copies: fix duplicatecopies() with overlay context

The reasoning for this check is in 78d760aa3607 (duplicatecopies: do
not mark items not in the dirstate as copies, 2013-03-28). The check
was then moved to workingfilectx in 754b5117622f (context: add
workingfilectx.markcopied, 2017-10-15) and no corresponding check was
added later when overlayworkingfilectx was added. Rather than adding
the check there, this patch adds a more generic check on the callers
side and removes the check in workingfilectx.markcopied().

Differential Revision: https://phab.mercurial-scm.org/D6380

mercurial/context.py

@@ -1,2571 +1,2570 b''
 # context.py - changeset and file context objects for mercurial
 #
 # Copyright 2006, 2007 Matt 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 errno
 import filecmp
 import os
 import stat
 
 from .i18n import _
 from .node import (
     addednodeid,
     hex,
     modifiednodeid,
     nullid,
     nullrev,
     short,
     wdirfilenodeids,
     wdirhex,
 )
 from . import (
     dagop,
     encoding,
     error,
     fileset,
     match as matchmod,
     obsolete as obsmod,
     patch,
     pathutil,
     phases,
     pycompat,
     repoview,
     scmutil,
     sparse,
     subrepo,
     subrepoutil,
     util,
 )
 from .utils import (
     dateutil,
     stringutil,
 )
 
 propertycache = util.propertycache
 
 class basectx(object):
     """A basectx object represents the common logic for its children:
     changectx: read-only context that is already present in the repo,
     workingctx: a context that represents the working directory and can
                 be committed,
     memctx: a context that represents changes in-memory and can also
             be committed."""
 
     def __init__(self, repo):
         self._repo = repo
 
     def __bytes__(self):
         return short(self.node())
 
     __str__ = encoding.strmethod(__bytes__)
 
     def __repr__(self):
         return r"<%s %s>" % (type(self).__name__, str(self))
 
     def __eq__(self, other):
         try:
             return type(self) == type(other) and self._rev == other._rev
         except AttributeError:
             return False
 
     def __ne__(self, other):
         return not (self == other)
 
     def __contains__(self, key):
         return key in self._manifest
 
     def __getitem__(self, key):
         return self.filectx(key)
 
     def __iter__(self):
         return iter(self._manifest)
 
     def _buildstatusmanifest(self, status):
         """Builds a manifest that includes the given status results, if this is
         a working copy context. For non-working copy contexts, it just returns
         the normal manifest."""
         return self.manifest()
 
     def _matchstatus(self, other, match):
         """This internal method provides a way for child objects to override the
         match operator.
         """
         return match
 
     def _buildstatus(self, other, s, match, listignored, listclean,
                      listunknown):
         """build a status with respect to another context"""
         # Load earliest manifest first for caching reasons. More specifically,
         # if you have revisions 1000 and 1001, 1001 is probably stored as a
         # delta against 1000. Thus, if you read 1000 first, we'll reconstruct
         # 1000 and cache it so that when you read 1001, we just need to apply a
         # delta to what's in the cache. So that's one full reconstruction + one
         # delta application.
         mf2 = None
         if self.rev() is not None and self.rev() < other.rev():
             mf2 = self._buildstatusmanifest(s)
         mf1 = other._buildstatusmanifest(s)
         if mf2 is None:
             mf2 = self._buildstatusmanifest(s)
 
         modified, added = [], []
         removed = []
         clean = []
         deleted, unknown, ignored = s.deleted, s.unknown, s.ignored
         deletedset = set(deleted)
         d = mf1.diff(mf2, match=match, clean=listclean)
         for fn, value in d.iteritems():
             if fn in deletedset:
                 continue
             if value is None:
                 clean.append(fn)
                 continue
             (node1, flag1), (node2, flag2) = value
             if node1 is None:
                 added.append(fn)
             elif node2 is None:
                 removed.append(fn)
             elif flag1 != flag2:
                 modified.append(fn)
             elif node2 not in wdirfilenodeids:
                 # When comparing files between two commits, we save time by
                 # not comparing the file contents when the nodeids differ.
                 # Note that this means we incorrectly report a reverted change
                 # to a file as a modification.
                 modified.append(fn)
             elif self[fn].cmp(other[fn]):
                 modified.append(fn)
             else:
                 clean.append(fn)
 
         if removed:
             # need to filter files if they are already reported as removed
             unknown = [fn for fn in unknown if fn not in mf1 and
                                                (not match or match(fn))]
             ignored = [fn for fn in ignored if fn not in mf1 and
                                                (not match or match(fn))]
             # if they're deleted, don't report them as removed
             removed = [fn for fn in removed if fn not in deletedset]
 
         return scmutil.status(modified, added, removed, deleted, unknown,
                               ignored, clean)
 
     @propertycache
     def substate(self):
         return subrepoutil.state(self, self._repo.ui)
 
     def subrev(self, subpath):
         return self.substate[subpath][1]
 
     def rev(self):
         return self._rev
     def node(self):
         return self._node
     def hex(self):
         return hex(self.node())
     def manifest(self):
         return self._manifest
     def manifestctx(self):
         return self._manifestctx
     def repo(self):
         return self._repo
     def phasestr(self):
         return phases.phasenames[self.phase()]
     def mutable(self):
         return self.phase() > phases.public
 
     def matchfileset(self, expr, badfn=None):
         return fileset.match(self, expr, badfn=badfn)
 
     def obsolete(self):
         """True if the changeset is obsolete"""
         return self.rev() in obsmod.getrevs(self._repo, 'obsolete')
 
     def extinct(self):
         """True if the changeset is extinct"""
         return self.rev() in obsmod.getrevs(self._repo, 'extinct')
 
     def orphan(self):
         """True if the changeset is not obsolete, but its ancestor is"""
         return self.rev() in obsmod.getrevs(self._repo, 'orphan')
 
     def phasedivergent(self):
         """True if the changeset tries to be a successor of a public changeset
 
         Only non-public and non-obsolete changesets may be phase-divergent.
         """
         return self.rev() in obsmod.getrevs(self._repo, 'phasedivergent')
 
     def contentdivergent(self):
         """Is a successor of a changeset with multiple possible successor sets
 
         Only non-public and non-obsolete changesets may be content-divergent.
         """
         return self.rev() in obsmod.getrevs(self._repo, 'contentdivergent')
 
     def isunstable(self):
         """True if the changeset is either orphan, phase-divergent or
         content-divergent"""
         return self.orphan() or self.phasedivergent() or self.contentdivergent()
 
     def instabilities(self):
         """return the list of instabilities affecting this changeset.
 
         Instabilities are returned as strings. possible values are:
         - orphan,
         - phase-divergent,
         - content-divergent.
         """
         instabilities = []
         if self.orphan():
             instabilities.append('orphan')
         if self.phasedivergent():
             instabilities.append('phase-divergent')
         if self.contentdivergent():
             instabilities.append('content-divergent')
         return instabilities
 
     def parents(self):
         """return contexts for each parent changeset"""
         return self._parents
 
     def p1(self):
         return self._parents[0]
 
     def p2(self):
         parents = self._parents
         if len(parents) == 2:
             return parents[1]
         return self._repo[nullrev]
 
     def _fileinfo(self, path):
         if r'_manifest' in self.__dict__:
             try:
                 return self._manifest[path], self._manifest.flags(path)
             except KeyError:
                 raise error.ManifestLookupError(self._node, path,
                                                 _('not found in manifest'))
         if r'_manifestdelta' in self.__dict__ or path in self.files():
             if path in self._manifestdelta:
                 return (self._manifestdelta[path],
                         self._manifestdelta.flags(path))
         mfl = self._repo.manifestlog
         try:
             node, flag = mfl[self._changeset.manifest].find(path)
         except KeyError:
             raise error.ManifestLookupError(self._node, path,
                                             _('not found in manifest'))
 
         return node, flag
 
     def filenode(self, path):
         return self._fileinfo(path)[0]
 
     def flags(self, path):
         try:
             return self._fileinfo(path)[1]
         except error.LookupError:
             return ''
 
     @propertycache
     def _copies(self):
         p1copies = {}
         p2copies = {}
         p1 = self.p1()
         p2 = self.p2()
         narrowmatch = self._repo.narrowmatch()
         for dst in self.files():
             if not narrowmatch(dst) or dst not in self:
                 continue
             copied = self[dst].renamed()
             if not copied:
                 continue
             src, srcnode = copied
             if src in p1 and p1[src].filenode() == srcnode:
                 p1copies[dst] = src
             elif src in p2 and p2[src].filenode() == srcnode:
                 p2copies[dst] = src
         return p1copies, p2copies
     def p1copies(self):
         return self._copies[0]
     def p2copies(self):
         return self._copies[1]
 
     def sub(self, path, allowcreate=True):
         '''return a subrepo for the stored revision of path, never wdir()'''
         return subrepo.subrepo(self, path, allowcreate=allowcreate)
 
     def nullsub(self, path, pctx):
         return subrepo.nullsubrepo(self, path, pctx)
 
     def workingsub(self, path):
         '''return a subrepo for the stored revision, or wdir if this is a wdir
         context.
         '''
         return subrepo.subrepo(self, path, allowwdir=True)
 
     def match(self, pats=None, include=None, exclude=None, default='glob',
               listsubrepos=False, badfn=None):
         r = self._repo
         return matchmod.match(r.root, r.getcwd(), pats,
                               include, exclude, default,
                               auditor=r.nofsauditor, ctx=self,
                               listsubrepos=listsubrepos, badfn=badfn)
 
     def diff(self, ctx2=None, match=None, changes=None, opts=None,
              losedatafn=None, pathfn=None, copy=None,
              copysourcematch=None, hunksfilterfn=None):
         """Returns a diff generator for the given contexts and matcher"""
         if ctx2 is None:
             ctx2 = self.p1()
         if ctx2 is not None:
             ctx2 = self._repo[ctx2]
         return patch.diff(self._repo, ctx2, self, match=match, changes=changes,
                           opts=opts, losedatafn=losedatafn, pathfn=pathfn,
                           copy=copy, copysourcematch=copysourcematch,
                           hunksfilterfn=hunksfilterfn)
 
     def dirs(self):
         return self._manifest.dirs()
 
     def hasdir(self, dir):
         return self._manifest.hasdir(dir)
 
     def status(self, other=None, match=None, listignored=False,
                listclean=False, listunknown=False, listsubrepos=False):
         """return status of files between two nodes or node and working
         directory.
 
         If other is None, compare this node with working directory.
 
         returns (modified, added, removed, deleted, unknown, ignored, clean)
         """
 
         ctx1 = self
         ctx2 = self._repo[other]
 
         # This next code block is, admittedly, fragile logic that tests for
         # reversing the contexts and wouldn't need to exist if it weren't for
         # the fast (and common) code path of comparing the working directory
         # with its first parent.
         #
         # What we're aiming for here is the ability to call:
         #
         # workingctx.status(parentctx)
         #
         # If we always built the manifest for each context and compared those,
         # then we'd be done. But the special case of the above call means we
         # just copy the manifest of the parent.
         reversed = False
         if (not isinstance(ctx1, changectx)
             and isinstance(ctx2, changectx)):
             reversed = True
             ctx1, ctx2 = ctx2, ctx1
 
         match = self._repo.narrowmatch(match)
         match = ctx2._matchstatus(ctx1, match)
         r = scmutil.status([], [], [], [], [], [], [])
         r = ctx2._buildstatus(ctx1, r, match, listignored, listclean,
                               listunknown)
 
         if reversed:
             # Reverse added and removed. Clear deleted, unknown and ignored as
             # these make no sense to reverse.
             r = scmutil.status(r.modified, r.removed, r.added, [], [], [],
                                r.clean)
 
         if listsubrepos:
             for subpath, sub in scmutil.itersubrepos(ctx1, ctx2):
                 try:
                     rev2 = ctx2.subrev(subpath)
                 except KeyError:
                     # A subrepo that existed in node1 was deleted between
                     # node1 and node2 (inclusive). Thus, ctx2's substate
                     # won't contain that subpath. The best we can do ignore it.
                     rev2 = None
                 submatch = matchmod.subdirmatcher(subpath, match)
                 s = sub.status(rev2, match=submatch, ignored=listignored,
                                clean=listclean, unknown=listunknown,
                                listsubrepos=True)
                 for rfiles, sfiles in zip(r, s):
                     rfiles.extend("%s/%s" % (subpath, f) for f in sfiles)
 
         for l in r:
             l.sort()
 
         return r
 
 class changectx(basectx):
     """A changecontext object makes access to data related to a particular
     changeset convenient. It represents a read-only context already present in
     the repo."""
     def __init__(self, repo, rev, node):
         super(changectx, self).__init__(repo)
         self._rev = rev
         self._node = node
 
     def __hash__(self):
         try:
             return hash(self._rev)
         except AttributeError:
             return id(self)
 
     def __nonzero__(self):
         return self._rev != nullrev
 
     __bool__ = __nonzero__
 
     @propertycache
     def _changeset(self):
         return self._repo.changelog.changelogrevision(self.rev())
 
     @propertycache
     def _manifest(self):
         return self._manifestctx.read()
 
     @property
     def _manifestctx(self):
         return self._repo.manifestlog[self._changeset.manifest]
 
     @propertycache
     def _manifestdelta(self):
         return self._manifestctx.readdelta()
 
     @propertycache
     def _parents(self):
         repo = self._repo
         p1, p2 = repo.changelog.parentrevs(self._rev)
         if p2 == nullrev:
             return [repo[p1]]
         return [repo[p1], repo[p2]]
 
     def changeset(self):
         c = self._changeset
         return (
             c.manifest,
             c.user,
             c.date,
             c.files,
             c.description,
             c.extra,
         )
     def manifestnode(self):
         return self._changeset.manifest
 
     def user(self):
         return self._changeset.user
     def date(self):
         return self._changeset.date
     def files(self):
         return self._changeset.files
     @propertycache
     def _copies(self):
         source = self._repo.ui.config('experimental', 'copies.read-from')
         p1copies = self._changeset.p1copies
         p2copies = self._changeset.p2copies
         # If config says to get copy metadata only from changeset, then return
         # that, defaulting to {} if there was no copy metadata.
         # In compatibility mode, we return copy data from the changeset if
         # it was recorded there, and otherwise we fall back to getting it from
         # the filelogs (below).
         if (source == 'changeset-only' or
             (source == 'compatibility' and p1copies is not None)):
             return p1copies or {}, p2copies or {}
 
         # Otherwise (config said to read only from filelog, or we are in
         # compatiblity mode and there is not data in the changeset), we get
         # the copy metadata from the filelogs.
         return super(changectx, self)._copies
     def description(self):
         return self._changeset.description
     def branch(self):
         return encoding.tolocal(self._changeset.extra.get("branch"))
     def closesbranch(self):
         return 'close' in self._changeset.extra
     def extra(self):
         """Return a dict of extra information."""
         return self._changeset.extra
     def tags(self):
         """Return a list of byte tag names"""
         return self._repo.nodetags(self._node)
     def bookmarks(self):
         """Return a list of byte bookmark names."""
         return self._repo.nodebookmarks(self._node)
     def phase(self):
         return self._repo._phasecache.phase(self._repo, self._rev)
     def hidden(self):
         return self._rev in repoview.filterrevs(self._repo, 'visible')
 
     def isinmemory(self):
         return False
 
     def children(self):
         """return list of changectx contexts for each child changeset.
 
         This returns only the immediate child changesets. Use descendants() to
         recursively walk children.
         """
         c = self._repo.changelog.children(self._node)
         return [self._repo[x] for x in c]
 
     def ancestors(self):
         for a in self._repo.changelog.ancestors([self._rev]):
             yield self._repo[a]
 
     def descendants(self):
         """Recursively yield all children of the changeset.
 
