graftcopies: remove `skip` and `repo` arguments

The `skip` argument was added in 2ba6c9b4e0eb (rebase: fix bug that
caused transitive copy records to disappear (issue4192), 2014-06-07)
in order to fix https://bz.mercurial-scm.org/show_bug.cgi?id=4192. I
ran tests at that commit without the `skiprev` argument and the only
difference I noticed was that `test-rebase-collapse.t` failed
differently, in the call that is now on line 501. Without the
`skiprev` argument, that call would end up creating another commit
because it tried to record an invalid copy. With the previous patch in
this series, such invalid copies are no longer recorded, so it seems
we don't need the `skip` argument anymore.

I also removed the `repo` argument since that also becomes unused with
the removal of the `skip` argument.

Differential Revision: https://phab.mercurial-scm.org/D7860

hgext/fix.py

@@ -1,870 +1,869 b''
 # fix - rewrite file content in changesets and working copy
 #
 # Copyright 2018 Google LLC.
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 """rewrite file content in changesets or working copy (EXPERIMENTAL)
 
 Provides a command that runs configured tools on the contents of modified files,
 writing back any fixes to the working copy or replacing changesets.
 
 Here is an example configuration that causes :hg:`fix` to apply automatic
 formatting fixes to modified lines in C++ code::
 
   [fix]
   clang-format:command=clang-format --assume-filename={rootpath}
   clang-format:linerange=--lines={first}:{last}
   clang-format:pattern=set:**.cpp or **.hpp
 
 The :command suboption forms the first part of the shell command that will be
 used to fix a file. The content of the file is passed on standard input, and the
 fixed file content is expected on standard output. Any output on standard error
 will be displayed as a warning. If the exit status is not zero, the file will
 not be affected. A placeholder warning is displayed if there is a non-zero exit
 status but no standard error output. Some values may be substituted into the
 command::
 
   {rootpath}  The path of the file being fixed, relative to the repo root
   {basename}  The name of the file being fixed, without the directory path
 
 If the :linerange suboption is set, the tool will only be run if there are
 changed lines in a file. The value of this suboption is appended to the shell
 command once for every range of changed lines in the file. Some values may be
 substituted into the command::
 
   {first}   The 1-based line number of the first line in the modified range
   {last}    The 1-based line number of the last line in the modified range
 
 Deleted sections of a file will be ignored by :linerange, because there is no
 corresponding line range in the version being fixed.
 
 By default, tools that set :linerange will only be executed if there is at least
 one changed line range. This is meant to prevent accidents like running a code
 formatter in such a way that it unexpectedly reformats the whole file. If such a
 tool needs to operate on unchanged files, it should set the :skipclean suboption
 to false.
 
 The :pattern suboption determines which files will be passed through each
 configured tool. See :hg:`help patterns` for possible values. However, all
 patterns are relative to the repo root, even if that text says they are relative
 to the current working directory. If there are file arguments to :hg:`fix`, the
 intersection of these patterns is used.
 
 There is also a configurable limit for the maximum size of file that will be
 processed by :hg:`fix`::
 
   [fix]
   maxfilesize = 2MB
 
 Normally, execution of configured tools will continue after a failure (indicated
 by a non-zero exit status). It can also be configured to abort after the first
 such failure, so that no files will be affected if any tool fails. This abort
 will also cause :hg:`fix` to exit with a non-zero status::
 
   [fix]
   failure = abort
 
 When multiple tools are configured to affect a file, they execute in an order
 defined by the :priority suboption. The priority suboption has a default value
 of zero for each tool. Tools are executed in order of descending priority. The
 execution order of tools with equal priority is unspecified. For example, you
 could use the 'sort' and 'head' utilities to keep only the 10 smallest numbers
 in a text file by ensuring that 'sort' runs before 'head'::
 
   [fix]
   sort:command = sort -n
   head:command = head -n 10
   sort:pattern = numbers.txt
   head:pattern = numbers.txt
   sort:priority = 2
   head:priority = 1
 
 To account for changes made by each tool, the line numbers used for incremental
 formatting are recomputed before executing the next tool. So, each tool may see
 different values for the arguments added by the :linerange suboption.
 
 Each fixer tool is allowed to return some metadata in addition to the fixed file
 content. The metadata must be placed before the file content on stdout,
 separated from the file content by a zero byte. The metadata is parsed as a JSON
 value (so, it should be UTF-8 encoded and contain no zero bytes). A fixer tool
 is expected to produce this metadata encoding if and only if the :metadata
 suboption is true::
 
   [fix]
   tool:command = tool --prepend-json-metadata
   tool:metadata = true
 
 The metadata values are passed to hooks, which can be used to print summaries or
 perform other post-fixing work. The supported hooks are::
 
   "postfixfile"
     Run once for each file in each revision where any fixer tools made changes
     to the file content. Provides "$HG_REV" and "$HG_PATH" to identify the file,
     and "$HG_METADATA" with a map of fixer names to metadata values from fixer
     tools that affected the file. Fixer tools that didn't affect the file have a
     valueof None. Only fixer tools that executed are present in the metadata.
 
   "postfix"
     Run once after all files and revisions have been handled. Provides
     "$HG_REPLACEMENTS" with information about what revisions were created and
     made obsolete. Provides a boolean "$HG_WDIRWRITTEN" to indicate whether any
     files in the working copy were updated. Provides a list "$HG_METADATA"
     mapping fixer tool names to lists of metadata values returned from
     executions that modified a file. This aggregates the same metadata
     previously passed to the "postfixfile" hook.
 
 Fixer tools are run the in repository's root directory. This allows them to read
 configuration files from the working copy, or even write to the working copy.
 The working copy is not updated to match the revision being fixed. In fact,
 several revisions may be fixed in parallel. Writes to the working copy are not
 amended into the revision being fixed; fixer tools should always write fixed
 file content back to stdout as documented above.
 """
 
 from __future__ import absolute_import
 
 import collections
 import itertools
 import os
 import re
 import subprocess
 
 from mercurial.i18n import _
 from mercurial.node import nullrev
 from mercurial.node import wdirrev
 
 from mercurial.utils import procutil
 
 from mercurial import (
     cmdutil,
     context,
     copies,
     error,
     match as matchmod,
     mdiff,
     merge,
     pycompat,
     registrar,
     rewriteutil,
     scmutil,
     util,
     worker,
 )
 
 # Note for extension authors: ONLY specify testedwith = 'ships-with-hg-core' for
 # extensions which SHIP WITH MERCURIAL. Non-mainline extensions should
 # be specifying the version(s) of Mercurial they are tested with, or
 # leave the attribute unspecified.
 testedwith = b'ships-with-hg-core'
 
 cmdtable = {}
 command = registrar.command(cmdtable)
 
 configtable = {}
 configitem = registrar.configitem(configtable)
 
 # Register the suboptions allowed for each configured fixer, and default values.
 FIXER_ATTRS = {
     b'command': None,
     b'linerange': None,
     b'pattern': None,
     b'priority': 0,
     b'metadata': False,
     b'skipclean': True,
     b'enabled': True,
 }
 
 for key, default in FIXER_ATTRS.items():
     configitem(b'fix', b'.*:%s$' % key, default=default, generic=True)
 
 # A good default size allows most source code files to be fixed, but avoids
 # letting fixer tools choke on huge inputs, which could be surprising to the
 # user.
 configitem(b'fix', b'maxfilesize', default=b'2MB')
 
 # Allow fix commands to exit non-zero if an executed fixer tool exits non-zero.
 # This helps users do shell scripts that stop when a fixer tool signals a
 # problem.
 configitem(b'fix', b'failure', default=b'continue')
 
 
 def checktoolfailureaction(ui, message, hint=None):
     """Abort with 'message' if fix.failure=abort"""
     action = ui.config(b'fix', b'failure')
     if action not in (b'continue', b'abort'):
         raise error.Abort(
             _(b'unknown fix.failure action: %s') % (action,),
             hint=_(b'use "continue" or "abort"'),
         )
     if action == b'abort':
         raise error.Abort(message, hint=hint)
 
 
 allopt = (b'', b'all', False, _(b'fix all non-public non-obsolete revisions'))
 baseopt = (
     b'',
     b'base',
     [],
     _(
         b'revisions to diff against (overrides automatic '
         b'selection, and applies to every revision being '
         b'fixed)'
     ),
     _(b'REV'),
 )
 revopt = (b'r', b'rev', [], _(b'revisions to fix'), _(b'REV'))
 wdiropt = (b'w', b'working-dir', False, _(b'fix the working directory'))
 wholeopt = (b'', b'whole', False, _(b'always fix every line of a file'))
 usage = _(b'[OPTION]... [FILE]...')
 
 
 @command(
     b'fix',
     [allopt, baseopt, revopt, wdiropt, wholeopt],
     usage,
     helpcategory=command.CATEGORY_FILE_CONTENTS,
 )
 def fix(ui, repo, *pats, **opts):
     """rewrite file content in changesets or working directory
 
     Runs any configured tools to fix the content of files. Only affects files
     with changes, unless file arguments are provided. Only affects changed lines
     of files, unless the --whole flag is used. Some tools may always affect the
     whole file regardless of --whole.
 
     If revisions are specified with --rev, those revisions will be checked, and
     they may be replaced with new revisions that have fixed file content.  It is
     desirable to specify all descendants of each specified revision, so that the
     fixes propagate to the descendants. If all descendants are fixed at the same
     time, no merging, rebasing, or evolution will be required.
 
     If --working-dir is used, files with uncommitted changes in the working copy
     will be fixed. If the checked-out revision is also fixed, the working
     directory will update to the replacement revision.
 
     When determining what lines of each file to fix at each revision, the whole
     set of revisions being fixed is considered, so that fixes to earlier
     revisions are not forgotten in later ones. The --base flag can be used to
     override this default behavior, though it is not usually desirable to do so.
     """
     opts = pycompat.byteskwargs(opts)
     cmdutil.check_at_most_one_arg(opts, b'all', b'rev')
     if opts[b'all']:
         opts[b'rev'] = [b'not public() and not obsolete()']
         opts[b'working_dir'] = True
     with repo.wlock(), repo.lock(), repo.transaction(b'fix'):
         revstofix = getrevstofix(ui, repo, opts)
         basectxs = getbasectxs(repo, opts, revstofix)
         workqueue, numitems = getworkqueue(
             ui, repo, pats, opts, revstofix, basectxs
         )
         fixers = getfixers(ui)
 
         # There are no data dependencies between the workers fixing each file
         # revision, so we can use all available parallelism.
         def getfixes(items):
             for rev, path in items:
                 ctx = repo[rev]
                 olddata = ctx[path].data()
                 metadata, newdata = fixfile(
                     ui, repo, opts, fixers, ctx, path, basectxs[rev]
                 )
                 # Don't waste memory/time passing unchanged content back, but
                 # produce one result per item either way.
                 yield (
                     rev,
                     path,
                     metadata,
                     newdata if newdata != olddata else None,
                 )
 
         results = worker.worker(
             ui, 1.0, getfixes, tuple(), workqueue, threadsafe=False
         )
 
         # We have to hold on to the data for each successor revision in memory
         # until all its parents are committed. We ensure this by committing and
         # freeing memory for the revisions in some topological order. This
         # leaves a little bit of memory efficiency on the table, but also makes
         # the tests deterministic. It might also be considered a feature since
         # it makes the results more easily reproducible.
         filedata = collections.defaultdict(dict)
         aggregatemetadata = collections.defaultdict(list)
         replacements = {}
         wdirwritten = False
         commitorder = sorted(revstofix, reverse=True)
         with ui.makeprogress(
             topic=_(b'fixing'), unit=_(b'files'), total=sum(numitems.values())
         ) as progress:
             for rev, path, filerevmetadata, newdata in results:
                 progress.increment(item=path)
                 for fixername, fixermetadata in filerevmetadata.items():
                     aggregatemetadata[fixername].append(fixermetadata)
                 if newdata is not None:
                     filedata[rev][path] = newdata
                     hookargs = {
                         b'rev': rev,
                         b'path': path,
                         b'metadata': filerevmetadata,
                     }
                     repo.hook(
                         b'postfixfile',
                         throw=False,
                         **pycompat.strkwargs(hookargs)
                     )
                 numitems[rev] -= 1
                 # Apply the fixes for this and any other revisions that are
                 # ready and sitting at the front of the queue. Using a loop here
                 # prevents the queue from being blocked by the first revision to
                 # be ready out of order.
                 while commitorder and not numitems[commitorder[-1]]:
                     rev = commitorder.pop()
                     ctx = repo[rev]
                     if rev == wdirrev:
                         writeworkingdir(repo, ctx, filedata[rev], replacements)
                         wdirwritten = bool(filedata[rev])
                     else:
                         replacerev(ui, repo, ctx, filedata[rev], replacements)
                     del filedata[rev]
 
         cleanup(repo, replacements, wdirwritten)
         hookargs = {
             b'replacements': replacements,
             b'wdirwritten': wdirwritten,
             b'metadata': aggregatemetadata,
         }
         repo.hook(b'postfix', throw=True, **pycompat.strkwargs(hookargs))
 
 
 def cleanup(repo, replacements, wdirwritten):
     """Calls scmutil.cleanupnodes() with the given replacements.
 
     "replacements" is a dict from nodeid to nodeid, with one key and one value
     for every revision that was affected by fixing. This is slightly different
     from cleanupnodes().
 
     "wdirwritten" is a bool which tells whether the working copy was affected by
     fixing, since it has no entry in "replacements".
 
     Useful as a hook point for extending "hg fix" with output summarizing the
     effects of the command, though we choose not to output anything here.
     """
     replacements = {
         prec: [succ] for prec, succ in pycompat.iteritems(replacements)
     }
     scmutil.cleanupnodes(repo, replacements, b'fix', fixphase=True)
 
 
 def getworkqueue(ui, repo, pats, opts, revstofix, basectxs):
     """"Constructs the list of files to be fixed at specific revisions
 
     It is up to the caller how to consume the work items, and the only
     dependence between them is that replacement revisions must be committed in
     topological order. Each work item represents a file in the working copy or
     in some revision that should be fixed and written back to the working copy
     or into a replacement revision.
 
     Work items for the same revision are grouped together, so that a worker
     pool starting with the first N items in parallel is likely to finish the
     first revision's work before other revisions. This can allow us to write
     the result to disk and reduce memory footprint. At time of writing, the
     partition strategy in worker.py seems favorable to this. We also sort the
     items by ascending revision number to match the order in which we commit
     the fixes later.
     """
     workqueue = []
     numitems = collections.defaultdict(int)
     maxfilesize = ui.configbytes(b'fix', b'maxfilesize')
     for rev in sorted(revstofix):
         fixctx = repo[rev]
         match = scmutil.match(fixctx, pats, opts)
         for path in sorted(
             pathstofix(ui, repo, pats, opts, match, basectxs[rev], fixctx)
         ):
             fctx = fixctx[path]
             if fctx.islink():
                 continue
             if fctx.size() > maxfilesize:
                 ui.warn(
                     _(b'ignoring file larger than %s: %s\n')
                     % (util.bytecount(maxfilesize), path)
                 )
                 continue
             workqueue.append((rev, path))
             numitems[rev] += 1
     return workqueue, numitems
 
 
 def getrevstofix(ui, repo, opts):
     """Returns the set of revision numbers that should be fixed"""
     revs = set(scmutil.revrange(repo, opts[b'rev']))
     for rev in revs:
         checkfixablectx(ui, repo, repo[rev])
     if revs:
         cmdutil.checkunfinished(repo)
         rewriteutil.precheck(repo, revs, b'fix')
     if opts.get(b'working_dir'):
         revs.add(wdirrev)
         if list(merge.mergestate.read(repo).unresolved()):
             raise error.Abort(b'unresolved conflicts', hint=b"use 'hg resolve'")
     if not revs:
         raise error.Abort(
             b'no changesets specified', hint=b'use --rev or --working-dir'
         )
     return revs
 
 
 def checkfixablectx(ui, repo, ctx):
     """Aborts if the revision shouldn't be replaced with a fixed one."""
     if ctx.obsolete():
         # It would be better to actually check if the revision has a successor.
         allowdivergence = ui.configbool(
             b'experimental', b'evolution.allowdivergence'
         )
         if not allowdivergence:
             raise error.Abort(
                 b'fixing obsolete revision could cause divergence'
             )
 
 
 def pathstofix(ui, repo, pats, opts, match, basectxs, fixctx):
     """Returns the set of files that should be fixed in a context
 
     The result depends on the base contexts; we include any file that has
     changed relative to any of the base contexts. Base contexts should be
     ancestors of the context being fixed.
     """
     files = set()
     for basectx in basectxs:
         stat = basectx.status(
             fixctx, match=match, listclean=bool(pats), listunknown=bool(pats)
         )
         files.update(
             set(
                 itertools.chain(
                     stat.added, stat.modified, stat.clean, stat.unknown
                 )
             )
         )
     return files
 
 
 def lineranges(opts, path, basectxs, fixctx, content2):
     """Returns the set of line ranges that should be fixed in a file
 
     Of the form [(10, 20), (30, 40)].
 
     This depends on the given base contexts; we must consider lines that have
     changed versus any of the base contexts, and whether the file has been
     renamed versus any of them.
 
     Another way to understand this is that we exclude line ranges that are
     common to the file in all base contexts.
     """
     if opts.get(b'whole'):
         # Return a range containing all lines. Rely on the diff implementation's
         # idea of how many lines are in the file, instead of reimplementing it.
         return difflineranges(b'', content2)
 
     rangeslist = []
     for basectx in basectxs:
         basepath = copies.pathcopies(basectx, fixctx).get(path, path)
         if basepath in basectx:
             content1 = basectx[basepath].data()
         else:
             content1 = b''
         rangeslist.extend(difflineranges(content1, content2))
     return unionranges(rangeslist)
 
 
 def unionranges(rangeslist):
     """Return the union of some closed intervals
 
     >>> unionranges([])
     []
     >>> unionranges([(1, 100)])
     [(1, 100)]
     >>> unionranges([(1, 100), (1, 100)])
     [(1, 100)]
     >>> unionranges([(1, 100), (2, 100)])
     [(1, 100)]
     >>> unionranges([(1, 99), (1, 100)])
     [(1, 100)]
     >>> unionranges([(1, 100), (40, 60)])
     [(1, 100)]
     >>> unionranges([(1, 49), (50, 100)])
     [(1, 100)]
     >>> unionranges([(1, 48), (50, 100)])
     [(1, 48), (50, 100)]
     >>> unionranges([(1, 2), (3, 4), (5, 6)])
     [(1, 6)]
     """
     rangeslist = sorted(set(rangeslist))
     unioned = []
     if rangeslist:
         unioned, rangeslist = [rangeslist[0]], rangeslist[1:]
     for a, b in rangeslist:
         c, d = unioned[-1]
         if a > d + 1:
             unioned.append((a, b))
         else:
             unioned[-1] = (c, max(b, d))
     return unioned
 
 
 def difflineranges(content1, content2):
     """Return list of line number ranges in content2 that differ from content1.
 
     Line numbers are 1-based. The numbers are the first and last line contained
     in the range. Single-line ranges have the same line number for the first and
     last line. Excludes any empty ranges that result from lines that are only
     present in content1. Relies on mdiff's idea of where the line endings are in
     the string.
 
     >>> from mercurial import pycompat
     >>> lines = lambda s: b'\\n'.join([c for c in pycompat.iterbytestr(s)])
     >>> difflineranges2 = lambda a, b: difflineranges(lines(a), lines(b))
     >>> difflineranges2(b'', b'')
     []
     >>> difflineranges2(b'a', b'')
     []
     >>> difflineranges2(b'', b'A')
     [(1, 1)]
     >>> difflineranges2(b'a', b'a')
     []
     >>> difflineranges2(b'a', b'A')
     [(1, 1)]
     >>> difflineranges2(b'ab', b'')
     []
     >>> difflineranges2(b'', b'AB')
     [(1, 2)]
     >>> difflineranges2(b'abc', b'ac')
     []
     >>> difflineranges2(b'ab', b'aCb')
     [(2, 2)]
     >>> difflineranges2(b'abc', b'aBc')
     [(2, 2)]
     >>> difflineranges2(b'ab', b'AB')
     [(1, 2)]
     >>> difflineranges2(b'abcde', b'aBcDe')
     [(2, 2), (4, 4)]
     >>> difflineranges2(b'abcde', b'aBCDe')
     [(2, 4)]
     """
     ranges = []
     for lines, kind in mdiff.allblocks(content1, content2):
         firstline, lastline = lines[2:4]
         if kind == b'!' and firstline != lastline:
             ranges.append((firstline + 1, lastline))
     return ranges
 
 
 def getbasectxs(repo, opts, revstofix):
     """Returns a map of the base contexts for each revision
 
     The base contexts determine which lines are considered modified when we
     attempt to fix just the modified lines in a file. It also determines which
     files we attempt to fix, so it is important to compute this even when
     --whole is used.
     """
     # The --base flag overrides the usual logic, and we give every revision
     # exactly the set of baserevs that the user specified.
     if opts.get(b'base'):
         baserevs = set(scmutil.revrange(repo, opts.get(b'base')))
         if not baserevs:
             baserevs = {nullrev}
         basectxs = {repo[rev] for rev in baserevs}
         return {rev: basectxs for rev in revstofix}
 
     # Proceed in topological order so that we can easily determine each
     # revision's baserevs by looking at its parents and their baserevs.
     basectxs = collections.defaultdict(set)
     for rev in sorted(revstofix):
         ctx = repo[rev]
         for pctx in ctx.parents():
             if pctx.rev() in basectxs:
                 basectxs[rev].update(basectxs[pctx.rev()])
             else:
                 basectxs[rev].add(pctx)
     return basectxs
 
 
 def fixfile(ui, repo, opts, fixers, fixctx, path, basectxs):
     """Run any configured fixers that should affect the file in this context
 
     Returns the file content that results from applying the fixers in some order
     starting with the file's content in the fixctx. Fixers that support line
     ranges will affect lines that have changed relative to any of the basectxs
     (i.e. they will only avoid lines that are common to all basectxs).
 
     A fixer tool's stdout will become the file's new content if and only if it
     exits with code zero. The fixer tool's working directory is the repository's
     root.
     """
     metadata = {}
     newdata = fixctx[path].data()
     for fixername, fixer in pycompat.iteritems(fixers):
         if fixer.affects(opts, fixctx, path):
             ranges = lineranges(opts, path, basectxs, fixctx, newdata)
             command = fixer.command(ui, path, ranges)
             if command is None:
                 continue
             ui.debug(b'subprocess: %s\n' % (command,))
             proc = subprocess.Popen(
                 procutil.tonativestr(command),
                 shell=True,
                 cwd=procutil.tonativestr(repo.root),
                 stdin=subprocess.PIPE,
                 stdout=subprocess.PIPE,
                 stderr=subprocess.PIPE,
             )
             stdout, stderr = proc.communicate(newdata)
             if stderr:
                 showstderr(ui, fixctx.rev(), fixername, stderr)
             newerdata = stdout
             if fixer.shouldoutputmetadata():
                 try:
                     metadatajson, newerdata = stdout.split(b'\0', 1)
                     metadata[fixername] = pycompat.json_loads(metadatajson)
                 except ValueError:
                     ui.warn(
                         _(b'ignored invalid output from fixer tool: %s\n')
                         % (fixername,)
                     )
                     continue
             else:
                 metadata[fixername] = None
             if proc.returncode == 0:
                 newdata = newerdata
             else:
                 if not stderr:
                     message = _(b'exited with status %d\n') % (proc.returncode,)
                     showstderr(ui, fixctx.rev(), fixername, message)
                 checktoolfailureaction(
                     ui,
                     _(b'no fixes will be applied'),
                     hint=_(
                         b'use --config fix.failure=continue to apply any '
                         b'successful fixes anyway'
                     ),
                 )
     return metadata, newdata
 
 
 def showstderr(ui, rev, fixername, stderr):
     """Writes the lines of the stderr string as warnings on the ui
 
     Uses the revision number and fixername to give more context to each line of
     the error message. Doesn't include file names, since those take up a lot of
     space and would tend to be included in the error message if they were
     relevant.
     """
     for line in re.split(b'[\r\n]+', stderr):
         if line:
             ui.warn(b'[')
             if rev is None:
                 ui.warn(_(b'wdir'), label=b'evolve.rev')
             else:
                 ui.warn(b'%d' % rev, label=b'evolve.rev')
             ui.warn(b'] %s: %s\n' % (fixername, line))
 
 
 def writeworkingdir(repo, ctx, filedata, replacements):
     """Write new content to the working copy and check out the new p1 if any
 
     We check out a new revision if and only if we fixed something in both the
     working directory and its parent revision. This avoids the need for a full
     update/merge, and means that the working directory simply isn't affected
     unless the --working-dir flag is given.
 
     Directly updates the dirstate for the affected files.
     """
     for path, data in pycompat.iteritems(filedata):
         fctx = ctx[path]
         fctx.write(data, fctx.flags())
         if repo.dirstate[path] == b'n':
             repo.dirstate.normallookup(path)
 
     oldparentnodes = repo.dirstate.parents()
     newparentnodes = [replacements.get(n, n) for n in oldparentnodes]
     if newparentnodes != oldparentnodes:
         repo.setparents(*newparentnodes)
 
 
 def replacerev(ui, repo, ctx, filedata, replacements):
     """Commit a new revision like the given one, but with file content changes
 
     "ctx" is the original revision to be replaced by a modified one.
 
