subrepo: adjust subrepo prefix before calling subrepo.addremove() (API)

Differential Revision: https://phab.mercurial-scm.org/D5885

mercurial/scmutil.py

@@ -1,1841 +1,1842 b''
 # scmutil.py - Mercurial core utility functions
 #
 #  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 errno
 import glob
 import hashlib
 import os
 import re
 import subprocess
 import weakref
 
 from .i18n import _
 from .node import (
     bin,
     hex,
     nullid,
     nullrev,
     short,
     wdirid,
     wdirrev,
 )
 
 from . import (
     encoding,
     error,
     match as matchmod,
     obsolete,
     obsutil,
     pathutil,
     phases,
     policy,
     pycompat,
     revsetlang,
     similar,
     smartset,
     url,
     util,
     vfs,
 )
 
 from .utils import (
     procutil,
     stringutil,
 )
 
 if pycompat.iswindows:
     from . import scmwindows as scmplatform
 else:
     from . import scmposix as scmplatform
 
 parsers = policy.importmod(r'parsers')
 
 termsize = scmplatform.termsize
 
 class status(tuple):
     '''Named tuple with a list of files per status. The 'deleted', 'unknown'
        and 'ignored' properties are only relevant to the working copy.
     '''
 
     __slots__ = ()
 
     def __new__(cls, modified, added, removed, deleted, unknown, ignored,
                 clean):
         return tuple.__new__(cls, (modified, added, removed, deleted, unknown,
                                    ignored, clean))
 
     @property
     def modified(self):
         '''files that have been modified'''
         return self[0]
 
     @property
     def added(self):
         '''files that have been added'''
         return self[1]
 
     @property
     def removed(self):
         '''files that have been removed'''
         return self[2]
 
     @property
     def deleted(self):
         '''files that are in the dirstate, but have been deleted from the
            working copy (aka "missing")
         '''
         return self[3]
 
     @property
     def unknown(self):
         '''files not in the dirstate that are not ignored'''
         return self[4]
 
     @property
     def ignored(self):
         '''files not in the dirstate that are ignored (by _dirignore())'''
         return self[5]
 
     @property
     def clean(self):
         '''files that have not been modified'''
         return self[6]
 
     def __repr__(self, *args, **kwargs):
         return ((r'<status modified=%s, added=%s, removed=%s, deleted=%s, '
                  r'unknown=%s, ignored=%s, clean=%s>') %
                 tuple(pycompat.sysstr(stringutil.pprint(v)) for v in self))
 
 def itersubrepos(ctx1, ctx2):
     """find subrepos in ctx1 or ctx2"""
     # Create a (subpath, ctx) mapping where we prefer subpaths from
     # ctx1. The subpaths from ctx2 are important when the .hgsub file
     # has been modified (in ctx2) but not yet committed (in ctx1).
     subpaths = dict.fromkeys(ctx2.substate, ctx2)
     subpaths.update(dict.fromkeys(ctx1.substate, ctx1))
 
     missing = set()
 
     for subpath in ctx2.substate:
         if subpath not in ctx1.substate:
             del subpaths[subpath]
             missing.add(subpath)
 
     for subpath, ctx in sorted(subpaths.iteritems()):
         yield subpath, ctx.sub(subpath)
 
     # Yield an empty subrepo based on ctx1 for anything only in ctx2.  That way,
     # status and diff will have an accurate result when it does
     # 'sub.{status|diff}(rev2)'.  Otherwise, the ctx2 subrepo is compared
     # against itself.
     for subpath in missing:
         yield subpath, ctx2.nullsub(subpath, ctx1)
 
 def nochangesfound(ui, repo, excluded=None):
     '''Report no changes for push/pull, excluded is None or a list of
     nodes excluded from the push/pull.
     '''
     secretlist = []
     if excluded:
         for n in excluded:
             ctx = repo[n]
             if ctx.phase() >= phases.secret and not ctx.extinct():
                 secretlist.append(n)
 
     if secretlist:
         ui.status(_("no changes found (ignored %d secret changesets)\n")
                   % len(secretlist))
     else:
         ui.status(_("no changes found\n"))
 
 def callcatch(ui, func):
     """call func() with global exception handling
 
     return func() if no exception happens. otherwise do some error handling
     and return an exit code accordingly. does not handle all exceptions.
     """
     try:
         try:
             return func()
         except: # re-raises
             ui.traceback()
             raise
     # Global exception handling, alphabetically
     # Mercurial-specific first, followed by built-in and library exceptions
     except error.LockHeld as inst:
         if inst.errno == errno.ETIMEDOUT:
             reason = _('timed out waiting for lock held by %r') % (
                 pycompat.bytestr(inst.locker))
         else:
             reason = _('lock held by %r') % inst.locker
         ui.error(_("abort: %s: %s\n") % (
             inst.desc or stringutil.forcebytestr(inst.filename), reason))
         if not inst.locker:
             ui.error(_("(lock might be very busy)\n"))
     except error.LockUnavailable as inst:
         ui.error(_("abort: could not lock %s: %s\n") %
                  (inst.desc or stringutil.forcebytestr(inst.filename),
                   encoding.strtolocal(inst.strerror)))
     except error.OutOfBandError as inst:
         if inst.args:
             msg = _("abort: remote error:\n")
         else:
             msg = _("abort: remote error\n")
         ui.error(msg)
         if inst.args:
             ui.error(''.join(inst.args))
         if inst.hint:
             ui.error('(%s)\n' % inst.hint)
     except error.RepoError as inst:
         ui.error(_("abort: %s!\n") % inst)
         if inst.hint:
             ui.error(_("(%s)\n") % inst.hint)
     except error.ResponseError as inst:
         ui.error(_("abort: %s") % inst.args[0])
         msg = inst.args[1]
         if isinstance(msg, type(u'')):
             msg = pycompat.sysbytes(msg)
         if not isinstance(msg, bytes):
             ui.error(" %r\n" % (msg,))
         elif not msg:
             ui.error(_(" empty string\n"))
         else:
             ui.error("\n%r\n" % pycompat.bytestr(stringutil.ellipsis(msg)))
     except error.CensoredNodeError as inst:
         ui.error(_("abort: file censored %s!\n") % inst)
     except error.StorageError as inst:
         ui.error(_("abort: %s!\n") % inst)
         if inst.hint:
             ui.error(_("(%s)\n") % inst.hint)
     except error.InterventionRequired as inst:
         ui.error("%s\n" % inst)
         if inst.hint:
             ui.error(_("(%s)\n") % inst.hint)
         return 1
     except error.WdirUnsupported:
         ui.error(_("abort: working directory revision cannot be specified\n"))
     except error.Abort as inst:
         ui.error(_("abort: %s\n") % inst)
         if inst.hint:
             ui.error(_("(%s)\n") % inst.hint)
     except ImportError as inst:
         ui.error(_("abort: %s!\n") % stringutil.forcebytestr(inst))
         m = stringutil.forcebytestr(inst).split()[-1]
         if m in "mpatch bdiff".split():
             ui.error(_("(did you forget to compile extensions?)\n"))
         elif m in "zlib".split():
             ui.error(_("(is your Python install correct?)\n"))
     except (IOError, OSError) as inst:
         if util.safehasattr(inst, "code"): # HTTPError
             ui.error(_("abort: %s\n") % stringutil.forcebytestr(inst))
         elif util.safehasattr(inst, "reason"): # URLError or SSLError
             try: # usually it is in the form (errno, strerror)
                 reason = inst.reason.args[1]
             except (AttributeError, IndexError):
                 # it might be anything, for example a string
                 reason = inst.reason
             if isinstance(reason, pycompat.unicode):
                 # SSLError of Python 2.7.9 contains a unicode
                 reason = encoding.unitolocal(reason)
             ui.error(_("abort: error: %s\n") % reason)
         elif (util.safehasattr(inst, "args")
               and inst.args and inst.args[0] == errno.EPIPE):
             pass
         elif getattr(inst, "strerror", None): # common IOError or OSError
             if getattr(inst, "filename", None) is not None:
                 ui.error(_("abort: %s: '%s'\n") % (
                     encoding.strtolocal(inst.strerror),
                     stringutil.forcebytestr(inst.filename)))
             else:
                 ui.error(_("abort: %s\n") % encoding.strtolocal(inst.strerror))
         else: # suspicious IOError
             raise
     except MemoryError:
         ui.error(_("abort: out of memory\n"))
     except SystemExit as inst:
         # Commands shouldn't sys.exit directly, but give a return code.
         # Just in case catch this and and pass exit code to caller.
         return inst.code
 
