py3: explicitly convert dict.keys() and dict.items() into a list

Differential Revision: https://phab.mercurial-scm.org/D853

mercurial/copies.py

@@ -1,867 +1,867 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 . import (
     match as matchmod,
     node,
     pathutil,
     phases,
     scmutil,
     util,
 )
 
 def _findlimit(repo, a, b):
     """
     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.
     None if no such revision exists.
     """
 
     # 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
     working = len(cl) # pseudo rev for the working directory
     if a is None:
         a = working
     if b is None:
         b = working
 
     side = {a: -1, b: 1}
     visit = [-a, -b]
     heapq.heapify(visit)
     interesting = len(visit)
     hascommonancestor = False
     limit = working
 
     while interesting:
         r = -heapq.heappop(visit)
         if r == working:
             parents = [cl.rev(p) for p in repo.dirstate.parents()]
         else:
             parents = cl.parentrevs(r)
         for p in parents:
             if p < 0:
                 continue
             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
                 hascommonancestor = True
         if side[r]:
             limit = r # lowest rev visited
             interesting -= 1
 
     if not hascommonancestor:
         return None
 
     # 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->b'''
     t = a.copy()
     for k, v in b.iteritems():
         if v in t:
             # found a chain
             if t[v] != k:
                 # file wasn't renamed back to itself
                 t[k] = t[v]
             if v not in dst:
                 # chain was a rename, not a copy
                 del t[v]
         if v in src:
             # file is a copy of an existing file
             t[k] = v
 
     # remove criss-crossed copies
     for k, v in t.items():
         if k in src and v in dst:
             del t[k]
 
     return t
 
 def _tracefile(fctx, am, limit=-1):
     '''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 limit >= 0 and f.linkrev() < limit and f.rev() < limit:
             return None
 
 def _dirstatecopies(d):
     ds = d._repo.dirstate
     c = ds.copies().copy()
-    for k in c.keys():
+    for k in list(c):
         if ds[k] not in 'anm':
             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 _forwardcopies(a, b, match=None):
     '''find {dst@b: src@a} copy mapping where a is an ancestor of b'''
 
     # check for working copy
     w = None
     if b.rev() is None:
         w = b
         b = w.p1()
         if a == b:
             # short-circuit to avoid issues with merge states
             return _dirstatecopies(w)
 
     # files might have to be traced back to the fctx parent of the last
     # one-side-only changeset, but not further back than that
     limit = _findlimit(a._repo, a.rev(), b.rev())
     if limit is None:
         limit = -1
     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 = scmutil.matchfiles(a._repo, b.files())
         forwardmissingmatch = matchmod.intersectmatchers(match, filesmatcher)
     missing = _computeforwardmissing(a, b, match=forwardmissingmatch)
 
     ancestrycontext = a._repo.changelog.ancestors([b.rev()], inclusive=True)
     for f in missing:
         fctx = b[f]
         fctx._ancestrycontext = ancestrycontext
         ofctx = _tracefile(fctx, am, limit)
         if ofctx:
             cm[f] = ofctx.path()
 
     # combine copies from dirstate if necessary
     if w is not None:
         cm = _chain(a, w, cm, _dirstatecopies(w))
 
     return cm
 
 def _backwardrenames(a, b):
     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.
     f = _forwardcopies(b, a)
     r = {}
     for k, v in sorted(f.iteritems()):
         # 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'''
     if x == y or not x or not y:
         return {}
     a = y.ancestor(x)
     if a == x:
         return _forwardcopies(x, y, match=match)
     if a == y:
         return _backwardrenames(x, y)
     return _chain(x, y, _backwardrenames(x, a),
                   _forwardcopies(a, y, match=match))
 
 def _computenonoverlap(repo, c1, c2, addedinm1, addedinm2, baselabel=''):
     """Computes, based on addedinm1 and addedinm2, the files exclusive to c1
     and c2. This is its own function so extensions can easily wrap this call
     to see what files mergecopies is about to process.
 
     Even though c1 and c2 are not used in this function, they are useful in
     other extensions for being able to read the file nodes of the changed files.
 
     "baselabel" can be passed to help distinguish the multiple computations
     done in the graft case.
     """
     u1 = sorted(addedinm1 - addedinm2)
     u2 = sorted(addedinm2 - addedinm1)
 
     header = "  unmatched files in %s"
     if baselabel:
         header += ' (from %s)' % baselabel
     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)))
     return u1, u2
 
 def _makegetfctx(ctx):
     """return a 'getfctx' function suitable for _checkcopies usage
 
     We have to re-setup the function building 'filectx' for each
     '_checkcopies' to ensure the linkrev adjustment is properly setup for
     each. Linkrev adjustment is important to avoid bug in rename
     detection. Moreover, having a proper '_ancestrycontext' setup ensures
     the performance impact of this adjustment is kept limited. Without it,
     each file could do a full dag traversal making the time complexity of
     the operation explode (see issue4537).
 
     This function exists here mostly to limit the impact on stable. Feel
     free to refactor on default.
     """
     rev = ctx.rev()
     repo = ctx._repo
     ac = getattr(ctx, '_ancestrycontext', None)
     if ac is None:
         revs = [rev]
         if rev is None:
             revs = [p.rev() for p in ctx.parents()]
         ac = repo.changelog.ancestors(revs, inclusive=True)
         ctx._ancestrycontext = ac
     def makectx(f, n):
         if n in node.wdirnodes:  # in a working context?
             if ctx.rev() is None:
                 return ctx.filectx(f)
             return repo[None][f]
         fctx = repo.filectx(f, fileid=n)
         # setup only needed for filectx not create from a changectx
         fctx._ancestrycontext = ac
         fctx._descendantrev = rev
         return fctx
     return util.lrucachefunc(makectx)
 
 def _combinecopies(copyfrom, copyto, finalcopy, diverge, incompletediverge):
     """combine partial copy paths"""
     remainder = {}
     for f in copyfrom:
         if f in copyto:
             finalcopy[copyto[f]] = copyfrom[f]
             del copyto[f]
     for f in incompletediverge:
         assert f not in diverge
         ic = incompletediverge[f]
         if ic[0] in copyto:
             diverge[f] = [copyto[ic[0]], ic[1]]
         else:
             remainder[f] = ic
     return remainder
 
 def mergecopies(repo, c1, c2, base):
     """
     The function calling different copytracing algorithms on the basis of config
     which find 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.
     """
     # avoid silly behavior for update from empty dir
     if not c1 or not c2 or c1 == c2:
         return {}, {}, {}, {}, {}
 
     # avoid silly behavior for parent -> working dir
     if c2.node() is None and c1.node() == repo.dirstate.p1():
         return repo.dirstate.copies(), {}, {}, {}, {}
 
     copytracing = repo.ui.config('experimental', 'copytrace')
 
     # 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 == 'off':
         return {}, {}, {}, {}, {}
     elif copytracing == 'heuristics':
         # Do full copytracing if only drafts are involved as that will be fast
         # enough and will also cover the copies which can 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 draft branches, if
     yes let's use the full copytrace algorithm for increased capabilities since
     it will be fast enough.
     """
     if c1.rev() is None:
         c1 = c1.p1()
 
     nonpublicphases = set([phases.draft, phases.secret])
 
     if (c1.phase() in nonpublicphases) and (base.phase() in nonpublicphases):
         sourcecommitlimit = repo.ui.configint('experimental',
                                               'copytrace.sourcecommitlimit')
         commits = len(repo.revs('%d::%d', base.rev(), c1.rev()))
         return commits < sourcecommitlimit
     return False
 
 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.
     """
     # In certain scenarios (e.g. graft, update or rebase), base can be
     # overridden We still need to know a real common ancestor in this case We
     # can't just compute _c1.ancestor(_c2) and compare it to ca, because there
     # can be multiple common ancestors, e.g. in case of bidmerge.  Because our
     # caller may not know if the revision passed in lieu of the CA is a genuine
     # common ancestor or not without explicitly checking it, it's better to
     # determine that here.
     #
     # base.descendant(wc) and base.descendant(base) are False, work around that
     _c1 = c1.p1() if c1.rev() is None else c1
     _c2 = c2.p1() if c2.rev() is None else c2
     # an endpoint is "dirty" if it isn't a descendant of the merge base
     # if we have a dirty endpoint, we need to trigger graft logic, and also
     # keep track of which endpoint is dirty
     dirtyc1 = not (base == _c1 or base.descendant(_c1))
     dirtyc2 = not (base == _c2 or base.descendant(_c2))
     graft = dirtyc1 or dirtyc2
     tca = base
     if graft:
         tca = _c1.ancestor(_c2)
 
     limit = _findlimit(repo, c1.rev(), c2.rev())
     if limit is None:
         # no common ancestor, no copies
         return {}, {}, {}, {}, {}
     repo.ui.debug("  searching for copies back to rev %d\n" % limit)
 
     m1 = c1.manifest()
     m2 = c2.manifest()
     mb = base.manifest()
 
     # gather data from _checkcopies:
     # - diverge = record all diverges in this dict
     # - copy = record all non-divergent copies in this dict
     # - fullcopy = record all copies in this dict
     # - incomplete = record non-divergent partial copies here
     # - incompletediverge = record divergent partial copies here
     diverge = {} # divergence data is shared
     incompletediverge  = {}
     data1 = {'copy': {},
              'fullcopy': {},
              'incomplete': {},
              'diverge': diverge,
              'incompletediverge': incompletediverge,
             }
     data2 = {'copy': {},
              'fullcopy': {},
              'incomplete': {},
              'diverge': diverge,
              'incompletediverge': incompletediverge,
             }
 