         For just the immediate children, use children()
         """
         for d in self._repo.changelog.descendants([self._rev]):
             yield self._repo[d]
 
     def filectx(self, path, fileid=None, filelog=None):
         """get a file context from this changeset"""
         if fileid is None:
             fileid = self.filenode(path)
         return filectx(self._repo, path, fileid=fileid,
                        changectx=self, filelog=filelog)
 
     def ancestor(self, c2, warn=False):
         """return the "best" ancestor context of self and c2
 
         If there are multiple candidates, it will show a message and check
         merge.preferancestor configuration before falling back to the
         revlog ancestor."""
         # deal with workingctxs
         n2 = c2._node
         if n2 is None:
             n2 = c2._parents[0]._node
         cahs = self._repo.changelog.commonancestorsheads(self._node, n2)
         if not cahs:
             anc = nullid
         elif len(cahs) == 1:
             anc = cahs[0]
         else:
             # experimental config: merge.preferancestor
             for r in self._repo.ui.configlist('merge', 'preferancestor'):
                 try:
                     ctx = scmutil.revsymbol(self._repo, r)
                 except error.RepoLookupError:
                     continue
                 anc = ctx.node()
                 if anc in cahs:
                     break
             else:
                 anc = self._repo.changelog.ancestor(self._node, n2)
             if warn:
                 self._repo.ui.status(
                     (_("note: using %s as ancestor of %s and %s\n") %
                      (short(anc), short(self._node), short(n2))) +
                     ''.join(_("      alternatively, use --config "
                               "merge.preferancestor=%s\n") %
                             short(n) for n in sorted(cahs) if n != anc))
         return self._repo[anc]
 
     def isancestorof(self, other):
         """True if this changeset is an ancestor of other"""
         return self._repo.changelog.isancestorrev(self._rev, other._rev)
 
     def walk(self, match):
         '''Generates matching file names.'''
 
         # Wrap match.bad method to have message with nodeid
         def bad(fn, msg):
             # The manifest doesn't know about subrepos, so don't complain about
             # paths into valid subrepos.
             if any(fn == s or fn.startswith(s + '/')
                    for s in self.substate):
                 return
             match.bad(fn, _('no such file in rev %s') % self)
 
         m = matchmod.badmatch(self._repo.narrowmatch(match), bad)
         return self._manifest.walk(m)
 
     def matches(self, match):
         return self.walk(match)
 
 class basefilectx(object):
     """A filecontext object represents the common logic for its children:
     filectx: read-only access to a filerevision that is already present
              in the repo,
     workingfilectx: a filecontext that represents files from the working
                     directory,
     memfilectx: a filecontext that represents files in-memory,
     """
     @propertycache
     def _filelog(self):
         return self._repo.file(self._path)
 
     @propertycache
     def _changeid(self):
         if r'_changectx' in self.__dict__:
             return self._changectx.rev()
         elif r'_descendantrev' in self.__dict__:
             # this file context was created from a revision with a known
             # descendant, we can (lazily) correct for linkrev aliases
             return self._adjustlinkrev(self._descendantrev)
         else:
             return self._filelog.linkrev(self._filerev)
 
     @propertycache
     def _filenode(self):
         if r'_fileid' in self.__dict__:
             return self._filelog.lookup(self._fileid)
         else:
             return self._changectx.filenode(self._path)
 
     @propertycache
     def _filerev(self):
         return self._filelog.rev(self._filenode)
 
     @propertycache
     def _repopath(self):
         return self._path
 
     def __nonzero__(self):
         try:
             self._filenode
             return True
         except error.LookupError:
             # file is missing
             return False
 
     __bool__ = __nonzero__
 
     def __bytes__(self):
         try:
             return "%s@%s" % (self.path(), self._changectx)
         except error.LookupError:
             return "%s@???" % self.path()
 
     __str__ = encoding.strmethod(__bytes__)
 
     def __repr__(self):
         return r"<%s %s>" % (type(self).__name__, str(self))
 
     def __hash__(self):
         try:
             return hash((self._path, self._filenode))
         except AttributeError:
             return id(self)
 
     def __eq__(self, other):
         try:
             return (type(self) == type(other) and self._path == other._path
                     and self._filenode == other._filenode)
         except AttributeError:
             return False
 
     def __ne__(self, other):
         return not (self == other)
 
     def filerev(self):
         return self._filerev
     def filenode(self):
         return self._filenode
     @propertycache
     def _flags(self):
         return self._changectx.flags(self._path)
     def flags(self):
         return self._flags
     def filelog(self):
         return self._filelog
     def rev(self):
         return self._changeid
     def linkrev(self):
         return self._filelog.linkrev(self._filerev)
     def node(self):
         return self._changectx.node()
     def hex(self):
         return self._changectx.hex()
     def user(self):
         return self._changectx.user()
     def date(self):
         return self._changectx.date()
     def files(self):
         return self._changectx.files()
     def description(self):
         return self._changectx.description()
     def branch(self):
         return self._changectx.branch()
     def extra(self):
         return self._changectx.extra()
     def phase(self):
         return self._changectx.phase()
     def phasestr(self):
         return self._changectx.phasestr()
     def obsolete(self):
         return self._changectx.obsolete()
     def instabilities(self):
         return self._changectx.instabilities()
     def manifest(self):
         return self._changectx.manifest()
     def changectx(self):
         return self._changectx
     def renamed(self):
         return self._copied
     def copysource(self):
         return self._copied and self._copied[0]
     def repo(self):
         return self._repo
     def size(self):
         return len(self.data())
 
     def path(self):
         return self._path
 
     def isbinary(self):
         try:
             return stringutil.binary(self.data())
         except IOError:
             return False
     def isexec(self):
         return 'x' in self.flags()
     def islink(self):
         return 'l' in self.flags()
 
     def isabsent(self):
         """whether this filectx represents a file not in self._changectx
 
         This is mainly for merge code to detect change/delete conflicts. This is
         expected to be True for all subclasses of basectx."""
         return False
 
     _customcmp = False
     def cmp(self, fctx):
         """compare with other file context
 
         returns True if different than fctx.
         """
         if fctx._customcmp:
             return fctx.cmp(self)
 
         if self._filenode is None:
             raise error.ProgrammingError(
                 'filectx.cmp() must be reimplemented if not backed by revlog')
 
         if fctx._filenode is None:
             if self._repo._encodefilterpats:
                 # can't rely on size() because wdir content may be decoded
                 return self._filelog.cmp(self._filenode, fctx.data())
             if self.size() - 4 == fctx.size():
                 # size() can match:
                 # if file data starts with '\1\n', empty metadata block is
                 # prepended, which adds 4 bytes to filelog.size().
                 return self._filelog.cmp(self._filenode, fctx.data())
         if self.size() == fctx.size():
             # size() matches: need to compare content
             return self._filelog.cmp(self._filenode, fctx.data())
 
         # size() differs
         return True
 
     def _adjustlinkrev(self, srcrev, inclusive=False, stoprev=None):
         """return the first ancestor of <srcrev> introducing <fnode>
 
         If the linkrev of the file revision does not point to an ancestor of
         srcrev, we'll walk down the ancestors until we find one introducing
         this file revision.
 
         :srcrev: the changeset revision we search ancestors from
         :inclusive: if true, the src revision will also be checked
         :stoprev: an optional revision to stop the walk at. If no introduction
                   of this file content could be found before this floor
                   revision, the function will returns "None" and stops its
                   iteration.
         """
         repo = self._repo
         cl = repo.unfiltered().changelog
         mfl = repo.manifestlog
         # fetch the linkrev
         lkr = self.linkrev()
         if srcrev == lkr:
             return lkr
         # hack to reuse ancestor computation when searching for renames
         memberanc = getattr(self, '_ancestrycontext', None)
         iteranc = None
         if srcrev is None:
             # wctx case, used by workingfilectx during mergecopy
             revs = [p.rev() for p in self._repo[None].parents()]
             inclusive = True # we skipped the real (revless) source
         else:
             revs = [srcrev]
         if memberanc is None:
             memberanc = iteranc = cl.ancestors(revs, lkr,
                                                inclusive=inclusive)
         # check if this linkrev is an ancestor of srcrev
         if lkr not in memberanc:
             if iteranc is None:
                 iteranc = cl.ancestors(revs, lkr, inclusive=inclusive)
             fnode = self._filenode
             path = self._path
             for a in iteranc:
                 if stoprev is not None and a < stoprev:
                     return None
                 ac = cl.read(a) # get changeset data (we avoid object creation)
                 if path in ac[3]: # checking the 'files' field.
                     # The file has been touched, check if the content is
                     # similar to the one we search for.
                     if fnode == mfl[ac[0]].readfast().get(path):
                         return a
             # In theory, we should never get out of that loop without a result.
             # But if manifest uses a buggy file revision (not children of the
             # one it replaces) we could. Such a buggy situation will likely
             # result is crash somewhere else at to some point.
         return lkr
 
     def isintroducedafter(self, changelogrev):
         """True if a filectx has been introduced after a given floor revision
         """
         if self.linkrev() >= changelogrev:
             return True
         introrev = self._introrev(stoprev=changelogrev)
         if introrev is None:
             return False
         return introrev >= changelogrev
 
     def introrev(self):
         """return the rev of the changeset which introduced this file revision
 
         This method is different from linkrev because it take into account the
         changeset the filectx was created from. It ensures the returned
         revision is one of its ancestors. This prevents bugs from
         'linkrev-shadowing' when a file revision is used by multiple
         changesets.
         """
         return self._introrev()
 
     def _introrev(self, stoprev=None):
         """
         Same as `introrev` but, with an extra argument to limit changelog
         iteration range in some internal usecase.
 
         If `stoprev` is set, the `introrev` will not be searched past that
         `stoprev` revision and "None" might be returned. This is useful to
         limit the iteration range.
         """
         toprev = None
         attrs = vars(self)
         if r'_changeid' in attrs:
             # We have a cached value already
             toprev = self._changeid
         elif r'_changectx' in attrs:
             # We know which changelog entry we are coming from
             toprev = self._changectx.rev()
 
         if toprev is not None:
             return self._adjustlinkrev(toprev, inclusive=True, stoprev=stoprev)
         elif r'_descendantrev' in attrs:
             introrev = self._adjustlinkrev(self._descendantrev, stoprev=stoprev)
             # be nice and cache the result of the computation
             if introrev is not None:
                 self._changeid = introrev
             return introrev
         else:
             return self.linkrev()
 
     def introfilectx(self):
         """Return filectx having identical contents, but pointing to the
         changeset revision where this filectx was introduced"""
         introrev = self.introrev()
         if self.rev() == introrev:
             return self
         return self.filectx(self.filenode(), changeid=introrev)
 
     def _parentfilectx(self, path, fileid, filelog):
         """create parent filectx keeping ancestry info for _adjustlinkrev()"""
         fctx = filectx(self._repo, path, fileid=fileid, filelog=filelog)
         if r'_changeid' in vars(self) or r'_changectx' in vars(self):
             # If self is associated with a changeset (probably explicitly
             # fed), ensure the created filectx is associated with a
             # changeset that is an ancestor of self.changectx.
             # This lets us later use _adjustlinkrev to get a correct link.
             fctx._descendantrev = self.rev()
             fctx._ancestrycontext = getattr(self, '_ancestrycontext', None)
         elif r'_descendantrev' in vars(self):
             # Otherwise propagate _descendantrev if we have one associated.
             fctx._descendantrev = self._descendantrev
             fctx._ancestrycontext = getattr(self, '_ancestrycontext', None)
         return fctx
 
     def parents(self):
         _path = self._path
         fl = self._filelog
         parents = self._filelog.parents(self._filenode)
         pl = [(_path, node, fl) for node in parents if node != nullid]
 
         r = fl.renamed(self._filenode)
         if r:
             # - In the simple rename case, both parent are nullid, pl is empty.
             # - In case of merge, only one of the parent is null id and should
             # be replaced with the rename information. This parent is -always-
             # the first one.
             #
             # As null id have always been filtered out in the previous list
             # comprehension, inserting to 0 will always result in "replacing
             # first nullid parent with rename information.
             pl.insert(0, (r[0], r[1], self._repo.file(r[0])))
 
         return [self._parentfilectx(path, fnode, l) for path, fnode, l in pl]
 
     def p1(self):
         return self.parents()[0]
 
     def p2(self):
         p = self.parents()
         if len(p) == 2:
             return p[1]
         return filectx(self._repo, self._path, fileid=-1, filelog=self._filelog)
 
     def annotate(self, follow=False, skiprevs=None, diffopts=None):
         """Returns a list of annotateline objects for each line in the file
 
         - line.fctx is the filectx of the node where that line was last changed
         - line.lineno is the line number at the first appearance in the managed
           file
         - line.text is the data on that line (including newline character)
         """
         getlog = util.lrucachefunc(lambda x: self._repo.file(x))
 
         def parents(f):
             # Cut _descendantrev here to mitigate the penalty of lazy linkrev
             # adjustment. Otherwise, p._adjustlinkrev() would walk changelog
             # from the topmost introrev (= srcrev) down to p.linkrev() if it
             # isn't an ancestor of the srcrev.
             f._changeid
             pl = f.parents()
 
             # Don't return renamed parents if we aren't following.
             if not follow:
                 pl = [p for p in pl if p.path() == f.path()]
 
             # renamed filectx won't have a filelog yet, so set it
             # from the cache to save time
             for p in pl:
                 if not r'_filelog' in p.__dict__:
                     p._filelog = getlog(p.path())
 
             return pl
 
         # use linkrev to find the first changeset where self appeared
         base = self.introfilectx()
         if getattr(base, '_ancestrycontext', None) is None:
             cl = self._repo.changelog
             if base.rev() is None:
                 # wctx is not inclusive, but works because _ancestrycontext
                 # is used to test filelog revisions
                 ac = cl.ancestors([p.rev() for p in base.parents()],
                                   inclusive=True)
             else:
                 ac = cl.ancestors([base.rev()], inclusive=True)
             base._ancestrycontext = ac
 
         return dagop.annotate(base, parents, skiprevs=skiprevs,
                               diffopts=diffopts)
 
     def ancestors(self, followfirst=False):
         visit = {}
         c = self
         if followfirst:
             cut = 1
         else:
             cut = None
 
         while True:
             for parent in c.parents()[:cut]:
                 visit[(parent.linkrev(), parent.filenode())] = parent
             if not visit:
                 break
             c = visit.pop(max(visit))
             yield c
 
     def decodeddata(self):
         """Returns `data()` after running repository decoding filters.
 
         This is often equivalent to how the data would be expressed on disk.
         """
         return self._repo.wwritedata(self.path(), self.data())
 
 class filectx(basefilectx):
     """A filecontext object makes access to data related to a particular
        filerevision convenient."""
     def __init__(self, repo, path, changeid=None, fileid=None,
                  filelog=None, changectx=None):
         """changeid must be a revision number, if specified.
            fileid can be a file revision or node."""
         self._repo = repo
         self._path = path
 
         assert (changeid is not None
                 or fileid is not None
                 or changectx is not None), (
                     "bad args: changeid=%r, fileid=%r, changectx=%r"
                     % (changeid, fileid, changectx))
 
         if filelog is not None:
             self._filelog = filelog
 
         if changeid is not None:
             self._changeid = changeid
         if changectx is not None:
             self._changectx = changectx
         if fileid is not None:
             self._fileid = fileid
 
     @propertycache
     def _changectx(self):
         try:
             return self._repo[self._changeid]
         except error.FilteredRepoLookupError:
             # Linkrev may point to any revision in the repository.  When the
             # repository is filtered this may lead to `filectx` trying to build
             # `changectx` for filtered revision. In such case we fallback to
             # creating `changectx` on the unfiltered version of the reposition.
             # This fallback should not be an issue because `changectx` from
             # `filectx` are not used in complex operations that care about
             # filtering.
             #
             # This fallback is a cheap and dirty fix that prevent several
             # crashes. It does not ensure the behavior is correct. However the
             # behavior was not correct before filtering either and "incorrect
             # behavior" is seen as better as "crash"
             #
             # Linkrevs have several serious troubles with filtering that are
             # complicated to solve. Proper handling of the issue here should be
             # considered when solving linkrev issue are on the table.
             return self._repo.unfiltered()[self._changeid]
 
     def filectx(self, fileid, changeid=None):
         '''opens an arbitrary revision of the file without
         opening a new filelog'''
         return filectx(self._repo, self._path, fileid=fileid,
                        filelog=self._filelog, changeid=changeid)
 
     def rawdata(self):
         return self._filelog.revision(self._filenode, raw=True)
 
     def rawflags(self):
         """low-level revlog flags"""
         return self._filelog.flags(self._filerev)
 
     def data(self):
         try:
             return self._filelog.read(self._filenode)
         except error.CensoredNodeError:
             if self._repo.ui.config("censor", "policy") == "ignore":
                 return ""
             raise error.Abort(_("censored node: %s") % short(self._filenode),
                              hint=_("set censor.policy to ignore errors"))
 
     def size(self):
         return self._filelog.size(self._filerev)
 