     return -1
 
 def checknewlabel(repo, lbl, kind):
     # Do not use the "kind" parameter in ui output.
     # It makes strings difficult to translate.
     if lbl in ['tip', '.', 'null']:
         raise error.Abort(_("the name '%s' is reserved") % lbl)
     for c in (':', '\0', '\n', '\r'):
         if c in lbl:
             raise error.Abort(
                 _("%r cannot be used in a name") % pycompat.bytestr(c))
     try:
         int(lbl)
         raise error.Abort(_("cannot use an integer as a name"))
     except ValueError:
         pass
     if lbl.strip() != lbl:
         raise error.Abort(_("leading or trailing whitespace in name %r") % lbl)
 
 def checkfilename(f):
     '''Check that the filename f is an acceptable filename for a tracked file'''
     if '\r' in f or '\n' in f:
         raise error.Abort(_("'\\n' and '\\r' disallowed in filenames: %r")
                           % pycompat.bytestr(f))
 
 def checkportable(ui, f):
     '''Check if filename f is portable and warn or abort depending on config'''
     checkfilename(f)
     abort, warn = checkportabilityalert(ui)
     if abort or warn:
         msg = util.checkwinfilename(f)
         if msg:
             msg = "%s: %s" % (msg, procutil.shellquote(f))
             if abort:
                 raise error.Abort(msg)
             ui.warn(_("warning: %s\n") % msg)
 
 def checkportabilityalert(ui):
     '''check if the user's config requests nothing, a warning, or abort for
     non-portable filenames'''
     val = ui.config('ui', 'portablefilenames')
     lval = val.lower()
     bval = stringutil.parsebool(val)
     abort = pycompat.iswindows or lval == 'abort'
     warn = bval or lval == 'warn'
     if bval is None and not (warn or abort or lval == 'ignore'):
         raise error.ConfigError(
             _("ui.portablefilenames value is invalid ('%s')") % val)
     return abort, warn
 
 class casecollisionauditor(object):
     def __init__(self, ui, abort, dirstate):
         self._ui = ui
         self._abort = abort
         allfiles = '\0'.join(dirstate._map)
         self._loweredfiles = set(encoding.lower(allfiles).split('\0'))
         self._dirstate = dirstate
         # The purpose of _newfiles is so that we don't complain about
         # case collisions if someone were to call this object with the
         # same filename twice.
         self._newfiles = set()
 
     def __call__(self, f):
         if f in self._newfiles:
             return
         fl = encoding.lower(f)
         if fl in self._loweredfiles and f not in self._dirstate:
             msg = _('possible case-folding collision for %s') % f
             if self._abort:
                 raise error.Abort(msg)
             self._ui.warn(_("warning: %s\n") % msg)
         self._loweredfiles.add(fl)
         self._newfiles.add(f)
 
 def filteredhash(repo, maxrev):
     """build hash of filtered revisions in the current repoview.
 
     Multiple caches perform up-to-date validation by checking that the
     tiprev and tipnode stored in the cache file match the current repository.
     However, this is not sufficient for validating repoviews because the set
     of revisions in the view may change without the repository tiprev and
     tipnode changing.
 
     This function hashes all the revs filtered from the view and returns
     that SHA-1 digest.
     """
     cl = repo.changelog
     if not cl.filteredrevs:
         return None
     key = None
     revs = sorted(r for r in cl.filteredrevs if r <= maxrev)
     if revs:
         s = hashlib.sha1()
         for rev in revs:
             s.update('%d;' % rev)
         key = s.digest()
     return key
 
 def walkrepos(path, followsym=False, seen_dirs=None, recurse=False):
     '''yield every hg repository under path, always recursively.
     The recurse flag will only control recursion into repo working dirs'''
     def errhandler(err):
         if err.filename == path:
             raise err
     samestat = getattr(os.path, 'samestat', None)
     if followsym and samestat is not None:
         def adddir(dirlst, dirname):
             dirstat = os.stat(dirname)
             match = any(samestat(dirstat, lstdirstat) for lstdirstat in dirlst)
             if not match:
                 dirlst.append(dirstat)
             return not match
     else:
         followsym = False
 
     if (seen_dirs is None) and followsym:
         seen_dirs = []
         adddir(seen_dirs, path)
     for root, dirs, files in os.walk(path, topdown=True, onerror=errhandler):
         dirs.sort()
         if '.hg' in dirs:
             yield root # found a repository
             qroot = os.path.join(root, '.hg', 'patches')
             if os.path.isdir(os.path.join(qroot, '.hg')):
                 yield qroot # we have a patch queue repo here
             if recurse:
                 # avoid recursing inside the .hg directory
                 dirs.remove('.hg')
             else:
                 dirs[:] = [] # don't descend further
         elif followsym:
             newdirs = []
             for d in dirs:
                 fname = os.path.join(root, d)
                 if adddir(seen_dirs, fname):
                     if os.path.islink(fname):
                         for hgname in walkrepos(fname, True, seen_dirs):
                             yield hgname
                     else:
                         newdirs.append(d)
             dirs[:] = newdirs
 
 def binnode(ctx):
     """Return binary node id for a given basectx"""
     node = ctx.node()
     if node is None:
         return wdirid
     return node
 
 def intrev(ctx):
     """Return integer for a given basectx that can be used in comparison or
     arithmetic operation"""
     rev = ctx.rev()
     if rev is None:
         return wdirrev
     return rev
 
 def formatchangeid(ctx):
     """Format changectx as '{rev}:{node|formatnode}', which is the default
     template provided by logcmdutil.changesettemplater"""
     repo = ctx.repo()
     return formatrevnode(repo.ui, intrev(ctx), binnode(ctx))
 
 def formatrevnode(ui, rev, node):
     """Format given revision and node depending on the current verbosity"""
     if ui.debugflag:
         hexfunc = hex
     else:
         hexfunc = short
     return '%d:%s' % (rev, hexfunc(node))
 
 def resolvehexnodeidprefix(repo, prefix):
     if (prefix.startswith('x') and
         repo.ui.configbool('experimental', 'revisions.prefixhexnode')):
         prefix = prefix[1:]
     try:
         # Uses unfiltered repo because it's faster when prefix is ambiguous/
         # This matches the shortesthexnodeidprefix() function below.
         node = repo.unfiltered().changelog._partialmatch(prefix)
     except error.AmbiguousPrefixLookupError:
         revset = repo.ui.config('experimental', 'revisions.disambiguatewithin')
         if revset:
             # Clear config to avoid infinite recursion
             configoverrides = {('experimental',
                                 'revisions.disambiguatewithin'): None}
             with repo.ui.configoverride(configoverrides):
                 revs = repo.anyrevs([revset], user=True)
                 matches = []
                 for rev in revs:
                     node = repo.changelog.node(rev)
                     if hex(node).startswith(prefix):
                         matches.append(node)
                 if len(matches) == 1:
                     return matches[0]
         raise
     if node is None:
         return
     repo.changelog.rev(node)  # make sure node isn't filtered
     return node
 
 def mayberevnum(repo, prefix):
     """Checks if the given prefix may be mistaken for a revision number"""
     try:
         i = int(prefix)
         # if we are a pure int, then starting with zero will not be
         # confused as a rev; or, obviously, if the int is larger
         # than the value of the tip rev. We still need to disambiguate if
         # prefix == '0', since that *is* a valid revnum.
         if (prefix != b'0' and prefix[0:1] == b'0') or i >= len(repo):
             return False
         return True
     except ValueError:
         return False
 
 def shortesthexnodeidprefix(repo, node, minlength=1, cache=None):
     """Find the shortest unambiguous prefix that matches hexnode.
 