     # find interesting file sets from manifests
     addedinm1 = m1.filesnotin(mb)
     addedinm2 = m2.filesnotin(mb)
     bothnew = sorted(addedinm1 & addedinm2)
     if tca == base:
         # unmatched file from base
         u1r, u2r = _computenonoverlap(repo, c1, c2, addedinm1, addedinm2)
         u1u, u2u = u1r, u2r
     else:
         # unmatched file from base (DAG rotation in the graft case)
         u1r, u2r = _computenonoverlap(repo, c1, c2, addedinm1, addedinm2,
                                       baselabel='base')
         # unmatched file from topological common ancestors (no DAG rotation)
         # need to recompute this for directory move handling when grafting
         mta = tca.manifest()
         u1u, u2u = _computenonoverlap(repo, c1, c2, m1.filesnotin(mta),
                                                     m2.filesnotin(mta),
                                       baselabel='topological common ancestor')
 
     for f in u1u:
         _checkcopies(c1, c2, f, base, tca, dirtyc1, limit, data1)
 
     for f in u2u:
         _checkcopies(c2, c1, f, base, tca, dirtyc2, limit, data2)
 
     copy = dict(data1['copy'])
     copy.update(data2['copy'])
     fullcopy = dict(data1['fullcopy'])
     fullcopy.update(data2['fullcopy'])
 
     if dirtyc1:
         _combinecopies(data2['incomplete'], data1['incomplete'], copy, diverge,
                        incompletediverge)
     else:
         _combinecopies(data1['incomplete'], data2['incomplete'], copy, diverge,
                        incompletediverge)
 
     renamedelete = {}
     renamedeleteset = set()
     divergeset = set()
-    for of, fl in diverge.items():
+    for of, fl in list(diverge.items()):
         if len(fl) == 1 or of in c1 or of in c2:
             del diverge[of] # not actually divergent, or not a rename
             if of not in c1 and of not in c2:
                 # renamed on one side, deleted on the other side, but filter
                 # out files that have been renamed and then deleted
                 renamedelete[of] = [f for f in fl if f in c1 or f in c2]
                 renamedeleteset.update(fl) # reverse map for below
         else:
             divergeset.update(fl) # reverse map for below
 
     if bothnew:
         repo.ui.debug("  unmatched files new in both:\n   %s\n"
                       % "\n   ".join(bothnew))
     bothdiverge = {}
     bothincompletediverge = {}
     remainder = {}
     both1 = {'copy': {},
              'fullcopy': {},
              'incomplete': {},
              'diverge': bothdiverge,
              'incompletediverge': bothincompletediverge
             }
     both2 = {'copy': {},
              'fullcopy': {},
              'incomplete': {},
              'diverge': bothdiverge,
              'incompletediverge': bothincompletediverge
             }
     for f in bothnew:
         _checkcopies(c1, c2, f, base, tca, dirtyc1, limit, both1)
         _checkcopies(c2, c1, f, base, tca, dirtyc2, limit, both2)
     if dirtyc1:
         # incomplete copies may only be found on the "dirty" side for bothnew
         assert not both2['incomplete']
         remainder = _combinecopies({}, both1['incomplete'], copy, bothdiverge,
                                    bothincompletediverge)
     elif dirtyc2:
         assert not both1['incomplete']
         remainder = _combinecopies({}, both2['incomplete'], copy, bothdiverge,
                                    bothincompletediverge)
     else:
         # incomplete copies and divergences can't happen outside grafts
         assert not both1['incomplete']
         assert not both2['incomplete']
         assert not bothincompletediverge
     for f in remainder:
         assert f not in bothdiverge
         ic = remainder[f]
         if ic[0] in (m1 if dirtyc1 else m2):
             # backed-out rename on one side, but watch out for deleted files
             bothdiverge[f] = ic
     for of, fl in bothdiverge.items():
         if len(fl) == 2 and fl[0] == fl[1]:
             copy[fl[0]] = of # not actually divergent, just matching renames
 
     if fullcopy and 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
 
     if not fullcopy:
         return copy, {}, diverge, renamedelete, {}
 
     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, {}
 
     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 u1r + u2r:
         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
     """
 
     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.
     missingfiles = filter(lambda f: f not in m1 and f in base and f in c2,
                           changedfiles)
 
     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)
 
         # in case of a rebase/graft, base may not be a common ancestor
         anc = c1.ancestor(c2)
 
         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)
             for candidate in movecandidates:
                 f1 = c1.filectx(candidate)
                 if _related(f1, f2, anc.rev()):
                     # 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, limit):
     """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 f1 # 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)
             elif f1 == f2:
                 return f1 # a match
             elif f1r == f2r or f1r < limit or f2r < limit:
                 return False # copy no longer relevant
     except StopIteration:
         return False
 
 def _checkcopies(srcctx, dstctx, f, base, tca, remotebase, limit, data):
     """
     check possible copies of f from msrc to mdst
 
     srcctx = starting context for f in msrc
     dstctx = destination context for f in mdst
     f = the filename to check (as in msrc)
     base = the changectx used as a merge base
     tca = topological common ancestor for graft-like scenarios
     remotebase = True if base is outside tca::srcctx, False otherwise
     limit = the rev number to not search beyond
     data = dictionary of dictionary to store copy data. (see mergecopies)
 
     note: limit is only an optimization, and provides no guarantee that
     irrelevant revisions will not be visited
     there is no easy way to make this algorithm stop in a guaranteed way
     once it "goes behind a certain revision".
     """
 
     msrc = srcctx.manifest()
     mdst = dstctx.manifest()
     mb = base.manifest()
     mta = tca.manifest()
     # Might be true if this call is about finding backward renames,
     # This happens in the case of grafts because the DAG is then rotated.
     # If the file exists in both the base and the source, we are not looking
     # for a rename on the source side, but on the part of the DAG that is
     # traversed backwards.
     #
     # In the case there is both backward and forward renames (before and after
     # the base) this is more complicated as we must detect a divergence.
     # We use 'backwards = False' in that case.
     backwards = not remotebase and base != tca and f in mb
     getsrcfctx = _makegetfctx(srcctx)
     getdstfctx = _makegetfctx(dstctx)
 
     if msrc[f] == mb.get(f) and not remotebase:
         # Nothing to merge
         return
 
     of = None
     seen = {f}
     for oc in getsrcfctx(f, msrc[f]).ancestors():
         ocr = oc.linkrev()
         of = oc.path()
         if of in seen:
             # check limit late - grab last rename before
             if ocr < limit:
                 break
             continue
         seen.add(of)
 
         # remember for dir rename detection
         if backwards:
             data['fullcopy'][of] = f # grafting backwards through renames
         else:
             data['fullcopy'][f] = of
         if of not in mdst:
             continue # no match, keep looking
         if mdst[of] == mb.get(of):
             return # no merge needed, quit early
         c2 = getdstfctx(of, mdst[of])
         # c2 might be a plain new file on added on destination side that is
         # unrelated to the droids we are looking for.
         cr = _related(oc, c2, tca.rev())
         if cr and (of == f or of == c2.path()): # non-divergent
             if backwards:
                 data['copy'][of] = f
             elif of in mb:
                 data['copy'][f] = of
             elif remotebase: # special case: a <- b <- a -> b "ping-pong" rename
                 data['copy'][of] = f
                 del data['fullcopy'][f]
                 data['fullcopy'][of] = f
             else: # divergence w.r.t. graft CA on one side of topological CA
                 for sf in seen:
                     if sf in mb:
                         assert sf not in data['diverge']
                         data['diverge'][sf] = [f, of]
                         break
             return
 
     if of in mta:
         if backwards or remotebase:
             data['incomplete'][of] = f
         else:
             for sf in seen:
                 if sf in mb:
                     if tca == base:
                         data['diverge'].setdefault(sf, []).append(f)
                     else:
                         data['incompletediverge'][sf] = [of, f]
                     return
 
 def duplicatecopies(repo, 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 = {}
     if (skiprev is not None and
         repo.ui.config('experimental', 'copytrace') != 'off'):
         # 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():
         # copies.pathcopies returns backward renames, so dst might not
         # actually be in the dirstate
         if dst in exclude:
             continue
         if repo.dirstate[dst] in "nma":
             repo.dirstate.copy(src, dst)

mercurial/merge.py

@@ -1,1774 +1,1774 b''
 # merge.py - directory-level update/merge handling 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 hashlib
 import shutil
 import struct
 
 from .i18n import _
 from .node import (
     addednodeid,
     bin,
     hex,
     modifiednodeid,
     nullhex,
     nullid,
     nullrev,
 )
 from . import (
     copies,
     error,
     filemerge,
     match as matchmod,
     obsutil,
     pycompat,
     scmutil,
     subrepo,
     util,
     worker,
 )
 
 _pack = struct.pack
 _unpack = struct.unpack
 
 def _droponode(data):
     # used for compatibility for v1
     bits = data.split('\0')
     bits = bits[:-2] + bits[-1:]
     return '\0'.join(bits)
 
 class mergestate(object):
     '''track 3-way merge state of individual files
 
     The merge state is stored on disk when needed. Two files are used: one with
     an old format (version 1), and one with a new format (version 2). Version 2
     stores a superset of the data in version 1, including new kinds of records
     in the future. For more about the new format, see the documentation for
     `_readrecordsv2`.
 
     Each record can contain arbitrary content, and has an associated type. This
     `type` should be a letter. If `type` is uppercase, the record is mandatory:
     versions of Mercurial that don't support it should abort. If `type` is
     lowercase, the record can be safely ignored.
 
     Currently known records:
 
     L: the node of the "local" part of the merge (hexified version)
     O: the node of the "other" part of the merge (hexified version)
     F: a file to be merged entry
     C: a change/delete or delete/change conflict
     D: a file that the external merge driver will merge internally
        (experimental)
     m: the external merge driver defined for this merge plus its run state
        (experimental)
     f: a (filename, dictionary) tuple of optional values for a given file
     X: unsupported mandatory record type (used in tests)
     x: unsupported advisory record type (used in tests)
     l: the labels for the parts of the merge.
 