     "filedata" is a dict that maps paths to their new file content. All other
     paths will be recreated from the original revision without changes.
     "filedata" may contain paths that didn't exist in the original revision;
     they will be added.
 
     "replacements" is a dict that maps a single node to a single node, and it is
     updated to indicate the original revision is replaced by the newly created
     one. No entry is added if the replacement's node already exists.
 
     The new revision has the same parents as the old one, unless those parents
     have already been replaced, in which case those replacements are the parents
     of this new revision. Thus, if revisions are replaced in topological order,
     there is no need to rebase them into the original topology later.
     """
 
     p1rev, p2rev = repo.changelog.parentrevs(ctx.rev())
     p1ctx, p2ctx = repo[p1rev], repo[p2rev]
     newp1node = replacements.get(p1ctx.node(), p1ctx.node())
     newp2node = replacements.get(p2ctx.node(), p2ctx.node())
 
     # We don't want to create a revision that has no changes from the original,
     # but we should if the original revision's parent has been replaced.
     # Otherwise, we would produce an orphan that needs no actual human
     # intervention to evolve. We can't rely on commit() to avoid creating the
     # un-needed revision because the extra field added below produces a new hash
     # regardless of file content changes.
     if (
         not filedata
         and p1ctx.node() not in replacements
         and p2ctx.node() not in replacements
     ):
         return
 
     extra = ctx.extra().copy()
     extra[b'fix_source'] = ctx.hex()
 
     wctx = context.overlayworkingctx(repo)
-    newp1ctx = repo[newp1node]
-    wctx.setbase(newp1ctx)
+    wctx.setbase(repo[newp1node])
     merge.update(
         repo,
         ctx.rev(),
         branchmerge=False,
         force=True,
         ancestor=p1rev,
         mergeancestor=False,
         wc=wctx,
     )
-    copies.graftcopies(repo, wctx, ctx, ctx.p1(), skip=newp1ctx)
+    copies.graftcopies(wctx, ctx, ctx.p1())
 
     for path in filedata.keys():
         fctx = ctx[path]
         copysource = fctx.copysource()
         wctx.write(path, filedata[path], flags=fctx.flags())
         if copysource:
             wctx.markcopied(path, copysource)
 
     memctx = wctx.tomemctx(
         text=ctx.description(),
         branch=ctx.branch(),
         extra=extra,
         date=ctx.date(),
         parents=(newp1node, newp2node),
         user=ctx.user(),
     )
 
     sucnode = memctx.commit()
     prenode = ctx.node()
     if prenode == sucnode:
         ui.debug(b'node %s already existed\n' % (ctx.hex()))
     else:
         replacements[ctx.node()] = sucnode
 
 
 def getfixers(ui):
     """Returns a map of configured fixer tools indexed by their names
 
     Each value is a Fixer object with methods that implement the behavior of the
     fixer's config suboptions. Does not validate the config values.
     """
     fixers = {}
     for name in fixernames(ui):
         enabled = ui.configbool(b'fix', name + b':enabled')
         command = ui.config(b'fix', name + b':command')
         pattern = ui.config(b'fix', name + b':pattern')
         linerange = ui.config(b'fix', name + b':linerange')
         priority = ui.configint(b'fix', name + b':priority')
         metadata = ui.configbool(b'fix', name + b':metadata')
         skipclean = ui.configbool(b'fix', name + b':skipclean')
         # Don't use a fixer if it has no pattern configured. It would be
         # dangerous to let it affect all files. It would be pointless to let it
         # affect no files. There is no reasonable subset of files to use as the
         # default.
         if command is None:
             ui.warn(
                 _(b'fixer tool has no command configuration: %s\n') % (name,)
             )
         elif pattern is None:
             ui.warn(
                 _(b'fixer tool has no pattern configuration: %s\n') % (name,)
             )
         elif not enabled:
             ui.debug(b'ignoring disabled fixer tool: %s\n' % (name,))
         else:
             fixers[name] = Fixer(
                 command, pattern, linerange, priority, metadata, skipclean
             )
     return collections.OrderedDict(
         sorted(fixers.items(), key=lambda item: item[1]._priority, reverse=True)
     )
 
 
 def fixernames(ui):
     """Returns the names of [fix] config options that have suboptions"""
     names = set()
     for k, v in ui.configitems(b'fix'):
         if b':' in k:
             names.add(k.split(b':', 1)[0])
     return names
 
 
 class Fixer(object):
     """Wraps the raw config values for a fixer with methods"""
 
     def __init__(
         self, command, pattern, linerange, priority, metadata, skipclean
     ):
         self._command = command
         self._pattern = pattern
         self._linerange = linerange
         self._priority = priority
         self._metadata = metadata
         self._skipclean = skipclean
 
     def affects(self, opts, fixctx, path):
         """Should this fixer run on the file at the given path and context?"""
         repo = fixctx.repo()
         matcher = matchmod.match(
             repo.root, repo.root, [self._pattern], ctx=fixctx
         )
         return matcher(path)
 
     def shouldoutputmetadata(self):
         """Should the stdout of this fixer start with JSON and a null byte?"""
         return self._metadata
 
     def command(self, ui, path, ranges):
         """A shell command to use to invoke this fixer on the given file/lines
 
         May return None if there is no appropriate command to run for the given
         parameters.
         """
         expand = cmdutil.rendercommandtemplate
         parts = [
             expand(
                 ui,
                 self._command,
                 {b'rootpath': path, b'basename': os.path.basename(path)},
             )
         ]
         if self._linerange:
             if self._skipclean and not ranges:
                 # No line ranges to fix, so don't run the fixer.
                 return None
             for first, last in ranges:
                 parts.append(
                     expand(
                         ui, self._linerange, {b'first': first, b'last': last}
                     )
                 )
         return b' '.join(parts)

hgext/rebase.py

@@ -1,2265 +1,2262 b''
 # rebase.py - rebasing feature for mercurial
 #
 # Copyright 2008 Stefano Tortarolo <stefano.tortarolo at gmail dot com>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 '''command to move sets of revisions to a different ancestor
 
 This extension lets you rebase changesets in an existing Mercurial
 repository.
 
 For more information:
 https://mercurial-scm.org/wiki/RebaseExtension
 '''
 
 from __future__ import absolute_import
 
 import errno
 import os
 
 from mercurial.i18n import _
 from mercurial.node import (
     nullrev,
     short,
 )
 from mercurial.pycompat import open
 from mercurial import (
     bookmarks,
     cmdutil,
     commands,
     copies,
     destutil,
     dirstateguard,
     error,
     extensions,
     hg,
     merge as mergemod,
     mergeutil,
     obsolete,
     obsutil,
     patch,
     phases,
     pycompat,
     registrar,
     repair,
     revset,
     revsetlang,
     rewriteutil,
     scmutil,
     smartset,
     state as statemod,
     util,
 )
 
 # The following constants are used throughout the rebase module. The ordering of
 # their values must be maintained.
 
 # Indicates that a revision needs to be rebased
 revtodo = -1
 revtodostr = b'-1'
 
 # legacy revstates no longer needed in current code
 # -2: nullmerge, -3: revignored, -4: revprecursor, -5: revpruned
 legacystates = {b'-2', b'-3', b'-4', b'-5'}
 
 cmdtable = {}
 command = registrar.command(cmdtable)
 # Note for extension authors: ONLY specify testedwith = 'ships-with-hg-core' for
 # extensions which SHIP WITH MERCURIAL. Non-mainline extensions should
 # be specifying the version(s) of Mercurial they are tested with, or
 # leave the attribute unspecified.
 testedwith = b'ships-with-hg-core'
 
 
 def _nothingtorebase():
     return 1
 
 
 def _savegraft(ctx, extra):
     s = ctx.extra().get(b'source', None)
     if s is not None:
         extra[b'source'] = s
     s = ctx.extra().get(b'intermediate-source', None)
     if s is not None:
         extra[b'intermediate-source'] = s
 
 
 def _savebranch(ctx, extra):
     extra[b'branch'] = ctx.branch()
 
 
 def _destrebase(repo, sourceset, destspace=None):
     """small wrapper around destmerge to pass the right extra args
 
     Please wrap destutil.destmerge instead."""
     return destutil.destmerge(
         repo,
         action=b'rebase',
         sourceset=sourceset,
         onheadcheck=False,
         destspace=destspace,
     )
 
 
 revsetpredicate = registrar.revsetpredicate()
 
 
 @revsetpredicate(b'_destrebase')
 def _revsetdestrebase(repo, subset, x):
     # ``_rebasedefaultdest()``
 
     # default destination for rebase.
     # # XXX: Currently private because I expect the signature to change.
     # # XXX: - bailing out in case of ambiguity vs returning all data.
     # i18n: "_rebasedefaultdest" is a keyword
     sourceset = None
     if x is not None:
         sourceset = revset.getset(repo, smartset.fullreposet(repo), x)
     return subset & smartset.baseset([_destrebase(repo, sourceset)])
 
 
 @revsetpredicate(b'_destautoorphanrebase')
 def _revsetdestautoorphanrebase(repo, subset, x):
     # ``_destautoorphanrebase()``
 
     # automatic rebase destination for a single orphan revision.
     unfi = repo.unfiltered()
     obsoleted = unfi.revs(b'obsolete()')
 
     src = revset.getset(repo, subset, x).first()
 
     # Empty src or already obsoleted - Do not return a destination
     if not src or src in obsoleted:
         return smartset.baseset()
     dests = destutil.orphanpossibledestination(repo, src)
     if len(dests) > 1:
         raise error.Abort(
             _(b"ambiguous automatic rebase: %r could end up on any of %r")
             % (src, dests)
         )
     # We have zero or one destination, so we can just return here.
     return smartset.baseset(dests)
 
 
 def _ctxdesc(ctx):
     """short description for a context"""
     desc = b'%d:%s "%s"' % (
         ctx.rev(),
         ctx,
         ctx.description().split(b'\n', 1)[0],
     )
     repo = ctx.repo()
     names = []
     for nsname, ns in pycompat.iteritems(repo.names):
         if nsname == b'branches':
             continue
         names.extend(ns.names(repo, ctx.node()))
     if names:
         desc += b' (%s)' % b' '.join(names)
     return desc
 
 
 class rebaseruntime(object):
     """This class is a container for rebase runtime state"""
 
     def __init__(self, repo, ui, inmemory=False, opts=None):
         if opts is None:
             opts = {}
 
         # prepared: whether we have rebasestate prepared or not. Currently it
         # decides whether "self.repo" is unfiltered or not.
         # The rebasestate has explicit hash to hash instructions not depending
         # on visibility. If rebasestate exists (in-memory or on-disk), use
         # unfiltered repo to avoid visibility issues.
         # Before knowing rebasestate (i.e. when starting a new rebase (not
         # --continue or --abort)), the original repo should be used so
         # visibility-dependent revsets are correct.
         self.prepared = False
         self._repo = repo
 
         self.ui = ui
         self.opts = opts
         self.originalwd = None
         self.external = nullrev
         # Mapping between the old revision id and either what is the new rebased
         # revision or what needs to be done with the old revision. The state
         # dict will be what contains most of the rebase progress state.
         self.state = {}
         self.activebookmark = None
         self.destmap = {}
         self.skipped = set()
 
         self.collapsef = opts.get(b'collapse', False)
         self.collapsemsg = cmdutil.logmessage(ui, opts)
         self.date = opts.get(b'date', None)
 
         e = opts.get(b'extrafn')  # internal, used by e.g. hgsubversion
         self.extrafns = [_savegraft]
         if e:
             self.extrafns = [e]
 
         self.backupf = ui.configbool(b'rewrite', b'backup-bundle')
         self.keepf = opts.get(b'keep', False)
         self.keepbranchesf = opts.get(b'keepbranches', False)
         self.obsoletenotrebased = {}
         self.obsoletewithoutsuccessorindestination = set()
         self.inmemory = inmemory
         self.stateobj = statemod.cmdstate(repo, b'rebasestate')
 
     @property
     def repo(self):
         if self.prepared:
             return self._repo.unfiltered()
         else:
             return self._repo
 
     def storestatus(self, tr=None):
         """Store the current status to allow recovery"""
         if tr:
             tr.addfilegenerator(
                 b'rebasestate',
                 (b'rebasestate',),
                 self._writestatus,
                 location=b'plain',
             )
         else:
             with self.repo.vfs(b"rebasestate", b"w") as f:
                 self._writestatus(f)
 
     def _writestatus(self, f):
         repo = self.repo
         assert repo.filtername is None
         f.write(repo[self.originalwd].hex() + b'\n')
         # was "dest". we now write dest per src root below.
         f.write(b'\n')
         f.write(repo[self.external].hex() + b'\n')
         f.write(b'%d\n' % int(self.collapsef))
         f.write(b'%d\n' % int(self.keepf))
         f.write(b'%d\n' % int(self.keepbranchesf))
         f.write(b'%s\n' % (self.activebookmark or b''))
         destmap = self.destmap
         for d, v in pycompat.iteritems(self.state):
             oldrev = repo[d].hex()
             if v >= 0:
                 newrev = repo[v].hex()
             else:
                 newrev = b"%d" % v
             destnode = repo[destmap[d]].hex()
             f.write(b"%s:%s:%s\n" % (oldrev, newrev, destnode))
         repo.ui.debug(b'rebase status stored\n')
 
     def restorestatus(self):
         """Restore a previously stored status"""
         if not self.stateobj.exists():
             cmdutil.wrongtooltocontinue(self.repo, _(b'rebase'))
 
         data = self._read()
         self.repo.ui.debug(b'rebase status resumed\n')
 
         self.originalwd = data[b'originalwd']
         self.destmap = data[b'destmap']
         self.state = data[b'state']
         self.skipped = data[b'skipped']
         self.collapsef = data[b'collapse']
         self.keepf = data[b'keep']
         self.keepbranchesf = data[b'keepbranches']
         self.external = data[b'external']
         self.activebookmark = data[b'activebookmark']
 
     def _read(self):
         self.prepared = True
         repo = self.repo
         assert repo.filtername is None
         data = {
             b'keepbranches': None,
             b'collapse': None,
             b'activebookmark': None,
             b'external': nullrev,
             b'keep': None,
             b'originalwd': None,
         }
         legacydest = None
         state = {}
         destmap = {}
 
         if True:
             f = repo.vfs(b"rebasestate")
             for i, l in enumerate(f.read().splitlines()):
                 if i == 0:
                     data[b'originalwd'] = repo[l].rev()
                 elif i == 1:
                     # this line should be empty in newer version. but legacy
                     # clients may still use it
                     if l:
                         legacydest = repo[l].rev()
                 elif i == 2:
                     data[b'external'] = repo[l].rev()
                 elif i == 3:
                     data[b'collapse'] = bool(int(l))
                 elif i == 4:
                     data[b'keep'] = bool(int(l))
                 elif i == 5:
                     data[b'keepbranches'] = bool(int(l))
                 elif i == 6 and not (len(l) == 81 and b':' in l):
                     # line 6 is a recent addition, so for backwards
                     # compatibility check that the line doesn't look like the
                     # oldrev:newrev lines
                     data[b'activebookmark'] = l
                 else:
                     args = l.split(b':')
                     oldrev = repo[args[0]].rev()
                     newrev = args[1]
                     if newrev in legacystates:
                         continue
                     if len(args) > 2:
                         destrev = repo[args[2]].rev()
                     else:
                         destrev = legacydest
                     destmap[oldrev] = destrev
                     if newrev == revtodostr:
                         state[oldrev] = revtodo
                         # Legacy compat special case
                     else:
                         state[oldrev] = repo[newrev].rev()
 
         if data[b'keepbranches'] is None:
             raise error.Abort(_(b'.hg/rebasestate is incomplete'))
 
         data[b'destmap'] = destmap
         data[b'state'] = state
         skipped = set()
         # recompute the set of skipped revs
         if not data[b'collapse']:
             seen = set(destmap.values())
             for old, new in sorted(state.items()):
                 if new != revtodo and new in seen:
                     skipped.add(old)
                 seen.add(new)
         data[b'skipped'] = skipped
         repo.ui.debug(
             b'computed skipped revs: %s\n'
             % (b' '.join(b'%d' % r for r in sorted(skipped)) or b'')
         )
 
         return data
 
     def _handleskippingobsolete(self, obsoleterevs, destmap):
         """Compute structures necessary for skipping obsolete revisions
 
         obsoleterevs:   iterable of all obsolete revisions in rebaseset
         destmap:        {srcrev: destrev} destination revisions
         """
         self.obsoletenotrebased = {}
         if not self.ui.configbool(b'experimental', b'rebaseskipobsolete'):
             return
         obsoleteset = set(obsoleterevs)
         (
             self.obsoletenotrebased,
             self.obsoletewithoutsuccessorindestination,
             obsoleteextinctsuccessors,
         ) = _computeobsoletenotrebased(self.repo, obsoleteset, destmap)
         skippedset = set(self.obsoletenotrebased)
         skippedset.update(self.obsoletewithoutsuccessorindestination)
         skippedset.update(obsoleteextinctsuccessors)
         _checkobsrebase(self.repo, self.ui, obsoleteset, skippedset)
 
     def _prepareabortorcontinue(self, isabort, backup=True, suppwarns=False):
         try:
             self.restorestatus()
             self.collapsemsg = restorecollapsemsg(self.repo, isabort)
         except error.RepoLookupError:
             if isabort:
                 clearstatus(self.repo)
                 clearcollapsemsg(self.repo)
                 self.repo.ui.warn(
                     _(
                         b'rebase aborted (no revision is removed,'
                         b' only broken state is cleared)\n'
                     )
                 )
                 return 0
             else:
                 msg = _(b'cannot continue inconsistent rebase')
                 hint = _(b'use "hg rebase --abort" to clear broken state')
                 raise error.Abort(msg, hint=hint)
 
         if isabort:
             backup = backup and self.backupf
             return self._abort(backup=backup, suppwarns=suppwarns)
 
     def _preparenewrebase(self, destmap):
         if not destmap:
             return _nothingtorebase()
 
         rebaseset = destmap.keys()
         if not self.keepf:
             try:
                 rewriteutil.precheck(self.repo, rebaseset, action=b'rebase')
             except error.Abort as e:
                 if e.hint is None:
                     e.hint = _(b'use --keep to keep original changesets')
                 raise e
 
         result = buildstate(self.repo, destmap, self.collapsef)
 
         if not result:
             # Empty state built, nothing to rebase
             self.ui.status(_(b'nothing to rebase\n'))
             return _nothingtorebase()
 
         (self.originalwd, self.destmap, self.state) = result
         if self.collapsef:
             dests = set(self.destmap.values())
             if len(dests) != 1:
                 raise error.Abort(
                     _(b'--collapse does not work with multiple destinations')
                 )
             destrev = next(iter(dests))
             destancestors = self.repo.changelog.ancestors(
                 [destrev], inclusive=True
             )
             self.external = externalparent(self.repo, self.state, destancestors)
 
         for destrev in sorted(set(destmap.values())):
             dest = self.repo[destrev]
             if dest.closesbranch() and not self.keepbranchesf:
                 self.ui.status(_(b'reopening closed branch head %s\n') % dest)
 
         self.prepared = True
 
     def _assignworkingcopy(self):
         if self.inmemory:
             from mercurial.context import overlayworkingctx
 
             self.wctx = overlayworkingctx(self.repo)
             self.repo.ui.debug(b"rebasing in-memory\n")
         else:
             self.wctx = self.repo[None]
             self.repo.ui.debug(b"rebasing on disk\n")
         self.repo.ui.log(
             b"rebase",
             b"using in-memory rebase: %r\n",
             self.inmemory,
             rebase_imm_used=self.inmemory,
         )
 
     def _performrebase(self, tr):
         self._assignworkingcopy()
         repo, ui = self.repo, self.ui
         if self.keepbranchesf:
             # insert _savebranch at the start of extrafns so if
             # there's a user-provided extrafn it can clobber branch if
             # desired
             self.extrafns.insert(0, _savebranch)
             if self.collapsef:
                 branches = set()
                 for rev in self.state:
                     branches.add(repo[rev].branch())
                     if len(branches) > 1:
                         raise error.Abort(
                             _(b'cannot collapse multiple named branches')
                         )
 
         # Calculate self.obsoletenotrebased
         obsrevs = _filterobsoleterevs(self.repo, self.state)
         self._handleskippingobsolete(obsrevs, self.destmap)
 
         # Keep track of the active bookmarks in order to reset them later
         self.activebookmark = self.activebookmark or repo._activebookmark
         if self.activebookmark:
             bookmarks.deactivate(repo)
 
         # Store the state before we begin so users can run 'hg rebase --abort'
         # if we fail before the transaction closes.
         self.storestatus()
         if tr:
             # When using single transaction, store state when transaction
             # commits.
             self.storestatus(tr)
 
         cands = [k for k, v in pycompat.iteritems(self.state) if v == revtodo]
         p = repo.ui.makeprogress(
             _(b"rebasing"), unit=_(b'changesets'), total=len(cands)
         )
 
         def progress(ctx):
             p.increment(item=(b"%d:%s" % (ctx.rev(), ctx)))
 
         allowdivergence = self.ui.configbool(
             b'experimental', b'evolution.allowdivergence'
         )
         for subset in sortsource(self.destmap):
             sortedrevs = self.repo.revs(b'sort(%ld, -topo)', subset)
             if not allowdivergence:
                 sortedrevs -= self.repo.revs(
                     b'descendants(%ld) and not %ld',
                     self.obsoletewithoutsuccessorindestination,
                     self.obsoletewithoutsuccessorindestination,
                 )
             for rev in sortedrevs:
                 self._rebasenode(tr, rev, allowdivergence, progress)
         p.complete()
         ui.note(_(b'rebase merging completed\n'))
 
     def _concludenode(self, rev, p1, p2, editor, commitmsg=None):
         '''Commit the wd changes with parents p1 and p2.
 
