp-compression-%s' % name)
             if engine.name() == b'zstd':
                 supported.add(b'revlog-compression-zstd')
     return supported
 
 
 def allowednewrequirements(repo):
     """Obtain requirements that can be added to a repository during upgrade.
 
     This is used to disallow proposed requirements from being added when
     they weren't present before.
 
     We use a list of allowed requirement additions instead of a list of known
     bad additions because the whitelist approach is safer and will prevent
     future, unknown requirements from accidentally being added.
     """
     supported = {
         requirements.DOTENCODE_REQUIREMENT,
         requirements.FNCACHE_REQUIREMENT,
         requirements.GENERALDELTA_REQUIREMENT,
         requirements.SPARSEREVLOG_REQUIREMENT,
         requirements.COPIESSDC_REQUIREMENT,
         requirements.NODEMAP_REQUIREMENT,
         requirements.SHARESAFE_REQUIREMENT,
         requirements.REVLOGV1_REQUIREMENT,
         requirements.REVLOGV2_REQUIREMENT,
         requirements.CHANGELOGV2_REQUIREMENT,
+        requirements.DIRSTATE_V2_REQUIREMENT,
     }
     for name in compression.compengines:
         engine = compression.compengines[name]
         if engine.available() and engine.revlogheader():
             supported.add(b'exp-compression-%s' % name)
             if engine.name() == b'zstd':
                 supported.add(b'revlog-compression-zstd')
     return supported
 
 
 def check_requirements_changes(repo, new_reqs):
     old_reqs = repo.requirements
     check_revlog_version(repo.requirements)
     support_removal = supportremovedrequirements(repo)
     no_remove_reqs = old_reqs - new_reqs - support_removal
     if no_remove_reqs:
         msg = _(b'cannot upgrade repository; requirement would be removed: %s')
         no_remove_reqs = b', '.join(sorted(no_remove_reqs))
         raise error.Abort(msg % no_remove_reqs)
 
     support_addition = allowednewrequirements(repo)
     no_add_reqs = new_reqs - old_reqs - support_addition
     if no_add_reqs:
         m = _(b'cannot upgrade repository; do not support adding requirement: ')
         no_add_reqs = b', '.join(sorted(no_add_reqs))
         raise error.Abort(m + no_add_reqs)
 
     supported = supporteddestrequirements(repo)
     unsupported_reqs = new_reqs - supported
     if unsupported_reqs:
         msg = _(
             b'cannot upgrade repository; do not support destination '
             b'requirement: %s'
         )
         unsupported_reqs = b', '.join(sorted(unsupported_reqs))
         raise error.Abort(msg % unsupported_reqs)

mercurial/upgrade_utils/engine.py

@@ -1,593 +1,632
 # upgrade.py - functions for in place upgrade of Mercurial repository
 #
 # Copyright (c) 2016-present, Gregory Szorc
 #
 # 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 stat
 
 from ..i18n import _
 from ..pycompat import getattr
 from .. import (
     changelog,
     error,
     filelog,
     manifest,
     metadata,
     pycompat,
     requirements,
     scmutil,
     store,
     util,
     vfs as vfsmod,
 )
 from ..revlogutils import (
     constants as revlogconst,
     flagutil,
     nodemap,
     sidedata as sidedatamod,
 )
+from . import actions as upgrade_actions
 
 
 def get_sidedata_helpers(srcrepo, dstrepo):
     use_w = srcrepo.ui.configbool(b'experimental', b'worker.repository-upgrade')
     sequential = pycompat.iswindows or not use_w
     if not sequential:
         srcrepo.register_sidedata_computer(
             revlogconst.KIND_CHANGELOG,
             sidedatamod.SD_FILES,
             (sidedatamod.SD_FILES,),
             metadata._get_worker_sidedata_adder(srcrepo, dstrepo),
             flagutil.REVIDX_HASCOPIESINFO,
             replace=True,
         )
     return sidedatamod.get_sidedata_helpers(srcrepo, dstrepo._wanted_sidedata)
 
 
 def _revlogfrompath(repo, rl_type, path):
     """Obtain a revlog from a repo path.
 
     An instance of the appropriate class is returned.
     """
     if rl_type & store.FILEFLAGS_CHANGELOG:
         return changelog.changelog(repo.svfs)
     elif rl_type & store.FILEFLAGS_MANIFESTLOG:
         mandir = b''
         if b'/' in path:
             mandir = path.rsplit(b'/', 1)[0]
         return manifest.manifestrevlog(
             repo.nodeconstants, repo.svfs, tree=mandir
         )
     else:
         # drop the extension and the `data/` prefix
         path = path.rsplit(b'.', 1)[0].split(b'/', 1)[1]
         return filelog.filelog(repo.svfs, path)
 
 
 def _copyrevlog(tr, destrepo, oldrl, rl_type, unencodedname):
     """copy all relevant files for `oldrl` into `destrepo` store
 
     Files are copied "as is" without any transformation. The copy is performed
     without extra checks. Callers are responsible for making sure the copied
     content is compatible with format of the destination repository.
     """
     oldrl = getattr(oldrl, '_revlog', oldrl)
     newrl = _revlogfrompath(destrepo, rl_type, unencodedname)
     newrl = getattr(newrl, '_revlog', newrl)
 
     oldvfs = oldrl.opener
     newvfs = newrl.opener
     oldindex = oldvfs.join(oldrl._indexfile)
     newindex = newvfs.join(newrl._indexfile)
     olddata = oldvfs.join(oldrl._datafile)
     newdata = newvfs.join(newrl._datafile)
 
     with newvfs(newrl._indexfile, b'w'):
         pass  # create all the directories
 
     util.copyfile(oldindex, newindex)
     copydata = oldrl.opener.exists(oldrl._datafile)
     if copydata:
         util.copyfile(olddata, newdata)
 
     if rl_type & store.FILEFLAGS_FILELOG:
         destrepo.svfs.fncache.add(unencodedname)
         if copydata:
             destrepo.svfs.fncache.add(unencodedname[:-2] + b'.d')
 
 
 UPGRADE_CHANGELOG = b"changelog"
 UPGRADE_MANIFEST = b"manifest"
 UPGRADE_FILELOGS = b"all-filelogs"
 
 UPGRADE_ALL_REVLOGS = frozenset(
     [UPGRADE_CHANGELOG, UPGRADE_MANIFEST, UPGRADE_FILELOGS]
 )
 
 
 def matchrevlog(revlogfilter, rl_type):
     """check if a revlog is selected for cloning.
 
     In other words, are there any updates which need to be done on revlog
     or it can be blindly copied.
 
     The store entry is checked against the passed filter"""
     if rl_type & store.FILEFLAGS_CHANGELOG:
         return UPGRADE_CHANGELOG in revlogfilter
     elif rl_type & store.FILEFLAGS_MANIFESTLOG:
         return UPGRADE_MANIFEST in revlogfilter
     assert rl_type & store.FILEFLAGS_FILELOG
     return UPGRADE_FILELOGS in revlogfilter
 
 
 def _perform_clone(
     ui,
     dstrepo,
     tr,
     old_revlog,
     rl_type,
     unencoded,
     upgrade_op,
     sidedata_helpers,
     oncopiedrevision,
 ):
     """returns the new revlog object created"""
     newrl = None
     if matchrevlog(upgrade_op.revlogs_to_process, rl_type):
         ui.note(
             _(b'cloning %d revisions from %s\n') % (len(old_revlog), unencoded)
         )
         newrl = _revlogfrompath(dstrepo, rl_type, unencoded)
         old_revlog.clone(
             tr,
             newrl,
             addrevisioncb=oncopiedrevision,
             deltareuse=upgrade_op.delta_reuse_mode,
             forcedeltabothparents=upgrade_op.force_re_delta_both_parents,
             sidedata_helpers=sidedata_helpers,
         )
     else:
         msg = _(b'blindly copying %s containing %i revisions\n')
         ui.note(msg % (unencoded, len(old_revlog)))
         _copyrevlog(tr, dstrepo, old_revlog, rl_type, unencoded)
 
         newrl = _revlogfrompath(dstrepo, rl_type, unencoded)
     return newrl
 
 
 def _clonerevlogs(
     ui,
     srcrepo,
     dstrepo,
     tr,
     upgrade_op,
 ):
     """Copy revlogs between 2 repos."""
     revcount = 0
     srcsize = 0
     srcrawsize = 0
     dstsize = 0
     fcount = 0
     frevcount = 0
     fsrcsize = 0
     frawsize = 0
     fdstsize = 0
     mcount = 0
     mrevcount = 0
     msrcsize = 0
     mrawsize = 0
     mdstsize = 0
     crevcount = 0
     csrcsize = 0
     crawsize = 0
     cdstsize = 0
 
     alldatafiles = list(srcrepo.store.walk())
     # mapping of data files which needs to be cloned
     # key is unencoded filename
     # value is revlog_object_from_srcrepo
     manifests = {}
     changelogs = {}
     filelogs = {}
 
     # Perform a pass to collect metadata. This validates we can open all
     # source files and allows a unified progress bar to be displayed.
     for rl_type, unencoded, encoded, size in alldatafiles:
         if not rl_type & store.FILEFLAGS_REVLOG_MAIN:
             continue
 
         rl = _revlogfrompath(srcrepo, rl_type, unencoded)
 
         info = rl.storageinfo(
             exclusivefiles=True,
             revisionscount=True,
             trackedsize=True,
             storedsize=True,
         )
 
         revcount += info[b'revisionscount'] or 0
         datasize = info[b'storedsize'] or 0
         rawsize = info[b'trackedsize'] or 0
 
         srcsize += datasize
         srcrawsize += rawsize
 
         # This is for the separate progress bars.
         if rl_type & store.FILEFLAGS_CHANGELOG:
             changelogs[unencoded] = (rl_type, rl)
             crevcount += len(rl)
             csrcsize += datasize
             crawsize += rawsize
         elif rl_type & store.FILEFLAGS_MANIFESTLOG:
             manifests[unencoded] = (rl_type, rl)
             mcount += 1
             mrevcount += len(rl)
             msrcsize += datasize
             mrawsize += rawsize
         elif rl_type & store.FILEFLAGS_FILELOG:
             filelogs[unencoded] = (rl_type, rl)
             fcount += 1
             frevcount += len(rl)
             fsrcsize += datasize
             frawsize += rawsize
         else:
             error.ProgrammingError(b'unknown revlog type')
 
     if not revcount:
         return
 
     ui.status(
         _(
             b'migrating %d total revisions (%d in filelogs, %d in manifests, '
             b'%d in changelog)\n'
         )
         % (revcount, frevcount, mrevcount, crevcount)
     )
     ui.status(
         _(b'migrating %s in store; %s tracked data\n')
         % ((util.bytecount(srcsize), util.bytecount(srcrawsize)))
     )
 
     # Used to keep track of progress.
     progress = None
 
     def oncopiedrevision(rl, rev, node):
         progress.increment()
 
     sidedata_helpers = get_sidedata_helpers(srcrepo, dstrepo)
 
     # Migrating filelogs
     ui.status(
         _(
             b'migrating %d filelogs containing %d revisions '
             b'(%s in store; %s tracked data)\n'
         )
         % (
             fcount,
             frevcount,
             util.bytecount(fsrcsize),
             util.bytecount(frawsize),
         )
     )
     progress = srcrepo.ui.makeprogress(_(b'file revisions'), total=frevcount)
     for unencoded, (rl_type, oldrl) in sorted(filelogs.items()):
         newrl = _perform_clone(
             ui,
             dstrepo,
             tr,
             oldrl,
             rl_type,
             unencoded,
             upgrade_op,
             sidedata_helpers,
             oncopiedrevision,
         )
         info = newrl.storageinfo(storedsize=True)
         fdstsize += info[b'storedsize'] or 0
     ui.status(
         _(
             b'finished migrating %d filelog revisions across %d '
             b'filelogs; change in size: %s\n'
         )
         % (frevcount, fcount, util.bytecount(fdstsize - fsrcsize))
     )
 
     # Migrating manifests
     ui.status(
         _(
             b'migrating %d manifests containing %d revisions '
             b'(%s in store; %s tracked data)\n'
         )
         % (
             mcount,
             mrevcount,
             util.bytecount(msrcsize),
             util.bytecount(mrawsize),
         )
     )
     if progress:
         progress.complete()
     progress = srcrepo.ui.makeprogress(
         _(b'manifest revisions'), total=mrevcount
     )
     for unencoded, (rl_type, oldrl) in sorted(manifests.items()):
         newrl = _perform_clone(
             ui,
             dstrepo,
             tr,
             oldrl,
             rl_type,
             unencoded,
             upgrade_op,
             sidedata_helpers,
             oncopiedrevision,
         )
         info = newrl.storageinfo(storedsize=True)
         mdstsize += info[b'storedsize'] or 0
     ui.status(
         _(
             b'finished migrating %d manifest revisions across %d '
             b'manifests; change in size: %s\n'
         )
         % (mrevcount, mcount, util.bytecount(mdstsize - msrcsize))
     )
 
     # Migrating changelog
     ui.status(
         _(
             b'migrating changelog containing %d revisions '
             b'(%s in store; %s tracked data)\n'
         )
         % (
             crevcount,
             util.bytecount(csrcsize),
             util.bytecount(crawsize),
         )
     )
     if progress:
         progress.complete()
     progress = srcrepo.ui.makeprogress(
         _(b'changelog revisions'), total=crevcount
     )
     for unencoded, (rl_type, oldrl) in sorted(changelogs.items()):
         newrl = _perform_clone(
             ui,
             dstrepo,
             tr,
             oldrl,
             rl_type,
             unencoded,
             upgrade_op,
             sidedata_helpers,
             oncopiedrevision,
         )
         info = newrl.storageinfo(storedsize=True)
         cdstsize += info[b'storedsize'] or 0
     progress.complete()
     ui.status(
         _(
             b'finished migrating %d changelog revisions; change in size: '
             b'%s\n'
         )
         % (crevcount, util.bytecount(cdstsize - csrcsize))
     )
 
     dstsize = fdstsize + mdstsize + cdstsize
     ui.status(
         _(
             b'finished migrating %d total revisions; total change in store '
             b'size: %s\n'
         )
         % (revcount, util.bytecount(dstsize - srcsize))
     )
 
 
 def _files_to_copy_post_revlog_clone(srcrepo):
     """yields files which should be copied to destination after revlogs
     are cloned"""
     for path, kind, st in sorted(srcrepo.store.vfs.readdir(b'', stat=True)):
         # don't copy revlogs as they are already cloned
         if store.revlog_type(path) is not None:
             continue
         # Skip transaction related files.
         if path.startswith(b'undo'):
             continue
         # Only copy regular files.
         if kind != stat.S_IFREG:
             continue
         # Skip other skipped files.
         if path in (b'lock', b'fncache'):
             continue
         # TODO: should we skip cache too?
 
         yield path
 
 
 def _replacestores(currentrepo, upgradedrepo, backupvfs, upgrade_op):
     """Replace the stores after current repository is upgraded
 
     Creates a backup of current repository store at backup path
     Replaces upgraded store files in current repo from upgraded one
 
     Arguments:
       currentrepo: repo object of current repository
       upgradedrepo: repo object of the upgraded data
       backupvfs: vfs object for the backup path
       upgrade_op: upgrade operation object
                   to be used to decide what all is upgraded
     """
     # TODO: don't blindly rename everything in store
     # There can be upgrades where store is not touched at all
     if upgrade_op.backup_store:
         util.rename(currentrepo.spath, backupvfs.join(b'store'))
     else:
         currentrepo.vfs.rmtree(b'store', forcibly=True)
     util.rename(upgradedrepo.spath, currentrepo.spath)
 
 
 def finishdatamigration(ui, srcrepo, dstrepo, requirements):
     """Hook point for extensions to perform additional actions during upgrade.
 
     This function is called after revlogs and store files have been copied but
     before the new store is swapped into the original location.
     """
 
 
 def upgrade(ui, srcrepo, dstrepo, upgrade_op):
     """Do the low-level work of upgrading a repository.
 
     The upgrade is effectively performed as a copy between a source
     repository and a temporary destination repository.
 
     The source repository is unmodified for as long as possible so the
     upgrade can abort at any time without causing loss of service for
     readers and without corrupting the source repository.
     """
     assert srcrepo.currentwlock()
     assert dstrepo.currentwlock()
     backuppath = None
     backupvfs = None
 
     ui.status(
         _(
             b'(it is safe to interrupt this process any time before '
             b'data migration completes)\n'
         )
     )
 
+    if upgrade_actions.dirstatev2 in upgrade_op.upgrade_actions:
+        ui.status(_(b'upgrading to dirstate-v2 from v1\n'))
+        upgrade_dirstate(ui, srcrepo, upgrade_op, b'v1', b'v2')
+        upgrade_op.upgrade_actions.remove(upgrade_actions.dirstatev2)
+
+    if upgrade_actions.dirstatev2 in upgrade_op.removed_actions:
+        ui.status(_(b'downgrading from dirstate-v2 to v1\n'))
+        upgrade_dirstate(ui, srcrepo, upgrade_op, b'v2', b'v1')
+        upgrade_op.removed_actions.remove(upgrade_actions.dirstatev2)
+
+    if not (upgrade_op.upgrade_actions or upgrade_op.removed_actions):
+        return
+
     if upgrade_op.requirements_only:
         ui.status(_(b'upgrading repository requirements\n'))
         scmutil.writereporequirements(srcrepo, upgrade_op.new_requirements)
     # if there is only one action and that is persistent nodemap upgrade
     # directly write the nodemap file and update requirements instead of going
     # through the whole cloning process
     elif (
         len(upgrade_op.upgrade_actions) == 1
-        and b'persistent-nodemap' in upgrade_op._upgrade_actions_names
+        and b'persistent-nodemap' in upgrade_op.upgrade_actions_names
         and not upgrade_op.removed_actions
     ):
         ui.status(
             _(b'upgrading repository to use persistent nodemap feature\n')
         )
         with srcrepo.transaction(b'upgrade') as tr:
             unfi = srcrepo.unfiltered()
             cl = unfi.changelog
             nodemap.persist_nodemap(tr, cl, force=True)
             # we want to directly operate on the underlying revlog to force
             # create a nodemap file. This is fine since this is upgrade code
             # and it heavily relies on repository being revlog based
             # hence accessing private attributes can be justified
             nodemap.persist_nodemap(
                 tr, unfi.manifestlog._rootstore._revlog, force=True
             )
         scmutil.writereporequirements(srcrepo, upgrade_op.new_requirements)
     elif (
         len(upgrade_op.removed_actions) == 1
         and [
             x
             for x in upgrade_op.removed_actions
             if x.name == b'persistent-nodemap'
         ]
         and not upgrade_op.upgrade_actions
     ):
         ui.status(
             _(b'downgrading repository to not use persistent nodemap feature\n')
         )
         with srcrepo.transaction(b'upgrade') as tr:
             unfi = srcrepo.unfiltered()
             cl = unfi.changelog
             nodemap.delete_nodemap(tr, srcrepo, cl)
             # check comment 20 lines above for accessing private attributes
             nodemap.delete_nodemap(
                 tr, srcrepo, unfi.manifestlog._rootstore._revlog
             )
         scmutil.writereporequirements(srcrepo, upgrade_op.new_requirements)
     else:
         with dstrepo.transaction(b'upgrade') as tr:
             _clonerevlogs(
                 ui,
                 srcrepo,
                 dstrepo,
                 tr,
                 upgrade_op,
             )
 
         # Now copy other files in the store directory.
         for p in _files_to_copy_post_revlog_clone(srcrepo):
             srcrepo.ui.status(_(b'copying %s\n') % p)
             src = srcrepo.store.rawvfs.join(p)
             dst = dstrepo.store.rawvfs.join(p)
             util.copyfile(src, dst, copystat=True)
 
         finishdatamigration(ui, srcrepo, dstrepo, requirements)
 
         ui.status(_(b'data fully upgraded in a temporary repository\n'))
 
         if upgrade_op.backup_store:
             backuppath = pycompat.mkdtemp(
                 prefix=b'upgradebackup.', dir=srcrepo.path
             )
             backupvfs = vfsmod.vfs(backuppath)
 
             # Make a backup of requires file first, as it is the first to be modified.
             util.copyfile(
                 srcrepo.vfs.join(b'requires'), backupvfs.join(b'requires')
             )
 
         # We install an arbitrary requirement that clients must not support
         # as a mechanism to lock out new clients during the data swap. This is
         # better than allowing a client to continue while the repository is in
         # an inconsistent state.
         ui.status(
             _(
                 b'marking source repository as being upgraded; clients will be '
                 b'unable to read from repository\n'
             )
         )
         scmutil.writereporequirements(
             srcrepo, srcrepo.requirements | {b'upgradeinprogress'}
         )
 
         ui.status(_(b'starting in-place swap of repository data\n'))
         if upgrade_op.backup_store:
             ui.status(
                 _(b'replaced files will be backed up at %s\n') % backuppath
             )
 
         # Now swap in the new store directory. Doing it as a rename should make
         # the operation nearly instantaneous and atomic (at least in well-behaved
         # environments).
         ui.status(_(b'replacing store...\n'))
         tstart = util.timer()
         _replacestores(srcrepo, dstrepo, backupvfs, upgrade_op)
         elapsed = util.timer() - tstart
         ui.status(
             _(
                 b'store replacement complete; repository was inconsistent for '
                 b'%0.1fs\n'
             )
             % elapsed
         )
 
         # We first write the requirements file. Any new requirements will lock
         # out legacy clients.
         ui.status(
             _(
                 b'finalizing requirements file and making repository readable '
                 b'again\n'
             )
         )
         scmutil.writereporequirements(srcrepo, upgrade_op.new_requirements)
 
         if upgrade_op.backup_store:
             # The lock file from the old store won't be removed because nothing has a
             # reference to its new location. So clean it up manually. Alternatively, we
             # could update srcrepo.svfs and other variables to point to the new
             # location. This is simpler.
             assert backupvfs is not None  # help pytype
             backupvfs.unlink(b'store/lock')
 
     return backuppath
+
+
+def upgrade_dirstate(ui, srcrepo, upgrade_op, old, new):
+    if upgrade_op.backup_store:
+        backuppath = pycompat.mkdtemp(
+            prefix=b'upgradebackup.', dir=srcrepo.path
+        )
+        ui.status(_(b'replaced files will be backed up at %s\n') % backuppath)
+        backupvfs = vfsmod.vfs(backuppath)
+        util.copyfile(
+            srcrepo.vfs.join(b'requires'), backupvfs.join(b'requires')
+        )
+        util.copyfile(
+            srcrepo.vfs.join(b'dirstate'), backupvfs.join(b'dirstate')
+        )
+
+    assert srcrepo.dirstate._use_dirstate_v2 == (old == b'v2')
+    srcrepo.dirstate._map._use_dirstate_tree = True
+    srcrepo.dirstate._map.preload()
+    srcrepo.dirstate._use_dirstate_v2 = new == b'v2'
+    srcrepo.dirstate._map._use_dirstate_v2 = srcrepo.dirstate._use_dirstate_v2
+    srcrepo.dirstate._dirty = True
+    srcrepo.dirstate.write(None)
+
+    scmutil.writereporequirements(srcrepo, upgrade_op.new_requirements)

tests/test-copies-chain-merge.t

@@ -1,3872 +1,3874
 #testcases filelog compatibility changeset sidedata upgraded upgraded-parallel pull push pull-upgrade push-upgrade
 
 =====================================================
 Test Copy tracing for chain of copies involving merge
 =====================================================
 
 This test files covers copies/rename case for a chains of commit where merges
 are involved. It cheks we do not have unwanted update of behavior and that the
 different options to retrieve copies behave correctly.
 
 
 Setup
 =====
 
 use git diff to see rename
 
   $ cat << EOF >> ./no-linkrev
   > #!$PYTHON
   > # filter out linkrev part of the debugindex command
   > import sys
   > for line in sys.stdin:
   >     if " linkrev " in line:
   >         print(line.rstrip())
   >     else:
   >         l = "%s       *%s" % (line[:6], line[14:].rstrip())
   >         print(l)
   > EOF
 
   $ cat << EOF >> $HGRCPATH
   > [diff]
   > git=yes
   > [command-templates]
   > log={desc}\n
   > EOF
 
 #if compatibility
   $ cat >> $HGRCPATH << EOF
   > [experimental]
   > copies.read-from = compatibility
   > EOF
 #endif
 
 #if changeset
   $ cat >> $HGRCPATH << EOF
   > [experimental]
   > copies.read-from = changeset-only
   > copies.write-to = changeset-only
   > EOF
 #endif
 
 #if sidedata
   $ cat >> $HGRCPATH << EOF
   > [format]
   > exp-use-copies-side-data-changeset = yes
   > EOF
 #endif
 
 #if pull
   $ cat >> $HGRCPATH << EOF
   > [format]
   > exp-use-copies-side-data-changeset = yes
   > EOF
 #endif
 
 #if push
   $ cat >> $HGRCPATH << EOF
   > [format]
   > exp-use-copies-side-data-changeset = yes
   > EOF
 #endif
 
 #if pull-upgrade
   $ cat >> $HGRCPATH << EOF
   > [format]
   > exp-use-copies-side-data-changeset = no
   > [experimental]
   > changegroup4 = yes
   > EOF
 #endif
 
 #if push-upgrade
   $ cat >> $HGRCPATH << EOF
   > [format]
   > exp-use-copies-side-data-changeset = no
   > [experimental]
   > changegroup4 = yes
   > EOF
 #endif
 
   $ cat > same-content.txt << EOF
   > Here is some content that will be the same accros multiple file.
   > 
   > This is done on purpose so that we end up in some merge situation, were the
   > resulting content is the same as in the parent(s), but a new filenodes still
   > need to be created to record some file history information (especially
   > about copies).
   > EOF
 
   $ hg init repo-chain
   $ cd repo-chain
 
 Add some linear rename initialy
 
   $ cp ../same-content.txt a
   $ cp ../same-content.txt b
   $ cp ../same-content.txt h
   $ echo "original content for P" > p
   $ echo "original content for Q" > q
   $ echo "original content for R" > r
   $ hg ci -Am 'i-0 initial commit: a b h p q r'
   adding a
   adding b
   adding h
   adding p
   adding q
   adding r
   $ hg mv a c
   $ hg mv p s
   $ hg ci -Am 'i-1: a -move-> c, p -move-> s'
   $ hg mv c d
   $ hg mv s t
   $ hg ci -Am 'i-2: c -move-> d, s -move-> t'
   $ hg log -G
   @  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 And having another branch with renames on the other side
 
   $ hg mv d e
   $ hg ci -Am 'a-1: d -move-> e'
   $ hg mv e f
   $ hg ci -Am 'a-2: e -move-> f'
   $ hg log -G --rev '::.'
   @  a-2: e -move-> f
   |
   o  a-1: d -move-> e
   |
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 Have a branching with nothing on one side
 
   $ hg up 'desc("i-2")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved
   $ echo foo > b
   $ hg ci -m 'b-1: b update'
   created new head
   $ hg log -G --rev '::.'
   @  b-1: b update
   |
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 Create a branch that delete a file previous renamed
 
   $ hg up 'desc("i-2")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg rm d
   $ hg ci -m 'c-1 delete d'
   created new head
   $ hg log -G --rev '::.'
   @  c-1 delete d
   |
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 Create a branch that delete a file previous renamed and recreate it
 
   $ hg up 'desc("i-2")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg rm d
   $ hg ci -m 'd-1 delete d'
   created new head
   $ echo bar > d
   $ hg add d
   $ hg ci -m 'd-2 re-add d'
   $ hg log -G --rev '::.'
   @  d-2 re-add d
   |
   o  d-1 delete d
   |
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 Having another branch renaming a different file to the same filename as another
 
   $ hg up 'desc("i-2")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg mv b g
   $ hg ci -m 'e-1 b -move-> g'
   created new head
   $ hg mv g f
   $ hg ci -m 'e-2 g -move-> f'
   $ hg log -G --rev '::.'
   @  e-2 g -move-> f
   |
   o  e-1 b -move-> g
   |
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
   $ hg up -q null
 
 Having a branch similar to the 'a' one, but moving the 'p' file around.
 