     If "cache" is not None, it must be a dictionary that can be used for
     caching between calls to this method.
     """
     # _partialmatch() of filtered changelog could take O(len(repo)) time,
     # which would be unacceptably slow. so we look for hash collision in
     # unfiltered space, which means some hashes may be slightly longer.
 
     minlength=max(minlength, 1)
 
     def disambiguate(prefix):
         """Disambiguate against revnums."""
         if repo.ui.configbool('experimental', 'revisions.prefixhexnode'):
             if mayberevnum(repo, prefix):
                 return 'x' + prefix
             else:
                 return prefix
 
         hexnode = hex(node)
         for length in range(len(prefix), len(hexnode) + 1):
             prefix = hexnode[:length]
             if not mayberevnum(repo, prefix):
                 return prefix
 
     cl = repo.unfiltered().changelog
     revset = repo.ui.config('experimental', 'revisions.disambiguatewithin')
     if revset:
         revs = None
         if cache is not None:
             revs = cache.get('disambiguationrevset')
         if revs is None:
             revs = repo.anyrevs([revset], user=True)
             if cache is not None:
                 cache['disambiguationrevset'] = revs
         if cl.rev(node) in revs:
             hexnode = hex(node)
             nodetree = None
             if cache is not None:
                 nodetree = cache.get('disambiguationnodetree')
             if not nodetree:
                 try:
                     nodetree = parsers.nodetree(cl.index, len(revs))
                 except AttributeError:
                     # no native nodetree
                     pass
                 else:
                     for r in revs:
                         nodetree.insert(r)
                     if cache is not None:
                         cache['disambiguationnodetree'] = nodetree
             if nodetree is not None:
                 length = max(nodetree.shortest(node), minlength)
                 prefix = hexnode[:length]
                 return disambiguate(prefix)
             for length in range(minlength, len(hexnode) + 1):
                 matches = []
                 prefix = hexnode[:length]
                 for rev in revs:
                     otherhexnode = repo[rev].hex()
                     if prefix == otherhexnode[:length]:
                         matches.append(otherhexnode)
                 if len(matches) == 1:
                     return disambiguate(prefix)
 
     try:
         return disambiguate(cl.shortest(node, minlength))
     except error.LookupError:
         raise error.RepoLookupError()
 
 def isrevsymbol(repo, symbol):
     """Checks if a symbol exists in the repo.
 
     See revsymbol() for details. Raises error.AmbiguousPrefixLookupError if the
     symbol is an ambiguous nodeid prefix.
     """
     try:
         revsymbol(repo, symbol)
         return True
     except error.RepoLookupError:
         return False
 
 def revsymbol(repo, symbol):
     """Returns a context given a single revision symbol (as string).
 
     This is similar to revsingle(), but accepts only a single revision symbol,
     i.e. things like ".", "tip", "1234", "deadbeef", "my-bookmark" work, but
     not "max(public())".
     """
     if not isinstance(symbol, bytes):
         msg = ("symbol (%s of type %s) was not a string, did you mean "
                "repo[symbol]?" % (symbol, type(symbol)))
         raise error.ProgrammingError(msg)
     try:
         if symbol in ('.', 'tip', 'null'):
             return repo[symbol]
 
         try:
             r = int(symbol)
             if '%d' % r != symbol:
                 raise ValueError
             l = len(repo.changelog)
             if r < 0:
                 r += l
             if r < 0 or r >= l and r != wdirrev:
                 raise ValueError
             return repo[r]
         except error.FilteredIndexError:
             raise
         except (ValueError, OverflowError, IndexError):
             pass
 
         if len(symbol) == 40:
             try:
                 node = bin(symbol)
                 rev = repo.changelog.rev(node)
                 return repo[rev]
             except error.FilteredLookupError:
                 raise
             except (TypeError, LookupError):
                 pass
 
         # look up bookmarks through the name interface
         try:
             node = repo.names.singlenode(repo, symbol)
             rev = repo.changelog.rev(node)
             return repo[rev]
         except KeyError:
             pass
 
         node = resolvehexnodeidprefix(repo, symbol)
         if node is not None:
             rev = repo.changelog.rev(node)
             return repo[rev]
 
         raise error.RepoLookupError(_("unknown revision '%s'") % symbol)
 
     except error.WdirUnsupported:
         return repo[None]
     except (error.FilteredIndexError, error.FilteredLookupError,
             error.FilteredRepoLookupError):
         raise _filterederror(repo, symbol)
 
 def _filterederror(repo, changeid):
     """build an exception to be raised about a filtered changeid
 
     This is extracted in a function to help extensions (eg: evolve) to
     experiment with various message variants."""
     if repo.filtername.startswith('visible'):
 
         # Check if the changeset is obsolete
         unfilteredrepo = repo.unfiltered()
         ctx = revsymbol(unfilteredrepo, changeid)
 
         # If the changeset is obsolete, enrich the message with the reason
         # that made this changeset not visible
         if ctx.obsolete():
             msg = obsutil._getfilteredreason(repo, changeid, ctx)
         else:
             msg = _("hidden revision '%s'") % changeid
 
         hint = _('use --hidden to access hidden revisions')
 
         return error.FilteredRepoLookupError(msg, hint=hint)
     msg = _("filtered revision '%s' (not in '%s' subset)")
     msg %= (changeid, repo.filtername)
     return error.FilteredRepoLookupError(msg)
 
 def revsingle(repo, revspec, default='.', localalias=None):
     if not revspec and revspec != 0:
         return repo[default]
 
     l = revrange(repo, [revspec], localalias=localalias)
     if not l:
         raise error.Abort(_('empty revision set'))
     return repo[l.last()]
 
 def _pairspec(revspec):
     tree = revsetlang.parse(revspec)
     return tree and tree[0] in ('range', 'rangepre', 'rangepost', 'rangeall')
 
 def revpair(repo, revs):
     if not revs:
         return repo['.'], repo[None]
 
     l = revrange(repo, revs)
 
     if not l:
         raise error.Abort(_('empty revision range'))
 
     first = l.first()
     second = l.last()
 
     if (first == second and len(revs) >= 2
         and not all(revrange(repo, [r]) for r in revs)):
         raise error.Abort(_('empty revision on one side of range'))
 
     # if top-level is range expression, the result must always be a pair
     if first == second and len(revs) == 1 and not _pairspec(revs[0]):
         return repo[first], repo[None]
 
     return repo[first], repo[second]
 
 def revrange(repo, specs, localalias=None):
     """Execute 1 to many revsets and return the union.
 
     This is the preferred mechanism for executing revsets using user-specified
     config options, such as revset aliases.
 
     The revsets specified by ``specs`` will be executed via a chained ``OR``
     expression. If ``specs`` is empty, an empty result is returned.
 
     ``specs`` can contain integers, in which case they are assumed to be
     revision numbers.
 
     It is assumed the revsets are already formatted. If you have arguments
     that need to be expanded in the revset, call ``revsetlang.formatspec()``
     and pass the result as an element of ``specs``.
 
     Specifying a single revset is allowed.
 
     Returns a ``revset.abstractsmartset`` which is a list-like interface over
     integer revisions.
     """
     allspecs = []
     for spec in specs:
         if isinstance(spec, int):
             spec = revsetlang.formatspec('%d', spec)
         allspecs.append(spec)
     return repo.anyrevs(allspecs, user=True, localalias=localalias)
 
 def meaningfulparents(repo, ctx):
     """Return list of meaningful (or all if debug) parentrevs for rev.
 
     For merges (two non-nullrev revisions) both parents are meaningful.
     Otherwise the first parent revision is considered meaningful if it
     is not the preceding revision.
     """
     parents = ctx.parents()
     if len(parents) > 1:
         return parents
     if repo.ui.debugflag:
         return [parents[0], repo[nullrev]]
     if parents[0].rev() >= intrev(ctx) - 1:
         return []
     return parents
 
 def getuipathfn(repo, legacyrelativevalue=False, forcerelativevalue=None):
     """Return a function that produced paths for presenting to the user.
 
     The returned function takes a repo-relative path and produces a path
     that can be presented in the UI.
 