         Reuse commit info from rev but also store useful information in extra.
         Return node of committed revision.'''
         repo = self.repo
         ctx = repo[rev]
         if commitmsg is None:
             commitmsg = ctx.description()
         date = self.date
         if date is None:
             date = ctx.date()
         extra = {b'rebase_source': ctx.hex()}
         for c in self.extrafns:
             c(ctx, extra)
         keepbranch = self.keepbranchesf and repo[p1].branch() != ctx.branch()
         destphase = max(ctx.phase(), phases.draft)
         overrides = {(b'phases', b'new-commit'): destphase}
         if keepbranch:
             overrides[(b'ui', b'allowemptycommit')] = True
         with repo.ui.configoverride(overrides, b'rebase'):
             if self.inmemory:
                 newnode = commitmemorynode(
                     repo,
                     p1,
                     p2,
                     wctx=self.wctx,
                     extra=extra,
                     commitmsg=commitmsg,
                     editor=editor,
                     user=ctx.user(),
                     date=date,
                 )
                 mergemod.mergestate.clean(repo)
             else:
                 newnode = commitnode(
                     repo,
                     p1,
                     p2,
                     extra=extra,
                     commitmsg=commitmsg,
                     editor=editor,
                     user=ctx.user(),
                     date=date,
                 )
 
             if newnode is None:
                 # If it ended up being a no-op commit, then the normal
                 # merge state clean-up path doesn't happen, so do it
                 # here. Fix issue5494
                 mergemod.mergestate.clean(repo)
             return newnode
 
     def _rebasenode(self, tr, rev, allowdivergence, progressfn):
         repo, ui, opts = self.repo, self.ui, self.opts
         dest = self.destmap[rev]
         ctx = repo[rev]
         desc = _ctxdesc(ctx)
         if self.state[rev] == rev:
             ui.status(_(b'already rebased %s\n') % desc)
         elif (
             not allowdivergence
             and rev in self.obsoletewithoutsuccessorindestination
         ):
             msg = (
                 _(
                     b'note: not rebasing %s and its descendants as '
                     b'this would cause divergence\n'
                 )
                 % desc
             )
             repo.ui.status(msg)
             self.skipped.add(rev)
         elif rev in self.obsoletenotrebased:
             succ = self.obsoletenotrebased[rev]
             if succ is None:
                 msg = _(b'note: not rebasing %s, it has no successor\n') % desc
             else:
                 succdesc = _ctxdesc(repo[succ])
                 msg = _(
                     b'note: not rebasing %s, already in destination as %s\n'
                 ) % (desc, succdesc)
             repo.ui.status(msg)
             # Make clearrebased aware state[rev] is not a true successor
             self.skipped.add(rev)
             # Record rev as moved to its desired destination in self.state.
             # This helps bookmark and working parent movement.
             dest = max(
                 adjustdest(repo, rev, self.destmap, self.state, self.skipped)
             )
             self.state[rev] = dest
         elif self.state[rev] == revtodo:
             ui.status(_(b'rebasing %s\n') % desc)
             progressfn(ctx)
             p1, p2, base = defineparents(
                 repo,
                 rev,
                 self.destmap,
                 self.state,
                 self.skipped,
                 self.obsoletenotrebased,
             )
             if not self.inmemory and len(repo[None].parents()) == 2:
                 repo.ui.debug(b'resuming interrupted rebase\n')
             else:
                 overrides = {(b'ui', b'forcemerge'): opts.get(b'tool', b'')}
                 with ui.configoverride(overrides, b'rebase'):
                     stats = rebasenode(
                         repo,
                         rev,
                         p1,
                         base,
                         self.collapsef,
                         dest,
                         wctx=self.wctx,
                     )
                     if stats.unresolvedcount > 0:
                         if self.inmemory:
                             raise error.InMemoryMergeConflictsError()
                         else:
                             raise error.InterventionRequired(
                                 _(
                                     b'unresolved conflicts (see hg '
                                     b'resolve, then hg rebase --continue)'
                                 )
                             )
             if not self.collapsef:
                 merging = p2 != nullrev
                 editform = cmdutil.mergeeditform(merging, b'rebase')
                 editor = cmdutil.getcommiteditor(
                     editform=editform, **pycompat.strkwargs(opts)
                 )
                 newnode = self._concludenode(rev, p1, p2, editor)
             else:
                 # Skip commit if we are collapsing
                 if self.inmemory:
                     self.wctx.setbase(repo[p1])
                 else:
                     repo.setparents(repo[p1].node())
                 newnode = None
             # Update the state
             if newnode is not None:
                 self.state[rev] = repo[newnode].rev()
                 ui.debug(b'rebased as %s\n' % short(newnode))
             else:
                 if not self.collapsef:
                     ui.warn(
                         _(
                             b'note: not rebasing %s, its destination already '
                             b'has all its changes\n'
                         )
                         % desc
                     )
                     self.skipped.add(rev)
                 self.state[rev] = p1
                 ui.debug(b'next revision set to %d\n' % p1)
         else:
             ui.status(
                 _(b'already rebased %s as %s\n') % (desc, repo[self.state[rev]])
             )
         if not tr:
             # When not using single transaction, store state after each
             # commit is completely done. On InterventionRequired, we thus
             # won't store the status. Instead, we'll hit the "len(parents) == 2"
             # case and realize that the commit was in progress.
             self.storestatus()
 
     def _finishrebase(self):
         repo, ui, opts = self.repo, self.ui, self.opts
         fm = ui.formatter(b'rebase', opts)
         fm.startitem()
         if self.collapsef:
             p1, p2, _base = defineparents(
                 repo,
                 min(self.state),
                 self.destmap,
                 self.state,
                 self.skipped,
                 self.obsoletenotrebased,
             )
             editopt = opts.get(b'edit')
             editform = b'rebase.collapse'
             if self.collapsemsg:
                 commitmsg = self.collapsemsg
             else:
                 commitmsg = b'Collapsed revision'
                 for rebased in sorted(self.state):
                     if rebased not in self.skipped:
                         commitmsg += b'\n* %s' % repo[rebased].description()
                 editopt = True
             editor = cmdutil.getcommiteditor(edit=editopt, editform=editform)
             revtoreuse = max(self.state)
 
             newnode = self._concludenode(
                 revtoreuse, p1, self.external, editor, commitmsg=commitmsg
             )
 
             if newnode is not None:
                 newrev = repo[newnode].rev()
                 for oldrev in self.state:
                     self.state[oldrev] = newrev
 
         if b'qtip' in repo.tags():
             updatemq(repo, self.state, self.skipped, **pycompat.strkwargs(opts))
 
         # restore original working directory
         # (we do this before stripping)
         newwd = self.state.get(self.originalwd, self.originalwd)
         if newwd < 0:
             # original directory is a parent of rebase set root or ignored
             newwd = self.originalwd
         if newwd not in [c.rev() for c in repo[None].parents()]:
             ui.note(_(b"update back to initial working directory parent\n"))
             hg.updaterepo(repo, newwd, overwrite=False)
 
         collapsedas = None
         if self.collapsef and not self.keepf:
             collapsedas = newnode
         clearrebased(
             ui,
             repo,
             self.destmap,
             self.state,
             self.skipped,
             collapsedas,
             self.keepf,
             fm=fm,
             backup=self.backupf,
         )
 
         clearstatus(repo)
         clearcollapsemsg(repo)
 
         ui.note(_(b"rebase completed\n"))
         util.unlinkpath(repo.sjoin(b'undo'), ignoremissing=True)
         if self.skipped:
             skippedlen = len(self.skipped)
             ui.note(_(b"%d revisions have been skipped\n") % skippedlen)
         fm.end()
 
         if (
             self.activebookmark
             and self.activebookmark in repo._bookmarks
             and repo[b'.'].node() == repo._bookmarks[self.activebookmark]
         ):
             bookmarks.activate(repo, self.activebookmark)
 
     def _abort(self, backup=True, suppwarns=False):
         '''Restore the repository to its original state.'''
 
         repo = self.repo
         try:
             # If the first commits in the rebased set get skipped during the
             # rebase, their values within the state mapping will be the dest
             # rev id. The rebased list must must not contain the dest rev
             # (issue4896)
             rebased = [
                 s
                 for r, s in self.state.items()
                 if s >= 0 and s != r and s != self.destmap[r]
             ]
             immutable = [d for d in rebased if not repo[d].mutable()]
             cleanup = True
             if immutable:
                 repo.ui.warn(
                     _(b"warning: can't clean up public changesets %s\n")
                     % b', '.join(bytes(repo[r]) for r in immutable),
                     hint=_(b"see 'hg help phases' for details"),
                 )
                 cleanup = False
 
             descendants = set()
             if rebased:
                 descendants = set(repo.changelog.descendants(rebased))
             if descendants - set(rebased):
                 repo.ui.warn(
                     _(
                         b"warning: new changesets detected on "
                         b"destination branch, can't strip\n"
                     )
                 )
                 cleanup = False
 
             if cleanup:
                 if rebased:
                     strippoints = [
                         c.node() for c in repo.set(b'roots(%ld)', rebased)
                     ]
 
                 updateifonnodes = set(rebased)
                 updateifonnodes.update(self.destmap.values())
                 updateifonnodes.add(self.originalwd)
                 shouldupdate = repo[b'.'].rev() in updateifonnodes
 
                 # Update away from the rebase if necessary
                 if shouldupdate:
                     mergemod.update(
                         repo, self.originalwd, branchmerge=False, force=True
                     )
 
                 # Strip from the first rebased revision
                 if rebased:
                     repair.strip(repo.ui, repo, strippoints, backup=backup)
 
             if self.activebookmark and self.activebookmark in repo._bookmarks:
                 bookmarks.activate(repo, self.activebookmark)
 
         finally:
             clearstatus(repo)
             clearcollapsemsg(repo)
             if not suppwarns:
                 repo.ui.warn(_(b'rebase aborted\n'))
         return 0
 
 
 @command(
     b'rebase',
     [
         (
             b's',
             b'source',
             b'',
             _(b'rebase the specified changeset and descendants'),
             _(b'REV'),
         ),
         (
             b'b',
             b'base',
             b'',
             _(b'rebase everything from branching point of specified changeset'),
             _(b'REV'),
         ),
         (b'r', b'rev', [], _(b'rebase these revisions'), _(b'REV')),
         (
             b'd',
             b'dest',
             b'',
             _(b'rebase onto the specified changeset'),
             _(b'REV'),
         ),
         (b'', b'collapse', False, _(b'collapse the rebased changesets')),
         (
             b'm',
             b'message',
             b'',
             _(b'use text as collapse commit message'),
             _(b'TEXT'),
         ),
         (b'e', b'edit', False, _(b'invoke editor on commit messages')),
         (
             b'l',
             b'logfile',
             b'',
             _(b'read collapse commit message from file'),
             _(b'FILE'),
         ),
         (b'k', b'keep', False, _(b'keep original changesets')),
         (b'', b'keepbranches', False, _(b'keep original branch names')),
         (b'D', b'detach', False, _(b'(DEPRECATED)')),
         (b'i', b'interactive', False, _(b'(DEPRECATED)')),
         (b't', b'tool', b'', _(b'specify merge tool')),
         (b'', b'stop', False, _(b'stop interrupted rebase')),
         (b'c', b'continue', False, _(b'continue an interrupted rebase')),
         (b'a', b'abort', False, _(b'abort an interrupted rebase')),
         (
             b'',
             b'auto-orphans',
             b'',
             _(
                 b'automatically rebase orphan revisions '
                 b'in the specified revset (EXPERIMENTAL)'
             ),
         ),
     ]
     + cmdutil.dryrunopts
     + cmdutil.formatteropts
     + cmdutil.confirmopts,
     _(b'[-s REV | -b REV] [-d REV] [OPTION]'),
     helpcategory=command.CATEGORY_CHANGE_MANAGEMENT,
 )
 def rebase(ui, repo, **opts):
     """move changeset (and descendants) to a different branch
 
     Rebase uses repeated merging to graft changesets from one part of
     history (the source) onto another (the destination). This can be
     useful for linearizing *local* changes relative to a master
     development tree.
 
     Published commits cannot be rebased (see :hg:`help phases`).
     To copy commits, see :hg:`help graft`.
 
     If you don't specify a destination changeset (``-d/--dest``), rebase
     will use the same logic as :hg:`merge` to pick a destination.  if
     the current branch contains exactly one other head, the other head
     is merged with by default.  Otherwise, an explicit revision with
     which to merge with must be provided.  (destination changeset is not
     modified by rebasing, but new changesets are added as its
     descendants.)
 
     Here are the ways to select changesets:
 
       1. Explicitly select them using ``--rev``.
 
       2. Use ``--source`` to select a root changeset and include all of its
          descendants.
 
       3. Use ``--base`` to select a changeset; rebase will find ancestors
          and their descendants which are not also ancestors of the destination.
 
       4. If you do not specify any of ``--rev``, ``--source``, or ``--base``,
          rebase will use ``--base .`` as above.
 
     If ``--source`` or ``--rev`` is used, special names ``SRC`` and ``ALLSRC``
     can be used in ``--dest``. Destination would be calculated per source
     revision with ``SRC`` substituted by that single source revision and
     ``ALLSRC`` substituted by all source revisions.
 
     Rebase will destroy original changesets unless you use ``--keep``.
     It will also move your bookmarks (even if you do).
 
     Some changesets may be dropped if they do not contribute changes
     (e.g. merges from the destination branch).
 
     Unlike ``merge``, rebase will do nothing if you are at the branch tip of
     a named branch with two heads. You will need to explicitly specify source
     and/or destination.
 
     If you need to use a tool to automate merge/conflict decisions, you
     can specify one with ``--tool``, see :hg:`help merge-tools`.
     As a caveat: the tool will not be used to mediate when a file was
     deleted, there is no hook presently available for this.
 
     If a rebase is interrupted to manually resolve a conflict, it can be
     continued with --continue/-c, aborted with --abort/-a, or stopped with
     --stop.
 
     .. container:: verbose
 
       Examples:
 
       - move "local changes" (current commit back to branching point)
         to the current branch tip after a pull::
 
           hg rebase
 
       - move a single changeset to the stable branch::
 
           hg rebase -r 5f493448 -d stable
 
       - splice a commit and all its descendants onto another part of history::
 
           hg rebase --source c0c3 --dest 4cf9
 
       - rebase everything on a branch marked by a bookmark onto the
         default branch::
 
           hg rebase --base myfeature --dest default
 
       - collapse a sequence of changes into a single commit::
 
           hg rebase --collapse -r 1520:1525 -d .
 
       - move a named branch while preserving its name::
 
           hg rebase -r "branch(featureX)" -d 1.3 --keepbranches
 
       - stabilize orphaned changesets so history looks linear::
 
           hg rebase -r 'orphan()-obsolete()'\
  -d 'first(max((successors(max(roots(ALLSRC) & ::SRC)^)-obsolete())::) +\
  max(::((roots(ALLSRC) & ::SRC)^)-obsolete()))'
 
     Configuration Options:
 
     You can make rebase require a destination if you set the following config
     option::
 
       [commands]
       rebase.requiredest = True
 
     By default, rebase will close the transaction after each commit. For
     performance purposes, you can configure rebase to use a single transaction
     across the entire rebase. WARNING: This setting introduces a significant
     risk of losing the work you've done in a rebase if the rebase aborts
     unexpectedly::
 
       [rebase]
       singletransaction = True
 
     By default, rebase writes to the working copy, but you can configure it to
     run in-memory for better performance. When the rebase is not moving the
     parent(s) of the working copy (AKA the "currently checked out changesets"),
     this may also allow it to run even if the working copy is dirty::
 
       [rebase]
       experimental.inmemory = True
 
     Return Values:
 
     Returns 0 on success, 1 if nothing to rebase or there are
     unresolved conflicts.
 
     """
     opts = pycompat.byteskwargs(opts)
     inmemory = ui.configbool(b'rebase', b'experimental.inmemory')
     action = cmdutil.check_at_most_one_arg(opts, b'abort', b'stop', b'continue')
     if action:
         cmdutil.check_incompatible_arguments(
             opts, action, b'confirm', b'dry_run'
         )
         cmdutil.check_incompatible_arguments(
             opts, action, b'rev', b'source', b'base', b'dest'
         )
     cmdutil.check_at_most_one_arg(opts, b'confirm', b'dry_run')
     cmdutil.check_at_most_one_arg(opts, b'rev', b'source', b'base')
 
     if action or repo.currenttransaction() is not None:
         # in-memory rebase is not compatible with resuming rebases.
         # (Or if it is run within a transaction, since the restart logic can
         # fail the entire transaction.)
         inmemory = False
 
     if opts.get(b'auto_orphans'):
         disallowed_opts = set(opts) - {b'auto_orphans'}
         cmdutil.check_incompatible_arguments(
             opts, b'auto_orphans', *disallowed_opts
         )
 
         userrevs = list(repo.revs(opts.get(b'auto_orphans')))
         opts[b'rev'] = [revsetlang.formatspec(b'%ld and orphan()', userrevs)]
         opts[b'dest'] = b'_destautoorphanrebase(SRC)'
 
     if opts.get(b'dry_run') or opts.get(b'confirm'):
         return _dryrunrebase(ui, repo, action, opts)
     elif action == b'stop':
         rbsrt = rebaseruntime(repo, ui)
         with repo.wlock(), repo.lock():
             rbsrt.restorestatus()
             if rbsrt.collapsef:
                 raise error.Abort(_(b"cannot stop in --collapse session"))
             allowunstable = obsolete.isenabled(repo, obsolete.allowunstableopt)
             if not (rbsrt.keepf or allowunstable):
                 raise error.Abort(
                     _(
                         b"cannot remove original changesets with"
                         b" unrebased descendants"
                     ),
                     hint=_(
                         b'either enable obsmarkers to allow unstable '
                         b'revisions or use --keep to keep original '
                         b'changesets'
                     ),
                 )
             # update to the current working revision
             # to clear interrupted merge
             hg.updaterepo(repo, rbsrt.originalwd, overwrite=True)
             rbsrt._finishrebase()
             return 0
     elif inmemory:
         try:
             # in-memory merge doesn't support conflicts, so if we hit any, abort
             # and re-run as an on-disk merge.
             overrides = {(b'rebase', b'singletransaction'): True}
             with ui.configoverride(overrides, b'rebase'):
                 return _dorebase(ui, repo, action, opts, inmemory=inmemory)
         except error.InMemoryMergeConflictsError:
             ui.warn(
                 _(
                     b'hit merge conflicts; re-running rebase without in-memory'
                     b' merge\n'
                 )
             )
             # TODO: Make in-memory merge not use the on-disk merge state, so
             # we don't have to clean it here
             mergemod.mergestate.clean(repo)
             clearstatus(repo)
             clearcollapsemsg(repo)
             return _dorebase(ui, repo, action, opts, inmemory=False)
     else:
         return _dorebase(ui, repo, action, opts)
 
 
 def _dryrunrebase(ui, repo, action, opts):
     rbsrt = rebaseruntime(repo, ui, inmemory=True, opts=opts)
     confirm = opts.get(b'confirm')
     if confirm:
         ui.status(_(b'starting in-memory rebase\n'))
     else:
         ui.status(
             _(b'starting dry-run rebase; repository will not be changed\n')
         )
     with repo.wlock(), repo.lock():
         needsabort = True
         try:
             overrides = {(b'rebase', b'singletransaction'): True}
             with ui.configoverride(overrides, b'rebase'):
                 _origrebase(
                     ui,
                     repo,
                     action,
                     opts,
                     rbsrt,
                     inmemory=True,
                     leaveunfinished=True,
                 )
         except error.InMemoryMergeConflictsError:
             ui.status(_(b'hit a merge conflict\n'))
             return 1
         except error.Abort:
             needsabort = False
             raise
         else:
             if confirm:
                 ui.status(_(b'rebase completed successfully\n'))
                 if not ui.promptchoice(_(b'apply changes (yn)?$$ &Yes $$ &No')):
                     # finish unfinished rebase
                     rbsrt._finishrebase()
                 else:
                     rbsrt._prepareabortorcontinue(
                         isabort=True, backup=False, suppwarns=True
                     )
                 needsabort = False
             else:
                 ui.status(
                     _(
                         b'dry-run rebase completed successfully; run without'
                         b' -n/--dry-run to perform this rebase\n'
                     )
                 )
             return 0
         finally:
             if needsabort:
                 # no need to store backup in case of dryrun
                 rbsrt._prepareabortorcontinue(
                     isabort=True, backup=False, suppwarns=True
                 )
 
 
 def _dorebase(ui, repo, action, opts, inmemory=False):
     rbsrt = rebaseruntime(repo, ui, inmemory, opts)
     return _origrebase(ui, repo, action, opts, rbsrt, inmemory=inmemory)
 
 
 def _origrebase(
     ui, repo, action, opts, rbsrt, inmemory=False, leaveunfinished=False
 ):
     assert action != b'stop'
     with repo.wlock(), repo.lock():
         if opts.get(b'interactive'):
             try:
                 if extensions.find(b'histedit'):
                     enablehistedit = b''
             except KeyError:
                 enablehistedit = b" --config extensions.histedit="
             help = b"hg%s help -e histedit" % enablehistedit
             msg = (
                 _(
                     b"interactive history editing is supported by the "
                     b"'histedit' extension (see \"%s\")"
                 )
                 % help
             )
             raise error.Abort(msg)
 
         if rbsrt.collapsemsg and not rbsrt.collapsef:
             raise error.Abort(_(b'message can only be specified with collapse'))
 
         if action:
             if rbsrt.collapsef:
                 raise error.Abort(
                     _(b'cannot use collapse with continue or abort')
                 )
             if action == b'abort' and opts.get(b'tool', False):
                 ui.warn(_(b'tool option will be ignored\n'))
             if action == b'continue':
                 ms = mergemod.mergestate.read(repo)
                 mergeutil.checkunresolved(ms)
 
             retcode = rbsrt._prepareabortorcontinue(
                 isabort=(action == b'abort')
             )
             if retcode is not None:
                 return retcode
         else:
             # search default destination in this space
             # used in the 'hg pull --rebase' case, see issue 5214.
             destspace = opts.get(b'_destspace')
             destmap = _definedestmap(
                 ui,
                 repo,
                 inmemory,
                 opts.get(b'dest', None),
                 opts.get(b'source', None),
                 opts.get(b'base', None),
                 opts.get(b'rev', []),
                 destspace=destspace,
             )
             retcode = rbsrt._preparenewrebase(destmap)
             if retcode is not None:
                 return retcode
             storecollapsemsg(repo, rbsrt.collapsemsg)
 
         tr = None
 
         singletr = ui.configbool(b'rebase', b'singletransaction')
         if singletr:
             tr = repo.transaction(b'rebase')
 
         # If `rebase.singletransaction` is enabled, wrap the entire operation in
         # one transaction here. Otherwise, transactions are obtained when
         # committing each node, which is slower but allows partial success.
         with util.acceptintervention(tr):
             # Same logic for the dirstate guard, except we don't create one when
             # rebasing in-memory (it's not needed).
             dsguard = None
             if singletr and not inmemory:
                 dsguard = dirstateguard.dirstateguard(repo, b'rebase')
             with util.acceptintervention(dsguard):
                 rbsrt._performrebase(tr)
                 if not leaveunfinished:
                     rbsrt._finishrebase()
 
 
 def _definedestmap(
     ui,
     repo,
     inmemory,
     destf=None,
     srcf=None,
     basef=None,
     revf=None,
     destspace=None,
 ):
     """use revisions argument to define destmap {srcrev: destrev}"""
     if revf is None:
         revf = []
 
     # destspace is here to work around issues with `hg pull --rebase` see
     # issue5214 for details
 
     cmdutil.checkunfinished(repo)
     if not inmemory:
         cmdutil.bailifchanged(repo)
 
     if ui.configbool(b'commands', b'rebase.requiredest') and not destf:
         raise error.Abort(
             _(b'you must specify a destination'),
             hint=_(b'use: hg rebase -d REV'),
         )
 
     dest = None
 
     if revf:
         rebaseset = scmutil.revrange(repo, revf)
         if not rebaseset:
             ui.status(_(b'empty "rev" revision set - nothing to rebase\n'))
             return None
     elif srcf:
         src = scmutil.revrange(repo, [srcf])
         if not src:
             ui.status(_(b'empty "source" revision set - nothing to rebase\n'))
             return None
         rebaseset = repo.revs(b'(%ld)::', src)
         assert rebaseset
     else:
         base = scmutil.revrange(repo, [basef or b'.'])
         if not base:
             ui.status(
                 _(b'empty "base" revision set - ' b"can't compute rebase set\n")
             )
             return None
         if destf:
             # --base does not support multiple destinations
             dest = scmutil.revsingle(repo, destf)
         else:
             dest = repo[_destrebase(repo, base, destspace=destspace)]
             destf = bytes(dest)
 
         roots = []  # selected children of branching points
         bpbase = {}  # {branchingpoint: [origbase]}
         for b in base:  # group bases by branching points
             bp = repo.revs(b'ancestor(%d, %d)', b, dest.rev()).first()
             bpbase[bp] = bpbase.get(bp, []) + [b]
         if None in bpbase:
             # emulate the old behavior, showing "nothing to rebase" (a better
             # behavior may be abort with "cannot find branching point" error)
             bpbase.clear()
         for bp, bs in pycompat.iteritems(bpbase):  # calculate roots
             roots += list(repo.revs(b'children(%d) & ancestors(%ld)', bp, bs))
 
         rebaseset = repo.revs(b'%ld::', roots)
 
         if not rebaseset:
             # transform to list because smartsets are not comparable to
             # lists. This should be improved to honor laziness of
             # smartset.
             if list(base) == [dest.rev()]:
                 if basef:
                     ui.status(
                         _(
                             b'nothing to rebase - %s is both "base"'
                             b' and destination\n'
                         )
                         % dest
                     )
                 else:
                     ui.status(
                         _(
                             b'nothing to rebase - working directory '
                             b'parent is also destination\n'
                         )
                     )
             elif not repo.revs(b'%ld - ::%d', base, dest.rev()):
                 if basef:
                     ui.status(
                         _(
                             b'nothing to rebase - "base" %s is '
                             b'already an ancestor of destination '
                             b'%s\n'
                         )
                         % (b'+'.join(bytes(repo[r]) for r in base), dest)
                     )
                 else:
                     ui.status(
                         _(
                             b'nothing to rebase - working '
                             b'directory parent is already an '
                             b'ancestor of destination %s\n'
                         )
                         % dest
                     )
             else:  # can it happen?
                 ui.status(
                     _(b'nothing to rebase from %s to %s\n')
                     % (b'+'.join(bytes(repo[r]) for r in base), dest)
                 )
             return None
 
     rebasingwcp = repo[b'.'].rev() in rebaseset
     ui.log(
         b"rebase",
         b"rebasing working copy parent: %r\n",
         rebasingwcp,
         rebase_rebasing_wcp=rebasingwcp,
     )
     if inmemory and rebasingwcp:
         # Check these since we did not before.
         cmdutil.checkunfinished(repo)
         cmdutil.bailifchanged(repo)
 
     if not destf:
         dest = repo[_destrebase(repo, rebaseset, destspace=destspace)]
         destf = bytes(dest)
 
     allsrc = revsetlang.formatspec(b'%ld', rebaseset)
     alias = {b'ALLSRC': allsrc}
 
     if dest is None:
         try:
             # fast path: try to resolve dest without SRC alias
             dest = scmutil.revsingle(repo, destf, localalias=alias)
         except error.RepoLookupError:
             # multi-dest path: resolve dest for each SRC separately
             destmap = {}
             for r in rebaseset:
                 alias[b'SRC'] = revsetlang.formatspec(b'%d', r)
                 # use repo.anyrevs instead of scmutil.revsingle because we
                 # don't want to abort if destset is empty.
                 destset = repo.anyrevs([destf], user=True, localalias=alias)
                 size = len(destset)
                 if size == 1:
                     destmap[r] = destset.first()
                 elif size == 0:
                     ui.note(_(b'skipping %s - empty destination\n') % repo[r])
                 else:
                     raise error.Abort(
                         _(b'rebase destination for %s is not unique') % repo[r]
                     )
 
     if dest is not None:
         # single-dest case: assign dest to each rev in rebaseset
         destrev = dest.rev()
         destmap = {r: destrev for r in rebaseset}  # {srcrev: destrev}
 
     if not destmap:
         ui.status(_(b'nothing to rebase - empty destination\n'))
         return None
 
     return destmap
 
 
 def externalparent(repo, state, destancestors):
     """Return the revision that should be used as the second parent
     when the revisions in state is collapsed on top of destancestors.
     Abort if there is more than one parent.
     """
     parents = set()
     source = min(state)
     for rev in state:
         if rev == source:
             continue
         for p in repo[rev].parents():
             if p.rev() not in state and p.rev() not in destancestors:
                 parents.add(p.rev())
     if not parents:
         return nullrev
     if len(parents) == 1:
         return parents.pop()
     raise error.Abort(
         _(
             b'unable to collapse on top of %d, there is more '
             b'than one external parent: %s'
         )
         % (max(destancestors), b', '.join(b"%d" % p for p in sorted(parents)))
     )
 
 
 def commitmemorynode(repo, p1, p2, wctx, editor, extra, user, date, commitmsg):
     '''Commit the memory changes with parents p1 and p2.
     Return node of committed revision.'''
     # Replicates the empty check in ``repo.commit``.
     if wctx.isempty() and not repo.ui.configbool(b'ui', b'allowemptycommit'):
         return None
 