     Merge driver run states (experimental):
     u: driver-resolved files unmarked -- needs to be run next time we're about
        to resolve or commit
     m: driver-resolved files marked -- only needs to be run before commit
     s: success/skipped -- does not need to be run any more
 
     '''
     statepathv1 = 'merge/state'
     statepathv2 = 'merge/state2'
 
     @staticmethod
     def clean(repo, node=None, other=None, labels=None):
         """Initialize a brand new merge state, removing any existing state on
         disk."""
         ms = mergestate(repo)
         ms.reset(node, other, labels)
         return ms
 
     @staticmethod
     def read(repo):
         """Initialize the merge state, reading it from disk."""
         ms = mergestate(repo)
         ms._read()
         return ms
 
     def __init__(self, repo):
         """Initialize the merge state.
 
         Do not use this directly! Instead call read() or clean()."""
         self._repo = repo
         self._dirty = False
         self._labels = None
 
     def reset(self, node=None, other=None, labels=None):
         self._state = {}
         self._stateextras = {}
         self._local = None
         self._other = None
         self._labels = labels
         for var in ('localctx', 'otherctx'):
             if var in vars(self):
                 delattr(self, var)
         if node:
             self._local = node
             self._other = other
         self._readmergedriver = None
         if self.mergedriver:
             self._mdstate = 's'
         else:
             self._mdstate = 'u'
         shutil.rmtree(self._repo.vfs.join('merge'), True)
         self._results = {}
         self._dirty = False
 
     def _read(self):
         """Analyse each record content to restore a serialized state from disk
 
         This function process "record" entry produced by the de-serialization
         of on disk file.
         """
         self._state = {}
         self._stateextras = {}
         self._local = None
         self._other = None
         for var in ('localctx', 'otherctx'):
             if var in vars(self):
                 delattr(self, var)
         self._readmergedriver = None
         self._mdstate = 's'
         unsupported = set()
         records = self._readrecords()
         for rtype, record in records:
             if rtype == 'L':
                 self._local = bin(record)
             elif rtype == 'O':
                 self._other = bin(record)
             elif rtype == 'm':
                 bits = record.split('\0', 1)
                 mdstate = bits[1]
                 if len(mdstate) != 1 or mdstate not in 'ums':
                     # the merge driver should be idempotent, so just rerun it
                     mdstate = 'u'
 
                 self._readmergedriver = bits[0]
                 self._mdstate = mdstate
             elif rtype in 'FDC':
                 bits = record.split('\0')
                 self._state[bits[0]] = bits[1:]
             elif rtype == 'f':
                 filename, rawextras = record.split('\0', 1)
                 extraparts = rawextras.split('\0')
                 extras = {}
                 i = 0
                 while i < len(extraparts):
                     extras[extraparts[i]] = extraparts[i + 1]
                     i += 2
 
                 self._stateextras[filename] = extras
             elif rtype == 'l':
                 labels = record.split('\0', 2)
                 self._labels = [l for l in labels if len(l) > 0]
             elif not rtype.islower():
                 unsupported.add(rtype)
         self._results = {}
         self._dirty = False
 
         if unsupported:
             raise error.UnsupportedMergeRecords(unsupported)
 
     def _readrecords(self):
         """Read merge state from disk and return a list of record (TYPE, data)
 
         We read data from both v1 and v2 files and decide which one to use.
 
         V1 has been used by version prior to 2.9.1 and contains less data than
         v2. We read both versions and check if no data in v2 contradicts
         v1. If there is not contradiction we can safely assume that both v1
         and v2 were written at the same time and use the extract data in v2. If
         there is contradiction we ignore v2 content as we assume an old version
         of Mercurial has overwritten the mergestate file and left an old v2
         file around.
 
         returns list of record [(TYPE, data), ...]"""
         v1records = self._readrecordsv1()
         v2records = self._readrecordsv2()
         if self._v1v2match(v1records, v2records):
             return v2records
         else:
             # v1 file is newer than v2 file, use it
             # we have to infer the "other" changeset of the merge
             # we cannot do better than that with v1 of the format
             mctx = self._repo[None].parents()[-1]
             v1records.append(('O', mctx.hex()))
             # add place holder "other" file node information
             # nobody is using it yet so we do no need to fetch the data
             # if mctx was wrong `mctx[bits[-2]]` may fails.
             for idx, r in enumerate(v1records):
                 if r[0] == 'F':
                     bits = r[1].split('\0')
                     bits.insert(-2, '')
                     v1records[idx] = (r[0], '\0'.join(bits))
             return v1records
 
     def _v1v2match(self, v1records, v2records):
         oldv2 = set() # old format version of v2 record
         for rec in v2records:
             if rec[0] == 'L':
                 oldv2.add(rec)
             elif rec[0] == 'F':
                 # drop the onode data (not contained in v1)
                 oldv2.add(('F', _droponode(rec[1])))
         for rec in v1records:
             if rec not in oldv2:
                 return False
         else:
             return True
 
     def _readrecordsv1(self):
         """read on disk merge state for version 1 file
 
         returns list of record [(TYPE, data), ...]
 
         Note: the "F" data from this file are one entry short
               (no "other file node" entry)
         """
         records = []
         try:
             f = self._repo.vfs(self.statepathv1)
             for i, l in enumerate(f):
                 if i == 0:
                     records.append(('L', l[:-1]))
                 else:
                     records.append(('F', l[:-1]))
             f.close()
         except IOError as err:
             if err.errno != errno.ENOENT:
                 raise
         return records
 
     def _readrecordsv2(self):
         """read on disk merge state for version 2 file
 
         This format is a list of arbitrary records of the form:
 
           [type][length][content]
 
         `type` is a single character, `length` is a 4 byte integer, and
         `content` is an arbitrary byte sequence of length `length`.
 
         Mercurial versions prior to 3.7 have a bug where if there are
         unsupported mandatory merge records, attempting to clear out the merge
         state with hg update --clean or similar aborts. The 't' record type
         works around that by writing out what those versions treat as an
         advisory record, but later versions interpret as special: the first
         character is the 'real' record type and everything onwards is the data.
 
         Returns list of records [(TYPE, data), ...]."""
         records = []
         try:
             f = self._repo.vfs(self.statepathv2)
             data = f.read()
             off = 0
             end = len(data)
             while off < end:
                 rtype = data[off]
                 off += 1
                 length = _unpack('>I', data[off:(off + 4)])[0]
                 off += 4
                 record = data[off:(off + length)]
                 off += length
                 if rtype == 't':
                     rtype, record = record[0], record[1:]
                 records.append((rtype, record))
             f.close()
         except IOError as err:
             if err.errno != errno.ENOENT:
                 raise
         return records
 
     @util.propertycache
     def mergedriver(self):
         # protect against the following:
         # - A configures a malicious merge driver in their hgrc, then
         #   pauses the merge
         # - A edits their hgrc to remove references to the merge driver
         # - A gives a copy of their entire repo, including .hg, to B
         # - B inspects .hgrc and finds it to be clean
         # - B then continues the merge and the malicious merge driver
         #  gets invoked
         configmergedriver = self._repo.ui.config('experimental', 'mergedriver')
         if (self._readmergedriver is not None
             and self._readmergedriver != configmergedriver):
             raise error.ConfigError(
                 _("merge driver changed since merge started"),
                 hint=_("revert merge driver change or abort merge"))
 
         return configmergedriver
 
     @util.propertycache
     def localctx(self):
         if self._local is None:
             msg = "localctx accessed but self._local isn't set"
             raise error.ProgrammingError(msg)
         return self._repo[self._local]
 
     @util.propertycache
     def otherctx(self):
         if self._other is None:
             msg = "otherctx accessed but self._other isn't set"
             raise error.ProgrammingError(msg)
         return self._repo[self._other]
 
     def active(self):
         """Whether mergestate is active.
 
         Returns True if there appears to be mergestate. This is a rough proxy
         for "is a merge in progress."
         """
         # Check local variables before looking at filesystem for performance
         # reasons.
         return bool(self._local) or bool(self._state) or \
                self._repo.vfs.exists(self.statepathv1) or \
                self._repo.vfs.exists(self.statepathv2)
 
     def commit(self):
         """Write current state on disk (if necessary)"""
         if self._dirty:
             records = self._makerecords()
             self._writerecords(records)
             self._dirty = False
 
     def _makerecords(self):
         records = []
         records.append(('L', hex(self._local)))
         records.append(('O', hex(self._other)))
         if self.mergedriver:
             records.append(('m', '\0'.join([
                 self.mergedriver, self._mdstate])))
         for d, v in self._state.iteritems():
             if v[0] == 'd':
                 records.append(('D', '\0'.join([d] + v)))
             # v[1] == local ('cd'), v[6] == other ('dc') -- not supported by
             # older versions of Mercurial
             elif v[1] == nullhex or v[6] == nullhex:
                 records.append(('C', '\0'.join([d] + v)))
             else:
                 records.append(('F', '\0'.join([d] + v)))
         for filename, extras in sorted(self._stateextras.iteritems()):
             rawextras = '\0'.join('%s\0%s' % (k, v) for k, v in
                                   extras.iteritems())
             records.append(('f', '%s\0%s' % (filename, rawextras)))
         if self._labels is not None:
             labels = '\0'.join(self._labels)
             records.append(('l', labels))
         return records
 
     def _writerecords(self, records):
         """Write current state on disk (both v1 and v2)"""
         self._writerecordsv1(records)
         self._writerecordsv2(records)
 
     def _writerecordsv1(self, records):
         """Write current state on disk in a version 1 file"""
         f = self._repo.vfs(self.statepathv1, 'w')
         irecords = iter(records)
         lrecords = next(irecords)
         assert lrecords[0] == 'L'
         f.write(hex(self._local) + '\n')
         for rtype, data in irecords:
             if rtype == 'F':
                 f.write('%s\n' % _droponode(data))
         f.close()
 
     def _writerecordsv2(self, records):
         """Write current state on disk in a version 2 file
 