     Depending on the value of ui.relative-paths, either a repo-relative or
     cwd-relative path will be produced.
 
     legacyrelativevalue is the value to use if ui.relative-paths=legacy
 
     If forcerelativevalue is not None, then that value will be used regardless
     of what ui.relative-paths is set to.
     """
     if forcerelativevalue is not None:
         relative = forcerelativevalue
     else:
         config = repo.ui.config('ui', 'relative-paths')
         if config == 'legacy':
             relative = legacyrelativevalue
         else:
             relative = stringutil.parsebool(config)
             if relative is None:
                 raise error.ConfigError(
                     _("ui.relative-paths is not a boolean ('%s')") % config)
 
     if relative:
         cwd = repo.getcwd()
         pathto = repo.pathto
         return lambda f: pathto(f, cwd)
     else:
         return lambda f: f
 
 def expandpats(pats):
     '''Expand bare globs when running on windows.
     On posix we assume it already has already been done by sh.'''
     if not util.expandglobs:
         return list(pats)
     ret = []
     for kindpat in pats:
         kind, pat = matchmod._patsplit(kindpat, None)
         if kind is None:
             try:
                 globbed = glob.glob(pat)
             except re.error:
                 globbed = [pat]
             if globbed:
                 ret.extend(globbed)
                 continue
         ret.append(kindpat)
     return ret
 
 def matchandpats(ctx, pats=(), opts=None, globbed=False, default='relpath',
                  badfn=None):
     '''Return a matcher and the patterns that were used.
     The matcher will warn about bad matches, unless an alternate badfn callback
     is provided.'''
     if pats == ("",):
         pats = []
     if opts is None:
         opts = {}
     if not globbed and default == 'relpath':
         pats = expandpats(pats or [])
 
     def bad(f, msg):
         ctx.repo().ui.warn("%s: %s\n" % (m.rel(f), msg))
 
     if badfn is None:
         badfn = bad
 
     m = ctx.match(pats, opts.get('include'), opts.get('exclude'),
                   default, listsubrepos=opts.get('subrepos'), badfn=badfn)
 
     if m.always():
         pats = []
     return m, pats
 
 def match(ctx, pats=(), opts=None, globbed=False, default='relpath',
           badfn=None):
     '''Return a matcher that will warn about bad matches.'''
     return matchandpats(ctx, pats, opts, globbed, default, badfn=badfn)[0]
 
 def matchall(repo):
     '''Return a matcher that will efficiently match everything.'''
     return matchmod.always(repo.root, repo.getcwd())
 
 def matchfiles(repo, files, badfn=None):
     '''Return a matcher that will efficiently match exactly these files.'''
     return matchmod.exact(repo.root, repo.getcwd(), files, badfn=badfn)
 
 def parsefollowlinespattern(repo, rev, pat, msg):
     """Return a file name from `pat` pattern suitable for usage in followlines
     logic.
     """
     if not matchmod.patkind(pat):
         return pathutil.canonpath(repo.root, repo.getcwd(), pat)
     else:
         ctx = repo[rev]
         m = matchmod.match(repo.root, repo.getcwd(), [pat], ctx=ctx)
         files = [f for f in ctx if m(f)]
         if len(files) != 1:
             raise error.ParseError(msg)
         return files[0]
 
 def getorigvfs(ui, repo):
     """return a vfs suitable to save 'orig' file
 
     return None if no special directory is configured"""
     origbackuppath = ui.config('ui', 'origbackuppath')
     if not origbackuppath:
         return None
     return vfs.vfs(repo.wvfs.join(origbackuppath))
 
 def backuppath(ui, repo, filepath):
     '''customize where working copy backup files (.orig files) are created
 
     Fetch user defined path from config file: [ui] origbackuppath = <path>
     Fall back to default (filepath with .orig suffix) if not specified
 
     filepath is repo-relative
 
     Returns an absolute path
     '''
     origvfs = getorigvfs(ui, repo)
     if origvfs is None:
         return repo.wjoin(filepath + ".orig")
 
     origbackupdir = origvfs.dirname(filepath)
     if not origvfs.isdir(origbackupdir) or origvfs.islink(origbackupdir):
         ui.note(_('creating directory: %s\n') % origvfs.join(origbackupdir))
 
         # Remove any files that conflict with the backup file's path
         for f in reversed(list(util.finddirs(filepath))):
             if origvfs.isfileorlink(f):
                 ui.note(_('removing conflicting file: %s\n')
                         % origvfs.join(f))
                 origvfs.unlink(f)
                 break
 
         origvfs.makedirs(origbackupdir)
 
     if origvfs.isdir(filepath) and not origvfs.islink(filepath):
         ui.note(_('removing conflicting directory: %s\n')
                 % origvfs.join(filepath))
         origvfs.rmtree(filepath, forcibly=True)
 
     return origvfs.join(filepath)
 
 class _containsnode(object):
     """proxy __contains__(node) to container.__contains__ which accepts revs"""
 
     def __init__(self, repo, revcontainer):
         self._torev = repo.changelog.rev
         self._revcontains = revcontainer.__contains__
 
     def __contains__(self, node):
         return self._revcontains(self._torev(node))
 
 def cleanupnodes(repo, replacements, operation, moves=None, metadata=None,
                  fixphase=False, targetphase=None, backup=True):
     """do common cleanups when old nodes are replaced by new nodes
 
     That includes writing obsmarkers or stripping nodes, and moving bookmarks.
     (we might also want to move working directory parent in the future)
 
     By default, bookmark moves are calculated automatically from 'replacements',
     but 'moves' can be used to override that. Also, 'moves' may include
     additional bookmark moves that should not have associated obsmarkers.
 
     replacements is {oldnode: [newnode]} or a iterable of nodes if they do not
     have replacements. operation is a string, like "rebase".
 
     metadata is dictionary containing metadata to be stored in obsmarker if
     obsolescence is enabled.
     """
     assert fixphase or targetphase is None
     if not replacements and not moves:
         return
 
     # translate mapping's other forms
     if not util.safehasattr(replacements, 'items'):
         replacements = {(n,): () for n in replacements}
     else:
         # upgrading non tuple "source" to tuple ones for BC
         repls = {}
         for key, value in replacements.items():
             if not isinstance(key, tuple):
                 key = (key,)
             repls[key] = value
         replacements = repls
 
     # Unfiltered repo is needed since nodes in replacements might be hidden.
     unfi = repo.unfiltered()
 
     # Calculate bookmark movements
     if moves is None:
         moves = {}
         for oldnodes, newnodes in replacements.items():
             for oldnode in oldnodes:
                 if oldnode in moves:
                     continue
                 if len(newnodes) > 1:
                     # usually a split, take the one with biggest rev number
                     newnode = next(unfi.set('max(%ln)', newnodes)).node()
                 elif len(newnodes) == 0:
                     # move bookmark backwards
                     allreplaced = []
                     for rep in replacements:
                         allreplaced.extend(rep)
                     roots = list(unfi.set('max((::%n) - %ln)', oldnode,
                                           allreplaced))
                     if roots:
                         newnode = roots[0].node()
                     else:
                         newnode = nullid
                 else:
                     newnode = newnodes[0]
                 moves[oldnode] = newnode
 
     allnewnodes = [n for ns in replacements.values() for n in ns]
     toretract = {}
     toadvance = {}
     if fixphase:
         precursors = {}
         for oldnodes, newnodes in replacements.items():
             for oldnode in oldnodes:
                 for newnode in newnodes:
                     precursors.setdefault(newnode, []).append(oldnode)
 
         allnewnodes.sort(key=lambda n: unfi[n].rev())
         newphases = {}
         def phase(ctx):
             return newphases.get(ctx.node(), ctx.phase())
         for newnode in allnewnodes:
             ctx = unfi[newnode]
             parentphase = max(phase(p) for p in ctx.parents())
             if targetphase is None:
                 oldphase = max(unfi[oldnode].phase()
                                for oldnode in precursors[newnode])
                 newphase = max(oldphase, parentphase)
             else:
                 newphase = max(targetphase, parentphase)
             newphases[newnode] = newphase
             if newphase > ctx.phase():
                 toretract.setdefault(newphase, []).append(newnode)
             elif newphase < ctx.phase():
                 toadvance.setdefault(newphase, []).append(newnode)
 
     with repo.transaction('cleanup') as tr:
         # Move bookmarks
         bmarks = repo._bookmarks
         bmarkchanges = []
         for oldnode, newnode in moves.items():
             oldbmarks = repo.nodebookmarks(oldnode)
             if not oldbmarks:
                 continue
             from . import bookmarks # avoid import cycle
             repo.ui.debug('moving bookmarks %r from %s to %s\n' %
                           (pycompat.rapply(pycompat.maybebytestr, oldbmarks),
                            hex(oldnode), hex(newnode)))
             # Delete divergent bookmarks being parents of related newnodes
             deleterevs = repo.revs('parents(roots(%ln & (::%n))) - parents(%n)',
                                    allnewnodes, newnode, oldnode)
             deletenodes = _containsnode(repo, deleterevs)
             for name in oldbmarks:
                 bmarkchanges.append((name, newnode))
                 for b in bookmarks.divergent2delete(repo, deletenodes, name):
                     bmarkchanges.append((b, None))
 
         if bmarkchanges:
             bmarks.applychanges(repo, tr, bmarkchanges)
 
         for phase, nodes in toretract.items():
             phases.retractboundary(repo, tr, phase, nodes)
         for phase, nodes in toadvance.items():
             phases.advanceboundary(repo, tr, phase, nodes)
 