     # By convention, ``extra['branch']`` (set by extrafn) clobbers
     # ``branch`` (used when passing ``--keepbranches``).
     branch = None
     if b'branch' in extra:
         branch = extra[b'branch']
 
     wctx.setparents(repo[p1].node(), repo[p2].node())
     memctx = wctx.tomemctx(
         commitmsg,
         date=date,
         extra=extra,
         user=user,
         branch=branch,
         editor=editor,
     )
     commitres = repo.commitctx(memctx)
     wctx.clean()  # Might be reused
     return commitres
 
 
 def commitnode(repo, p1, p2, editor, extra, user, date, commitmsg):
     '''Commit the wd changes with parents p1 and p2.
     Return node of committed revision.'''
     dsguard = util.nullcontextmanager()
     if not repo.ui.configbool(b'rebase', b'singletransaction'):
         dsguard = dirstateguard.dirstateguard(repo, b'rebase')
     with dsguard:
         repo.setparents(repo[p1].node(), repo[p2].node())
 
         # Commit might fail if unresolved files exist
         newnode = repo.commit(
             text=commitmsg, user=user, date=date, extra=extra, editor=editor
         )
 
         repo.dirstate.setbranch(repo[newnode].branch())
         return newnode
 
 
 def rebasenode(repo, rev, p1, base, collapse, dest, wctx):
     """Rebase a single revision rev on top of p1 using base as merge ancestor"""
     # Merge phase
     # Update to destination and merge it with local
     if wctx.isinmemory():
         wctx.setbase(repo[p1])
     else:
         if repo[b'.'].rev() != p1:
             repo.ui.debug(b" update to %d:%s\n" % (p1, repo[p1]))
             mergemod.update(repo, p1, branchmerge=False, force=True)
         else:
             repo.ui.debug(b" already in destination\n")
         # This is, alas, necessary to invalidate workingctx's manifest cache,
         # as well as other data we litter on it in other places.
         wctx = repo[None]
         repo.dirstate.write(repo.currenttransaction())
     ctx = repo[rev]
     repo.ui.debug(b" merge against %d:%s\n" % (rev, ctx))
     if base is not None:
         repo.ui.debug(b"   detach base %d:%s\n" % (base, repo[base]))
     # When collapsing in-place, the parent is the common ancestor, we
     # have to allow merging with it.
     stats = mergemod.update(
         repo,
         rev,
         branchmerge=True,
         force=True,
         ancestor=base,
         mergeancestor=collapse,
         labels=[b'dest', b'source'],
         wc=wctx,
     )
-    destctx = repo[dest]
     if collapse:
-        copies.graftcopies(repo, wctx, ctx, destctx)
+        copies.graftcopies(wctx, ctx, repo[dest])
     else:
         # If we're not using --collapse, we need to
         # duplicate copies between the revision we're
-        # rebasing and its first parent, but *not*
-        # duplicate any copies that have already been
-        # performed in the destination.
-        copies.graftcopies(repo, wctx, ctx, ctx.p1(), skip=destctx)
+        # rebasing and its first parent.
+        copies.graftcopies(wctx, ctx, ctx.p1())
     return stats
 
 
 def adjustdest(repo, rev, destmap, state, skipped):
     r"""adjust rebase destination given the current rebase state
 
     rev is what is being rebased. Return a list of two revs, which are the
     adjusted destinations for rev's p1 and p2, respectively. If a parent is
     nullrev, return dest without adjustment for it.
 
     For example, when doing rebasing B+E to F, C to G, rebase will first move B
     to B1, and E's destination will be adjusted from F to B1.
 
         B1 <- written during rebasing B
         |
         F <- original destination of B, E
         |
         | E <- rev, which is being rebased
         | |
         | D <- prev, one parent of rev being checked
         | |
         | x <- skipped, ex. no successor or successor in (::dest)
         | |
         | C <- rebased as C', different destination
         | |
         | B <- rebased as B1     C'
         |/                       |
         A                        G <- destination of C, different
 
     Another example about merge changeset, rebase -r C+G+H -d K, rebase will
     first move C to C1, G to G1, and when it's checking H, the adjusted
     destinations will be [C1, G1].
 
             H       C1 G1
            /|       | /
           F G       |/
         K | |  ->   K
         | C D       |
         | |/        |
         | B         | ...
         |/          |/
         A           A
 
     Besides, adjust dest according to existing rebase information. For example,
 
       B C D    B needs to be rebased on top of C, C needs to be rebased on top
        \|/     of D. We will rebase C first.
         A
 
           C'   After rebasing C, when considering B's destination, use C'
           |    instead of the original C.
       B   D
        \ /
         A
     """
     # pick already rebased revs with same dest from state as interesting source
     dest = destmap[rev]
     source = [
         s
         for s, d in state.items()
         if d > 0 and destmap[s] == dest and s not in skipped
     ]
 
     result = []
     for prev in repo.changelog.parentrevs(rev):
         adjusted = dest
         if prev != nullrev:
             candidate = repo.revs(b'max(%ld and (::%d))', source, prev).first()
             if candidate is not None:
                 adjusted = state[candidate]
         if adjusted == dest and dest in state:
             adjusted = state[dest]
             if adjusted == revtodo:
                 # sortsource should produce an order that makes this impossible
                 raise error.ProgrammingError(
                     b'rev %d should be rebased already at this time' % dest
                 )
         result.append(adjusted)
     return result
 
 
 def _checkobsrebase(repo, ui, rebaseobsrevs, rebaseobsskipped):
     """
     Abort if rebase will create divergence or rebase is noop because of markers
 
     `rebaseobsrevs`: set of obsolete revision in source
     `rebaseobsskipped`: set of revisions from source skipped because they have
     successors in destination or no non-obsolete successor.
     """
     # Obsolete node with successors not in dest leads to divergence
     divergenceok = ui.configbool(b'experimental', b'evolution.allowdivergence')
     divergencebasecandidates = rebaseobsrevs - rebaseobsskipped
 
     if divergencebasecandidates and not divergenceok:
         divhashes = (bytes(repo[r]) for r in divergencebasecandidates)
         msg = _(b"this rebase will cause divergences from: %s")
         h = _(
             b"to force the rebase please set "
             b"experimental.evolution.allowdivergence=True"
         )
         raise error.Abort(msg % (b",".join(divhashes),), hint=h)
 
 
 def successorrevs(unfi, rev):
     """yield revision numbers for successors of rev"""
     assert unfi.filtername is None
     get_rev = unfi.changelog.index.get_rev
     for s in obsutil.allsuccessors(unfi.obsstore, [unfi[rev].node()]):
         r = get_rev(s)
         if r is not None:
             yield r
 
 
 def defineparents(repo, rev, destmap, state, skipped, obsskipped):
     """Return new parents and optionally a merge base for rev being rebased
 
     The destination specified by "dest" cannot always be used directly because
     previously rebase result could affect destination. For example,
 
           D E    rebase -r C+D+E -d B
           |/     C will be rebased to C'
         B C      D's new destination will be C' instead of B
         |/       E's new destination will be C' instead of B
         A
 
     The new parents of a merge is slightly more complicated. See the comment
     block below.
     """
     # use unfiltered changelog since successorrevs may return filtered nodes
     assert repo.filtername is None
     cl = repo.changelog
     isancestor = cl.isancestorrev
 
     dest = destmap[rev]
     oldps = repo.changelog.parentrevs(rev)  # old parents
     newps = [nullrev, nullrev]  # new parents
     dests = adjustdest(repo, rev, destmap, state, skipped)
     bases = list(oldps)  # merge base candidates, initially just old parents
 
     if all(r == nullrev for r in oldps[1:]):
         # For non-merge changeset, just move p to adjusted dest as requested.
         newps[0] = dests[0]
     else:
         # For merge changeset, if we move p to dests[i] unconditionally, both
         # parents may change and the end result looks like "the merge loses a
         # parent", which is a surprise. This is a limit because "--dest" only
         # accepts one dest per src.
         #
         # Therefore, only move p with reasonable conditions (in this order):
         #   1. use dest, if dest is a descendent of (p or one of p's successors)
         #   2. use p's rebased result, if p is rebased (state[p] > 0)
         #
         # Comparing with adjustdest, the logic here does some additional work:
         #   1. decide which parents will not be moved towards dest
         #   2. if the above decision is "no", should a parent still be moved
         #      because it was rebased?
         #
         # For example:
         #
         #     C    # "rebase -r C -d D" is an error since none of the parents
         #    /|    # can be moved. "rebase -r B+C -d D" will move C's parent
         #   A B D  # B (using rule "2."), since B will be rebased.
         #
         # The loop tries to be not rely on the fact that a Mercurial node has
         # at most 2 parents.
         for i, p in enumerate(oldps):
             np = p  # new parent
             if any(isancestor(x, dests[i]) for x in successorrevs(repo, p)):
                 np = dests[i]
             elif p in state and state[p] > 0:
                 np = state[p]
 
             # "bases" only record "special" merge bases that cannot be
             # calculated from changelog DAG (i.e. isancestor(p, np) is False).
             # For example:
             #
             #   B'   # rebase -s B -d D, when B was rebased to B'. dest for C
             #   | C  # is B', but merge base for C is B, instead of
             #   D |  # changelog.ancestor(C, B') == A. If changelog DAG and
             #   | B  # "state" edges are merged (so there will be an edge from
             #   |/   # B to B'), the merge base is still ancestor(C, B') in
             #   A    # the merged graph.
             #
             # Also see https://bz.mercurial-scm.org/show_bug.cgi?id=1950#c8
             # which uses "virtual null merge" to explain this situation.
             if isancestor(p, np):
                 bases[i] = nullrev
 
             # If one parent becomes an ancestor of the other, drop the ancestor
             for j, x in enumerate(newps[:i]):
                 if x == nullrev:
                     continue
                 if isancestor(np, x):  # CASE-1
                     np = nullrev
                 elif isancestor(x, np):  # CASE-2
                     newps[j] = np
                     np = nullrev
                     # New parents forming an ancestor relationship does not
                     # mean the old parents have a similar relationship. Do not
                     # set bases[x] to nullrev.
                     bases[j], bases[i] = bases[i], bases[j]
 
             newps[i] = np
 
         # "rebasenode" updates to new p1, and the old p1 will be used as merge
         # base. If only p2 changes, merging using unchanged p1 as merge base is
         # suboptimal. Therefore swap parents to make the merge sane.
         if newps[1] != nullrev and oldps[0] == newps[0]:
             assert len(newps) == 2 and len(oldps) == 2
             newps.reverse()
             bases.reverse()
 
         # No parent change might be an error because we fail to make rev a
         # descendent of requested dest. This can happen, for example:
         #
         #     C    # rebase -r C -d D
         #    /|    # None of A and B will be changed to D and rebase fails.
         #   A B D
         if set(newps) == set(oldps) and dest not in newps:
             raise error.Abort(
                 _(
                     b'cannot rebase %d:%s without '
                     b'moving at least one of its parents'
                 )
                 % (rev, repo[rev])
             )
 
     # Source should not be ancestor of dest. The check here guarantees it's
     # impossible. With multi-dest, the initial check does not cover complex
     # cases since we don't have abstractions to dry-run rebase cheaply.
     if any(p != nullrev and isancestor(rev, p) for p in newps):
         raise error.Abort(_(b'source is ancestor of destination'))
 
     # "rebasenode" updates to new p1, use the corresponding merge base.
     if bases[0] != nullrev:
         base = bases[0]
     else:
         base = None
 
     # Check if the merge will contain unwanted changes. That may happen if
     # there are multiple special (non-changelog ancestor) merge bases, which
     # cannot be handled well by the 3-way merge algorithm. For example:
     #
     #     F
     #    /|
     #   D E  # "rebase -r D+E+F -d Z", when rebasing F, if "D" was chosen
     #   | |  # as merge base, the difference between D and F will include
     #   B C  # C, so the rebased F will contain C surprisingly. If "E" was
     #   |/   #  chosen, the rebased F will contain B.
     #   A Z
     #
     # But our merge base candidates (D and E in above case) could still be
     # better than the default (ancestor(F, Z) == null). Therefore still
     # pick one (so choose p1 above).
     if sum(1 for b in set(bases) if b != nullrev) > 1:
         unwanted = [None, None]  # unwanted[i]: unwanted revs if choose bases[i]
         for i, base in enumerate(bases):
             if base == nullrev:
                 continue
             # Revisions in the side (not chosen as merge base) branch that
             # might contain "surprising" contents
             siderevs = list(
                 repo.revs(b'((%ld-%d) %% (%d+%d))', bases, base, base, dest)
             )
 
             # If those revisions are covered by rebaseset, the result is good.
             # A merge in rebaseset would be considered to cover its ancestors.
             if siderevs:
                 rebaseset = [
                     r for r, d in state.items() if d > 0 and r not in obsskipped
                 ]
                 merges = [
                     r for r in rebaseset if cl.parentrevs(r)[1] != nullrev
                 ]
                 unwanted[i] = list(
                     repo.revs(
                         b'%ld - (::%ld) - %ld', siderevs, merges, rebaseset
                     )
                 )
 
         # Choose a merge base that has a minimal number of unwanted revs.
         l, i = min(
             (len(revs), i)
             for i, revs in enumerate(unwanted)
             if revs is not None
         )
         base = bases[i]
 
         # newps[0] should match merge base if possible. Currently, if newps[i]
         # is nullrev, the only case is newps[i] and newps[j] (j < i), one is
         # the other's ancestor. In that case, it's fine to not swap newps here.
         # (see CASE-1 and CASE-2 above)
         if i != 0 and newps[i] != nullrev:
             newps[0], newps[i] = newps[i], newps[0]
 
         # The merge will include unwanted revisions. Abort now. Revisit this if
         # we have a more advanced merge algorithm that handles multiple bases.
         if l > 0:
             unwanteddesc = _(b' or ').join(
                 (
                     b', '.join(b'%d:%s' % (r, repo[r]) for r in revs)
                     for revs in unwanted
                     if revs is not None
                 )
             )
             raise error.Abort(
                 _(b'rebasing %d:%s will include unwanted changes from %s')
                 % (rev, repo[rev], unwanteddesc)
             )
 
     repo.ui.debug(b" future parents are %d and %d\n" % tuple(newps))
 
     return newps[0], newps[1], base
 
 
 def isagitpatch(repo, patchname):
     """Return true if the given patch is in git format"""
     mqpatch = os.path.join(repo.mq.path, patchname)
     for line in patch.linereader(open(mqpatch, b'rb')):
         if line.startswith(b'diff --git'):
             return True
     return False
 
 
 def updatemq(repo, state, skipped, **opts):
     """Update rebased mq patches - finalize and then import them"""
     mqrebase = {}
     mq = repo.mq
     original_series = mq.fullseries[:]
     skippedpatches = set()
 
     for p in mq.applied:
         rev = repo[p.node].rev()
         if rev in state:
             repo.ui.debug(
                 b'revision %d is an mq patch (%s), finalize it.\n'
                 % (rev, p.name)
             )
             mqrebase[rev] = (p.name, isagitpatch(repo, p.name))
         else:
             # Applied but not rebased, not sure this should happen
             skippedpatches.add(p.name)
 
     if mqrebase:
         mq.finish(repo, mqrebase.keys())
 
         # We must start import from the newest revision
         for rev in sorted(mqrebase, reverse=True):
             if rev not in skipped:
                 name, isgit = mqrebase[rev]
                 repo.ui.note(
                     _(b'updating mq patch %s to %d:%s\n')
                     % (name, state[rev], repo[state[rev]])
                 )
                 mq.qimport(
                     repo,
                     (),
                     patchname=name,
                     git=isgit,
                     rev=[b"%d" % state[rev]],
                 )
             else:
                 # Rebased and skipped
                 skippedpatches.add(mqrebase[rev][0])
 
         # Patches were either applied and rebased and imported in
         # order, applied and removed or unapplied. Discard the removed
         # ones while preserving the original series order and guards.
         newseries = [
             s
             for s in original_series
             if mq.guard_re.split(s, 1)[0] not in skippedpatches
         ]
         mq.fullseries[:] = newseries
         mq.seriesdirty = True
         mq.savedirty()
 
 
 def storecollapsemsg(repo, collapsemsg):
     """Store the collapse message to allow recovery"""
     collapsemsg = collapsemsg or b''
     f = repo.vfs(b"last-message.txt", b"w")
     f.write(b"%s\n" % collapsemsg)
     f.close()
 
 
 def clearcollapsemsg(repo):
     """Remove collapse message file"""
     repo.vfs.unlinkpath(b"last-message.txt", ignoremissing=True)
 
 
 def restorecollapsemsg(repo, isabort):
     """Restore previously stored collapse message"""
     try:
         f = repo.vfs(b"last-message.txt")
         collapsemsg = f.readline().strip()
         f.close()
     except IOError as err:
         if err.errno != errno.ENOENT:
             raise
         if isabort:
             # Oh well, just abort like normal
             collapsemsg = b''
         else:
             raise error.Abort(_(b'missing .hg/last-message.txt for rebase'))
     return collapsemsg
 
 
 def clearstatus(repo):
     """Remove the status files"""
     # Make sure the active transaction won't write the state file
     tr = repo.currenttransaction()
     if tr:
         tr.removefilegenerator(b'rebasestate')
     repo.vfs.unlinkpath(b"rebasestate", ignoremissing=True)
 
 
 def sortsource(destmap):
     """yield source revisions in an order that we only rebase things once
 
     If source and destination overlaps, we should filter out revisions
     depending on other revisions which hasn't been rebased yet.
 
     Yield a sorted list of revisions each time.
 
     For example, when rebasing A to B, B to C. This function yields [B], then
     [A], indicating B needs to be rebased first.
 
     Raise if there is a cycle so the rebase is impossible.
     """
     srcset = set(destmap)
     while srcset:
         srclist = sorted(srcset)
         result = []
         for r in srclist:
             if destmap[r] not in srcset:
                 result.append(r)
         if not result:
             raise error.Abort(_(b'source and destination form a cycle'))
         srcset -= set(result)
         yield result
 
 
 def buildstate(repo, destmap, collapse):
     '''Define which revisions are going to be rebased and where
 
     repo: repo
     destmap: {srcrev: destrev}
     '''
     rebaseset = destmap.keys()
     originalwd = repo[b'.'].rev()
 
     # This check isn't strictly necessary, since mq detects commits over an
     # applied patch. But it prevents messing up the working directory when
     # a partially completed rebase is blocked by mq.
     if b'qtip' in repo.tags():
         mqapplied = set(repo[s.node].rev() for s in repo.mq.applied)
         if set(destmap.values()) & mqapplied:
             raise error.Abort(_(b'cannot rebase onto an applied mq patch'))
 
     # Get "cycle" error early by exhausting the generator.
     sortedsrc = list(sortsource(destmap))  # a list of sorted revs
     if not sortedsrc:
         raise error.Abort(_(b'no matching revisions'))
 
     # Only check the first batch of revisions to rebase not depending on other
     # rebaseset. This means "source is ancestor of destination" for the second
     # (and following) batches of revisions are not checked here. We rely on
     # "defineparents" to do that check.
     roots = list(repo.set(b'roots(%ld)', sortedsrc[0]))
     if not roots:
         raise error.Abort(_(b'no matching revisions'))
 
     def revof(r):
         return r.rev()
 
     roots = sorted(roots, key=revof)
     state = dict.fromkeys(rebaseset, revtodo)
     emptyrebase = len(sortedsrc) == 1
     for root in roots:
         dest = repo[destmap[root.rev()]]
         commonbase = root.ancestor(dest)
         if commonbase == root:
             raise error.Abort(_(b'source is ancestor of destination'))
         if commonbase == dest:
             wctx = repo[None]
             if dest == wctx.p1():
                 # when rebasing to '.', it will use the current wd branch name
                 samebranch = root.branch() == wctx.branch()
             else:
                 samebranch = root.branch() == dest.branch()
             if not collapse and samebranch and dest in root.parents():
                 # mark the revision as done by setting its new revision
                 # equal to its old (current) revisions
                 state[root.rev()] = root.rev()
                 repo.ui.debug(b'source is a child of destination\n')
                 continue
 
         emptyrebase = False
         repo.ui.debug(b'rebase onto %s starting from %s\n' % (dest, root))
     if emptyrebase:
         return None
     for rev in sorted(state):
         parents = [p for p in repo.changelog.parentrevs(rev) if p != nullrev]
         # if all parents of this revision are done, then so is this revision
         if parents and all((state.get(p) == p for p in parents)):
             state[rev] = rev
     return originalwd, destmap, state
 
 
 def clearrebased(
     ui,
     repo,
     destmap,
     state,
     skipped,
     collapsedas=None,
     keepf=False,
     fm=None,
     backup=True,
 ):
     """dispose of rebased revision at the end of the rebase
 
     If `collapsedas` is not None, the rebase was a collapse whose result if the
     `collapsedas` node.
 
     If `keepf` is not True, the rebase has --keep set and no nodes should be
     removed (but bookmarks still need to be moved).
 
     If `backup` is False, no backup will be stored when stripping rebased
     revisions.
     """
     tonode = repo.changelog.node
     replacements = {}
     moves = {}
     stripcleanup = not obsolete.isenabled(repo, obsolete.createmarkersopt)
 
     collapsednodes = []
     for rev, newrev in sorted(state.items()):
         if newrev >= 0 and newrev != rev:
             oldnode = tonode(rev)
             newnode = collapsedas or tonode(newrev)
             moves[oldnode] = newnode
             succs = None
             if rev in skipped:
                 if stripcleanup or not repo[rev].obsolete():
                     succs = ()
             elif collapsedas:
                 collapsednodes.append(oldnode)
             else:
                 succs = (newnode,)
             if succs is not None:
                 replacements[(oldnode,)] = succs
     if collapsednodes:
         replacements[tuple(collapsednodes)] = (collapsedas,)
     if fm:
         hf = fm.hexfunc
         fl = fm.formatlist
         fd = fm.formatdict
         changes = {}
         for oldns, newn in pycompat.iteritems(replacements):
             for oldn in oldns:
                 changes[hf(oldn)] = fl([hf(n) for n in newn], name=b'node')
         nodechanges = fd(changes, key=b"oldnode", value=b"newnodes")
         fm.data(nodechanges=nodechanges)
     if keepf:
         replacements = {}
     scmutil.cleanupnodes(repo, replacements, b'rebase', moves, backup=backup)
 
 
 def pullrebase(orig, ui, repo, *args, **opts):
     """Call rebase after pull if the latter has been invoked with --rebase"""
     if opts.get('rebase'):
         if ui.configbool(b'commands', b'rebase.requiredest'):
             msg = _(b'rebase destination required by configuration')
             hint = _(b'use hg pull followed by hg rebase -d DEST')
             raise error.Abort(msg, hint=hint)
 
         with repo.wlock(), repo.lock():
             if opts.get('update'):
                 del opts['update']
                 ui.debug(
                     b'--update and --rebase are not compatible, ignoring '
                     b'the update flag\n'
                 )
 
             cmdutil.checkunfinished(repo, skipmerge=True)
             cmdutil.bailifchanged(
                 repo,
                 hint=_(
                     b'cannot pull with rebase: '
                     b'please commit or shelve your changes first'
                 ),
             )
 
             revsprepull = len(repo)
             origpostincoming = commands.postincoming
 
             def _dummy(*args, **kwargs):
                 pass
 
             commands.postincoming = _dummy
             try:
                 ret = orig(ui, repo, *args, **opts)
             finally:
                 commands.postincoming = origpostincoming
             revspostpull = len(repo)
             if revspostpull > revsprepull:
                 # --rev option from pull conflict with rebase own --rev
                 # dropping it
                 if 'rev' in opts:
                     del opts['rev']
                 # positional argument from pull conflicts with rebase's own
                 # --source.
                 if 'source' in opts:
                     del opts['source']
                 # revsprepull is the len of the repo, not revnum of tip.
                 destspace = list(repo.changelog.revs(start=revsprepull))
                 opts['_destspace'] = destspace
                 try:
                     rebase(ui, repo, **opts)
                 except error.NoMergeDestAbort:
                     # we can maybe update instead
                     rev, _a, _b = destutil.destupdate(repo)
                     if rev == repo[b'.'].rev():
                         ui.status(_(b'nothing to rebase\n'))
                     else:
                         ui.status(_(b'nothing to rebase - updating instead\n'))
                         # not passing argument to get the bare update behavior
                         # with warning and trumpets
                         commands.update(ui, repo)
     else:
         if opts.get('tool'):
             raise error.Abort(_(b'--tool can only be used with --rebase'))
         ret = orig(ui, repo, *args, **opts)
 
     return ret
 
 
 def _filterobsoleterevs(repo, revs):
     """returns a set of the obsolete revisions in revs"""
     return set(r for r in revs if repo[r].obsolete())
 
 
 def _computeobsoletenotrebased(repo, rebaseobsrevs, destmap):
     """Return (obsoletenotrebased, obsoletewithoutsuccessorindestination).
 