         See the docstring for _readrecordsv2 for why we use 't'."""
         # these are the records that all version 2 clients can read
         whitelist = 'LOF'
         f = self._repo.vfs(self.statepathv2, 'w')
         for key, data in records:
             assert len(key) == 1
             if key not in whitelist:
                 key, data = 't', '%s%s' % (key, data)
             format = '>sI%is' % len(data)
             f.write(_pack(format, key, len(data), data))
         f.close()
 
     def add(self, fcl, fco, fca, fd):
         """add a new (potentially?) conflicting file the merge state
         fcl: file context for local,
         fco: file context for remote,
         fca: file context for ancestors,
         fd:  file path of the resulting merge.
 
         note: also write the local version to the `.hg/merge` directory.
         """
         if fcl.isabsent():
             hash = nullhex
         else:
             hash = hex(hashlib.sha1(fcl.path()).digest())
             self._repo.vfs.write('merge/' + hash, fcl.data())
         self._state[fd] = ['u', hash, fcl.path(),
                            fca.path(), hex(fca.filenode()),
                            fco.path(), hex(fco.filenode()),
                            fcl.flags()]
         self._stateextras[fd] = { 'ancestorlinknode' : hex(fca.node()) }
         self._dirty = True
 
     def __contains__(self, dfile):
         return dfile in self._state
 
     def __getitem__(self, dfile):
         return self._state[dfile][0]
 
     def __iter__(self):
         return iter(sorted(self._state))
 
     def files(self):
         return self._state.keys()
 
     def mark(self, dfile, state):
         self._state[dfile][0] = state
         self._dirty = True
 
     def mdstate(self):
         return self._mdstate
 
     def unresolved(self):
         """Obtain the paths of unresolved files."""
 
         for f, entry in self._state.iteritems():
             if entry[0] == 'u':
                 yield f
 
     def driverresolved(self):
         """Obtain the paths of driver-resolved files."""
 
         for f, entry in self._state.items():
             if entry[0] == 'd':
                 yield f
 
     def extras(self, filename):
         return self._stateextras.setdefault(filename, {})
 
     def _resolve(self, preresolve, dfile, wctx):
         """rerun merge process for file path `dfile`"""
         if self[dfile] in 'rd':
             return True, 0
         stateentry = self._state[dfile]
         state, hash, lfile, afile, anode, ofile, onode, flags = stateentry
         octx = self._repo[self._other]
         extras = self.extras(dfile)
         anccommitnode = extras.get('ancestorlinknode')
         if anccommitnode:
             actx = self._repo[anccommitnode]
         else:
             actx = None
         fcd = self._filectxorabsent(hash, wctx, dfile)
         fco = self._filectxorabsent(onode, octx, ofile)
         # TODO: move this to filectxorabsent
         fca = self._repo.filectx(afile, fileid=anode, changeid=actx)
         # "premerge" x flags
         flo = fco.flags()
         fla = fca.flags()
         if 'x' in flags + flo + fla and 'l' not in flags + flo + fla:
             if fca.node() == nullid and flags != flo:
                 if preresolve:
                     self._repo.ui.warn(
                         _('warning: cannot merge flags for %s '
                           'without common ancestor - keeping local flags\n')
                         % afile)
             elif flags == fla:
                 flags = flo
         if preresolve:
             # restore local
             if hash != nullhex:
                 f = self._repo.vfs('merge/' + hash)
                 wctx[dfile].write(f.read(), flags)
                 f.close()
             else:
                 wctx[dfile].remove(ignoremissing=True)
             complete, r, deleted = filemerge.premerge(self._repo, wctx,
                                                       self._local, lfile, fcd,
                                                       fco, fca,
                                                       labels=self._labels)
         else:
             complete, r, deleted = filemerge.filemerge(self._repo, wctx,
                                                        self._local, lfile, fcd,
                                                        fco, fca,
                                                        labels=self._labels)
         if r is None:
             # no real conflict
             del self._state[dfile]
             self._stateextras.pop(dfile, None)
             self._dirty = True
         elif not r:
             self.mark(dfile, 'r')
 
         if complete:
             action = None
             if deleted:
                 if fcd.isabsent():
                     # dc: local picked. Need to drop if present, which may
                     # happen on re-resolves.
                     action = 'f'
                 else:
                     # cd: remote picked (or otherwise deleted)
                     action = 'r'
             else:
                 if fcd.isabsent(): # dc: remote picked
                     action = 'g'
                 elif fco.isabsent(): # cd: local picked
                     if dfile in self.localctx:
                         action = 'am'
                     else:
                         action = 'a'
                 # else: regular merges (no action necessary)
             self._results[dfile] = r, action
 
         return complete, r
 
     def _filectxorabsent(self, hexnode, ctx, f):
         if hexnode == nullhex:
             return filemerge.absentfilectx(ctx, f)
         else:
             return ctx[f]
 
     def preresolve(self, dfile, wctx):
         """run premerge process for dfile
 
         Returns whether the merge is complete, and the exit code."""
         return self._resolve(True, dfile, wctx)
 
     def resolve(self, dfile, wctx):
         """run merge process (assuming premerge was run) for dfile
 
         Returns the exit code of the merge."""
         return self._resolve(False, dfile, wctx)[1]
 
     def counts(self):
         """return counts for updated, merged and removed files in this
         session"""
         updated, merged, removed = 0, 0, 0
         for r, action in self._results.itervalues():
             if r is None:
                 updated += 1
             elif r == 0:
                 if action == 'r':
                     removed += 1
                 else:
                     merged += 1
         return updated, merged, removed
 
     def unresolvedcount(self):
         """get unresolved count for this merge (persistent)"""
         return len(list(self.unresolved()))
 
     def actions(self):
         """return lists of actions to perform on the dirstate"""
         actions = {'r': [], 'f': [], 'a': [], 'am': [], 'g': []}
         for f, (r, action) in self._results.iteritems():
             if action is not None:
                 actions[action].append((f, None, "merge result"))
         return actions
 
     def recordactions(self):
         """record remove/add/get actions in the dirstate"""
         branchmerge = self._repo.dirstate.p2() != nullid
         recordupdates(self._repo, self.actions(), branchmerge)
 
     def queueremove(self, f):
         """queues a file to be removed from the dirstate
 
         Meant for use by custom merge drivers."""
         self._results[f] = 0, 'r'
 
     def queueadd(self, f):
         """queues a file to be added to the dirstate
 
         Meant for use by custom merge drivers."""
         self._results[f] = 0, 'a'
 
     def queueget(self, f):
         """queues a file to be marked modified in the dirstate
 
         Meant for use by custom merge drivers."""
         self._results[f] = 0, 'g'
 
 def _getcheckunknownconfig(repo, section, name):
     config = repo.ui.config(section, name, default='abort')
     valid = ['abort', 'ignore', 'warn']
     if config not in valid:
         validstr = ', '.join(["'" + v + "'" for v in valid])
         raise error.ConfigError(_("%s.%s not valid "
                                   "('%s' is none of %s)")
                                 % (section, name, config, validstr))
     return config
 
 def _checkunknownfile(repo, wctx, mctx, f, f2=None):
     if f2 is None:
         f2 = f
     return (repo.wvfs.audit.check(f)
         and repo.wvfs.isfileorlink(f)
         and repo.dirstate.normalize(f) not in repo.dirstate
         and mctx[f2].cmp(wctx[f]))
 
 def _checkunknownfiles(repo, wctx, mctx, force, actions, mergeforce):
     """
     Considers any actions that care about the presence of conflicting unknown
     files. For some actions, the result is to abort; for others, it is to
     choose a different action.
     """
     conflicts = set()
     warnconflicts = set()
     abortconflicts = set()
     unknownconfig = _getcheckunknownconfig(repo, 'merge', 'checkunknown')
     ignoredconfig = _getcheckunknownconfig(repo, 'merge', 'checkignored')
     if not force:
         def collectconflicts(conflicts, config):
             if config == 'abort':
                 abortconflicts.update(conflicts)
             elif config == 'warn':
                 warnconflicts.update(conflicts)
 
         for f, (m, args, msg) in actions.iteritems():
             if m in ('c', 'dc'):
                 if _checkunknownfile(repo, wctx, mctx, f):
                     conflicts.add(f)
             elif m == 'dg':
                 if _checkunknownfile(repo, wctx, mctx, f, args[0]):
                     conflicts.add(f)
 
         ignoredconflicts = set([c for c in conflicts
                                 if repo.dirstate._ignore(c)])
         unknownconflicts = conflicts - ignoredconflicts
         collectconflicts(ignoredconflicts, ignoredconfig)
         collectconflicts(unknownconflicts, unknownconfig)
     else:
         for f, (m, args, msg) in actions.iteritems():
             if m == 'cm':
                 fl2, anc = args
                 different = _checkunknownfile(repo, wctx, mctx, f)
                 if repo.dirstate._ignore(f):
                     config = ignoredconfig
                 else:
                     config = unknownconfig
 
                 # The behavior when force is True is described by this table:
                 #  config  different  mergeforce  |    action    backup
                 #    *         n          *       |      get        n
                 #    *         y          y       |     merge       -
                 #   abort      y          n       |     merge       -   (1)
                 #   warn       y          n       |  warn + get     y
                 #  ignore      y          n       |      get        y
                 #
                 # (1) this is probably the wrong behavior here -- we should
                 #     probably abort, but some actions like rebases currently
                 #     don't like an abort happening in the middle of
                 #     merge.update.
                 if not different:
                     actions[f] = ('g', (fl2, False), "remote created")
                 elif mergeforce or config == 'abort':
                     actions[f] = ('m', (f, f, None, False, anc),
                                   "remote differs from untracked local")
                 elif config == 'abort':
                     abortconflicts.add(f)
                 else:
                     if config == 'warn':
                         warnconflicts.add(f)
                     actions[f] = ('g', (fl2, True), "remote created")
 
     for f in sorted(abortconflicts):
         repo.ui.warn(_("%s: untracked file differs\n") % f)
     if abortconflicts:
         raise error.Abort(_("untracked files in working directory "
                             "differ from files in requested revision"))
 
     for f in sorted(warnconflicts):
         repo.ui.warn(_("%s: replacing untracked file\n") % f)
 
     for f, (m, args, msg) in actions.iteritems():
         backup = f in conflicts
         if m == 'c':
             flags, = args
             actions[f] = ('g', (flags, backup), msg)
 
 def _forgetremoved(wctx, mctx, branchmerge):
     """
     Forget removed files
 
     If we're jumping between revisions (as opposed to merging), and if
     neither the working directory nor the target rev has the file,
     then we need to remove it from the dirstate, to prevent the
     dirstate from listing the file when it is no longer in the
     manifest.
 