     @propertycache
     def _copied(self):
         """check if file was actually renamed in this changeset revision
 
         If rename logged in file revision, we report copy for changeset only
         if file revisions linkrev points back to the changeset in question
         or both changeset parents contain different file revisions.
         """
 
         renamed = self._filelog.renamed(self._filenode)
         if not renamed:
             return None
 
         if self.rev() == self.linkrev():
             return renamed
 
         name = self.path()
         fnode = self._filenode
         for p in self._changectx.parents():
             try:
                 if fnode == p.filenode(name):
                     return None
             except error.LookupError:
                 pass
         return renamed
 
     def children(self):
         # hard for renames
         c = self._filelog.children(self._filenode)
         return [filectx(self._repo, self._path, fileid=x,
                         filelog=self._filelog) for x in c]
 
 class committablectx(basectx):
     """A committablectx object provides common functionality for a context that
     wants the ability to commit, e.g. workingctx or memctx."""
     def __init__(self, repo, text="", user=None, date=None, extra=None,
                  changes=None, branch=None):
         super(committablectx, self).__init__(repo)
         self._rev = None
         self._node = None
         self._text = text
         if date:
             self._date = dateutil.parsedate(date)
         if user:
             self._user = user
         if changes:
             self._status = changes
 
         self._extra = {}
         if extra:
             self._extra = extra.copy()
         if branch is not None:
             self._extra['branch'] = encoding.fromlocal(branch)
         if not self._extra.get('branch'):
             self._extra['branch'] = 'default'
 
     def __bytes__(self):
         return bytes(self._parents[0]) + "+"
 
     __str__ = encoding.strmethod(__bytes__)
 
     def __nonzero__(self):
         return True
 
     __bool__ = __nonzero__
 
     @propertycache
     def _status(self):
         return self._repo.status()
 
     @propertycache
     def _user(self):
         return self._repo.ui.username()
 
     @propertycache
     def _date(self):
         ui = self._repo.ui
         date = ui.configdate('devel', 'default-date')
         if date is None:
             date = dateutil.makedate()
         return date
 
     def subrev(self, subpath):
         return None
 
     def manifestnode(self):
         return None
     def user(self):
         return self._user or self._repo.ui.username()
     def date(self):
         return self._date
     def description(self):
         return self._text
     def files(self):
         return sorted(self._status.modified + self._status.added +
                       self._status.removed)
     def modified(self):
         return self._status.modified
     def added(self):
         return self._status.added
     def removed(self):
         return self._status.removed
     def deleted(self):
         return self._status.deleted
     def branch(self):
         return encoding.tolocal(self._extra['branch'])
     def closesbranch(self):
         return 'close' in self._extra
     def extra(self):
         return self._extra
 
     def isinmemory(self):
         return False
 
     def tags(self):
         return []
 
     def bookmarks(self):
         b = []
         for p in self.parents():
             b.extend(p.bookmarks())
         return b
 
     def phase(self):
         phase = phases.draft # default phase to draft
         for p in self.parents():
             phase = max(phase, p.phase())
         return phase
 
     def hidden(self):
         return False
 
     def children(self):
         return []
 
     def ancestor(self, c2):
         """return the "best" ancestor context of self and c2"""
         return self._parents[0].ancestor(c2) # punt on two parents for now
 
     def ancestors(self):
         for p in self._parents:
             yield p
         for a in self._repo.changelog.ancestors(
             [p.rev() for p in self._parents]):
             yield self._repo[a]
 
     def markcommitted(self, node):
         """Perform post-commit cleanup necessary after committing this ctx
 
         Specifically, this updates backing stores this working context
         wraps to reflect the fact that the changes reflected by this
         workingctx have been committed.  For example, it marks
         modified and added files as normal in the dirstate.
 
         """
 
     def dirty(self, missing=False, merge=True, branch=True):
         return False
 
 class workingctx(committablectx):
     """A workingctx object makes access to data related to
     the current working directory convenient.
     date - any valid date string or (unixtime, offset), or None.
     user - username string, or None.
     extra - a dictionary of extra values, or None.
     changes - a list of file lists as returned by localrepo.status()
                or None to use the repository status.
     """
     def __init__(self, repo, text="", user=None, date=None, extra=None,
                  changes=None):
         branch = None
         if not extra or 'branch' not in extra:
             try:
                 branch = repo.dirstate.branch()
             except UnicodeDecodeError:
                 raise error.Abort(_('branch name not in UTF-8!'))
         super(workingctx, self).__init__(repo, text, user, date, extra, changes,
                                          branch=branch)
 
     def __iter__(self):
         d = self._repo.dirstate
         for f in d:
             if d[f] != 'r':
                 yield f
 
     def __contains__(self, key):
         return self._repo.dirstate[key] not in "?r"
 
     def hex(self):
         return wdirhex
 
     @propertycache
     def _parents(self):
         p = self._repo.dirstate.parents()
         if p[1] == nullid:
             p = p[:-1]
         # use unfiltered repo to delay/avoid loading obsmarkers
         unfi = self._repo.unfiltered()
         return [changectx(self._repo, unfi.changelog.rev(n), n) for n in p]
 
     def _fileinfo(self, path):
         # populate __dict__['_manifest'] as workingctx has no _manifestdelta
         self._manifest
         return super(workingctx, self)._fileinfo(path)
 
     def _buildflagfunc(self):
         # Create a fallback function for getting file flags when the
         # filesystem doesn't support them
 
         copiesget = self._repo.dirstate.copies().get
         parents = self.parents()
         if len(parents) < 2:
             # when we have one parent, it's easy: copy from parent
             man = parents[0].manifest()
             def func(f):
                 f = copiesget(f, f)
                 return man.flags(f)
         else:
             # merges are tricky: we try to reconstruct the unstored
             # result from the merge (issue1802)
             p1, p2 = parents
             pa = p1.ancestor(p2)
             m1, m2, ma = p1.manifest(), p2.manifest(), pa.manifest()
 
             def func(f):
                 f = copiesget(f, f) # may be wrong for merges with copies
                 fl1, fl2, fla = m1.flags(f), m2.flags(f), ma.flags(f)
                 if fl1 == fl2:
                     return fl1
                 if fl1 == fla:
                     return fl2
                 if fl2 == fla:
                     return fl1
                 return '' # punt for conflicts
 
         return func
 
     @propertycache
     def _flagfunc(self):
         return self._repo.dirstate.flagfunc(self._buildflagfunc)
 
     def flags(self, path):
         if r'_manifest' in self.__dict__:
             try:
                 return self._manifest.flags(path)
             except KeyError:
                 return ''
 
         try:
             return self._flagfunc(path)
         except OSError:
             return ''
 
     def filectx(self, path, filelog=None):
         """get a file context from the working directory"""
         return workingfilectx(self._repo, path, workingctx=self,
                               filelog=filelog)
 
     def dirty(self, missing=False, merge=True, branch=True):
         "check whether a working directory is modified"
         # check subrepos first
         for s in sorted(self.substate):
             if self.sub(s).dirty(missing=missing):
                 return True
         # check current working dir
         return ((merge and self.p2()) or
                 (branch and self.branch() != self.p1().branch()) or
                 self.modified() or self.added() or self.removed() or
                 (missing and self.deleted()))
 
     def add(self, list, prefix=""):
         with self._repo.wlock():
             ui, ds = self._repo.ui, self._repo.dirstate
             uipath = lambda f: ds.pathto(pathutil.join(prefix, f))
             rejected = []
             lstat = self._repo.wvfs.lstat
             for f in list:
                 # ds.pathto() returns an absolute file when this is invoked from
                 # the keyword extension.  That gets flagged as non-portable on
                 # Windows, since it contains the drive letter and colon.
                 scmutil.checkportable(ui, os.path.join(prefix, f))
                 try:
                     st = lstat(f)
                 except OSError:
                     ui.warn(_("%s does not exist!\n") % uipath(f))
                     rejected.append(f)
                     continue
                 limit = ui.configbytes('ui', 'large-file-limit')
                 if limit != 0 and st.st_size > limit:
                     ui.warn(_("%s: up to %d MB of RAM may be required "
                               "to manage this file\n"
                               "(use 'hg revert %s' to cancel the "
                               "pending addition)\n")
                             % (f, 3 * st.st_size // 1000000, uipath(f)))
                 if not (stat.S_ISREG(st.st_mode) or stat.S_ISLNK(st.st_mode)):
                     ui.warn(_("%s not added: only files and symlinks "
                               "supported currently\n") % uipath(f))
                     rejected.append(f)
                 elif ds[f] in 'amn':
                     ui.warn(_("%s already tracked!\n") % uipath(f))
                 elif ds[f] == 'r':
                     ds.normallookup(f)
                 else:
                     ds.add(f)
             return rejected
 
     def forget(self, files, prefix=""):
         with self._repo.wlock():
             ds = self._repo.dirstate
             uipath = lambda f: ds.pathto(pathutil.join(prefix, f))
             rejected = []
             for f in files:
                 if f not in ds:
                     self._repo.ui.warn(_("%s not tracked!\n") % uipath(f))
                     rejected.append(f)
                 elif ds[f] != 'a':
                     ds.remove(f)
                 else:
                     ds.drop(f)
             return rejected
 
     def copy(self, source, dest):
         try:
             st = self._repo.wvfs.lstat(dest)
         except OSError as err:
             if err.errno != errno.ENOENT:
                 raise
             self._repo.ui.warn(_("%s does not exist!\n")
                                % self._repo.dirstate.pathto(dest))
             return
         if not (stat.S_ISREG(st.st_mode) or stat.S_ISLNK(st.st_mode)):
             self._repo.ui.warn(_("copy failed: %s is not a file or a "
                                  "symbolic link\n")
                                % self._repo.dirstate.pathto(dest))
         else:
             with self._repo.wlock():
                 ds = self._repo.dirstate
                 if ds[dest] in '?':
                     ds.add(dest)
                 elif ds[dest] in 'r':
                     ds.normallookup(dest)
                 ds.copy(source, dest)
 
     def match(self, pats=None, include=None, exclude=None, default='glob',
               listsubrepos=False, badfn=None):
         r = self._repo
 
         # Only a case insensitive filesystem needs magic to translate user input
         # to actual case in the filesystem.
         icasefs = not util.fscasesensitive(r.root)
         return matchmod.match(r.root, r.getcwd(), pats, include, exclude,
                               default, auditor=r.auditor, ctx=self,
                               listsubrepos=listsubrepos, badfn=badfn,
                               icasefs=icasefs)
 
     def _filtersuspectsymlink(self, files):
         if not files or self._repo.dirstate._checklink:
             return files
 
         # Symlink placeholders may get non-symlink-like contents
         # via user error or dereferencing by NFS or Samba servers,
         # so we filter out any placeholders that don't look like a
         # symlink
         sane = []
         for f in files:
             if self.flags(f) == 'l':
                 d = self[f].data()
                 if (d == '' or len(d) >= 1024 or '\n' in d
                     or stringutil.binary(d)):
                     self._repo.ui.debug('ignoring suspect symlink placeholder'
                                         ' "%s"\n' % f)
                     continue
             sane.append(f)
         return sane
 
     def _checklookup(self, files):
         # check for any possibly clean files
         if not files:
             return [], [], []
 
         modified = []
         deleted = []
         fixup = []
         pctx = self._parents[0]
         # do a full compare of any files that might have changed
         for f in sorted(files):
             try:
                 # This will return True for a file that got replaced by a
                 # directory in the interim, but fixing that is pretty hard.
                 if (f not in pctx or self.flags(f) != pctx.flags(f)
                     or pctx[f].cmp(self[f])):
                     modified.append(f)
                 else:
                     fixup.append(f)
             except (IOError, OSError):
                 # A file become inaccessible in between? Mark it as deleted,
                 # matching dirstate behavior (issue5584).
                 # The dirstate has more complex behavior around whether a
                 # missing file matches a directory, etc, but we don't need to
                 # bother with that: if f has made it to this point, we're sure
                 # it's in the dirstate.
                 deleted.append(f)
 
         return modified, deleted, fixup
 
     def _poststatusfixup(self, status, fixup):
         """update dirstate for files that are actually clean"""
         poststatus = self._repo.postdsstatus()
         if fixup or poststatus:
             try:
                 oldid = self._repo.dirstate.identity()
 
                 # updating the dirstate is optional
                 # so we don't wait on the lock
                 # wlock can invalidate the dirstate, so cache normal _after_
                 # taking the lock
                 with self._repo.wlock(False):
                     if self._repo.dirstate.identity() == oldid:
                         if fixup:
                             normal = self._repo.dirstate.normal
                             for f in fixup:
                                 normal(f)
                             # write changes out explicitly, because nesting
                             # wlock at runtime may prevent 'wlock.release()'
                             # after this block from doing so for subsequent
                             # changing files
                             tr = self._repo.currenttransaction()
                             self._repo.dirstate.write(tr)
 
                         if poststatus:
                             for ps in poststatus:
                                 ps(self, status)
                     else:
                         # in this case, writing changes out breaks
                         # consistency, because .hg/dirstate was
                         # already changed simultaneously after last
                         # caching (see also issue5584 for detail)
                         self._repo.ui.debug('skip updating dirstate: '
                                             'identity mismatch\n')
             except error.LockError:
                 pass
             finally:
                 # Even if the wlock couldn't be grabbed, clear out the list.
                 self._repo.clearpostdsstatus()
 
     def _dirstatestatus(self, match, ignored=False, clean=False, unknown=False):
         '''Gets the status from the dirstate -- internal use only.'''
         subrepos = []
         if '.hgsub' in self:
             subrepos = sorted(self.substate)
         cmp, s = self._repo.dirstate.status(match, subrepos, ignored=ignored,
                                             clean=clean, unknown=unknown)
 
         # check for any possibly clean files
         fixup = []
         if cmp:
             modified2, deleted2, fixup = self._checklookup(cmp)
             s.modified.extend(modified2)
             s.deleted.extend(deleted2)
 
             if fixup and clean:
                 s.clean.extend(fixup)
 
         self._poststatusfixup(s, fixup)
 
         if match.always():
             # cache for performance
             if s.unknown or s.ignored or s.clean:
                 # "_status" is cached with list*=False in the normal route
                 self._status = scmutil.status(s.modified, s.added, s.removed,
                                               s.deleted, [], [], [])
             else:
                 self._status = s
 
         return s
 
     @propertycache
     def _copies(self):
         p1copies = {}
         p2copies = {}
         parents = self._repo.dirstate.parents()
         p1manifest = self._repo[parents[0]].manifest()
         p2manifest = self._repo[parents[1]].manifest()
         narrowmatch = self._repo.narrowmatch()
         for dst, src in self._repo.dirstate.copies().items():
             if not narrowmatch(dst):
                 continue
             if src in p1manifest:
                 p1copies[dst] = src
             elif src in p2manifest:
                 p2copies[dst] = src
         return p1copies, p2copies
     def p1copies(self):
         return self._copies[0]
     def p2copies(self):
         return self._copies[1]
 
     @propertycache
     def _manifest(self):
         """generate a manifest corresponding to the values in self._status
 
         This reuse the file nodeid from parent, but we use special node
         identifiers for added and modified files. This is used by manifests
         merge to see that files are different and by update logic to avoid
         deleting newly added files.
         """
         return self._buildstatusmanifest(self._status)
 
     def _buildstatusmanifest(self, status):
         """Builds a manifest that includes the given status results."""
         parents = self.parents()
 
         man = parents[0].manifest().copy()
 
         ff = self._flagfunc
         for i, l in ((addednodeid, status.added),
                      (modifiednodeid, status.modified)):
             for f in l:
                 man[f] = i
                 try:
                     man.setflag(f, ff(f))
                 except OSError:
                     pass
 
         for f in status.deleted + status.removed:
             if f in man:
                 del man[f]
 
         return man
 
     def _buildstatus(self, other, s, match, listignored, listclean,
                      listunknown):
         """build a status with respect to another context
 
         This includes logic for maintaining the fast path of status when
         comparing the working directory against its parent, which is to skip
         building a new manifest if self (working directory) is not comparing
         against its parent (repo['.']).
         """
         s = self._dirstatestatus(match, listignored, listclean, listunknown)
         # Filter out symlinks that, in the case of FAT32 and NTFS filesystems,
         # might have accidentally ended up with the entire contents of the file
         # they are supposed to be linking to.
         s.modified[:] = self._filtersuspectsymlink(s.modified)
         if other != self._repo['.']:
             s = super(workingctx, self)._buildstatus(other, s, match,
                                                      listignored, listclean,
                                                      listunknown)
         return s
 
     def _matchstatus(self, other, match):
         """override the match method with a filter for directory patterns
 
         We use inheritance to customize the match.bad method only in cases of
         workingctx since it belongs only to the working directory when
         comparing against the parent changeset.
 