   $ hg up 'desc("i-2")'
   6 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg mv t u
   $ hg ci -Am 'p-1: t -move-> u'
   created new head
   $ hg mv u v
   $ hg ci -Am 'p-2: u -move-> v'
   $ hg log -G --rev '::.'
   @  p-2: u -move-> v
   |
   o  p-1: t -move-> u
   |
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
   $ hg up -q null
 
 Having another branch renaming a different file to the same filename as another
 
   $ hg up 'desc("i-2")'
   6 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg mv r w
   $ hg ci -m 'q-1 r -move-> w'
   created new head
   $ hg mv w v
   $ hg ci -m 'q-2 w -move-> v'
   $ hg log -G --rev '::.'
   @  q-2 w -move-> v
   |
   o  q-1 r -move-> w
   |
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
   $ hg up -q null
 
 Setup all merge
 ===============
 
 This is done beforehand to validate that the upgrade process creates valid copy
 information.
 
 merging with unrelated change does not interfere with the renames
 ---------------------------------------------------------------
 
 - rename on one side
 - unrelated change on the other side
 
   $ case_desc="simple merge - A side: multiple renames, B side: unrelated update"
 
   $ hg up 'desc("b-1")'
   6 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg merge 'desc("a-2")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mBAm-0 $case_desc - one way"
   $ hg up 'desc("a-2")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg merge 'desc("b-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mABm-0 $case_desc - the other way"
   created new head
   $ hg log -G --rev '::(desc("mABm")+desc("mBAm"))'
   @    mABm-0 simple merge - A side: multiple renames, B side: unrelated update - the other way
   |\
   +---o  mBAm-0 simple merge - A side: multiple renames, B side: unrelated update - one way
   | |/
   | o  b-1: b update
   | |
   o |  a-2: e -move-> f
   | |
   o |  a-1: d -move-> e
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 
 merging with the side having a delete
 -------------------------------------
 
 case summary:
 - one with change to an unrelated file
 - one deleting the change
 and recreate an unrelated file after the merge
 
   $ case_desc="simple merge - C side: delete a file with copies history , B side: unrelated update"
 
   $ hg up 'desc("b-1")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved
   $ hg merge 'desc("c-1")'
   0 files updated, 0 files merged, 1 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mBCm-0 $case_desc - one way"
   $ echo bar > d
   $ hg add d
   $ hg ci -m 'mBCm-1 re-add d'
   $ hg up 'desc("c-1")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved
   $ hg merge 'desc("b-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mCBm-0 $case_desc - the other way"
   created new head
   $ echo bar > d
   $ hg add d
   $ hg ci -m 'mCBm-1 re-add d'
   $ hg log -G --rev '::(desc("mCBm")+desc("mBCm"))'
   @  mCBm-1 re-add d
   |
   o    mCBm-0 simple merge - C side: delete a file with copies history , B side: unrelated update - the other way
   |\
   | | o  mBCm-1 re-add d
   | | |
   +---o  mBCm-0 simple merge - C side: delete a file with copies history , B side: unrelated update - one way
   | |/
   | o  c-1 delete d
   | |
   o |  b-1: b update
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 Comparing with a merge re-adding the file afterward
 ---------------------------------------------------
 
 Merge:
 - one with change to an unrelated file
 - one deleting and recreating the change
 
   $ case_desc="simple merge - B side: unrelated update, D side: delete and recreate a file (with different content)"
 
   $ hg up 'desc("b-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg merge 'desc("d-2")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mBDm-0 $case_desc - one way"
   $ hg up 'desc("d-2")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg merge 'desc("b-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mDBm-0 $case_desc - the other way"
   created new head
   $ hg log -G --rev '::(desc("mDBm")+desc("mBDm"))'
   @    mDBm-0 simple merge - B side: unrelated update, D side: delete and recreate a file (with different content) - the other way
   |\
   +---o  mBDm-0 simple merge - B side: unrelated update, D side: delete and recreate a file (with different content) - one way
   | |/
   | o  d-2 re-add d
   | |
   | o  d-1 delete d
   | |
   o |  b-1: b update
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 
 Comparing with a merge with colliding rename
 --------------------------------------------
 
 Subcase: new copy information on both side
 ``````````````````````````````````````````
 
 - the "e-" branch renaming b to f (through 'g')
 - the "a-" branch renaming d to f (through e)
 
   $ case_desc="merge with copies info on both side - A side: rename d to f, E side: b to f, (same content for f)"
 
   $ hg up 'desc("a-2")'
   2 files updated, 0 files merged, 1 files removed, 0 files unresolved
   $ hg merge 'desc("e-2")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved (no-changeset !)
   0 files updated, 0 files merged, 1 files removed, 0 files unresolved (changeset !)
   (branch merge, don't forget to commit)
   $ hg ci -m "mAEm-0 $case_desc - one way"
   $ hg up 'desc("e-2")'
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved (no-changeset !)
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved (changeset !)
   $ hg merge 'desc("a-2")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved (no-changeset !)
   0 files updated, 0 files merged, 1 files removed, 0 files unresolved (changeset !)
   (branch merge, don't forget to commit)
   $ hg ci -m "mEAm-0 $case_desc - the other way"
   created new head
   $ hg log -G --rev '::(desc("mAEm")+desc("mEAm"))'
   @    mEAm-0 merge with copies info on both side - A side: rename d to f, E side: b to f, (same content for f) - the other way
   |\
   +---o  mAEm-0 merge with copies info on both side - A side: rename d to f, E side: b to f, (same content for f) - one way
   | |/
   | o  e-2 g -move-> f
   | |
   | o  e-1 b -move-> g
   | |
   o |  a-2: e -move-> f
   | |
   o |  a-1: d -move-> e
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 Subcase: new copy information on both side with an actual merge happening
 `````````````````````````````````````````````````````````````````````````
 
 - the "p-" branch renaming 't' to 'v' (through 'u')
 - the "q-" branch renaming 'r' to 'v' (through 'w')
 
   $ case_desc="merge with copies info on both side - P side: rename t to v, Q side: r to v, (different content)"
 
   $ hg up 'desc("p-2")'
   3 files updated, 0 files merged, 2 files removed, 0 files unresolved
   $ hg merge 'desc("q-2")' --tool ':union'
   merging v
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mPQm-0 $case_desc - one way"
   $ hg up 'desc("q-2")'
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg merge 'desc("p-2")' --tool ':union'
   merging v
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mQPm-0 $case_desc - the other way"
   created new head
   $ hg log -G --rev '::(desc("mAEm")+desc("mEAm"))'
   o    mEAm-0 merge with copies info on both side - A side: rename d to f, E side: b to f, (same content for f) - the other way
   |\
   +---o  mAEm-0 merge with copies info on both side - A side: rename d to f, E side: b to f, (same content for f) - one way
   | |/
   | o  e-2 g -move-> f
   | |
   | o  e-1 b -move-> g
   | |
   o |  a-2: e -move-> f
   | |
   o |  a-1: d -move-> e
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 Subcase: existing copy information overwritten on one branch
 ````````````````````````````````````````````````````````````
 
 Merge:
 - one with change to an unrelated file (b)
 - one overwriting a file (d) with a rename (from h to i to d)
 
   $ case_desc="simple merge - B side: unrelated change, F side: overwrite d with a copy (from h->i->d)"
 
   $ hg up 'desc("i-2")'
   2 files updated, 0 files merged, 1 files removed, 0 files unresolved
   $ hg mv h i
   $ hg commit -m "f-1: rename h -> i"
   created new head
   $ hg mv --force i d
   $ hg commit -m "f-2: rename i -> d"
   $ hg debugindex d | "$PYTHON" ../no-linkrev
      rev linkrev nodeid       p1           p2
        0       * d8252ab2e760 000000000000 000000000000 (no-changeset !)
        0       * ae258f702dfe 000000000000 000000000000 (changeset !)
        1       * b004912a8510 000000000000 000000000000
        2       * 7b79e2fe0c89 000000000000 000000000000 (no-changeset !)
   $ hg up 'desc("b-1")'
   3 files updated, 0 files merged, 0 files removed, 0 files unresolved (no-changeset !)
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved (changeset !)
   $ hg merge 'desc("f-2")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved (no-changeset !)
   0 files updated, 0 files merged, 1 files removed, 0 files unresolved (changeset !)
   (branch merge, don't forget to commit)
   $ hg ci -m "mBFm-0 $case_desc - one way"
   $ hg up 'desc("f-2")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg merge 'desc("b-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mFBm-0 $case_desc - the other way"
   created new head
   $ hg up null --quiet
   $ hg log -G --rev '::(desc("mBFm")+desc("mFBm"))'
   o    mFBm-0 simple merge - B side: unrelated change, F side: overwrite d with a copy (from h->i->d) - the other way
   |\
   +---o  mBFm-0 simple merge - B side: unrelated change, F side: overwrite d with a copy (from h->i->d) - one way
   | |/
   | o  f-2: rename i -> d
   | |
   | o  f-1: rename h -> i
   | |
   o |  b-1: b update
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 Subcase: existing copy information overwritten on one branch, with different content)
 `````````````````````````````````````````````````````````````````````````````````````
 
 Merge:
 - one with change to an unrelated file (b)
 - one overwriting a file (t) with a rename (from r to x to t), v content is not the same as on the other branch
 
   $ case_desc="simple merge - B side: unrelated change, R side: overwrite d with a copy (from r->x->t) different content"
 
   $ hg up 'desc("i-2")'
   6 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg mv r x
   $ hg commit -m "r-1: rename r -> x"
   created new head
   $ hg mv --force x t
   $ hg commit -m "r-2: rename t -> x"
   $ hg debugindex t | "$PYTHON" ../no-linkrev
      rev linkrev nodeid       p1           p2
        0       * d74efbf65309 000000000000 000000000000 (no-changeset !)
        1       * 02a930b9d7ad 000000000000 000000000000 (no-changeset !)
        0       * 5aed6a8dbff0 000000000000 000000000000 (changeset !)
        1       * a38b2fa17021 000000000000 000000000000 (changeset !)
   $ hg up 'desc("b-1")'
   3 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg merge 'desc("r-2")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mBRm-0 $case_desc - one way"
   $ hg up 'desc("r-2")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg merge 'desc("b-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mRBm-0 $case_desc - the other way"
   created new head
   $ hg up null --quiet
   $ hg log -G --rev '::(desc("mBRm")+desc("mRBm"))'
   o    mRBm-0 simple merge - B side: unrelated change, R side: overwrite d with a copy (from r->x->t) different content - the other way
   |\
   +---o  mBRm-0 simple merge - B side: unrelated change, R side: overwrite d with a copy (from r->x->t) different content - one way
   | |/
   | o  r-2: rename t -> x
   | |
   | o  r-1: rename r -> x
   | |
   o |  b-1: b update
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 
 Subcase: reset of the copy history on one side
 ``````````````````````````````````````````````
 
 Merge:
 - one with change to a file
 - one deleting and recreating the file
 
 Unlike in the 'BD/DB' cases, an actual merge happened here. So we should
 consider history and rename on both branch of the merge.
 
   $ case_desc="actual content merge, copies on one side - D side: delete and re-add (different content), G side: update content"
 
   $ hg up 'desc("i-2")'
   6 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ echo "some update" >> d
   $ hg commit -m "g-1: update d"
   created new head
   $ hg up 'desc("d-2")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg merge 'desc("g-1")' --tool :union
   merging d
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mDGm-0 $case_desc - one way"
   $ hg up 'desc("g-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg merge 'desc("d-2")' --tool :union
   merging d
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mGDm-0 $case_desc - the other way"
   created new head
   $ hg log -G --rev '::(desc("mDGm")+desc("mGDm"))'
   @    mGDm-0 actual content merge, copies on one side - D side: delete and re-add (different content), G side: update content - the other way
   |\
   +---o  mDGm-0 actual content merge, copies on one side - D side: delete and re-add (different content), G side: update content - one way
   | |/
   | o  g-1: update d
   | |
   o |  d-2 re-add d
   | |
   o |  d-1 delete d
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 Subcase: merging a change to a file with a "copy overwrite" to that file from another branch
 ````````````````````````````````````````````````````````````````````````````````````````````
 
 Merge:
 - one with change to a file (d)
 - one overwriting that file with a rename (from h to i, to d)
 
 This case is similar to BF/FB, but an actual merge happens, so both side of the
 history are relevant.
 
 Note:
 | In this case, the merge get conflicting information since on one side we have
 | "a -> c -> d". and one the other one we have "h -> i -> d".
 |
 | The current code arbitrarily pick one side
 
   $ case_desc="merge - G side: content change, F side: copy overwrite, no content change"
 
   $ hg up 'desc("f-2")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved
   $ hg merge 'desc("g-1")' --tool :union
   merging d (no-changeset !)
   0 files updated, 1 files merged, 0 files removed, 0 files unresolved (no-changeset !)
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved (changeset !)
   (branch merge, don't forget to commit)
   $ hg ci -m "mFGm-0 $case_desc - one way"
   created new head
   $ hg up 'desc("g-1")'
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved (no-changeset !)
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved (changeset !)
   $ hg merge 'desc("f-2")' --tool :union
   merging d (no-changeset !)
   0 files updated, 1 files merged, 1 files removed, 0 files unresolved (no-changeset !)
   0 files updated, 0 files merged, 1 files removed, 0 files unresolved (changeset !)
   (branch merge, don't forget to commit)
   $ hg ci -m "mGFm-0 $case_desc - the other way"
   created new head
   $ hg log -G --rev '::(desc("mGFm")+desc("mFGm"))'
   @    mGFm-0 merge - G side: content change, F side: copy overwrite, no content change - the other way
   |\
   +---o  mFGm-0 merge - G side: content change, F side: copy overwrite, no content change - one way
   | |/
   | o  g-1: update d
   | |
   o |  f-2: rename i -> d
   | |
   o |  f-1: rename h -> i
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 
 Comparing with merging with a deletion (and keeping the file)
 -------------------------------------------------------------
 
 Merge:
 - one removing a file (d)
 - one updating that file
 - the merge keep the modified version of the file (canceling the delete)
 
 In this case, the file keep on living after the merge. So we should not drop its
 copy tracing chain.
 
   $ case_desc="merge updated/deleted - revive the file (updated content)"
 
   $ hg up 'desc("c-1")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved
   $ hg merge 'desc("g-1")'
   file 'd' was deleted in local [working copy] but was modified in other [merge rev].
   You can use (c)hanged version, leave (d)eleted, or leave (u)nresolved.
   What do you want to do? u
   0 files updated, 0 files merged, 0 files removed, 1 files unresolved
   use 'hg resolve' to retry unresolved file merges or 'hg merge --abort' to abandon
   [1]
   $ hg resolve -t :other d
   (no more unresolved files)
   $ hg ci -m "mCGm-0 $case_desc - one way"
   created new head
 
   $ hg up 'desc("g-1")'
   0 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg merge 'desc("c-1")'
   file 'd' was deleted in other [merge rev] but was modified in local [working copy].
   You can use (c)hanged version, (d)elete, or leave (u)nresolved.
   What do you want to do? u
   0 files updated, 0 files merged, 0 files removed, 1 files unresolved
   use 'hg resolve' to retry unresolved file merges or 'hg merge --abort' to abandon
   [1]
   $ hg resolve -t :local d
   (no more unresolved files)
   $ hg ci -m "mGCm-0 $case_desc - the other way"
   created new head
 
   $ hg log -G --rev '::(desc("mCGm")+desc("mGCm"))'
   @    mGCm-0 merge updated/deleted - revive the file (updated content) - the other way
   |\
   +---o  mCGm-0 merge updated/deleted - revive the file (updated content) - one way
   | |/
   | o  g-1: update d
   | |
   o |  c-1 delete d
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 
 
 Comparing with merge restoring an untouched deleted file
 --------------------------------------------------------
 
 Merge:
 - one removing a file (d)
 - one leaving the file untouched
 - the merge actively restore the file to the same content.
 
 In this case, the file keep on living after the merge. So we should not drop its
 copy tracing chain.
 
   $ case_desc="merge explicitely revive deleted file - B side: unrelated change, C side: delete d (restored by merge)"
 
   $ hg up 'desc("c-1")'
   0 files updated, 0 files merged, 1 files removed, 0 files unresolved
   $ hg merge 'desc("b-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg revert --rev 'desc("b-1")' d
   $ hg ci -m "mCB-revert-m-0 $case_desc - one way"
   created new head
 
   $ hg up 'desc("b-1")'
   0 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg merge 'desc("c-1")'
   0 files updated, 0 files merged, 1 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg revert --rev 'desc("b-1")' d
   $ hg ci -m "mBC-revert-m-0 $case_desc - the other way"
   created new head
 
   $ hg log -G --rev '::(desc("mCB-revert-m")+desc("mBC-revert-m"))'
   @    mBC-revert-m-0 merge explicitely revive deleted file - B side: unrelated change, C side: delete d (restored by merge) - the other way
   |\
   +---o  mCB-revert-m-0 merge explicitely revive deleted file - B side: unrelated change, C side: delete d (restored by merge) - one way
   | |/
   | o  c-1 delete d
   | |
   o |  b-1: b update
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 
   $ hg up null --quiet
 
 Merging a branch where a rename was deleted with a branch where the same file was renamed
 ------------------------------------------------------------------------------------------
 
 Create a "conflicting" merge where `d` get removed on one branch before its
 rename information actually conflict with the other branch.
 
 (the copy information from the branch that was not deleted should win).
 
   $ case_desc="simple merge - C side: d is the results of renames then deleted, H side: d is result of another rename (same content as the other branch)"
 
   $ hg up 'desc("i-0")'
   6 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg mv b d
   $ hg ci -m "h-1: b -(move)-> d"
   created new head
 
   $ hg up 'desc("c-1")'
   2 files updated, 0 files merged, 3 files removed, 0 files unresolved
   $ hg merge 'desc("h-1")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mCH-delete-before-conflict-m-0 $case_desc - one way"
 
   $ hg up 'desc("h-1")'
   2 files updated, 0 files merged, 1 files removed, 0 files unresolved
   $ hg merge 'desc("c-1")'
   1 files updated, 0 files merged, 2 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mHC-delete-before-conflict-m-0 $case_desc - the other way"
   created new head
   $ hg log -G --rev '::(desc("mCH-delete-before-conflict-m")+desc("mHC-delete-before-conflict-m"))'
   @    mHC-delete-before-conflict-m-0 simple merge - C side: d is the results of renames then deleted, H side: d is result of another rename (same content as the other branch) - the other way
   |\
   +---o  mCH-delete-before-conflict-m-0 simple merge - C side: d is the results of renames then deleted, H side: d is result of another rename (same content as the other branch) - one way
   | |/
   | o  h-1: b -(move)-> d
   | |
   o |  c-1 delete d
   | |
   o |  i-2: c -move-> d, s -move-> t
   | |
   o |  i-1: a -move-> c, p -move-> s
   |/
   o  i-0 initial commit: a b h p q r
   
 
 Variant of previous with extra changes introduced by the merge
 --------------------------------------------------------------
 
 Multiple cases above explicitely test cases where content are the same on both side during merge. In this section we will introduce variants for theses cases where new change are introduced to these file content during the merges.
 
 
 Subcase: merge has same initial content on both side, but merge introduced a change
 ```````````````````````````````````````````````````````````````````````````````````
 
 Same as `mAEm` and `mEAm` but with extra change to the file before commiting
 
 - the "e-" branch renaming b to f (through 'g')
 - the "a-" branch renaming d to f (through e)
 
   $ case_desc="merge with file update and copies info on both side - A side: rename d to f, E side: b to f, (same content for f in parent)"
 
   $ hg up 'desc("a-2")'
   2 files updated, 0 files merged, 1 files removed, 0 files unresolved
   $ hg merge 'desc("e-2")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved (no-changeset !)
   0 files updated, 0 files merged, 1 files removed, 0 files unresolved (changeset !)
   (branch merge, don't forget to commit)
   $ echo "content change for mAE-change-m" > f
   $ hg ci -m "mAE-change-m-0 $case_desc - one way"
   created new head
   $ hg up 'desc("e-2")'
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg merge 'desc("a-2")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved (no-changeset !)
   0 files updated, 0 files merged, 1 files removed, 0 files unresolved (changeset !)
   (branch merge, don't forget to commit)
   $ echo "content change for mEA-change-m" > f
   $ hg ci -m "mEA-change-m-0 $case_desc - the other way"
   created new head
   $ hg log -G --rev '::(desc("mAE-change-m")+desc("mEA-change-m"))'
   @    mEA-change-m-0 merge with file update and copies info on both side - A side: rename d to f, E side: b to f, (same content for f in parent) - the other way
   |\
   +---o  mAE-change-m-0 merge with file update and copies info on both side - A side: rename d to f, E side: b to f, (same content for f in parent) - one way
   | |/
   | o  e-2 g -move-> f
   | |
   | o  e-1 b -move-> g
   | |
   o |  a-2: e -move-> f
   | |
   o |  a-1: d -move-> e
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 Subcase: merge overwrite common copy information, but with extra change during the merge
 ````````````````````````````````````````````````````````````````````````````````````````
 
 Merge:
 - one with change to an unrelated file (b)
 - one overwriting a file (d) with a rename (from h to i to d)
 - the merge update f content
 
   $ case_desc="merge with extra change - B side: unrelated change, F side: overwrite d with a copy (from h->i->d)"
 
   $ hg up 'desc("f-2")'
   2 files updated, 0 files merged, 2 files removed, 0 files unresolved
 #if no-changeset
   $ hg debugindex d | "$PYTHON" ../no-linkrev
      rev linkrev nodeid       p1           p2
        0       * d8252ab2e760 000000000000 000000000000
        1       * b004912a8510 000000000000 000000000000
        2       * 7b79e2fe0c89 000000000000 000000000000
        3       * 17ec97e60577 d8252ab2e760 000000000000
        4       * 06dabf50734c b004912a8510 17ec97e60577
        5       * 19c0e3924691 17ec97e60577 b004912a8510
        6       * 89c873a01d97 7b79e2fe0c89 17ec97e60577
        7       * d55cb4e9ef57 000000000000 000000000000
 #else
   $ hg debugindex d | "$PYTHON" ../no-linkrev
      rev linkrev nodeid       p1           p2
        0       * ae258f702dfe 000000000000 000000000000
        1       * b004912a8510 000000000000 000000000000
        2       * 5cce88bf349f ae258f702dfe 000000000000
        3       * cc269dd788c8 b004912a8510 5cce88bf349f
        4       * 51c91a115080 5cce88bf349f b004912a8510
 #endif
   $ hg up 'desc("b-1")'
   3 files updated, 0 files merged, 0 files removed, 0 files unresolved (no-changeset !)
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved (changeset !)
   $ hg merge 'desc("f-2")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved (no-changeset !)
   0 files updated, 0 files merged, 1 files removed, 0 files unresolved (changeset !)
   (branch merge, don't forget to commit)
   $ echo "extra-change to (formelly h) during the merge" > d
   $ hg ci -m "mBF-change-m-0 $case_desc - one way"
   created new head
   $ hg manifest --rev . --debug | grep "  d"
   1c334238bd42ec85c6a0d83fd1b2a898a6a3215d 644   d (no-changeset !)
   cea2d99c0fde64672ef61953786fdff34f16e230 644   d (changeset !)
 
   $ hg up 'desc("f-2")'
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg merge 'desc("b-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ echo "extra-change to (formelly h) during the merge" > d
   $ hg ci -m "mFB-change-m-0 $case_desc - the other way"
   created new head
   $ hg manifest --rev . --debug | grep "  d"
   1c334238bd42ec85c6a0d83fd1b2a898a6a3215d 644   d (no-changeset !)
   cea2d99c0fde64672ef61953786fdff34f16e230 644   d (changeset !)
 #if no-changeset
   $ hg debugindex d | "$PYTHON" ../no-linkrev
      rev linkrev nodeid       p1           p2
        0       * d8252ab2e760 000000000000 000000000000
        1       * b004912a8510 000000000000 000000000000
        2       * 7b79e2fe0c89 000000000000 000000000000
        3       * 17ec97e60577 d8252ab2e760 000000000000
        4       * 06dabf50734c b004912a8510 17ec97e60577
        5       * 19c0e3924691 17ec97e60577 b004912a8510
        6       * 89c873a01d97 7b79e2fe0c89 17ec97e60577
        7       * d55cb4e9ef57 000000000000 000000000000
        8       * 1c334238bd42 7b79e2fe0c89 000000000000
 #else
   $ hg debugindex d | "$PYTHON" ../no-linkrev
      rev linkrev nodeid       p1           p2
        0       * ae258f702dfe 000000000000 000000000000
        1       * b004912a8510 000000000000 000000000000
        2       * 5cce88bf349f ae258f702dfe 000000000000
        3       * cc269dd788c8 b004912a8510 5cce88bf349f
        4       * 51c91a115080 5cce88bf349f b004912a8510
        5       * cea2d99c0fde ae258f702dfe 000000000000
 #endif
   $ hg log -G --rev '::(desc("mBF-change-m")+desc("mFB-change-m"))'
   @    mFB-change-m-0 merge with extra change - B side: unrelated change, F side: overwrite d with a copy (from h->i->d) - the other way
   |\
   +---o  mBF-change-m-0 merge with extra change - B side: unrelated change, F side: overwrite d with a copy (from h->i->d) - one way
   | |/
   | o  f-2: rename i -> d
   | |
   | o  f-1: rename h -> i
   | |
   o |  b-1: b update
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 Subcase: restoring and untouched deleted file, while touching it
 ````````````````````````````````````````````````````````````````
 
 Merge:
 - one removing a file (d)
 - one leaving the file untouched
 - the merge actively restore the file to the same content.
 
 In this case, the file keep on living after the merge. So we should not drop its
 copy tracing chain.
 
   $ case_desc="merge explicitely revive deleted file - B side: unrelated change, C side: delete d (restored by merge)"
 
   $ hg up 'desc("c-1")'
   2 files updated, 0 files merged, 1 files removed, 0 files unresolved
   $ hg merge 'desc("b-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg revert --rev 'desc("b-1")' d
   $ echo "new content for d after the revert" > d
   $ hg ci -m "mCB-change-m-0 $case_desc - one way"
   created new head
   $ hg manifest --rev . --debug | grep "  d"
   e333780c17752a3b0dd15e3ad48aa4e5c745f621 644   d (no-changeset !)
   4b540a18ad699234b2b2aa18cb69555ac9c4b1df 644   d (changeset !)
 
   $ hg up 'desc("b-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg merge 'desc("c-1")'
   0 files updated, 0 files merged, 1 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg revert --rev 'desc("b-1")' d
   $ echo "new content for d after the revert" > d
   $ hg ci -m "mBC-change-m-0 $case_desc - the other way"
   created new head
   $ hg manifest --rev . --debug | grep "  d"
   e333780c17752a3b0dd15e3ad48aa4e5c745f621 644   d (no-changeset !)
   4b540a18ad699234b2b2aa18cb69555ac9c4b1df 644   d (changeset !)
 
 
   $ hg up null --quiet
   $ hg log -G --rev '::(desc("mCB-change-m")+desc("mBC-change-m"))'
   o    mBC-change-m-0 merge explicitely revive deleted file - B side: unrelated change, C side: delete d (restored by merge) - the other way
   |\
   +---o  mCB-change-m-0 merge explicitely revive deleted file - B side: unrelated change, C side: delete d (restored by merge) - one way
   | |/
   | o  c-1 delete d
   | |
   o |  b-1: b update
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 Decision from previous merge are properly chained with later merge
 ------------------------------------------------------------------
 
 Subcase: chaining conflicting rename resolution
 ```````````````````````````````````````````````
 
 The "mAEm" and "mEAm" case create a rename tracking conflict on file 'f'. We
 add more change on the respective branch and merge again. These second merge
 does not involve the file 'f' and the arbitration done within "mAEm" and "mEA"
 about that file should stay unchanged.
 
 We also touch J during some of the merge to check for unrelated change to new file during merge.
 
   $ case_desc="chained merges (conflict -> simple) - same content everywhere"
 
 (extra unrelated changes)
 
   $ hg up 'desc("a-2")'
   6 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ echo j > unrelated-j
   $ hg add unrelated-j
   $ hg ci -m 'j-1: unrelated changes (based on the "a" series of changes)'
   created new head
 
   $ hg up 'desc("e-2")'
   2 files updated, 0 files merged, 2 files removed, 0 files unresolved (no-changeset !)
   1 files updated, 0 files merged, 2 files removed, 0 files unresolved (changeset !)
   $ echo k > unrelated-k
   $ hg add unrelated-k
   $ hg ci -m 'k-1: unrelated changes (based on "e" changes)'
   created new head
 
 (merge variant 1)
 
   $ hg up 'desc("mAEm")'
   1 files updated, 0 files merged, 2 files removed, 0 files unresolved (no-changeset !)
   0 files updated, 0 files merged, 2 files removed, 0 files unresolved (changeset !)
   $ hg merge 'desc("k-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mAE,Km: $case_desc"
 
 (merge variant 2)
 
   $ hg up 'desc("k-1")'
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved (no-changeset !)
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved (changeset !)
 
   $ hg merge 'desc("mAEm")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved (no-changeset !)
   0 files updated, 0 files merged, 1 files removed, 0 files unresolved (changeset !)
   (branch merge, don't forget to commit)
   $ hg ci -m "mK,AEm: $case_desc"
   created new head
 
 (merge variant 3)
 
   $ hg up 'desc("mEAm")'
   0 files updated, 0 files merged, 1 files removed, 0 files unresolved
   $ hg merge 'desc("j-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ echo jj > unrelated-j
   $ hg ci -m "mEA,Jm: $case_desc"
 
 (merge variant 4)
 
   $ hg up 'desc("j-1")'
   3 files updated, 0 files merged, 0 files removed, 0 files unresolved (no-changeset !)
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved (changeset !)
   $ hg merge 'desc("mEAm")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved (no-changeset !)
   0 files updated, 0 files merged, 1 files removed, 0 files unresolved (changeset !)
   (branch merge, don't forget to commit)
   $ echo jj > unrelated-j
   $ hg ci -m "mJ,EAm: $case_desc"
   created new head
 
 
   $ hg log -G --rev '::(desc("mAE,Km") + desc("mK,AEm") + desc("mEA,Jm") + desc("mJ,EAm"))'
   @    mJ,EAm: chained merges (conflict -> simple) - same content everywhere
   |\
   +---o  mEA,Jm: chained merges (conflict -> simple) - same content everywhere
   | |/
   | | o    mK,AEm: chained merges (conflict -> simple) - same content everywhere
   | | |\
   | | +---o  mAE,Km: chained merges (conflict -> simple) - same content everywhere
   | | | |/
   | | | o  k-1: unrelated changes (based on "e" changes)
   | | | |
   | o | |  j-1: unrelated changes (based on the "a" series of changes)
   | | | |
   o-----+  mEAm-0 merge with copies info on both side - A side: rename d to f, E side: b to f, (same content for f) - the other way
   |/ / /
   | o /  mAEm-0 merge with copies info on both side - A side: rename d to f, E side: b to f, (same content for f) - one way
   |/|/
   | o  e-2 g -move-> f
   | |
   | o  e-1 b -move-> g
   | |
   o |  a-2: e -move-> f
   | |
   o |  a-1: d -move-> e
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 Subcase: chaining conflicting rename resolution, with actual merging happening
 ``````````````````````````````````````````````````````````````````````````````
 
 The "mPQm" and "mQPm" case create a rename tracking conflict on file 't'. We
 add more change on the respective branch and merge again. These second merge
 does not involve the file 't' and the arbitration done within "mPQm" and "mQP"
 about that file should stay unchanged.
 
   $ case_desc="chained merges (conflict -> simple) - different content"
 
 (extra unrelated changes)
 
   $ hg up 'desc("p-2")'
   3 files updated, 0 files merged, 3 files removed, 0 files unresolved
   $ echo s > unrelated-s
   $ hg add unrelated-s
   $ hg ci -m 's-1: unrelated changes (based on the "p" series of changes)'
   created new head
 
   $ hg up 'desc("q-2")'
   2 files updated, 0 files merged, 2 files removed, 0 files unresolved
   $ echo t > unrelated-t
   $ hg add unrelated-t
   $ hg ci -m 't-1: unrelated changes (based on "q" changes)'
   created new head
 
 (merge variant 1)
 
   $ hg up 'desc("mPQm")'
   1 files updated, 0 files merged, 2 files removed, 0 files unresolved
   $ hg merge 'desc("t-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mPQ,Tm: $case_desc"
 
 (merge variant 2)
 
   $ hg up 'desc("t-1")'
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
 
   $ hg merge 'desc("mPQm")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mT,PQm: $case_desc"
   created new head
 
 (merge variant 3)
 
   $ hg up 'desc("mQPm")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved
   $ hg merge 'desc("s-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mQP,Sm: $case_desc"
 
 (merge variant 4)
 
   $ hg up 'desc("s-1")'
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg merge 'desc("mQPm")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mS,QPm: $case_desc"
   created new head
   $ hg up null --quiet
 
 
   $ hg log -G --rev '::(desc("mPQ,Tm") + desc("mT,PQm") + desc("mQP,Sm") + desc("mS,QPm"))'
   o    mS,QPm: chained merges (conflict -> simple) - different content
   |\
   +---o  mQP,Sm: chained merges (conflict -> simple) - different content
   | |/
   | | o    mT,PQm: chained merges (conflict -> simple) - different content
   | | |\
   | | +---o  mPQ,Tm: chained merges (conflict -> simple) - different content
   | | | |/
   | | | o  t-1: unrelated changes (based on "q" changes)
   | | | |
   | o | |  s-1: unrelated changes (based on the "p" series of changes)
   | | | |
   o-----+  mQPm-0 merge with copies info on both side - P side: rename t to v, Q side: r to v, (different content) - the other way
   |/ / /
   | o /  mPQm-0 merge with copies info on both side - P side: rename t to v, Q side: r to v, (different content) - one way
   |/|/
   | o  q-2 w -move-> v
   | |
   | o  q-1 r -move-> w
   | |
   o |  p-2: u -move-> v
   | |
   o |  p-1: t -move-> u
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 Subcase: chaining salvage information during a merge
 ````````````````````````````````````````````````````
 
 We add more change on the branch were the file was deleted. merging again
 should preserve the fact eh file was salvaged.
 
   $ case_desc="chained merges (salvaged -> simple) - same content (when the file exists)"
 
 (creating the change)
 
   $ hg up 'desc("c-1")'
   5 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ echo l > unrelated-l
   $ hg add unrelated-l
   $ hg ci -m 'l-1: unrelated changes (based on "c" changes)'
   created new head
 
 (Merge variant 1)
 
   $ hg up 'desc("mBC-revert-m")'
   2 files updated, 0 files merged, 1 files removed, 0 files unresolved
   $ hg merge 'desc("l-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mBC+revert,Lm: $case_desc"
 
 (Merge variant 2)
 
   $ hg up 'desc("mCB-revert-m")'
   0 files updated, 0 files merged, 1 files removed, 0 files unresolved
   $ hg merge 'desc("l-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mCB+revert,Lm: $case_desc"
 
 (Merge variant 3)
 
   $ hg up 'desc("l-1")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved
 
   $ hg merge 'desc("mBC-revert-m")'
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mL,BC+revertm: $case_desc"
   created new head
 
 (Merge variant 4)
 
   $ hg up 'desc("l-1")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved
 
   $ hg merge 'desc("mCB-revert-m")'
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mL,CB+revertm: $case_desc"
   created new head
 
   $ hg log -G --rev '::(desc("mBC+revert,Lm") + desc("mCB+revert,Lm") + desc("mL,BC+revertm") + desc("mL,CB+revertm"))'
   @    mL,CB+revertm: chained merges (salvaged -> simple) - same content (when the file exists)
   |\
   | | o  mL,BC+revertm: chained merges (salvaged -> simple) - same content (when the file exists)
   | |/|
   +-+---o  mCB+revert,Lm: chained merges (salvaged -> simple) - same content (when the file exists)
   | | |
   | +---o  mBC+revert,Lm: chained merges (salvaged -> simple) - same content (when the file exists)
   | | |/
   | o |  l-1: unrelated changes (based on "c" changes)
   | | |
   | | o  mBC-revert-m-0 merge explicitely revive deleted file - B side: unrelated change, C side: delete d (restored by merge) - the other way
   | |/|
   o---+  mCB-revert-m-0 merge explicitely revive deleted file - B side: unrelated change, C side: delete d (restored by merge) - one way
   |/ /
   o |  c-1 delete d
   | |
   | o  b-1: b update
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 
 Subcase: chaining "merged" information during a merge
 ``````````````````````````````````````````````````````
 
 When a non-rename change are merged with a copy overwrite, the merge pick the copy source from (p1) as the reference. We should preserve this information in subsequent merges.
 
   $ case_desc="chained merges (copy-overwrite -> simple) - same content"
 
 (extra unrelated changes)
 
   $ hg up 'desc("f-2")'
   2 files updated, 0 files merged, 2 files removed, 0 files unresolved (no-changeset !)
   1 files updated, 0 files merged, 2 files removed, 0 files unresolved (changeset !)
   $ echo n > unrelated-n
   $ hg add unrelated-n
   $ hg ci -m 'n-1: unrelated changes (based on the "f" series of changes)'
   created new head
 
   $ hg up 'desc("g-1")'
   2 files updated, 0 files merged, 1 files removed, 0 files unresolved
   $ echo o > unrelated-o
   $ hg add unrelated-o
   $ hg ci -m 'o-1: unrelated changes (based on "g" changes)'
   created new head
 
 (merge variant 1)
 
   $ hg up 'desc("mFGm")'
   1 files updated, 0 files merged, 2 files removed, 0 files unresolved (no-changeset !)
   0 files updated, 0 files merged, 2 files removed, 0 files unresolved (changeset !)
   $ hg merge 'desc("o-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mFG,Om: $case_desc"
 
 (merge variant 2)
 
   $ hg up 'desc("o-1")'
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved (no-changeset !)
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved (changeset !)
   $ hg merge 'desc("FGm")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved (no-changeset !)
   0 files updated, 0 files merged, 1 files removed, 0 files unresolved (changeset !)
   (branch merge, don't forget to commit)
   $ hg ci -m "mO,FGm: $case_desc"
   created new head
 
 (merge variant 3)
 
   $ hg up 'desc("mGFm")'
   0 files updated, 0 files merged, 1 files removed, 0 files unresolved
   $ hg merge 'desc("n-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mGF,Nm: $case_desc"
 
 (merge variant 4)
 
   $ hg up 'desc("n-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg merge 'desc("mGFm")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mN,GFm: $case_desc"
   created new head
 
   $ hg log -G --rev '::(desc("mFG,Om") + desc("mO,FGm") + desc("mGF,Nm") + desc("mN,GFm"))'
   @    mN,GFm: chained merges (copy-overwrite -> simple) - same content
   |\
   +---o  mGF,Nm: chained merges (copy-overwrite -> simple) - same content
   | |/
   | | o    mO,FGm: chained merges (copy-overwrite -> simple) - same content
   | | |\
   | | +---o  mFG,Om: chained merges (copy-overwrite -> simple) - same content
   | | | |/
   | | | o  o-1: unrelated changes (based on "g" changes)
   | | | |
   | o | |  n-1: unrelated changes (based on the "f" series of changes)
   | | | |
   o-----+  mGFm-0 merge - G side: content change, F side: copy overwrite, no content change - the other way
   |/ / /
   | o /  mFGm-0 merge - G side: content change, F side: copy overwrite, no content change - one way
   |/|/
   | o  g-1: update d
   | |
   o |  f-2: rename i -> d
   | |
   o |  f-1: rename h -> i
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 Subcase: chaining conflicting rename resolution, with extra change during the merge
 ```````````````````````````````````````````````````````````````````````````````````
 
 The "mEA-change-m-0" and "mAE-change-m-0" case create a rename tracking conflict on file 'f'. We
 add more change on the respective branch and merge again. These second merge
 does not involve the file 'f' and the arbitration done within "mAEm" and "mEA"
 about that file should stay unchanged.
 