     `obsoletenotrebased` is a mapping mapping obsolete => successor for all
     obsolete nodes to be rebased given in `rebaseobsrevs`.
 
     `obsoletewithoutsuccessorindestination` is a set with obsolete revisions
     without a successor in destination.
 
     `obsoleteextinctsuccessors` is a set of obsolete revisions with only
     obsolete successors.
     """
     obsoletenotrebased = {}
     obsoletewithoutsuccessorindestination = set()
     obsoleteextinctsuccessors = set()
 
     assert repo.filtername is None
     cl = repo.changelog
     get_rev = cl.index.get_rev
     extinctrevs = set(repo.revs(b'extinct()'))
     for srcrev in rebaseobsrevs:
         srcnode = cl.node(srcrev)
         # XXX: more advanced APIs are required to handle split correctly
         successors = set(obsutil.allsuccessors(repo.obsstore, [srcnode]))
         # obsutil.allsuccessors includes node itself
         successors.remove(srcnode)
         succrevs = {get_rev(s) for s in successors}
         succrevs.discard(None)
         if succrevs.issubset(extinctrevs):
             # all successors are extinct
             obsoleteextinctsuccessors.add(srcrev)
         if not successors:
             # no successor
             obsoletenotrebased[srcrev] = None
         else:
             dstrev = destmap[srcrev]
             for succrev in succrevs:
                 if cl.isancestorrev(succrev, dstrev):
                     obsoletenotrebased[srcrev] = succrev
                     break
             else:
                 # If 'srcrev' has a successor in rebase set but none in
                 # destination (which would be catched above), we shall skip it
                 # and its descendants to avoid divergence.
                 if srcrev in extinctrevs or any(s in destmap for s in succrevs):
                     obsoletewithoutsuccessorindestination.add(srcrev)
 
     return (
         obsoletenotrebased,
         obsoletewithoutsuccessorindestination,
         obsoleteextinctsuccessors,
     )
 
 
 def abortrebase(ui, repo):
     with repo.wlock(), repo.lock():
         rbsrt = rebaseruntime(repo, ui)
         rbsrt._prepareabortorcontinue(isabort=True)
 
 
 def continuerebase(ui, repo):
     with repo.wlock(), repo.lock():
         rbsrt = rebaseruntime(repo, ui)
         ms = mergemod.mergestate.read(repo)
         mergeutil.checkunresolved(ms)
         retcode = rbsrt._prepareabortorcontinue(isabort=False)
         if retcode is not None:
             return retcode
         rbsrt._performrebase(None)
         rbsrt._finishrebase()
 
 
 def summaryhook(ui, repo):
     if not repo.vfs.exists(b'rebasestate'):
         return
     try:
         rbsrt = rebaseruntime(repo, ui, {})
         rbsrt.restorestatus()
         state = rbsrt.state
     except error.RepoLookupError:
         # i18n: column positioning for "hg summary"
         msg = _(b'rebase: (use "hg rebase --abort" to clear broken state)\n')
         ui.write(msg)
         return
     numrebased = len([i for i in pycompat.itervalues(state) if i >= 0])
     # i18n: column positioning for "hg summary"
     ui.write(
         _(b'rebase: %s, %s (rebase --continue)\n')
         % (
             ui.label(_(b'%d rebased'), b'rebase.rebased') % numrebased,
             ui.label(_(b'%d remaining'), b'rebase.remaining')
             % (len(state) - numrebased),
         )
     )
 
 
 def uisetup(ui):
     # Replace pull with a decorator to provide --rebase option
     entry = extensions.wrapcommand(commands.table, b'pull', pullrebase)
     entry[1].append(
         (b'', b'rebase', None, _(b"rebase working directory to branch head"))
     )
     entry[1].append((b't', b'tool', b'', _(b"specify merge tool for rebase")))
     cmdutil.summaryhooks.add(b'rebase', summaryhook)
     statemod.addunfinished(
         b'rebase',
         fname=b'rebasestate',
         stopflag=True,
         continueflag=True,
         abortfunc=abortrebase,
         continuefunc=continuerebase,
     )

mercurial/copies.py

@@ -1,1130 +1,1111 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 multiprocessing
 import os
 
 from .i18n import _
 
 
 from .revlogutils.flagutil import REVIDX_SIDEDATA
 
 from . import (
     error,
     match as matchmod,
     node,
     pathutil,
     pycompat,
     util,
 )
 
 from .revlogutils import sidedata as sidedatamod
 
 from .utils import stringutil
 
 
 def _filter(src, dst, t):
     """filters out invalid copies after chaining"""
 
     # When _chain()'ing copies in 'a' (from 'src' via some other commit 'mid')
     # with copies in 'b' (from 'mid' to 'dst'), we can get the different cases
     # in the following table (not including trivial cases). For example, case 2
     # is where a file existed in 'src' and remained under that name in 'mid' and
     # then was renamed between 'mid' and 'dst'.
     #
     # case src mid dst result
     #   1   x   y   -    -
     #   2   x   y   y   x->y
     #   3   x   y   x    -
     #   4   x   y   z   x->z
     #   5   -   x   y    -
     #   6   x   x   y   x->y
     #
     # _chain() takes care of chaining the copies in 'a' and 'b', but it
     # cannot tell the difference between cases 1 and 2, between 3 and 4, or
     # between 5 and 6, so it includes all cases in its result.
     # Cases 1, 3, and 5 are then removed by _filter().
 
     for k, v in list(t.items()):
         # remove copies from files that didn't exist
         if v not in src:
             del t[k]
         # remove criss-crossed copies
         elif k in src and v in dst:
             del t[k]
         # remove copies to files that were then removed
         elif k not in dst:
             del t[k]
 
 
 def _chain(prefix, suffix):
     """chain two sets of copies 'prefix' and 'suffix'"""
     result = prefix.copy()
     for key, value in pycompat.iteritems(suffix):
         result[key] = prefix.get(value, value)
     return result
 
 
 def _tracefile(fctx, am, basemf):
     """return file context that is the ancestor of fctx present in ancestor
     manifest am
 
     Note: we used to try and stop after a given limit, however checking if that
     limit is reached turned out to be very expensive. we are better off
     disabling that feature."""
 
     for f in fctx.ancestors():
         path = f.path()
         if am.get(path, None) == f.filenode():
             return path
         if basemf and basemf.get(path, None) == f.filenode():
             return path
 
 
 def _dirstatecopies(repo, match=None):
     ds = repo.dirstate
     c = ds.copies().copy()
     for k in list(c):
         if ds[k] not in b'anm' or (match and not match(k)):
             del c[k]
     return c
 
 
 def _computeforwardmissing(a, b, match=None):
     """Computes which files are in b but not a.
     This is its own function so extensions can easily wrap this call to see what
     files _forwardcopies is about to process.
     """
     ma = a.manifest()
     mb = b.manifest()
     return mb.filesnotin(ma, match=match)
 
 
 def usechangesetcentricalgo(repo):
     """Checks if we should use changeset-centric copy algorithms"""
     if repo.filecopiesmode == b'changeset-sidedata':
         return True
     readfrom = repo.ui.config(b'experimental', b'copies.read-from')
     changesetsource = (b'changeset-only', b'compatibility')
     return readfrom in changesetsource
 
 
 def _committedforwardcopies(a, b, base, match):
     """Like _forwardcopies(), but b.rev() cannot be None (working copy)"""
     # files might have to be traced back to the fctx parent of the last
     # one-side-only changeset, but not further back than that
     repo = a._repo
 
     if usechangesetcentricalgo(repo):
         return _changesetforwardcopies(a, b, match)
 
     debug = repo.ui.debugflag and repo.ui.configbool(b'devel', b'debug.copies')
     dbg = repo.ui.debug
     if debug:
         dbg(b'debug.copies:    looking into rename from %s to %s\n' % (a, b))
     am = a.manifest()
     basemf = None if base is None else base.manifest()
 
     # find where new files came from
     # we currently don't try to find where old files went, too expensive
     # this means we can miss a case like 'hg rm b; hg cp a b'
     cm = {}
 
     # Computing the forward missing is quite expensive on large manifests, since
     # it compares the entire manifests. We can optimize it in the common use
     # case of computing what copies are in a commit versus its parent (like
     # during a rebase or histedit). Note, we exclude merge commits from this
     # optimization, since the ctx.files() for a merge commit is not correct for
     # this comparison.
     forwardmissingmatch = match
     if b.p1() == a and b.p2().node() == node.nullid:
         filesmatcher = matchmod.exact(b.files())
         forwardmissingmatch = matchmod.intersectmatchers(match, filesmatcher)
     missing = _computeforwardmissing(a, b, match=forwardmissingmatch)
 
     ancestrycontext = a._repo.changelog.ancestors([b.rev()], inclusive=True)
 
     if debug:
         dbg(b'debug.copies:      missing files to search: %d\n' % len(missing))
 
     for f in sorted(missing):
         if debug:
             dbg(b'debug.copies:        tracing file: %s\n' % f)
         fctx = b[f]
         fctx._ancestrycontext = ancestrycontext
 
         if debug:
             start = util.timer()
         opath = _tracefile(fctx, am, basemf)
         if opath:
             if debug:
                 dbg(b'debug.copies:          rename of: %s\n' % opath)
             cm[f] = opath
         if debug:
             dbg(
                 b'debug.copies:          time: %f seconds\n'
                 % (util.timer() - start)
             )
     return cm
 
 
 def _revinfogetter(repo):
     """return a function that return multiple data given a <rev>"i
 
     * p1: revision number of first parent
     * p2: revision number of first parent
     * p1copies: mapping of copies from p1
     * p2copies: mapping of copies from p2
     * removed: a list of removed files
     """
     cl = repo.changelog
     parents = cl.parentrevs
 
     if repo.filecopiesmode == b'changeset-sidedata':
         changelogrevision = cl.changelogrevision
         flags = cl.flags
 
         # A small cache to avoid doing the work twice for merges
         #
         # In the vast majority of cases, if we ask information for a revision
         # about 1 parent, we'll later ask it for the other. So it make sense to
         # keep the information around when reaching the first parent of a merge
         # and dropping it after it was provided for the second parents.
         #
         # It exists cases were only one parent of the merge will be walked. It
         # happens when the "destination" the copy tracing is descendant from a
         # new root, not common with the "source". In that case, we will only walk
         # through merge parents that are descendant of changesets common
         # between "source" and "destination".
         #
         # With the current case implementation if such changesets have a copy
         # information, we'll keep them in memory until the end of
         # _changesetforwardcopies. We don't expect the case to be frequent
         # enough to matters.
         #
         # In addition, it would be possible to reach pathological case, were
         # many first parent are met before any second parent is reached. In
         # that case the cache could grow. If this even become an issue one can
         # safely introduce a maximum cache size. This would trade extra CPU/IO
         # time to save memory.
         merge_caches = {}
 
         def revinfo(rev):
             p1, p2 = parents(rev)
             if flags(rev) & REVIDX_SIDEDATA:
                 e = merge_caches.pop(rev, None)
                 if e is not None:
                     return e
                 c = changelogrevision(rev)
                 p1copies = c.p1copies
                 p2copies = c.p2copies
                 removed = c.filesremoved
                 if p1 != node.nullrev and p2 != node.nullrev:
                     # XXX some case we over cache, IGNORE
                     merge_caches[rev] = (p1, p2, p1copies, p2copies, removed)
             else:
                 p1copies = {}
                 p2copies = {}
                 removed = []
             return p1, p2, p1copies, p2copies, removed
 
     else:
 
         def revinfo(rev):
             p1, p2 = parents(rev)
             ctx = repo[rev]
             p1copies, p2copies = ctx._copies
             removed = ctx.filesremoved()
             return p1, p2, p1copies, p2copies, removed
 
     return revinfo
 
 
 def _changesetforwardcopies(a, b, match):
     if a.rev() in (node.nullrev, b.rev()):
         return {}
 
     repo = a.repo().unfiltered()
     children = {}
     revinfo = _revinfogetter(repo)
 
     cl = repo.changelog
     missingrevs = cl.findmissingrevs(common=[a.rev()], heads=[b.rev()])
     mrset = set(missingrevs)
     roots = set()
     for r in missingrevs:
         for p in cl.parentrevs(r):
             if p == node.nullrev:
                 continue
             if p not in children:
                 children[p] = [r]
             else:
                 children[p].append(r)
             if p not in mrset:
                 roots.add(p)
     if not roots:
         # no common revision to track copies from
         return {}
     min_root = min(roots)
 
     from_head = set(
         cl.reachableroots(min_root, [b.rev()], list(roots), includepath=True)
     )
 
     iterrevs = set(from_head)
     iterrevs &= mrset
     iterrevs.update(roots)
     iterrevs.remove(b.rev())
     revs = sorted(iterrevs)
     return _combinechangesetcopies(revs, children, b.rev(), revinfo, match)
 
 
 def _combinechangesetcopies(revs, children, targetrev, revinfo, match):
     """combine the copies information for each item of iterrevs
 
     revs: sorted iterable of revision to visit
     children: a {parent: [children]} mapping.
     targetrev: the final copies destination revision (not in iterrevs)
     revinfo(rev): a function that return (p1, p2, p1copies, p2copies, removed)
     match: a matcher
 
     It returns the aggregated copies information for `targetrev`.
     """
     all_copies = {}
     alwaysmatch = match.always()
     for r in revs:
         copies = all_copies.pop(r, None)
         if copies is None:
             # this is a root
             copies = {}
         for i, c in enumerate(children[r]):
             p1, p2, p1copies, p2copies, removed = revinfo(c)
             if r == p1:
                 parent = 1
                 childcopies = p1copies
             else:
                 assert r == p2
                 parent = 2
                 childcopies = p2copies
             if not alwaysmatch:
                 childcopies = {
                     dst: src for dst, src in childcopies.items() if match(dst)
                 }
             newcopies = copies
             if childcopies:
                 newcopies = _chain(newcopies, childcopies)
                 # _chain makes a copies, we can avoid doing so in some
                 # simple/linear cases.
                 assert newcopies is not copies
             for f in removed:
                 if f in newcopies:
                     if newcopies is copies:
                         # copy on write to avoid affecting potential other
                         # branches.  when there are no other branches, this
                         # could be avoided.
                         newcopies = copies.copy()
                     del newcopies[f]
             othercopies = all_copies.get(c)
             if othercopies is None:
                 all_copies[c] = newcopies
             else:
                 # we are the second parent to work on c, we need to merge our
                 # work with the other.
                 #
                 # Unlike when copies are stored in the filelog, we consider
                 # it a copy even if the destination already existed on the
                 # other branch. It's simply too expensive to check if the
                 # file existed in the manifest.
                 #
                 # In case of conflict, parent 1 take precedence over parent 2.
                 # This is an arbitrary choice made anew when implementing
                 # changeset based copies. It was made without regards with
                 # potential filelog related behavior.
                 if parent == 1:
                     othercopies.update(newcopies)
                 else:
                     newcopies.update(othercopies)
                     all_copies[c] = newcopies
     return all_copies[targetrev]
 
 
 def _forwardcopies(a, b, base=None, match=None):
     """find {dst@b: src@a} copy mapping where a is an ancestor of b"""
 
     if base is None:
         base = a
     match = a.repo().narrowmatch(match)
     # check for working copy
     if b.rev() is None:
         cm = _committedforwardcopies(a, b.p1(), base, match)
         # combine copies from dirstate if necessary
         copies = _chain(cm, _dirstatecopies(b._repo, match))
     else:
         copies = _committedforwardcopies(a, b, base, match)
     return copies
 
 
 def _backwardrenames(a, b, match):
     if a._repo.ui.config(b'experimental', b'copytrace') == b'off':
         return {}
 
     # Even though we're not taking copies into account, 1:n rename situations
     # can still exist (e.g. hg cp a b; hg mv a c). In those cases we
     # arbitrarily pick one of the renames.
     # We don't want to pass in "match" here, since that would filter
     # the destination by it. Since we're reversing the copies, we want
     # to filter the source instead.
     f = _forwardcopies(b, a)
     r = {}
     for k, v in sorted(pycompat.iteritems(f)):
         if match and not match(v):
             continue
         # remove copies
         if v in a:
             continue
         r[v] = k
     return r
 
 
 def pathcopies(x, y, match=None):
     """find {dst@y: src@x} copy mapping for directed compare"""
     repo = x._repo
     debug = repo.ui.debugflag and repo.ui.configbool(b'devel', b'debug.copies')
     if debug:
         repo.ui.debug(
             b'debug.copies: searching copies from %s to %s\n' % (x, y)
         )
     if x == y or not x or not y:
         return {}
     a = y.ancestor(x)
     if a == x:
         if debug:
             repo.ui.debug(b'debug.copies: search mode: forward\n')
         if y.rev() is None and x == y.p1():
             # short-circuit to avoid issues with merge states
             return _dirstatecopies(repo, match)
         copies = _forwardcopies(x, y, match=match)
     elif a == y:
         if debug:
             repo.ui.debug(b'debug.copies: search mode: backward\n')
         copies = _backwardrenames(x, y, match=match)
     else:
         if debug:
             repo.ui.debug(b'debug.copies: search mode: combined\n')
         base = None
         if a.rev() != node.nullrev:
             base = x
         copies = _chain(
             _backwardrenames(x, a, match=match),
             _forwardcopies(a, y, base, match=match),
         )
     _filter(x, y, copies)
     return copies
 
 
 def mergecopies(repo, c1, c2, base):
     """
     Finds moves and copies between context c1 and c2 that are relevant for
     merging. 'base' will be used as the merge base.
 
     Copytracing is used in commands like rebase, merge, unshelve, etc to merge
     files that were moved/ copied in one merge parent and modified in another.
     For example:
 
     o          ---> 4 another commit
     |
     |   o      ---> 3 commit that modifies a.txt
     |  /
     o /        ---> 2 commit that moves a.txt to b.txt
     |/
     o          ---> 1 merge base
 
     If we try to rebase revision 3 on revision 4, since there is no a.txt in
     revision 4, and if user have copytrace disabled, we prints the following
     message:
 
     ```other changed <file> which local deleted```
 
     Returns five dicts: "copy", "movewithdir", "diverge", "renamedelete" and
     "dirmove".
 
     "copy" is a mapping from destination name -> source name,
     where source is in c1 and destination is in c2 or vice-versa.
 
     "movewithdir" is a mapping from source name -> destination name,
     where the file at source present in one context but not the other
     needs to be moved to destination by the merge process, because the
     other context moved the directory it is in.
 
     "diverge" is a mapping of source name -> list of destination names
     for divergent renames.
 
     "renamedelete" is a mapping of source name -> list of destination
     names for files deleted in c1 that were renamed in c2 or vice-versa.
 
     "dirmove" is a mapping of detected source dir -> destination dir renames.
     This is needed for handling changes to new files previously grafted into
     renamed directories.
 
     This function calls different copytracing algorithms based on config.
     """
     # avoid silly behavior for update from empty dir
     if not c1 or not c2 or c1 == c2:
         return {}, {}, {}, {}, {}
 
     narrowmatch = c1.repo().narrowmatch()
 
     # avoid silly behavior for parent -> working dir
     if c2.node() is None and c1.node() == repo.dirstate.p1():
         return _dirstatecopies(repo, narrowmatch), {}, {}, {}, {}
 
     copytracing = repo.ui.config(b'experimental', b'copytrace')
     if stringutil.parsebool(copytracing) is False:
         # stringutil.parsebool() returns None when it is unable to parse the
         # value, so we should rely on making sure copytracing is on such cases
         return {}, {}, {}, {}, {}
 
     if usechangesetcentricalgo(repo):
         # The heuristics don't make sense when we need changeset-centric algos
         return _fullcopytracing(repo, c1, c2, base)
 
     # Copy trace disabling is explicitly below the node == p1 logic above
     # because the logic above is required for a simple copy to be kept across a
     # rebase.
     if copytracing == b'heuristics':
         # Do full copytracing if only non-public revisions are involved as
         # that will be fast enough and will also cover the copies which could
         # be missed by heuristics
         if _isfullcopytraceable(repo, c1, base):
             return _fullcopytracing(repo, c1, c2, base)
         return _heuristicscopytracing(repo, c1, c2, base)
     else:
         return _fullcopytracing(repo, c1, c2, base)
 
 
 def _isfullcopytraceable(repo, c1, base):
     """ Checks that if base, source and destination are all no-public branches,
     if yes let's use the full copytrace algorithm for increased capabilities
     since it will be fast enough.
 
     `experimental.copytrace.sourcecommitlimit` can be used to set a limit for
     number of changesets from c1 to base such that if number of changesets are
     more than the limit, full copytracing algorithm won't be used.
     """
     if c1.rev() is None:
         c1 = c1.p1()
     if c1.mutable() and base.mutable():
         sourcecommitlimit = repo.ui.configint(
             b'experimental', b'copytrace.sourcecommitlimit'
         )
         commits = len(repo.revs(b'%d::%d', base.rev(), c1.rev()))
         return commits < sourcecommitlimit
     return False
 
 
 def _checksinglesidecopies(
     src, dsts1, m1, m2, mb, c2, base, copy, renamedelete
 ):
     if src not in m2:
         # deleted on side 2
         if src not in m1:
             # renamed on side 1, deleted on side 2
             renamedelete[src] = dsts1
     elif m2[src] != mb[src]:
         if not _related(c2[src], base[src]):
             return
         # modified on side 2
         for dst in dsts1:
             if dst not in m2:
                 # dst not added on side 2 (handle as regular
                 # "both created" case in manifestmerge otherwise)
                 copy[dst] = src
 
 
 def _fullcopytracing(repo, c1, c2, base):
     """ The full copytracing algorithm which finds all the new files that were
     added from merge base up to the top commit and for each file it checks if
     this file was copied from another file.
 