         # Obsolete or strip nodes
         if obsolete.isenabled(repo, obsolete.createmarkersopt):
             # If a node is already obsoleted, and we want to obsolete it
             # without a successor, skip that obssolete request since it's
             # unnecessary. That's the "if s or not isobs(n)" check below.
             # Also sort the node in topology order, that might be useful for
             # some obsstore logic.
             # NOTE: the sorting might belong to createmarkers.
             torev = unfi.changelog.rev
             sortfunc = lambda ns: torev(ns[0][0])
             rels = []
             for ns, s in sorted(replacements.items(), key=sortfunc):
                 rel = (tuple(unfi[n] for n in ns), tuple(unfi[m] for m in s))
                 rels.append(rel)
             if rels:
                 obsolete.createmarkers(repo, rels, operation=operation,
                                        metadata=metadata)
         else:
             from . import repair # avoid import cycle
             tostrip = list(n for ns in replacements for n in ns)
             if tostrip:
                 repair.delayedstrip(repo.ui, repo, tostrip, operation,
                                     backup=backup)
 
 def addremove(repo, matcher, prefix, opts=None):
     if opts is None:
         opts = {}
     m = matcher
     dry_run = opts.get('dry_run')
     try:
         similarity = float(opts.get('similarity') or 0)
     except ValueError:
         raise error.Abort(_('similarity must be a number'))
     if similarity < 0 or similarity > 100:
         raise error.Abort(_('similarity must be between 0 and 100'))
     similarity /= 100.0
 
     ret = 0
     join = lambda f: os.path.join(prefix, f)
 
     wctx = repo[None]
     for subpath in sorted(wctx.substate):
         submatch = matchmod.subdirmatcher(subpath, m)
         if opts.get('subrepos') or m.exact(subpath) or any(submatch.files()):
             sub = wctx.sub(subpath)
+            subprefix = repo.wvfs.reljoin(prefix, subpath)
             try:
-                if sub.addremove(submatch, prefix, opts):
+                if sub.addremove(submatch, subprefix, opts):
                     ret = 1
             except error.LookupError:
                 repo.ui.status(_("skipping missing subrepository: %s\n")
                                  % join(subpath))
 
     rejected = []
     def badfn(f, msg):
         if f in m.files():
             m.bad(f, msg)
         rejected.append(f)
 
     badmatch = matchmod.badmatch(m, badfn)
     added, unknown, deleted, removed, forgotten = _interestingfiles(repo,
                                                                     badmatch)
 
     unknownset = set(unknown + forgotten)
     toprint = unknownset.copy()
     toprint.update(deleted)
     for abs in sorted(toprint):
         if repo.ui.verbose or not m.exact(abs):
             if abs in unknownset:
                 status = _('adding %s\n') % m.uipath(abs)
                 label = 'ui.addremove.added'
             else:
                 status = _('removing %s\n') % m.uipath(abs)
                 label = 'ui.addremove.removed'
             repo.ui.status(status, label=label)
 
     renames = _findrenames(repo, m, added + unknown, removed + deleted,
                            similarity)
 
     if not dry_run:
         _markchanges(repo, unknown + forgotten, deleted, renames)
 
     for f in rejected:
         if f in m.files():
             return 1
     return ret
 
 def marktouched(repo, files, similarity=0.0):
     '''Assert that files have somehow been operated upon. files are relative to
     the repo root.'''
     m = matchfiles(repo, files, badfn=lambda x, y: rejected.append(x))
     rejected = []
 
     added, unknown, deleted, removed, forgotten = _interestingfiles(repo, m)
 
     if repo.ui.verbose:
         unknownset = set(unknown + forgotten)
         toprint = unknownset.copy()
         toprint.update(deleted)
         for abs in sorted(toprint):
             if abs in unknownset:
                 status = _('adding %s\n') % abs
             else:
                 status = _('removing %s\n') % abs
             repo.ui.status(status)
 
     renames = _findrenames(repo, m, added + unknown, removed + deleted,
                            similarity)
 
     _markchanges(repo, unknown + forgotten, deleted, renames)
 
     for f in rejected:
         if f in m.files():
             return 1
     return 0
 
 def _interestingfiles(repo, matcher):
     '''Walk dirstate with matcher, looking for files that addremove would care
     about.
 
     This is different from dirstate.status because it doesn't care about
     whether files are modified or clean.'''
     added, unknown, deleted, removed, forgotten = [], [], [], [], []
     audit_path = pathutil.pathauditor(repo.root, cached=True)
 
     ctx = repo[None]
     dirstate = repo.dirstate
     matcher = repo.narrowmatch(matcher, includeexact=True)
     walkresults = dirstate.walk(matcher, subrepos=sorted(ctx.substate),
                                 unknown=True, ignored=False, full=False)
     for abs, st in walkresults.iteritems():
         dstate = dirstate[abs]
         if dstate == '?' and audit_path.check(abs):
             unknown.append(abs)
         elif dstate != 'r' and not st:
             deleted.append(abs)
         elif dstate == 'r' and st:
             forgotten.append(abs)
         # for finding renames
         elif dstate == 'r' and not st:
             removed.append(abs)
         elif dstate == 'a':
             added.append(abs)
 
     return added, unknown, deleted, removed, forgotten
 
 def _findrenames(repo, matcher, added, removed, similarity):
     '''Find renames from removed files to added ones.'''
     renames = {}
     if similarity > 0:
         for old, new, score in similar.findrenames(repo, added, removed,
                                                    similarity):
             if (repo.ui.verbose or not matcher.exact(old)
                 or not matcher.exact(new)):
                 repo.ui.status(_('recording removal of %s as rename to %s '
                                  '(%d%% similar)\n') %
                                (matcher.rel(old), matcher.rel(new),
                                 score * 100))
             renames[new] = old
     return renames
 
 def _markchanges(repo, unknown, deleted, renames):
     '''Marks the files in unknown as added, the files in deleted as removed,
     and the files in renames as copied.'''
     wctx = repo[None]
     with repo.wlock():
         wctx.forget(deleted)
         wctx.add(unknown)
         for new, old in renames.iteritems():
             wctx.copy(old, new)
 
 def dirstatecopy(ui, repo, wctx, src, dst, dryrun=False, cwd=None):
     """Update the dirstate to reflect the intent of copying src to dst. For
     different reasons it might not end with dst being marked as copied from src.
     """
     origsrc = repo.dirstate.copied(src) or src
     if dst == origsrc: # copying back a copy?
         if repo.dirstate[dst] not in 'mn' and not dryrun:
             repo.dirstate.normallookup(dst)
     else:
         if repo.dirstate[origsrc] == 'a' and origsrc == src:
             if not ui.quiet:
                 ui.warn(_("%s has not been committed yet, so no copy "
                           "data will be stored for %s.\n")
                         % (repo.pathto(origsrc, cwd), repo.pathto(dst, cwd)))
             if repo.dirstate[dst] in '?r' and not dryrun:
                 wctx.add([dst])
         elif not dryrun:
             wctx.copy(origsrc, dst)
 
 def writerequires(opener, requirements):
     with opener('requires', 'w', atomictemp=True) as fp:
         for r in sorted(requirements):
             fp.write("%s\n" % r)
 
 class filecachesubentry(object):
     def __init__(self, path, stat):
         self.path = path
         self.cachestat = None
         self._cacheable = None
 
         if stat:
             self.cachestat = filecachesubentry.stat(self.path)
 
             if self.cachestat:
                 self._cacheable = self.cachestat.cacheable()
             else:
                 # None means we don't know yet
                 self._cacheable = None
 
     def refresh(self):
         if self.cacheable():
             self.cachestat = filecachesubentry.stat(self.path)
 
     def cacheable(self):
         if self._cacheable is not None:
             return self._cacheable
 