         If we aren't comparing against the working directory's parent, then we
         just use the default match object sent to us.
         """
         if other != self._repo['.']:
             def bad(f, msg):
                 # 'f' may be a directory pattern from 'match.files()',
                 # so 'f not in ctx1' is not enough
                 if f not in other and not other.hasdir(f):
                     self._repo.ui.warn('%s: %s\n' %
                                        (self._repo.dirstate.pathto(f), msg))
             match.bad = bad
         return match
 
     def walk(self, match):
         '''Generates matching file names.'''
         return sorted(self._repo.dirstate.walk(self._repo.narrowmatch(match),
                                                subrepos=sorted(self.substate),
                                                unknown=True, ignored=False))
 
     def matches(self, match):
         match = self._repo.narrowmatch(match)
         ds = self._repo.dirstate
         return sorted(f for f in ds.matches(match) if ds[f] != 'r')
 
     def markcommitted(self, node):
         with self._repo.dirstate.parentchange():
             for f in self.modified() + self.added():
                 self._repo.dirstate.normal(f)
             for f in self.removed():
                 self._repo.dirstate.drop(f)
             self._repo.dirstate.setparents(node)
 
         # write changes out explicitly, because nesting wlock at
         # runtime may prevent 'wlock.release()' in 'repo.commit()'
         # from immediately doing so for subsequent changing files
         self._repo.dirstate.write(self._repo.currenttransaction())
 
         sparse.aftercommit(self._repo, node)
 
 class committablefilectx(basefilectx):
     """A committablefilectx provides common functionality for a file context
     that wants the ability to commit, e.g. workingfilectx or memfilectx."""
     def __init__(self, repo, path, filelog=None, ctx=None):
         self._repo = repo
         self._path = path
         self._changeid = None
         self._filerev = self._filenode = None
 
         if filelog is not None:
             self._filelog = filelog
         if ctx:
             self._changectx = ctx
 
     def __nonzero__(self):
         return True
 
     __bool__ = __nonzero__
 
     def linkrev(self):
         # linked to self._changectx no matter if file is modified or not
         return self.rev()
 
     def renamed(self):
         path = self.copysource()
         if not path:
             return None
         return path, self._changectx._parents[0]._manifest.get(path, nullid)
 
     def parents(self):
         '''return parent filectxs, following copies if necessary'''
         def filenode(ctx, path):
             return ctx._manifest.get(path, nullid)
 
         path = self._path
         fl = self._filelog
         pcl = self._changectx._parents
         renamed = self.renamed()
 
         if renamed:
             pl = [renamed + (None,)]
         else:
             pl = [(path, filenode(pcl[0], path), fl)]
 
         for pc in pcl[1:]:
             pl.append((path, filenode(pc, path), fl))
 
         return [self._parentfilectx(p, fileid=n, filelog=l)
                 for p, n, l in pl if n != nullid]
 
     def children(self):
         return []
 
 class workingfilectx(committablefilectx):
     """A workingfilectx object makes access to data related to a particular
        file in the working directory convenient."""
     def __init__(self, repo, path, filelog=None, workingctx=None):
         super(workingfilectx, self).__init__(repo, path, filelog, workingctx)
 
     @propertycache
     def _changectx(self):
         return workingctx(self._repo)
 
     def data(self):
         return self._repo.wread(self._path)
     def copysource(self):
         return self._repo.dirstate.copied(self._path)
 
     def size(self):
         return self._repo.wvfs.lstat(self._path).st_size
     def date(self):
         t, tz = self._changectx.date()
         try:
             return (self._repo.wvfs.lstat(self._path)[stat.ST_MTIME], tz)
         except OSError as err:
             if err.errno != errno.ENOENT:
                 raise
             return (t, tz)
 
     def exists(self):
         return self._repo.wvfs.exists(self._path)
 
     def lexists(self):
         return self._repo.wvfs.lexists(self._path)
 
     def audit(self):
         return self._repo.wvfs.audit(self._path)
 
     def cmp(self, fctx):
         """compare with other file context
 
         returns True if different than fctx.
         """
         # fctx should be a filectx (not a workingfilectx)
         # invert comparison to reuse the same code path
         return fctx.cmp(self)
 
     def remove(self, ignoremissing=False):
         """wraps unlink for a repo's working directory"""
         rmdir = self._repo.ui.configbool('experimental', 'removeemptydirs')
         self._repo.wvfs.unlinkpath(self._path, ignoremissing=ignoremissing,
                                    rmdir=rmdir)
 
     def write(self, data, flags, backgroundclose=False, **kwargs):
         """wraps repo.wwrite"""
         self._repo.wwrite(self._path, data, flags,
                           backgroundclose=backgroundclose,
                           **kwargs)
 
     def markcopied(self, src):
         """marks this file a copy of `src`"""
-        if self._repo.dirstate[self._path] in "nma":
             self._repo.dirstate.copy(src, self._path)
 
     def clearunknown(self):
         """Removes conflicting items in the working directory so that
         ``write()`` can be called successfully.
         """
         wvfs = self._repo.wvfs
         f = self._path
         wvfs.audit(f)
         if self._repo.ui.configbool('experimental', 'merge.checkpathconflicts'):
             # remove files under the directory as they should already be
             # warned and backed up
             if wvfs.isdir(f) and not wvfs.islink(f):
                 wvfs.rmtree(f, forcibly=True)
             for p in reversed(list(util.finddirs(f))):
                 if wvfs.isfileorlink(p):
                     wvfs.unlink(p)
                     break
         else:
             # don't remove files if path conflicts are not processed
             if wvfs.isdir(f) and not wvfs.islink(f):
                 wvfs.removedirs(f)
 
     def setflags(self, l, x):
         self._repo.wvfs.setflags(self._path, l, x)
 
 class overlayworkingctx(committablectx):
     """Wraps another mutable context with a write-back cache that can be
     converted into a commit context.
 
     self._cache[path] maps to a dict with keys: {
         'exists': bool?
         'date': date?
         'data': str?
         'flags': str?
         'copied': str? (path or None)
     }
     If `exists` is True, `flags` must be non-None and 'date' is non-None. If it
     is `False`, the file was deleted.
     """
 
     def __init__(self, repo):
         super(overlayworkingctx, self).__init__(repo)
         self.clean()
 
     def setbase(self, wrappedctx):
         self._wrappedctx = wrappedctx
         self._parents = [wrappedctx]
         # Drop old manifest cache as it is now out of date.
         # This is necessary when, e.g., rebasing several nodes with one
         # ``overlayworkingctx`` (e.g. with --collapse).
         util.clearcachedproperty(self, '_manifest')
 
     def data(self, path):
         if self.isdirty(path):
             if self._cache[path]['exists']:
                 if self._cache[path]['data'] is not None:
                     return self._cache[path]['data']
                 else:
                     # Must fallback here, too, because we only set flags.
                     return self._wrappedctx[path].data()
             else:
                 raise error.ProgrammingError("No such file or directory: %s" %
                                              path)
         else:
             return self._wrappedctx[path].data()
 
     @propertycache
     def _manifest(self):
         parents = self.parents()
         man = parents[0].manifest().copy()
 
         flag = self._flagfunc
         for path in self.added():
             man[path] = addednodeid
             man.setflag(path, flag(path))
         for path in self.modified():
             man[path] = modifiednodeid
             man.setflag(path, flag(path))
         for path in self.removed():
             del man[path]
         return man
 
     @propertycache
     def _flagfunc(self):
         def f(path):
             return self._cache[path]['flags']
         return f
 
     def files(self):
         return sorted(self.added() + self.modified() + self.removed())
 
     def modified(self):
         return [f for f in self._cache.keys() if self._cache[f]['exists'] and
                 self._existsinparent(f)]
 
     def added(self):
         return [f for f in self._cache.keys() if self._cache[f]['exists'] and
                 not self._existsinparent(f)]
 
     def removed(self):
         return [f for f in self._cache.keys() if
                 not self._cache[f]['exists'] and self._existsinparent(f)]
 
     def p1copies(self):
         copies = self._repo._wrappedctx.p1copies().copy()
         narrowmatch = self._repo.narrowmatch()
         for f in self._cache.keys():
             if not narrowmatch(f):
                 continue
             copies.pop(f, None) # delete if it exists
             source = self._cache[f]['copied']
             if source:
                 copies[f] = source
         return copies
 
     def p2copies(self):
         copies = self._repo._wrappedctx.p2copies().copy()
         narrowmatch = self._repo.narrowmatch()
         for f in self._cache.keys():
             if not narrowmatch(f):
                 continue
             copies.pop(f, None) # delete if it exists
             source = self._cache[f]['copied']
             if source:
                 copies[f] = source
         return copies
 
     def isinmemory(self):
         return True
 
     def filedate(self, path):
         if self.isdirty(path):
             return self._cache[path]['date']
         else:
             return self._wrappedctx[path].date()
 
     def markcopied(self, path, origin):
         self._markdirty(path, exists=True, date=self.filedate(path),
                         flags=self.flags(path), copied=origin)
 
     def copydata(self, path):
         if self.isdirty(path):
             return self._cache[path]['copied']
         else:
             return None
 
     def flags(self, path):
         if self.isdirty(path):
             if self._cache[path]['exists']:
                 return self._cache[path]['flags']
             else:
                 raise error.ProgrammingError("No such file or directory: %s" %
                                              self._path)
         else:
             return self._wrappedctx[path].flags()
 
     def __contains__(self, key):
         if key in self._cache:
             return self._cache[key]['exists']
         return key in self.p1()
 
     def _existsinparent(self, path):
         try:
             # ``commitctx` raises a ``ManifestLookupError`` if a path does not
             # exist, unlike ``workingctx``, which returns a ``workingfilectx``
             # with an ``exists()`` function.
             self._wrappedctx[path]
             return True
         except error.ManifestLookupError:
             return False
 
     def _auditconflicts(self, path):
         """Replicates conflict checks done by wvfs.write().
 
         Since we never write to the filesystem and never call `applyupdates` in
         IMM, we'll never check that a path is actually writable -- e.g., because
         it adds `a/foo`, but `a` is actually a file in the other commit.
         """
         def fail(path, component):
             # p1() is the base and we're receiving "writes" for p2()'s
             # files.
             if 'l' in self.p1()[component].flags():
                 raise error.Abort("error: %s conflicts with symlink %s "
                                   "in %d." % (path, component,
                                               self.p1().rev()))
             else:
                 raise error.Abort("error: '%s' conflicts with file '%s' in "
                                   "%d." % (path, component,
                                            self.p1().rev()))
 
         # Test that each new directory to be created to write this path from p2
         # is not a file in p1.
         components = path.split('/')
         for i in pycompat.xrange(len(components)):
             component = "/".join(components[0:i])
             if component in self:
                 fail(path, component)
 
         # Test the other direction -- that this path from p2 isn't a directory
         # in p1 (test that p1 doesn't have any paths matching `path/*`).
         match = self.match([path], default=b'path')
         matches = self.p1().manifest().matches(match)
         mfiles = matches.keys()
         if len(mfiles) > 0:
             if len(mfiles) == 1 and mfiles[0] == path:
                 return
             # omit the files which are deleted in current IMM wctx
             mfiles = [m for m in mfiles if m in self]
             if not mfiles:
                 return
             raise error.Abort("error: file '%s' cannot be written because "
                               " '%s/' is a directory in %s (containing %d "
                               "entries: %s)"
                               % (path, path, self.p1(), len(mfiles),
                                  ', '.join(mfiles)))
 
     def write(self, path, data, flags='', **kwargs):
         if data is None:
             raise error.ProgrammingError("data must be non-None")
         self._auditconflicts(path)
         self._markdirty(path, exists=True, data=data, date=dateutil.makedate(),
                         flags=flags)
 
     def setflags(self, path, l, x):
         flag = ''
         if l:
             flag = 'l'
         elif x:
             flag = 'x'
         self._markdirty(path, exists=True, date=dateutil.makedate(),
                         flags=flag)
 
     def remove(self, path):
         self._markdirty(path, exists=False)
 
     def exists(self, path):
         """exists behaves like `lexists`, but needs to follow symlinks and
         return False if they are broken.
         """
         if self.isdirty(path):
             # If this path exists and is a symlink, "follow" it by calling
             # exists on the destination path.
             if (self._cache[path]['exists'] and
                         'l' in self._cache[path]['flags']):
                 return self.exists(self._cache[path]['data'].strip())
             else:
                 return self._cache[path]['exists']
 
         return self._existsinparent(path)
 
     def lexists(self, path):
         """lexists returns True if the path exists"""
         if self.isdirty(path):
             return self._cache[path]['exists']
 
         return self._existsinparent(path)
 
     def size(self, path):
         if self.isdirty(path):
             if self._cache[path]['exists']:
                 return len(self._cache[path]['data'])
             else:
                 raise error.ProgrammingError("No such file or directory: %s" %
                                              self._path)
         return self._wrappedctx[path].size()
 
     def tomemctx(self, text, branch=None, extra=None, date=None, parents=None,
                  user=None, editor=None):
         """Converts this ``overlayworkingctx`` into a ``memctx`` ready to be
         committed.
 
         ``text`` is the commit message.
         ``parents`` (optional) are rev numbers.
         """
         # Default parents to the wrapped contexts' if not passed.
         if parents is None:
             parents = self._wrappedctx.parents()
             if len(parents) == 1:
                 parents = (parents[0], None)
 
         # ``parents`` is passed as rev numbers; convert to ``commitctxs``.
         if parents[1] is None:
             parents = (self._repo[parents[0]], None)
         else:
             parents = (self._repo[parents[0]], self._repo[parents[1]])
 
         files = self.files()
         def getfile(repo, memctx, path):
             if self._cache[path]['exists']:
                 return memfilectx(repo, memctx, path,
                                   self._cache[path]['data'],
                                   'l' in self._cache[path]['flags'],
                                   'x' in self._cache[path]['flags'],
                                   self._cache[path]['copied'])
             else:
                 # Returning None, but including the path in `files`, is
                 # necessary for memctx to register a deletion.
                 return None
         return memctx(self._repo, parents, text, files, getfile, date=date,
                       extra=extra, user=user, branch=branch, editor=editor)
 
     def isdirty(self, path):
         return path in self._cache
 
     def isempty(self):
         # We need to discard any keys that are actually clean before the empty
         # commit check.
         self._compact()
         return len(self._cache) == 0
 
     def clean(self):
         self._cache = {}
 
     def _compact(self):
         """Removes keys from the cache that are actually clean, by comparing
         them with the underlying context.
 