     This is pretty slow when a lot of changesets are involved but will track all
     the copies.
     """
     m1 = c1.manifest()
     m2 = c2.manifest()
     mb = base.manifest()
 
     copies1 = pathcopies(base, c1)
     copies2 = pathcopies(base, c2)
 
     inversecopies1 = {}
     inversecopies2 = {}
     for dst, src in copies1.items():
         inversecopies1.setdefault(src, []).append(dst)
     for dst, src in copies2.items():
         inversecopies2.setdefault(src, []).append(dst)
 
     copy = {}
     diverge = {}
     renamedelete = {}
     allsources = set(inversecopies1) | set(inversecopies2)
     for src in allsources:
         dsts1 = inversecopies1.get(src)
         dsts2 = inversecopies2.get(src)
         if dsts1 and dsts2:
             # copied/renamed on both sides
             if src not in m1 and src not in m2:
                 # renamed on both sides
                 dsts1 = set(dsts1)
                 dsts2 = set(dsts2)
                 # If there's some overlap in the rename destinations, we
                 # consider it not divergent. For example, if side 1 copies 'a'
                 # to 'b' and 'c' and deletes 'a', and side 2 copies 'a' to 'c'
                 # and 'd' and deletes 'a'.
                 if dsts1 & dsts2:
                     for dst in dsts1 & dsts2:
                         copy[dst] = src
                 else:
                     diverge[src] = sorted(dsts1 | dsts2)
             elif src in m1 and src in m2:
                 # copied on both sides
                 dsts1 = set(dsts1)
                 dsts2 = set(dsts2)
                 for dst in dsts1 & dsts2:
                     copy[dst] = src
             # TODO: Handle cases where it was renamed on one side and copied
             # on the other side
         elif dsts1:
             # copied/renamed only on side 1
             _checksinglesidecopies(
                 src, dsts1, m1, m2, mb, c2, base, copy, renamedelete
             )
         elif dsts2:
             # copied/renamed only on side 2
             _checksinglesidecopies(
                 src, dsts2, m2, m1, mb, c1, base, copy, renamedelete
             )
 
     renamedeleteset = set()
     divergeset = set()
     for dsts in diverge.values():
         divergeset.update(dsts)
     for dsts in renamedelete.values():
         renamedeleteset.update(dsts)
 
     # find interesting file sets from manifests
     addedinm1 = m1.filesnotin(mb, repo.narrowmatch())
     addedinm2 = m2.filesnotin(mb, repo.narrowmatch())
     u1 = sorted(addedinm1 - addedinm2)
     u2 = sorted(addedinm2 - addedinm1)
 
     header = b"  unmatched files in %s"
     if u1:
         repo.ui.debug(b"%s:\n   %s\n" % (header % b'local', b"\n   ".join(u1)))
     if u2:
         repo.ui.debug(b"%s:\n   %s\n" % (header % b'other', b"\n   ".join(u2)))
 
     fullcopy = copies1.copy()
     fullcopy.update(copies2)
     if not fullcopy:
         return copy, {}, diverge, renamedelete, {}
 
     if repo.ui.debugflag:
         repo.ui.debug(
             b"  all copies found (* = to merge, ! = divergent, "
             b"% = renamed and deleted):\n"
         )
         for f in sorted(fullcopy):
             note = b""
             if f in copy:
                 note += b"*"
             if f in divergeset:
                 note += b"!"
             if f in renamedeleteset:
                 note += b"%"
             repo.ui.debug(
                 b"   src: '%s' -> dst: '%s' %s\n" % (fullcopy[f], f, note)
             )
     del divergeset
 
     repo.ui.debug(b"  checking for directory renames\n")
 
     # generate a directory move map
     d1, d2 = c1.dirs(), c2.dirs()
     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 pycompat.iteritems(fullcopy):
         dsrc, ddst = pathutil.dirname(src), pathutil.dirname(dst)
         if dsrc in invalid:
             # already seen to be uninteresting
             continue
         elif dsrc in d1 and ddst in d1:
             # directory wasn't entirely moved locally
             invalid.add(dsrc)
         elif dsrc in d2 and ddst in d2:
             # directory wasn't entirely moved remotely
             invalid.add(dsrc)
         elif dsrc in dirmove and dirmove[dsrc] != ddst:
             # files from the same directory moved to two different places
             invalid.add(dsrc)
         else:
             # looks good so far
             dirmove[dsrc] = ddst
 
     for i in invalid:
         if i in dirmove:
             del dirmove[i]
     del d1, d2, invalid
 
     if not dirmove:
         return copy, {}, diverge, renamedelete, {}
 
     dirmove = {k + b"/": v + b"/" for k, v in pycompat.iteritems(dirmove)}
 
     for d in dirmove:
         repo.ui.debug(
             b"   discovered dir src: '%s' -> dst: '%s'\n" % (d, dirmove[d])
         )
 
     movewithdir = {}
     # check unaccounted nonoverlapping files against directory moves
     for f in u1 + u2:
         if f not in fullcopy:
             for d in dirmove:
                 if f.startswith(d):
                     # new file added in a directory that was moved, move it
                     df = dirmove[d] + f[len(d) :]
                     if df not in copy:
                         movewithdir[f] = df
                         repo.ui.debug(
                             b"   pending file src: '%s' -> dst: '%s'\n"
                             % (f, df)
                         )
                     break
 
     return copy, movewithdir, diverge, renamedelete, dirmove
 
 
 def _heuristicscopytracing(repo, c1, c2, base):
     """ Fast copytracing using filename heuristics
 
     Assumes that moves or renames are of following two types:
 
     1) Inside a directory only (same directory name but different filenames)
     2) Move from one directory to another
                     (same filenames but different directory names)
 
     Works only when there are no merge commits in the "source branch".
     Source branch is commits from base up to c2 not including base.
 
     If merge is involved it fallbacks to _fullcopytracing().
 
     Can be used by setting the following config:
 
         [experimental]
         copytrace = heuristics
 
     In some cases the copy/move candidates found by heuristics can be very large
     in number and that will make the algorithm slow. The number of possible
     candidates to check can be limited by using the config
     `experimental.copytrace.movecandidateslimit` which defaults to 100.
     """
 
     if c1.rev() is None:
         c1 = c1.p1()
     if c2.rev() is None:
         c2 = c2.p1()
 
     copies = {}
 
     changedfiles = set()
     m1 = c1.manifest()
     if not repo.revs(b'%d::%d', base.rev(), c2.rev()):
         # If base is not in c2 branch, we switch to fullcopytracing
         repo.ui.debug(
             b"switching to full copytracing as base is not "
             b"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(b"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 pycompat.iteritems(cp):
         if src in m1:
             copies[dst] = src
 
     # file is missing if it isn't present in the destination, but is present in
     # the base and present in the source.
     # Presence in the base is important to exclude added files, presence in the
     # source is important to exclude removed files.
     filt = lambda f: f not in m1 and f in base and f in c2
     missingfiles = [f for f in changedfiles if filt(f)]
 
     if missingfiles:
         basenametofilename = collections.defaultdict(list)
         dirnametofilename = collections.defaultdict(list)
 
         for f in m1.filesnotin(base.manifest()):
             basename = os.path.basename(f)
             dirname = os.path.dirname(f)
             basenametofilename[basename].append(f)
             dirnametofilename[dirname].append(f)
 
         for f in missingfiles:
             basename = os.path.basename(f)
             dirname = os.path.dirname(f)
             samebasename = basenametofilename[basename]
             samedirname = dirnametofilename[dirname]
             movecandidates = samebasename + samedirname
             # f is guaranteed to be present in c2, that's why
             # c2.filectx(f) won't fail
             f2 = c2.filectx(f)
             # we can have a lot of candidates which can slow down the heuristics
             # config value to limit the number of candidates moves to check
             maxcandidates = repo.ui.configint(
                 b'experimental', b'copytrace.movecandidateslimit'
             )
 
             if len(movecandidates) > maxcandidates:
                 repo.ui.status(
                     _(
                         b"skipping copytracing for '%s', more "
                         b"candidates than the limit: %d\n"
                     )
                     % (f, len(movecandidates))
                 )
                 continue
 
             for candidate in movecandidates:
                 f1 = c1.filectx(candidate)
                 if _related(f1, f2):
                     # if there are a few related copies then we'll merge
                     # changes into all of them. This matches the behaviour
                     # of upstream copytracing
                     copies[candidate] = f
 
     return copies, {}, {}, {}, {}
 
 
 def _related(f1, f2):
     """return True if f1 and f2 filectx have a common ancestor
 
     Walk back to common ancestor to see if the two files originate
     from the same file. Since workingfilectx's rev() is None it messes
     up the integer comparison logic, hence the pre-step check for
     None (f1 and f2 can only be workingfilectx's initially).
     """
 
     if f1 == f2:
         return True  # a match
 
     g1, g2 = f1.ancestors(), f2.ancestors()
     try:
         f1r, f2r = f1.linkrev(), f2.linkrev()
 
         if f1r is None:
             f1 = next(g1)
         if f2r is None:
             f2 = next(g2)
 
         while True:
             f1r, f2r = f1.linkrev(), f2.linkrev()
             if f1r > f2r:
                 f1 = next(g1)
             elif f2r > f1r:
                 f2 = next(g2)
             else:  # f1 and f2 point to files in the same linkrev
                 return f1 == f2  # true if they point to the same file
     except StopIteration:
         return False
 
 
-def graftcopies(repo, wctx, ctx, base, skip=None):
-    """reproduce copies between base and ctx in the wctx
-
-    If skip is specified, it's a revision that should be used to
-    filter copy records. Any copies that occur between base and
-    skip will not be duplicated, even if they appear in the set of
-    copies between base and ctx.
-    """
-    exclude = {}
-    ctraceconfig = repo.ui.config(b'experimental', b'copytrace')
-    bctrace = stringutil.parsebool(ctraceconfig)
-    if skip is not None and (
-        ctraceconfig == b'heuristics' or bctrace or bctrace is None
-    ):
-        # copytrace='off' skips this line, but not the entire function because
-        # the line below is O(size of the repo) during a rebase, while the rest
-        # of the function is much faster (and is required for carrying copy
-        # metadata across the rebase anyway).
-        exclude = pathcopies(base, skip)
+def graftcopies(wctx, ctx, base):
+    """reproduce copies between base and ctx in the wctx"""
     new_copies = pathcopies(base, ctx)
     _filter(wctx.p1(), wctx, new_copies)
     for dst, src in pycompat.iteritems(new_copies):
-        if dst in exclude:
-            continue
         wctx[dst].markcopied(src)
 
 
 def computechangesetfilesadded(ctx):
     """return the list of files added in a changeset
     """
     added = []
     for f in ctx.files():
         if not any(f in p for p in ctx.parents()):
             added.append(f)
     return added
 
 
 def computechangesetfilesremoved(ctx):
     """return the list of files removed in a changeset
     """
     removed = []
     for f in ctx.files():
         if f not in ctx:
             removed.append(f)
     return removed
 
 
 def computechangesetcopies(ctx):
     """return the copies data for a changeset
 
     The copies data are returned as a pair of dictionnary (p1copies, p2copies).
 
     Each dictionnary are in the form: `{newname: oldname}`
     """
     p1copies = {}
     p2copies = {}
     p1 = ctx.p1()
     p2 = ctx.p2()
     narrowmatch = ctx._repo.narrowmatch()
     for dst in ctx.files():
         if not narrowmatch(dst) or dst not in ctx:
             continue
         copied = ctx[dst].renamed()
         if not copied:
             continue
         src, srcnode = copied
         if src in p1 and p1[src].filenode() == srcnode:
             p1copies[dst] = src
         elif src in p2 and p2[src].filenode() == srcnode:
             p2copies[dst] = src
     return p1copies, p2copies
 
 
 def encodecopies(files, copies):
     items = []
     for i, dst in enumerate(files):
         if dst in copies:
             items.append(b'%d\0%s' % (i, copies[dst]))
     if len(items) != len(copies):
         raise error.ProgrammingError(
             b'some copy targets missing from file list'
         )
     return b"\n".join(items)
 
 
 def decodecopies(files, data):
     try:
         copies = {}
         if not data:
             return copies
         for l in data.split(b'\n'):
             strindex, src = l.split(b'\0')
             i = int(strindex)
             dst = files[i]
             copies[dst] = src
         return copies
     except (ValueError, IndexError):
         # Perhaps someone had chosen the same key name (e.g. "p1copies") and
         # used different syntax for the value.
         return None
 
 
 def encodefileindices(files, subset):
     subset = set(subset)
     indices = []
     for i, f in enumerate(files):
         if f in subset:
             indices.append(b'%d' % i)
     return b'\n'.join(indices)
 
 
 def decodefileindices(files, data):
     try:
         subset = []
         if not data:
             return subset
         for strindex in data.split(b'\n'):
             i = int(strindex)
             if i < 0 or i >= len(files):
                 return None
             subset.append(files[i])
         return subset
     except (ValueError, IndexError):
         # Perhaps someone had chosen the same key name (e.g. "added") and
         # used different syntax for the value.
         return None
 
 
 def _getsidedata(srcrepo, rev):
     ctx = srcrepo[rev]
     filescopies = computechangesetcopies(ctx)
     filesadded = computechangesetfilesadded(ctx)
     filesremoved = computechangesetfilesremoved(ctx)
     sidedata = {}
     if any([filescopies, filesadded, filesremoved]):
         sortedfiles = sorted(ctx.files())
         p1copies, p2copies = filescopies
         p1copies = encodecopies(sortedfiles, p1copies)
         p2copies = encodecopies(sortedfiles, p2copies)
         filesadded = encodefileindices(sortedfiles, filesadded)
         filesremoved = encodefileindices(sortedfiles, filesremoved)
         if p1copies:
             sidedata[sidedatamod.SD_P1COPIES] = p1copies
         if p2copies:
             sidedata[sidedatamod.SD_P2COPIES] = p2copies
         if filesadded:
             sidedata[sidedatamod.SD_FILESADDED] = filesadded
         if filesremoved:
             sidedata[sidedatamod.SD_FILESREMOVED] = filesremoved
     return sidedata
 
 
 def getsidedataadder(srcrepo, destrepo):
     use_w = srcrepo.ui.configbool(b'experimental', b'worker.repository-upgrade')
     if pycompat.iswindows or not use_w:
         return _get_simple_sidedata_adder(srcrepo, destrepo)
     else:
         return _get_worker_sidedata_adder(srcrepo, destrepo)
 
 
 def _sidedata_worker(srcrepo, revs_queue, sidedata_queue, tokens):
     """The function used by worker precomputing sidedata
 
     It read an input queue containing revision numbers
     It write in an output queue containing (rev, <sidedata-map>)
 
     The `None` input value is used as a stop signal.
 
     The `tokens` semaphore is user to avoid having too many unprocessed
     entries. The workers needs to acquire one token before fetching a task.
     They will be released by the consumer of the produced data.
     """
     tokens.acquire()
     rev = revs_queue.get()
     while rev is not None:
         data = _getsidedata(srcrepo, rev)
         sidedata_queue.put((rev, data))
         tokens.acquire()
         rev = revs_queue.get()
     # processing of `None` is completed, release the token.
     tokens.release()
 
 
 BUFF_PER_WORKER = 50
 
 
 def _get_worker_sidedata_adder(srcrepo, destrepo):
     """The parallel version of the sidedata computation
 
     This code spawn a pool of worker that precompute a buffer of sidedata
     before we actually need them"""
     # avoid circular import copies -> scmutil -> worker -> copies
     from . import worker
 
     nbworkers = worker._numworkers(srcrepo.ui)
 
     tokens = multiprocessing.BoundedSemaphore(nbworkers * BUFF_PER_WORKER)
     revsq = multiprocessing.Queue()
     sidedataq = multiprocessing.Queue()
 
     assert srcrepo.filtername is None
     # queue all tasks beforehand, revision numbers are small and it make
     # synchronisation simpler
     #
     # Since the computation for each node can be quite expensive, the overhead
     # of using a single queue is not revelant. In practice, most computation
     # are fast but some are very expensive and dominate all the other smaller
     # cost.
     for r in srcrepo.changelog.revs():
         revsq.put(r)
     # queue the "no more tasks" markers
     for i in range(nbworkers):
         revsq.put(None)
 
     allworkers = []
     for i in range(nbworkers):
         args = (srcrepo, revsq, sidedataq, tokens)
         w = multiprocessing.Process(target=_sidedata_worker, args=args)
         allworkers.append(w)
         w.start()
 
     # dictionnary to store results for revision higher than we one we are
     # looking for. For example, if we need the sidedatamap for 42, and 43 is
     # received, when shelve 43 for later use.
     staging = {}
 
     def sidedata_companion(revlog, rev):
         sidedata = {}
         if util.safehasattr(revlog, b'filteredrevs'):  # this is a changelog
             # Is the data previously shelved ?
             sidedata = staging.pop(rev, None)
             if sidedata is None:
                 # look at the queued result until we find the one we are lookig
                 # for (shelve the other ones)
                 r, sidedata = sidedataq.get()
                 while r != rev:
                     staging[r] = sidedata
                     r, sidedata = sidedataq.get()
             tokens.release()
         return False, (), sidedata
 
     return sidedata_companion
 
 
 def _get_simple_sidedata_adder(srcrepo, destrepo):
     """The simple version of the sidedata computation
 
     It just compute it in the same thread on request"""
 
     def sidedatacompanion(revlog, rev):
         sidedata = {}
         if util.safehasattr(revlog, 'filteredrevs'):  # this is a changelog
             sidedata = _getsidedata(srcrepo, rev)
         return False, (), sidedata
 
     return sidedatacompanion
 
 
 def getsidedataremover(srcrepo, destrepo):
     def sidedatacompanion(revlog, rev):
         f = ()
         if util.safehasattr(revlog, 'filteredrevs'):  # this is a changelog
             if revlog.flags(rev) & REVIDX_SIDEDATA:
                 f = (
                     sidedatamod.SD_P1COPIES,
                     sidedatamod.SD_P2COPIES,
                     sidedatamod.SD_FILESADDED,
                     sidedatamod.SD_FILESREMOVED,
                 )
         return False, f, {}
 
     return sidedatacompanion

mercurial/merge.py

@@ -1,2713 +1,2713 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 shutil
 import stat
 import struct
 
 from .i18n import _
 from .node import (
     addednodeid,
     bin,
     hex,
     modifiednodeid,
     nullhex,
     nullid,
     nullrev,
 )
 from .pycompat import delattr
 from .thirdparty import attr
 from . import (
     copies,
     encoding,
     error,
     filemerge,
     match as matchmod,
     obsutil,
     pathutil,
     pycompat,
     scmutil,
     subrepoutil,
     util,
     worker,
 )
 from .utils import hashutil
 
 _pack = struct.pack
 _unpack = struct.unpack
 
 
 def _droponode(data):
     # used for compatibility for v1
     bits = data.split(b'\0')
     bits = bits[:-2] + bits[-1:]
     return b'\0'.join(bits)
 
 
 # Merge state record types. See ``mergestate`` docs for more.
 RECORD_LOCAL = b'L'
 RECORD_OTHER = b'O'
 RECORD_MERGED = b'F'
 RECORD_CHANGEDELETE_CONFLICT = b'C'
 RECORD_MERGE_DRIVER_MERGE = b'D'
 RECORD_PATH_CONFLICT = b'P'
 RECORD_MERGE_DRIVER_STATE = b'm'
 RECORD_FILE_VALUES = b'f'
 RECORD_LABELS = b'l'
 RECORD_OVERRIDE = b't'
 RECORD_UNSUPPORTED_MANDATORY = b'X'
 RECORD_UNSUPPORTED_ADVISORY = b'x'
 
 MERGE_DRIVER_STATE_UNMARKED = b'u'
 MERGE_DRIVER_STATE_MARKED = b'm'
 MERGE_DRIVER_STATE_SUCCESS = b's'
 
 MERGE_RECORD_UNRESOLVED = b'u'
 MERGE_RECORD_RESOLVED = b'r'
 MERGE_RECORD_UNRESOLVED_PATH = b'pu'
 MERGE_RECORD_RESOLVED_PATH = b'pr'
 MERGE_RECORD_DRIVER_RESOLVED = b'd'
 
 ACTION_FORGET = b'f'
 ACTION_REMOVE = b'r'
 ACTION_ADD = b'a'
 ACTION_GET = b'g'
 ACTION_PATH_CONFLICT = b'p'
 ACTION_PATH_CONFLICT_RESOLVE = b'pr'
 ACTION_ADD_MODIFIED = b'am'
 ACTION_CREATED = b'c'
 ACTION_DELETED_CHANGED = b'dc'
 ACTION_CHANGED_DELETED = b'cd'
 ACTION_MERGE = b'm'
 ACTION_LOCAL_DIR_RENAME_GET = b'dg'
 ACTION_DIR_RENAME_MOVE_LOCAL = b'dm'
 ACTION_KEEP = b'k'
 ACTION_EXEC = b'e'
 ACTION_CREATED_MERGE = b'cm'
 
 
 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)
     P: a path conflict (file vs directory)
     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
 
     Merge record states (stored in self._state, indexed by filename):
     u: unresolved conflict
     r: resolved conflict
     pu: unresolved path conflict (file conflicts with directory)
     pr: resolved path conflict
     d: driver-resolved conflict
 
     The resolve command transitions between 'u' and 'r' for conflicts and
     'pu' and 'pr' for path conflicts.
     '''
 
     statepathv1 = b'merge/state'
     statepathv2 = b'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 = MERGE_DRIVER_STATE_SUCCESS
         else:
             self._mdstate = MERGE_DRIVER_STATE_UNMARKED
         shutil.rmtree(self._repo.vfs.join(b'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 = MERGE_DRIVER_STATE_SUCCESS
         unsupported = set()
         records = self._readrecords()
         for rtype, record in records:
             if rtype == RECORD_LOCAL:
                 self._local = bin(record)
             elif rtype == RECORD_OTHER:
                 self._other = bin(record)
             elif rtype == RECORD_MERGE_DRIVER_STATE:
                 bits = record.split(b'\0', 1)
                 mdstate = bits[1]
                 if len(mdstate) != 1 or mdstate not in (
                     MERGE_DRIVER_STATE_UNMARKED,
                     MERGE_DRIVER_STATE_MARKED,
                     MERGE_DRIVER_STATE_SUCCESS,
                 ):
                     # the merge driver should be idempotent, so just rerun it
                     mdstate = MERGE_DRIVER_STATE_UNMARKED
 
                 self._readmergedriver = bits[0]
                 self._mdstate = mdstate
             elif rtype in (
                 RECORD_MERGED,
                 RECORD_CHANGEDELETE_CONFLICT,
                 RECORD_PATH_CONFLICT,
                 RECORD_MERGE_DRIVER_MERGE,
             ):
                 bits = record.split(b'\0')
                 self._state[bits[0]] = bits[1:]
             elif rtype == RECORD_FILE_VALUES:
                 filename, rawextras = record.split(b'\0', 1)
                 extraparts = rawextras.split(b'\0')
                 extras = {}
                 i = 0
                 while i < len(extraparts):
                     extras[extraparts[i]] = extraparts[i + 1]
                     i += 2
 
                 self._stateextras[filename] = extras
             elif rtype == RECORD_LABELS:
                 labels = record.split(b'\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((RECORD_OTHER, 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] == RECORD_MERGED:
                     bits = r[1].split(b'\0')
                     bits.insert(-2, b'')
                     v1records[idx] = (r[0], b'\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] == RECORD_LOCAL:
                 oldv2.add(rec)
             elif rec[0] == RECORD_MERGED:
                 # drop the onode data (not contained in v1)
                 oldv2.add((RECORD_MERGED, _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((RECORD_LOCAL, l[:-1]))
                 else:
                     records.append((RECORD_MERGED, 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]
                 off += 1
                 length = _unpack(b'>I', data[off : (off + 4)])[0]
                 off += 4
                 record = data[off : (off + length)]
                 off += length
                 if rtype == RECORD_OVERRIDE:
                     rtype, record = record[0:1], 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(
             b'experimental', b'mergedriver'
         )
         if (
             self._readmergedriver is not None
             and self._readmergedriver != configmergedriver
         ):
             raise error.ConfigError(
                 _(b"merge driver changed since merge started"),
                 hint=_(b"revert merge driver change or abort merge"),
             )
 
         return configmergedriver
 
     @util.propertycache
     def localctx(self):
         if self._local is None:
             msg = b"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 = b"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((RECORD_LOCAL, hex(self._local)))
         records.append((RECORD_OTHER, hex(self._other)))
         if self.mergedriver:
             records.append(
                 (
                     RECORD_MERGE_DRIVER_STATE,
                     b'\0'.join([self.mergedriver, self._mdstate]),
                 )
             )
         # Write out state items. In all cases, the value of the state map entry
         # is written as the contents of the record. The record type depends on
         # the type of state that is stored, and capital-letter records are used
         # to prevent older versions of Mercurial that do not support the feature
         # from loading them.
         for filename, v in pycompat.iteritems(self._state):
             if v[0] == MERGE_RECORD_DRIVER_RESOLVED:
                 # Driver-resolved merge. These are stored in 'D' records.
                 records.append(
                     (RECORD_MERGE_DRIVER_MERGE, b'\0'.join([filename] + v))
                 )
             elif v[0] in (
                 MERGE_RECORD_UNRESOLVED_PATH,
                 MERGE_RECORD_RESOLVED_PATH,
             ):
                 # Path conflicts. These are stored in 'P' records.  The current
                 # resolution state ('pu' or 'pr') is stored within the record.
                 records.append(
                     (RECORD_PATH_CONFLICT, b'\0'.join([filename] + v))
                 )
             elif v[1] == nullhex or v[6] == nullhex:
                 # Change/Delete or Delete/Change conflicts. These are stored in
                 # 'C' records. v[1] is the local file, and is nullhex when the
                 # file is deleted locally ('dc'). v[6] is the remote file, and
                 # is nullhex when the file is deleted remotely ('cd').
                 records.append(
                     (RECORD_CHANGEDELETE_CONFLICT, b'\0'.join([filename] + v))
                 )
             else:
                 # Normal files.  These are stored in 'F' records.
                 records.append((RECORD_MERGED, b'\0'.join([filename] + v)))
         for filename, extras in sorted(pycompat.iteritems(self._stateextras)):
             rawextras = b'\0'.join(
                 b'%s\0%s' % (k, v) for k, v in pycompat.iteritems(extras)
             )
             records.append(
                 (RECORD_FILE_VALUES, b'%s\0%s' % (filename, rawextras))
             )
         if self._labels is not None:
             labels = b'\0'.join(self._labels)
             records.append((RECORD_LABELS, 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, b'wb')
         irecords = iter(records)
         lrecords = next(irecords)
         assert lrecords[0] == RECORD_LOCAL
         f.write(hex(self._local) + b'\n')
         for rtype, data in irecords:
             if rtype == RECORD_MERGED:
                 f.write(b'%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
         allowlist = (RECORD_LOCAL, RECORD_OTHER, RECORD_MERGED)
         f = self._repo.vfs(self.statepathv2, b'wb')
         for key, data in records:
             assert len(key) == 1
             if key not in allowlist:
                 key, data = RECORD_OVERRIDE, b'%s%s' % (key, data)
             format = b'>sI%is' % len(data)
             f.write(_pack(format, key, len(data), data))
         f.close()
 
     @staticmethod
     def getlocalkey(path):
         """hash the path of a local file context for storage in the .hg/merge
         directory."""
 