   $ case_desc="chained merges (conflict+change -> simple) - same content on both branch in the initial merge"
 
 
 (merge variant 1)
 
   $ hg up 'desc("mAE-change-m")'
   2 files updated, 0 files merged, 3 files removed, 0 files unresolved
   $ hg merge 'desc("k-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mAE-change,Km: $case_desc"
 
 (merge variant 2)
 
   $ hg up 'desc("k-1")'
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
 
   $ hg merge 'desc("mAE-change-m")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mK,AE-change-m: $case_desc"
   created new head
 
 (merge variant 3)
 
   $ hg up 'desc("mEA-change-m")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved
   $ hg merge 'desc("j-1")'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mEA-change,Jm: $case_desc"
 
 (merge variant 4)
 
   $ hg up 'desc("j-1")'
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg merge 'desc("mEA-change-m")'
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m "mJ,EA-change-m: $case_desc"
   created new head
 
 
   $ hg log -G --rev '::(desc("mAE-change,Km") + desc("mK,AE-change-m") + desc("mEA-change,Jm") + desc("mJ,EA-change-m"))'
   @    mJ,EA-change-m: chained merges (conflict+change -> simple) - same content on both branch in the initial merge
   |\
   +---o  mEA-change,Jm: chained merges (conflict+change -> simple) - same content on both branch in the initial merge
   | |/
   | | o    mK,AE-change-m: chained merges (conflict+change -> simple) - same content on both branch in the initial merge
   | | |\
   | | +---o  mAE-change,Km: chained merges (conflict+change -> simple) - same content on both branch in the initial merge
   | | | |/
   | | | o  k-1: unrelated changes (based on "e" changes)
   | | | |
   | o | |  j-1: unrelated changes (based on the "a" series of changes)
   | | | |
   o-----+  mEA-change-m-0 merge with file update and copies info on both side - A side: rename d to f, E side: b to f, (same content for f in parent) - the other way
   |/ / /
   | o /  mAE-change-m-0 merge with file update and copies info on both side - A side: rename d to f, E side: b to f, (same content for f in parent) - one way
   |/|/
   | o  e-2 g -move-> f
   | |
   | o  e-1 b -move-> g
   | |
   o |  a-2: e -move-> f
   | |
   o |  a-1: d -move-> e
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 Summary of all created cases
 ----------------------------
 
   $ hg up --quiet null
 
 (This exists to help keeping a compact list of the various cases we have built)
 
   $ hg log -T '{desc|firstline}\n'| sort
   a-1: d -move-> e
   a-2: e -move-> f
   b-1: b update
   c-1 delete d
   d-1 delete d
   d-2 re-add d
   e-1 b -move-> g
   e-2 g -move-> f
   f-1: rename h -> i
   f-2: rename i -> d
   g-1: update d
   h-1: b -(move)-> d
   i-0 initial commit: a b h p q r
   i-1: a -move-> c, p -move-> s
   i-2: c -move-> d, s -move-> t
   j-1: unrelated changes (based on the "a" series of changes)
   k-1: unrelated changes (based on "e" changes)
   l-1: unrelated changes (based on "c" changes)
   mABm-0 simple merge - A side: multiple renames, B side: unrelated update - the other way
   mAE,Km: chained merges (conflict -> simple) - same content everywhere
   mAE-change,Km: chained merges (conflict+change -> simple) - same content on both branch in the initial merge
   mAE-change-m-0 merge with file update and copies info on both side - A side: rename d to f, E side: b to f, (same content for f in parent) - one way
   mAEm-0 merge with copies info on both side - A side: rename d to f, E side: b to f, (same content for f) - one way
   mBAm-0 simple merge - A side: multiple renames, B side: unrelated update - one way
   mBC+revert,Lm: chained merges (salvaged -> simple) - same content (when the file exists)
   mBC-change-m-0 merge explicitely revive deleted file - B side: unrelated change, C side: delete d (restored by merge) - the other way
   mBC-revert-m-0 merge explicitely revive deleted file - B side: unrelated change, C side: delete d (restored by merge) - the other way
   mBCm-0 simple merge - C side: delete a file with copies history , B side: unrelated update - one way
   mBCm-1 re-add d
   mBDm-0 simple merge - B side: unrelated update, D side: delete and recreate a file (with different content) - one way
   mBF-change-m-0 merge with extra change - B side: unrelated change, F side: overwrite d with a copy (from h->i->d) - one way
   mBFm-0 simple merge - B side: unrelated change, F side: overwrite d with a copy (from h->i->d) - one way
   mBRm-0 simple merge - B side: unrelated change, R side: overwrite d with a copy (from r->x->t) different content - one way
   mCB+revert,Lm: chained merges (salvaged -> simple) - same content (when the file exists)
   mCB-change-m-0 merge explicitely revive deleted file - B side: unrelated change, C side: delete d (restored by merge) - one way
   mCB-revert-m-0 merge explicitely revive deleted file - B side: unrelated change, C side: delete d (restored by merge) - one way
   mCBm-0 simple merge - C side: delete a file with copies history , B side: unrelated update - the other way
   mCBm-1 re-add d
   mCGm-0 merge updated/deleted - revive the file (updated content) - one way
   mCH-delete-before-conflict-m-0 simple merge - C side: d is the results of renames then deleted, H side: d is result of another rename (same content as the other branch) - one way
   mDBm-0 simple merge - B side: unrelated update, D side: delete and recreate a file (with different content) - the other way
   mDGm-0 actual content merge, copies on one side - D side: delete and re-add (different content), G side: update content - one way
   mEA,Jm: chained merges (conflict -> simple) - same content everywhere
   mEA-change,Jm: chained merges (conflict+change -> simple) - same content on both branch in the initial merge
   mEA-change-m-0 merge with file update and copies info on both side - A side: rename d to f, E side: b to f, (same content for f in parent) - the other way
   mEAm-0 merge with copies info on both side - A side: rename d to f, E side: b to f, (same content for f) - the other way
   mFB-change-m-0 merge with extra change - B side: unrelated change, F side: overwrite d with a copy (from h->i->d) - the other way
   mFBm-0 simple merge - B side: unrelated change, F side: overwrite d with a copy (from h->i->d) - the other way
   mFG,Om: chained merges (copy-overwrite -> simple) - same content
   mFGm-0 merge - G side: content change, F side: copy overwrite, no content change - one way
   mGCm-0 merge updated/deleted - revive the file (updated content) - the other way
   mGDm-0 actual content merge, copies on one side - D side: delete and re-add (different content), G side: update content - the other way
   mGF,Nm: chained merges (copy-overwrite -> simple) - same content
   mGFm-0 merge - G side: content change, F side: copy overwrite, no content change - the other way
   mHC-delete-before-conflict-m-0 simple merge - C side: d is the results of renames then deleted, H side: d is result of another rename (same content as the other branch) - the other way
   mJ,EA-change-m: chained merges (conflict+change -> simple) - same content on both branch in the initial merge
   mJ,EAm: chained merges (conflict -> simple) - same content everywhere
   mK,AE-change-m: chained merges (conflict+change -> simple) - same content on both branch in the initial merge
   mK,AEm: chained merges (conflict -> simple) - same content everywhere
   mL,BC+revertm: chained merges (salvaged -> simple) - same content (when the file exists)
   mL,CB+revertm: chained merges (salvaged -> simple) - same content (when the file exists)
   mN,GFm: chained merges (copy-overwrite -> simple) - same content
   mO,FGm: chained merges (copy-overwrite -> simple) - same content
   mPQ,Tm: chained merges (conflict -> simple) - different content
   mPQm-0 merge with copies info on both side - P side: rename t to v, Q side: r to v, (different content) - one way
   mQP,Sm: chained merges (conflict -> simple) - different content
   mQPm-0 merge with copies info on both side - P side: rename t to v, Q side: r to v, (different content) - the other way
   mRBm-0 simple merge - B side: unrelated change, R side: overwrite d with a copy (from r->x->t) different content - the other way
   mS,QPm: chained merges (conflict -> simple) - different content
   mT,PQm: chained merges (conflict -> simple) - different content
   n-1: unrelated changes (based on the "f" series of changes)
   o-1: unrelated changes (based on "g" changes)
   p-1: t -move-> u
   p-2: u -move-> v
   q-1 r -move-> w
   q-2 w -move-> v
   r-1: rename r -> x
   r-2: rename t -> x
   s-1: unrelated changes (based on the "p" series of changes)
   t-1: unrelated changes (based on "q" changes)
 
 
 Test that sidedata computations during upgrades are correct
 ===========================================================
 
 We upgrade a repository that is not using sidedata (the filelog case) and
  check that the same side data have been generated as if they were computed at
  commit time.
 
 
 #if upgraded
   $ cat >> $HGRCPATH << EOF
   > [format]
   > exp-use-copies-side-data-changeset = yes
   > EOF
   $ hg debugformat -v
   format-variant     repo config default
   fncache:            yes    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   persistent-nodemap:  no     no      no (no-rust !)
   persistent-nodemap: yes    yes      no (rust !)
   copies-sdc:          no    yes      no
   revlog-v2:           no     no      no
   changelog-v2:        no    yes      no
   plain-cl-delta:     yes    yes     yes
   compression:        * (glob)
   compression-level:  default default default
   $ hg debugupgraderepo --run --quiet
   upgrade will perform the following actions:
   
   requirements
      preserved: * (glob)
      added: exp-changelog-v2, exp-copies-sidedata-changeset
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
 #endif
 
 #if upgraded-parallel
   $ cat >> $HGRCPATH << EOF
   > [format]
   > exp-use-copies-side-data-changeset = yes
   > [experimental]
   > worker.repository-upgrade=yes
   > [worker]
   > enabled=yes
   > numcpus=8
   > EOF
   $ hg debugformat -v
   format-variant     repo config default
   fncache:            yes    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   persistent-nodemap:  no     no      no (no-rust !)
   persistent-nodemap: yes    yes      no (rust !)
   copies-sdc:          no    yes      no
   revlog-v2:           no     no      no
   changelog-v2:        no    yes      no
   plain-cl-delta:     yes    yes     yes
   compression:        * (glob)
   compression-level:  default default default
   $ hg debugupgraderepo --run --quiet
   upgrade will perform the following actions:
   
   requirements
      preserved: * (glob)
      added: exp-changelog-v2, exp-copies-sidedata-changeset
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
 #endif
 
 #if pull
   $ cd ..
   $ mv repo-chain repo-source
   $ hg init repo-chain
   $ cd repo-chain
   $ hg pull ../repo-source
   pulling from ../repo-source
   requesting all changes
   adding changesets
   adding manifests
   adding file changes
   added 80 changesets with 44 changes to 25 files (+39 heads)
   new changesets a3a31bbefea6:908ce9259ffa
   (run 'hg heads' to see heads, 'hg merge' to merge)
 #endif
 
 #if pull-upgrade
   $ cat >> $HGRCPATH << EOF
   > [format]
   > exp-use-copies-side-data-changeset = yes
   > [experimental]
   > changegroup4 = yes
   > EOF
   $ cd ..
   $ mv repo-chain repo-source
   $ hg init repo-chain
   $ cd repo-chain
   $ hg pull ../repo-source
   pulling from ../repo-source
   requesting all changes
   adding changesets
   adding manifests
   adding file changes
   added 80 changesets with 44 changes to 25 files (+39 heads)
   new changesets a3a31bbefea6:908ce9259ffa
   (run 'hg heads' to see heads, 'hg merge' to merge)
 #endif
 
 #if push
   $ cd ..
   $ mv repo-chain repo-source
   $ hg init repo-chain
   $ cd repo-source
   $ hg push ../repo-chain
   pushing to ../repo-chain
   searching for changes
   adding changesets
   adding manifests
   adding file changes
   added 80 changesets with 44 changes to 25 files (+39 heads)
   $ cd ../repo-chain
 #endif
 
 #if push-upgrade
   $ cat >> $HGRCPATH << EOF
   > [format]
   > exp-use-copies-side-data-changeset = yes
   > [experimental]
   > changegroup4 = yes
   > EOF
   $ cd ..
   $ mv repo-chain repo-source
   $ hg init repo-chain
   $ cd repo-source
   $ hg push ../repo-chain
   pushing to ../repo-chain
   searching for changes
   adding changesets
   adding manifests
   adding file changes
   added 80 changesets with 44 changes to 25 files (+39 heads)
   $ cd ../repo-chain
 #endif
 
 #if no-compatibility no-filelog no-changeset
 
   $ hg debugchangedfiles --compute 0
   added      : a, ;
   added      : b, ;
   added      : h, ;
   added      : p, ;
   added      : q, ;
   added      : r, ;
 
   $ for rev in `hg log --rev 'all()' -T '{rev}\n'`; do
   >     case_id=`hg log -r $rev -T '{word(0, desc, ":")}\n'`
   >     echo "##### revision \"$case_id\" #####"
   >     hg debugsidedata -c -v -- $rev
   >     hg debugchangedfiles $rev
   > done
   ##### revision "i-0 initial commit" #####
   1 sidedata entries
    entry-0014 size 64
     '\x00\x00\x00\x06\x04\x00\x00\x00\x01\x00\x00\x00\x00\x04\x00\x00\x00\x02\x00\x00\x00\x00\x04\x00\x00\x00\x03\x00\x00\x00\x00\x04\x00\x00\x00\x04\x00\x00\x00\x00\x04\x00\x00\x00\x05\x00\x00\x00\x00\x04\x00\x00\x00\x06\x00\x00\x00\x00abhpqr'
   added      : a, ;
   added      : b, ;
   added      : h, ;
   added      : p, ;
   added      : q, ;
   added      : r, ;
   ##### revision "i-1" #####
   1 sidedata entries
    entry-0014 size 44
     '\x00\x00\x00\x04\x0c\x00\x00\x00\x01\x00\x00\x00\x00\x06\x00\x00\x00\x02\x00\x00\x00\x00\x0c\x00\x00\x00\x03\x00\x00\x00\x00\x06\x00\x00\x00\x04\x00\x00\x00\x02acps'
   removed    : a, ;
   added    p1: c, a;
   removed    : p, ;
   added    p1: s, p;
   ##### revision "i-2" #####
   1 sidedata entries
    entry-0014 size 44
     '\x00\x00\x00\x04\x0c\x00\x00\x00\x01\x00\x00\x00\x00\x06\x00\x00\x00\x02\x00\x00\x00\x00\x0c\x00\x00\x00\x03\x00\x00\x00\x00\x06\x00\x00\x00\x04\x00\x00\x00\x02cdst'
   removed    : c, ;
   added    p1: d, c;
   removed    : s, ;
   added    p1: t, s;
   ##### revision "a-1" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x02\x0c\x00\x00\x00\x01\x00\x00\x00\x00\x06\x00\x00\x00\x02\x00\x00\x00\x00de'
   removed    : d, ;
   added    p1: e, d;
   ##### revision "a-2" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x02\x0c\x00\x00\x00\x01\x00\x00\x00\x00\x06\x00\x00\x00\x02\x00\x00\x00\x00ef'
   removed    : e, ;
   added    p1: f, e;
   ##### revision "b-1" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x14\x00\x00\x00\x01\x00\x00\x00\x00b'
   touched    : b, ;
   ##### revision "c-1 delete d" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x0c\x00\x00\x00\x01\x00\x00\x00\x00d'
   removed    : d, ;
   ##### revision "d-1 delete d" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x0c\x00\x00\x00\x01\x00\x00\x00\x00d'
   removed    : d, ;
   ##### revision "d-2 re-add d" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x04\x00\x00\x00\x01\x00\x00\x00\x00d'
   added      : d, ;
   ##### revision "e-1 b -move-> g" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x02\x0c\x00\x00\x00\x01\x00\x00\x00\x00\x06\x00\x00\x00\x02\x00\x00\x00\x00bg'
   removed    : b, ;
   added    p1: g, b;
   ##### revision "e-2 g -move-> f" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x02\x06\x00\x00\x00\x01\x00\x00\x00\x01\x0c\x00\x00\x00\x02\x00\x00\x00\x00fg'
   added    p1: f, g;
   removed    : g, ;
   ##### revision "p-1" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x02\x0c\x00\x00\x00\x01\x00\x00\x00\x00\x06\x00\x00\x00\x02\x00\x00\x00\x00tu'
   removed    : t, ;
   added    p1: u, t;
   ##### revision "p-2" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x02\x0c\x00\x00\x00\x01\x00\x00\x00\x00\x06\x00\x00\x00\x02\x00\x00\x00\x00uv'
   removed    : u, ;
   added    p1: v, u;
   ##### revision "q-1 r -move-> w" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x02\x0c\x00\x00\x00\x01\x00\x00\x00\x00\x06\x00\x00\x00\x02\x00\x00\x00\x00rw'
   removed    : r, ;
   added    p1: w, r;
   ##### revision "q-2 w -move-> v" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x02\x06\x00\x00\x00\x01\x00\x00\x00\x01\x0c\x00\x00\x00\x02\x00\x00\x00\x00vw'
   added    p1: v, w;
   removed    : w, ;
   ##### revision "mBAm-0 simple merge - A side" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mABm-0 simple merge - A side" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mBCm-0 simple merge - C side" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mBCm-1 re-add d" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x04\x00\x00\x00\x01\x00\x00\x00\x00d'
   added      : d, ;
   ##### revision "mCBm-0 simple merge - C side" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mCBm-1 re-add d" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x04\x00\x00\x00\x01\x00\x00\x00\x00d'
   added      : d, ;
   ##### revision "mBDm-0 simple merge - B side" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mDBm-0 simple merge - B side" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mAEm-0 merge with copies info on both side - A side" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x08\x00\x00\x00\x01\x00\x00\x00\x00f'
   merged     : f, ;
   ##### revision "mEAm-0 merge with copies info on both side - A side" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x08\x00\x00\x00\x01\x00\x00\x00\x00f'
   merged     : f, ;
   ##### revision "mPQm-0 merge with copies info on both side - P side" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x08\x00\x00\x00\x01\x00\x00\x00\x00v'
   merged     : v, ;
   ##### revision "mQPm-0 merge with copies info on both side - P side" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x08\x00\x00\x00\x01\x00\x00\x00\x00v'
   merged     : v, ;
   ##### revision "f-1" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x02\x0c\x00\x00\x00\x01\x00\x00\x00\x00\x06\x00\x00\x00\x02\x00\x00\x00\x00hi'
   removed    : h, ;
   added    p1: i, h;
   ##### revision "f-2" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x02\x16\x00\x00\x00\x01\x00\x00\x00\x01\x0c\x00\x00\x00\x02\x00\x00\x00\x00di'
   touched  p1: d, i;
   removed    : i, ;
   ##### revision "mBFm-0 simple merge - B side" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mFBm-0 simple merge - B side" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "r-1" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x02\x0c\x00\x00\x00\x01\x00\x00\x00\x00\x06\x00\x00\x00\x02\x00\x00\x00\x00rx'
   removed    : r, ;
   added    p1: x, r;
   ##### revision "r-2" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x02\x16\x00\x00\x00\x01\x00\x00\x00\x01\x0c\x00\x00\x00\x02\x00\x00\x00\x00tx'
   touched  p1: t, x;
   removed    : x, ;
   ##### revision "mBRm-0 simple merge - B side" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mRBm-0 simple merge - B side" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "g-1" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x14\x00\x00\x00\x01\x00\x00\x00\x00d'
   touched    : d, ;
   ##### revision "mDGm-0 actual content merge, copies on one side - D side" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x08\x00\x00\x00\x01\x00\x00\x00\x00d'
   merged     : d, ;
   ##### revision "mGDm-0 actual content merge, copies on one side - D side" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x08\x00\x00\x00\x01\x00\x00\x00\x00d'
   merged     : d, ;
   ##### revision "mFGm-0 merge - G side" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x08\x00\x00\x00\x01\x00\x00\x00\x00d'
   merged     : d, ;
   ##### revision "mGFm-0 merge - G side" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x08\x00\x00\x00\x01\x00\x00\x00\x00d'
   merged     : d, ;
   ##### revision "mCGm-0 merge updated/deleted - revive the file (updated content) - one way" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x10\x00\x00\x00\x01\x00\x00\x00\x00d'
   salvaged   : d, ;
   ##### revision "mGCm-0 merge updated/deleted - revive the file (updated content) - the other way" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x10\x00\x00\x00\x01\x00\x00\x00\x00d'
   salvaged   : d, ;
   ##### revision "mCB-revert-m-0 merge explicitely revive deleted file - B side" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x10\x00\x00\x00\x01\x00\x00\x00\x00d'
   salvaged   : d, ;
   ##### revision "mBC-revert-m-0 merge explicitely revive deleted file - B side" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x10\x00\x00\x00\x01\x00\x00\x00\x00d'
   salvaged   : d, ;
   ##### revision "h-1" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x02\x0c\x00\x00\x00\x01\x00\x00\x00\x00\x06\x00\x00\x00\x02\x00\x00\x00\x00bd'
   removed    : b, ;
   added    p1: d, b;
   ##### revision "mCH-delete-before-conflict-m-0 simple merge - C side" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mHC-delete-before-conflict-m-0 simple merge - C side" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mAE-change-m-0 merge with file update and copies info on both side - A side" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x08\x00\x00\x00\x01\x00\x00\x00\x00f'
   merged     : f, ;
   ##### revision "mEA-change-m-0 merge with file update and copies info on both side - A side" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x08\x00\x00\x00\x01\x00\x00\x00\x00f'
   merged     : f, ;
   ##### revision "mBF-change-m-0 merge with extra change - B side" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x14\x00\x00\x00\x01\x00\x00\x00\x00d'
   touched    : d, ;
   ##### revision "mFB-change-m-0 merge with extra change - B side" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x14\x00\x00\x00\x01\x00\x00\x00\x00d'
   touched    : d, ;
   ##### revision "mCB-change-m-0 merge explicitely revive deleted file - B side" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x10\x00\x00\x00\x01\x00\x00\x00\x00d'
   salvaged   : d, ;
   ##### revision "mBC-change-m-0 merge explicitely revive deleted file - B side" #####
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x10\x00\x00\x00\x01\x00\x00\x00\x00d'
   salvaged   : d, ;
   ##### revision "j-1" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x01\x04\x00\x00\x00\x0b\x00\x00\x00\x00unrelated-j'
   added      : unrelated-j, ;
   ##### revision "k-1" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x01\x04\x00\x00\x00\x0b\x00\x00\x00\x00unrelated-k'
   added      : unrelated-k, ;
   ##### revision "mAE,Km" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mK,AEm" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mEA,Jm" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x01\x14\x00\x00\x00\x0b\x00\x00\x00\x00unrelated-j'
   touched    : unrelated-j, ;
   ##### revision "mJ,EAm" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x01\x14\x00\x00\x00\x0b\x00\x00\x00\x00unrelated-j'
   touched    : unrelated-j, ;
   ##### revision "s-1" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x01\x04\x00\x00\x00\x0b\x00\x00\x00\x00unrelated-s'
   added      : unrelated-s, ;
   ##### revision "t-1" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x01\x04\x00\x00\x00\x0b\x00\x00\x00\x00unrelated-t'
   added      : unrelated-t, ;
   ##### revision "mPQ,Tm" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mT,PQm" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mQP,Sm" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mS,QPm" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "l-1" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x01\x04\x00\x00\x00\x0b\x00\x00\x00\x00unrelated-l'
   added      : unrelated-l, ;
   ##### revision "mBC+revert,Lm" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mCB+revert,Lm" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mL,BC+revertm" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mL,CB+revertm" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "n-1" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x01\x04\x00\x00\x00\x0b\x00\x00\x00\x00unrelated-n'
   added      : unrelated-n, ;
   ##### revision "o-1" #####
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x01\x04\x00\x00\x00\x0b\x00\x00\x00\x00unrelated-o'
   added      : unrelated-o, ;
   ##### revision "mFG,Om" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mO,FGm" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mGF,Nm" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mN,GFm" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mAE-change,Km" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mK,AE-change-m" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mEA-change,Jm" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
   ##### revision "mJ,EA-change-m" #####
   1 sidedata entries
    entry-0014 size 4
     '\x00\x00\x00\x00'
 
 #endif
 
 
 Test copy information chaining
 ==============================
 
 Check that matching only affect the destination and not intermediate path
 -------------------------------------------------------------------------
 
 The two status call should give the same value for f
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("a-2")'
   A f
     a
   A t
     p
   R a
   R p
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("a-2")' f
   A f
     a (no-changeset no-compatibility !)
 
 merging with unrelated change does not interfere with the renames
 ---------------------------------------------------------------
 
 - rename on one side
 - unrelated change on the other side
 
   $ hg log -G --rev '::(desc("mABm")+desc("mBAm"))'
   o    mABm-0 simple merge - A side: multiple renames, B side: unrelated update - the other way
   |\
   +---o  mBAm-0 simple merge - A side: multiple renames, B side: unrelated update - one way
   | |/
   | o  b-1: b update
   | |
   o |  a-2: e -move-> f
   | |
   o |  a-1: d -move-> e
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
   $ hg status --copies --rev 'desc("b-1")' --rev 'desc("mABm")'
   A f
     d
   R d
   $ hg status --copies --rev 'desc("b-1")' --rev 'desc("mBAm")'
   A f
     d
   R d
   $ hg status --copies --rev 'desc("a-2")' --rev 'desc("mABm")'
   M b
   $ hg status --copies --rev 'desc("a-2")' --rev 'desc("mBAm")'
   M b
   $ hg status --copies --rev 'desc("i-2")' --rev 'desc("mABm")'
   M b
   A f
     d
   R d
   $ hg status --copies --rev 'desc("i-2")' --rev 'desc("mBAm")'
   M b
   A f
     d
   R d
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mABm")'
   M b
   A f
     a
   A t
     p
   R a
   R p
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mBAm")'
   M b
   A f
     a
   A t
     p
   R a
   R p
 
 merging with the side having a delete
 -------------------------------------
 
 case summary:
 - one with change to an unrelated file
 - one deleting the change
 and recreate an unrelated file after the merge
 
   $ hg log -G --rev '::(desc("mCBm")+desc("mBCm"))'
   o  mCBm-1 re-add d
   |
   o    mCBm-0 simple merge - C side: delete a file with copies history , B side: unrelated update - the other way
   |\
   | | o  mBCm-1 re-add d
   | | |
   +---o  mBCm-0 simple merge - C side: delete a file with copies history , B side: unrelated update - one way
   | |/
   | o  c-1 delete d
   | |
   o |  b-1: b update
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 - comparing from the merge
 
   $ hg status --copies --rev 'desc("b-1")' --rev 'desc("mBCm-0")'
   R d
   $ hg status --copies --rev 'desc("b-1")' --rev 'desc("mCBm-0")'
   R d
   $ hg status --copies --rev 'desc("c-1")' --rev 'desc("mBCm-0")'
   M b
   $ hg status --copies --rev 'desc("c-1")' --rev 'desc("mCBm-0")'
   M b
   $ hg status --copies --rev 'desc("i-2")' --rev 'desc("mBCm-0")'
   M b
   R d
   $ hg status --copies --rev 'desc("i-2")' --rev 'desc("mCBm-0")'
   M b
   R d
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mBCm-0")'
   M b
   A t
     p
   R a
   R p
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mCBm-0")'
   M b
   A t
     p
   R a
   R p
 
 - comparing with the merge children re-adding the file
 
   $ hg status --copies --rev 'desc("b-1")' --rev 'desc("mBCm-1")'
   M d
   $ hg status --copies --rev 'desc("b-1")' --rev 'desc("mCBm-1")'
   M d
   $ hg status --copies --rev 'desc("c-1")' --rev 'desc("mBCm-1")'
   M b
   A d
   $ hg status --copies --rev 'desc("c-1")' --rev 'desc("mCBm-1")'
   M b
   A d
   $ hg status --copies --rev 'desc("i-2")' --rev 'desc("mBCm-1")'
   M b
   M d
   $ hg status --copies --rev 'desc("i-2")' --rev 'desc("mCBm-1")'
   M b
   M d
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mBCm-1")'
   M b
   A d
   A t
     p
   R a
   R p
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mCBm-1")'
   M b
   A d
   A t
     p
   R a
   R p
 
 Comparing with a merge re-adding the file afterward
 ---------------------------------------------------
 
 Merge:
 - one with change to an unrelated file
 - one deleting and recreating the change
 
   $ hg log -G --rev '::(desc("mDBm")+desc("mBDm"))'
   o    mDBm-0 simple merge - B side: unrelated update, D side: delete and recreate a file (with different content) - the other way
   |\
   +---o  mBDm-0 simple merge - B side: unrelated update, D side: delete and recreate a file (with different content) - one way
   | |/
   | o  d-2 re-add d
   | |
   | o  d-1 delete d
   | |
   o |  b-1: b update
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
   $ hg status --copies --rev 'desc("b-1")' --rev 'desc("mBDm-0")'
   M d
   $ hg status --copies --rev 'desc("b-1")' --rev 'desc("mDBm-0")'
   M d
   $ hg status --copies --rev 'desc("d-2")' --rev 'desc("mBDm-0")'
   M b
   $ hg status --copies --rev 'desc("d-2")' --rev 'desc("mDBm-0")'
   M b
   $ hg status --copies --rev 'desc("i-2")' --rev 'desc("mBDm-0")'
   M b
   M d
   $ hg status --copies --rev 'desc("i-2")' --rev 'desc("mDBm-0")'
   M b
   M d
 
 The bugs makes recorded copy is different depending of where we started the merge from since
 
   $ hg manifest --debug --rev 'desc("mBDm-0")' | grep '644   d'
   b004912a8510032a0350a74daa2803dadfb00e12 644   d
   $ hg manifest --debug --rev 'desc("mDBm-0")' | grep '644   d'
   b004912a8510032a0350a74daa2803dadfb00e12 644   d
 
   $ hg manifest --debug --rev 'desc("d-2")' | grep '644   d'
   b004912a8510032a0350a74daa2803dadfb00e12 644   d
   $ hg manifest --debug --rev 'desc("b-1")' | grep '644   d'
   d8252ab2e760b0d4e5288fd44cbd15a0fa567e16 644   d (no-changeset !)
   ae258f702dfeca05bf9b6a22a97a4b5645570f11 644   d (changeset !)
   $ hg debugindex d | head -n 4 | "$PYTHON" ../no-linkrev
      rev linkrev nodeid       p1           p2
        0       * d8252ab2e760 000000000000 000000000000 (no-changeset !)
        0       * ae258f702dfe 000000000000 000000000000 (changeset !)
        1       * b004912a8510 000000000000 000000000000
        2       * 7b79e2fe0c89 000000000000 000000000000 (no-changeset !)
        2       * 5cce88bf349f ae258f702dfe 000000000000 (changeset !)
 