     If we're merging, and the other revision has removed a file
     that is not present in the working directory, we need to mark it
     as removed.
     """
 
     actions = {}
     m = 'f'
     if branchmerge:
         m = 'r'
     for f in wctx.deleted():
         if f not in mctx:
             actions[f] = m, None, "forget deleted"
 
     if not branchmerge:
         for f in wctx.removed():
             if f not in mctx:
                 actions[f] = 'f', None, "forget removed"
 
     return actions
 
 def _checkcollision(repo, wmf, actions):
     # build provisional merged manifest up
     pmmf = set(wmf)
 
     if actions:
         # k, dr, e and rd are no-op
         for m in 'a', 'am', 'f', 'g', 'cd', 'dc':
             for f, args, msg in actions[m]:
                 pmmf.add(f)
         for f, args, msg in actions['r']:
             pmmf.discard(f)
         for f, args, msg in actions['dm']:
             f2, flags = args
             pmmf.discard(f2)
             pmmf.add(f)
         for f, args, msg in actions['dg']:
             pmmf.add(f)
         for f, args, msg in actions['m']:
             f1, f2, fa, move, anc = args
             if move:
                 pmmf.discard(f1)
             pmmf.add(f)
 
     # check case-folding collision in provisional merged manifest
     foldmap = {}
     for f in pmmf:
         fold = util.normcase(f)
         if fold in foldmap:
             raise error.Abort(_("case-folding collision between %s and %s")
                              % (f, foldmap[fold]))
         foldmap[fold] = f
 
     # check case-folding of directories
     foldprefix = unfoldprefix = lastfull = ''
     for fold, f in sorted(foldmap.items()):
         if fold.startswith(foldprefix) and not f.startswith(unfoldprefix):
             # the folded prefix matches but actual casing is different
             raise error.Abort(_("case-folding collision between "
                                 "%s and directory of %s") % (lastfull, f))
         foldprefix = fold + '/'
         unfoldprefix = f + '/'
         lastfull = f
 
 def driverpreprocess(repo, ms, wctx, labels=None):
     """run the preprocess step of the merge driver, if any
 
     This is currently not implemented -- it's an extension point."""
     return True
 
 def driverconclude(repo, ms, wctx, labels=None):
     """run the conclude step of the merge driver, if any
 
     This is currently not implemented -- it's an extension point."""
     return True
 
 def manifestmerge(repo, wctx, p2, pa, branchmerge, force, matcher,
                   acceptremote, followcopies, forcefulldiff=False):
     """
     Merge wctx and p2 with ancestor pa and generate merge action list
 
     branchmerge and force are as passed in to update
     matcher = matcher to filter file lists
     acceptremote = accept the incoming changes without prompting
     """
     if matcher is not None and matcher.always():
         matcher = None
 
     copy, movewithdir, diverge, renamedelete, dirmove = {}, {}, {}, {}, {}
 
     # manifests fetched in order are going to be faster, so prime the caches
     [x.manifest() for x in
      sorted(wctx.parents() + [p2, pa], key=scmutil.intrev)]
 
     if followcopies:
         ret = copies.mergecopies(repo, wctx, p2, pa)
         copy, movewithdir, diverge, renamedelete, dirmove = ret
 
     boolbm = pycompat.bytestr(bool(branchmerge))
     boolf = pycompat.bytestr(bool(force))
     boolm = pycompat.bytestr(bool(matcher))
     repo.ui.note(_("resolving manifests\n"))
     repo.ui.debug(" branchmerge: %s, force: %s, partial: %s\n"
                   % (boolbm, boolf, boolm))
     repo.ui.debug(" ancestor: %s, local: %s, remote: %s\n" % (pa, wctx, p2))
 
     m1, m2, ma = wctx.manifest(), p2.manifest(), pa.manifest()
     copied = set(copy.values())
     copied.update(movewithdir.values())
 
     if '.hgsubstate' in m1:
         # check whether sub state is modified
         if any(wctx.sub(s).dirty() for s in wctx.substate):
             m1['.hgsubstate'] = modifiednodeid
 
     # Don't use m2-vs-ma optimization if:
     # - ma is the same as m1 or m2, which we're just going to diff again later
     # - The caller specifically asks for a full diff, which is useful during bid
     #   merge.
     if (pa not in ([wctx, p2] + wctx.parents()) and not forcefulldiff):
         # Identify which files are relevant to the merge, so we can limit the
         # total m1-vs-m2 diff to just those files. This has significant
         # performance benefits in large repositories.
         relevantfiles = set(ma.diff(m2).keys())
 
         # For copied and moved files, we need to add the source file too.
         for copykey, copyvalue in copy.iteritems():
             if copyvalue in relevantfiles:
                 relevantfiles.add(copykey)
         for movedirkey in movewithdir:
             relevantfiles.add(movedirkey)
         filesmatcher = scmutil.matchfiles(repo, relevantfiles)
         matcher = matchmod.intersectmatchers(matcher, filesmatcher)
 
     diff = m1.diff(m2, match=matcher)
 
     if matcher is None:
         matcher = matchmod.always('', '')
 
     actions = {}
     for f, ((n1, fl1), (n2, fl2)) in diff.iteritems():
         if n1 and n2: # file exists on both local and remote side
             if f not in ma:
                 fa = copy.get(f, None)
                 if fa is not None:
                     actions[f] = ('m', (f, f, fa, False, pa.node()),
                                   "both renamed from " + fa)
                 else:
                     actions[f] = ('m', (f, f, None, False, pa.node()),
                                   "both created")
             else:
                 a = ma[f]
                 fla = ma.flags(f)
                 nol = 'l' not in fl1 + fl2 + fla
                 if n2 == a and fl2 == fla:
                     actions[f] = ('k' , (), "remote unchanged")
                 elif n1 == a and fl1 == fla: # local unchanged - use remote
                     if n1 == n2: # optimization: keep local content
                         actions[f] = ('e', (fl2,), "update permissions")
                     else:
                         actions[f] = ('g', (fl2, False), "remote is newer")
                 elif nol and n2 == a: # remote only changed 'x'
                     actions[f] = ('e', (fl2,), "update permissions")
                 elif nol and n1 == a: # local only changed 'x'
                     actions[f] = ('g', (fl1, False), "remote is newer")
                 else: # both changed something
                     actions[f] = ('m', (f, f, f, False, pa.node()),
                                    "versions differ")
         elif n1: # file exists only on local side
             if f in copied:
                 pass # we'll deal with it on m2 side
             elif f in movewithdir: # directory rename, move local
                 f2 = movewithdir[f]
                 if f2 in m2:
                     actions[f2] = ('m', (f, f2, None, True, pa.node()),
                                    "remote directory rename, both created")
                 else:
                     actions[f2] = ('dm', (f, fl1),
                                    "remote directory rename - move from " + f)
             elif f in copy:
                 f2 = copy[f]
                 actions[f] = ('m', (f, f2, f2, False, pa.node()),
                               "local copied/moved from " + f2)
             elif f in ma: # clean, a different, no remote
                 if n1 != ma[f]:
                     if acceptremote:
                         actions[f] = ('r', None, "remote delete")
                     else:
                         actions[f] = ('cd', (f, None, f, False, pa.node()),
                                       "prompt changed/deleted")
                 elif n1 == addednodeid:
                     # This extra 'a' is added by working copy manifest to mark
                     # the file as locally added. We should forget it instead of
                     # deleting it.
                     actions[f] = ('f', None, "remote deleted")
                 else:
                     actions[f] = ('r', None, "other deleted")
         elif n2: # file exists only on remote side
             if f in copied:
                 pass # we'll deal with it on m1 side
             elif f in movewithdir:
                 f2 = movewithdir[f]
                 if f2 in m1:
                     actions[f2] = ('m', (f2, f, None, False, pa.node()),
                                    "local directory rename, both created")
                 else:
                     actions[f2] = ('dg', (f, fl2),
                                    "local directory rename - get from " + f)
             elif f in copy:
                 f2 = copy[f]
                 if f2 in m2:
                     actions[f] = ('m', (f2, f, f2, False, pa.node()),
                                   "remote copied from " + f2)
                 else:
                     actions[f] = ('m', (f2, f, f2, True, pa.node()),
                                   "remote moved from " + f2)
             elif f not in ma:
                 # local unknown, remote created: the logic is described by the
                 # following table:
                 #
                 # force  branchmerge  different  |  action
                 #   n         *           *      |   create
                 #   y         n           *      |   create
                 #   y         y           n      |   create
                 #   y         y           y      |   merge
                 #
                 # Checking whether the files are different is expensive, so we
                 # don't do that when we can avoid it.
                 if not force:
                     actions[f] = ('c', (fl2,), "remote created")
                 elif not branchmerge:
                     actions[f] = ('c', (fl2,), "remote created")
                 else:
                     actions[f] = ('cm', (fl2, pa.node()),
                                   "remote created, get or merge")
             elif n2 != ma[f]:
                 df = None
                 for d in dirmove:
                     if f.startswith(d):
                         # new file added in a directory that was moved
                         df = dirmove[d] + f[len(d):]
                         break
                 if df is not None and df in m1:
                     actions[df] = ('m', (df, f, f, False, pa.node()),
                             "local directory rename - respect move from " + f)
                 elif acceptremote:
                     actions[f] = ('c', (fl2,), "remote recreating")
                 else:
                     actions[f] = ('dc', (None, f, f, False, pa.node()),
                                   "prompt deleted/changed")
 
     return actions, diverge, renamedelete
 
 def _resolvetrivial(repo, wctx, mctx, ancestor, actions):
     """Resolves false conflicts where the nodeid changed but the content
        remained the same."""
 