         return hex(hashutil.sha1(path).digest())
 
     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():
             localkey = nullhex
         else:
             localkey = mergestate.getlocalkey(fcl.path())
             self._repo.vfs.write(b'merge/' + localkey, fcl.data())
         self._state[fd] = [
             MERGE_RECORD_UNRESOLVED,
             localkey,
             fcl.path(),
             fca.path(),
             hex(fca.filenode()),
             fco.path(),
             hex(fco.filenode()),
             fcl.flags(),
         ]
         self._stateextras[fd] = {b'ancestorlinknode': hex(fca.node())}
         self._dirty = True
 
     def addpath(self, path, frename, forigin):
         """add a new conflicting path to the merge state
         path:    the path that conflicts
         frename: the filename the conflicting file was renamed to
         forigin: origin of the file ('l' or 'r' for local/remote)
         """
         self._state[path] = [MERGE_RECORD_UNRESOLVED_PATH, frename, forigin]
         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 pycompat.iteritems(self._state):
             if entry[0] in (
                 MERGE_RECORD_UNRESOLVED,
                 MERGE_RECORD_UNRESOLVED_PATH,
             ):
                 yield f
 
     def driverresolved(self):
         """Obtain the paths of driver-resolved files."""
 
         for f, entry in self._state.items():
             if entry[0] == MERGE_RECORD_DRIVER_RESOLVED:
                 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 (MERGE_RECORD_RESOLVED, MERGE_RECORD_DRIVER_RESOLVED):
             return True, 0
         stateentry = self._state[dfile]
         state, localkey, lfile, afile, anode, ofile, onode, flags = stateentry
         octx = self._repo[self._other]
         extras = self.extras(dfile)
         anccommitnode = extras.get(b'ancestorlinknode')
         if anccommitnode:
             actx = self._repo[anccommitnode]
         else:
             actx = None
         fcd = self._filectxorabsent(localkey, wctx, dfile)
         fco = self._filectxorabsent(onode, octx, ofile)
         # TODO: move this to filectxorabsent
         fca = self._repo.filectx(afile, fileid=anode, changectx=actx)
         # "premerge" x flags
         flo = fco.flags()
         fla = fca.flags()
         if b'x' in flags + flo + fla and b'l' not in flags + flo + fla:
             if fca.node() == nullid and flags != flo:
                 if preresolve:
                     self._repo.ui.warn(
                         _(
                             b'warning: cannot merge flags for %s '
                             b'without common ancestor - keeping local flags\n'
                         )
                         % afile
                     )
             elif flags == fla:
                 flags = flo
         if preresolve:
             # restore local
             if localkey != nullhex:
                 f = self._repo.vfs(b'merge/' + localkey)
                 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, MERGE_RECORD_RESOLVED)
 
         if complete:
             action = None
             if deleted:
                 if fcd.isabsent():
                     # dc: local picked. Need to drop if present, which may
                     # happen on re-resolves.
                     action = ACTION_FORGET
                 else:
                     # cd: remote picked (or otherwise deleted)
                     action = ACTION_REMOVE
             else:
                 if fcd.isabsent():  # dc: remote picked
                     action = ACTION_GET
                 elif fco.isabsent():  # cd: local picked
                     if dfile in self.localctx:
                         action = ACTION_ADD_MODIFIED
                     else:
                         action = ACTION_ADD
                 # 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 pycompat.itervalues(self._results):
             if r is None:
                 updated += 1
             elif r == 0:
                 if action == ACTION_REMOVE:
                     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 = {
             ACTION_REMOVE: [],
             ACTION_FORGET: [],
             ACTION_ADD: [],
             ACTION_ADD_MODIFIED: [],
             ACTION_GET: [],
         }
         for f, (r, action) in pycompat.iteritems(self._results):
             if action is not None:
                 actions[action].append((f, None, b"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, None)
 
     def queueremove(self, f):
         """queues a file to be removed from the dirstate
 
         Meant for use by custom merge drivers."""
         self._results[f] = 0, ACTION_REMOVE
 
     def queueadd(self, f):
         """queues a file to be added to the dirstate
 
         Meant for use by custom merge drivers."""
         self._results[f] = 0, ACTION_ADD
 
     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, ACTION_GET
 
 
 def _getcheckunknownconfig(repo, section, name):
     config = repo.ui.config(section, name)
     valid = [b'abort', b'ignore', b'warn']
     if config not in valid:
         validstr = b', '.join([b"'" + v + b"'" for v in valid])
         raise error.ConfigError(
             _(b"%s.%s not valid ('%s' is none of %s)")
             % (section, name, config, validstr)
         )
     return config
 
 
 def _checkunknownfile(repo, wctx, mctx, f, f2=None):
     if wctx.isinmemory():
         # Nothing to do in IMM because nothing in the "working copy" can be an
         # unknown file.
         #
         # Note that we should bail out here, not in ``_checkunknownfiles()``,
         # because that function does other useful work.
         return False
 
     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])
     )
 
 
 class _unknowndirschecker(object):
     """
     Look for any unknown files or directories that may have a path conflict
     with a file.  If any path prefix of the file exists as a file or link,
     then it conflicts.  If the file itself is a directory that contains any
     file that is not tracked, then it conflicts.
 
     Returns the shortest path at which a conflict occurs, or None if there is
     no conflict.
     """
 
     def __init__(self):
         # A set of paths known to be good.  This prevents repeated checking of
         # dirs.  It will be updated with any new dirs that are checked and found
         # to be safe.
         self._unknowndircache = set()
 
         # A set of paths that are known to be absent.  This prevents repeated
         # checking of subdirectories that are known not to exist. It will be
         # updated with any new dirs that are checked and found to be absent.
         self._missingdircache = set()
 
     def __call__(self, repo, wctx, f):
         if wctx.isinmemory():
             # Nothing to do in IMM for the same reason as ``_checkunknownfile``.
             return False
 
         # Check for path prefixes that exist as unknown files.
         for p in reversed(list(pathutil.finddirs(f))):
             if p in self._missingdircache:
                 return
             if p in self._unknowndircache:
                 continue
             if repo.wvfs.audit.check(p):
                 if (
                     repo.wvfs.isfileorlink(p)
                     and repo.dirstate.normalize(p) not in repo.dirstate
                 ):
                     return p
                 if not repo.wvfs.lexists(p):
                     self._missingdircache.add(p)
                     return
                 self._unknowndircache.add(p)
 
         # Check if the file conflicts with a directory containing unknown files.
         if repo.wvfs.audit.check(f) and repo.wvfs.isdir(f):
             # Does the directory contain any files that are not in the dirstate?
             for p, dirs, files in repo.wvfs.walk(f):
                 for fn in files:
                     relf = util.pconvert(repo.wvfs.reljoin(p, fn))
                     relf = repo.dirstate.normalize(relf, isknown=True)
                     if relf not in repo.dirstate:
                         return f
         return None
 
 
 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.
     """
     fileconflicts = set()
     pathconflicts = set()
     warnconflicts = set()
     abortconflicts = set()
     unknownconfig = _getcheckunknownconfig(repo, b'merge', b'checkunknown')
     ignoredconfig = _getcheckunknownconfig(repo, b'merge', b'checkignored')
     pathconfig = repo.ui.configbool(
         b'experimental', b'merge.checkpathconflicts'
     )
     if not force:
 
         def collectconflicts(conflicts, config):
             if config == b'abort':
                 abortconflicts.update(conflicts)
             elif config == b'warn':
                 warnconflicts.update(conflicts)
 
         checkunknowndirs = _unknowndirschecker()
         for f, (m, args, msg) in pycompat.iteritems(actions):
             if m in (ACTION_CREATED, ACTION_DELETED_CHANGED):
                 if _checkunknownfile(repo, wctx, mctx, f):
                     fileconflicts.add(f)
                 elif pathconfig and f not in wctx:
                     path = checkunknowndirs(repo, wctx, f)
                     if path is not None:
                         pathconflicts.add(path)
             elif m == ACTION_LOCAL_DIR_RENAME_GET:
                 if _checkunknownfile(repo, wctx, mctx, f, args[0]):
                     fileconflicts.add(f)
 
         allconflicts = fileconflicts | pathconflicts
         ignoredconflicts = {c for c in allconflicts if repo.dirstate._ignore(c)}
         unknownconflicts = allconflicts - ignoredconflicts
         collectconflicts(ignoredconflicts, ignoredconfig)
         collectconflicts(unknownconflicts, unknownconfig)
     else:
         for f, (m, args, msg) in pycompat.iteritems(actions):
             if m == ACTION_CREATED_MERGE:
                 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] = (ACTION_GET, (fl2, False), b'remote created')
                 elif mergeforce or config == b'abort':
                     actions[f] = (
                         ACTION_MERGE,
                         (f, f, None, False, anc),
                         b'remote differs from untracked local',
                     )
                 elif config == b'abort':
                     abortconflicts.add(f)
                 else:
                     if config == b'warn':
                         warnconflicts.add(f)
                     actions[f] = (ACTION_GET, (fl2, True), b'remote created')
 
     for f in sorted(abortconflicts):
         warn = repo.ui.warn
         if f in pathconflicts:
             if repo.wvfs.isfileorlink(f):
                 warn(_(b"%s: untracked file conflicts with directory\n") % f)
             else:
                 warn(_(b"%s: untracked directory conflicts with file\n") % f)
         else:
             warn(_(b"%s: untracked file differs\n") % f)
     if abortconflicts:
         raise error.Abort(
             _(
                 b"untracked files in working directory "
                 b"differ from files in requested revision"
             )
         )
 
     for f in sorted(warnconflicts):
         if repo.wvfs.isfileorlink(f):
             repo.ui.warn(_(b"%s: replacing untracked file\n") % f)
         else:
             repo.ui.warn(_(b"%s: replacing untracked files in directory\n") % f)
 
     for f, (m, args, msg) in pycompat.iteritems(actions):
         if m == ACTION_CREATED:
             backup = (
                 f in fileconflicts
                 or f in pathconflicts
                 or any(p in pathconflicts for p in pathutil.finddirs(f))
             )
             (flags,) = args
             actions[f] = (ACTION_GET, (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 = ACTION_FORGET
     if branchmerge:
         m = ACTION_REMOVE
     for f in wctx.deleted():
         if f not in mctx:
             actions[f] = m, None, b"forget deleted"
 
     if not branchmerge:
         for f in wctx.removed():
             if f not in mctx:
                 actions[f] = ACTION_FORGET, None, b"forget removed"
 
     return actions
 
 
 def _checkcollision(repo, wmf, actions):
     """
     Check for case-folding collisions.
     """
 
     # If the repo is narrowed, filter out files outside the narrowspec.
     narrowmatch = repo.narrowmatch()
     if not narrowmatch.always():
         wmf = wmf.matches(narrowmatch)
         if actions:
             narrowactions = {}
             for m, actionsfortype in pycompat.iteritems(actions):
                 narrowactions[m] = []
                 for (f, args, msg) in actionsfortype:
                     if narrowmatch(f):
                         narrowactions[m].append((f, args, msg))
             actions = narrowactions
 
     # build provisional merged manifest up
     pmmf = set(wmf)
 
     if actions:
         # KEEP and EXEC are no-op
         for m in (
             ACTION_ADD,
             ACTION_ADD_MODIFIED,
             ACTION_FORGET,
             ACTION_GET,
             ACTION_CHANGED_DELETED,
             ACTION_DELETED_CHANGED,
         ):
             for f, args, msg in actions[m]:
                 pmmf.add(f)
         for f, args, msg in actions[ACTION_REMOVE]:
             pmmf.discard(f)
         for f, args, msg in actions[ACTION_DIR_RENAME_MOVE_LOCAL]:
             f2, flags = args
             pmmf.discard(f2)
             pmmf.add(f)
         for f, args, msg in actions[ACTION_LOCAL_DIR_RENAME_GET]:
             pmmf.add(f)
         for f, args, msg in actions[ACTION_MERGE]:
             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(
                 _(b"case-folding collision between %s and %s")
                 % (f, foldmap[fold])
             )
         foldmap[fold] = f
 
     # check case-folding of directories
     foldprefix = unfoldprefix = lastfull = b''
     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(
                 _(b"case-folding collision between %s and directory of %s")
                 % (lastfull, f)
             )
         foldprefix = fold + b'/'
         unfoldprefix = f + b'/'
         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 _filesindirs(repo, manifest, dirs):
     """
     Generator that yields pairs of all the files in the manifest that are found
     inside the directories listed in dirs, and which directory they are found
     in.
     """
     for f in manifest:
         for p in pathutil.finddirs(f):
             if p in dirs:
                 yield f, p
                 break
 
 
 def checkpathconflicts(repo, wctx, mctx, actions):
     """
     Check if any actions introduce path conflicts in the repository, updating
     actions to record or handle the path conflict accordingly.
     """
     mf = wctx.manifest()
 
     # The set of local files that conflict with a remote directory.
     localconflicts = set()
 
     # The set of directories that conflict with a remote file, and so may cause
     # conflicts if they still contain any files after the merge.
     remoteconflicts = set()
 
     # The set of directories that appear as both a file and a directory in the
     # remote manifest.  These indicate an invalid remote manifest, which
     # can't be updated to cleanly.
     invalidconflicts = set()
 
     # The set of directories that contain files that are being created.
     createdfiledirs = set()
 
     # The set of files deleted by all the actions.
     deletedfiles = set()
 
     for f, (m, args, msg) in actions.items():
         if m in (
             ACTION_CREATED,
             ACTION_DELETED_CHANGED,
             ACTION_MERGE,
             ACTION_CREATED_MERGE,
         ):
             # This action may create a new local file.
             createdfiledirs.update(pathutil.finddirs(f))
             if mf.hasdir(f):
                 # The file aliases a local directory.  This might be ok if all
                 # the files in the local directory are being deleted.  This
                 # will be checked once we know what all the deleted files are.
                 remoteconflicts.add(f)
         # Track the names of all deleted files.
         if m == ACTION_REMOVE:
             deletedfiles.add(f)
         if m == ACTION_MERGE:
             f1, f2, fa, move, anc = args
             if move:
                 deletedfiles.add(f1)
         if m == ACTION_DIR_RENAME_MOVE_LOCAL:
             f2, flags = args
             deletedfiles.add(f2)
 
     # Check all directories that contain created files for path conflicts.
     for p in createdfiledirs:
         if p in mf:
             if p in mctx:
                 # A file is in a directory which aliases both a local
                 # and a remote file.  This is an internal inconsistency
                 # within the remote manifest.
                 invalidconflicts.add(p)
             else:
                 # A file is in a directory which aliases a local file.
                 # We will need to rename the local file.
                 localconflicts.add(p)
         if p in actions and actions[p][0] in (
             ACTION_CREATED,
             ACTION_DELETED_CHANGED,
             ACTION_MERGE,
             ACTION_CREATED_MERGE,
         ):
             # The file is in a directory which aliases a remote file.
             # This is an internal inconsistency within the remote
             # manifest.
             invalidconflicts.add(p)
 
     # Rename all local conflicting files that have not been deleted.
     for p in localconflicts:
         if p not in deletedfiles:
             ctxname = bytes(wctx).rstrip(b'+')
             pnew = util.safename(p, ctxname, wctx, set(actions.keys()))
             actions[pnew] = (
                 ACTION_PATH_CONFLICT_RESOLVE,
                 (p,),
                 b'local path conflict',
             )
             actions[p] = (ACTION_PATH_CONFLICT, (pnew, b'l'), b'path conflict')
 
     if remoteconflicts:
         # Check if all files in the conflicting directories have been removed.
         ctxname = bytes(mctx).rstrip(b'+')
         for f, p in _filesindirs(repo, mf, remoteconflicts):
             if f not in deletedfiles:
                 m, args, msg = actions[p]
                 pnew = util.safename(p, ctxname, wctx, set(actions.keys()))
                 if m in (ACTION_DELETED_CHANGED, ACTION_MERGE):
                     # Action was merge, just update target.
                     actions[pnew] = (m, args, msg)
                 else:
                     # Action was create, change to renamed get action.
                     fl = args[0]
                     actions[pnew] = (
                         ACTION_LOCAL_DIR_RENAME_GET,
                         (p, fl),
                         b'remote path conflict',
                     )
                 actions[p] = (
                     ACTION_PATH_CONFLICT,
                     (pnew, ACTION_REMOVE),
                     b'path conflict',
                 )
                 remoteconflicts.remove(p)
                 break
 
     if invalidconflicts:
         for p in invalidconflicts:
             repo.ui.warn(_(b"%s: is both a file and a directory\n") % p)
         raise error.Abort(_(b"destination manifest contains path conflicts"))
 
 
 def _filternarrowactions(narrowmatch, branchmerge, actions):
     """
     Filters out actions that can ignored because the repo is narrowed.
 
     Raise an exception if the merge cannot be completed because the repo is
     narrowed.
     """
     nooptypes = {b'k'}  # TODO: handle with nonconflicttypes
     nonconflicttypes = set(b'a am c cm f g r e'.split())
     # We mutate the items in the dict during iteration, so iterate
     # over a copy.
     for f, action in list(actions.items()):
         if narrowmatch(f):
             pass
         elif not branchmerge:
             del actions[f]  # just updating, ignore changes outside clone
         elif action[0] in nooptypes:
             del actions[f]  # merge does not affect file
         elif action[0] in nonconflicttypes:
             raise error.Abort(
                 _(
                     b'merge affects file \'%s\' outside narrow, '
                     b'which is not yet supported'
                 )
                 % f,
                 hint=_(b'merging in the other direction may work'),
             )
         else:
             raise error.Abort(
                 _(b'conflict in file \'%s\' is outside narrow clone') % f
             )
 
 
 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(_(b"resolving manifests\n"))
     repo.ui.debug(
         b" branchmerge: %s, force: %s, partial: %s\n" % (boolbm, boolf, boolm)
     )
     repo.ui.debug(b" 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 b'.hgsubstate' in m1 and wctx.rev() is None:
         # Check whether sub state is modified, and overwrite the manifest
         # to flag the change. If wctx is a committed revision, we shouldn't
         # care for the dirty state of the working directory.
         if any(wctx.sub(s).dirty() for s in wctx.substate):
             m1[b'.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 pycompat.iteritems(copy):
             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)
 
     actions = {}
     for f, ((n1, fl1), (n2, fl2)) in pycompat.iteritems(diff):
         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] = (
                         ACTION_MERGE,
                         (f, f, fa, False, pa.node()),
                         b'both renamed from %s' % fa,
                     )
                 else:
                     actions[f] = (
                         ACTION_MERGE,
                         (f, f, None, False, pa.node()),
                         b'both created',
                     )
             else:
                 a = ma[f]
                 fla = ma.flags(f)
                 nol = b'l' not in fl1 + fl2 + fla
                 if n2 == a and fl2 == fla:
                     actions[f] = (ACTION_KEEP, (), b'remote unchanged')
                 elif n1 == a and fl1 == fla:  # local unchanged - use remote
                     if n1 == n2:  # optimization: keep local content
                         actions[f] = (
                             ACTION_EXEC,
                             (fl2,),
                             b'update permissions',
                         )
                     else:
                         actions[f] = (
                             ACTION_GET,
                             (fl2, False),
                             b'remote is newer',
                         )
                 elif nol and n2 == a:  # remote only changed 'x'
                     actions[f] = (ACTION_EXEC, (fl2,), b'update permissions')
                 elif nol and n1 == a:  # local only changed 'x'
                     actions[f] = (ACTION_GET, (fl1, False), b'remote is newer')
                 else:  # both changed something
                     actions[f] = (
                         ACTION_MERGE,
                         (f, f, f, False, pa.node()),
                         b'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] = (
                         ACTION_MERGE,
                         (f, f2, None, True, pa.node()),
                         b'remote directory rename, both created',
                     )
                 else:
                     actions[f2] = (
                         ACTION_DIR_RENAME_MOVE_LOCAL,
                         (f, fl1),
                         b'remote directory rename - move from %s' % f,
                     )
             elif f in copy:
                 f2 = copy[f]
                 actions[f] = (
                     ACTION_MERGE,
                     (f, f2, f2, False, pa.node()),
                     b'local copied/moved from %s' % f2,
                 )
             elif f in ma:  # clean, a different, no remote
                 if n1 != ma[f]:
                     if acceptremote:
                         actions[f] = (ACTION_REMOVE, None, b'remote delete')
                     else:
                         actions[f] = (
                             ACTION_CHANGED_DELETED,
                             (f, None, f, False, pa.node()),
                             b'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] = (ACTION_FORGET, None, b'remote deleted')
                 else:
                     actions[f] = (ACTION_REMOVE, None, b'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] = (
                         ACTION_MERGE,
                         (f2, f, None, False, pa.node()),
                         b'local directory rename, both created',
                     )
                 else:
                     actions[f2] = (
                         ACTION_LOCAL_DIR_RENAME_GET,
                         (f, fl2),
                         b'local directory rename - get from %s' % f,
                     )
             elif f in copy:
                 f2 = copy[f]
                 if f2 in m2:
                     actions[f] = (
                         ACTION_MERGE,
                         (f2, f, f2, False, pa.node()),
                         b'remote copied from %s' % f2,
                     )
                 else:
                     actions[f] = (
                         ACTION_MERGE,
                         (f2, f, f2, True, pa.node()),
                         b'remote moved from %s' % 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] = (ACTION_CREATED, (fl2,), b'remote created')
                 elif not branchmerge:
                     actions[f] = (ACTION_CREATED, (fl2,), b'remote created')
                 else:
                     actions[f] = (
                         ACTION_CREATED_MERGE,
                         (fl2, pa.node()),
                         b'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] = (
                         ACTION_MERGE,
                         (df, f, f, False, pa.node()),
                         b'local directory rename - respect move '
                         b'from %s' % f,
                     )
                 elif acceptremote:
                     actions[f] = (ACTION_CREATED, (fl2,), b'remote recreating')
                 else:
                     actions[f] = (
                         ACTION_DELETED_CHANGED,
                         (None, f, f, False, pa.node()),
                         b'prompt deleted/changed',
                     )
 
     if repo.ui.configbool(b'experimental', b'merge.checkpathconflicts'):
         # If we are merging, look for path conflicts.
         checkpathconflicts(repo, wctx, p2, actions)
 
     narrowmatch = repo.narrowmatch()
     if not narrowmatch.always():
         # Updates "actions" in place
         _filternarrowactions(narrowmatch, branchmerge, actions)
 
     return actions, diverge, renamedelete
 
 
 def _resolvetrivial(repo, wctx, mctx, ancestor, actions):
     """Resolves false conflicts where the nodeid changed but the content
        remained the same."""
     # We force a copy of actions.items() because we're going to mutate
     # actions as we resolve trivial conflicts.
     for f, (m, args, msg) in list(actions.items()):
         if (
             m == ACTION_CHANGED_DELETED
             and f in ancestor
             and not wctx[f].cmp(ancestor[f])
         ):
             # local did change but ended up with same content
             actions[f] = ACTION_REMOVE, None, b'prompt same'
         elif (
             m == ACTION_DELETED_CHANGED
             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(
             _(b"note: merging %s and %s using bids from ancestors %s\n")
             % (
                 wctx,
                 mctx,
                 _(b' and ').join(pycompat.bytestr(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(_(b'\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(pycompat.iteritems(actions)):
                 m, args, msg = a
                 repo.ui.debug(b' %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(_(b'\nauction for merging merge bids\n'))
         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 = list(bids.items())[0]
                 if all(a == l[0] for a in l[1:]):  # len(bids) is > 1
                     repo.ui.note(_(b" %s: consensus for %s\n") % (f, m))
                     actions[f] = l[0]
                     continue
             # If keep is an option, just do it.
             if ACTION_KEEP in bids:
                 repo.ui.note(_(b" %s: picking 'keep' action\n") % f)
                 actions[f] = bids[ACTION_KEEP][0]
                 continue
             # If there are gets and they all agree [how could they not?], do it.
             if ACTION_GET in bids:
                 ga0 = bids[ACTION_GET][0]
                 if all(a == ga0 for a in bids[ACTION_GET][1:]):
                     repo.ui.note(_(b" %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(_(b' %s: multiple bids for merge action:\n') % f)
             for m, l in sorted(bids.items()):
                 for _f, args, msg in l:
                     repo.ui.note(b'  %s -> %s\n' % (msg, m))
             # Pick random action. TODO: Instead, prompt user when resolving
             m, l = list(bids.items())[0]
             repo.ui.warn(
                 _(b' %s: ambiguous merge - picked %s action\n') % (f, m)
             )
             actions[f] = l[0]
             continue
         repo.ui.note(_(b'end of auction\n\n'))
 
     if wctx.rev() is None:
         fractions = _forgetremoved(wctx, mctx, branchmerge)
         actions.update(fractions)
 
     prunedactions = sparse.filterupdatesactions(
         repo, wctx, mctx, branchmerge, actions
     )
     _resolvetrivial(repo, wctx, mctx, ancestors[0], actions)
 
     return prunedactions, diverge, renamedelete
 
 
 def _getcwd():
     try:
         return encoding.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(b" %s: %s -> r\n" % (f, msg))
         if verbose:
             repo.ui.note(_(b"removing %s\n") % f)
         wctx[f].audit()
         try:
             wctx[f].remove(ignoremissing=True)
         except OSError as inst:
             repo.ui.warn(
                 _(b"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(
             _(
                 b"current directory was removed\n"
                 b"(consider changing to repo root: %s)\n"
             )
             % repo.root
         )
 
 
 def batchget(repo, mctx, wctx, wantfiledata, actions):
     """apply gets to the working directory
 
     mctx is the context to get from
 
     Yields arbitrarily many (False, tuple) for progress updates, followed by
     exactly one (True, filedata). When wantfiledata is false, filedata is an
     empty dict. When wantfiledata is true, filedata[f] is a triple (mode, size,
     mtime) of the file f written for each action.
     """
     filedata = {}
     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(b" %s: %s -> g\n" % (f, msg))
             if verbose:
                 repo.ui.note(_(b"getting %s\n") % f)
 
             if backup:
                 # If a file or directory exists with the same name, back that
                 # up.  Otherwise, look to see if there is a file that conflicts
                 # with a directory this file is in, and if so, back that up.
                 conflicting = f
                 if not repo.wvfs.lexists(f):
                     for p in pathutil.finddirs(f):
                         if repo.wvfs.isfileorlink(p):
                             conflicting = p
                             break
                 if repo.wvfs.lexists(conflicting):
                     orig = scmutil.backuppath(ui, repo, conflicting)
                     util.rename(repo.wjoin(conflicting), orig)
             wfctx = wctx[f]
             wfctx.clearunknown()
             atomictemp = ui.configbool(b"experimental", b"update.atomic-file")
             size = wfctx.write(
                 fctx(f).data(),
                 flags,
                 backgroundclose=True,
                 atomictemp=atomictemp,
             )
             if wantfiledata:
                 s = wfctx.lstat()
                 mode = s.st_mode
                 mtime = s[stat.ST_MTIME]
                 filedata[f] = (mode, size, mtime)  # for dirstate.normal
             if i == 100:
                 yield False, (i, f)
                 i = 0
             i += 1
     if i > 0:
         yield False, (i, f)
     yield True, filedata
 
 
 def _prefetchfiles(repo, ctx, actions):
     """Invoke ``scmutil.prefetchfiles()`` for the files relevant to the dict
     of merge actions.  ``ctx`` is the context being merged in."""
 