         # we don't know yet, assume it is for now
         return True
 
     def changed(self):
         # no point in going further if we can't cache it
         if not self.cacheable():
             return True
 
         newstat = filecachesubentry.stat(self.path)
 
         # we may not know if it's cacheable yet, check again now
         if newstat and self._cacheable is None:
             self._cacheable = newstat.cacheable()
 
             # check again
             if not self._cacheable:
                 return True
 
         if self.cachestat != newstat:
             self.cachestat = newstat
             return True
         else:
             return False
 
     @staticmethod
     def stat(path):
         try:
             return util.cachestat(path)
         except OSError as e:
             if e.errno != errno.ENOENT:
                 raise
 
 class filecacheentry(object):
     def __init__(self, paths, stat=True):
         self._entries = []
         for path in paths:
             self._entries.append(filecachesubentry(path, stat))
 
     def changed(self):
         '''true if any entry has changed'''
         for entry in self._entries:
             if entry.changed():
                 return True
         return False
 
     def refresh(self):
         for entry in self._entries:
             entry.refresh()
 
 class filecache(object):
     """A property like decorator that tracks files under .hg/ for updates.
 
     On first access, the files defined as arguments are stat()ed and the
     results cached. The decorated function is called. The results are stashed
     away in a ``_filecache`` dict on the object whose method is decorated.
 
     On subsequent access, the cached result is used as it is set to the
     instance dictionary.
 
     On external property set/delete operations, the caller must update the
     corresponding _filecache entry appropriately. Use __class__.<attr>.set()
     instead of directly setting <attr>.
 
     When using the property API, the cached data is always used if available.
     No stat() is performed to check if the file has changed.
 
     Others can muck about with the state of the ``_filecache`` dict. e.g. they
     can populate an entry before the property's getter is called. In this case,
     entries in ``_filecache`` will be used during property operations,
     if available. If the underlying file changes, it is up to external callers
     to reflect this by e.g. calling ``delattr(obj, attr)`` to remove the cached
     method result as well as possibly calling ``del obj._filecache[attr]`` to
     remove the ``filecacheentry``.
     """
 
     def __init__(self, *paths):
         self.paths = paths
 
     def join(self, obj, fname):
         """Used to compute the runtime path of a cached file.
 
         Users should subclass filecache and provide their own version of this
         function to call the appropriate join function on 'obj' (an instance
         of the class that its member function was decorated).
         """
         raise NotImplementedError
 
     def __call__(self, func):
         self.func = func
         self.sname = func.__name__
         self.name = pycompat.sysbytes(self.sname)
         return self
 
     def __get__(self, obj, type=None):
         # if accessed on the class, return the descriptor itself.
         if obj is None:
             return self
 
         assert self.sname not in obj.__dict__
 
         entry = obj._filecache.get(self.name)
 
         if entry:
             if entry.changed():
                 entry.obj = self.func(obj)
         else:
             paths = [self.join(obj, path) for path in self.paths]
 
             # We stat -before- creating the object so our cache doesn't lie if
             # a writer modified between the time we read and stat
             entry = filecacheentry(paths, True)
             entry.obj = self.func(obj)
 
             obj._filecache[self.name] = entry
 
         obj.__dict__[self.sname] = entry.obj
         return entry.obj
 
     # don't implement __set__(), which would make __dict__ lookup as slow as
     # function call.
 
     def set(self, obj, value):
         if self.name not in obj._filecache:
             # we add an entry for the missing value because X in __dict__
             # implies X in _filecache
             paths = [self.join(obj, path) for path in self.paths]
             ce = filecacheentry(paths, False)
             obj._filecache[self.name] = ce
         else:
             ce = obj._filecache[self.name]
 
         ce.obj = value # update cached copy
         obj.__dict__[self.sname] = value # update copy returned by obj.x
 
 def extdatasource(repo, source):
     """Gather a map of rev -> value dict from the specified source
 
     A source spec is treated as a URL, with a special case shell: type
     for parsing the output from a shell command.
 
     The data is parsed as a series of newline-separated records where
     each record is a revision specifier optionally followed by a space
     and a freeform string value. If the revision is known locally, it
     is converted to a rev, otherwise the record is skipped.
 
     Note that both key and value are treated as UTF-8 and converted to
     the local encoding. This allows uniformity between local and
     remote data sources.
     """
 
     spec = repo.ui.config("extdata", source)
     if not spec:
         raise error.Abort(_("unknown extdata source '%s'") % source)
 
     data = {}
     src = proc = None
     try:
         if spec.startswith("shell:"):
             # external commands should be run relative to the repo root
             cmd = spec[6:]
             proc = subprocess.Popen(procutil.tonativestr(cmd),
                                     shell=True, bufsize=-1,
                                     close_fds=procutil.closefds,
                                     stdout=subprocess.PIPE,
                                     cwd=procutil.tonativestr(repo.root))
             src = proc.stdout
         else:
             # treat as a URL or file
             src = url.open(repo.ui, spec)
         for l in src:
             if " " in l:
                 k, v = l.strip().split(" ", 1)
             else:
                 k, v = l.strip(), ""
 
             k = encoding.tolocal(k)
             try:
                 data[revsingle(repo, k).rev()] = encoding.tolocal(v)
             except (error.LookupError, error.RepoLookupError):
                 pass # we ignore data for nodes that don't exist locally
     finally:
         if proc:
             proc.communicate()
         if src:
             src.close()
     if proc and proc.returncode != 0:
         raise error.Abort(_("extdata command '%s' failed: %s")
                           % (cmd, procutil.explainexit(proc.returncode)))
 
     return data
 
 def _locksub(repo, lock, envvar, cmd, environ=None, *args, **kwargs):
     if lock is None:
         raise error.LockInheritanceContractViolation(
             'lock can only be inherited while held')
     if environ is None:
         environ = {}
     with lock.inherit() as locker:
         environ[envvar] = locker
         return repo.ui.system(cmd, environ=environ, *args, **kwargs)
 
 def wlocksub(repo, cmd, *args, **kwargs):
     """run cmd as a subprocess that allows inheriting repo's wlock
 
     This can only be called while the wlock is held. This takes all the
     arguments that ui.system does, and returns the exit code of the
     subprocess."""
     return _locksub(repo, repo.currentwlock(), 'HG_WLOCK_LOCKER', cmd, *args,
                     **kwargs)
 
 class progress(object):
     def __init__(self, ui, updatebar, topic, unit="", total=None):
         self.ui = ui
         self.pos = 0
         self.topic = topic
         self.unit = unit
         self.total = total
         self.debug = ui.configbool('progress', 'debug')
         self._updatebar = updatebar
 
     def __enter__(self):
         return self
 
     def __exit__(self, exc_type, exc_value, exc_tb):
         self.complete()
 
     def update(self, pos, item="", total=None):
         assert pos is not None
         if total:
             self.total = total
         self.pos = pos
         self._updatebar(self.topic, self.pos, item, self.unit, self.total)
         if self.debug:
             self._printdebug(item)
 
     def increment(self, step=1, item="", total=None):
         self.update(self.pos + step, item, total)
 
     def complete(self):
         self.pos = None
         self.unit = ""
         self.total = None
         self._updatebar(self.topic, self.pos, "", self.unit, self.total)
 
     def _printdebug(self, item):
         if self.unit:
             unit = ' ' + self.unit
         if item:
             item = ' ' + item
 
         if self.total:
             pct = 100.0 * self.pos / self.total
             self.ui.debug('%s:%s %d/%d%s (%4.2f%%)\n'
                        % (self.topic, item, self.pos, self.total, unit, pct))
         else:
             self.ui.debug('%s:%s %d%s\n' % (self.topic, item, self.pos, unit))
 
 def gdinitconfig(ui):
     """helper function to know if a repo should be created as general delta
     """
     # experimental config: format.generaldelta
     return (ui.configbool('format', 'generaldelta')
             or ui.configbool('format', 'usegeneraldelta'))
 
 def gddeltaconfig(ui):
     """helper function to know if incoming delta should be optimised
     """
     # experimental config: format.generaldelta
     return ui.configbool('format', 'generaldelta')
 
 class simplekeyvaluefile(object):
     """A simple file with key=value lines
 
     Keys must be alphanumerics and start with a letter, values must not
     contain '\n' characters"""
     firstlinekey = '__firstline'
 
     def __init__(self, vfs, path, keys=None):
         self.vfs = vfs
         self.path = path
 
     def read(self, firstlinenonkeyval=False):
         """Read the contents of a simple key-value file
 