     for f, (m, args, msg) in actions.items():
         if m == 'cd' and f in ancestor and not wctx[f].cmp(ancestor[f]):
             # local did change but ended up with same content
             actions[f] = 'r', None, "prompt same"
         elif m == 'dc' and f in ancestor and not mctx[f].cmp(ancestor[f]):
             # remote did change but ended up with same content
             del actions[f] # don't get = keep local deleted
 
 def calculateupdates(repo, wctx, mctx, ancestors, branchmerge, force,
                      acceptremote, followcopies, matcher=None,
                      mergeforce=False):
     """Calculate the actions needed to merge mctx into wctx using ancestors"""
     # Avoid cycle.
     from . import sparse
 
     if len(ancestors) == 1: # default
         actions, diverge, renamedelete = manifestmerge(
             repo, wctx, mctx, ancestors[0], branchmerge, force, matcher,
             acceptremote, followcopies)
         _checkunknownfiles(repo, wctx, mctx, force, actions, mergeforce)
 
     else: # only when merge.preferancestor=* - the default
         repo.ui.note(
             _("note: merging %s and %s using bids from ancestors %s\n") %
             (wctx, mctx, _(' and ').join(str(anc) for anc in ancestors)))
 
         # Call for bids
         fbids = {} # mapping filename to bids (action method to list af actions)
         diverge, renamedelete = None, None
         for ancestor in ancestors:
             repo.ui.note(_('\ncalculating bids for ancestor %s\n') % ancestor)
             actions, diverge1, renamedelete1 = manifestmerge(
                 repo, wctx, mctx, ancestor, branchmerge, force, matcher,
                 acceptremote, followcopies, forcefulldiff=True)
             _checkunknownfiles(repo, wctx, mctx, force, actions, mergeforce)
 
             # Track the shortest set of warning on the theory that bid
             # merge will correctly incorporate more information
             if diverge is None or len(diverge1) < len(diverge):
                 diverge = diverge1
             if renamedelete is None or len(renamedelete) < len(renamedelete1):
                 renamedelete = renamedelete1
 
             for f, a in sorted(actions.iteritems()):
                 m, args, msg = a
                 repo.ui.debug(' %s: %s -> %s\n' % (f, msg, m))
                 if f in fbids:
                     d = fbids[f]
                     if m in d:
                         d[m].append(a)
                     else:
                         d[m] = [a]
                 else:
                     fbids[f] = {m: [a]}
 
         # Pick the best bid for each file
         repo.ui.note(_('\nauction for merging merge bids\n'))
         actions = {}
         dms = [] # filenames that have dm actions
         for f, bids in sorted(fbids.items()):
             # bids is a mapping from action method to list af actions
             # Consensus?
             if len(bids) == 1: # all bids are the same kind of method
-                m, l = bids.items()[0]
+                m, l = list(bids.items())[0]
                 if all(a == l[0] for a in l[1:]): # len(bids) is > 1
                     repo.ui.note(_(" %s: consensus for %s\n") % (f, m))
                     actions[f] = l[0]
                     if m == 'dm':
                         dms.append(f)
                     continue
             # If keep is an option, just do it.
             if 'k' in bids:
                 repo.ui.note(_(" %s: picking 'keep' action\n") % f)
                 actions[f] = bids['k'][0]
                 continue
             # If there are gets and they all agree [how could they not?], do it.
             if 'g' in bids:
                 ga0 = bids['g'][0]
                 if all(a == ga0 for a in bids['g'][1:]):
                     repo.ui.note(_(" %s: picking 'get' action\n") % f)
                     actions[f] = ga0
                     continue
             # TODO: Consider other simple actions such as mode changes
             # Handle inefficient democrazy.
             repo.ui.note(_(' %s: multiple bids for merge action:\n') % f)
             for m, l in sorted(bids.items()):
                 for _f, args, msg in l:
                     repo.ui.note('  %s -> %s\n' % (msg, m))
             # Pick random action. TODO: Instead, prompt user when resolving
-            m, l = bids.items()[0]
+            m, l = list(bids.items())[0]
             repo.ui.warn(_(' %s: ambiguous merge - picked %s action\n') %
                          (f, m))
             actions[f] = l[0]
             if m == 'dm':
                 dms.append(f)
             continue
         # Work around 'dm' that can cause multiple actions for the same file
         for f in dms:
             dm, (f0, flags), msg = actions[f]
             assert dm == 'dm', dm
             if f0 in actions and actions[f0][0] == 'r':
                 # We have one bid for removing a file and another for moving it.
                 # These two could be merged as first move and then delete ...
                 # but instead drop moving and just delete.
                 del actions[f]
         repo.ui.note(_('end of auction\n\n'))
 
     _resolvetrivial(repo, wctx, mctx, ancestors[0], actions)
 
     if wctx.rev() is None:
         fractions = _forgetremoved(wctx, mctx, branchmerge)
         actions.update(fractions)
 
     prunedactions = sparse.filterupdatesactions(repo, wctx, mctx, branchmerge,
                                                 actions)
 
     return prunedactions, diverge, renamedelete
 
 def _getcwd():
     try:
         return pycompat.getcwd()
     except OSError as err:
         if err.errno == errno.ENOENT:
             return None
         raise
 
 def batchremove(repo, wctx, actions):
     """apply removes to the working directory
 
     yields tuples for progress updates
     """
     verbose = repo.ui.verbose
     cwd = _getcwd()
     i = 0
     for f, args, msg in actions:
         repo.ui.debug(" %s: %s -> r\n" % (f, msg))
         if verbose:
             repo.ui.note(_("removing %s\n") % f)
         wctx[f].audit()
         try:
             wctx[f].remove(ignoremissing=True)
         except OSError as inst:
             repo.ui.warn(_("update failed to remove %s: %s!\n") %
                          (f, inst.strerror))
         if i == 100:
             yield i, f
             i = 0
         i += 1
     if i > 0:
         yield i, f
 
     if cwd and not _getcwd():
         # cwd was removed in the course of removing files; print a helpful
         # warning.
         repo.ui.warn(_("current directory was removed\n"
                        "(consider changing to repo root: %s)\n") % repo.root)
 
     # It's necessary to flush here in case we're inside a worker fork and will
     # quit after this function.
     wctx.flushall()
 
 def batchget(repo, mctx, wctx, actions):
     """apply gets to the working directory
 
     mctx is the context to get from
 
     yields tuples for progress updates
     """
     verbose = repo.ui.verbose
     fctx = mctx.filectx
     ui = repo.ui
     i = 0
     with repo.wvfs.backgroundclosing(ui, expectedcount=len(actions)):
         for f, (flags, backup), msg in actions:
             repo.ui.debug(" %s: %s -> g\n" % (f, msg))
             if verbose:
                 repo.ui.note(_("getting %s\n") % f)
 
             if backup:
                 absf = repo.wjoin(f)
                 orig = scmutil.origpath(ui, repo, absf)
                 try:
                     if repo.wvfs.isfileorlink(f):
                         util.rename(absf, orig)
                 except OSError as e:
                     if e.errno != errno.ENOENT:
                         raise
             wctx[f].clearunknown()
             wctx[f].write(fctx(f).data(), flags, backgroundclose=True)
             if i == 100:
                 yield i, f
                 i = 0
             i += 1
     if i > 0:
         yield i, f
 
     # It's necessary to flush here in case we're inside a worker fork and will
     # quit after this function.
     wctx.flushall()
 
 def applyupdates(repo, actions, wctx, mctx, overwrite, labels=None):
     """apply the merge action list to the working directory
 
     wctx is the working copy context
     mctx is the context to be merged into the working copy
 
     Return a tuple of counts (updated, merged, removed, unresolved) that
     describes how many files were affected by the update.
     """
 
     updated, merged, removed = 0, 0, 0
     ms = mergestate.clean(repo, wctx.p1().node(), mctx.node(), labels)
     moves = []
     for m, l in actions.items():
         l.sort()
 
     # 'cd' and 'dc' actions are treated like other merge conflicts
     mergeactions = sorted(actions['cd'])
     mergeactions.extend(sorted(actions['dc']))
     mergeactions.extend(actions['m'])
     for f, args, msg in mergeactions:
         f1, f2, fa, move, anc = args
         if f == '.hgsubstate': # merged internally
             continue
         if f1 is None:
             fcl = filemerge.absentfilectx(wctx, fa)
         else:
             repo.ui.debug(" preserving %s for resolve of %s\n" % (f1, f))
             fcl = wctx[f1]
         if f2 is None:
             fco = filemerge.absentfilectx(mctx, fa)
         else:
             fco = mctx[f2]
         actx = repo[anc]
         if fa in actx:
             fca = actx[fa]
         else:
             # TODO: move to absentfilectx
             fca = repo.filectx(f1, fileid=nullrev)
         ms.add(fcl, fco, fca, f)
         if f1 != f and move:
             moves.append(f1)
 
     _updating = _('updating')
     _files = _('files')
     progress = repo.ui.progress
 
     # remove renamed files after safely stored
     for f in moves:
         if wctx[f].lexists():
             repo.ui.debug("removing %s\n" % f)
             wctx[f].audit()
             wctx[f].remove()
 
     numupdates = sum(len(l) for m, l in actions.items() if m != 'k')
 
     if [a for a in actions['r'] if a[0] == '.hgsubstate']:
         subrepo.submerge(repo, wctx, mctx, wctx, overwrite, labels)
 
     # remove in parallel (must come first)
     z = 0
     prog = worker.worker(repo.ui, 0.001, batchremove, (repo, wctx),
                          actions['r'])
     for i, item in prog:
         z += i
         progress(_updating, z, item=item, total=numupdates, unit=_files)
     removed = len(actions['r'])
 
     # We should flush before forking into worker processes, since those workers
     # flush when they complete, and we don't want to duplicate work.
     wctx.flushall()
 