     # Skipping 'a', 'am', 'f', 'r', 'dm', 'e', 'k', 'p' and 'pr', because they
     # don't touch the context to be merged in.  'cd' is skipped, because
     # changed/deleted never resolves to something from the remote side.
     oplist = [
         actions[a]
         for a in (
             ACTION_GET,
             ACTION_DELETED_CHANGED,
             ACTION_LOCAL_DIR_RENAME_GET,
             ACTION_MERGE,
         )
     ]
     prefetch = scmutil.prefetchfiles
     matchfiles = scmutil.matchfiles
     prefetch(
         repo,
         [ctx.rev()],
         matchfiles(repo, [f for sublist in oplist for f, args, msg in sublist]),
     )
 
 
 @attr.s(frozen=True)
 class updateresult(object):
     updatedcount = attr.ib()
     mergedcount = attr.ib()
     removedcount = attr.ib()
     unresolvedcount = attr.ib()
 
     def isempty(self):
         return not (
             self.updatedcount
             or self.mergedcount
             or self.removedcount
             or self.unresolvedcount
         )
 
 
 def emptyactions():
     """create an actions dict, to be populated and passed to applyupdates()"""
     return dict(
         (m, [])
         for m in (
             ACTION_ADD,
             ACTION_ADD_MODIFIED,
             ACTION_FORGET,
             ACTION_GET,
             ACTION_CHANGED_DELETED,
             ACTION_DELETED_CHANGED,
             ACTION_REMOVE,
             ACTION_DIR_RENAME_MOVE_LOCAL,
             ACTION_LOCAL_DIR_RENAME_GET,
             ACTION_MERGE,
             ACTION_EXEC,
             ACTION_KEEP,
             ACTION_PATH_CONFLICT,
             ACTION_PATH_CONFLICT_RESOLVE,
         )
     )
 
 
 def applyupdates(
     repo, actions, wctx, mctx, overwrite, wantfiledata, 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, filedata), where counts is a tuple
     (updated, merged, removed, unresolved) that describes how many
     files were affected by the update, and filedata is as described in
     batchget.
     """
 
     _prefetchfiles(repo, mctx, actions)
 
     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[ACTION_CHANGED_DELETED])
     mergeactions.extend(sorted(actions[ACTION_DELETED_CHANGED]))
     mergeactions.extend(actions[ACTION_MERGE])
     for f, args, msg in mergeactions:
         f1, f2, fa, move, anc = args
         if f == b'.hgsubstate':  # merged internally
             continue
         if f1 is None:
             fcl = filemerge.absentfilectx(wctx, fa)
         else:
             repo.ui.debug(b" 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)
 
     # remove renamed files after safely stored
     for f in moves:
         if wctx[f].lexists():
             repo.ui.debug(b"removing %s\n" % f)
             wctx[f].audit()
             wctx[f].remove()
 
     numupdates = sum(len(l) for m, l in actions.items() if m != ACTION_KEEP)
     progress = repo.ui.makeprogress(
         _(b'updating'), unit=_(b'files'), total=numupdates
     )
 
     if [a for a in actions[ACTION_REMOVE] if a[0] == b'.hgsubstate']:
         subrepoutil.submerge(repo, wctx, mctx, wctx, overwrite, labels)
 
     # record path conflicts
     for f, args, msg in actions[ACTION_PATH_CONFLICT]:
         f1, fo = args
         s = repo.ui.status
         s(
             _(
                 b"%s: path conflict - a file or link has the same name as a "
                 b"directory\n"
             )
             % f
         )
         if fo == b'l':
             s(_(b"the local file has been renamed to %s\n") % f1)
         else:
             s(_(b"the remote file has been renamed to %s\n") % f1)
         s(_(b"resolve manually then use 'hg resolve --mark %s'\n") % f)
         ms.addpath(f, f1, fo)
         progress.increment(item=f)
 
     # When merging in-memory, we can't support worker processes, so set the
     # per-item cost at 0 in that case.
     cost = 0 if wctx.isinmemory() else 0.001
 
     # remove in parallel (must come before resolving path conflicts and getting)
     prog = worker.worker(
         repo.ui, cost, batchremove, (repo, wctx), actions[ACTION_REMOVE]
     )
     for i, item in prog:
         progress.increment(step=i, item=item)
     removed = len(actions[ACTION_REMOVE])
 
     # resolve path conflicts (must come before getting)
     for f, args, msg in actions[ACTION_PATH_CONFLICT_RESOLVE]:
         repo.ui.debug(b" %s: %s -> pr\n" % (f, msg))
         (f0,) = args
         if wctx[f0].lexists():
             repo.ui.note(_(b"moving %s to %s\n") % (f0, f))
             wctx[f].audit()
             wctx[f].write(wctx.filectx(f0).data(), wctx.filectx(f0).flags())
             wctx[f0].remove()
         progress.increment(item=f)
 
     # get in parallel.
     threadsafe = repo.ui.configbool(
         b'experimental', b'worker.wdir-get-thread-safe'
     )
     prog = worker.worker(
         repo.ui,
         cost,
         batchget,
         (repo, mctx, wctx, wantfiledata),
         actions[ACTION_GET],
         threadsafe=threadsafe,
         hasretval=True,
     )
     getfiledata = {}
     for final, res in prog:
         if final:
             getfiledata = res
         else:
             i, item = res
             progress.increment(step=i, item=item)
     updated = len(actions[ACTION_GET])
 
     if [a for a in actions[ACTION_GET] if a[0] == b'.hgsubstate']:
         subrepoutil.submerge(repo, wctx, mctx, wctx, overwrite, labels)
 
     # forget (manifest only, just log it) (must come first)
     for f, args, msg in actions[ACTION_FORGET]:
         repo.ui.debug(b" %s: %s -> f\n" % (f, msg))
         progress.increment(item=f)
 
     # re-add (manifest only, just log it)
     for f, args, msg in actions[ACTION_ADD]:
         repo.ui.debug(b" %s: %s -> a\n" % (f, msg))
         progress.increment(item=f)
 
     # re-add/mark as modified (manifest only, just log it)
     for f, args, msg in actions[ACTION_ADD_MODIFIED]:
         repo.ui.debug(b" %s: %s -> am\n" % (f, msg))
         progress.increment(item=f)
 
     # keep (noop, just log it)
     for f, args, msg in actions[ACTION_KEEP]:
         repo.ui.debug(b" %s: %s -> k\n" % (f, msg))
         # no progress
 
     # directory rename, move local
     for f, args, msg in actions[ACTION_DIR_RENAME_MOVE_LOCAL]:
         repo.ui.debug(b" %s: %s -> dm\n" % (f, msg))
         progress.increment(item=f)
         f0, flags = args
         repo.ui.note(_(b"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[ACTION_LOCAL_DIR_RENAME_GET]:
         repo.ui.debug(b" %s: %s -> dg\n" % (f, msg))
         progress.increment(item=f)
         f0, flags = args
         repo.ui.note(_(b"getting %s to %s\n") % (f0, f))
         wctx[f].write(mctx.filectx(f0).data(), flags)
         updated += 1
 
     # exec
     for f, args, msg in actions[ACTION_EXEC]:
         repo.ui.debug(b" %s: %s -> e\n" % (f, msg))
         progress.increment(item=f)
         (flags,) = args
         wctx[f].audit()
         wctx[f].setflags(b'l' in flags, b'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:
         if wctx.isinmemory():
             raise error.InMemoryMergeConflictsError(
                 b"in-memory merge does not support mergedriver"
             )
         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 updateresult(
                 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
 
     try:
         # premerge
         tocomplete = []
         for f, args, msg in mergeactions:
             repo.ui.debug(b" %s: %s -> m (premerge)\n" % (f, msg))
             progress.increment(item=f)
             if f == b'.hgsubstate':  # subrepo states need updating
                 subrepoutil.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(b" %s: %s -> m (merge)\n" % (f, msg))
             progress.increment(item=f, total=numupdates)
             ms.resolve(f, wctx)
 
     finally:
         ms.commit()
 
     unresolved = ms.unresolvedcount()
 
     if (
         usemergedriver
         and not unresolved
         and ms.mdstate() != MERGE_DRIVER_STATE_SUCCESS
     ):
         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[ACTION_MERGE])
         for k, acts in pycompat.iteritems(extraactions):
             actions[k].extend(acts)
             if k == ACTION_GET and wantfiledata:
                 # no filedata until mergestate is updated to provide it
                 for a in acts:
                     getfiledata[a[0]] = None
             # Remove these files from actions[ACTION_MERGE] as well. This is
             # important because in recordupdates, files in actions[ACTION_MERGE]
             # 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[ACTION_REMOVE] (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[ACTION_MERGE] = [
             a for a in actions[ACTION_MERGE] if a[0] in mfiles
         ]
 
     progress.complete()
     assert len(getfiledata) == (len(actions[ACTION_GET]) if wantfiledata else 0)
     return updateresult(updated, merged, removed, unresolved), getfiledata
 
 
 def recordupdates(repo, actions, branchmerge, getfiledata):
     """record merge actions to the dirstate"""
     # remove (must come first)
     for f, args, msg in actions.get(ACTION_REMOVE, []):
         if branchmerge:
             repo.dirstate.remove(f)
         else:
             repo.dirstate.drop(f)
 
     # forget (must come first)
     for f, args, msg in actions.get(ACTION_FORGET, []):
         repo.dirstate.drop(f)
 
     # resolve path conflicts
     for f, args, msg in actions.get(ACTION_PATH_CONFLICT_RESOLVE, []):
         (f0,) = args
         origf0 = repo.dirstate.copied(f0) or f0
         repo.dirstate.add(f)
         repo.dirstate.copy(origf0, f)
         if f0 == origf0:
             repo.dirstate.remove(f0)
         else:
             repo.dirstate.drop(f0)
 
     # re-add
     for f, args, msg in actions.get(ACTION_ADD, []):
         repo.dirstate.add(f)
 
     # re-add/mark as modified
     for f, args, msg in actions.get(ACTION_ADD_MODIFIED, []):
         if branchmerge:
             repo.dirstate.normallookup(f)
         else:
             repo.dirstate.add(f)
 
     # exec change
     for f, args, msg in actions.get(ACTION_EXEC, []):
         repo.dirstate.normallookup(f)
 
     # keep
     for f, args, msg in actions.get(ACTION_KEEP, []):
         pass
 
     # get
     for f, args, msg in actions.get(ACTION_GET, []):
         if branchmerge:
             repo.dirstate.otherparent(f)
         else:
             parentfiledata = getfiledata[f] if getfiledata else None
             repo.dirstate.normal(f, parentfiledata=parentfiledata)
 
     # merge
     for f, args, msg in actions.get(ACTION_MERGE, []):
         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(ACTION_DIR_RENAME_MOVE_LOCAL, []):
         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(ACTION_LOCAL_DIR_RENAME_GET, []):
         f0, flag = args
         if branchmerge:
             repo.dirstate.add(f)
             repo.dirstate.copy(f0, f)
         else:
             repo.dirstate.normal(f)
 
 
 UPDATECHECK_ABORT = b'abort'  # handled at higher layers
 UPDATECHECK_NONE = b'none'
 UPDATECHECK_LINEAR = b'linear'
 UPDATECHECK_NO_CONFLICT = b'noconflict'
 
 
 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/--check and -C/--clean 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 = UPDATECHECK_LINEAR
         if updatecheck not in (
             UPDATECHECK_NONE,
             UPDATECHECK_LINEAR,
             UPDATECHECK_NO_CONFLICT,
         ):
             raise ValueError(
                 r'Invalid updatecheck %r (can accept %r)'
                 % (
                     updatecheck,
                     (
                         UPDATECHECK_NONE,
                         UPDATECHECK_LINEAR,
                         UPDATECHECK_NO_CONFLICT,
                     ),
                 )
             )
     # 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]
         p2 = repo[node]
         if ancestor is not None:
             pas = [repo[ancestor]]
         else:
             if repo.ui.configlist(b'merge', b'preferancestor') == [b'*']:
                 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(), bytes(p1), bytes(p2)
 
         overwrite = force and not branchmerge
         ### check phase
         if not overwrite:
             if len(pl) > 1:
                 raise error.Abort(_(b"outstanding uncommitted merge"))
             ms = mergestate.read(repo)
             if list(ms.unresolved()):
                 raise error.Abort(
                     _(b"outstanding merge conflicts"),
                     hint=_(b"use 'hg resolve' to resolve"),
                 )
         if branchmerge:
             if pas == [p2]:
                 raise error.Abort(
                     _(
                         b"merging with a working directory ancestor"
                         b" has no effect"
                     )
                 )
             elif pas == [p1]:
                 if not mergeancestor and wc.branch() == p2.branch():
                     raise error.Abort(
                         _(b"nothing to merge"),
                         hint=_(b"use 'hg update' or check 'hg heads'"),
                     )
             if not force and (wc.files() or wc.deleted()):
                 raise error.Abort(
                     _(b"uncommitted changes"),
                     hint=_(b"use 'hg status' to list changes"),
                 )
             if not wc.isinmemory():
                 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(b'preupdate', throw=True, parent1=xp2, parent2=b'')
                 repo.hook(b'update', parent1=xp2, parent2=b'', error=0)
                 return updateresult(0, 0, 0, 0)
 
             if updatecheck == 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 = _(b"uncommitted changes")
                         hint = _(b"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(b'merge', b'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 == UPDATECHECK_NO_CONFLICT:
             for f, (m, args, msg) in pycompat.iteritems(actionbyfile):
                 if m not in (
                     ACTION_GET,
                     ACTION_KEEP,
                     ACTION_EXEC,
                     ACTION_REMOVE,
                     ACTION_PATH_CONFLICT_RESOLVE,
                 ):
                     msg = _(b"conflicting changes")
                     hint = _(b"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
         # subrepoutil.submerge yet so we have to keep prompting for it.
         if b'.hgsubstate' in actionbyfile:
             f = b'.hgsubstate'
             m, args, msg = actionbyfile[f]
             prompts = filemerge.partextras(labels)
             prompts[b'f'] = f
             if m == ACTION_CHANGED_DELETED:
                 if repo.ui.promptchoice(
                     _(
                         b"local%(l)s changed %(f)s which other%(o)s deleted\n"
                         b"use (c)hanged version or (d)elete?"
                         b"$$ &Changed $$ &Delete"
                     )
                     % prompts,
                     0,
                 ):
                     actionbyfile[f] = (ACTION_REMOVE, None, b'prompt delete')
                 elif f in p1:
                     actionbyfile[f] = (
                         ACTION_ADD_MODIFIED,
                         None,
                         b'prompt keep',
                     )
                 else:
                     actionbyfile[f] = (ACTION_ADD, None, b'prompt keep')
             elif m == ACTION_DELETED_CHANGED:
                 f1, f2, fa, move, anc = args
                 flags = p2[f2].flags()
                 if (
                     repo.ui.promptchoice(
                         _(
                             b"other%(o)s changed %(f)s which local%(l)s deleted\n"
                             b"use (c)hanged version or leave (d)eleted?"
                             b"$$ &Changed $$ &Deleted"
                         )
                         % prompts,
                         0,
                     )
                     == 0
                 ):
                     actionbyfile[f] = (
                         ACTION_GET,
                         (flags, False),
                         b'prompt recreating',
                     )
                 else:
                     del actionbyfile[f]
 
         # Convert to dictionary-of-lists format
         actions = emptyactions()
         for f, (m, args, msg) in pycompat.iteritems(actionbyfile):
             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(pycompat.iteritems(diverge)):
             repo.ui.warn(
                 _(
                     b"note: possible conflict - %s was renamed "
                     b"multiple times to:\n"
                 )
                 % f
             )
             for nf in sorted(fl):
                 repo.ui.warn(b" %s\n" % nf)
 
         # rename and delete
         for f, fl in sorted(pycompat.iteritems(renamedelete)):
             repo.ui.warn(
                 _(
                     b"note: possible conflict - %s was deleted "
                     b"and renamed to:\n"
                 )
                 % f
             )
             for nf in sorted(fl):
                 repo.ui.warn(b" %s\n" % nf)
 
         ### apply phase
         if not branchmerge:  # just jump to the new rev
             fp1, fp2, xp1, xp2 = fp2, nullid, xp2, b''
         if not partial and not wc.isinmemory():
             repo.hook(b'preupdate', throw=True, parent1=xp1, parent2=xp2)
             # note that we're in the middle of an update
             repo.vfs.write(b'updatestate', p2.hex())
 
         # Advertise fsmonitor when its presence could be useful.
         #
         # We only advertise when performing an update from an empty working
         # directory. This typically only occurs during initial clone.
         #
         # We give users a mechanism to disable the warning in case it is
         # annoying.
         #
         # We only allow on Linux and MacOS because that's where fsmonitor is
         # considered stable.
         fsmonitorwarning = repo.ui.configbool(b'fsmonitor', b'warn_when_unused')
         fsmonitorthreshold = repo.ui.configint(
             b'fsmonitor', b'warn_update_file_count'
         )
         try:
             # avoid cycle: extensions -> cmdutil -> merge
             from . import extensions
 
             extensions.find(b'fsmonitor')
             fsmonitorenabled = repo.ui.config(b'fsmonitor', b'mode') != b'off'
             # We intentionally don't look at whether fsmonitor has disabled
             # itself because a) fsmonitor may have already printed a warning
             # b) we only care about the config state here.
         except KeyError:
             fsmonitorenabled = False
 
         if (
             fsmonitorwarning
             and not fsmonitorenabled
             and p1.node() == nullid
             and len(actions[ACTION_GET]) >= fsmonitorthreshold
             and pycompat.sysplatform.startswith((b'linux', b'darwin'))
         ):
             repo.ui.warn(
                 _(
                     b'(warning: large working directory being used without '
                     b'fsmonitor enabled; enable fsmonitor to improve performance; '
                     b'see "hg help -e fsmonitor")\n'
                 )
             )
 
         updatedirstate = not partial and not wc.isinmemory()
         wantfiledata = updatedirstate and not branchmerge
         stats, getfiledata = applyupdates(
             repo, actions, wc, p2, overwrite, wantfiledata, labels=labels
         )
 
         if updatedirstate:
             with repo.dirstate.parentchange():
                 repo.setparents(fp1, fp2)
                 recordupdates(repo, actions, branchmerge, getfiledata)
                 # update completed, clear state
                 util.unlink(repo.vfs.join(b'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(
             b'update', parent1=xp1, parent2=xp2, error=stats.unresolvedcount
         )
     return stats
 
 
 def graft(
     repo, ctx, base, labels=None, keepparent=False, keepconflictparent=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
     base - merge base, usually ctx.p1()
     labels - merge labels eg ['local', 'graft']
     keepparent - keep second parent if any
     keepconflictparent - if unresolved, keep parent used for the merge
 
     """
     # 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".
     wctx = repo[None]
     pctx = wctx.p1()
     mergeancestor = repo.changelog.isancestor(pctx.node(), ctx.node())
 
     stats = update(
         repo,
         ctx.node(),
         True,
         True,
         base.node(),
         mergeancestor=mergeancestor,
         labels=labels,
     )
 
     if keepconflictparent and stats.unresolvedcount:
         pother = ctx.node()
     else:
         pother = nullid
         parents = ctx.parents()
         if keepparent and len(parents) == 2 and base in parents:
             parents.remove(base)
             pother = parents[0].node()
     # Never set both parents equal to each other
     if pother == pctx.node():
         pother = nullid
 
     with repo.dirstate.parentchange():
         repo.setparents(pctx.node(), pother)
         repo.dirstate.write(repo.currenttransaction())
         # fix up dirstate for copies and renames
-        copies.graftcopies(repo, wctx, ctx, base)
+        copies.graftcopies(wctx, ctx, base)
     return stats
 
 
 def purge(
     repo,
     matcher,
     ignored=False,
     removeemptydirs=True,
     removefiles=True,
     abortonerror=False,
     noop=False,
 ):
     """Purge the working directory of untracked files.
 
     ``matcher`` is a matcher configured to scan the working directory -
     potentially a subset.
 
     ``ignored`` controls whether ignored files should also be purged.
 
     ``removeemptydirs`` controls whether empty directories should be removed.
 
     ``removefiles`` controls whether files are removed.
 
     ``abortonerror`` causes an exception to be raised if an error occurs
     deleting a file or directory.
 
     ``noop`` controls whether to actually remove files. If not defined, actions
     will be taken.
 
     Returns an iterable of relative paths in the working directory that were
     or would be removed.
     """
 
     def remove(removefn, path):
         try:
             removefn(path)
         except OSError:
             m = _(b'%s cannot be removed') % path
             if abortonerror:
                 raise error.Abort(m)
             else:
                 repo.ui.warn(_(b'warning: %s\n') % m)
 
     # There's no API to copy a matcher. So mutate the passed matcher and
     # restore it when we're done.
     oldtraversedir = matcher.traversedir
 
     res = []
 
     try:
         if removeemptydirs:
             directories = []
             matcher.traversedir = directories.append
 
         status = repo.status(match=matcher, ignored=ignored, unknown=True)
 
         if removefiles:
             for f in sorted(status.unknown + status.ignored):
                 if not noop:
                     repo.ui.note(_(b'removing file %s\n') % f)
                     remove(repo.wvfs.unlink, f)
                 res.append(f)
 
         if removeemptydirs:
             for f in sorted(directories, reverse=True):
                 if matcher(f) and not repo.wvfs.listdir(f):
                     if not noop:
                         repo.ui.note(_(b'removing directory %s\n') % f)
                         remove(repo.wvfs.rmdir, f)
                     res.append(f)
 
         return res
 
     finally:
         matcher.traversedir = oldtraversedir

relnotes/next

@@ -1,34 +1,35 b''
 == New Features ==
 
  * Windows will process hgrc files in %PROGRAMDATA%\Mercurial\hgrc.d.
 
 
 == New Experimental Features ==
 
 
 == Bug Fixes  ==
 
  * The `indent()` template function was documented to not indent empty lines,
    but it still indented the first line even if it was empty. It no longer does
    that.
 
 == Backwards Compatibility Changes ==
 
 
 == Internal API Changes ==
 
  * Matcher instances no longer have a `explicitdir` property. Consider
    rewriting your code to use `repo.wvfs.isdir()` and/or
    `ctx.hasdir()` instead. Also, the `traversedir` property is now
    also called when only `explicitdir` used to be called. That may
    mean that you can simply remove the use of `explicitdir` if you
    were already using `traversedir`.
 
  * The `revlog.nodemap` object have been merged into the `revlog.index` object.
    * `n in revlog.nodemap` becomes `revlog.index.has_node(n)`,
    * `revlog.nodemap[n]` becomes `revlog.index.rev(n)`,
    * `revlog.nodemap.get(n)` becomes `revlog.index.get_rev(n)`.
 
  * `copies.duplicatecopies()` was renamed to
    `copies.graftcopies()`. Its arguments changed from revision numbers
-   to context objects.
+   to context objects. It also lost its `repo` and `skip` arguments
+   (they should no longer be needed).