         This can occur during the merge process, e.g. by passing --tool :local
         to resolve a conflict.
         """
         keys = []
         # This won't be perfect, but can help performance significantly when
         # using things like remotefilelog.
         scmutil.prefetchfiles(
             self.repo(), [self.p1().rev()],
             scmutil.matchfiles(self.repo(), self._cache.keys()))
 
         for path in self._cache.keys():
             cache = self._cache[path]
             try:
                 underlying = self._wrappedctx[path]
                 if (underlying.data() == cache['data'] and
                             underlying.flags() == cache['flags']):
                     keys.append(path)
             except error.ManifestLookupError:
                 # Path not in the underlying manifest (created).
                 continue
 
         for path in keys:
             del self._cache[path]
         return keys
 
     def _markdirty(self, path, exists, data=None, date=None, flags='',
         copied=None):
         # data not provided, let's see if we already have some; if not, let's
         # grab it from our underlying context, so that we always have data if
         # the file is marked as existing.
         if exists and data is None:
             oldentry = self._cache.get(path) or {}
             data = oldentry.get('data') or self._wrappedctx[path].data()
 
         self._cache[path] = {
             'exists': exists,
             'data': data,
             'date': date,
             'flags': flags,
             'copied': copied,
         }
 
     def filectx(self, path, filelog=None):
         return overlayworkingfilectx(self._repo, path, parent=self,
                                      filelog=filelog)
 
 class overlayworkingfilectx(committablefilectx):
     """Wrap a ``workingfilectx`` but intercepts all writes into an in-memory
     cache, which can be flushed through later by calling ``flush()``."""
 
     def __init__(self, repo, path, filelog=None, parent=None):
         super(overlayworkingfilectx, self).__init__(repo, path, filelog,
                                                     parent)
         self._repo = repo
         self._parent = parent
         self._path = path
 
     def cmp(self, fctx):
         return self.data() != fctx.data()
 
     def changectx(self):
         return self._parent
 
     def data(self):
         return self._parent.data(self._path)
 
     def date(self):
         return self._parent.filedate(self._path)
 
     def exists(self):
         return self.lexists()
 
     def lexists(self):
         return self._parent.exists(self._path)
 
     def copysource(self):
         return self._parent.copydata(self._path)
 
     def size(self):
         return self._parent.size(self._path)
 
     def markcopied(self, origin):
         self._parent.markcopied(self._path, origin)
 
     def audit(self):
         pass
 
     def flags(self):
         return self._parent.flags(self._path)
 
     def setflags(self, islink, isexec):
         return self._parent.setflags(self._path, islink, isexec)
 
     def write(self, data, flags, backgroundclose=False, **kwargs):
         return self._parent.write(self._path, data, flags, **kwargs)
 
     def remove(self, ignoremissing=False):
         return self._parent.remove(self._path)
 
     def clearunknown(self):
         pass
 
 class workingcommitctx(workingctx):
     """A workingcommitctx object makes access to data related to
     the revision being committed convenient.
 
     This hides changes in the working directory, if they aren't
     committed in this context.
     """
     def __init__(self, repo, changes,
                  text="", user=None, date=None, extra=None):
         super(workingcommitctx, self).__init__(repo, text, user, date, extra,
                                                changes)
 
     def _dirstatestatus(self, match, ignored=False, clean=False, unknown=False):
         """Return matched files only in ``self._status``
 
         Uncommitted files appear "clean" via this context, even if
         they aren't actually so in the working directory.
         """
         if clean:
             clean = [f for f in self._manifest if f not in self._changedset]
         else:
             clean = []
         return scmutil.status([f for f in self._status.modified if match(f)],
                               [f for f in self._status.added if match(f)],
                               [f for f in self._status.removed if match(f)],
                               [], [], [], clean)
 
     @propertycache
     def _changedset(self):
         """Return the set of files changed in this context
         """
         changed = set(self._status.modified)
         changed.update(self._status.added)
         changed.update(self._status.removed)
         return changed
 
 def makecachingfilectxfn(func):
     """Create a filectxfn that caches based on the path.
 
     We can't use util.cachefunc because it uses all arguments as the cache
     key and this creates a cycle since the arguments include the repo and
     memctx.
     """
     cache = {}
 
     def getfilectx(repo, memctx, path):
         if path not in cache:
             cache[path] = func(repo, memctx, path)
         return cache[path]
 
     return getfilectx
 
 def memfilefromctx(ctx):
     """Given a context return a memfilectx for ctx[path]
 
     This is a convenience method for building a memctx based on another
     context.
     """
     def getfilectx(repo, memctx, path):
         fctx = ctx[path]
         copysource = fctx.copysource()
         return memfilectx(repo, memctx, path, fctx.data(),
                           islink=fctx.islink(), isexec=fctx.isexec(),
                           copysource=copysource)
 
     return getfilectx
 
 def memfilefrompatch(patchstore):
     """Given a patch (e.g. patchstore object) return a memfilectx
 
     This is a convenience method for building a memctx based on a patchstore.
     """
     def getfilectx(repo, memctx, path):
         data, mode, copysource = patchstore.getfile(path)
         if data is None:
             return None
         islink, isexec = mode
         return memfilectx(repo, memctx, path, data, islink=islink,
                           isexec=isexec, copysource=copysource)
 
     return getfilectx
 
 class memctx(committablectx):
     """Use memctx to perform in-memory commits via localrepo.commitctx().
 
     Revision information is supplied at initialization time while
     related files data and is made available through a callback
     mechanism.  'repo' is the current localrepo, 'parents' is a
     sequence of two parent revisions identifiers (pass None for every
     missing parent), 'text' is the commit message and 'files' lists
     names of files touched by the revision (normalized and relative to
     repository root).
 
     filectxfn(repo, memctx, path) is a callable receiving the
     repository, the current memctx object and the normalized path of
     requested file, relative to repository root. It is fired by the
     commit function for every file in 'files', but calls order is
     undefined. If the file is available in the revision being
     committed (updated or added), filectxfn returns a memfilectx
     object. If the file was removed, filectxfn return None for recent
     Mercurial. Moved files are represented by marking the source file
     removed and the new file added with copy information (see
     memfilectx).
 
     user receives the committer name and defaults to current
     repository username, date is the commit date in any format
     supported by dateutil.parsedate() and defaults to current date, extra
     is a dictionary of metadata or is left empty.
     """
 
     # Mercurial <= 3.1 expects the filectxfn to raise IOError for missing files.
     # Extensions that need to retain compatibility across Mercurial 3.1 can use
     # this field to determine what to do in filectxfn.
     _returnnoneformissingfiles = True
 
     def __init__(self, repo, parents, text, files, filectxfn, user=None,
                  date=None, extra=None, branch=None, editor=False):
         super(memctx, self).__init__(repo, text, user, date, extra,
                                      branch=branch)
         self._rev = None
         self._node = None
         parents = [(p or nullid) for p in parents]
         p1, p2 = parents
         self._parents = [self._repo[p] for p in (p1, p2)]
         files = sorted(set(files))
         self._files = files
         self.substate = {}
 
         if isinstance(filectxfn, patch.filestore):
             filectxfn = memfilefrompatch(filectxfn)
         elif not callable(filectxfn):
             # if store is not callable, wrap it in a function
             filectxfn = memfilefromctx(filectxfn)
 
         # memoizing increases performance for e.g. vcs convert scenarios.
         self._filectxfn = makecachingfilectxfn(filectxfn)
 
         if editor:
             self._text = editor(self._repo, self, [])
             self._repo.savecommitmessage(self._text)
 
     def filectx(self, path, filelog=None):
         """get a file context from the working directory
 
         Returns None if file doesn't exist and should be removed."""
         return self._filectxfn(self._repo, self, path)
 
     def commit(self):
         """commit context to the repo"""
         return self._repo.commitctx(self)
 
     @propertycache
     def _manifest(self):
         """generate a manifest based on the return values of filectxfn"""
 
         # keep this simple for now; just worry about p1
         pctx = self._parents[0]
         man = pctx.manifest().copy()
 
         for f in self._status.modified:
             man[f] = modifiednodeid
 
         for f in self._status.added:
             man[f] = addednodeid
 
         for f in self._status.removed:
             if f in man:
                 del man[f]
 
         return man
 
     @propertycache
     def _status(self):
         """Calculate exact status from ``files`` specified at construction
         """
         man1 = self.p1().manifest()
         p2 = self._parents[1]
         # "1 < len(self._parents)" can't be used for checking
         # existence of the 2nd parent, because "memctx._parents" is
         # explicitly initialized by the list, of which length is 2.
         if p2.node() != nullid:
             man2 = p2.manifest()
             managing = lambda f: f in man1 or f in man2
         else:
             managing = lambda f: f in man1
 
         modified, added, removed = [], [], []
         for f in self._files:
             if not managing(f):
                 added.append(f)
             elif self[f]:
                 modified.append(f)
             else:
                 removed.append(f)
 
         return scmutil.status(modified, added, removed, [], [], [], [])
 
 class memfilectx(committablefilectx):
     """memfilectx represents an in-memory file to commit.
 
     See memctx and committablefilectx for more details.
     """
     def __init__(self, repo, changectx, path, data, islink=False,
                  isexec=False, copysource=None):
         """
         path is the normalized file path relative to repository root.
         data is the file content as a string.
         islink is True if the file is a symbolic link.
         isexec is True if the file is executable.
         copied is the source file path if current file was copied in the
         revision being committed, or None."""
         super(memfilectx, self).__init__(repo, path, None, changectx)
         self._data = data
         if islink:
             self._flags = 'l'
         elif isexec:
             self._flags = 'x'
         else:
             self._flags = ''
         self._copysource = copysource
 
     def copysource(self):
         return self._copysource
 
     def cmp(self, fctx):
         return self.data() != fctx.data()
 
     def data(self):
         return self._data
 
     def remove(self, ignoremissing=False):
         """wraps unlink for a repo's working directory"""
         # need to figure out what to do here
         del self._changectx[self._path]
 
     def write(self, data, flags, **kwargs):
         """wraps repo.wwrite"""
         self._data = data
 
 
 class metadataonlyctx(committablectx):
     """Like memctx but it's reusing the manifest of different commit.
     Intended to be used by lightweight operations that are creating
     metadata-only changes.
 
     Revision information is supplied at initialization time.  'repo' is the
     current localrepo, 'ctx' is original revision which manifest we're reuisng
     'parents' is a sequence of two parent revisions identifiers (pass None for
     every missing parent), 'text' is the commit.
 
     user receives the committer name and defaults to current repository
     username, date is the commit date in any format supported by
     dateutil.parsedate() and defaults to current date, extra is a dictionary of
     metadata or is left empty.
     """
     def __init__(self, repo, originalctx, parents=None, text=None, user=None,
                  date=None, extra=None, editor=False):
         if text is None:
             text = originalctx.description()
         super(metadataonlyctx, self).__init__(repo, text, user, date, extra)
         self._rev = None
         self._node = None
         self._originalctx = originalctx
         self._manifestnode = originalctx.manifestnode()
         if parents is None:
             parents = originalctx.parents()
         else:
             parents = [repo[p] for p in parents if p is not None]
         parents = parents[:]
         while len(parents) < 2:
             parents.append(repo[nullid])
         p1, p2 = self._parents = parents
 
         # sanity check to ensure that the reused manifest parents are
         # manifests of our commit parents
         mp1, mp2 = self.manifestctx().parents
         if p1 != nullid and p1.manifestnode() != mp1:
             raise RuntimeError(r"can't reuse the manifest: its p1 "
                                r"doesn't match the new ctx p1")
         if p2 != nullid and p2.manifestnode() != mp2:
             raise RuntimeError(r"can't reuse the manifest: "
                                r"its p2 doesn't match the new ctx p2")
 
         self._files = originalctx.files()
         self.substate = {}
 
         if editor:
             self._text = editor(self._repo, self, [])
             self._repo.savecommitmessage(self._text)
 
     def manifestnode(self):
         return self._manifestnode
 
     @property
     def _manifestctx(self):
         return self._repo.manifestlog[self._manifestnode]
 
     def filectx(self, path, filelog=None):
         return self._originalctx.filectx(path, filelog=filelog)
 
     def commit(self):
         """commit context to the repo"""
         return self._repo.commitctx(self)
 
     @property
     def _manifest(self):
         return self._originalctx.manifest()
 
     @propertycache
     def _status(self):
         """Calculate exact status from ``files`` specified in the ``origctx``
         and parents manifests.
         """
         man1 = self.p1().manifest()
         p2 = self._parents[1]
         # "1 < len(self._parents)" can't be used for checking
         # existence of the 2nd parent, because "metadataonlyctx._parents" is
         # explicitly initialized by the list, of which length is 2.
         if p2.node() != nullid:
             man2 = p2.manifest()
             managing = lambda f: f in man1 or f in man2
         else:
             managing = lambda f: f in man1
 
         modified, added, removed = [], [], []
         for f in self._files:
             if not managing(f):
                 added.append(f)
             elif f in self:
                 modified.append(f)
             else:
                 removed.append(f)
 
         return scmutil.status(modified, added, removed, [], [], [], [])
 
 class arbitraryfilectx(object):
     """Allows you to use filectx-like functions on a file in an arbitrary
     location on disk, possibly not in the working directory.
     """
     def __init__(self, path, repo=None):
         # Repo is optional because contrib/simplemerge uses this class.
         self._repo = repo
         self._path = path
 
     def cmp(self, fctx):
         # filecmp follows symlinks whereas `cmp` should not, so skip the fast
         # path if either side is a symlink.
         symlinks = ('l' in self.flags() or 'l' in fctx.flags())
         if not symlinks and isinstance(fctx, workingfilectx) and self._repo:
             # Add a fast-path for merge if both sides are disk-backed.
             # Note that filecmp uses the opposite return values (True if same)
             # from our cmp functions (True if different).
             return not filecmp.cmp(self.path(), self._repo.wjoin(fctx.path()))
         return self.data() != fctx.data()
 
     def path(self):
         return self._path
 
     def flags(self):
         return ''
 
     def data(self):
         return util.readfile(self._path)
 
     def decodeddata(self):
         with open(self._path, "rb") as f:
             return f.read()
 
     def remove(self):
         util.unlink(self._path)
 
     def write(self, data, flags, **kwargs):
         assert not flags
         with open(self._path, "wb") as f:
             f.write(data)

mercurial/copies.py

@@ -1,801 +1,802 b''
 # copies.py - copy detection for Mercurial
 #
 # Copyright 2008 Matt 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 heapq
 import os
 
 from .i18n import _
 
 from . import (
     match as matchmod,
     node,
     pathutil,
     util,
 )
 from .utils import (
     stringutil,
 )
 
 def _findlimit(repo, ctxa, ctxb):
     """
     Find the last revision that needs to be checked to ensure that a full
     transitive closure for file copies can be properly calculated.
     Generally, this means finding the earliest revision number that's an
     ancestor of a or b but not both, except when a or b is a direct descendent
     of the other, in which case we can return the minimum revnum of a and b.
     """
 
     # basic idea:
     # - mark a and b with different sides
     # - if a parent's children are all on the same side, the parent is
     #   on that side, otherwise it is on no side
     # - walk the graph in topological order with the help of a heap;
     #   - add unseen parents to side map
     #   - clear side of any parent that has children on different sides
     #   - track number of interesting revs that might still be on a side
     #   - track the lowest interesting rev seen
     #   - quit when interesting revs is zero
 
     cl = repo.changelog
     wdirparents = None
     a = ctxa.rev()
     b = ctxb.rev()
     if a is None:
         wdirparents = (ctxa.p1(), ctxa.p2())
         a = node.wdirrev
     if b is None:
         assert not wdirparents
         wdirparents = (ctxb.p1(), ctxb.p2())
         b = node.wdirrev
 
     side = {a: -1, b: 1}
     visit = [-a, -b]
     heapq.heapify(visit)
     interesting = len(visit)
     limit = node.wdirrev
 
     while interesting:
         r = -heapq.heappop(visit)
         if r == node.wdirrev:
             parents = [pctx.rev() for pctx in wdirparents]
         else:
             parents = cl.parentrevs(r)
         if parents[1] == node.nullrev:
             parents = parents[:1]
         for p in parents:
             if p not in side:
                 # first time we see p; add it to visit
                 side[p] = side[r]
                 if side[p]:
                     interesting += 1
                 heapq.heappush(visit, -p)
             elif side[p] and side[p] != side[r]:
                 # p was interesting but now we know better
                 side[p] = 0
                 interesting -= 1
         if side[r]:
             limit = r # lowest rev visited
             interesting -= 1
 
     # Consider the following flow (see test-commit-amend.t under issue4405):
     # 1/ File 'a0' committed
     # 2/ File renamed from 'a0' to 'a1' in a new commit (call it 'a1')
     # 3/ Move back to first commit
     # 4/ Create a new commit via revert to contents of 'a1' (call it 'a1-amend')
     # 5/ Rename file from 'a1' to 'a2' and commit --amend 'a1-msg'
     #
     # During the amend in step five, we will be in this state:
     #
     # @  3 temporary amend commit for a1-amend
     # |
     # o  2 a1-amend
     # |
     # | o  1 a1
     # |/
     # o  0 a0
     #
     # When _findlimit is called, a and b are revs 3 and 0, so limit will be 2,
     # yet the filelog has the copy information in rev 1 and we will not look
     # back far enough unless we also look at the a and b as candidates.
     # This only occurs when a is a descendent of b or visa-versa.
     return min(limit, a, b)
 
 def _chain(src, dst, a, b):
     """chain two sets of copies 'a' and 'b'"""
 