 Log output should not include a merge commit as it did not happen
 
   $ hg log -Gfr 'desc("mBDm-0")' d
   o  d-2 re-add d
   |
   ~
 
   $ hg log -Gfr 'desc("mDBm-0")' d
   o  d-2 re-add d
   |
   ~
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mBDm-0")'
   M b
   A d
   A t
     p
   R a
   R p
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mDBm-0")'
   M b
   A d
   A t
     p
   R a
   R p
 
 
 Comparing with a merge with colliding rename
 --------------------------------------------
 
 Subcase: new copy information on both side
 ``````````````````````````````````````````
 
 - the "e-" branch renaming b to f (through 'g')
 - the "a-" branch renaming d to f (through e)
 
   $ hg log -G --rev '::(desc("mAEm")+desc("mEAm"))'
   o    mEAm-0 merge with copies info on both side - A side: rename d to f, E side: b to f, (same content for f) - the other way
   |\
   +---o  mAEm-0 merge with copies info on both side - A side: rename d to f, E side: b to f, (same content for f) - one way
   | |/
   | o  e-2 g -move-> f
   | |
   | o  e-1 b -move-> g
   | |
   o |  a-2: e -move-> f
   | |
   o |  a-1: d -move-> e
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 #if no-changeset
   $ hg manifest --debug --rev 'desc("mAEm-0")' | grep '644   f'
   2ff93c643948464ee1f871867910ae43a45b0bea 644   f
   $ hg manifest --debug --rev 'desc("mEAm-0")' | grep '644   f'
   2ff93c643948464ee1f871867910ae43a45b0bea 644   f
   $ hg manifest --debug --rev 'desc("a-2")' | grep '644   f'
   b76eb76580df486c3d51d63c5c210d4dd43a8ac7 644   f
   $ hg manifest --debug --rev 'desc("e-2")' | grep '644   f'
   e8825b386367b29fec957283a80bb47b47483fe1 644   f
   $ hg debugindex f | "$PYTHON" ../no-linkrev
      rev linkrev nodeid       p1           p2
        0       * b76eb76580df 000000000000 000000000000
        1       * e8825b386367 000000000000 000000000000
        2       * 2ff93c643948 b76eb76580df e8825b386367
        3       * 2f649fba7eb2 b76eb76580df e8825b386367
        4       * 774e7c1637d5 e8825b386367 b76eb76580df
 #else
   $ hg manifest --debug --rev 'desc("mAEm-0")' | grep '644   f'
   ae258f702dfeca05bf9b6a22a97a4b5645570f11 644   f
   $ hg manifest --debug --rev 'desc("mEAm-0")' | grep '644   f'
   ae258f702dfeca05bf9b6a22a97a4b5645570f11 644   f
   $ hg manifest --debug --rev 'desc("a-2")' | grep '644   f'
   ae258f702dfeca05bf9b6a22a97a4b5645570f11 644   f
   $ hg manifest --debug --rev 'desc("e-2")' | grep '644   f'
   ae258f702dfeca05bf9b6a22a97a4b5645570f11 644   f
   $ hg debugindex f | "$PYTHON" ../no-linkrev
      rev linkrev nodeid       p1           p2
        0       * ae258f702dfe 000000000000 000000000000
        1       * d3613c1ec831 ae258f702dfe 000000000000
        2       * 05e03c868bbc ae258f702dfe 000000000000
 #endif
 
 # Here the filelog based implementation is not looking at the rename
 # information (because the file exist on both side). However the changelog
 # based on works fine. We have different output.
 
   $ hg status --copies --rev 'desc("a-2")' --rev 'desc("mAEm-0")'
   M f (no-changeset !)
     b (no-filelog no-changeset !)
   R b
   $ hg status --copies --rev 'desc("a-2")' --rev 'desc("mEAm-0")'
   M f (no-changeset !)
     b (no-filelog no-changeset !)
   R b
   $ hg status --copies --rev 'desc("e-2")' --rev 'desc("mAEm-0")'
   M f (no-changeset !)
     d (no-filelog no-changeset !)
   R d
   $ hg status --copies --rev 'desc("e-2")' --rev 'desc("mEAm-0")'
   M f (no-changeset !)
     d (no-filelog no-changeset !)
   R d
   $ hg status --copies --rev 'desc("i-2")' --rev 'desc("a-2")'
   A f
     d
   R d
   $ hg status --copies --rev 'desc("i-2")' --rev 'desc("e-2")'
   A f
     b
   R b
 
 # From here, we run status against revision where both source file exists.
 #
 # The filelog based implementation picks an arbitrary side based on revision
 # numbers. So the same side "wins" whatever the parents order is. This is
 # sub-optimal because depending on revision numbers means the result can be
 # different from one repository to the next.
 #
 # The changeset based algorithm use the parent order to break tie on conflicting
 # information and will have a different order depending on who is p1 and p2.
 # That order is stable accross repositories. (data from p1 prevails)
 
   $ hg status --copies --rev 'desc("i-2")' --rev 'desc("mAEm-0")'
   A f
     d
   R b
   R d
   $ hg status --copies --rev 'desc("i-2")' --rev 'desc("mEAm-0")'
   A f
     d (filelog !)
     b (no-filelog !)
   R b
   R d
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mAEm-0")'
   A f
     a
   A t
     p
   R a
   R b
   R p
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mEAm-0")'
   A f
     a (filelog !)
     b (no-filelog !)
   A t
     p
   R a
   R b
   R p
 
 
 Subcase: existing copy information overwritten on one branch
 ````````````````````````````````````````````````````````````
 
 Note:
 | In this case, one of the merge wrongly record a merge while there is none.
 | This lead to bad copy tracing information to be dug up.
 
 
 Merge:
 - one with change to an unrelated file (b)
 - one overwriting a file (d) with a rename (from h to i to d)
 
   $ hg log -G --rev '::(desc("mBFm")+desc("mFBm"))'
   o    mFBm-0 simple merge - B side: unrelated change, F side: overwrite d with a copy (from h->i->d) - the other way
   |\
   +---o  mBFm-0 simple merge - B side: unrelated change, F side: overwrite d with a copy (from h->i->d) - one way
   | |/
   | o  f-2: rename i -> d
   | |
   | o  f-1: rename h -> i
   | |
   o |  b-1: b update
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mBFm-0")'
   M b
   A d
     h
   A t
     p
   R a
   R h
   R p
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mFBm-0")'
   M b
   A d
     h
   A t
     p
   R a
   R h
   R p
   $ hg status --copies --rev 'desc("b-1")' --rev 'desc("mBFm-0")'
   M d (no-changeset !)
     h (no-filelog no-changeset !)
   R h
   $ hg status --copies --rev 'desc("f-2")' --rev 'desc("mBFm-0")'
   M b
   $ hg status --copies --rev 'desc("f-1")' --rev 'desc("mBFm-0")'
   M b
   M d (no-changeset !)
     i (no-filelog no-changeset !)
   R i
   $ hg status --copies --rev 'desc("b-1")' --rev 'desc("mFBm-0")'
   M d (no-changeset !)
     h (no-filelog no-changeset !)
   R h
   $ hg status --copies --rev 'desc("f-2")' --rev 'desc("mFBm-0")'
   M b
   $ hg status --copies --rev 'desc("f-1")' --rev 'desc("mFBm-0")'
   M b
   M d (no-changeset !)
     i (no-filelog no-changeset !)
   R i
 
 #if no-changeset
   $ hg log -Gfr 'desc("mBFm-0")' d
   o  f-2: rename i -> d
   |
   o  f-1: rename h -> i
   :
   o  i-0 initial commit: a b h p q r
   
 #else
 BROKEN: `hg log --follow <file>` relies on filelog metadata to work
   $ hg log -Gfr 'desc("mBFm-0")' d
   o  i-2: c -move-> d, s -move-> t
   |
   ~
 #endif
 
 #if no-changeset
   $ hg log -Gfr 'desc("mFBm-0")' d
   o  f-2: rename i -> d
   |
   o  f-1: rename h -> i
   :
   o  i-0 initial commit: a b h p q r
   
 #else
 BROKEN: `hg log --follow <file>` relies on filelog metadata to work
   $ hg log -Gfr 'desc("mFBm-0")' d
   o  i-2: c -move-> d, s -move-> t
   |
   ~
 #endif
 
 
 Subcase: existing copy information overwritten on one branch, with different content)
 `````````````````````````````````````````````````````````````````````````````````````
 
 Merge:
 - one with change to an unrelated file (b)
 - one overwriting a file (t) with a rename (from r to x to t), v content is not the same as on the other branch
 
   $ hg log -G --rev '::(desc("mBRm")+desc("mRBm"))'
   o    mRBm-0 simple merge - B side: unrelated change, R side: overwrite d with a copy (from r->x->t) different content - the other way
   |\
   +---o  mBRm-0 simple merge - B side: unrelated change, R side: overwrite d with a copy (from r->x->t) different content - one way
   | |/
   | o  r-2: rename t -> x
   | |
   | o  r-1: rename r -> x
   | |
   o |  b-1: b update
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mBRm-0")'
   M b
   A d
     a
   A t
     r
   R a
   R p
   R r
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mRBm-0")'
   M b
   A d
     a
   A t
     r
   R a
   R p
   R r
   $ hg status --copies --rev 'desc("b-1")' --rev 'desc("mBRm-0")'
   M t
     r (no-filelog !)
   R r
   $ hg status --copies --rev 'desc("r-2")' --rev 'desc("mBRm-0")'
   M b
   $ hg status --copies --rev 'desc("r-1")' --rev 'desc("mBRm-0")'
   M b
   M t
     x (no-filelog !)
   R x
   $ hg status --copies --rev 'desc("b-1")' --rev 'desc("mRBm-0")'
   M t
     r (no-filelog !)
   R r
   $ hg status --copies --rev 'desc("r-2")' --rev 'desc("mRBm-0")'
   M b
   $ hg status --copies --rev 'desc("r-1")' --rev 'desc("mRBm-0")'
   M b
   M t
     x (no-filelog !)
   R x
 
 #if no-changeset
   $ hg log -Gfr 'desc("mBRm-0")' d
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 #else
 BROKEN: `hg log --follow <file>` relies on filelog metadata to work
   $ hg log -Gfr 'desc("mBRm-0")' d
   o  i-2: c -move-> d, s -move-> t
   |
   ~
 #endif
 
 #if no-changeset
   $ hg log -Gfr 'desc("mRBm-0")' d
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 #else
 BROKEN: `hg log --follow <file>` relies on filelog metadata to work
   $ hg log -Gfr 'desc("mRBm-0")' d
   o  i-2: c -move-> d, s -move-> t
   |
   ~
 #endif
 
 Subcase: reset of the copy history on one side
 ``````````````````````````````````````````````
 
 Merge:
 - one with change to a file
 - one deleting and recreating the file
 
 Unlike in the 'BD/DB' cases, an actual merge happened here. So we should
 consider history and rename on both branch of the merge.
 
   $ hg log -G --rev '::(desc("mDGm")+desc("mGDm"))'
   o    mGDm-0 actual content merge, copies on one side - D side: delete and re-add (different content), G side: update content - the other way
   |\
   +---o  mDGm-0 actual content merge, copies on one side - D side: delete and re-add (different content), G side: update content - one way
   | |/
   | o  g-1: update d
   | |
   o |  d-2 re-add d
   | |
   o |  d-1 delete d
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 One side of the merge have a long history with rename. The other side of the
 merge point to a new file with a smaller history. Each side is "valid".
 
 (and again the filelog based algorithm only explore one, with a pick based on
 revision numbers)
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mDGm-0")'
   A d
     a (filelog !)
   A t
     p
   R a
   R p
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mGDm-0")'
   A d
     a
   A t
     p
   R a
   R p
   $ hg status --copies --rev 'desc("d-2")' --rev 'desc("mDGm-0")'
   M d
   $ hg status --copies --rev 'desc("d-2")' --rev 'desc("mGDm-0")'
   M d
   $ hg status --copies --rev 'desc("g-1")' --rev 'desc("mDGm-0")'
   M d
   $ hg status --copies --rev 'desc("g-1")' --rev 'desc("mGDm-0")'
   M d
 
 #if no-changeset
   $ hg log -Gfr 'desc("mDGm-0")' d
   o    mDGm-0 actual content merge, copies on one side - D side: delete and re-add (different content), G side: update content - one way
   |\
   | o  g-1: update d
   | |
   o |  d-2 re-add d
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 #else
 BROKEN: `hg log --follow <file>` relies on filelog metadata to work
   $ hg log -Gfr 'desc("mDGm-0")' d
   o    mDGm-0 actual content merge, copies on one side - D side: delete and re-add (different content), G side: update content - one way
   |\
   | o  g-1: update d
   | |
   o |  d-2 re-add d
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   ~
 #endif
 
 
 #if no-changeset
   $ hg log -Gfr 'desc("mDGm-0")' d
   o    mDGm-0 actual content merge, copies on one side - D side: delete and re-add (different content), G side: update content - one way
   |\
   | o  g-1: update d
   | |
   o |  d-2 re-add d
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 #else
 BROKEN: `hg log --follow <file>` relies on filelog metadata to work
   $ hg log -Gfr 'desc("mDGm-0")' d
   o    mDGm-0 actual content merge, copies on one side - D side: delete and re-add (different content), G side: update content - one way
   |\
   | o  g-1: update d
   | |
   o |  d-2 re-add d
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   ~
 #endif
 
 Subcase: merging a change to a file with a "copy overwrite" to that file from another branch
 ````````````````````````````````````````````````````````````````````````````````````````````
 
 Merge:
 - one with change to a file (d)
 - one overwriting that file with a rename (from h to i, to d)
 
 This case is similar to BF/FB, but an actual merge happens, so both side of the
 history are relevant.
 
 
   $ hg log -G --rev '::(desc("mGFm")+desc("mFGm"))'
   o    mGFm-0 merge - G side: content change, F side: copy overwrite, no content change - the other way
   |\
   +---o  mFGm-0 merge - G side: content change, F side: copy overwrite, no content change - one way
   | |/
   | o  g-1: update d
   | |
   o |  f-2: rename i -> d
   | |
   o |  f-1: rename h -> i
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 Note:
 | In this case, the merge get conflicting information since on one side we have
 | "a -> c -> d". and one the other one we have "h -> i -> d".
 |
 | The current code arbitrarily pick one side depending the ordering of the merged hash:
 
 In this case, the file hash from "f-2" is lower, so it will be `p1` of the resulting filenode its copy tracing information will win (and trace back to "h"):
 
 Details on this hash ordering pick:
 
   $ hg manifest --debug 'desc("g-1")' | egrep 'd$'
   17ec97e605773eb44a117d1136b3849bcdc1924f 644   d (no-changeset !)
   5cce88bf349f7c742bb440f2c53f81db9c294279 644   d (changeset !)
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("g-1")' d
   A d
     a (no-changeset no-compatibility !)
 
   $ hg manifest --debug 'desc("f-2")' | egrep 'd$'
   7b79e2fe0c8924e0e598a82f048a7b024afa4d96 644   d (no-changeset !)
   ae258f702dfeca05bf9b6a22a97a4b5645570f11 644   d (changeset !)
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("f-2")' d
   A d
     h (no-changeset no-compatibility !)
 
 Copy tracing data on the resulting merge:
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mFGm-0")'
   A d
     h (no-filelog !)
     a (filelog !)
   A t
     p
   R a
   R h
   R p
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mGFm-0")'
   A d
     a (no-changeset !)
     h (changeset !)
   A t
     p
   R a
   R h
   R p
   $ hg status --copies --rev 'desc("f-2")' --rev 'desc("mFGm-0")'
   M d
   $ hg status --copies --rev 'desc("f-2")' --rev 'desc("mGFm-0")'
   M d
   $ hg status --copies --rev 'desc("f-1")' --rev 'desc("mFGm-0")'
   M d
     i (no-filelog !)
   R i
   $ hg status --copies --rev 'desc("f-1")' --rev 'desc("mGFm-0")'
   M d
     i (no-filelog !)
   R i
   $ hg status --copies --rev 'desc("g-1")' --rev 'desc("mFGm-0")'
   M d (no-changeset !)
     h (no-filelog no-changeset !)
   R h
   $ hg status --copies --rev 'desc("g-1")' --rev 'desc("mGFm-0")'
   M d (no-changeset !)
     h (no-filelog no-changeset !)
   R h
 
 #if no-changeset
   $ hg log -Gfr 'desc("mFGm-0")' d
   o    mFGm-0 merge - G side: content change, F side: copy overwrite, no content change - one way
   |\
   | o  g-1: update d
   | |
   o |  f-2: rename i -> d
   | |
   o |  f-1: rename h -> i
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 #else
 BROKEN: `hg log --follow <file>` relies on filelog metadata to work
   $ hg log -Gfr 'desc("mFGm-0")' d
   o  g-1: update d
   |
   o  i-2: c -move-> d, s -move-> t
   |
   ~
 #endif
 
 #if no-changeset
   $ hg log -Gfr 'desc("mGFm-0")' d
   o    mGFm-0 merge - G side: content change, F side: copy overwrite, no content change - the other way
   |\
   | o  g-1: update d
   | |
   o |  f-2: rename i -> d
   | |
   o |  f-1: rename h -> i
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 #else
 BROKEN: `hg log --follow <file>` relies on filelog metadata to work
   $ hg log -Gfr 'desc("mGFm-0")' d
   o  g-1: update d
   |
   o  i-2: c -move-> d, s -move-> t
   |
   ~
 #endif
 
 Subcase: new copy information on both side with an actual merge happening
 `````````````````````````````````````````````````````````````````````````
 
 - the "p-" branch renaming 't' to 'v' (through 'u')
 - the "q-" branch renaming 'r' to 'v' (through 'w')
 
 
   $ hg log -G --rev '::(desc("mPQm")+desc("mQPm"))'
   o    mQPm-0 merge with copies info on both side - P side: rename t to v, Q side: r to v, (different content) - the other way
   |\
   +---o  mPQm-0 merge with copies info on both side - P side: rename t to v, Q side: r to v, (different content) - one way
   | |/
   | o  q-2 w -move-> v
   | |
   | o  q-1 r -move-> w
   | |
   o |  p-2: u -move-> v
   | |
   o |  p-1: t -move-> u
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 #if no-changeset
   $ hg manifest --debug --rev 'desc("mPQm-0")' | grep '644   v'
   0946c662ef16e4e67397fd717389eb6693d41749 644   v
   $ hg manifest --debug --rev 'desc("mQPm-0")' | grep '644   v'
   0db3aad7fcc1ec27fab57060e327b9e864ea0cc9 644   v
   $ hg manifest --debug --rev 'desc("p-2")' | grep '644   v'
   3f91841cd75cadc9a1f1b4e7c1aa6d411f76032e 644   v
   $ hg manifest --debug --rev 'desc("q-2")' | grep '644   v'
   c43c088b811fd27983c0a9aadf44f3343cd4cd7e 644   v
   $ hg debugindex v | "$PYTHON" ../no-linkrev
      rev linkrev nodeid       p1           p2
        0       * 3f91841cd75c 000000000000 000000000000
        1       * c43c088b811f 000000000000 000000000000
        2       * 0946c662ef16 3f91841cd75c c43c088b811f
        3       * 0db3aad7fcc1 c43c088b811f 3f91841cd75c
 #else
   $ hg manifest --debug --rev 'desc("mPQm-0")' | grep '644   v'
   65fde9f6e4d4da23b3f610e07b53673ea9541d75 644   v
   $ hg manifest --debug --rev 'desc("mQPm-0")' | grep '644   v'
   a098dda6413aecf154eefc976afc38b295acb7e5 644   v
   $ hg manifest --debug --rev 'desc("p-2")' | grep '644   v'
   5aed6a8dbff0301328c08360d24354d3d064cf0d 644   v
   $ hg manifest --debug --rev 'desc("q-2")' | grep '644   v'
   a38b2fa170219750dac9bc7d19df831f213ba708 644   v
   $ hg debugindex v | "$PYTHON" ../no-linkrev
      rev linkrev nodeid       p1           p2
        0       * 5aed6a8dbff0 000000000000 000000000000
        1       * a38b2fa17021 000000000000 000000000000
        2       * 65fde9f6e4d4 5aed6a8dbff0 a38b2fa17021
        3       * a098dda6413a a38b2fa17021 5aed6a8dbff0
 #endif
 
 # Here the filelog based implementation is not looking at the rename
 # information (because the file exist on both side). However the changelog
 # based on works fine. We have different output.
 
   $ hg status --copies --rev 'desc("p-2")' --rev 'desc("mPQm-0")'
   M v
     r (no-filelog !)
   R r
   $ hg status --copies --rev 'desc("p-2")' --rev 'desc("mQPm-0")'
   M v
     r (no-filelog !)
   R r
   $ hg status --copies --rev 'desc("q-2")' --rev 'desc("mPQm-0")'
   M v
     t (no-filelog !)
   R t
   $ hg status --copies --rev 'desc("q-2")' --rev 'desc("mQPm-0")'
   M v
     t (no-filelog !)
   R t
   $ hg status --copies --rev 'desc("i-2")' --rev 'desc("p-2")'
   A v
     t
   R t
   $ hg status --copies --rev 'desc("i-2")' --rev 'desc("q-2")'
   A v
     r
   R r
 
 # From here, we run status against revision where both source file exists.
 #
 # The filelog based implementation picks an arbitrary side based on revision
 # numbers. So the same side "wins" whatever the parents order is. This is
 # sub-optimal because depending on revision numbers means the result can be
 # different from one repository to the next.
 #
 # The changeset based algorithm use the parent order to break tie on conflicting
 # information and will have a different order depending on who is p1 and p2.
 # That order is stable accross repositories. (data from p1 prevails)
 
   $ hg status --copies --rev 'desc("i-2")' --rev 'desc("mPQm-0")'
   A v
     t
   R r
   R t
   $ hg status --copies --rev 'desc("i-2")' --rev 'desc("mQPm-0")'
   A v
     t (filelog !)
     r (no-filelog !)
   R r
   R t
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mPQm-0")'
   A d
     a
   A v
     r (filelog !)
     p (no-filelog !)
   R a
   R p
   R r
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mQPm-0")'
   A d
     a
   A v
     r
   R a
   R p
   R r
 
 
 Comparing with merging with a deletion (and keeping the file)
 -------------------------------------------------------------
 
 Merge:
 - one removing a file (d)
 - one updating that file
 - the merge keep the modified version of the file (canceling the delete)
 
 In this case, the file keep on living after the merge. So we should not drop its
 copy tracing chain.
 
   $ hg log -G --rev '::(desc("mCGm")+desc("mGCm"))'
   o    mGCm-0 merge updated/deleted - revive the file (updated content) - the other way
   |\
   +---o  mCGm-0 merge updated/deleted - revive the file (updated content) - one way
   | |/
   | o  g-1: update d
   | |
   o |  c-1 delete d
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 'a' is the copy source of 'd'
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mCGm-0")'
   A d
     a (no-compatibility no-changeset !)
   A t
     p
   R a
   R p
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mGCm-0")'
   A d
     a (no-compatibility no-changeset !)
   A t
     p
   R a
   R p
   $ hg status --copies --rev 'desc("c-1")' --rev 'desc("mCGm-0")'
   A d
   $ hg status --copies --rev 'desc("c-1")' --rev 'desc("mGCm-0")'
   A d
   $ hg status --copies --rev 'desc("g-1")' --rev 'desc("mCGm-0")'
   $ hg status --copies --rev 'desc("g-1")' --rev 'desc("mGCm-0")'
 
 
 Comparing with merge restoring an untouched deleted file
 --------------------------------------------------------
 
 Merge:
 - one removing a file (d)
 - one leaving the file untouched
 - the merge actively restore the file to the same content.
 
 In this case, the file keep on living after the merge. So we should not drop its
 copy tracing chain.
 
   $ hg log -G --rev '::(desc("mCB-revert-m")+desc("mBC-revert-m"))'
   o    mBC-revert-m-0 merge explicitely revive deleted file - B side: unrelated change, C side: delete d (restored by merge) - the other way
   |\
   +---o  mCB-revert-m-0 merge explicitely revive deleted file - B side: unrelated change, C side: delete d (restored by merge) - one way
   | |/
   | o  c-1 delete d
   | |
   o |  b-1: b update
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 'a' is the the copy source of 'd'
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mCB-revert-m-0")'
   M b
   A d
     a (no-compatibility no-changeset !)
   A t
     p
   R a
   R p
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mBC-revert-m-0")'
   M b
   A d
     a (no-compatibility no-changeset !)
   A t
     p
   R a
   R p
   $ hg status --copies --rev 'desc("c-1")' --rev 'desc("mCB-revert-m-0")'
   M b
   A d
   $ hg status --copies --rev 'desc("c-1")' --rev 'desc("mBC-revert-m-0")'
   M b
   A d
   $ hg status --copies --rev 'desc("b-1")' --rev 'desc("mCB-revert-m-0")'
   $ hg status --copies --rev 'desc("b-1")' --rev 'desc("mBC-revert-m-0")'
 
 
 Merging a branch where a rename was deleted with a branch where the same file was renamed
 ------------------------------------------------------------------------------------------
 
 Create a "conflicting" merge where `d` get removed on one branch before its
 rename information actually conflict with the other branch.
 
 (the copy information from the branch that was not deleted should win).
 
   $ hg log -G --rev '::(desc("mCH-delete-before-conflict-m")+desc("mHC-delete-before-conflict-m"))'
   o    mHC-delete-before-conflict-m-0 simple merge - C side: d is the results of renames then deleted, H side: d is result of another rename (same content as the other branch) - the other way
   |\
   +---o  mCH-delete-before-conflict-m-0 simple merge - C side: d is the results of renames then deleted, H side: d is result of another rename (same content as the other branch) - one way
   | |/
   | o  h-1: b -(move)-> d
   | |
   o |  c-1 delete d
   | |
   o |  i-2: c -move-> d, s -move-> t
   | |
   o |  i-1: a -move-> c, p -move-> s
   |/
   o  i-0 initial commit: a b h p q r
   
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mCH-delete-before-conflict-m")'
   A d
     b (no-compatibility no-changeset !)
   A t
     p
   R a
   R b
   R p
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mHC-delete-before-conflict-m")'
   A d
     b
   A t
     p
   R a
   R b
   R p
   $ hg status --copies --rev 'desc("c-1")' --rev 'desc("mCH-delete-before-conflict-m")'
   A d
     b
   R b
   $ hg status --copies --rev 'desc("c-1")' --rev 'desc("mHC-delete-before-conflict-m")'
   A d
     b
   R b
   $ hg status --copies --rev 'desc("h-1")' --rev 'desc("mCH-delete-before-conflict-m")'
   A t
     p
   R a
   R p
   $ hg status --copies --rev 'desc("h-1")' --rev 'desc("mHC-delete-before-conflict-m")'
   A t
     p
   R a
   R p
 
 Variant of previous with extra changes introduced by the merge
 --------------------------------------------------------------
 
 (see case declaration for details)
 
 Subcase: merge has same initial content on both side, but merge introduced a change
 ```````````````````````````````````````````````````````````````````````````````````
 
 - the "e-" branch renaming b to f (through 'g')
 - the "a-" branch renaming d to f (through e)
 - the merge add new change to b
 
   $ hg log -G --rev '::(desc("mAE-change-m")+desc("mEA-change-m"))'
   o    mEA-change-m-0 merge with file update and copies info on both side - A side: rename d to f, E side: b to f, (same content for f in parent) - the other way
   |\
   +---o  mAE-change-m-0 merge with file update and copies info on both side - A side: rename d to f, E side: b to f, (same content for f in parent) - one way
   | |/
   | o  e-2 g -move-> f
   | |
   | o  e-1 b -move-> g
   | |
   o |  a-2: e -move-> f
   | |
   o |  a-1: d -move-> e
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 #if no-changeset
   $ hg manifest --debug --rev 'desc("mAE-change-m-0")' | grep '644   f'
   2f649fba7eb284e720d02b61f0546fcef694c045 644   f
   $ hg manifest --debug --rev 'desc("mEA-change-m-0")' | grep '644   f'
   774e7c1637d536b99e2d8ef16fd731f87a82bd09 644   f
   $ hg manifest --debug --rev 'desc("a-2")' | grep '644   f'
   b76eb76580df486c3d51d63c5c210d4dd43a8ac7 644   f
   $ hg manifest --debug --rev 'desc("e-2")' | grep '644   f'
   e8825b386367b29fec957283a80bb47b47483fe1 644   f
   $ hg debugindex f | "$PYTHON" ../no-linkrev
      rev linkrev nodeid       p1           p2
        0       * b76eb76580df 000000000000 000000000000
        1       * e8825b386367 000000000000 000000000000
        2       * 2ff93c643948 b76eb76580df e8825b386367
        3       * 2f649fba7eb2 b76eb76580df e8825b386367
        4       * 774e7c1637d5 e8825b386367 b76eb76580df
 #else
   $ hg manifest --debug --rev 'desc("mAE-change-m-0")' | grep '644   f'
   d3613c1ec8310a812ac4268fd853ac576b6caea5 644   f
   $ hg manifest --debug --rev 'desc("mEA-change-m-0")' | grep '644   f'
   05e03c868bbcab4a649cb33a238d7aa07398a469 644   f
   $ hg manifest --debug --rev 'desc("a-2")' | grep '644   f'
   ae258f702dfeca05bf9b6a22a97a4b5645570f11 644   f
   $ hg manifest --debug --rev 'desc("e-2")' | grep '644   f'
   ae258f702dfeca05bf9b6a22a97a4b5645570f11 644   f
   $ hg debugindex f | "$PYTHON" ../no-linkrev
      rev linkrev nodeid       p1           p2
        0       * ae258f702dfe 000000000000 000000000000
        1       * d3613c1ec831 ae258f702dfe 000000000000
        2       * 05e03c868bbc ae258f702dfe 000000000000
 #endif
 
 # Here the filelog based implementation is not looking at the rename
 # information (because the file exist on both side). However the changelog
 # based on works fine. We have different output.
 
   $ hg status --copies --rev 'desc("a-2")' --rev 'desc("mAE-change-m-0")'
   M f
     b (no-filelog !)
   R b
   $ hg status --copies --rev 'desc("a-2")' --rev 'desc("mEA-change-m-0")'
   M f
     b (no-filelog !)
   R b
   $ hg status --copies --rev 'desc("e-2")' --rev 'desc("mAE-change-m-0")'
   M f
     d (no-filelog !)
   R d
   $ hg status --copies --rev 'desc("e-2")' --rev 'desc("mEA-change-m-0")'
   M f
     d (no-filelog !)
   R d
   $ hg status --copies --rev 'desc("i-2")' --rev 'desc("a-2")'
   A f
     d
   R d
   $ hg status --copies --rev 'desc("i-2")' --rev 'desc("e-2")'
   A f
     b
   R b
 
 # From here, we run status against revision where both source file exists.
 #
 # The filelog based implementation picks an arbitrary side based on revision
 # numbers. So the same side "wins" whatever the parents order is. This is
 # sub-optimal because depending on revision numbers means the result can be
 # different from one repository to the next.
 #
 # The changeset based algorithm use the parent order to break tie on conflicting
 # information and will have a different order depending on who is p1 and p2.
 # That order is stable accross repositories. (data from p1 prevails)
 
   $ hg status --copies --rev 'desc("i-2")' --rev 'desc("mAE-change-m-0")'
   A f
     d
   R b
   R d
   $ hg status --copies --rev 'desc("i-2")' --rev 'desc("mEA-change-m-0")'
   A f
     d (filelog !)
     b (no-filelog !)
   R b
   R d
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mAE-change-m-0")'
   A f
     a
   A t
     p
   R a
   R b
   R p
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mEA-change-m-0")'
   A f
     a (filelog !)
     b (no-filelog !)
   A t
     p
   R a
   R b
   R p
 
 
 Subcase: merge overwrite common copy information, but with extra change during the merge
 ```````````````````````````````````````````````````````````````````````````````````
 
 Merge:
 - one with change to an unrelated file (b)
 - one overwriting a file (d) with a rename (from h to i to d)
 
   $ hg log -G --rev '::(desc("mBF-change-m")+desc("mFB-change-m"))'
   o    mFB-change-m-0 merge with extra change - B side: unrelated change, F side: overwrite d with a copy (from h->i->d) - the other way
   |\
   +---o  mBF-change-m-0 merge with extra change - B side: unrelated change, F side: overwrite d with a copy (from h->i->d) - one way
   | |/
   | o  f-2: rename i -> d
   | |
   | o  f-1: rename h -> i
   | |
   o |  b-1: b update
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mBF-change-m-0")'
   M b
   A d
     h
   A t
     p
   R a
   R h
   R p
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mFB-change-m-0")'
   M b
   A d
     h
   A t
     p
   R a
   R h
   R p
   $ hg status --copies --rev 'desc("b-1")' --rev 'desc("mBF-change-m-0")'
   M d
     h (no-filelog !)
   R h
   $ hg status --copies --rev 'desc("f-2")' --rev 'desc("mBF-change-m-0")'
   M b
   M d
   $ hg status --copies --rev 'desc("f-1")' --rev 'desc("mBF-change-m-0")'
   M b
   M d
     i (no-filelog !)
   R i
   $ hg status --copies --rev 'desc("b-1")' --rev 'desc("mFB-change-m-0")'
   M d
     h (no-filelog !)
   R h
   $ hg status --copies --rev 'desc("f-2")' --rev 'desc("mFB-change-m-0")'
   M b
   M d
   $ hg status --copies --rev 'desc("f-1")' --rev 'desc("mFB-change-m-0")'
   M b
   M d
     i (no-filelog !)
   R i
 
 #if no-changeset
   $ hg log -Gfr 'desc("mBF-change-m-0")' d
   o    mBF-change-m-0 merge with extra change - B side: unrelated change, F side: overwrite d with a copy (from h->i->d) - one way
   |\
   o :  f-2: rename i -> d
   | :
   o :  f-1: rename h -> i
   :/
   o  i-0 initial commit: a b h p q r
   
 #else
 BROKEN: `hg log --follow <file>` relies on filelog metadata to work
   $ hg log -Gfr 'desc("mBF-change-m-0")' d
   o  mBF-change-m-0 merge with extra change - B side: unrelated change, F side: overwrite d with a copy (from h->i->d) - one way
   :
   o  i-2: c -move-> d, s -move-> t
   |
   ~
 #endif
 
 #if no-changeset
   $ hg log -Gfr 'desc("mFB-change-m-0")' d
   o    mFB-change-m-0 merge with extra change - B side: unrelated change, F side: overwrite d with a copy (from h->i->d) - the other way
   |\
   o :  f-2: rename i -> d
   | :
   o :  f-1: rename h -> i
   :/
   o  i-0 initial commit: a b h p q r
   
 #else
 BROKEN: `hg log --follow <file>` relies on filelog metadata to work
   $ hg log -Gfr 'desc("mFB-change-m-0")' d
   o  mFB-change-m-0 merge with extra change - B side: unrelated change, F side: overwrite d with a copy (from h->i->d) - the other way
   :
   o  i-2: c -move-> d, s -move-> t
   |
   ~
 #endif
 
 
 Subcase: restoring and untouched deleted file, while touching it
 ````````````````````````````````````````````````````````````````
 
 Merge:
 - one removing a file (d)
 - one leaving the file untouched
 - the merge actively restore the file to the same content.
 
 In this case, the file keep on living after the merge. So we should not drop its
 copy tracing chain.
 
   $ hg log -G --rev '::(desc("mCB-change-m")+desc("mBC-change-m"))'
   o    mBC-change-m-0 merge explicitely revive deleted file - B side: unrelated change, C side: delete d (restored by merge) - the other way
   |\
   +---o  mCB-change-m-0 merge explicitely revive deleted file - B side: unrelated change, C side: delete d (restored by merge) - one way
   | |/
   | o  c-1 delete d
   | |
   o |  b-1: b update
   |/
   o  i-2: c -move-> d, s -move-> t
   |
   o  i-1: a -move-> c, p -move-> s
   |
   o  i-0 initial commit: a b h p q r
   
 
 'a' is the the copy source of 'd'
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mCB-change-m-0")'
   M b
   A d
     a (no-compatibility no-changeset !)
   A t
     p
   R a
   R p
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mBC-change-m-0")'
   M b
   A d
     a (no-compatibility no-changeset !)
   A t
     p
   R a
   R p
   $ hg status --copies --rev 'desc("c-1")' --rev 'desc("mCB-change-m-0")'
   M b
   A d
   $ hg status --copies --rev 'desc("c-1")' --rev 'desc("mBC-change-m-0")'
   M b
   A d
   $ hg status --copies --rev 'desc("b-1")' --rev 'desc("mCB-change-m-0")'
   M d
   $ hg status --copies --rev 'desc("b-1")' --rev 'desc("mBC-change-m-0")'
   M d
 
 
 Decision from previous merge are properly chained with later merge
 ------------------------------------------------------------------
 
 
 Subcase: chaining conflicting rename resolution
 ```````````````````````````````````````````````
 
 The "mAEm" and "mEAm" case create a rename tracking conflict on file 'f'. We
 add more change on the respective branch and merge again. These second merge
 does not involve the file 'f' and the arbitration done within "mAEm" and "mEA"
 about that file should stay unchanged.
 