         'firstlinenonkeyval' indicates whether the first line of file should
         be treated as a key-value pair or reuturned fully under the
         __firstline key."""
         lines = self.vfs.readlines(self.path)
         d = {}
         if firstlinenonkeyval:
             if not lines:
                 e = _("empty simplekeyvalue file")
                 raise error.CorruptedState(e)
             # we don't want to include '\n' in the __firstline
             d[self.firstlinekey] = lines[0][:-1]
             del lines[0]
 
         try:
             # the 'if line.strip()' part prevents us from failing on empty
             # lines which only contain '\n' therefore are not skipped
             # by 'if line'
             updatedict = dict(line[:-1].split('=', 1) for line in lines
                                                       if line.strip())
             if self.firstlinekey in updatedict:
                 e = _("%r can't be used as a key")
                 raise error.CorruptedState(e % self.firstlinekey)
             d.update(updatedict)
         except ValueError as e:
             raise error.CorruptedState(str(e))
         return d
 
     def write(self, data, firstline=None):
         """Write key=>value mapping to a file
         data is a dict. Keys must be alphanumerical and start with a letter.
         Values must not contain newline characters.
 
         If 'firstline' is not None, it is written to file before
         everything else, as it is, not in a key=value form"""
         lines = []
         if firstline is not None:
             lines.append('%s\n' % firstline)
 
         for k, v in data.items():
             if k == self.firstlinekey:
                 e = "key name '%s' is reserved" % self.firstlinekey
                 raise error.ProgrammingError(e)
             if not k[0:1].isalpha():
                 e = "keys must start with a letter in a key-value file"
                 raise error.ProgrammingError(e)
             if not k.isalnum():
                 e = "invalid key name in a simple key-value file"
                 raise error.ProgrammingError(e)
             if '\n' in v:
                 e = "invalid value in a simple key-value file"
                 raise error.ProgrammingError(e)
             lines.append("%s=%s\n" % (k, v))
         with self.vfs(self.path, mode='wb', atomictemp=True) as fp:
             fp.write(''.join(lines))
 
 _reportobsoletedsource = [
     'debugobsolete',
     'pull',
     'push',
     'serve',
     'unbundle',
 ]
 
 _reportnewcssource = [
     'pull',
     'unbundle',
 ]
 
 def prefetchfiles(repo, revs, match):
     """Invokes the registered file prefetch functions, allowing extensions to
     ensure the corresponding files are available locally, before the command
     uses them."""
     if match:
         # The command itself will complain about files that don't exist, so
         # don't duplicate the message.
         match = matchmod.badmatch(match, lambda fn, msg: None)
     else:
         match = matchall(repo)
 
     fileprefetchhooks(repo, revs, match)
 
 # a list of (repo, revs, match) prefetch functions
 fileprefetchhooks = util.hooks()
 
 # A marker that tells the evolve extension to suppress its own reporting
 _reportstroubledchangesets = True
 
 def registersummarycallback(repo, otr, txnname=''):
     """register a callback to issue a summary after the transaction is closed
     """
     def txmatch(sources):
         return any(txnname.startswith(source) for source in sources)
 
     categories = []
 
     def reportsummary(func):
         """decorator for report callbacks."""
         # The repoview life cycle is shorter than the one of the actual
         # underlying repository. So the filtered object can die before the
         # weakref is used leading to troubles. We keep a reference to the
         # unfiltered object and restore the filtering when retrieving the
         # repository through the weakref.
         filtername = repo.filtername
         reporef = weakref.ref(repo.unfiltered())
         def wrapped(tr):
             repo = reporef()
             if filtername:
                 repo = repo.filtered(filtername)
             func(repo, tr)
         newcat = '%02i-txnreport' % len(categories)
         otr.addpostclose(newcat, wrapped)
         categories.append(newcat)
         return wrapped
 
     if txmatch(_reportobsoletedsource):
         @reportsummary
         def reportobsoleted(repo, tr):
             obsoleted = obsutil.getobsoleted(repo, tr)
             if obsoleted:
                 repo.ui.status(_('obsoleted %i changesets\n')
                                % len(obsoleted))
 
     if (obsolete.isenabled(repo, obsolete.createmarkersopt) and
         repo.ui.configbool('experimental', 'evolution.report-instabilities')):
         instabilitytypes = [
             ('orphan', 'orphan'),
             ('phase-divergent', 'phasedivergent'),
             ('content-divergent', 'contentdivergent'),
         ]
 
         def getinstabilitycounts(repo):
             filtered = repo.changelog.filteredrevs
             counts = {}
             for instability, revset in instabilitytypes:
                 counts[instability] = len(set(obsolete.getrevs(repo, revset)) -
                                           filtered)
             return counts
 
         oldinstabilitycounts = getinstabilitycounts(repo)
         @reportsummary
         def reportnewinstabilities(repo, tr):
             newinstabilitycounts = getinstabilitycounts(repo)
             for instability, revset in instabilitytypes:
                 delta = (newinstabilitycounts[instability] -
                          oldinstabilitycounts[instability])
                 msg = getinstabilitymessage(delta, instability)
                 if msg:
                     repo.ui.warn(msg)
 
     if txmatch(_reportnewcssource):
         @reportsummary
         def reportnewcs(repo, tr):
             """Report the range of new revisions pulled/unbundled."""
             origrepolen = tr.changes.get('origrepolen', len(repo))
             unfi = repo.unfiltered()
             if origrepolen >= len(unfi):
                 return
 
             # Compute the bounds of new visible revisions' range.
             revs = smartset.spanset(repo, start=origrepolen)
             if revs:
                 minrev, maxrev = repo[revs.min()], repo[revs.max()]
 
                 if minrev == maxrev:
                     revrange = minrev
                 else:
                     revrange = '%s:%s' % (minrev, maxrev)
                 draft = len(repo.revs('%ld and draft()', revs))
                 secret = len(repo.revs('%ld and secret()', revs))
                 if not (draft or secret):
                     msg = _('new changesets %s\n') % revrange
                 elif draft and secret:
                     msg = _('new changesets %s (%d drafts, %d secrets)\n')
                     msg %= (revrange, draft, secret)
                 elif draft:
                     msg = _('new changesets %s (%d drafts)\n')
                     msg %= (revrange, draft)
                 elif secret:
                     msg = _('new changesets %s (%d secrets)\n')
                     msg %= (revrange, secret)
                 else:
                     errormsg = 'entered unreachable condition'
                     raise error.ProgrammingError(errormsg)
                 repo.ui.status(msg)
 
             # search new changesets directly pulled as obsolete
             duplicates = tr.changes.get('revduplicates', ())
             obsadded = unfi.revs('(%d: + %ld) and obsolete()',
                                  origrepolen, duplicates)
             cl = repo.changelog
             extinctadded = [r for r in obsadded if r not in cl]
             if extinctadded:
                 # They are not just obsolete, but obsolete and invisible
                 # we call them "extinct" internally but the terms have not been
                 # exposed to users.
                 msg = '(%d other changesets obsolete on arrival)\n'
                 repo.ui.status(msg % len(extinctadded))
 