     # When chaining copies in 'a' (from 'src' via some other commit 'mid') with
     # copies in 'b' (from 'mid' to 'dst'), we can get the different cases in the
     # following table (not including trivial cases). For example, case 2 is
     # where a file existed in 'src' and remained under that name in 'mid' and
     # then was renamed between 'mid' and 'dst'.
     #
     # case src mid dst result
     #   1   x   y   -    -
     #   2   x   y   y   x->y
     #   3   x   y   x    -
     #   4   x   y   z   x->z
     #   5   -   x   y    -
     #   6   x   x   y   x->y
 
     # Initialize result ('t') from 'a'. This catches cases 1 & 2. We'll remove
     # case 1 later. We'll also catch cases 3 & 4 here. Case 4 will be
     # overwritten later, and case 3 will be removed later.
     t = a.copy()
     for k, v in b.iteritems():
         if v in t:
             # Found a chain, i.e. cases 3 & 4. We'll remove case 3 later.
             t[k] = t[v]
         else:
             # Renamed only in 'b', i.e. cases 5 & 6. We'll remove case 5 later.
             t[k] = v
 
     for k, v in list(t.items()):
         # remove copies from files that didn't exist, i.e. case 5
         if v not in src:
             del t[k]
         # remove criss-crossed copies, i.e. case 3
         elif k in src and v in dst:
             del t[k]
         # remove copies to files that were then removed, i.e. case 1
         # and file 'y' in cases 3 & 4 (in case of rename)
         elif k not in dst:
             del t[k]
 
     return t
 
 def _tracefile(fctx, am, limit):
     """return file context that is the ancestor of fctx present in ancestor
     manifest am, stopping after the first ancestor lower than limit"""
 
     for f in fctx.ancestors():
         if am.get(f.path(), None) == f.filenode():
             return f
         if not f.isintroducedafter(limit):
             return None
 
 def _dirstatecopies(repo, match=None):
     ds = repo.dirstate
     c = ds.copies().copy()
     for k in list(c):
         if ds[k] not in 'anm' or (match and not match(k)):
             del c[k]
     return c
 
 def _computeforwardmissing(a, b, match=None):
     """Computes which files are in b but not a.
     This is its own function so extensions can easily wrap this call to see what
     files _forwardcopies is about to process.
     """
     ma = a.manifest()
     mb = b.manifest()
     return mb.filesnotin(ma, match=match)
 
 def usechangesetcentricalgo(repo):
     """Checks if we should use changeset-centric copy algorithms"""
     return (repo.ui.config('experimental', 'copies.read-from') in
             ('changeset-only', 'compatibility'))
 
 def _committedforwardcopies(a, b, match):
     """Like _forwardcopies(), but b.rev() cannot be None (working copy)"""
     # files might have to be traced back to the fctx parent of the last
     # one-side-only changeset, but not further back than that
     repo = a._repo
 
     if usechangesetcentricalgo(repo):
         return _changesetforwardcopies(a, b, match)
 
     debug = repo.ui.debugflag and repo.ui.configbool('devel', 'debug.copies')
     dbg = repo.ui.debug
     if debug:
         dbg('debug.copies:    looking into rename from %s to %s\n'
             % (a, b))
     limit = _findlimit(repo, a, b)
     if debug:
         dbg('debug.copies:      search limit: %d\n' % limit)
     am = a.manifest()
 
     # find where new files came from
     # we currently don't try to find where old files went, too expensive
     # this means we can miss a case like 'hg rm b; hg cp a b'
     cm = {}
 
     # Computing the forward missing is quite expensive on large manifests, since
     # it compares the entire manifests. We can optimize it in the common use
     # case of computing what copies are in a commit versus its parent (like
     # during a rebase or histedit). Note, we exclude merge commits from this
     # optimization, since the ctx.files() for a merge commit is not correct for
     # this comparison.
     forwardmissingmatch = match
     if b.p1() == a and b.p2().node() == node.nullid:
         filesmatcher = matchmod.exact(b.files())
         forwardmissingmatch = matchmod.intersectmatchers(match, filesmatcher)
     missing = _computeforwardmissing(a, b, match=forwardmissingmatch)
 
     ancestrycontext = a._repo.changelog.ancestors([b.rev()], inclusive=True)
 
     if debug:
         dbg('debug.copies:      missing files to search: %d\n' % len(missing))
 
     for f in sorted(missing):
         if debug:
             dbg('debug.copies:        tracing file: %s\n' % f)
         fctx = b[f]
         fctx._ancestrycontext = ancestrycontext
 
         if debug:
             start = util.timer()
         ofctx = _tracefile(fctx, am, limit)
         if ofctx:
             if debug:
                 dbg('debug.copies:          rename of: %s\n' % ofctx._path)
             cm[f] = ofctx.path()
         if debug:
             dbg('debug.copies:          time: %f seconds\n'
                 % (util.timer() - start))
     return cm
 
 def _changesetforwardcopies(a, b, match):
     if a.rev() == node.nullrev:
         return {}
 
     repo = a.repo()
     children = {}
     cl = repo.changelog
     missingrevs = cl.findmissingrevs(common=[a.rev()], heads=[b.rev()])
     for r in missingrevs:
         for p in cl.parentrevs(r):
             if p == node.nullrev:
                 continue
             if p not in children:
                 children[p] = [r]
             else:
                 children[p].append(r)
 
     roots = set(children) - set(missingrevs)
     # 'work' contains 3-tuples of a (revision number, parent number, copies).
     # The parent number is only used for knowing which parent the copies dict
     # came from.
     work = [(r, 1, {}) for r in roots]
     heapq.heapify(work)
     while work:
         r, i1, copies1 = heapq.heappop(work)
         if work and work[0][0] == r:
             # We are tracing copies from both parents
             r, i2, copies2 = heapq.heappop(work)
             copies = {}
             ctx = repo[r]
             p1man, p2man = ctx.p1().manifest(), ctx.p2().manifest()
             allcopies = set(copies1) | set(copies2)
             # TODO: perhaps this filtering should be done as long as ctx
             # is merge, whether or not we're tracing from both parent.
             for dst in allcopies:
                 if not match(dst):
                     continue
                 if dst not in copies2:
                     # Copied on p1 side: mark as copy from p1 side if it didn't
                     # already exist on p2 side
                     if dst not in p2man:
                         copies[dst] = copies1[dst]
                 elif dst not in copies1:
                     # Copied on p2 side: mark as copy from p2 side if it didn't
                     # already exist on p1 side
                     if dst not in p1man:
                         copies[dst] = copies2[dst]
                 else:
                     # Copied on both sides: mark as copy from p1 side
                     copies[dst] = copies1[dst]
         else:
             copies = copies1
         if r == b.rev():
             return copies
         for c in children[r]:
             childctx = repo[c]
             if r == childctx.p1().rev():
                 parent = 1
                 childcopies = childctx.p1copies()
             else:
                 assert r == childctx.p2().rev()
                 parent = 2
                 childcopies = childctx.p2copies()
             if not match.always():
                 childcopies = {dst: src for dst, src in childcopies.items()
                                if match(dst)}
             childcopies = _chain(a, childctx, copies, childcopies)
             heapq.heappush(work, (c, parent, childcopies))
     assert False
 
 def _forwardcopies(a, b, match=None):
     """find {dst@b: src@a} copy mapping where a is an ancestor of b"""
 
     match = a.repo().narrowmatch(match)
     # check for working copy
     if b.rev() is None:
         if a == b.p1():
             # short-circuit to avoid issues with merge states
             return _dirstatecopies(b._repo, match)
 
         cm = _committedforwardcopies(a, b.p1(), match)
         # combine copies from dirstate if necessary
         return _chain(a, b, cm, _dirstatecopies(b._repo, match))
     return _committedforwardcopies(a, b, match)
 
 def _backwardrenames(a, b, match):
     if a._repo.ui.config('experimental', 'copytrace') == 'off':
         return {}
 
     # Even though we're not taking copies into account, 1:n rename situations
     # can still exist (e.g. hg cp a b; hg mv a c). In those cases we
     # arbitrarily pick one of the renames.
     # We don't want to pass in "match" here, since that would filter
     # the destination by it. Since we're reversing the copies, we want
     # to filter the source instead.
     f = _forwardcopies(b, a)
     r = {}
     for k, v in sorted(f.iteritems()):
         if match and not match(v):
             continue
         # remove copies
         if v in a:
             continue
         r[v] = k
     return r
 
 def pathcopies(x, y, match=None):
     """find {dst@y: src@x} copy mapping for directed compare"""
     repo = x._repo
     debug = repo.ui.debugflag and repo.ui.configbool('devel', 'debug.copies')
     if debug:
         repo.ui.debug('debug.copies: searching copies from %s to %s\n'
                       % (x, y))
     if x == y or not x or not y:
         return {}
     a = y.ancestor(x)
     if a == x:
         if debug:
             repo.ui.debug('debug.copies: search mode: forward\n')
         return _forwardcopies(x, y, match=match)
     if a == y:
         if debug:
             repo.ui.debug('debug.copies: search mode: backward\n')
         return _backwardrenames(x, y, match=match)
     if debug:
         repo.ui.debug('debug.copies: search mode: combined\n')
     return _chain(x, y, _backwardrenames(x, a, match=match),
                   _forwardcopies(a, y, match=match))
 
 def mergecopies(repo, c1, c2, base):
     """
     Finds moves and copies between context c1 and c2 that are relevant for
     merging. 'base' will be used as the merge base.
 
     Copytracing is used in commands like rebase, merge, unshelve, etc to merge
     files that were moved/ copied in one merge parent and modified in another.
     For example:
 
     o          ---> 4 another commit
     |
     |   o      ---> 3 commit that modifies a.txt
     |  /
     o /        ---> 2 commit that moves a.txt to b.txt
     |/
     o          ---> 1 merge base
 
     If we try to rebase revision 3 on revision 4, since there is no a.txt in
     revision 4, and if user have copytrace disabled, we prints the following
     message:
 
     ```other changed <file> which local deleted```
 
     Returns five dicts: "copy", "movewithdir", "diverge", "renamedelete" and
     "dirmove".
 
     "copy" is a mapping from destination name -> source name,
     where source is in c1 and destination is in c2 or vice-versa.
 
     "movewithdir" is a mapping from source name -> destination name,
     where the file at source present in one context but not the other
     needs to be moved to destination by the merge process, because the
     other context moved the directory it is in.
 
     "diverge" is a mapping of source name -> list of destination names
     for divergent renames.
 
     "renamedelete" is a mapping of source name -> list of destination
     names for files deleted in c1 that were renamed in c2 or vice-versa.
 
     "dirmove" is a mapping of detected source dir -> destination dir renames.
     This is needed for handling changes to new files previously grafted into
     renamed directories.
 
     This function calls different copytracing algorithms based on config.
     """
     # avoid silly behavior for update from empty dir
     if not c1 or not c2 or c1 == c2:
         return {}, {}, {}, {}, {}
 
     narrowmatch = c1.repo().narrowmatch()
 
     # avoid silly behavior for parent -> working dir
     if c2.node() is None and c1.node() == repo.dirstate.p1():
         return _dirstatecopies(repo, narrowmatch), {}, {}, {}, {}
 
     copytracing = repo.ui.config('experimental', 'copytrace')
     if stringutil.parsebool(copytracing) is False:
         # stringutil.parsebool() returns None when it is unable to parse the
         # value, so we should rely on making sure copytracing is on such cases
         return {}, {}, {}, {}, {}
 
     if usechangesetcentricalgo(repo):
         # The heuristics don't make sense when we need changeset-centric algos
         return _fullcopytracing(repo, c1, c2, base)
 
     # Copy trace disabling is explicitly below the node == p1 logic above
     # because the logic above is required for a simple copy to be kept across a
     # rebase.
     if copytracing == 'heuristics':
         # Do full copytracing if only non-public revisions are involved as
         # that will be fast enough and will also cover the copies which could
         # be missed by heuristics
         if _isfullcopytraceable(repo, c1, base):
             return _fullcopytracing(repo, c1, c2, base)
         return _heuristicscopytracing(repo, c1, c2, base)
     else:
         return _fullcopytracing(repo, c1, c2, base)
 
 def _isfullcopytraceable(repo, c1, base):
     """ Checks that if base, source and destination are all no-public branches,
     if yes let's use the full copytrace algorithm for increased capabilities
     since it will be fast enough.
 
     `experimental.copytrace.sourcecommitlimit` can be used to set a limit for
     number of changesets from c1 to base such that if number of changesets are
     more than the limit, full copytracing algorithm won't be used.
     """
     if c1.rev() is None:
         c1 = c1.p1()
     if c1.mutable() and base.mutable():
         sourcecommitlimit = repo.ui.configint('experimental',
                                               'copytrace.sourcecommitlimit')
         commits = len(repo.revs('%d::%d', base.rev(), c1.rev()))
         return commits < sourcecommitlimit
     return False
 
 def _checksinglesidecopies(src, dsts1, m1, m2, mb, c2, base,
                            copy, renamedelete):
     if src not in m2:
         # deleted on side 2
         if src not in m1:
             # renamed on side 1, deleted on side 2
             renamedelete[src] = dsts1
     elif m2[src] != mb[src]:
         if not _related(c2[src], base[src]):
             return
         # modified on side 2
         for dst in dsts1:
             if dst not in m2:
                 # dst not added on side 2 (handle as regular
                 # "both created" case in manifestmerge otherwise)
                 copy[dst] = src
 
 def _fullcopytracing(repo, c1, c2, base):
     """ The full copytracing algorithm which finds all the new files that were
     added from merge base up to the top commit and for each file it checks if
     this file was copied from another file.
 
     This is pretty slow when a lot of changesets are involved but will track all
     the copies.
     """
     m1 = c1.manifest()
     m2 = c2.manifest()
     mb = base.manifest()
 
     copies1 = pathcopies(base, c1)
     copies2 = pathcopies(base, c2)
 
     inversecopies1 = {}
     inversecopies2 = {}
     for dst, src in copies1.items():
         inversecopies1.setdefault(src, []).append(dst)
     for dst, src in copies2.items():
         inversecopies2.setdefault(src, []).append(dst)
 
     copy = {}
     diverge = {}
     renamedelete = {}
     allsources = set(inversecopies1) | set(inversecopies2)
     for src in allsources:
         dsts1 = inversecopies1.get(src)
         dsts2 = inversecopies2.get(src)
         if dsts1 and dsts2:
             # copied/renamed on both sides
             if src not in m1 and src not in m2:
                 # renamed on both sides
                 dsts1 = set(dsts1)
                 dsts2 = set(dsts2)
                 # If there's some overlap in the rename destinations, we
                 # consider it not divergent. For example, if side 1 copies 'a'
                 # to 'b' and 'c' and deletes 'a', and side 2 copies 'a' to 'c'
                 # and 'd' and deletes 'a'.
                 if dsts1 & dsts2:
                     for dst in (dsts1 & dsts2):
                         copy[dst] = src
                 else:
                     diverge[src] = sorted(dsts1 | dsts2)
             elif src in m1 and src in m2:
                 # copied on both sides
                 dsts1 = set(dsts1)
                 dsts2 = set(dsts2)
                 for dst in (dsts1 & dsts2):
                     copy[dst] = src
             # TODO: Handle cases where it was renamed on one side and copied
             # on the other side
         elif dsts1:
             # copied/renamed only on side 1
             _checksinglesidecopies(src, dsts1, m1, m2, mb, c2, base,
                                    copy, renamedelete)
         elif dsts2:
             # copied/renamed only on side 2
             _checksinglesidecopies(src, dsts2, m2, m1, mb, c1, base,
                                    copy, renamedelete)
 
     renamedeleteset = set()
     divergeset = set()
     for dsts in diverge.values():
         divergeset.update(dsts)
     for dsts in renamedelete.values():
         renamedeleteset.update(dsts)
 