 The result from mAEm is the same for the subsequent merge:
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mAEm")' f
   A f
     a (no-changeset no-compatibility !)
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mAE,Km")' f
   A f
     a (no-changeset no-compatibility !)
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mK,AEm")' f
   A f
     a (no-changeset no-compatibility !)
 
 
 The result from mEAm is the same for the subsequent merge:
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mEAm")' f
   A f
     a (filelog !)
     b (no-changeset no-compatibility no-filelog !)
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mEA,Jm")' f
   A f
     a (filelog !)
     b (no-changeset no-compatibility no-filelog !)
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mJ,EAm")' f
   A f
     a (filelog !)
     b (no-changeset no-compatibility no-filelog !)
 
 Subcase: chaining conflicting rename resolution
 ```````````````````````````````````````````````
 
 The "mPQm" and "mQPm" case create a rename tracking conflict on file 'v'. We
 add more change on the respective branch and merge again. These second merge
 does not involve the file 'v' and the arbitration done within "mPQm" and "mQP"
 about that file should stay unchanged.
 
 The result from mPQm is the same for the subsequent merge:
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mPQm")' v
   A v
     r (filelog !)
     p (no-changeset no-compatibility no-filelog !)
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mPQ,Tm")' v
   A v
     r (filelog !)
     p (no-changeset no-compatibility no-filelog !)
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mT,PQm")' v
   A v
     r (filelog !)
     p (no-changeset no-compatibility no-filelog !)
 
 The result from mQPm is the same for the subsequent merge:
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mQPm")' v
   A v
     r (no-changeset no-compatibility !)
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mQP,Sm")' v
   A v
     r (no-changeset no-compatibility !)
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mS,QPm")' v
   A v
     r (filelog !)
     r (no-changeset no-compatibility no-filelog !)
 
 
 Subcase: chaining salvage information during a merge
 ````````````````````````````````````````````````````
 
 We add more change on the branch were the file was deleted. merging again
 should preserve the fact eh file was salvaged.
 
 reference output:
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mCB-revert-m-0")'
   M b
   A d
     a (no-changeset no-compatibility !)
   A t
     p
   R a
   R p
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mBC-revert-m-0")'
   M b
   A d
     a (no-changeset no-compatibility !)
   A t
     p
   R a
   R p
 
 chained output
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mBC+revert,Lm")'
   M b
   A d
     a (no-changeset no-compatibility !)
   A t
     p
   A unrelated-l
   R a
   R p
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mCB+revert,Lm")'
   M b
   A d
     a (no-changeset no-compatibility !)
   A t
     p
   A unrelated-l
   R a
   R p
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mL,BC+revertm")'
   M b
   A d
     a (no-changeset no-compatibility !)
   A t
     p
   A unrelated-l
   R a
   R p
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mL,CB+revertm")'
   M b
   A d
     a (no-changeset no-compatibility !)
   A t
     p
   A unrelated-l
   R a
   R p
 
 Subcase: chaining "merged" information during a merge
 ``````````````````````````````````````````````````````
 
 When a non-rename change are merged with a copy overwrite, the merge pick the copy source from (p1) as the reference. We should preserve this information in subsequent merges.
 
 
 reference output:
 
  (for details about the filelog pick, check the mFGm/mGFm case)
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mFGm")' d
   A d
     a (filelog !)
     h (no-changeset no-compatibility no-filelog !)
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mGFm")' d
   A d
     a (filelog !)
     a (no-changeset no-compatibility no-filelog !)
 
 Chained output
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mO,FGm")' d
   A d
     a (filelog !)
     h (no-changeset no-compatibility no-filelog !)
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mFG,Om")' d
   A d
     a (filelog !)
     h (no-changeset no-compatibility no-filelog !)
 
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mGF,Nm")' d
   A d
     a (no-changeset no-compatibility !)
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mN,GFm")' d
   A d
     a (no-changeset no-compatibility !)
 
 
 Subcase: chaining conflicting rename resolution, with extra change during the merge
 ```````````````````````````````````````````````````````````````````````````````````
 
 The "mAEm" and "mEAm" case create a rename tracking conflict on file 'f'. We
 add more change on the respective branch and merge again. These second merge
 does not involve the file 'f' and the arbitration done within "mAEm" and "mEA"
 about that file should stay unchanged.
 
 The result from mAEm is the same for the subsequent merge:
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mAE-change-m")' f
   A f
     a (no-changeset no-compatibility !)
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mAE-change,Km")' f
   A f
     a (no-changeset no-compatibility !)
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mK,AE-change-m")' f
   A f
     a (no-changeset no-compatibility !)
 
 
 The result from mEAm is the same for the subsequent merge:
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mEA-change-m")' f
   A f
     a (filelog !)
     b (no-changeset no-compatibility no-filelog !)
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mEA-change,Jm")' f
   A f
     a (filelog !)
     b (no-changeset no-compatibility no-filelog !)
 
   $ hg status --copies --rev 'desc("i-0")' --rev 'desc("mJ,EA-change-m")' f
   A f
     a (filelog !)
     b (no-changeset no-compatibility no-filelog !)

tests/test-copies-in-changeset.t

@@ -1,509 +1,514
 #testcases extra sidedata
 
 #if extra
   $ cat >> $HGRCPATH << EOF
   > [experimental]
   > copies.write-to=changeset-only
   > copies.read-from=changeset-only
   > [alias]
   > changesetcopies = log -r . -T 'files: {files}
   >   {extras % "{ifcontains("files", key, "{key}: {value}\n")}"}
   >   {extras % "{ifcontains("copies", key, "{key}: {value}\n")}"}'
   > EOF
 #endif
 
 #if sidedata
   $ cat >> $HGRCPATH << EOF
   > [format]
   > exp-use-copies-side-data-changeset = yes
   > EOF
 #endif
 
   $ cat >> $HGRCPATH << EOF
   > [alias]
   > showcopies = log -r . -T '{file_copies % "{source} -> {name}\n"}'
   > [extensions]
   > rebase =
   > split =
   > EOF
 
 Check that copies are recorded correctly
 
   $ hg init repo
   $ cd repo
 #if sidedata
   $ hg debugformat -v
   format-variant     repo config default
   fncache:            yes    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   persistent-nodemap:  no     no      no (no-rust !)
   persistent-nodemap: yes    yes      no (rust !)
   copies-sdc:         yes    yes      no
   revlog-v2:           no     no      no
   changelog-v2:       yes    yes      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib (no-zstd !)
   compression:        zstd   zstd    zstd (zstd !)
   compression-level:  default default default
 #else
   $ hg debugformat -v
   format-variant     repo config default
   fncache:            yes    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   persistent-nodemap:  no     no      no (no-rust !)
   persistent-nodemap: yes    yes      no (rust !)
   copies-sdc:          no     no      no
   revlog-v2:           no     no      no
   changelog-v2:        no     no      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib (no-zstd !)
   compression:        zstd   zstd    zstd (zstd !)
   compression-level:  default default default
 #endif
   $ echo a > a
   $ hg add a
   $ hg ci -m initial
   $ hg cp a b
   $ hg cp a c
   $ hg cp a d
   $ hg ci -m 'copy a to b, c, and d'
 
 #if extra
 
   $ hg changesetcopies
   files: b c d
   filesadded: 0
   1
   2
   
   p1copies: 0\x00a (esc)
   1\x00a (esc)
   2\x00a (esc)
 #else
   $ hg debugsidedata -c -v -- -1
   1 sidedata entries
    entry-0014 size 44
     '\x00\x00\x00\x04\x00\x00\x00\x00\x01\x00\x00\x00\x00\x06\x00\x00\x00\x02\x00\x00\x00\x00\x06\x00\x00\x00\x03\x00\x00\x00\x00\x06\x00\x00\x00\x04\x00\x00\x00\x00abcd'
 #endif
 
   $ hg showcopies
   a -> b
   a -> c
   a -> d
 
 #if extra
 
   $ hg showcopies --config experimental.copies.read-from=compatibility
   a -> b
   a -> c
   a -> d
   $ hg showcopies --config experimental.copies.read-from=filelog-only
 
 #endif
 
 Check that renames are recorded correctly
 
   $ hg mv b b2
   $ hg ci -m 'rename b to b2'
 
 #if extra
 
   $ hg changesetcopies
   files: b b2
   filesadded: 1
   filesremoved: 0
   
   p1copies: 1\x00b (esc)
 
 #else
   $ hg debugsidedata -c -v -- -1
   1 sidedata entries
    entry-0014 size 25
     '\x00\x00\x00\x02\x0c\x00\x00\x00\x01\x00\x00\x00\x00\x06\x00\x00\x00\x03\x00\x00\x00\x00bb2'
 #endif
 
   $ hg showcopies
   b -> b2
 
 
 Rename onto existing file. This should get recorded in the changeset files list and in the extras,
 even though there is no filelog entry.
 
   $ hg cp b2 c --force
   $ hg st --copies
   M c
     b2
 
 #if extra
 
   $ hg debugindex c
      rev linkrev nodeid       p1           p2
        0       1 b789fdd96dc2 000000000000 000000000000
 
 #else
 
   $ hg debugindex c
      rev linkrev nodeid       p1           p2
        0       1 37d9b5d994ea 000000000000 000000000000
 
 #endif
 
 
   $ hg ci -m 'move b onto d'
 
 #if extra
 
   $ hg changesetcopies
   files: c
   
   p1copies: 0\x00b2 (esc)
 
 #else
   $ hg debugsidedata -c -v -- -1
   1 sidedata entries
    entry-0014 size 25
     '\x00\x00\x00\x02\x00\x00\x00\x00\x02\x00\x00\x00\x00\x16\x00\x00\x00\x03\x00\x00\x00\x00b2c'
 #endif
 
   $ hg showcopies
   b2 -> c
 
 #if extra
 
   $ hg debugindex c
      rev linkrev nodeid       p1           p2
        0       1 b789fdd96dc2 000000000000 000000000000
 
 #else
 
   $ hg debugindex c
      rev linkrev nodeid       p1           p2
        0       1 37d9b5d994ea 000000000000 000000000000
        1       3 029625640347 000000000000 000000000000
 
 #endif
 
 Create a merge commit with copying done during merge.
 
   $ hg co 0
   0 files updated, 0 files merged, 3 files removed, 0 files unresolved
   $ hg cp a e
   $ hg cp a f
   $ hg ci -m 'copy a to e and f'
   created new head
   $ hg merge 3
   3 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
 File 'a' exists on both sides, so 'g' could be recorded as being from p1 or p2, but we currently
 always record it as being from p1
   $ hg cp a g
 File 'd' exists only in p2, so 'h' should be from p2
   $ hg cp d h
 File 'f' exists only in p1, so 'i' should be from p1
   $ hg cp f i
   $ hg ci -m 'merge'
 
 #if extra
 
   $ hg changesetcopies
   files: g h i
   filesadded: 0
   1
   2
   
   p1copies: 0\x00a (esc)
   2\x00f (esc)
   p2copies: 1\x00d (esc)
 
 #else
   $ hg debugsidedata -c -v -- -1
   1 sidedata entries
    entry-0014 size 64
     '\x00\x00\x00\x06\x00\x00\x00\x00\x01\x00\x00\x00\x00\x00\x00\x00\x00\x02\x00\x00\x00\x00\x00\x00\x00\x00\x03\x00\x00\x00\x00\x06\x00\x00\x00\x04\x00\x00\x00\x00\x07\x00\x00\x00\x05\x00\x00\x00\x01\x06\x00\x00\x00\x06\x00\x00\x00\x02adfghi'
 #endif
 
   $ hg showcopies
   a -> g
   d -> h
   f -> i
 
 Test writing to both changeset and filelog
 
   $ hg cp a j
 #if extra
   $ hg ci -m 'copy a to j' --config experimental.copies.write-to=compatibility
   $ hg changesetcopies
   files: j
   filesadded: 0
   filesremoved: 
   
   p1copies: 0\x00a (esc)
   p2copies: 
 #else
   $ hg ci -m 'copy a to j'
   $ hg debugsidedata -c -v -- -1
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x02\x00\x00\x00\x00\x01\x00\x00\x00\x00\x06\x00\x00\x00\x02\x00\x00\x00\x00aj'
 #endif
   $ hg debugdata j 0
   \x01 (esc)
   copy: a
   copyrev: b789fdd96dc2f3bd229c1dd8eedf0fc60e2b68e3
   \x01 (esc)
   a
   $ hg showcopies
   a -> j
   $ hg showcopies --config experimental.copies.read-from=compatibility
   a -> j
   $ hg showcopies --config experimental.copies.read-from=filelog-only
   a -> j
 Existing copy information in the changeset gets removed on amend and writing
 copy information on to the filelog
 #if extra
   $ hg ci --amend -m 'copy a to j, v2' \
   > --config experimental.copies.write-to=filelog-only
   saved backup bundle to $TESTTMP/repo/.hg/strip-backup/*-*-amend.hg (glob)
   $ hg changesetcopies
   files: j
   
 #else
   $ hg ci --amend -m 'copy a to j, v2'
   saved backup bundle to $TESTTMP/repo/.hg/strip-backup/*-*-amend.hg (glob)
   $ hg debugsidedata -c -v -- -1
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x02\x00\x00\x00\x00\x01\x00\x00\x00\x00\x06\x00\x00\x00\x02\x00\x00\x00\x00aj'
 #endif
   $ hg showcopies --config experimental.copies.read-from=filelog-only
   a -> j
 The entries should be written to extras even if they're empty (so the client
 won't have to fall back to reading from filelogs)
   $ echo x >> j
 #if extra
   $ hg ci -m 'modify j' --config experimental.copies.write-to=compatibility
   $ hg changesetcopies
   files: j
   filesadded: 
   filesremoved: 
   
   p1copies: 
   p2copies: 
 #else
   $ hg ci -m 'modify j'
   $ hg debugsidedata -c -v -- -1
   1 sidedata entries
    entry-0014 size 14
     '\x00\x00\x00\x01\x14\x00\x00\x00\x01\x00\x00\x00\x00j'
 #endif
 
 Test writing only to filelog
 
   $ hg cp a k
 #if extra
   $ hg ci -m 'copy a to k' --config experimental.copies.write-to=filelog-only
 
   $ hg changesetcopies
   files: k
   
 #else
   $ hg ci -m 'copy a to k'
   $ hg debugsidedata -c -v -- -1
   1 sidedata entries
    entry-0014 size 24
     '\x00\x00\x00\x02\x00\x00\x00\x00\x01\x00\x00\x00\x00\x06\x00\x00\x00\x02\x00\x00\x00\x00ak'
 #endif
 
   $ hg debugdata k 0
   \x01 (esc)
   copy: a
   copyrev: b789fdd96dc2f3bd229c1dd8eedf0fc60e2b68e3
   \x01 (esc)
   a
 #if extra
   $ hg showcopies
 
   $ hg showcopies --config experimental.copies.read-from=compatibility
   a -> k
   $ hg showcopies --config experimental.copies.read-from=filelog-only
   a -> k
 #else
   $ hg showcopies
   a -> k
 #endif
 
   $ cd ..
 
 Test rebasing a commit with copy information
 
   $ hg init rebase-rename
   $ cd rebase-rename
   $ echo a > a
   $ hg ci -Aqm 'add a'
   $ echo a2 > a
   $ hg ci -m 'modify a'
   $ hg co -q 0
   $ hg mv a b
   $ hg ci -qm 'rename a to b'
 Not only do we want this to run in-memory, it shouldn't fall back to
 on-disk merge (no conflicts), so we force it to be in-memory
 with no fallback.
   $ hg rebase -d 1 --config rebase.experimental.inmemory=yes --config devel.rebase.force-in-memory-merge=yes
   rebasing 2:* tip "rename a to b" (glob)
   merging a and b to b
   saved backup bundle to $TESTTMP/rebase-rename/.hg/strip-backup/*-*-rebase.hg (glob)
   $ hg st --change . --copies
   A b
     a
   R a
   $ cd ..
 
 Test splitting a commit
 
   $ hg init split
   $ cd split
   $ echo a > a
   $ echo b > b
   $ hg ci -Aqm 'add a and b'
   $ echo a2 > a
   $ hg mv b c
   $ hg ci -m 'modify a, move b to c'
   $ hg --config ui.interactive=yes split <<EOF
   > y
   > y
   > n
   > y
   > EOF
   diff --git a/a b/a
   1 hunks, 1 lines changed
   examine changes to 'a'?
   (enter ? for help) [Ynesfdaq?] y
   
   @@ -1,1 +1,1 @@
   -a
   +a2
   record this change to 'a'?
   (enter ? for help) [Ynesfdaq?] y
   
   diff --git a/b b/c
   rename from b
   rename to c
   examine changes to 'b' and 'c'?
   (enter ? for help) [Ynesfdaq?] n
   
   created new head
   diff --git a/b b/c
   rename from b
   rename to c
   examine changes to 'b' and 'c'?
   (enter ? for help) [Ynesfdaq?] y
   
   saved backup bundle to $TESTTMP/split/.hg/strip-backup/*-*-split.hg (glob)
   $ cd ..
 
 Test committing half a rename
 
   $ hg init partial
   $ cd partial
   $ echo a > a
   $ hg ci -Aqm 'add a'
   $ hg mv a b
   $ hg ci -m 'remove a' a
 
 #if sidedata
 
 Test upgrading/downgrading to sidedata storage
 ==============================================
 
 downgrading
 
   $ hg debugformat -v
   format-variant     repo config default
   fncache:            yes    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   persistent-nodemap:  no     no      no (no-rust !)
   persistent-nodemap: yes    yes      no (rust !)
   copies-sdc:         yes    yes      no
   revlog-v2:           no     no      no
   changelog-v2:       yes    yes      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib (no-zstd !)
   compression:        zstd   zstd    zstd (zstd !)
   compression-level:  default default default
   $ hg debugsidedata -c -- 0
   1 sidedata entries
    entry-0014 size 14
   $ hg debugsidedata -c -- 1
   1 sidedata entries
    entry-0014 size 14
   $ hg debugsidedata -m -- 0
   $ cat << EOF > .hg/hgrc
   > [format]
   > exp-use-copies-side-data-changeset = no
   > [experimental]
   > revlogv2 = enable-unstable-format-and-corrupt-my-data
   > EOF
   $ hg debugupgraderepo --run --quiet --no-backup > /dev/null
   $ hg debugformat -v
   format-variant     repo config default
   fncache:            yes    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   persistent-nodemap:  no     no      no (no-rust !)
   persistent-nodemap: yes    yes      no (rust !)
   copies-sdc:          no     no      no
   revlog-v2:          yes    yes      no
   changelog-v2:        no     no      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib (no-zstd !)
   compression:        zstd   zstd    zstd (zstd !)
   compression-level:  default default default
   $ hg debugsidedata -c -- 0
   $ hg debugsidedata -c -- 1
   $ hg debugsidedata -m -- 0
 
 upgrading
 
   $ cat << EOF > .hg/hgrc
   > [format]
   > exp-use-copies-side-data-changeset = yes
   > EOF
   $ hg debugupgraderepo --run --quiet --no-backup > /dev/null
   $ hg debugformat -v
   format-variant     repo config default
   fncache:            yes    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   persistent-nodemap:  no     no      no (no-rust !)
   persistent-nodemap: yes    yes      no (rust !)
   copies-sdc:         yes    yes      no
   revlog-v2:           no     no      no
   changelog-v2:       yes    yes      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib (no-zstd !)
   compression:        zstd   zstd    zstd (zstd !)
   compression-level:  default default default
   $ hg debugsidedata -c -- 0
   1 sidedata entries
    entry-0014 size 14
   $ hg debugsidedata -c -- 1
   1 sidedata entries
    entry-0014 size 14
   $ hg debugsidedata -m -- 0
 
 #endif
 
   $ cd ..

tests/test-persistent-nodemap.t

@@ -1,1059 +1,1062
 ===================================
 Test the persistent on-disk nodemap
 ===================================
 
 
 #if no-rust
 
   $ cat << EOF >> $HGRCPATH
   > [format]
   > use-persistent-nodemap=yes
   > [devel]
   > persistent-nodemap=yes
   > EOF
 
 #endif
 
   $ hg init test-repo --config storage.revlog.persistent-nodemap.slow-path=allow
   $ cd test-repo
 
 Check handling of the default slow-path value
 
 #if no-pure no-rust
 
   $ hg id
   abort: accessing `persistent-nodemap` repository without associated fast implementation.
   (check `hg help config.format.use-persistent-nodemap` for details)
   [255]
 
 Unlock further check (we are here to test the feature)
 
   $ cat << EOF >> $HGRCPATH
   > [storage]
   > # to avoid spamming the test
   > revlog.persistent-nodemap.slow-path=allow
   > EOF
 
 #endif
 
 #if rust
 
 Regression test for a previous bug in Rust/C FFI for the `Revlog_CAPI` capsule:
 in places where `mercurial/cext/revlog.c` function signatures use `Py_ssize_t`
 (64 bits on Linux x86_64), corresponding declarations in `rust/hg-cpython/src/cindex.rs`
 incorrectly used `libc::c_int` (32 bits).
 As a result, -1 passed from Rust for the null revision became 4294967295 in C.
 
   $ hg log -r 00000000
   changeset:   -1:000000000000
   tag:         tip
   user:        
   date:        Thu Jan 01 00:00:00 1970 +0000
   
 
 #endif
 
 
   $ hg debugformat
   format-variant     repo
   fncache:            yes
+  dirstate-v2:         no
   dotencode:          yes
   generaldelta:       yes
   share-safe:          no
   sparserevlog:       yes
   persistent-nodemap: yes
   copies-sdc:          no
   revlog-v2:           no
   changelog-v2:        no
   plain-cl-delta:     yes
   compression:        zlib (no-zstd !)
   compression:        zstd (zstd !)
   compression-level:  default
   $ hg debugbuilddag .+5000 --new-file
 
   $ hg debugnodemap --metadata
   uid: ???????? (glob)
   tip-rev: 5000
   tip-node: 6b02b8c7b96654c25e86ba69eda198d7e6ad8b3c
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
   $ f --size .hg/store/00changelog.n
   .hg/store/00changelog.n: size=62
 
 Simple lookup works
 
   $ ANYNODE=`hg log --template '{node|short}\n' --rev tip`
   $ hg log -r "$ANYNODE" --template '{rev}\n'
   5000
 
 
 #if rust
 
   $ f --sha256 .hg/store/00changelog-*.nd
   .hg/store/00changelog-????????.nd: sha256=2e029d3200bd1a986b32784fc2ef1a3bd60dc331f025718bcf5ff44d93f026fd (glob)
 
   $ f --sha256 .hg/store/00manifest-*.nd
   .hg/store/00manifest-????????.nd: sha256=97117b1c064ea2f86664a124589e47db0e254e8d34739b5c5cc5bf31c9da2b51 (glob)
   $ hg debugnodemap --dump-new | f --sha256 --size
   size=121088, sha256=2e029d3200bd1a986b32784fc2ef1a3bd60dc331f025718bcf5ff44d93f026fd
   $ hg debugnodemap --dump-disk | f --sha256 --bytes=256 --hexdump --size
   size=121088, sha256=2e029d3200bd1a986b32784fc2ef1a3bd60dc331f025718bcf5ff44d93f026fd
   0000: 00 00 00 91 00 00 00 20 00 00 00 bb 00 00 00 e7 |....... ........|
   0010: 00 00 00 66 00 00 00 a1 00 00 01 13 00 00 01 22 |...f..........."|
   0020: 00 00 00 23 00 00 00 fc 00 00 00 ba 00 00 00 5e |...#...........^|
   0030: 00 00 00 df 00 00 01 4e 00 00 01 65 00 00 00 ab |.......N...e....|
   0040: 00 00 00 a9 00 00 00 95 00 00 00 73 00 00 00 38 |...........s...8|
   0050: 00 00 00 cc 00 00 00 92 00 00 00 90 00 00 00 69 |...............i|
   0060: 00 00 00 ec 00 00 00 8d 00 00 01 4f 00 00 00 12 |...........O....|
   0070: 00 00 02 0c 00 00 00 77 00 00 00 9c 00 00 00 8f |.......w........|
   0080: 00 00 00 d5 00 00 00 6b 00 00 00 48 00 00 00 b3 |.......k...H....|
   0090: 00 00 00 e5 00 00 00 b5 00 00 00 8e 00 00 00 ad |................|
   00a0: 00 00 00 7b 00 00 00 7c 00 00 00 0b 00 00 00 2b |...{...|.......+|
   00b0: 00 00 00 c6 00 00 00 1e 00 00 01 08 00 00 00 11 |................|
   00c0: 00 00 01 30 00 00 00 26 00 00 01 9c 00 00 00 35 |...0...&.......5|
   00d0: 00 00 00 b8 00 00 01 31 00 00 00 2c 00 00 00 55 |.......1...,...U|
   00e0: 00 00 00 8a 00 00 00 9a 00 00 00 0c 00 00 01 1e |................|
   00f0: 00 00 00 a4 00 00 00 83 00 00 00 c9 00 00 00 8c |................|
 
 
 #else
 
   $ f --sha256 .hg/store/00changelog-*.nd
   .hg/store/00changelog-????????.nd: sha256=f544f5462ff46097432caf6d764091f6d8c46d6121be315ead8576d548c9dd79 (glob)
   $ hg debugnodemap --dump-new | f --sha256 --size
   size=121088, sha256=f544f5462ff46097432caf6d764091f6d8c46d6121be315ead8576d548c9dd79
   $ hg debugnodemap --dump-disk | f --sha256 --bytes=256 --hexdump --size
   size=121088, sha256=f544f5462ff46097432caf6d764091f6d8c46d6121be315ead8576d548c9dd79
   0000: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff |................|
   0010: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff |................|
   0020: ff ff ff ff ff ff f5 06 ff ff ff ff ff ff f3 e7 |................|
   0030: ff ff ef ca ff ff ff ff ff ff ff ff ff ff ff ff |................|
   0040: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff |................|
   0050: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ed 08 |................|
   0060: ff ff ed 66 ff ff ff ff ff ff ff ff ff ff ff ff |...f............|
   0070: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff |................|
   0080: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff |................|
   0090: ff ff ff ff ff ff ff ff ff ff ff ff ff ff f6 ed |................|
   00a0: ff ff ff ff ff ff fe 61 ff ff ff ff ff ff ff ff |.......a........|
   00b0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff |................|
   00c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff |................|
   00d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff |................|
   00e0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff f1 02 |................|
   00f0: ff ff ff ff ff ff ed 1b ff ff ff ff ff ff ff ff |................|
 
 #endif
 
   $ hg debugnodemap --check
   revision in index:   5001
   revision in nodemap: 5001
 
 add a new commit
 
   $ hg up
   5001 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ echo foo > foo
   $ hg add foo
 
 
 Check slow-path config value handling
 -------------------------------------
 
 #if no-pure no-rust
 
   $ hg id --config "storage.revlog.persistent-nodemap.slow-path=invalid-value"
   unknown value for config "storage.revlog.persistent-nodemap.slow-path": "invalid-value"
   falling back to default value: abort
   abort: accessing `persistent-nodemap` repository without associated fast implementation.
   (check `hg help config.format.use-persistent-nodemap` for details)
   [255]
 
   $ hg log -r . --config "storage.revlog.persistent-nodemap.slow-path=warn"
   warning: accessing `persistent-nodemap` repository without associated fast implementation.
   (check `hg help config.format.use-persistent-nodemap` for details)
   changeset:   5000:6b02b8c7b966
   tag:         tip
   user:        debugbuilddag
   date:        Thu Jan 01 01:23:20 1970 +0000
   summary:     r5000
   
   $ hg ci -m 'foo' --config "storage.revlog.persistent-nodemap.slow-path=abort"
   abort: accessing `persistent-nodemap` repository without associated fast implementation.
   (check `hg help config.format.use-persistent-nodemap` for details)
   [255]
 
 #else
 
   $ hg id --config "storage.revlog.persistent-nodemap.slow-path=invalid-value"
   unknown value for config "storage.revlog.persistent-nodemap.slow-path": "invalid-value"
   falling back to default value: abort
   6b02b8c7b966+ tip
 
 #endif
 
   $ hg ci -m 'foo'
 
 #if no-pure no-rust
   $ hg debugnodemap --metadata
   uid: ???????? (glob)
   tip-rev: 5001
   tip-node: 16395c3cf7e231394735e6b1717823ada303fb0c
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
 #else
   $ hg debugnodemap --metadata
   uid: ???????? (glob)
   tip-rev: 5001
   tip-node: 16395c3cf7e231394735e6b1717823ada303fb0c
   data-length: 121344
   data-unused: 256
   data-unused: 0.211%
 #endif
 
   $ f --size .hg/store/00changelog.n
   .hg/store/00changelog.n: size=62
 
 (The pure code use the debug code that perform incremental update, the C code reencode from scratch)
 
 #if pure
   $ f --sha256 .hg/store/00changelog-*.nd --size
   .hg/store/00changelog-????????.nd: size=121344, sha256=cce54c5da5bde3ad72a4938673ed4064c86231b9c64376b082b163fdb20f8f66 (glob)
 #endif
 
 #if rust
   $ f --sha256 .hg/store/00changelog-*.nd --size
   .hg/store/00changelog-????????.nd: size=121344, sha256=952b042fcf614ceb37b542b1b723e04f18f83efe99bee4e0f5ccd232ef470e58 (glob)
 #endif
 
 #if no-pure no-rust
   $ f --sha256 .hg/store/00changelog-*.nd --size
   .hg/store/00changelog-????????.nd: size=121088, sha256=df7c06a035b96cb28c7287d349d603baef43240be7736fe34eea419a49702e17 (glob)
 #endif
 
   $ hg debugnodemap --check
   revision in index:   5002
   revision in nodemap: 5002
 
 Test code path without mmap
 ---------------------------
 
   $ echo bar > bar
   $ hg add bar
   $ hg ci -m 'bar' --config storage.revlog.persistent-nodemap.mmap=no
 
   $ hg debugnodemap --check --config storage.revlog.persistent-nodemap.mmap=yes
   revision in index:   5003
   revision in nodemap: 5003
   $ hg debugnodemap --check --config storage.revlog.persistent-nodemap.mmap=no
   revision in index:   5003
   revision in nodemap: 5003
 
 
 #if pure
   $ hg debugnodemap --metadata
   uid: ???????? (glob)
   tip-rev: 5002
   tip-node: 880b18d239dfa9f632413a2071bfdbcc4806a4fd
   data-length: 121600
   data-unused: 512
   data-unused: 0.421%
   $ f --sha256 .hg/store/00changelog-*.nd --size
   .hg/store/00changelog-????????.nd: size=121600, sha256=def52503d049ccb823974af313a98a935319ba61f40f3aa06a8be4d35c215054 (glob)
 #endif
 #if rust
   $ hg debugnodemap --metadata
   uid: ???????? (glob)
   tip-rev: 5002
   tip-node: 880b18d239dfa9f632413a2071bfdbcc4806a4fd
   data-length: 121600
   data-unused: 512
   data-unused: 0.421%
   $ f --sha256 .hg/store/00changelog-*.nd --size
   .hg/store/00changelog-????????.nd: size=121600, sha256=dacf5b5f1d4585fee7527d0e67cad5b1ba0930e6a0928f650f779aefb04ce3fb (glob)
 #endif
 #if no-pure no-rust
   $ hg debugnodemap --metadata
   uid: ???????? (glob)
   tip-rev: 5002
   tip-node: 880b18d239dfa9f632413a2071bfdbcc4806a4fd
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
   $ f --sha256 .hg/store/00changelog-*.nd --size
   .hg/store/00changelog-????????.nd: size=121088, sha256=59fcede3e3cc587755916ceed29e3c33748cd1aa7d2f91828ac83e7979d935e8 (glob)
 #endif
 
 Test force warming the cache
 
   $ rm .hg/store/00changelog.n
   $ hg debugnodemap --metadata
   $ hg debugupdatecache
 #if pure
   $ hg debugnodemap --metadata
   uid: ???????? (glob)
   tip-rev: 5002
   tip-node: 880b18d239dfa9f632413a2071bfdbcc4806a4fd
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
 #else
   $ hg debugnodemap --metadata
   uid: ???????? (glob)
   tip-rev: 5002
   tip-node: 880b18d239dfa9f632413a2071bfdbcc4806a4fd
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
 #endif
 
 Check out of sync nodemap
 =========================
 
 First copy old data on the side.
 