     # get in parallel
     prog = worker.worker(repo.ui, 0.001, batchget, (repo, mctx, wctx),
                          actions['g'])
     for i, item in prog:
         z += i
         progress(_updating, z, item=item, total=numupdates, unit=_files)
     updated = len(actions['g'])
 
     if [a for a in actions['g'] if a[0] == '.hgsubstate']:
         subrepo.submerge(repo, wctx, mctx, wctx, overwrite, labels)
 
     # forget (manifest only, just log it) (must come first)
     for f, args, msg in actions['f']:
         repo.ui.debug(" %s: %s -> f\n" % (f, msg))
         z += 1
         progress(_updating, z, item=f, total=numupdates, unit=_files)
 
     # re-add (manifest only, just log it)
     for f, args, msg in actions['a']:
         repo.ui.debug(" %s: %s -> a\n" % (f, msg))
         z += 1
         progress(_updating, z, item=f, total=numupdates, unit=_files)
 
     # re-add/mark as modified (manifest only, just log it)
     for f, args, msg in actions['am']:
         repo.ui.debug(" %s: %s -> am\n" % (f, msg))
         z += 1
         progress(_updating, z, item=f, total=numupdates, unit=_files)
 
     # keep (noop, just log it)
     for f, args, msg in actions['k']:
         repo.ui.debug(" %s: %s -> k\n" % (f, msg))
         # no progress
 
     # directory rename, move local
     for f, args, msg in actions['dm']:
         repo.ui.debug(" %s: %s -> dm\n" % (f, msg))
         z += 1
         progress(_updating, z, item=f, total=numupdates, unit=_files)
         f0, flags = args
         repo.ui.note(_("moving %s to %s\n") % (f0, f))
         wctx[f].audit()
         wctx[f].write(wctx.filectx(f0).data(), flags)
         wctx[f0].remove()
         updated += 1
 
     # local directory rename, get
     for f, args, msg in actions['dg']:
         repo.ui.debug(" %s: %s -> dg\n" % (f, msg))
         z += 1
         progress(_updating, z, item=f, total=numupdates, unit=_files)
         f0, flags = args
         repo.ui.note(_("getting %s to %s\n") % (f0, f))
         wctx[f].write(mctx.filectx(f0).data(), flags)
         updated += 1
 
     # exec
     for f, args, msg in actions['e']:
         repo.ui.debug(" %s: %s -> e\n" % (f, msg))
         z += 1
         progress(_updating, z, item=f, total=numupdates, unit=_files)
         flags, = args
         wctx[f].audit()
         wctx[f].setflags('l' in flags, 'x' in flags)
         updated += 1
 
     # the ordering is important here -- ms.mergedriver will raise if the merge
     # driver has changed, and we want to be able to bypass it when overwrite is
     # True
     usemergedriver = not overwrite and mergeactions and ms.mergedriver
 
     if usemergedriver:
         ms.commit()
         proceed = driverpreprocess(repo, ms, wctx, labels=labels)
         # the driver might leave some files unresolved
         unresolvedf = set(ms.unresolved())
         if not proceed:
             # XXX setting unresolved to at least 1 is a hack to make sure we
             # error out
             return updated, merged, removed, max(len(unresolvedf), 1)
         newactions = []
         for f, args, msg in mergeactions:
             if f in unresolvedf:
                 newactions.append((f, args, msg))
         mergeactions = newactions
 
     # premerge
     tocomplete = []
     for f, args, msg in mergeactions:
         repo.ui.debug(" %s: %s -> m (premerge)\n" % (f, msg))
         z += 1
         progress(_updating, z, item=f, total=numupdates, unit=_files)
         if f == '.hgsubstate': # subrepo states need updating
             subrepo.submerge(repo, wctx, mctx, wctx.ancestor(mctx),
                              overwrite, labels)
             continue
         wctx[f].audit()
         complete, r = ms.preresolve(f, wctx)
         if not complete:
             numupdates += 1
             tocomplete.append((f, args, msg))
 
     # merge
     for f, args, msg in tocomplete:
         repo.ui.debug(" %s: %s -> m (merge)\n" % (f, msg))
         z += 1
         progress(_updating, z, item=f, total=numupdates, unit=_files)
         ms.resolve(f, wctx)
 
     ms.commit()
 
     unresolved = ms.unresolvedcount()
 
     if usemergedriver and not unresolved and ms.mdstate() != 's':
         if not driverconclude(repo, ms, wctx, labels=labels):
             # XXX setting unresolved to at least 1 is a hack to make sure we
             # error out
             unresolved = max(unresolved, 1)
 
         ms.commit()
 
     msupdated, msmerged, msremoved = ms.counts()
     updated += msupdated
     merged += msmerged
     removed += msremoved
 
     extraactions = ms.actions()
     if extraactions:
         mfiles = set(a[0] for a in actions['m'])
         for k, acts in extraactions.iteritems():
             actions[k].extend(acts)
             # Remove these files from actions['m'] as well. This is important
             # because in recordupdates, files in actions['m'] are processed
             # after files in other actions, and the merge driver might add
             # files to those actions via extraactions above. This can lead to a
             # file being recorded twice, with poor results. This is especially
             # problematic for actions['r'] (currently only possible with the
             # merge driver in the initial merge process; interrupted merges
             # don't go through this flow).
             #
             # The real fix here is to have indexes by both file and action so
             # that when the action for a file is changed it is automatically
             # reflected in the other action lists. But that involves a more
             # complex data structure, so this will do for now.
             #
             # We don't need to do the same operation for 'dc' and 'cd' because
             # those lists aren't consulted again.
             mfiles.difference_update(a[0] for a in acts)
 
         actions['m'] = [a for a in actions['m'] if a[0] in mfiles]
 
     progress(_updating, None, total=numupdates, unit=_files)
 
     return updated, merged, removed, unresolved
 
 def recordupdates(repo, actions, branchmerge):
     "record merge actions to the dirstate"
     # remove (must come first)
     for f, args, msg in actions.get('r', []):
         if branchmerge:
             repo.dirstate.remove(f)
         else:
             repo.dirstate.drop(f)
 
     # forget (must come first)
     for f, args, msg in actions.get('f', []):
         repo.dirstate.drop(f)
 
     # re-add
     for f, args, msg in actions.get('a', []):
         repo.dirstate.add(f)
 
     # re-add/mark as modified
     for f, args, msg in actions.get('am', []):
         if branchmerge:
             repo.dirstate.normallookup(f)
         else:
             repo.dirstate.add(f)
 
     # exec change
     for f, args, msg in actions.get('e', []):
         repo.dirstate.normallookup(f)
 
     # keep
     for f, args, msg in actions.get('k', []):
         pass
 
     # get
     for f, args, msg in actions.get('g', []):
         if branchmerge:
             repo.dirstate.otherparent(f)
         else:
             repo.dirstate.normal(f)
 
     # merge
     for f, args, msg in actions.get('m', []):
         f1, f2, fa, move, anc = args
         if branchmerge:
             # We've done a branch merge, mark this file as merged
             # so that we properly record the merger later
             repo.dirstate.merge(f)
             if f1 != f2: # copy/rename
                 if move:
                     repo.dirstate.remove(f1)
                 if f1 != f:
                     repo.dirstate.copy(f1, f)
                 else:
                     repo.dirstate.copy(f2, f)
         else:
             # We've update-merged a locally modified file, so
             # we set the dirstate to emulate a normal checkout
             # of that file some time in the past. Thus our
             # merge will appear as a normal local file
             # modification.
             if f2 == f: # file not locally copied/moved
                 repo.dirstate.normallookup(f)
             if move:
                 repo.dirstate.drop(f1)
 
     # directory rename, move local
     for f, args, msg in actions.get('dm', []):
         f0, flag = args
         if branchmerge:
             repo.dirstate.add(f)
             repo.dirstate.remove(f0)
             repo.dirstate.copy(f0, f)
         else:
             repo.dirstate.normal(f)
             repo.dirstate.drop(f0)
 
     # directory rename, get
     for f, args, msg in actions.get('dg', []):
         f0, flag = args
         if branchmerge:
             repo.dirstate.add(f)
             repo.dirstate.copy(f0, f)
         else:
             repo.dirstate.normal(f)
 
 def update(repo, node, branchmerge, force, ancestor=None,
            mergeancestor=False, labels=None, matcher=None, mergeforce=False,
            updatecheck=None, wc=None):
     """
     Perform a merge between the working directory and the given node
 
     node = the node to update to
     branchmerge = whether to merge between branches
     force = whether to force branch merging or file overwriting
     matcher = a matcher to filter file lists (dirstate not updated)
     mergeancestor = whether it is merging with an ancestor. If true,
       we should accept the incoming changes for any prompts that occur.
       If false, merging with an ancestor (fast-forward) is only allowed
       between different named branches. This flag is used by rebase extension
       as a temporary fix and should be avoided in general.
     labels = labels to use for base, local and other
     mergeforce = whether the merge was run with 'merge --force' (deprecated): if
       this is True, then 'force' should be True as well.
 
     The table below shows all the behaviors of the update command
     given the -c and -C or no options, whether the working directory
     is dirty, whether a revision is specified, and the relationship of
     the parent rev to the target rev (linear or not). Match from top first. The
     -n option doesn't exist on the command line, but represents the
     experimental.updatecheck=noconflict option.
 
     This logic is tested by test-update-branches.t.
 