     # find interesting file sets from manifests
     addedinm1 = m1.filesnotin(mb, repo.narrowmatch())
     addedinm2 = m2.filesnotin(mb, repo.narrowmatch())
     u1 = sorted(addedinm1 - addedinm2)
     u2 = sorted(addedinm2 - addedinm1)
 
     header = "  unmatched files in %s"
     if u1:
         repo.ui.debug("%s:\n   %s\n" % (header % 'local', "\n   ".join(u1)))
     if u2:
         repo.ui.debug("%s:\n   %s\n" % (header % 'other', "\n   ".join(u2)))
 
     fullcopy = copies1.copy()
     fullcopy.update(copies2)
     if not fullcopy:
         return copy, {}, diverge, renamedelete, {}
 
     if repo.ui.debugflag:
         repo.ui.debug("  all copies found (* = to merge, ! = divergent, "
                       "% = renamed and deleted):\n")
         for f in sorted(fullcopy):
             note = ""
             if f in copy:
                 note += "*"
             if f in divergeset:
                 note += "!"
             if f in renamedeleteset:
                 note += "%"
             repo.ui.debug("   src: '%s' -> dst: '%s' %s\n" % (fullcopy[f], f,
                                                               note))
     del divergeset
 
     repo.ui.debug("  checking for directory renames\n")
 
     # generate a directory move map
     d1, d2 = c1.dirs(), c2.dirs()
     # Hack for adding '', which is not otherwise added, to d1 and d2
     d1.addpath('/')
     d2.addpath('/')
     invalid = set()
     dirmove = {}
 
     # examine each file copy for a potential directory move, which is
     # when all the files in a directory are moved to a new directory
     for dst, src in fullcopy.iteritems():
         dsrc, ddst = pathutil.dirname(src), pathutil.dirname(dst)
         if dsrc in invalid:
             # already seen to be uninteresting
             continue
         elif dsrc in d1 and ddst in d1:
             # directory wasn't entirely moved locally
             invalid.add(dsrc)
         elif dsrc in d2 and ddst in d2:
             # directory wasn't entirely moved remotely
             invalid.add(dsrc)
         elif dsrc in dirmove and dirmove[dsrc] != ddst:
             # files from the same directory moved to two different places
             invalid.add(dsrc)
         else:
             # looks good so far
             dirmove[dsrc] = ddst
 
     for i in invalid:
         if i in dirmove:
             del dirmove[i]
     del d1, d2, invalid
 
     if not dirmove:
         return copy, {}, diverge, renamedelete, {}
 
     dirmove = {k + "/": v + "/" for k, v in dirmove.iteritems()}
 
     for d in dirmove:
         repo.ui.debug("   discovered dir src: '%s' -> dst: '%s'\n" %
                       (d, dirmove[d]))
 
     movewithdir = {}
     # check unaccounted nonoverlapping files against directory moves
     for f in u1 + u2:
         if f not in fullcopy:
             for d in dirmove:
                 if f.startswith(d):
                     # new file added in a directory that was moved, move it
                     df = dirmove[d] + f[len(d):]
                     if df not in copy:
                         movewithdir[f] = df
                         repo.ui.debug(("   pending file src: '%s' -> "
                                        "dst: '%s'\n") % (f, df))
                     break
 
     return copy, movewithdir, diverge, renamedelete, dirmove
 
 def _heuristicscopytracing(repo, c1, c2, base):
     """ Fast copytracing using filename heuristics
 
     Assumes that moves or renames are of following two types:
 
     1) Inside a directory only (same directory name but different filenames)
     2) Move from one directory to another
                     (same filenames but different directory names)
 
     Works only when there are no merge commits in the "source branch".
     Source branch is commits from base up to c2 not including base.
 
     If merge is involved it fallbacks to _fullcopytracing().
 
     Can be used by setting the following config:
 
         [experimental]
         copytrace = heuristics
 
     In some cases the copy/move candidates found by heuristics can be very large
     in number and that will make the algorithm slow. The number of possible
     candidates to check can be limited by using the config
     `experimental.copytrace.movecandidateslimit` which defaults to 100.
     """
 
     if c1.rev() is None:
         c1 = c1.p1()
     if c2.rev() is None:
         c2 = c2.p1()
 
     copies = {}
 
     changedfiles = set()
     m1 = c1.manifest()
     if not repo.revs('%d::%d', base.rev(), c2.rev()):
         # If base is not in c2 branch, we switch to fullcopytracing
         repo.ui.debug("switching to full copytracing as base is not "
                       "an ancestor of c2\n")
         return _fullcopytracing(repo, c1, c2, base)
 
     ctx = c2
     while ctx != base:
         if len(ctx.parents()) == 2:
             # To keep things simple let's not handle merges
             repo.ui.debug("switching to full copytracing because of merges\n")
             return _fullcopytracing(repo, c1, c2, base)
         changedfiles.update(ctx.files())
         ctx = ctx.p1()
 
     cp = _forwardcopies(base, c2)
     for dst, src in cp.iteritems():
         if src in m1:
             copies[dst] = src
 
     # file is missing if it isn't present in the destination, but is present in
     # the base and present in the source.
     # Presence in the base is important to exclude added files, presence in the
     # source is important to exclude removed files.
     filt = lambda f: f not in m1 and f in base and f in c2
     missingfiles = [f for f in changedfiles if filt(f)]
 
     if missingfiles:
         basenametofilename = collections.defaultdict(list)
         dirnametofilename = collections.defaultdict(list)
 
         for f in m1.filesnotin(base.manifest()):
             basename = os.path.basename(f)
             dirname = os.path.dirname(f)
             basenametofilename[basename].append(f)
             dirnametofilename[dirname].append(f)
 
         for f in missingfiles:
             basename = os.path.basename(f)
             dirname = os.path.dirname(f)
             samebasename = basenametofilename[basename]
             samedirname = dirnametofilename[dirname]
             movecandidates = samebasename + samedirname
             # f is guaranteed to be present in c2, that's why
             # c2.filectx(f) won't fail
             f2 = c2.filectx(f)
             # we can have a lot of candidates which can slow down the heuristics
             # config value to limit the number of candidates moves to check
             maxcandidates = repo.ui.configint('experimental',
                                               'copytrace.movecandidateslimit')
 
             if len(movecandidates) > maxcandidates:
                 repo.ui.status(_("skipping copytracing for '%s', more "
                                  "candidates than the limit: %d\n")
                                % (f, len(movecandidates)))
                 continue
 
             for candidate in movecandidates:
                 f1 = c1.filectx(candidate)
                 if _related(f1, f2):
                     # if there are a few related copies then we'll merge
                     # changes into all of them. This matches the behaviour
                     # of upstream copytracing
                     copies[candidate] = f
 
     return copies, {}, {}, {}, {}
 
 def _related(f1, f2):
     """return True if f1 and f2 filectx have a common ancestor
 
     Walk back to common ancestor to see if the two files originate
     from the same file. Since workingfilectx's rev() is None it messes
     up the integer comparison logic, hence the pre-step check for
     None (f1 and f2 can only be workingfilectx's initially).
     """
 
     if f1 == f2:
         return True # a match
 
     g1, g2 = f1.ancestors(), f2.ancestors()
     try:
         f1r, f2r = f1.linkrev(), f2.linkrev()
 
         if f1r is None:
             f1 = next(g1)
         if f2r is None:
             f2 = next(g2)
 
         while True:
             f1r, f2r = f1.linkrev(), f2.linkrev()
             if f1r > f2r:
                 f1 = next(g1)
             elif f2r > f1r:
                 f2 = next(g2)
             else: # f1 and f2 point to files in the same linkrev
                 return f1 == f2 # true if they point to the same file
     except StopIteration:
         return False
 
 def duplicatecopies(repo, wctx, rev, fromrev, skiprev=None):
     """reproduce copies from fromrev to rev in the dirstate
 
     If skiprev is specified, it's a revision that should be used to
     filter copy records. Any copies that occur between fromrev and
     skiprev will not be duplicated, even if they appear in the set of
     copies between fromrev and rev.
     """
     exclude = {}
     ctraceconfig = repo.ui.config('experimental', 'copytrace')
     bctrace = stringutil.parsebool(ctraceconfig)
     if (skiprev is not None and
         (ctraceconfig == 'heuristics' or bctrace or bctrace is None)):
         # copytrace='off' skips this line, but not the entire function because
         # the line below is O(size of the repo) during a rebase, while the rest
         # of the function is much faster (and is required for carrying copy
         # metadata across the rebase anyway).
         exclude = pathcopies(repo[fromrev], repo[skiprev])
     for dst, src in pathcopies(repo[fromrev], repo[rev]).iteritems():
         if dst in exclude:
             continue
+        if dst in wctx:
         wctx[dst].markcopied(src)

tests/test-rebase-inmemory.t

@@ -1,845 +1,843 b''
 #require symlink execbit
   $ cat << EOF >> $HGRCPATH
   > [phases]
   > publish=False
   > [extensions]
   > amend=
   > rebase=
   > debugdrawdag=$TESTDIR/drawdag.py
   > strip=
   > [rebase]
   > experimental.inmemory=1
   > [diff]
   > git=1
   > [alias]
   > tglog = log -G --template "{rev}: {node|short} '{desc}'\n"
   > EOF
 
 Rebase a simple DAG:
   $ hg init repo1
   $ cd repo1
   $ hg debugdrawdag <<'EOS'
   > c b
   > |/
   > d
   > |
   > a
   > EOS
   $ hg up -C a
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg tglog
   o  3: 814f6bd05178 'c'
   |
   | o  2: db0e82a16a62 'b'
   |/
   o  1: 02952614a83d 'd'
   |
   @  0: b173517d0057 'a'
   
   $ hg cat -r 3 c
   c (no-eol)
   $ hg cat -r 2 b
   b (no-eol)
   $ hg rebase --debug -r b -d c | grep rebasing
   rebasing in-memory
   rebasing 2:db0e82a16a62 "b" (b)
   $ hg tglog
   o  3: ca58782ad1e4 'b'
   |
   o  2: 814f6bd05178 'c'
   |
   o  1: 02952614a83d 'd'
   |
   @  0: b173517d0057 'a'
   
   $ hg cat -r 3 b
   b (no-eol)
   $ hg cat -r 2 c
   c (no-eol)
   $ cd ..
 
 Case 2:
   $ hg init repo2
   $ cd repo2
   $ hg debugdrawdag <<'EOS'
   > c b
   > |/
   > d
   > |
   > a
   > EOS
 
 Add a symlink and executable file:
   $ hg up -C c
   3 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ ln -s somefile e
   $ echo f > f
   $ chmod +x f
   $ hg add e f
   $ hg amend -q
   $ hg up -Cq a
 
 Write files to the working copy, and ensure they're still there after the rebase
   $ echo "abc" > a
   $ ln -s def b
   $ echo "ghi" > c
   $ echo "jkl" > d
   $ echo "mno" > e
   $ hg tglog
   o  3: f56b71190a8f 'c'
   |
   | o  2: db0e82a16a62 'b'
   |/
   o  1: 02952614a83d 'd'
   |
   @  0: b173517d0057 'a'
   
   $ hg cat -r 3 c
   c (no-eol)
   $ hg cat -r 2 b
   b (no-eol)
   $ hg cat -r 3 e
   somefile (no-eol)
   $ hg rebase --debug -s b -d a | grep rebasing
   rebasing in-memory
   rebasing 2:db0e82a16a62 "b" (b)
   $ hg tglog
   o  3: fc055c3b4d33 'b'
   |
   | o  2: f56b71190a8f 'c'
   | |
   | o  1: 02952614a83d 'd'
   |/
   @  0: b173517d0057 'a'
   
   $ hg cat -r 2 c
   c (no-eol)
   $ hg cat -r 3 b
   b (no-eol)
   $ hg rebase --debug -s 1 -d 3 | grep rebasing
   rebasing in-memory
   rebasing 1:02952614a83d "d" (d)
   rebasing 2:f56b71190a8f "c"
   $ hg tglog
   o  3: 753feb6fd12a 'c'
   |
   o  2: 09c044d2cb43 'd'
   |
   o  1: fc055c3b4d33 'b'
   |
   @  0: b173517d0057 'a'
   
 Ensure working copy files are still there:
   $ cat a
   abc
   $ readlink.py b
   b -> def
   $ cat e
   mno
 
 Ensure symlink and executable files were rebased properly:
   $ hg up -Cq 3
   $ readlink.py e
   e -> somefile
   $ ls -l f | cut -c -10
   -rwxr-xr-x
 
 Rebase the working copy parent
   $ hg up -C 3
   0 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg rebase -r 3 -d 0 --debug | grep rebasing
   rebasing in-memory
   rebasing 3:753feb6fd12a "c" (tip)
   $ hg tglog
   @  3: 844a7de3e617 'c'
   |
   | o  2: 09c044d2cb43 'd'
   | |
   | o  1: fc055c3b4d33 'b'
   |/
   o  0: b173517d0057 'a'
   
 
 Test reporting of path conflicts
 
   $ hg rm a
   $ mkdir a
   $ touch a/a
   $ hg ci -Am "a/a"
   adding a/a
   $ hg tglog
   @  4: daf7dfc139cb 'a/a'
   |
   o  3: 844a7de3e617 'c'
   |
   | o  2: 09c044d2cb43 'd'
   | |
   | o  1: fc055c3b4d33 'b'
   |/
   o  0: b173517d0057 'a'
   
   $ hg rebase -r . -d 2
   rebasing 4:daf7dfc139cb "a/a" (tip)
   saved backup bundle to $TESTTMP/repo2/.hg/strip-backup/daf7dfc139cb-fdbfcf4f-rebase.hg
 
   $ hg tglog
   @  4: c6ad37a4f250 'a/a'
   |
   | o  3: 844a7de3e617 'c'
   | |
   o |  2: 09c044d2cb43 'd'
   | |
   o |  1: fc055c3b4d33 'b'
   |/
   o  0: b173517d0057 'a'
   
   $ echo foo > foo
   $ hg ci -Aqm "added foo"
   $ hg up '.^'
   0 files updated, 0 files merged, 1 files removed, 0 files unresolved
   $ echo bar > bar
   $ hg ci -Aqm "added bar"
   $ hg rm a/a
   $ echo a > a
   $ hg ci -Aqm "added a back!"
   $ hg tglog
   @  7: 855e9797387e 'added a back!'
   |
   o  6: d14530e5e3e6 'added bar'
   |
   | o  5: 9b94b9373deb 'added foo'
   |/
   o  4: c6ad37a4f250 'a/a'
   |
   | o  3: 844a7de3e617 'c'
   | |
   o |  2: 09c044d2cb43 'd'
   | |
   o |  1: fc055c3b4d33 'b'
   |/
   o  0: b173517d0057 'a'
   
   $ hg rebase -r . -d 5
   rebasing 7:855e9797387e "added a back!" (tip)
   saved backup bundle to $TESTTMP/repo2/.hg/strip-backup/855e9797387e-81ee4c5d-rebase.hg
 
   $ hg tglog
   @  7: bb3f02be2688 'added a back!'
   |
   | o  6: d14530e5e3e6 'added bar'
   | |
   o |  5: 9b94b9373deb 'added foo'
   |/
   o  4: c6ad37a4f250 'a/a'
   |
   | o  3: 844a7de3e617 'c'
   | |
   o |  2: 09c044d2cb43 'd'
   | |
   o |  1: fc055c3b4d33 'b'
   |/
   o  0: b173517d0057 'a'
   
   $ mkdir -p c/subdir
   $ echo c > c/subdir/file.txt
   $ hg add c/subdir/file.txt
   $ hg ci -m 'c/subdir/file.txt'
   $ hg rebase -r . -d 3 -n
   starting dry-run rebase; repository will not be changed
   rebasing 8:e147e6e3c490 "c/subdir/file.txt" (tip)
   abort: error: 'c/subdir/file.txt' conflicts with file 'c' in 3.
   [255]
   $ hg rebase -r 3 -d . -n
   starting dry-run rebase; repository will not be changed
   rebasing 3:844a7de3e617 "c"
   abort: error: file 'c' cannot be written because  'c/' is a directory in e147e6e3c490 (containing 1 entries: c/subdir/file.txt)
   [255]
 
   $ cd ..
 
 Test path auditing (issue5818)
 
   $ mkdir lib_
   $ ln -s lib_ lib
   $ hg init repo
   $ cd repo
   $ mkdir -p ".$TESTTMP/lib"
   $ touch ".$TESTTMP/lib/a"
   $ hg add ".$TESTTMP/lib/a"
   $ hg ci -m 'a'
 
   $ touch ".$TESTTMP/lib/b"
   $ hg add ".$TESTTMP/lib/b"
   $ hg ci -m 'b'
 
   $ hg up -q '.^'
   $ touch ".$TESTTMP/lib/c"
   $ hg add ".$TESTTMP/lib/c"
   $ hg ci -m 'c'
   created new head
   $ hg rebase -s 1 -d .
   rebasing 1:* "b" (glob)
   saved backup bundle to $TESTTMP/repo/.hg/strip-backup/*-rebase.hg (glob)
   $ cd ..
 