   $ mkdir ../tmp-copies
   $ cp .hg/store/00changelog-????????.nd .hg/store/00changelog.n ../tmp-copies
 
 Nodemap lagging behind
 ----------------------
 
 make a new commit
 
   $ echo bar2 > bar
   $ hg ci -m 'bar2'
   $ NODE=`hg log -r tip -T '{node}\n'`
   $ hg log -r "$NODE" -T '{rev}\n'
   5003
 
 If the nodemap is lagging behind, it can catch up fine
 
   $ hg debugnodemap --metadata
   uid: ???????? (glob)
   tip-rev: 5003
   tip-node: c9329770f979ade2d16912267c38ba5f82fd37b3
   data-length: 121344 (pure !)
   data-length: 121344 (rust !)
   data-length: 121152 (no-rust no-pure !)
   data-unused: 192 (pure !)
   data-unused: 192 (rust !)
   data-unused: 0 (no-rust no-pure !)
   data-unused: 0.158% (pure !)
   data-unused: 0.158% (rust !)
   data-unused: 0.000% (no-rust no-pure !)
   $ cp -f ../tmp-copies/* .hg/store/
   $ hg debugnodemap --metadata
   uid: ???????? (glob)
   tip-rev: 5002
   tip-node: 880b18d239dfa9f632413a2071bfdbcc4806a4fd
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
   $ hg log -r "$NODE" -T '{rev}\n'
   5003
 
 changelog altered
 -----------------
 
 If the nodemap is not gated behind a requirements, an unaware client can alter
 the repository so the revlog used to generate the nodemap is not longer
 compatible with the persistent nodemap. We need to detect that.
 
   $ hg up "$NODE~5"
   0 files updated, 0 files merged, 4 files removed, 0 files unresolved
   $ echo bar > babar
   $ hg add babar
   $ hg ci -m 'babar'
   created new head
   $ OTHERNODE=`hg log -r tip -T '{node}\n'`
   $ hg log -r "$OTHERNODE" -T '{rev}\n'
   5004
 
   $ hg --config extensions.strip= strip --rev "$NODE~1" --no-backup
 
 the nodemap should detect the changelog have been tampered with and recover.
 
   $ hg debugnodemap --metadata
   uid: ???????? (glob)
   tip-rev: 5002
   tip-node: b355ef8adce0949b8bdf6afc72ca853740d65944
   data-length: 121536 (pure !)
   data-length: 121088 (rust !)
   data-length: 121088 (no-pure no-rust !)
   data-unused: 448 (pure !)
   data-unused: 0 (rust !)
   data-unused: 0 (no-pure no-rust !)
   data-unused: 0.000% (rust !)
   data-unused: 0.369% (pure !)
   data-unused: 0.000% (no-pure no-rust !)
 
   $ cp -f ../tmp-copies/* .hg/store/
   $ hg debugnodemap --metadata
   uid: ???????? (glob)
   tip-rev: 5002
   tip-node: 880b18d239dfa9f632413a2071bfdbcc4806a4fd
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
   $ hg log -r "$OTHERNODE" -T '{rev}\n'
   5002
 
 missing data file
 -----------------
 
   $ UUID=`hg debugnodemap --metadata| grep 'uid:' | \
   > sed 's/uid: //'`
   $ FILE=.hg/store/00changelog-"${UUID}".nd
   $ mv $FILE ../tmp-data-file
   $ cp .hg/store/00changelog.n ../tmp-docket
 
 mercurial don't crash
 
   $ hg log -r .
   changeset:   5002:b355ef8adce0
   tag:         tip
   parent:      4998:d918ad6d18d3
   user:        test
   date:        Thu Jan 01 00:00:00 1970 +0000
   summary:     babar
   
   $ hg debugnodemap --metadata
 
   $ hg debugupdatecache
   $ hg debugnodemap --metadata
   uid: * (glob)
   tip-rev: 5002
   tip-node: b355ef8adce0949b8bdf6afc72ca853740d65944
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
   $ mv ../tmp-data-file $FILE
   $ mv ../tmp-docket .hg/store/00changelog.n
 
 Check transaction related property
 ==================================
 
 An up to date nodemap should be available to shell hooks,
 
   $ echo dsljfl > a
   $ hg add a
   $ hg ci -m a
   $ hg debugnodemap --metadata
   uid: ???????? (glob)
   tip-rev: 5003
   tip-node: a52c5079765b5865d97b993b303a18740113bbb2
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
   $ echo babar2 > babar
   $ hg ci -m 'babar2' --config "hooks.pretxnclose.nodemap-test=hg debugnodemap --metadata"
   uid: ???????? (glob)
   tip-rev: 5004
   tip-node: 2f5fb1c06a16834c5679d672e90da7c5f3b1a984
   data-length: 121280 (pure !)
   data-length: 121280 (rust !)
   data-length: 121088 (no-pure no-rust !)
   data-unused: 192 (pure !)
   data-unused: 192 (rust !)
   data-unused: 0 (no-pure no-rust !)
   data-unused: 0.158% (pure !)
   data-unused: 0.158% (rust !)
   data-unused: 0.000% (no-pure no-rust !)
   $ hg debugnodemap --metadata
   uid: ???????? (glob)
   tip-rev: 5004
   tip-node: 2f5fb1c06a16834c5679d672e90da7c5f3b1a984
   data-length: 121280 (pure !)
   data-length: 121280 (rust !)
   data-length: 121088 (no-pure no-rust !)
   data-unused: 192 (pure !)
   data-unused: 192 (rust !)
   data-unused: 0 (no-pure no-rust !)
   data-unused: 0.158% (pure !)
   data-unused: 0.158% (rust !)
   data-unused: 0.000% (no-pure no-rust !)
 
 Another process does not see the pending nodemap content during run.
 
   $ echo qpoasp > a
   $ hg ci -m a2 \
   > --config "hooks.pretxnclose=sh \"$RUNTESTDIR/testlib/wait-on-file\" 20 sync-repo-read sync-txn-pending" \
   > --config "hooks.txnclose=touch sync-txn-close" > output.txt 2>&1 &
 
 (read the repository while the commit transaction is pending)
 
   $ sh "$RUNTESTDIR/testlib/wait-on-file" 20 sync-txn-pending && \
   > hg debugnodemap --metadata && \
   > sh "$RUNTESTDIR/testlib/wait-on-file" 20 sync-txn-close sync-repo-read
   uid: ???????? (glob)
   tip-rev: 5004
   tip-node: 2f5fb1c06a16834c5679d672e90da7c5f3b1a984
   data-length: 121280 (pure !)
   data-length: 121280 (rust !)
   data-length: 121088 (no-pure no-rust !)
   data-unused: 192 (pure !)
   data-unused: 192 (rust !)
   data-unused: 0 (no-pure no-rust !)
   data-unused: 0.158% (pure !)
   data-unused: 0.158% (rust !)
   data-unused: 0.000% (no-pure no-rust !)
   $ hg debugnodemap --metadata
   uid: ???????? (glob)
   tip-rev: 5005
   tip-node: 90d5d3ba2fc47db50f712570487cb261a68c8ffe
   data-length: 121536 (pure !)
   data-length: 121536 (rust !)
   data-length: 121088 (no-pure no-rust !)
   data-unused: 448 (pure !)
   data-unused: 448 (rust !)
   data-unused: 0 (no-pure no-rust !)
   data-unused: 0.369% (pure !)
   data-unused: 0.369% (rust !)
   data-unused: 0.000% (no-pure no-rust !)
 
   $ cat output.txt
 
 Check that a failing transaction will properly revert the data
 
   $ echo plakfe > a
   $ f --size --sha256 .hg/store/00changelog-*.nd
   .hg/store/00changelog-????????.nd: size=121536, sha256=bb414468d225cf52d69132e1237afba34d4346ee2eb81b505027e6197b107f03 (glob) (pure !)
   .hg/store/00changelog-????????.nd: size=121536, sha256=909ac727bc4d1c0fda5f7bff3c620c98bd4a2967c143405a1503439e33b377da (glob) (rust !)
   .hg/store/00changelog-????????.nd: size=121088, sha256=342d36d30d86dde67d3cb6c002606c4a75bcad665595d941493845066d9c8ee0 (glob) (no-pure no-rust !)
   $ hg ci -m a3 --config "extensions.abort=$RUNTESTDIR/testlib/crash_transaction_late.py"
   transaction abort!
   rollback completed
   abort: This is a late abort
   [255]
   $ hg debugnodemap --metadata
   uid: ???????? (glob)
   tip-rev: 5005
   tip-node: 90d5d3ba2fc47db50f712570487cb261a68c8ffe
   data-length: 121536 (pure !)
   data-length: 121536 (rust !)
   data-length: 121088 (no-pure no-rust !)
   data-unused: 448 (pure !)
   data-unused: 448 (rust !)
   data-unused: 0 (no-pure no-rust !)
   data-unused: 0.369% (pure !)
   data-unused: 0.369% (rust !)
   data-unused: 0.000% (no-pure no-rust !)
   $ f --size --sha256 .hg/store/00changelog-*.nd
   .hg/store/00changelog-????????.nd: size=121536, sha256=bb414468d225cf52d69132e1237afba34d4346ee2eb81b505027e6197b107f03 (glob) (pure !)
   .hg/store/00changelog-????????.nd: size=121536, sha256=909ac727bc4d1c0fda5f7bff3c620c98bd4a2967c143405a1503439e33b377da (glob) (rust !)
   .hg/store/00changelog-????????.nd: size=121088, sha256=342d36d30d86dde67d3cb6c002606c4a75bcad665595d941493845066d9c8ee0 (glob) (no-pure no-rust !)
 
 Check that removing content does not confuse the nodemap
 --------------------------------------------------------
 
 removing data with rollback
 
   $ echo aso > a
   $ hg ci -m a4
   $ hg rollback
   repository tip rolled back to revision 5005 (undo commit)
   working directory now based on revision 5005
   $ hg id -r .
   90d5d3ba2fc4 tip
 
 roming data with strip
 
   $ echo aso > a
   $ hg ci -m a4
   $ hg --config extensions.strip= strip -r . --no-backup
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg id -r . --traceback
   90d5d3ba2fc4 tip
 
 Test upgrade / downgrade
 ========================
 
 downgrading
 
   $ cat << EOF >> .hg/hgrc
   > [format]
   > use-persistent-nodemap=no
   > EOF
   $ hg debugformat -v
   format-variant     repo config default
   fncache:            yes    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   persistent-nodemap: yes     no      no
   copies-sdc:          no     no      no
   revlog-v2:           no     no      no
   changelog-v2:        no     no      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib (no-zstd !)
   compression:        zstd   zstd    zstd (zstd !)
   compression-level:  default default default
   $ hg debugupgraderepo --run --no-backup
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, sparserevlog, store (no-zstd no-dirstate-v2 !)
      preserved: dotencode, fncache, generaldelta, revlog-compression-zstd, revlogv1, sparserevlog, store (zstd no-dirstate-v2 !)
      preserved: dotencode, exp-dirstate-v2, fncache, generaldelta, revlog-compression-zstd, revlogv1, sparserevlog, store (zstd dirstate-v2 !)
      removed: persistent-nodemap
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   beginning upgrade...
   repository locked and read-only
   creating temporary repository to stage upgraded data: $TESTTMP/test-repo/.hg/upgrade.* (glob)
   (it is safe to interrupt this process any time before data migration completes)
   downgrading repository to not use persistent nodemap feature
   removing temporary repository $TESTTMP/test-repo/.hg/upgrade.* (glob)
   $ ls -1 .hg/store/ | egrep '00(changelog|manifest)(\.n|-.*\.nd)'
   00changelog-*.nd (glob)
   00manifest-*.nd (glob)
   undo.backup.00changelog.n
   undo.backup.00manifest.n
   $ hg debugnodemap --metadata
 
 
 upgrading
 
   $ cat << EOF >> .hg/hgrc
   > [format]
   > use-persistent-nodemap=yes
   > EOF
   $ hg debugformat -v
   format-variant     repo config default
   fncache:            yes    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   persistent-nodemap:  no    yes      no
   copies-sdc:          no     no      no
   revlog-v2:           no     no      no
   changelog-v2:        no     no      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib (no-zstd !)
   compression:        zstd   zstd    zstd (zstd !)
   compression-level:  default default default
   $ hg debugupgraderepo --run --no-backup
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, sparserevlog, store (no-zstd no-dirstate-v2 !)
      preserved: dotencode, fncache, generaldelta, revlog-compression-zstd, revlogv1, sparserevlog, store (zstd no-dirstate-v2 !)
      preserved: dotencode, exp-dirstate-v2, fncache, generaldelta, revlog-compression-zstd, revlogv1, sparserevlog, store (zstd dirstate-v2 !)
      added: persistent-nodemap
   
   persistent-nodemap
      Speedup revision lookup by node id.
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   beginning upgrade...
   repository locked and read-only
   creating temporary repository to stage upgraded data: $TESTTMP/test-repo/.hg/upgrade.* (glob)
   (it is safe to interrupt this process any time before data migration completes)
   upgrading repository to use persistent nodemap feature
   removing temporary repository $TESTTMP/test-repo/.hg/upgrade.* (glob)
   $ ls -1 .hg/store/ | egrep '00(changelog|manifest)(\.n|-.*\.nd)'
   00changelog-*.nd (glob)
   00changelog.n
   00manifest-*.nd (glob)
   00manifest.n
   undo.backup.00changelog.n
   undo.backup.00manifest.n
 
   $ hg debugnodemap --metadata
   uid: * (glob)
   tip-rev: 5005
   tip-node: 90d5d3ba2fc47db50f712570487cb261a68c8ffe
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
 
 Running unrelated upgrade
 
   $ hg debugupgraderepo --run --no-backup --quiet --optimize re-delta-all
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlogv1, sparserevlog, store (no-zstd no-dirstate-v2 !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlog-compression-zstd, revlogv1, sparserevlog, store (zstd no-dirstate-v2 !)
      preserved: dotencode, exp-dirstate-v2, fncache, generaldelta, persistent-nodemap, revlog-compression-zstd, revlogv1, sparserevlog, store (zstd dirstate-v2 !)
   
   optimisations: re-delta-all
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   $ ls -1 .hg/store/ | egrep '00(changelog|manifest)(\.n|-.*\.nd)'
   00changelog-*.nd (glob)
   00changelog.n
   00manifest-*.nd (glob)
   00manifest.n
 
   $ hg debugnodemap --metadata
   uid: * (glob)
   tip-rev: 5005
   tip-node: 90d5d3ba2fc47db50f712570487cb261a68c8ffe
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
 
 Persistent nodemap and local/streaming clone
 ============================================
 
   $ cd ..
 
 standard clone
 --------------
 
 The persistent nodemap should exist after a streaming clone
 
   $ hg clone --pull --quiet -U test-repo standard-clone
   $ ls -1 standard-clone/.hg/store/ | egrep '00(changelog|manifest)(\.n|-.*\.nd)'
   00changelog-*.nd (glob)
   00changelog.n
   00manifest-*.nd (glob)
   00manifest.n
   $ hg -R standard-clone debugnodemap --metadata
   uid: * (glob)
   tip-rev: 5005
   tip-node: 90d5d3ba2fc47db50f712570487cb261a68c8ffe
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
 
 
 local clone
 ------------
 
 The persistent nodemap should exist after a streaming clone
 
   $ hg clone -U test-repo local-clone
   $ ls -1 local-clone/.hg/store/ | egrep '00(changelog|manifest)(\.n|-.*\.nd)'
   00changelog-*.nd (glob)
   00changelog.n
   00manifest-*.nd (glob)
   00manifest.n
   $ hg -R local-clone debugnodemap --metadata
   uid: * (glob)
   tip-rev: 5005
   tip-node: 90d5d3ba2fc47db50f712570487cb261a68c8ffe
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
 
 Test various corruption case
 ============================
 
 Missing datafile
 ----------------
 
 Test behavior with a missing datafile
 
   $ hg clone --quiet --pull test-repo corruption-test-repo
   $ ls -1 corruption-test-repo/.hg/store/00changelog*
   corruption-test-repo/.hg/store/00changelog-*.nd (glob)
   corruption-test-repo/.hg/store/00changelog.d
   corruption-test-repo/.hg/store/00changelog.i
   corruption-test-repo/.hg/store/00changelog.n
   $ rm corruption-test-repo/.hg/store/00changelog*.nd
   $ hg log -R corruption-test-repo -r .
   changeset:   5005:90d5d3ba2fc4
   tag:         tip
   user:        test
   date:        Thu Jan 01 00:00:00 1970 +0000
   summary:     a2
   
   $ ls -1 corruption-test-repo/.hg/store/00changelog*
   corruption-test-repo/.hg/store/00changelog.d
   corruption-test-repo/.hg/store/00changelog.i
   corruption-test-repo/.hg/store/00changelog.n
 
 Truncated data file
 -------------------
 
 Test behavior with a too short datafile
 
 rebuild the missing data
   $ hg -R corruption-test-repo debugupdatecache
   $ ls -1 corruption-test-repo/.hg/store/00changelog*
   corruption-test-repo/.hg/store/00changelog-*.nd (glob)
   corruption-test-repo/.hg/store/00changelog.d
   corruption-test-repo/.hg/store/00changelog.i
   corruption-test-repo/.hg/store/00changelog.n
 
 truncate the file
 
   $ datafilepath=`ls corruption-test-repo/.hg/store/00changelog*.nd`
   $ f -s $datafilepath
   corruption-test-repo/.hg/store/00changelog-*.nd: size=121088 (glob)
   $ dd if=$datafilepath bs=1000 count=10 of=$datafilepath-tmp status=noxfer
   10+0 records in
   10+0 records out
   $ mv $datafilepath-tmp $datafilepath
   $ f -s $datafilepath
   corruption-test-repo/.hg/store/00changelog-*.nd: size=10000 (glob)
 
 Check that Mercurial reaction to this event
 
   $ hg -R corruption-test-repo log -r . --traceback
   changeset:   5005:90d5d3ba2fc4
   tag:         tip
   user:        test
   date:        Thu Jan 01 00:00:00 1970 +0000
   summary:     a2
   
 
 
 stream clone
 ============
 
 The persistent nodemap should exist after a streaming clone
 
 Simple case
 -----------
 
 No race condition
 
   $ hg clone -U --stream --config ui.ssh="\"$PYTHON\" \"$TESTDIR/dummyssh\"" ssh://user@dummy/test-repo stream-clone --debug | egrep '00(changelog|manifest)'
   adding [s] 00manifest.n (62 bytes)
   adding [s] 00manifest-*.nd (118 KB) (glob)
   adding [s] 00changelog.n (62 bytes)
   adding [s] 00changelog-*.nd (118 KB) (glob)
   adding [s] 00manifest.d (452 KB) (no-zstd !)
   adding [s] 00manifest.d (491 KB) (zstd !)
   adding [s] 00changelog.d (360 KB) (no-zstd !)
   adding [s] 00changelog.d (368 KB) (zstd !)
   adding [s] 00manifest.i (313 KB)
   adding [s] 00changelog.i (313 KB)
   $ ls -1 stream-clone/.hg/store/ | egrep '00(changelog|manifest)(\.n|-.*\.nd)'
   00changelog-*.nd (glob)
   00changelog.n
   00manifest-*.nd (glob)
   00manifest.n
   $ hg -R stream-clone debugnodemap --metadata
   uid: * (glob)
   tip-rev: 5005
   tip-node: 90d5d3ba2fc47db50f712570487cb261a68c8ffe
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
 
 new data appened
 -----------------
 
 Other commit happening on the server during the stream clone
 
 setup the step-by-step stream cloning
 
   $ HG_TEST_STREAM_WALKED_FILE_1="$TESTTMP/sync_file_walked_1"
   $ export HG_TEST_STREAM_WALKED_FILE_1
   $ HG_TEST_STREAM_WALKED_FILE_2="$TESTTMP/sync_file_walked_2"
   $ export HG_TEST_STREAM_WALKED_FILE_2
   $ HG_TEST_STREAM_WALKED_FILE_3="$TESTTMP/sync_file_walked_3"
   $ export HG_TEST_STREAM_WALKED_FILE_3
   $ cat << EOF >> test-repo/.hg/hgrc
   > [extensions]
   > steps=$RUNTESTDIR/testlib/ext-stream-clone-steps.py
   > EOF
 
 Check and record file state beforehand
 
   $ f --size test-repo/.hg/store/00changelog*
   test-repo/.hg/store/00changelog-*.nd: size=121088 (glob)
   test-repo/.hg/store/00changelog.d: size=376891 (zstd !)
   test-repo/.hg/store/00changelog.d: size=368890 (no-zstd !)
   test-repo/.hg/store/00changelog.i: size=320384
   test-repo/.hg/store/00changelog.n: size=62
   $ hg -R test-repo debugnodemap --metadata | tee server-metadata.txt
   uid: * (glob)
   tip-rev: 5005
   tip-node: 90d5d3ba2fc47db50f712570487cb261a68c8ffe
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
 
 Prepare a commit
 
   $ echo foo >> test-repo/foo
   $ hg -R test-repo/ add test-repo/foo
 
 Do a mix of clone and commit at the same time so that the file listed on disk differ at actual transfer time.
 
   $ (hg clone -U --stream --config ui.ssh="\"$PYTHON\" \"$TESTDIR/dummyssh\"" ssh://user@dummy/test-repo stream-clone-race-1 --debug 2>> clone-output | egrep '00(changelog|manifest)' >> clone-output; touch $HG_TEST_STREAM_WALKED_FILE_3) &
   $ $RUNTESTDIR/testlib/wait-on-file 10 $HG_TEST_STREAM_WALKED_FILE_1
   $ hg -R test-repo/ commit -m foo
   $ touch $HG_TEST_STREAM_WALKED_FILE_2
   $ $RUNTESTDIR/testlib/wait-on-file 10 $HG_TEST_STREAM_WALKED_FILE_3
   $ cat clone-output
   adding [s] 00manifest.n (62 bytes)
   adding [s] 00manifest-*.nd (118 KB) (glob)
   adding [s] 00changelog.n (62 bytes)
   adding [s] 00changelog-*.nd (118 KB) (glob)
   adding [s] 00manifest.d (452 KB) (no-zstd !)
   adding [s] 00manifest.d (491 KB) (zstd !)
   adding [s] 00changelog.d (360 KB) (no-zstd !)
   adding [s] 00changelog.d (368 KB) (zstd !)
   adding [s] 00manifest.i (313 KB)
   adding [s] 00changelog.i (313 KB)
 
 Check the result state
 
   $ f --size stream-clone-race-1/.hg/store/00changelog*
   stream-clone-race-1/.hg/store/00changelog-*.nd: size=121088 (glob)
   stream-clone-race-1/.hg/store/00changelog.d: size=368890 (no-zstd !)
   stream-clone-race-1/.hg/store/00changelog.d: size=376891 (zstd !)
   stream-clone-race-1/.hg/store/00changelog.i: size=320384
   stream-clone-race-1/.hg/store/00changelog.n: size=62
 
   $ hg -R stream-clone-race-1 debugnodemap --metadata | tee client-metadata.txt
   uid: * (glob)
   tip-rev: 5005
   tip-node: 90d5d3ba2fc47db50f712570487cb261a68c8ffe
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
 
 We get a usable nodemap, so no rewrite would be needed and the metadata should be identical
 (ie: the following diff should be empty)
 
 This isn't the case for the `no-rust` `no-pure` implementation as it use a very minimal nodemap implementation that unconditionnaly rewrite the nodemap "all the time".
 
 #if no-rust no-pure
   $ diff -u server-metadata.txt client-metadata.txt
   --- server-metadata.txt	* (glob)
   +++ client-metadata.txt	* (glob)
   @@ -1,4 +1,4 @@
   -uid: * (glob)
   +uid: * (glob)
    tip-rev: 5005
    tip-node: 90d5d3ba2fc47db50f712570487cb261a68c8ffe
    data-length: 121088
   [1]
 #else
   $ diff -u server-metadata.txt client-metadata.txt
 #endif
 
 
 Clean up after the test.
 
   $ rm -f "$HG_TEST_STREAM_WALKED_FILE_1"
   $ rm -f "$HG_TEST_STREAM_WALKED_FILE_2"
   $ rm -f "$HG_TEST_STREAM_WALKED_FILE_3"
 
 full regeneration
 -----------------
 
 A full nodemap is generated
 
 (ideally this test would append enough data to make sure the nodemap data file
 get changed, however to make thing simpler we will force the regeneration for
 this test.
 
 Check the initial state
 
   $ f --size test-repo/.hg/store/00changelog*
   test-repo/.hg/store/00changelog-*.nd: size=121344 (glob) (rust !)
   test-repo/.hg/store/00changelog-*.nd: size=121344 (glob) (pure !)
   test-repo/.hg/store/00changelog-*.nd: size=121152 (glob) (no-rust no-pure !)
   test-repo/.hg/store/00changelog.d: size=376950 (zstd !)
   test-repo/.hg/store/00changelog.d: size=368949 (no-zstd !)
   test-repo/.hg/store/00changelog.i: size=320448
   test-repo/.hg/store/00changelog.n: size=62
   $ hg -R test-repo debugnodemap --metadata | tee server-metadata-2.txt
   uid: * (glob)
   tip-rev: 5006
   tip-node: ed2ec1eef9aa2a0ec5057c51483bc148d03e810b
   data-length: 121344 (rust !)
   data-length: 121344 (pure !)
   data-length: 121152 (no-rust no-pure !)
   data-unused: 192 (rust !)
   data-unused: 192 (pure !)
   data-unused: 0 (no-rust no-pure !)
   data-unused: 0.158% (rust !)
   data-unused: 0.158% (pure !)
   data-unused: 0.000% (no-rust no-pure !)
 
 Performe the mix of clone and full refresh of the nodemap, so that the files
 (and filenames) are different between listing time and actual transfer time.
 
   $ (hg clone -U --stream --config ui.ssh="\"$PYTHON\" \"$TESTDIR/dummyssh\"" ssh://user@dummy/test-repo stream-clone-race-2 --debug 2>> clone-output-2 | egrep '00(changelog|manifest)' >> clone-output-2; touch $HG_TEST_STREAM_WALKED_FILE_3) &
   $ $RUNTESTDIR/testlib/wait-on-file 10 $HG_TEST_STREAM_WALKED_FILE_1
   $ rm test-repo/.hg/store/00changelog.n
   $ rm test-repo/.hg/store/00changelog-*.nd
   $ hg -R test-repo/ debugupdatecache
   $ touch $HG_TEST_STREAM_WALKED_FILE_2
   $ $RUNTESTDIR/testlib/wait-on-file 10 $HG_TEST_STREAM_WALKED_FILE_3
   $ cat clone-output-2
   adding [s] 00manifest.n (62 bytes)
   adding [s] 00manifest-*.nd (118 KB) (glob)
   adding [s] 00changelog.n (62 bytes)
   adding [s] 00changelog-*.nd (118 KB) (glob)
   adding [s] 00manifest.d (492 KB) (zstd !)
   adding [s] 00manifest.d (452 KB) (no-zstd !)
   adding [s] 00changelog.d (360 KB) (no-zstd !)
   adding [s] 00changelog.d (368 KB) (zstd !)
   adding [s] 00manifest.i (313 KB)
   adding [s] 00changelog.i (313 KB)
 
 Check the result.
 
   $ f --size stream-clone-race-2/.hg/store/00changelog*
   stream-clone-race-2/.hg/store/00changelog-*.nd: size=121344 (glob) (rust !)
   stream-clone-race-2/.hg/store/00changelog-*.nd: size=121344 (glob) (pure !)
   stream-clone-race-2/.hg/store/00changelog-*.nd: size=121152 (glob) (no-rust no-pure !)
   stream-clone-race-2/.hg/store/00changelog.d: size=376950 (zstd !)
   stream-clone-race-2/.hg/store/00changelog.d: size=368949 (no-zstd !)
   stream-clone-race-2/.hg/store/00changelog.i: size=320448
   stream-clone-race-2/.hg/store/00changelog.n: size=62
 
   $ hg -R stream-clone-race-2 debugnodemap --metadata | tee client-metadata-2.txt
   uid: * (glob)
   tip-rev: 5006
   tip-node: ed2ec1eef9aa2a0ec5057c51483bc148d03e810b
   data-length: 121344 (rust !)
   data-unused: 192 (rust !)
   data-unused: 0.158% (rust !)
   data-length: 121152 (no-rust no-pure !)
   data-unused: 0 (no-rust no-pure !)
   data-unused: 0.000% (no-rust no-pure !)
   data-length: 121344 (pure !)
   data-unused: 192 (pure !)
   data-unused: 0.158% (pure !)
 