     -c  -C  -n  -m  dirty  rev  linear  |  result
      y   y   *   *    *     *     *     |    (1)
      y   *   y   *    *     *     *     |    (1)
      y   *   *   y    *     *     *     |    (1)
      *   y   y   *    *     *     *     |    (1)
      *   y   *   y    *     *     *     |    (1)
      *   *   y   y    *     *     *     |    (1)
      *   *   *   *    *     n     n     |     x
      *   *   *   *    n     *     *     |    ok
      n   n   n   n    y     *     y     |   merge
      n   n   n   n    y     y     n     |    (2)
      n   n   n   y    y     *     *     |   merge
      n   n   y   n    y     *     *     |  merge if no conflict
      n   y   n   n    y     *     *     |  discard
      y   n   n   n    y     *     *     |    (3)
 
     x = can't happen
     * = don't-care
     1 = incompatible options (checked in commands.py)
     2 = abort: uncommitted changes (commit or update --clean to discard changes)
     3 = abort: uncommitted changes (checked in commands.py)
 
     The merge is performed inside ``wc``, a workingctx-like objects. It defaults
     to repo[None] if None is passed.
 
     Return the same tuple as applyupdates().
     """
     # Avoid cycle.
     from . import sparse
 
     # This function used to find the default destination if node was None, but
     # that's now in destutil.py.
     assert node is not None
     if not branchmerge and not force:
         # TODO: remove the default once all callers that pass branchmerge=False
         # and force=False pass a value for updatecheck. We may want to allow
         # updatecheck='abort' to better suppport some of these callers.
         if updatecheck is None:
             updatecheck = 'linear'
         assert updatecheck in ('none', 'linear', 'noconflict')
     # If we're doing a partial update, we need to skip updating
     # the dirstate, so make a note of any partial-ness to the
     # update here.
     if matcher is None or matcher.always():
         partial = False
     else:
         partial = True
     with repo.wlock():
         if wc is None:
             wc = repo[None]
         pl = wc.parents()
         p1 = pl[0]
         pas = [None]
         if ancestor is not None:
             pas = [repo[ancestor]]
 
         overwrite = force and not branchmerge
 
         p2 = repo[node]
         if pas[0] is None:
             if repo.ui.configlist('merge', 'preferancestor', ['*']) == ['*']:
                 cahs = repo.changelog.commonancestorsheads(p1.node(), p2.node())
                 pas = [repo[anc] for anc in (sorted(cahs) or [nullid])]
             else:
                 pas = [p1.ancestor(p2, warn=branchmerge)]
 
         fp1, fp2, xp1, xp2 = p1.node(), p2.node(), str(p1), str(p2)
 
         ### check phase
         if not overwrite:
             if len(pl) > 1:
                 raise error.Abort(_("outstanding uncommitted merge"))
             ms = mergestate.read(repo)
             if list(ms.unresolved()):
                 raise error.Abort(_("outstanding merge conflicts"))
         if branchmerge:
             if pas == [p2]:
                 raise error.Abort(_("merging with a working directory ancestor"
                                    " has no effect"))
             elif pas == [p1]:
                 if not mergeancestor and wc.branch() == p2.branch():
                     raise error.Abort(_("nothing to merge"),
                                      hint=_("use 'hg update' "
                                             "or check 'hg heads'"))
             if not force and (wc.files() or wc.deleted()):
                 raise error.Abort(_("uncommitted changes"),
                                  hint=_("use 'hg status' to list changes"))
             for s in sorted(wc.substate):
                 wc.sub(s).bailifchanged()
 
         elif not overwrite:
             if p1 == p2: # no-op update
                 # call the hooks and exit early
                 repo.hook('preupdate', throw=True, parent1=xp2, parent2='')
                 repo.hook('update', parent1=xp2, parent2='', error=0)
                 return 0, 0, 0, 0
 
             if (updatecheck == 'linear' and
                     pas not in ([p1], [p2])):  # nonlinear
                 dirty = wc.dirty(missing=True)
                 if dirty:
                     # Branching is a bit strange to ensure we do the minimal
                     # amount of call to obsutil.foreground.
                     foreground = obsutil.foreground(repo, [p1.node()])
                     # note: the <node> variable contains a random identifier
                     if repo[node].node() in foreground:
                         pass # allow updating to successors
                     else:
                         msg = _("uncommitted changes")
                         hint = _("commit or update --clean to discard changes")
                         raise error.UpdateAbort(msg, hint=hint)
                 else:
                     # Allow jumping branches if clean and specific rev given
                     pass
 
         if overwrite:
             pas = [wc]
         elif not branchmerge:
             pas = [p1]
 
         # deprecated config: merge.followcopies
         followcopies = repo.ui.configbool('merge', 'followcopies')
         if overwrite:
             followcopies = False
         elif not pas[0]:
             followcopies = False
         if not branchmerge and not wc.dirty(missing=True):
             followcopies = False
 
         ### calculate phase
         actionbyfile, diverge, renamedelete = calculateupdates(
             repo, wc, p2, pas, branchmerge, force, mergeancestor,
             followcopies, matcher=matcher, mergeforce=mergeforce)
 
         if updatecheck == 'noconflict':
             for f, (m, args, msg) in actionbyfile.iteritems():
                 if m not in ('g', 'k', 'e', 'r'):
                     msg = _("conflicting changes")
                     hint = _("commit or update --clean to discard changes")
                     raise error.Abort(msg, hint=hint)
 
         # Prompt and create actions. Most of this is in the resolve phase
         # already, but we can't handle .hgsubstate in filemerge or
         # subrepo.submerge yet so we have to keep prompting for it.
         if '.hgsubstate' in actionbyfile:
             f = '.hgsubstate'
             m, args, msg = actionbyfile[f]
             prompts = filemerge.partextras(labels)
             prompts['f'] = f
             if m == 'cd':
                 if repo.ui.promptchoice(
                     _("local%(l)s changed %(f)s which other%(o)s deleted\n"
                       "use (c)hanged version or (d)elete?"
                       "$$ &Changed $$ &Delete") % prompts, 0):
                     actionbyfile[f] = ('r', None, "prompt delete")
                 elif f in p1:
                     actionbyfile[f] = ('am', None, "prompt keep")
                 else:
                     actionbyfile[f] = ('a', None, "prompt keep")
             elif m == 'dc':
                 f1, f2, fa, move, anc = args
                 flags = p2[f2].flags()
                 if repo.ui.promptchoice(
                     _("other%(o)s changed %(f)s which local%(l)s deleted\n"
                       "use (c)hanged version or leave (d)eleted?"
                       "$$ &Changed $$ &Deleted") % prompts, 0) == 0:
                     actionbyfile[f] = ('g', (flags, False), "prompt recreating")
                 else:
                     del actionbyfile[f]
 
         # Convert to dictionary-of-lists format
         actions = dict((m, []) for m in 'a am f g cd dc r dm dg m e k'.split())
         for f, (m, args, msg) in actionbyfile.iteritems():
             if m not in actions:
                 actions[m] = []
             actions[m].append((f, args, msg))
 
         if not util.fscasesensitive(repo.path):
             # check collision between files only in p2 for clean update
             if (not branchmerge and
                 (force or not wc.dirty(missing=True, branch=False))):
                 _checkcollision(repo, p2.manifest(), None)
             else:
                 _checkcollision(repo, wc.manifest(), actions)
 
         # divergent renames
         for f, fl in sorted(diverge.iteritems()):
             repo.ui.warn(_("note: possible conflict - %s was renamed "
                            "multiple times to:\n") % f)
             for nf in fl:
                 repo.ui.warn(" %s\n" % nf)
 
         # rename and delete
         for f, fl in sorted(renamedelete.iteritems()):
             repo.ui.warn(_("note: possible conflict - %s was deleted "
                            "and renamed to:\n") % f)
             for nf in fl:
                 repo.ui.warn(" %s\n" % nf)
 
         ### apply phase
         if not branchmerge: # just jump to the new rev
             fp1, fp2, xp1, xp2 = fp2, nullid, xp2, ''
         if not partial:
             repo.hook('preupdate', throw=True, parent1=xp1, parent2=xp2)
             # note that we're in the middle of an update
             repo.vfs.write('updatestate', p2.hex())
 
         stats = applyupdates(repo, actions, wc, p2, overwrite, labels=labels)
         wc.flushall()
 
         if not partial:
             with repo.dirstate.parentchange():
                 repo.setparents(fp1, fp2)
                 recordupdates(repo, actions, branchmerge)
                 # update completed, clear state
                 util.unlink(repo.vfs.join('updatestate'))
 
                 if not branchmerge:
                     repo.dirstate.setbranch(p2.branch())
 
     # If we're updating to a location, clean up any stale temporary includes
     # (ex: this happens during hg rebase --abort).
     if not branchmerge:
         sparse.prunetemporaryincludes(repo)
 
     if not partial:
         repo.hook('update', parent1=xp1, parent2=xp2, error=stats[3])
     return stats
 
 def graft(repo, ctx, pctx, labels, keepparent=False):
     """Do a graft-like merge.
 
     This is a merge where the merge ancestor is chosen such that one
     or more changesets are grafted onto the current changeset. In
     addition to the merge, this fixes up the dirstate to include only
     a single parent (if keepparent is False) and tries to duplicate any
     renames/copies appropriately.
 
     ctx - changeset to rebase
     pctx - merge base, usually ctx.p1()
     labels - merge labels eg ['local', 'graft']
     keepparent - keep second parent if any
 
     """
     # If we're grafting a descendant onto an ancestor, be sure to pass
     # mergeancestor=True to update. This does two things: 1) allows the merge if
     # the destination is the same as the parent of the ctx (so we can use graft
     # to copy commits), and 2) informs update that the incoming changes are
     # newer than the destination so it doesn't prompt about "remote changed foo
     # which local deleted".
     mergeancestor = repo.changelog.isancestor(repo['.'].node(), ctx.node())
 
     stats = update(repo, ctx.node(), True, True, pctx.node(),
                    mergeancestor=mergeancestor, labels=labels)
 
     pother = nullid
     parents = ctx.parents()
     if keepparent and len(parents) == 2 and pctx in parents:
         parents.remove(pctx)
         pother = parents[0].node()
 
     with repo.dirstate.parentchange():
         repo.setparents(repo['.'].node(), pother)
         repo.dirstate.write(repo.currenttransaction())
         # fix up dirstate for copies and renames
         copies.duplicatecopies(repo, ctx.rev(), pctx.rev())
     return stats