 Test dry-run rebasing
 
   $ hg init repo3
   $ cd repo3
   $ echo a>a
   $ hg ci -Aqma
   $ echo b>b
   $ hg ci -Aqmb
   $ echo c>c
   $ hg ci -Aqmc
   $ echo d>d
   $ hg ci -Aqmd
   $ echo e>e
   $ hg ci -Aqme
 
   $ hg up 1 -q
   $ echo f>f
   $ hg ci -Amf
   adding f
   created new head
   $ echo g>g
   $ hg ci -Aqmg
   $ hg log -G --template "{rev}:{short(node)} {person(author)}\n{firstline(desc)} {topic}\n\n"
   @  6:baf10c5166d4 test
   |  g
   |
   o  5:6343ca3eff20 test
   |  f
   |
   | o  4:e860deea161a test
   | |  e
   | |
   | o  3:055a42cdd887 test
   | |  d
   | |
   | o  2:177f92b77385 test
   |/   c
   |
   o  1:d2ae7f538514 test
   |  b
   |
   o  0:cb9a9f314b8b test
      a
   
 Make sure it throws error while passing --continue or --abort with --dry-run
   $ hg rebase -s 2 -d 6 -n --continue
   abort: cannot specify both --dry-run and --continue
   [255]
   $ hg rebase -s 2 -d 6 -n --abort
   abort: cannot specify both --dry-run and --abort
   [255]
 
 Check dryrun gives correct results when there is no conflict in rebasing
   $ hg rebase -s 2 -d 6 -n
   starting dry-run rebase; repository will not be changed
   rebasing 2:177f92b77385 "c"
   rebasing 3:055a42cdd887 "d"
   rebasing 4:e860deea161a "e"
   dry-run rebase completed successfully; run without -n/--dry-run to perform this rebase
 
   $ hg diff
   $ hg status
 
   $ hg log -G --template "{rev}:{short(node)} {person(author)}\n{firstline(desc)} {topic}\n\n"
   @  6:baf10c5166d4 test
   |  g
   |
   o  5:6343ca3eff20 test
   |  f
   |
   | o  4:e860deea161a test
   | |  e
   | |
   | o  3:055a42cdd887 test
   | |  d
   | |
   | o  2:177f92b77385 test
   |/   c
   |
   o  1:d2ae7f538514 test
   |  b
   |
   o  0:cb9a9f314b8b test
      a
   
 Check dryrun working with --collapse when there is no conflict
   $ hg rebase -s 2 -d 6 -n --collapse
   starting dry-run rebase; repository will not be changed
   rebasing 2:177f92b77385 "c"
   rebasing 3:055a42cdd887 "d"
   rebasing 4:e860deea161a "e"
   dry-run rebase completed successfully; run without -n/--dry-run to perform this rebase
 
 Check dryrun gives correct results when there is conflict in rebasing
 Make a conflict:
   $ hg up 6 -q
   $ echo conflict>e
   $ hg ci -Aqm "conflict with e"
   $ hg log -G --template "{rev}:{short(node)} {person(author)}\n{firstline(desc)} {topic}\n\n"
   @  7:d2c195b28050 test
   |  conflict with e
   |
   o  6:baf10c5166d4 test
   |  g
   |
   o  5:6343ca3eff20 test
   |  f
   |
   | o  4:e860deea161a test
   | |  e
   | |
   | o  3:055a42cdd887 test
   | |  d
   | |
   | o  2:177f92b77385 test
   |/   c
   |
   o  1:d2ae7f538514 test
   |  b
   |
   o  0:cb9a9f314b8b test
      a
   
   $ hg rebase -s 2 -d 7 -n
   starting dry-run rebase; repository will not be changed
   rebasing 2:177f92b77385 "c"
   rebasing 3:055a42cdd887 "d"
   rebasing 4:e860deea161a "e"
   merging e
   transaction abort!
   rollback completed
   hit a merge conflict
   [1]
   $ hg diff
   $ hg status
   $ hg log -G --template "{rev}:{short(node)} {person(author)}\n{firstline(desc)} {topic}\n\n"
   @  7:d2c195b28050 test
   |  conflict with e
   |
   o  6:baf10c5166d4 test
   |  g
   |
   o  5:6343ca3eff20 test
   |  f
   |
   | o  4:e860deea161a test
   | |  e
   | |
   | o  3:055a42cdd887 test
   | |  d
   | |
   | o  2:177f92b77385 test
   |/   c
   |
   o  1:d2ae7f538514 test
   |  b
   |
   o  0:cb9a9f314b8b test
      a
   
 Check dryrun working with --collapse when there is conflicts
   $ hg rebase -s 2 -d 7 -n --collapse
   starting dry-run rebase; repository will not be changed
   rebasing 2:177f92b77385 "c"
   rebasing 3:055a42cdd887 "d"
   rebasing 4:e860deea161a "e"
   merging e
   hit a merge conflict
   [1]
 
 In-memory rebase that fails due to merge conflicts
 
   $ hg rebase -s 2 -d 7
   rebasing 2:177f92b77385 "c"
   rebasing 3:055a42cdd887 "d"
   rebasing 4:e860deea161a "e"
   merging e
   transaction abort!
   rollback completed
   hit merge conflicts; re-running rebase without in-memory merge
   rebasing 2:177f92b77385 "c"
   rebasing 3:055a42cdd887 "d"
   rebasing 4:e860deea161a "e"
   merging e
   warning: conflicts while merging e! (edit, then use 'hg resolve --mark')
   unresolved conflicts (see hg resolve, then hg rebase --continue)
   [1]
   $ hg rebase --abort
   saved backup bundle to $TESTTMP/repo3/.hg/strip-backup/c1e524d4287c-f91f82e1-backup.hg
   rebase aborted
 
 Retrying without in-memory merge won't lose working copy changes
   $ cd ..
   $ hg clone repo3 repo3-dirty -q
   $ cd repo3-dirty
   $ echo dirty > a
   $ hg rebase -s 2 -d 7
   rebasing 2:177f92b77385 "c"
   rebasing 3:055a42cdd887 "d"
   rebasing 4:e860deea161a "e"
   merging e
   transaction abort!
   rollback completed
   hit merge conflicts; re-running rebase without in-memory merge
   abort: uncommitted changes
   [255]
   $ cat a
   dirty
 
 Retrying without in-memory merge won't lose merge state
   $ cd ..
   $ hg clone repo3 repo3-merge-state -q
   $ cd repo3-merge-state
   $ hg merge 4
   merging e
   warning: conflicts while merging e! (edit, then use 'hg resolve --mark')
   2 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 -l
   U e
   $ hg rebase -s 2 -d 7
   rebasing 2:177f92b77385 "c"
   abort: outstanding merge conflicts
   [255]
   $ hg resolve -l
   U e
 
 ==========================
 Test for --confirm option|
 ==========================
   $ cd ..
   $ hg clone repo3 repo4 -q
   $ cd repo4
   $ hg strip 7 -q
   $ hg log -G --template "{rev}:{short(node)} {person(author)}\n{firstline(desc)} {topic}\n\n"
   @  6:baf10c5166d4 test
   |  g
   |
   o  5:6343ca3eff20 test
   |  f
   |
   | o  4:e860deea161a test
   | |  e
   | |
   | o  3:055a42cdd887 test
   | |  d
   | |
   | o  2:177f92b77385 test
   |/   c
   |
   o  1:d2ae7f538514 test
   |  b
   |
   o  0:cb9a9f314b8b test
      a
   
 Check it gives error when both --dryrun and --confirm is used:
   $ hg rebase -s 2 -d . --confirm --dry-run
   abort: cannot specify both --confirm and --dry-run
   [255]
   $ hg rebase -s 2 -d . --confirm --abort
   abort: cannot specify both --confirm and --abort
   [255]
   $ hg rebase -s 2 -d . --confirm --continue
   abort: cannot specify both --confirm and --continue
   [255]
 
 Test --confirm option when there are no conflicts:
   $ hg rebase -s 2 -d . --keep --config ui.interactive=True --confirm << EOF
   > n
   > EOF
   starting in-memory rebase
   rebasing 2:177f92b77385 "c"
   rebasing 3:055a42cdd887 "d"
   rebasing 4:e860deea161a "e"
   rebase completed successfully
   apply changes (yn)? n
   $ hg log -G --template "{rev}:{short(node)} {person(author)}\n{firstline(desc)} {topic}\n\n"
   @  6:baf10c5166d4 test
   |  g
   |
   o  5:6343ca3eff20 test
   |  f
   |
   | o  4:e860deea161a test
   | |  e
   | |
   | o  3:055a42cdd887 test
   | |  d
   | |
   | o  2:177f92b77385 test
   |/   c
   |
   o  1:d2ae7f538514 test
   |  b
   |
   o  0:cb9a9f314b8b test
      a
   
   $ hg rebase -s 2 -d . --keep --config ui.interactive=True --confirm << EOF
   > y
   > EOF
   starting in-memory rebase
   rebasing 2:177f92b77385 "c"
   rebasing 3:055a42cdd887 "d"
   rebasing 4:e860deea161a "e"
   rebase completed successfully
   apply changes (yn)? y
   $ hg log -G --template "{rev}:{short(node)} {person(author)}\n{firstline(desc)} {topic}\n\n"
   o  9:9fd28f55f6dc test
   |  e
   |
   o  8:12cbf031f469 test
   |  d
   |
   o  7:c83b1da5b1ae test
   |  c
   |
   @  6:baf10c5166d4 test
   |  g
   |
   o  5:6343ca3eff20 test
   |  f
   |
   | o  4:e860deea161a test
   | |  e
   | |
   | o  3:055a42cdd887 test
   | |  d
   | |
   | o  2:177f92b77385 test
   |/   c
   |
   o  1:d2ae7f538514 test
   |  b
   |
   o  0:cb9a9f314b8b test
      a
   
 Test --confirm option when there is a conflict
   $ hg up tip -q
   $ echo ee>e
   $ hg ci --amend -m "conflict with e" -q
   $ hg log -G --template "{rev}:{short(node)} {person(author)}\n{firstline(desc)} {topic}\n\n"
   @  9:906d72f66a59 test
   |  conflict with e
   |
   o  8:12cbf031f469 test
   |  d
   |
   o  7:c83b1da5b1ae test
   |  c
   |
   o  6:baf10c5166d4 test
   |  g
   |
   o  5:6343ca3eff20 test
   |  f
   |
   | o  4:e860deea161a test
   | |  e
   | |
   | o  3:055a42cdd887 test
   | |  d
   | |
   | o  2:177f92b77385 test
   |/   c
   |
   o  1:d2ae7f538514 test
   |  b
   |
   o  0:cb9a9f314b8b test
      a
   
   $ hg rebase -s 4 -d . --keep --confirm
   starting in-memory rebase
   rebasing 4:e860deea161a "e"
   merging e
   hit a merge conflict
   [1]
   $ hg log -G --template "{rev}:{short(node)} {person(author)}\n{firstline(desc)} {topic}\n\n"
   @  9:906d72f66a59 test
   |  conflict with e
   |
   o  8:12cbf031f469 test
   |  d
   |
   o  7:c83b1da5b1ae test
   |  c
   |
   o  6:baf10c5166d4 test
   |  g
   |
   o  5:6343ca3eff20 test
   |  f
   |
   | o  4:e860deea161a test
   | |  e
   | |
   | o  3:055a42cdd887 test
   | |  d
   | |
   | o  2:177f92b77385 test
   |/   c
   |
   o  1:d2ae7f538514 test
   |  b
   |
   o  0:cb9a9f314b8b test
      a
   
 Test a metadata-only in-memory merge
   $ cd $TESTTMP
   $ hg init no_exception
   $ cd no_exception
 # Produce the following graph:
 #   o  'add +x to foo.txt'
 #   | o  r1  (adds bar.txt, just for something to rebase to)
 #   |/
 #   o  r0   (adds foo.txt, no +x)
   $ echo hi > foo.txt
   $ hg ci -qAm r0
   $ echo hi > bar.txt
   $ hg ci -qAm r1
   $ hg co -qr ".^"
   $ chmod +x foo.txt
   $ hg ci -qAm 'add +x to foo.txt'
 issue5960: this was raising an AttributeError exception
   $ hg rebase -r . -d 1
   rebasing 2:539b93e77479 "add +x to foo.txt" (tip)
   saved backup bundle to $TESTTMP/no_exception/.hg/strip-backup/*.hg (glob)
   $ hg diff -c tip
   diff --git a/foo.txt b/foo.txt
   old mode 100644
   new mode 100755
 
 Test rebasing a commit with copy information, but no content changes
 
   $ cd ..
   $ hg clone -q repo1 merge-and-rename
   $ cd merge-and-rename
   $ cat << EOF >> .hg/hgrc
   > [experimental]
   > evolution.createmarkers=True
   > evolution.allowunstable=True
   > EOF
   $ hg co -q 1
   $ hg mv d e
   $ hg ci -qm 'rename d to e'
   $ hg co -q 3
   $ hg merge -q 4
   $ hg ci -m 'merge'
   $ hg co -q 2
   $ mv d e
   $ hg addremove -qs 0
   $ hg ci -qm 'untracked rename of d to e'
   $ hg debugobsolete -q `hg log -T '{node}' -r 4` `hg log -T '{node}' -r .`
   1 new orphan changesets
   $ hg tglog
   @  6: 676538af172d 'untracked rename of d to e'
   |
   | *    5: 71cb43376053 'merge'
   | |\
   | | x  4: 2c8b5dad7956 'rename d to e'
   | | |
   | o |  3: ca58782ad1e4 'b'
   |/ /
   o /  2: 814f6bd05178 'c'
   |/
   o  1: 02952614a83d 'd'
   |
   o  0: b173517d0057 'a'
   
   $ hg rebase -b 5 -d tip
   rebasing 3:ca58782ad1e4 "b"
   rebasing 5:71cb43376053 "merge"
   note: not rebasing 5:71cb43376053 "merge", its destination already has all its changes
 
   $ 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'
   $ hg rebase -d 1
   rebasing 2:b977edf6f839 "rename a to b" (tip)
   merging a and b to b
   saved backup bundle to $TESTTMP/rebase-rename/.hg/strip-backup/b977edf6f839-0864f570-rebase.hg
   $ hg st --copies --change .
   A b
     a
   R a
   $ cd ..
 
 Rebase across a copy with --collapse
 
   $ hg init rebase-rename-collapse
   $ cd rebase-rename-collapse
   $ echo a > a
   $ hg ci -Aqm 'add a'
   $ hg mv a b
   $ hg ci -m 'rename a to b'
   $ hg co -q 0
   $ echo a2 > a
   $ hg ci -qm 'modify a'
-BROKEN: obviously...
   $ hg rebase -r . -d 1 --collapse
   rebasing 2:41c4ea50d4cf "modify a" (tip)
   merging b and a to b
-  abort: a@b977edf6f839: not found in manifest!
-  [255]
+  saved backup bundle to $TESTTMP/rebase-rename-collapse/.hg/strip-backup/41c4ea50d4cf-b90b7994-rebase.hg
   $ cd ..
 
 Test rebasing when the file we are merging in destination is empty
 
   $ hg init test
   $ cd test
   $ echo a > foo
   $ hg ci -Aqm 'added a to foo'
 
   $ rm foo
   $ touch foo
   $ hg di
   diff --git a/foo b/foo
   --- a/foo
   +++ b/foo
   @@ -1,1 +0,0 @@
   -a
 
   $ hg ci -m "make foo an empty file"
 
   $ hg up '.^'
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ echo b > foo
   $ hg di
   diff --git a/foo b/foo
   --- a/foo
   +++ b/foo
   @@ -1,1 +1,1 @@
   -a
   +b
   $ hg ci -m "add b to foo"
   created new head
 
   $ hg rebase -r . -d 1 --config ui.merge=internal:merge3
   rebasing 2:fb62b706688e "add b to foo" (tip)
   merging foo
   hit merge conflicts; re-running rebase without in-memory merge
   rebasing 2:fb62b706688e "add b to foo" (tip)
   merging foo
   warning: conflicts while merging foo! (edit, then use 'hg resolve --mark')
   unresolved conflicts (see hg resolve, then hg rebase --continue)
   [1]