 We get a usable nodemap, so no rewrite would be needed and the metadata should be identical
 (ie: the following diff should be empty)
 
 This isn't the case for the `no-rust` `no-pure` implementation as it use a very minimal nodemap implementation that unconditionnaly rewrite the nodemap "all the time".
 
 #if no-rust no-pure
   $ diff -u server-metadata-2.txt client-metadata-2.txt
   --- server-metadata-2.txt	* (glob)
   +++ client-metadata-2.txt	* (glob)
   @@ -1,4 +1,4 @@
   -uid: * (glob)
   +uid: * (glob)
    tip-rev: 5006
    tip-node: ed2ec1eef9aa2a0ec5057c51483bc148d03e810b
    data-length: 121152
   [1]
 #else
   $ diff -u server-metadata-2.txt client-metadata-2.txt
 #endif
 
 Clean up after the test
 
   $ rm -f $HG_TEST_STREAM_WALKED_FILE_1
   $ rm -f $HG_TEST_STREAM_WALKED_FILE_2
   $ rm -f $HG_TEST_STREAM_WALKED_FILE_3
 

tests/test-sidedata.t

@@ -1,122 +1,126
 ==========================================================
 Test file dedicated to checking side-data related behavior
 ==========================================================
 
 Check data can be written/read from sidedata
 ============================================
 
   $ cat << EOF >> $HGRCPATH
   > [extensions]
   > testsidedata=$TESTDIR/testlib/ext-sidedata.py
   > EOF
 
   $ hg init test-sidedata --config experimental.revlogv2=enable-unstable-format-and-corrupt-my-data
   $ cd test-sidedata
   $ echo aaa > a
   $ hg add a
   $ hg commit -m a --traceback
   $ echo aaa > b
   $ hg add b
   $ hg commit -m b
   $ echo xxx >> a
   $ hg commit -m aa
 
   $ hg debugsidedata -c 0
   2 sidedata entries
    entry-0001 size 4
    entry-0002 size 32
   $ hg debugsidedata -c 1 -v
   2 sidedata entries
    entry-0001 size 4
     '\x00\x00\x006'
    entry-0002 size 32
     '\x98\t\xf9\xc4v\xf0\xc5P\x90\xf7wRf\xe8\xe27e\xfc\xc1\x93\xa4\x96\xd0\x1d\x97\xaaG\x1d\xd7t\xfa\xde'
   $ hg debugsidedata -m 2
   2 sidedata entries
    entry-0001 size 4
    entry-0002 size 32
   $ hg debugsidedata a  1
   2 sidedata entries
    entry-0001 size 4
    entry-0002 size 32
 
 Check upgrade behavior
 ======================
 
 Right now, sidedata has not upgrade support
 
 Check that we can upgrade to sidedata
 -------------------------------------
 
   $ hg init up-no-side-data --config experimental.revlogv2=no
   $ hg debugformat -v -R up-no-side-data
   format-variant     repo config default
   fncache:            yes    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   persistent-nodemap:  no     no      no (no-rust !)
   persistent-nodemap: yes    yes      no (rust !)
   copies-sdc:          no     no      no
   revlog-v2:           no     no      no
   changelog-v2:        no     no      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib (no-zstd !)
   compression:        zstd   zstd    zstd (zstd !)
   compression-level:  default default default
   $ hg debugformat -v -R up-no-side-data --config experimental.revlogv2=enable-unstable-format-and-corrupt-my-data
   format-variant     repo config default
   fncache:            yes    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   persistent-nodemap:  no     no      no (no-rust !)
   persistent-nodemap: yes    yes      no (rust !)
   copies-sdc:          no     no      no
   revlog-v2:           no    yes      no
   changelog-v2:        no     no      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib (no-zstd !)
   compression:        zstd   zstd    zstd (zstd !)
   compression-level:  default default default
   $ hg debugupgraderepo -R up-no-side-data --config experimental.revlogv2=enable-unstable-format-and-corrupt-my-data > /dev/null
 
 Check that we can downgrade from sidedata
 -----------------------------------------
 
   $ hg init up-side-data --config experimental.revlogv2=enable-unstable-format-and-corrupt-my-data
   $ hg debugformat -v -R up-side-data
   format-variant     repo config default
   fncache:            yes    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   persistent-nodemap:  no     no      no (no-rust !)
   persistent-nodemap: yes    yes      no (rust !)
   copies-sdc:          no     no      no
   revlog-v2:          yes     no      no
   changelog-v2:        no     no      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib (no-zstd !)
   compression:        zstd   zstd    zstd (zstd !)
   compression-level:  default default default
   $ hg debugformat -v -R up-side-data --config experimental.revlogv2=no
   format-variant     repo config default
   fncache:            yes    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   persistent-nodemap:  no     no      no (no-rust !)
   persistent-nodemap: yes    yes      no (rust !)
   copies-sdc:          no     no      no
   revlog-v2:          yes     no      no
   changelog-v2:        no     no      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib (no-zstd !)
   compression:        zstd   zstd    zstd (zstd !)
   compression-level:  default default default
   $ hg debugupgraderepo -R up-side-data --config experimental.revlogv2=no > /dev/null

tests/test-upgrade-repo.t

@@ -1,1615 +1,1737
 #require no-reposimplestore
 
   $ cat >> $HGRCPATH << EOF
   > [extensions]
   > share =
   > [format]
   > # stabilize test accross variant
   > revlog-compression=zlib
   > EOF
 
 store and revlogv1 are required in source
 
   $ hg --config format.usestore=false init no-store
   $ hg -R no-store debugupgraderepo
   abort: cannot upgrade repository; requirement missing: store
   [255]
 
   $ hg init no-revlogv1
   $ cat > no-revlogv1/.hg/requires << EOF
   > dotencode
   > fncache
   > generaldelta
   > store
   > EOF
 
   $ hg -R no-revlogv1 debugupgraderepo
   abort: cannot upgrade repository; missing a revlog version
   [255]
 
 Cannot upgrade shared repositories
 
   $ hg init share-parent
   $ hg -q share share-parent share-child
 
   $ hg -R share-child debugupgraderepo
   abort: cannot upgrade repository; unsupported source requirement: shared
   [255]
 
 Do not yet support upgrading treemanifest repos
 
   $ hg --config experimental.treemanifest=true init treemanifest
   $ hg -R treemanifest debugupgraderepo
   abort: cannot upgrade repository; unsupported source requirement: treemanifest
   [255]
 
 Cannot add treemanifest requirement during upgrade
 
   $ hg init disallowaddedreq
   $ hg -R disallowaddedreq --config experimental.treemanifest=true debugupgraderepo
   abort: cannot upgrade repository; do not support adding requirement: treemanifest
   [255]
 
 An upgrade of a repository created with recommended settings only suggests optimizations
 
   $ hg init empty
   $ cd empty
   $ hg debugformat
   format-variant     repo
   fncache:            yes
+  dirstate-v2:         no
   dotencode:          yes
   generaldelta:       yes
   share-safe:          no
   sparserevlog:       yes
   persistent-nodemap:  no (no-rust !)
   persistent-nodemap: yes (rust !)
   copies-sdc:          no
   revlog-v2:           no
   changelog-v2:        no
   plain-cl-delta:     yes
   compression:        zlib
   compression-level:  default
   $ hg debugformat --verbose
   format-variant     repo config default
   fncache:            yes    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   persistent-nodemap:  no     no      no (no-rust !)
   persistent-nodemap: yes    yes      no (rust !)
   copies-sdc:          no     no      no
   revlog-v2:           no     no      no
   changelog-v2:        no     no      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib (no-zstd !)
   compression:        zlib   zlib    zstd (zstd !)
   compression-level:  default default default
   $ hg debugformat --verbose --config format.usefncache=no
   format-variant     repo config default
   fncache:            yes     no     yes
+  dirstate-v2:         no     no      no
   dotencode:          yes     no     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   persistent-nodemap:  no     no      no (no-rust !)
   persistent-nodemap: yes    yes      no (rust !)
   copies-sdc:          no     no      no
   revlog-v2:           no     no      no
   changelog-v2:        no     no      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib (no-zstd !)
   compression:        zlib   zlib    zstd (zstd !)
   compression-level:  default default default
   $ hg debugformat --verbose --config format.usefncache=no --color=debug
   format-variant     repo config default
   [formatvariant.name.mismatchconfig|fncache:           ][formatvariant.repo.mismatchconfig| yes][formatvariant.config.special|     no][formatvariant.default|     yes]
+  [formatvariant.name.uptodate|dirstate-v2:       ][formatvariant.repo.uptodate|  no][formatvariant.config.default|     no][formatvariant.default|      no]
   [formatvariant.name.mismatchconfig|dotencode:         ][formatvariant.repo.mismatchconfig| yes][formatvariant.config.special|     no][formatvariant.default|     yes]
   [formatvariant.name.uptodate|generaldelta:      ][formatvariant.repo.uptodate| yes][formatvariant.config.default|    yes][formatvariant.default|     yes]
   [formatvariant.name.uptodate|share-safe:        ][formatvariant.repo.uptodate|  no][formatvariant.config.default|     no][formatvariant.default|      no]
   [formatvariant.name.uptodate|sparserevlog:      ][formatvariant.repo.uptodate| yes][formatvariant.config.default|    yes][formatvariant.default|     yes]
   [formatvariant.name.uptodate|persistent-nodemap:][formatvariant.repo.uptodate|  no][formatvariant.config.default|     no][formatvariant.default|      no] (no-rust !)
   [formatvariant.name.mismatchdefault|persistent-nodemap:][formatvariant.repo.mismatchdefault| yes][formatvariant.config.special|    yes][formatvariant.default|      no] (rust !)
   [formatvariant.name.uptodate|copies-sdc:        ][formatvariant.repo.uptodate|  no][formatvariant.config.default|     no][formatvariant.default|      no]
   [formatvariant.name.uptodate|revlog-v2:         ][formatvariant.repo.uptodate|  no][formatvariant.config.default|     no][formatvariant.default|      no]
   [formatvariant.name.uptodate|changelog-v2:      ][formatvariant.repo.uptodate|  no][formatvariant.config.default|     no][formatvariant.default|      no]
   [formatvariant.name.uptodate|plain-cl-delta:    ][formatvariant.repo.uptodate| yes][formatvariant.config.default|    yes][formatvariant.default|     yes]
   [formatvariant.name.uptodate|compression:       ][formatvariant.repo.uptodate| zlib][formatvariant.config.default|   zlib][formatvariant.default|    zlib] (no-zstd !)
   [formatvariant.name.mismatchdefault|compression:       ][formatvariant.repo.mismatchdefault| zlib][formatvariant.config.special|   zlib][formatvariant.default|    zstd] (zstd !)
   [formatvariant.name.uptodate|compression-level: ][formatvariant.repo.uptodate| default][formatvariant.config.default| default][formatvariant.default| default]
   $ hg debugformat -Tjson
   [
    {
     "config": true,
     "default": true,
     "name": "fncache",
     "repo": true
    },
    {
+    "config": false,
+    "default": false,
+    "name": "dirstate-v2",
+    "repo": false
+   },
+   {
     "config": true,
     "default": true,
     "name": "dotencode",
     "repo": true
    },
    {
     "config": true,
     "default": true,
     "name": "generaldelta",
     "repo": true
    },
    {
     "config": false,
     "default": false,
     "name": "share-safe",
     "repo": false
    },
    {
     "config": true,
     "default": true,
     "name": "sparserevlog",
     "repo": true
    },
    {
     "config": false, (no-rust !)
     "config": true, (rust !)
     "default": false,
     "name": "persistent-nodemap",
     "repo": false (no-rust !)
     "repo": true (rust !)
    },
    {
     "config": false,
     "default": false,
     "name": "copies-sdc",
     "repo": false
    },
    {
     "config": false,
     "default": false,
     "name": "revlog-v2",
     "repo": false
    },
    {
     "config": false,
     "default": false,
     "name": "changelog-v2",
     "repo": false
    },
    {
     "config": true,
     "default": true,
     "name": "plain-cl-delta",
     "repo": true
    },
    {
     "config": "zlib",
     "default": "zlib", (no-zstd !)
     "default": "zstd", (zstd !)
     "name": "compression",
     "repo": "zlib"
    },
    {
     "config": "default",
     "default": "default",
     "name": "compression-level",
     "repo": "default"
    }
   ]
   $ hg debugupgraderepo
   (no format upgrades found in existing repository)
   performing an upgrade with "--run" will make the following changes:
   
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, sparserevlog, store (no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlogv1, sparserevlog, store (rust !)
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   additional optimizations are available by specifying "--optimize <name>":
   
   re-delta-parent
      deltas within internal storage will be recalculated to choose an optimal base revision where this was not already done; the size of the repository may shrink and various operations may become faster; the first time this optimization is performed could slow down upgrade execution considerably; subsequent invocations should not run noticeably slower
   
   re-delta-multibase
      deltas within internal storage will be recalculated against multiple base revision and the smallest difference will be used; the size of the repository may shrink significantly when there are many merges; this optimization will slow down execution in proportion to the number of merges in the repository and the amount of files in the repository; this slow down should not be significant unless there are tens of thousands of files and thousands of merges
   
   re-delta-all
      deltas within internal storage will always be recalculated without reusing prior deltas; this will likely make execution run several times slower; this optimization is typically not needed
   
   re-delta-fulladd
      every revision will be re-added as if it was new content. It will go through the full storage mechanism giving extensions a chance to process it (eg. lfs). This is similar to "re-delta-all" but even slower since more logic is involved.
   
 
   $ hg debugupgraderepo --quiet
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, sparserevlog, store (no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlogv1, sparserevlog, store (rust !)
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
 
 --optimize can be used to add optimizations
 
   $ hg debugupgrade --optimize 're-delta-parent'
   (no format upgrades found in existing repository)
   performing an upgrade with "--run" will make the following changes:
   
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, sparserevlog, store (no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlogv1, sparserevlog, store (rust !)
   
   optimisations: re-delta-parent
   
   re-delta-parent
      deltas within internal storage will choose a new base revision if needed
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   additional optimizations are available by specifying "--optimize <name>":
   
   re-delta-multibase
      deltas within internal storage will be recalculated against multiple base revision and the smallest difference will be used; the size of the repository may shrink significantly when there are many merges; this optimization will slow down execution in proportion to the number of merges in the repository and the amount of files in the repository; this slow down should not be significant unless there are tens of thousands of files and thousands of merges
   
   re-delta-all
      deltas within internal storage will always be recalculated without reusing prior deltas; this will likely make execution run several times slower; this optimization is typically not needed
   
   re-delta-fulladd
      every revision will be re-added as if it was new content. It will go through the full storage mechanism giving extensions a chance to process it (eg. lfs). This is similar to "re-delta-all" but even slower since more logic is involved.
   
 
 modern form of the option
 
   $ hg debugupgrade --optimize re-delta-parent
   (no format upgrades found in existing repository)
   performing an upgrade with "--run" will make the following changes:
   
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, sparserevlog, store (no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlogv1, sparserevlog, store (rust !)
   
   optimisations: re-delta-parent
   
   re-delta-parent
      deltas within internal storage will choose a new base revision if needed
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   additional optimizations are available by specifying "--optimize <name>":
   
   re-delta-multibase
      deltas within internal storage will be recalculated against multiple base revision and the smallest difference will be used; the size of the repository may shrink significantly when there are many merges; this optimization will slow down execution in proportion to the number of merges in the repository and the amount of files in the repository; this slow down should not be significant unless there are tens of thousands of files and thousands of merges
   
   re-delta-all
      deltas within internal storage will always be recalculated without reusing prior deltas; this will likely make execution run several times slower; this optimization is typically not needed
   
   re-delta-fulladd
      every revision will be re-added as if it was new content. It will go through the full storage mechanism giving extensions a chance to process it (eg. lfs). This is similar to "re-delta-all" but even slower since more logic is involved.
   
   $ hg debugupgrade --optimize re-delta-parent --quiet
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, sparserevlog, store (no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlogv1, sparserevlog, store (rust !)
   
   optimisations: re-delta-parent
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
 
 unknown optimization:
 
   $ hg debugupgrade --optimize foobar
   abort: unknown optimization action requested: foobar
   (run without arguments to see valid optimizations)
   [255]
 
 Various sub-optimal detections work
 
   $ cat > .hg/requires << EOF
   > revlogv1
   > store
   > EOF
 
   $ hg debugformat
   format-variant     repo
   fncache:             no
+  dirstate-v2:         no
   dotencode:           no
   generaldelta:        no
   share-safe:          no
   sparserevlog:        no
   persistent-nodemap:  no
   copies-sdc:          no
   revlog-v2:           no
   changelog-v2:        no
   plain-cl-delta:     yes
   compression:        zlib
   compression-level:  default
   $ hg debugformat --verbose
   format-variant     repo config default
   fncache:             no    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:           no    yes     yes
   generaldelta:        no    yes     yes
   share-safe:          no     no      no
   sparserevlog:        no    yes     yes
   persistent-nodemap:  no     no      no (no-rust !)
   persistent-nodemap:  no    yes      no (rust !)
   copies-sdc:          no     no      no
   revlog-v2:           no     no      no
   changelog-v2:        no     no      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib (no-zstd !)
   compression:        zlib   zlib    zstd (zstd !)
   compression-level:  default default default
   $ hg debugformat --verbose --config format.usegeneraldelta=no
   format-variant     repo config default
   fncache:             no    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:           no    yes     yes
   generaldelta:        no     no     yes
   share-safe:          no     no      no
   sparserevlog:        no     no     yes
   persistent-nodemap:  no     no      no (no-rust !)
   persistent-nodemap:  no    yes      no (rust !)
   copies-sdc:          no     no      no
   revlog-v2:           no     no      no
   changelog-v2:        no     no      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib (no-zstd !)
   compression:        zlib   zlib    zstd (zstd !)
   compression-level:  default default default
   $ hg debugformat --verbose --config format.usegeneraldelta=no --color=debug
   format-variant     repo config default
   [formatvariant.name.mismatchconfig|fncache:           ][formatvariant.repo.mismatchconfig|  no][formatvariant.config.default|    yes][formatvariant.default|     yes]
+  [formatvariant.name.uptodate|dirstate-v2:       ][formatvariant.repo.uptodate|  no][formatvariant.config.default|     no][formatvariant.default|      no]
   [formatvariant.name.mismatchconfig|dotencode:         ][formatvariant.repo.mismatchconfig|  no][formatvariant.config.default|    yes][formatvariant.default|     yes]
   [formatvariant.name.mismatchdefault|generaldelta:      ][formatvariant.repo.mismatchdefault|  no][formatvariant.config.special|     no][formatvariant.default|     yes]
   [formatvariant.name.uptodate|share-safe:        ][formatvariant.repo.uptodate|  no][formatvariant.config.default|     no][formatvariant.default|      no]
   [formatvariant.name.mismatchdefault|sparserevlog:      ][formatvariant.repo.mismatchdefault|  no][formatvariant.config.special|     no][formatvariant.default|     yes]
   [formatvariant.name.uptodate|persistent-nodemap:][formatvariant.repo.uptodate|  no][formatvariant.config.default|     no][formatvariant.default|      no] (no-rust !)
   [formatvariant.name.mismatchconfig|persistent-nodemap:][formatvariant.repo.mismatchconfig|  no][formatvariant.config.special|    yes][formatvariant.default|      no] (rust !)
   [formatvariant.name.uptodate|copies-sdc:        ][formatvariant.repo.uptodate|  no][formatvariant.config.default|     no][formatvariant.default|      no]
   [formatvariant.name.uptodate|revlog-v2:         ][formatvariant.repo.uptodate|  no][formatvariant.config.default|     no][formatvariant.default|      no]
   [formatvariant.name.uptodate|changelog-v2:      ][formatvariant.repo.uptodate|  no][formatvariant.config.default|     no][formatvariant.default|      no]
   [formatvariant.name.uptodate|plain-cl-delta:    ][formatvariant.repo.uptodate| yes][formatvariant.config.default|    yes][formatvariant.default|     yes]
   [formatvariant.name.uptodate|compression:       ][formatvariant.repo.uptodate| zlib][formatvariant.config.default|   zlib][formatvariant.default|    zlib] (no-zstd !)
   [formatvariant.name.mismatchdefault|compression:       ][formatvariant.repo.mismatchdefault| zlib][formatvariant.config.special|   zlib][formatvariant.default|    zstd] (zstd !)
   [formatvariant.name.uptodate|compression-level: ][formatvariant.repo.uptodate| default][formatvariant.config.default| default][formatvariant.default| default]
   $ hg debugupgraderepo
   repository lacks features recommended by current config options:
   
   fncache
      long and reserved filenames may not work correctly; repository performance is sub-optimal
   
   dotencode
      storage of filenames beginning with a period or space may not work correctly
   
   generaldelta
      deltas within internal storage are unable to choose optimal revisions; repository is larger and slower than it could be; interaction with other repositories may require extra network and CPU resources, making "hg push" and "hg pull" slower
   
   sparserevlog
      in order to limit disk reading and memory usage on older version, the span of a delta chain from its root to its end is limited, whatever the relevant data in this span. This can severly limit Mercurial ability to build good chain of delta resulting is much more storage space being taken and limit reusability of on disk delta during exchange.
   
   persistent-nodemap (rust !)
      persist the node -> rev mapping on disk to speedup lookup (rust !)
    (rust !)
   
   performing an upgrade with "--run" will make the following changes:
   
   requirements
      preserved: revlogv1, store
      added: dotencode, fncache, generaldelta, sparserevlog (no-rust !)
      added: dotencode, fncache, generaldelta, persistent-nodemap, sparserevlog (rust !)
   
   fncache
      repository will be more resilient to storing certain paths and performance of certain operations should be improved
   
   dotencode
      repository will be better able to store files beginning with a space or period
   
   generaldelta
      repository storage will be able to create optimal deltas; new repository data will be smaller and read times should decrease; interacting with other repositories using this storage model should require less network and CPU resources, making "hg push" and "hg pull" faster
   
   sparserevlog
      Revlog supports delta chain with more unused data between payload. These gaps will be skipped at read time. This allows for better delta chains, making a better compression and faster exchange with server.
   
   persistent-nodemap (rust !)
      Speedup revision lookup by node id. (rust !)
    (rust !)
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   additional optimizations are available by specifying "--optimize <name>":
   
   re-delta-parent
      deltas within internal storage will be recalculated to choose an optimal base revision where this was not already done; the size of the repository may shrink and various operations may become faster; the first time this optimization is performed could slow down upgrade execution considerably; subsequent invocations should not run noticeably slower
   
   re-delta-multibase
      deltas within internal storage will be recalculated against multiple base revision and the smallest difference will be used; the size of the repository may shrink significantly when there are many merges; this optimization will slow down execution in proportion to the number of merges in the repository and the amount of files in the repository; this slow down should not be significant unless there are tens of thousands of files and thousands of merges
   
   re-delta-all
      deltas within internal storage will always be recalculated without reusing prior deltas; this will likely make execution run several times slower; this optimization is typically not needed
   
   re-delta-fulladd
      every revision will be re-added as if it was new content. It will go through the full storage mechanism giving extensions a chance to process it (eg. lfs). This is similar to "re-delta-all" but even slower since more logic is involved.
   
   $ hg debugupgraderepo --quiet
   requirements
      preserved: revlogv1, store
      added: dotencode, fncache, generaldelta, sparserevlog (no-rust !)
      added: dotencode, fncache, generaldelta, persistent-nodemap, sparserevlog (rust !)
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
 
   $ hg --config format.dotencode=false debugupgraderepo
   repository lacks features recommended by current config options:
   
   fncache
      long and reserved filenames may not work correctly; repository performance is sub-optimal
   
   generaldelta
      deltas within internal storage are unable to choose optimal revisions; repository is larger and slower than it could be; interaction with other repositories may require extra network and CPU resources, making "hg push" and "hg pull" slower
   
   sparserevlog
      in order to limit disk reading and memory usage on older version, the span of a delta chain from its root to its end is limited, whatever the relevant data in this span. This can severly limit Mercurial ability to build good chain of delta resulting is much more storage space being taken and limit reusability of on disk delta during exchange.
   
   persistent-nodemap (rust !)
      persist the node -> rev mapping on disk to speedup lookup (rust !)
    (rust !)
   repository lacks features used by the default config options:
   
   dotencode
      storage of filenames beginning with a period or space may not work correctly
   
   
   performing an upgrade with "--run" will make the following changes:
   
   requirements
      preserved: revlogv1, store
      added: fncache, generaldelta, sparserevlog (no-rust !)
      added: fncache, generaldelta, persistent-nodemap, sparserevlog (rust !)
   
   fncache
      repository will be more resilient to storing certain paths and performance of certain operations should be improved
   
   generaldelta
      repository storage will be able to create optimal deltas; new repository data will be smaller and read times should decrease; interacting with other repositories using this storage model should require less network and CPU resources, making "hg push" and "hg pull" faster
   
   sparserevlog
      Revlog supports delta chain with more unused data between payload. These gaps will be skipped at read time. This allows for better delta chains, making a better compression and faster exchange with server.
   
   persistent-nodemap (rust !)
      Speedup revision lookup by node id. (rust !)
    (rust !)
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   additional optimizations are available by specifying "--optimize <name>":
   
   re-delta-parent
      deltas within internal storage will be recalculated to choose an optimal base revision where this was not already done; the size of the repository may shrink and various operations may become faster; the first time this optimization is performed could slow down upgrade execution considerably; subsequent invocations should not run noticeably slower
   
   re-delta-multibase
      deltas within internal storage will be recalculated against multiple base revision and the smallest difference will be used; the size of the repository may shrink significantly when there are many merges; this optimization will slow down execution in proportion to the number of merges in the repository and the amount of files in the repository; this slow down should not be significant unless there are tens of thousands of files and thousands of merges
   
   re-delta-all
      deltas within internal storage will always be recalculated without reusing prior deltas; this will likely make execution run several times slower; this optimization is typically not needed
   
   re-delta-fulladd
      every revision will be re-added as if it was new content. It will go through the full storage mechanism giving extensions a chance to process it (eg. lfs). This is similar to "re-delta-all" but even slower since more logic is involved.
   
 
   $ cd ..
 
 Upgrading a repository that is already modern essentially no-ops
 
   $ hg init modern
   $ hg -R modern debugupgraderepo --run
   nothing to do
 
 Upgrading a repository to generaldelta works
 
   $ hg --config format.usegeneraldelta=false init upgradegd
   $ cd upgradegd
   $ touch f0
   $ hg -q commit -A -m initial
   $ mkdir FooBarDirectory.d
   $ touch FooBarDirectory.d/f1
   $ hg -q commit -A -m 'add f1'
   $ hg -q up -r 0
   >>> from __future__ import absolute_import, print_function
   >>> import random
   >>> random.seed(0) # have a reproducible content
   >>> with open("f2", "wb") as f:
   ...     for i in range(100000):
   ...         f.write(b"%d\n" % random.randint(1000000000, 9999999999)) and None
   $ hg -q commit -A -m 'add f2'
 
 make sure we have a .d file
 
   $ ls -d .hg/store/data/*
   .hg/store/data/_foo_bar_directory.d.hg
   .hg/store/data/f0.i
   .hg/store/data/f2.d
   .hg/store/data/f2.i
 
   $ hg debugupgraderepo --run --config format.sparse-revlog=false
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, revlogv1, store (no-rust !)
      preserved: dotencode, fncache, persistent-nodemap, revlogv1, store (rust !)
      added: generaldelta
   
   generaldelta
      repository storage will be able to create optimal deltas; new repository data will be smaller and read times should decrease; interacting with other repositories using this storage model should require less network and CPU resources, making "hg push" and "hg pull" faster
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   beginning upgrade...
   repository locked and read-only
   creating temporary repository to stage upgraded data: $TESTTMP/upgradegd/.hg/upgrade.* (glob)
   (it is safe to interrupt this process any time before data migration completes)
   migrating 9 total revisions (3 in filelogs, 3 in manifests, 3 in changelog)
   migrating 519 KB in store; 1.05 MB tracked data
   migrating 3 filelogs containing 3 revisions (518 KB in store; 1.05 MB tracked data)
   finished migrating 3 filelog revisions across 3 filelogs; change in size: 0 bytes
   migrating 1 manifests containing 3 revisions (384 bytes in store; 238 bytes tracked data)
   finished migrating 3 manifest revisions across 1 manifests; change in size: -17 bytes
   migrating changelog containing 3 revisions (394 bytes in store; 199 bytes tracked data)
   finished migrating 3 changelog revisions; change in size: 0 bytes
   finished migrating 9 total revisions; total change in store size: -17 bytes
   copying phaseroots
   data fully upgraded in a temporary repository
   marking source repository as being upgraded; clients will be unable to read from repository
   starting in-place swap of repository data
   replaced files will be backed up at $TESTTMP/upgradegd/.hg/upgradebackup.* (glob)
   replacing store...
   store replacement complete; repository was inconsistent for *s (glob)
   finalizing requirements file and making repository readable again
   removing temporary repository $TESTTMP/upgradegd/.hg/upgrade.* (glob)
   copy of old repository backed up at $TESTTMP/upgradegd/.hg/upgradebackup.* (glob)
   the old repository will not be deleted; remove it to free up disk space once the upgraded repository is verified
 
 Original requirements backed up
 
   $ cat .hg/upgradebackup.*/requires
   dotencode
   fncache
   persistent-nodemap (rust !)
   revlogv1
   store
 
 generaldelta added to original requirements files
 
   $ cat .hg/requires
   dotencode
   fncache
   generaldelta
   persistent-nodemap (rust !)
   revlogv1
   store
 
 store directory has files we expect
 
   $ ls .hg/store
   00changelog.i
   00manifest.i
   data
   fncache
   phaseroots
   undo
   undo.backupfiles
   undo.phaseroots
 
 manifest should be generaldelta
 
   $ hg debugrevlog -m | grep flags
   flags  : inline, generaldelta
 
 verify should be happy
 
   $ hg verify
   checking changesets
   checking manifests
   crosschecking files in changesets and manifests
   checking files
   checked 3 changesets with 3 changes to 3 files
 
 old store should be backed up
 
   $ ls -d .hg/upgradebackup.*/
   .hg/upgradebackup.*/ (glob)
   $ ls .hg/upgradebackup.*/store
   00changelog.i
   00manifest.i
   data
   fncache
   phaseroots
   undo
   undo.backup.fncache
   undo.backupfiles
   undo.phaseroots
 
 unless --no-backup is passed
 
   $ rm -rf .hg/upgradebackup.*/
   $ hg debugupgraderepo --run --no-backup
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, store (no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlogv1, store (rust !)
      added: sparserevlog
   
   sparserevlog
      Revlog supports delta chain with more unused data between payload. These gaps will be skipped at read time. This allows for better delta chains, making a better compression and faster exchange with server.
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   beginning upgrade...
   repository locked and read-only
   creating temporary repository to stage upgraded data: $TESTTMP/upgradegd/.hg/upgrade.* (glob)
   (it is safe to interrupt this process any time before data migration completes)
   migrating 9 total revisions (3 in filelogs, 3 in manifests, 3 in changelog)
   migrating 519 KB in store; 1.05 MB tracked data
   migrating 3 filelogs containing 3 revisions (518 KB in store; 1.05 MB tracked data)
   finished migrating 3 filelog revisions across 3 filelogs; change in size: 0 bytes
   migrating 1 manifests containing 3 revisions (367 bytes in store; 238 bytes tracked data)
   finished migrating 3 manifest revisions across 1 manifests; change in size: 0 bytes
   migrating changelog containing 3 revisions (394 bytes in store; 199 bytes tracked data)
   finished migrating 3 changelog revisions; change in size: 0 bytes
   finished migrating 9 total revisions; total change in store size: 0 bytes
   copying phaseroots
   data fully upgraded in a temporary repository
   marking source repository as being upgraded; clients will be unable to read from repository
   starting in-place swap of repository data
   replacing store...
   store replacement complete; repository was inconsistent for * (glob)
   finalizing requirements file and making repository readable again
   removing temporary repository $TESTTMP/upgradegd/.hg/upgrade.* (glob)
   $ ls -1 .hg/ | grep upgradebackup
   [1]
 
 We can restrict optimization to some revlog:
 
   $ hg debugupgrade --optimize re-delta-parent --run --manifest --no-backup --debug --traceback
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, sparserevlog, store (no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlogv1, sparserevlog, store (rust !)
   
   optimisations: re-delta-parent
   
   re-delta-parent
      deltas within internal storage will choose a new base revision if needed
   
   processed revlogs:
     - manifest
   
   beginning upgrade...
   repository locked and read-only
   creating temporary repository to stage upgraded data: $TESTTMP/upgradegd/.hg/upgrade.* (glob)
   (it is safe to interrupt this process any time before data migration completes)
   migrating 9 total revisions (3 in filelogs, 3 in manifests, 3 in changelog)
   migrating 519 KB in store; 1.05 MB tracked data
   migrating 3 filelogs containing 3 revisions (518 KB in store; 1.05 MB tracked data)
   blindly copying data/FooBarDirectory.d/f1.i containing 1 revisions
   blindly copying data/f0.i containing 1 revisions
   blindly copying data/f2.i containing 1 revisions
   finished migrating 3 filelog revisions across 3 filelogs; change in size: 0 bytes
   migrating 1 manifests containing 3 revisions (367 bytes in store; 238 bytes tracked data)
   cloning 3 revisions from 00manifest.i
   finished migrating 3 manifest revisions across 1 manifests; change in size: 0 bytes
   migrating changelog containing 3 revisions (394 bytes in store; 199 bytes tracked data)
   blindly copying 00changelog.i containing 3 revisions
   finished migrating 3 changelog revisions; change in size: 0 bytes
   finished migrating 9 total revisions; total change in store size: 0 bytes
   copying phaseroots
   data fully upgraded in a temporary repository
   marking source repository as being upgraded; clients will be unable to read from repository
   starting in-place swap of repository data
   replacing store...
   store replacement complete; repository was inconsistent for *s (glob)
   finalizing requirements file and making repository readable again
   removing temporary repository $TESTTMP/upgradegd/.hg/upgrade.* (glob)
 
 Check that the repo still works fine
 
   $ hg log -G --stat
   @  changeset:   2:76d4395f5413 (no-py3 !)
   @  changeset:   2:fca376863211 (py3 !)
   |  tag:         tip
   |  parent:      0:ba592bf28da2
   |  user:        test
   |  date:        Thu Jan 01 00:00:00 1970 +0000
   |  summary:     add f2
   |
   |   f2 |  100000 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
   |   1 files changed, 100000 insertions(+), 0 deletions(-)
   |
   | o  changeset:   1:2029ce2354e2
   |/   user:        test
   |    date:        Thu Jan 01 00:00:00 1970 +0000
   |    summary:     add f1
   |
   |
   o  changeset:   0:ba592bf28da2
      user:        test
      date:        Thu Jan 01 00:00:00 1970 +0000
      summary:     initial
   
   
 
   $ hg verify
   checking changesets
   checking manifests
   crosschecking files in changesets and manifests
   checking files
   checked 3 changesets with 3 changes to 3 files
 
 Check we can select negatively
 
   $ hg debugupgrade --optimize re-delta-parent --run --no-manifest --no-backup --debug --traceback
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, sparserevlog, store (no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlogv1, sparserevlog, store (rust !)
   
   optimisations: re-delta-parent
   
   re-delta-parent
      deltas within internal storage will choose a new base revision if needed
   
   processed revlogs:
     - all-filelogs
     - changelog
   
   beginning upgrade...
   repository locked and read-only
   creating temporary repository to stage upgraded data: $TESTTMP/upgradegd/.hg/upgrade.* (glob)
   (it is safe to interrupt this process any time before data migration completes)
   migrating 9 total revisions (3 in filelogs, 3 in manifests, 3 in changelog)
   migrating 519 KB in store; 1.05 MB tracked data
   migrating 3 filelogs containing 3 revisions (518 KB in store; 1.05 MB tracked data)
   cloning 1 revisions from data/FooBarDirectory.d/f1.i
   cloning 1 revisions from data/f0.i
   cloning 1 revisions from data/f2.i
   finished migrating 3 filelog revisions across 3 filelogs; change in size: 0 bytes
   migrating 1 manifests containing 3 revisions (367 bytes in store; 238 bytes tracked data)
   blindly copying 00manifest.i containing 3 revisions
   finished migrating 3 manifest revisions across 1 manifests; change in size: 0 bytes
   migrating changelog containing 3 revisions (394 bytes in store; 199 bytes tracked data)
   cloning 3 revisions from 00changelog.i
   finished migrating 3 changelog revisions; change in size: 0 bytes
   finished migrating 9 total revisions; total change in store size: 0 bytes
   copying phaseroots
   data fully upgraded in a temporary repository
   marking source repository as being upgraded; clients will be unable to read from repository
   starting in-place swap of repository data
   replacing store...
   store replacement complete; repository was inconsistent for *s (glob)
   finalizing requirements file and making repository readable again
   removing temporary repository $TESTTMP/upgradegd/.hg/upgrade.* (glob)
   $ hg verify
   checking changesets
   checking manifests
   crosschecking files in changesets and manifests
   checking files
   checked 3 changesets with 3 changes to 3 files
 
 Check that we can select changelog only
 
   $ hg debugupgrade --optimize re-delta-parent --run --changelog --no-backup --debug --traceback
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, sparserevlog, store (no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlogv1, sparserevlog, store (rust !)
   
   optimisations: re-delta-parent
   
   re-delta-parent
      deltas within internal storage will choose a new base revision if needed
   
   processed revlogs:
     - changelog
   
   beginning upgrade...
   repository locked and read-only
   creating temporary repository to stage upgraded data: $TESTTMP/upgradegd/.hg/upgrade.* (glob)
   (it is safe to interrupt this process any time before data migration completes)
   migrating 9 total revisions (3 in filelogs, 3 in manifests, 3 in changelog)
   migrating 519 KB in store; 1.05 MB tracked data
   migrating 3 filelogs containing 3 revisions (518 KB in store; 1.05 MB tracked data)
   blindly copying data/FooBarDirectory.d/f1.i containing 1 revisions
   blindly copying data/f0.i containing 1 revisions
   blindly copying data/f2.i containing 1 revisions
   finished migrating 3 filelog revisions across 3 filelogs; change in size: 0 bytes
   migrating 1 manifests containing 3 revisions (367 bytes in store; 238 bytes tracked data)
   blindly copying 00manifest.i containing 3 revisions
   finished migrating 3 manifest revisions across 1 manifests; change in size: 0 bytes
   migrating changelog containing 3 revisions (394 bytes in store; 199 bytes tracked data)
   cloning 3 revisions from 00changelog.i
   finished migrating 3 changelog revisions; change in size: 0 bytes
   finished migrating 9 total revisions; total change in store size: 0 bytes
   copying phaseroots
   data fully upgraded in a temporary repository
   marking source repository as being upgraded; clients will be unable to read from repository
   starting in-place swap of repository data
   replacing store...
   store replacement complete; repository was inconsistent for *s (glob)
   finalizing requirements file and making repository readable again
   removing temporary repository $TESTTMP/upgradegd/.hg/upgrade.* (glob)
   $ hg verify
   checking changesets
   checking manifests
   crosschecking files in changesets and manifests
   checking files
   checked 3 changesets with 3 changes to 3 files
 
 Check that we can select filelog only
 
   $ hg debugupgrade --optimize re-delta-parent --run --no-changelog --no-manifest --no-backup --debug --traceback
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, sparserevlog, store (no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlogv1, sparserevlog, store (rust !)
   
   optimisations: re-delta-parent
   
   re-delta-parent
      deltas within internal storage will choose a new base revision if needed
   
   processed revlogs:
     - all-filelogs
   
   beginning upgrade...
   repository locked and read-only
   creating temporary repository to stage upgraded data: $TESTTMP/upgradegd/.hg/upgrade.* (glob)
   (it is safe to interrupt this process any time before data migration completes)
   migrating 9 total revisions (3 in filelogs, 3 in manifests, 3 in changelog)
   migrating 519 KB in store; 1.05 MB tracked data
   migrating 3 filelogs containing 3 revisions (518 KB in store; 1.05 MB tracked data)
   cloning 1 revisions from data/FooBarDirectory.d/f1.i
   cloning 1 revisions from data/f0.i
   cloning 1 revisions from data/f2.i
   finished migrating 3 filelog revisions across 3 filelogs; change in size: 0 bytes
   migrating 1 manifests containing 3 revisions (367 bytes in store; 238 bytes tracked data)
   blindly copying 00manifest.i containing 3 revisions
   finished migrating 3 manifest revisions across 1 manifests; change in size: 0 bytes
   migrating changelog containing 3 revisions (394 bytes in store; 199 bytes tracked data)
   blindly copying 00changelog.i containing 3 revisions
   finished migrating 3 changelog revisions; change in size: 0 bytes
   finished migrating 9 total revisions; total change in store size: 0 bytes
   copying phaseroots
   data fully upgraded in a temporary repository
   marking source repository as being upgraded; clients will be unable to read from repository
   starting in-place swap of repository data
   replacing store...
   store replacement complete; repository was inconsistent for *s (glob)
   finalizing requirements file and making repository readable again
   removing temporary repository $TESTTMP/upgradegd/.hg/upgrade.* (glob)
   $ hg verify
   checking changesets
   checking manifests
   crosschecking files in changesets and manifests
   checking files
   checked 3 changesets with 3 changes to 3 files
 
 
 Check you can't skip revlog clone during important format downgrade
 
   $ echo "[format]" > .hg/hgrc
   $ echo "sparse-revlog=no" >> .hg/hgrc
   $ hg debugupgrade --optimize re-delta-parent --run --manifest --no-backup --debug --traceback
   ignoring revlogs selection flags, format requirements change: sparserevlog
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, store (no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlogv1, store (rust !)
      removed: sparserevlog
   
   optimisations: re-delta-parent
   
   re-delta-parent
      deltas within internal storage will choose a new base revision if needed
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   beginning upgrade...
   repository locked and read-only
   creating temporary repository to stage upgraded data: $TESTTMP/upgradegd/.hg/upgrade.* (glob)
   (it is safe to interrupt this process any time before data migration completes)
   migrating 9 total revisions (3 in filelogs, 3 in manifests, 3 in changelog)
   migrating 519 KB in store; 1.05 MB tracked data
   migrating 3 filelogs containing 3 revisions (518 KB in store; 1.05 MB tracked data)
   cloning 1 revisions from data/FooBarDirectory.d/f1.i
   cloning 1 revisions from data/f0.i
   cloning 1 revisions from data/f2.i
   finished migrating 3 filelog revisions across 3 filelogs; change in size: 0 bytes
   migrating 1 manifests containing 3 revisions (367 bytes in store; 238 bytes tracked data)
   cloning 3 revisions from 00manifest.i
   finished migrating 3 manifest revisions across 1 manifests; change in size: 0 bytes
   migrating changelog containing 3 revisions (394 bytes in store; 199 bytes tracked data)
   cloning 3 revisions from 00changelog.i
   finished migrating 3 changelog revisions; change in size: 0 bytes
   finished migrating 9 total revisions; total change in store size: 0 bytes
   copying phaseroots
   data fully upgraded in a temporary repository
   marking source repository as being upgraded; clients will be unable to read from repository
   starting in-place swap of repository data
   replacing store...
   store replacement complete; repository was inconsistent for *s (glob)
   finalizing requirements file and making repository readable again
   removing temporary repository $TESTTMP/upgradegd/.hg/upgrade.* (glob)
   $ hg verify
   checking changesets
   checking manifests
   crosschecking files in changesets and manifests
   checking files
   checked 3 changesets with 3 changes to 3 files
 
 Check you can't skip revlog clone during important format upgrade
 
   $ echo "sparse-revlog=yes" >> .hg/hgrc
   $ hg debugupgrade --optimize re-delta-parent --run --manifest --no-backup --debug --traceback
   ignoring revlogs selection flags, format requirements change: sparserevlog
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, store (no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlogv1, store (rust !)
      added: sparserevlog
   
   optimisations: re-delta-parent
   
   sparserevlog
      Revlog supports delta chain with more unused data between payload. These gaps will be skipped at read time. This allows for better delta chains, making a better compression and faster exchange with server.
   
   re-delta-parent
      deltas within internal storage will choose a new base revision if needed
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   beginning upgrade...
   repository locked and read-only
   creating temporary repository to stage upgraded data: $TESTTMP/upgradegd/.hg/upgrade.* (glob)
   (it is safe to interrupt this process any time before data migration completes)
   migrating 9 total revisions (3 in filelogs, 3 in manifests, 3 in changelog)
   migrating 519 KB in store; 1.05 MB tracked data
   migrating 3 filelogs containing 3 revisions (518 KB in store; 1.05 MB tracked data)
   cloning 1 revisions from data/FooBarDirectory.d/f1.i
   cloning 1 revisions from data/f0.i
   cloning 1 revisions from data/f2.i
   finished migrating 3 filelog revisions across 3 filelogs; change in size: 0 bytes
   migrating 1 manifests containing 3 revisions (367 bytes in store; 238 bytes tracked data)
   cloning 3 revisions from 00manifest.i
   finished migrating 3 manifest revisions across 1 manifests; change in size: 0 bytes
   migrating changelog containing 3 revisions (394 bytes in store; 199 bytes tracked data)
   cloning 3 revisions from 00changelog.i
   finished migrating 3 changelog revisions; change in size: 0 bytes
   finished migrating 9 total revisions; total change in store size: 0 bytes
   copying phaseroots
   data fully upgraded in a temporary repository
   marking source repository as being upgraded; clients will be unable to read from repository
   starting in-place swap of repository data
   replacing store...
   store replacement complete; repository was inconsistent for *s (glob)
   finalizing requirements file and making repository readable again
   removing temporary repository $TESTTMP/upgradegd/.hg/upgrade.* (glob)
   $ hg verify
   checking changesets
   checking manifests
   crosschecking files in changesets and manifests
   checking files
   checked 3 changesets with 3 changes to 3 files
 
   $ cd ..
 
 store files with special filenames aren't encoded during copy
 
   $ hg init store-filenames
   $ cd store-filenames
   $ touch foo
   $ hg -q commit -A -m initial
   $ touch .hg/store/.XX_special_filename
 
   $ hg debugupgraderepo --run
   nothing to do
   $ hg debugupgraderepo --run --optimize 're-delta-fulladd'
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, sparserevlog, store (no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlogv1, sparserevlog, store (rust !)
   
   optimisations: re-delta-fulladd
   
   re-delta-fulladd
      each revision will be added as new content to the internal storage; this will likely drastically slow down execution time, but some extensions might need it
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   beginning upgrade...
   repository locked and read-only
   creating temporary repository to stage upgraded data: $TESTTMP/store-filenames/.hg/upgrade.* (glob)
   (it is safe to interrupt this process any time before data migration completes)
   migrating 3 total revisions (1 in filelogs, 1 in manifests, 1 in changelog)
   migrating 301 bytes in store; 107 bytes tracked data
   migrating 1 filelogs containing 1 revisions (64 bytes in store; 0 bytes tracked data)
   finished migrating 1 filelog revisions across 1 filelogs; change in size: 0 bytes
   migrating 1 manifests containing 1 revisions (110 bytes in store; 45 bytes tracked data)
   finished migrating 1 manifest revisions across 1 manifests; change in size: 0 bytes
   migrating changelog containing 1 revisions (127 bytes in store; 62 bytes tracked data)
   finished migrating 1 changelog revisions; change in size: 0 bytes
   finished migrating 3 total revisions; total change in store size: 0 bytes
   copying .XX_special_filename
   copying phaseroots
   data fully upgraded in a temporary repository
   marking source repository as being upgraded; clients will be unable to read from repository
   starting in-place swap of repository data
   replaced files will be backed up at $TESTTMP/store-filenames/.hg/upgradebackup.* (glob)
   replacing store...
   store replacement complete; repository was inconsistent for *s (glob)
   finalizing requirements file and making repository readable again
   removing temporary repository $TESTTMP/store-filenames/.hg/upgrade.* (glob)
   copy of old repository backed up at $TESTTMP/store-filenames/.hg/upgradebackup.* (glob)
   the old repository will not be deleted; remove it to free up disk space once the upgraded repository is verified
 
 fncache is valid after upgrade
 
   $ hg debugrebuildfncache
   fncache already up to date
 
   $ cd ..
 
 Check upgrading a large file repository
 ---------------------------------------
 
   $ hg init largefilesrepo
   $ cat << EOF >> largefilesrepo/.hg/hgrc
   > [extensions]
   > largefiles =
   > EOF
 
   $ cd largefilesrepo
   $ touch foo
   $ hg add --large foo
   $ hg -q commit -m initial
   $ cat .hg/requires
   dotencode
   fncache
   generaldelta
   largefiles
   persistent-nodemap (rust !)
   revlogv1
   sparserevlog
   store
 
   $ hg debugupgraderepo --run
   nothing to do
   $ cat .hg/requires
   dotencode
   fncache
   generaldelta
   largefiles
   persistent-nodemap (rust !)
   revlogv1
   sparserevlog
   store
 
   $ cat << EOF >> .hg/hgrc
   > [extensions]
   > lfs =
   > [lfs]
   > threshold = 10
   > EOF
   $ echo '123456789012345' > lfs.bin
   $ hg ci -Am 'lfs.bin'
   adding lfs.bin
   $ grep lfs .hg/requires
   lfs
   $ find .hg/store/lfs -type f
   .hg/store/lfs/objects/d0/beab232adff5ba365880366ad30b1edb85c4c5372442b5d2fe27adc96d653f
 
   $ hg debugupgraderepo --run
   nothing to do
 
   $ grep lfs .hg/requires
   lfs
   $ find .hg/store/lfs -type f
   .hg/store/lfs/objects/d0/beab232adff5ba365880366ad30b1edb85c4c5372442b5d2fe27adc96d653f
   $ hg verify
   checking changesets
   checking manifests
   crosschecking files in changesets and manifests
   checking files
   checked 2 changesets with 2 changes to 2 files
   $ hg debugdata lfs.bin 0
   version https://git-lfs.github.com/spec/v1
   oid sha256:d0beab232adff5ba365880366ad30b1edb85c4c5372442b5d2fe27adc96d653f
   size 16
   x-is-binary 0
 
   $ cd ..
 
 repository config is taken in account
 -------------------------------------
 
   $ cat << EOF >> $HGRCPATH
   > [format]
   > maxchainlen = 1
   > EOF
 
   $ hg init localconfig
   $ cd localconfig
   $ cat << EOF > file
   > some content
   > with some length
   > to make sure we get a delta
   > after changes
   > very long
   > very long
   > very long
   > very long
   > very long
   > very long
   > very long
   > very long
   > very long
   > very long
   > very long
   > EOF
   $ hg -q commit -A -m A
   $ echo "new line" >> file
   $ hg -q commit -m B
   $ echo "new line" >> file
   $ hg -q commit -m C
 
   $ cat << EOF >> .hg/hgrc
   > [format]
   > maxchainlen = 9001
   > EOF
   $ hg config format
   format.revlog-compression=$BUNDLE2_COMPRESSIONS$
   format.maxchainlen=9001
   $ hg debugdeltachain file
       rev  chain# chainlen     prev   delta       size    rawsize  chainsize     ratio   lindist extradist extraratio   readsize largestblk rddensity srchunks
         0       1        1       -1    base         77        182         77   0.42308        77         0    0.00000         77         77   1.00000        1
         1       1        2        0      p1         21        191         98   0.51309        98         0    0.00000         98         98   1.00000        1
         2       1        2        0   other         30        200        107   0.53500       128        21    0.19626        128        128   0.83594        1
 
   $ hg debugupgraderepo --run --optimize 're-delta-all'
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, sparserevlog, store (no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlogv1, sparserevlog, store (rust !)
   
   optimisations: re-delta-all
   
   re-delta-all
      deltas within internal storage will be fully recomputed; this will likely drastically slow down execution time
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   beginning upgrade...
   repository locked and read-only
   creating temporary repository to stage upgraded data: $TESTTMP/localconfig/.hg/upgrade.* (glob)
   (it is safe to interrupt this process any time before data migration completes)
   migrating 9 total revisions (3 in filelogs, 3 in manifests, 3 in changelog)
   migrating 1019 bytes in store; 882 bytes tracked data
   migrating 1 filelogs containing 3 revisions (320 bytes in store; 573 bytes tracked data)
   finished migrating 3 filelog revisions across 1 filelogs; change in size: -9 bytes
   migrating 1 manifests containing 3 revisions (333 bytes in store; 138 bytes tracked data)
   finished migrating 3 manifest revisions across 1 manifests; change in size: 0 bytes
   migrating changelog containing 3 revisions (366 bytes in store; 171 bytes tracked data)
   finished migrating 3 changelog revisions; change in size: 0 bytes
   finished migrating 9 total revisions; total change in store size: -9 bytes
   copying phaseroots
   data fully upgraded in a temporary repository
   marking source repository as being upgraded; clients will be unable to read from repository
   starting in-place swap of repository data
   replaced files will be backed up at $TESTTMP/localconfig/.hg/upgradebackup.* (glob)
   replacing store...
   store replacement complete; repository was inconsistent for *s (glob)
   finalizing requirements file and making repository readable again
   removing temporary repository $TESTTMP/localconfig/.hg/upgrade.* (glob)
   copy of old repository backed up at $TESTTMP/localconfig/.hg/upgradebackup.* (glob)
   the old repository will not be deleted; remove it to free up disk space once the upgraded repository is verified
   $ hg debugdeltachain file
       rev  chain# chainlen     prev   delta       size    rawsize  chainsize     ratio   lindist extradist extraratio   readsize largestblk rddensity srchunks
         0       1        1       -1    base         77        182         77   0.42308        77         0    0.00000         77         77   1.00000        1
         1       1        2        0      p1         21        191         98   0.51309        98         0    0.00000         98         98   1.00000        1
         2       1        3        1      p1         21        200        119   0.59500       119         0    0.00000        119        119   1.00000        1
   $ cd ..
 
   $ cat << EOF >> $HGRCPATH
   > [format]
   > maxchainlen = 9001
   > EOF
 
 Check upgrading a sparse-revlog repository
 ---------------------------------------
 
   $ hg init sparserevlogrepo --config format.sparse-revlog=no
   $ cd sparserevlogrepo
   $ touch foo
   $ hg add foo
   $ hg -q commit -m "foo"
   $ cat .hg/requires
   dotencode
   fncache
   generaldelta
   persistent-nodemap (rust !)
   revlogv1
   store
 
 Check that we can add the sparse-revlog format requirement
   $ hg --config format.sparse-revlog=yes debugupgraderepo --run --quiet
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, store (no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlogv1, store (rust !)
      added: sparserevlog
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   $ cat .hg/requires
   dotencode
   fncache
   generaldelta
   persistent-nodemap (rust !)
   revlogv1
   sparserevlog
   store
 
 Check that we can remove the sparse-revlog format requirement
   $ hg --config format.sparse-revlog=no debugupgraderepo --run --quiet
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, store (no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlogv1, store (rust !)
      removed: sparserevlog
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   $ cat .hg/requires
   dotencode
   fncache
   generaldelta
   persistent-nodemap (rust !)
   revlogv1
   store
 
 #if zstd
 
 Check upgrading to a zstd revlog
 --------------------------------
 
 upgrade
 
   $ hg --config format.revlog-compression=zstd debugupgraderepo --run  --no-backup --quiet
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, store (no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlogv1, store (rust !)
      added: revlog-compression-zstd, sparserevlog
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   $ hg debugformat -v
   format-variant     repo config default
   fncache:            yes    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   persistent-nodemap:  no     no      no (no-rust !)
   persistent-nodemap: yes    yes      no (rust !)
   copies-sdc:          no     no      no
   revlog-v2:           no     no      no
   changelog-v2:        no     no      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib (no-zstd !)
   compression:        zstd   zlib    zstd (zstd !)
   compression-level:  default default default
   $ cat .hg/requires
   dotencode
   fncache
   generaldelta
   persistent-nodemap (rust !)
   revlog-compression-zstd
   revlogv1
   sparserevlog
   store
 
 downgrade
 
   $ hg debugupgraderepo --run --no-backup --quiet
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, sparserevlog, store (no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlogv1, sparserevlog, store (rust !)
      removed: revlog-compression-zstd
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   $ hg debugformat -v
   format-variant     repo config default
   fncache:            yes    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   persistent-nodemap:  no     no      no (no-rust !)
   persistent-nodemap: yes    yes      no (rust !)
   copies-sdc:          no     no      no
   revlog-v2:           no     no      no
   changelog-v2:        no     no      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib (no-zstd !)
   compression:        zlib   zlib    zstd (zstd !)
   compression-level:  default default default
   $ cat .hg/requires
   dotencode
   fncache
   generaldelta
   persistent-nodemap (rust !)
   revlogv1
   sparserevlog
   store
 
 upgrade from hgrc
 
   $ cat >> .hg/hgrc << EOF
   > [format]
   > revlog-compression=zstd
   > EOF
   $ hg debugupgraderepo --run --no-backup --quiet
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, sparserevlog, store (no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlogv1, sparserevlog, store (rust !)
      added: revlog-compression-zstd
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   $ hg debugformat -v
   format-variant     repo config default
   fncache:            yes    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   persistent-nodemap:  no     no      no (no-rust !)
   persistent-nodemap: yes    yes      no (rust !)
   copies-sdc:          no     no      no
   revlog-v2:           no     no      no
   changelog-v2:        no     no      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib (no-zstd !)
   compression:        zstd   zstd    zstd (zstd !)
   compression-level:  default default default
   $ cat .hg/requires
   dotencode
   fncache
   generaldelta
   persistent-nodemap (rust !)
   revlog-compression-zstd
   revlogv1
   sparserevlog
   store
 
 #endif
 
 Check upgrading to a revlog format supporting sidedata
 ------------------------------------------------------
 
 upgrade
 
   $ hg debugsidedata -c 0
   $ hg --config experimental.revlogv2=enable-unstable-format-and-corrupt-my-data debugupgraderepo --run  --no-backup --config "extensions.sidedata=$TESTDIR/testlib/ext-sidedata.py" --quiet
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, store (no-zstd !)
      preserved: dotencode, fncache, generaldelta, revlog-compression-zstd, sparserevlog, store (zstd no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlog-compression-zstd, sparserevlog, store (rust !)
      removed: revlogv1
      added: exp-revlogv2.2 (zstd !)
      added: exp-revlogv2.2, sparserevlog (no-zstd !)
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   $ hg debugformat -v
   format-variant     repo config default
   fncache:            yes    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   persistent-nodemap:  no     no      no (no-rust !)
   persistent-nodemap: yes    yes      no (rust !)
   copies-sdc:          no     no      no
   revlog-v2:          yes     no      no
   changelog-v2:        no     no      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib (no-zstd !)
   compression:        zstd   zstd    zstd (zstd !)
   compression-level:  default default default
   $ cat .hg/requires
   dotencode
   exp-revlogv2.2
   fncache
   generaldelta
   persistent-nodemap (rust !)
   revlog-compression-zstd (zstd !)
   sparserevlog
   store
   $ hg debugsidedata -c 0
   2 sidedata entries
    entry-0001 size 4
    entry-0002 size 32
 
 downgrade
 
   $ hg debugupgraderepo --config experimental.revlogv2=no --run --no-backup --quiet
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, sparserevlog, store (no-zstd !)
      preserved: dotencode, fncache, generaldelta, revlog-compression-zstd, sparserevlog, store (zstd no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlog-compression-zstd, sparserevlog, store (rust !)
      removed: exp-revlogv2.2
      added: revlogv1
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   $ hg debugformat -v
   format-variant     repo config default
   fncache:            yes    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   persistent-nodemap:  no     no      no (no-rust !)
   persistent-nodemap: yes    yes      no (rust !)
   copies-sdc:          no     no      no
   revlog-v2:           no     no      no
   changelog-v2:        no     no      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib (no-zstd !)
   compression:        zstd   zstd    zstd (zstd !)
   compression-level:  default default default
   $ cat .hg/requires
   dotencode
   fncache
   generaldelta
   persistent-nodemap (rust !)
   revlog-compression-zstd (zstd !)
   revlogv1
   sparserevlog
   store
   $ hg debugsidedata -c 0
 
 upgrade from hgrc
 
   $ cat >> .hg/hgrc << EOF
   > [experimental]
   > revlogv2=enable-unstable-format-and-corrupt-my-data
   > EOF
   $ hg debugupgraderepo --run --no-backup --quiet
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, sparserevlog, store (no-zstd !)
      preserved: dotencode, fncache, generaldelta, revlog-compression-zstd, sparserevlog, store (zstd no-rust !)
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlog-compression-zstd, sparserevlog, store (rust !)
      removed: revlogv1
      added: exp-revlogv2.2
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   $ hg debugformat -v
   format-variant     repo config default
   fncache:            yes    yes     yes
+  dirstate-v2:         no     no      no
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   persistent-nodemap:  no     no      no (no-rust !)
   persistent-nodemap: yes    yes      no (rust !)
   copies-sdc:          no     no      no
   revlog-v2:          yes    yes      no
   changelog-v2:        no     no      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib (no-zstd !)
   compression:        zstd   zstd    zstd (zstd !)
   compression-level:  default default default
   $ cat .hg/requires
   dotencode
   exp-revlogv2.2
   fncache
   generaldelta
   persistent-nodemap (rust !)
   revlog-compression-zstd (zstd !)
   sparserevlog
   store
   $ hg debugsidedata -c 0
 
 Demonstrate that nothing to perform upgrade will still run all the way through
 
   $ hg debugupgraderepo --run
   nothing to do
+
+#if rust
+
+Upgrade to dirstate-v2
+
+  $ hg debugformat -v --config format.exp-dirstate-v2=1
+  format-variant     repo config default
+  fncache:            yes    yes     yes
+  dirstate-v2:         no    yes      no
+  dotencode:          yes    yes     yes
+  generaldelta:       yes    yes     yes
+  share-safe:          no     no      no
+  sparserevlog:       yes    yes     yes
+  persistent-nodemap: yes    yes      no
+  copies-sdc:          no     no      no
+  revlog-v2:          yes    yes      no
+  changelog-v2:        no     no      no
+  plain-cl-delta:     yes    yes     yes
+  compression:        zstd   zstd    zstd
+  compression-level:  default default default
+  $ hg debugupgraderepo --config format.exp-dirstate-v2=1 --run
+  upgrade will perform the following actions:
+  
+  requirements
+     preserved: dotencode, exp-revlogv2.2, fncache, generaldelta, persistent-nodemap, revlog-compression-zstd, sparserevlog, store
+     added: exp-dirstate-v2
+  
+  dirstate-v2
+     "hg status" will be faster
+  
+  processed revlogs:
+    - all-filelogs
+    - changelog
+    - manifest
+  
+  beginning upgrade...
+  repository locked and read-only
+  creating temporary repository to stage upgraded data: $TESTTMP/sparserevlogrepo/.hg/upgrade.* (glob)
+  (it is safe to interrupt this process any time before data migration completes)
+  upgrading to dirstate-v2 from v1
+  replaced files will be backed up at $TESTTMP/sparserevlogrepo/.hg/upgradebackup.* (glob)
+  removing temporary repository $TESTTMP/sparserevlogrepo/.hg/upgrade.* (glob)
+  $ ls .hg/upgradebackup.*/dirstate
+  .hg/upgradebackup.*/dirstate (glob)
+  $ hg debugformat -v
+  format-variant     repo config default
+  fncache:            yes    yes     yes
+  dirstate-v2:        yes     no      no
+  dotencode:          yes    yes     yes
+  generaldelta:       yes    yes     yes
+  share-safe:          no     no      no
+  sparserevlog:       yes    yes     yes
+  persistent-nodemap: yes    yes      no
+  copies-sdc:          no     no      no
+  revlog-v2:          yes    yes      no
+  changelog-v2:        no     no      no
+  plain-cl-delta:     yes    yes     yes
+  compression:        zstd   zstd    zstd
+  compression-level:  default default default
+  $ hg status
+  $ dd status=none bs=12 count=1 if=.hg/dirstate
+  dirstate-v2
+
+Downgrade from dirstate-v2
+
+  $ hg debugupgraderepo --run
+  upgrade will perform the following actions:
+  
+  requirements
+     preserved: dotencode, exp-revlogv2.2, fncache, generaldelta, persistent-nodemap, revlog-compression-zstd, sparserevlog, store
+     removed: exp-dirstate-v2
+  
+  processed revlogs:
+    - all-filelogs
+    - changelog
+    - manifest
+  
+  beginning upgrade...
+  repository locked and read-only
+  creating temporary repository to stage upgraded data: $TESTTMP/sparserevlogrepo/.hg/upgrade.* (glob)
+  (it is safe to interrupt this process any time before data migration completes)
+  downgrading from dirstate-v2 to v1
+  replaced files will be backed up at $TESTTMP/sparserevlogrepo/.hg/upgradebackup.* (glob)
+  removing temporary repository $TESTTMP/sparserevlogrepo/.hg/upgrade.* (glob)
+  $ hg debugformat -v
+  format-variant     repo config default
+  fncache:            yes    yes     yes
+  dirstate-v2:         no     no      no
+  dotencode:          yes    yes     yes
+  generaldelta:       yes    yes     yes
+  share-safe:          no     no      no
+  sparserevlog:       yes    yes     yes
+  persistent-nodemap: yes    yes      no
+  copies-sdc:          no     no      no
+  revlog-v2:          yes    yes      no
+  changelog-v2:        no     no      no
+  plain-cl-delta:     yes    yes     yes
+  compression:        zstd   zstd    zstd
+  compression-level:  default default default
+  $ hg status
+
+#endif