         @reportsummary
         def reportphasechanges(repo, tr):
             """Report statistics of phase changes for changesets pre-existing
             pull/unbundle.
             """
             origrepolen = tr.changes.get('origrepolen', len(repo))
             phasetracking = tr.changes.get('phases', {})
             if not phasetracking:
                 return
             published = [
                 rev for rev, (old, new) in phasetracking.iteritems()
                 if new == phases.public and rev < origrepolen
             ]
             if not published:
                 return
             repo.ui.status(_('%d local changesets published\n')
                            % len(published))
 
 def getinstabilitymessage(delta, instability):
     """function to return the message to show warning about new instabilities
 
     exists as a separate function so that extension can wrap to show more
     information like how to fix instabilities"""
     if delta > 0:
         return _('%i new %s changesets\n') % (delta, instability)
 
 def nodesummaries(repo, nodes, maxnumnodes=4):
     if len(nodes) <= maxnumnodes or repo.ui.verbose:
         return ' '.join(short(h) for h in nodes)
     first = ' '.join(short(h) for h in nodes[:maxnumnodes])
     return _("%s and %d others") % (first, len(nodes) - maxnumnodes)
 
 def enforcesinglehead(repo, tr, desc):
     """check that no named branch has multiple heads"""
     if desc in ('strip', 'repair'):
         # skip the logic during strip
         return
     visible = repo.filtered('visible')
     # possible improvement: we could restrict the check to affected branch
     for name, heads in visible.branchmap().iteritems():
         if len(heads) > 1:
             msg = _('rejecting multiple heads on branch "%s"')
             msg %= name
             hint = _('%d heads: %s')
             hint %= (len(heads), nodesummaries(repo, heads))
             raise error.Abort(msg, hint=hint)
 
 def wrapconvertsink(sink):
     """Allow extensions to wrap the sink returned by convcmd.convertsink()
     before it is used, whether or not the convert extension was formally loaded.
     """
     return sink
 
 def unhidehashlikerevs(repo, specs, hiddentype):
     """parse the user specs and unhide changesets whose hash or revision number
     is passed.
 
     hiddentype can be: 1) 'warn': warn while unhiding changesets
                        2) 'nowarn': don't warn while unhiding changesets
 
     returns a repo object with the required changesets unhidden
     """
     if not repo.filtername or not repo.ui.configbool('experimental',
                                                      'directaccess'):
         return repo
 
     if repo.filtername not in ('visible', 'visible-hidden'):
         return repo
 
     symbols = set()
     for spec in specs:
         try:
             tree = revsetlang.parse(spec)
         except error.ParseError: # will be reported by scmutil.revrange()
             continue
 
         symbols.update(revsetlang.gethashlikesymbols(tree))
 
     if not symbols:
         return repo
 
     revs = _getrevsfromsymbols(repo, symbols)
 
     if not revs:
         return repo
 
     if hiddentype == 'warn':
         unfi = repo.unfiltered()
         revstr = ", ".join([pycompat.bytestr(unfi[l]) for l in revs])
         repo.ui.warn(_("warning: accessing hidden changesets for write "
                        "operation: %s\n") % revstr)
 
     # we have to use new filtername to separate branch/tags cache until we can
     # disbale these cache when revisions are dynamically pinned.
     return repo.filtered('visible-hidden', revs)
 
 def _getrevsfromsymbols(repo, symbols):
     """parse the list of symbols and returns a set of revision numbers of hidden
     changesets present in symbols"""
     revs = set()
     unfi = repo.unfiltered()
     unficl = unfi.changelog
     cl = repo.changelog
     tiprev = len(unficl)
     allowrevnums = repo.ui.configbool('experimental', 'directaccess.revnums')
     for s in symbols:
         try:
             n = int(s)
             if n <= tiprev:
                 if not allowrevnums:
                     continue
                 else:
                     if n not in cl:
                         revs.add(n)
                     continue
         except ValueError:
             pass
 
         try:
             s = resolvehexnodeidprefix(unfi, s)
         except (error.LookupError, error.WdirUnsupported):
             s = None
 
         if s is not None:
             rev = unficl.rev(s)
             if rev not in cl:
                 revs.add(rev)
 
     return revs
 
 def bookmarkrevs(repo, mark):
     """
     Select revisions reachable by a given bookmark
     """
     return repo.revs("ancestors(bookmark(%s)) - "
                      "ancestors(head() and not bookmark(%s)) - "
                      "ancestors(bookmark() and not bookmark(%s))",
                      mark, mark, mark)

mercurial/subrepo.py

@@ -1,1844 +1,1843 b''
 # subrepo.py - sub-repository classes and factory
 #
 # Copyright 2009-2010 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 copy
 import errno
 import hashlib
 import os
 import posixpath
 import re
 import stat
 import subprocess
 import sys
 import tarfile
 import xml.dom.minidom
 
 from .i18n import _
 from . import (
     cmdutil,
     encoding,
     error,
     exchange,
     logcmdutil,
     match as matchmod,
     node,
     pathutil,
     phases,
     pycompat,
     scmutil,
     subrepoutil,
     util,
     vfs as vfsmod,
 )
 from .utils import (
     dateutil,
     procutil,
     stringutil,
 )
 
 hg = None
 reporelpath = subrepoutil.reporelpath
 subrelpath = subrepoutil.subrelpath
 _abssource = subrepoutil._abssource
 propertycache = util.propertycache
 
 def _expandedabspath(path):
     '''
     get a path or url and if it is a path expand it and return an absolute path
     '''
     expandedpath = util.urllocalpath(util.expandpath(path))
     u = util.url(expandedpath)
     if not u.scheme:
         path = util.normpath(os.path.abspath(u.path))
     return path
 
 def _getstorehashcachename(remotepath):
     '''get a unique filename for the store hash cache of a remote repository'''
     return node.hex(hashlib.sha1(_expandedabspath(remotepath)).digest())[0:12]
 
 class SubrepoAbort(error.Abort):
     """Exception class used to avoid handling a subrepo error more than once"""
     def __init__(self, *args, **kw):
         self.subrepo = kw.pop(r'subrepo', None)
         self.cause = kw.pop(r'cause', None)
         error.Abort.__init__(self, *args, **kw)
 
 def annotatesubrepoerror(func):
     def decoratedmethod(self, *args, **kargs):
         try:
             res = func(self, *args, **kargs)
         except SubrepoAbort as ex:
             # This exception has already been handled
             raise ex
         except error.Abort as ex:
             subrepo = subrelpath(self)
             errormsg = (stringutil.forcebytestr(ex) + ' '
                         + _('(in subrepository "%s")') % subrepo)
             # avoid handling this exception by raising a SubrepoAbort exception
             raise SubrepoAbort(errormsg, hint=ex.hint, subrepo=subrepo,
                                cause=sys.exc_info())
         return res
     return decoratedmethod
 
 def _updateprompt(ui, sub, dirty, local, remote):
     if dirty:
         msg = (_(' subrepository sources for %s differ\n'
                  'use (l)ocal source (%s) or (r)emote source (%s)?'
                  '$$ &Local $$ &Remote')
                % (subrelpath(sub), local, remote))
     else:
         msg = (_(' subrepository sources for %s differ (in checked out '
                  'version)\n'
                  'use (l)ocal source (%s) or (r)emote source (%s)?'
                  '$$ &Local $$ &Remote')
                % (subrelpath(sub), local, remote))
     return ui.promptchoice(msg, 0)
 
 def _sanitize(ui, vfs, ignore):
     for dirname, dirs, names in vfs.walk():
         for i, d in enumerate(dirs):
             if d.lower() == ignore:
                 del dirs[i]
                 break
         if vfs.basename(dirname).lower() != '.hg':
             continue
         for f in names:
             if f.lower() == 'hgrc':
                 ui.warn(_("warning: removing potentially hostile 'hgrc' "
                           "in '%s'\n") % vfs.join(dirname))
                 vfs.unlink(vfs.reljoin(dirname, f))
 
 def _auditsubrepopath(repo, path):
     # sanity check for potentially unsafe paths such as '~' and '$FOO'
     if path.startswith('~') or '$' in path or util.expandpath(path) != path:
         raise error.Abort(_('subrepo path contains illegal component: %s')
                           % path)
     # auditor doesn't check if the path itself is a symlink
     pathutil.pathauditor(repo.root)(path)
     if repo.wvfs.islink(path):
         raise error.Abort(_("subrepo '%s' traverses symbolic link") % path)
 
 SUBREPO_ALLOWED_DEFAULTS = {
     'hg': True,
     'git': False,
     'svn': False,
 }
 
 def _checktype(ui, kind):
     # subrepos.allowed is a master kill switch. If disabled, subrepos are
     # disabled period.
     if not ui.configbool('subrepos', 'allowed', True):
         raise error.Abort(_('subrepos not enabled'),
                           hint=_("see 'hg help config.subrepos' for details"))
 
     default = SUBREPO_ALLOWED_DEFAULTS.get(kind, False)
     if not ui.configbool('subrepos', '%s:allowed' % kind, default):
         raise error.Abort(_('%s subrepos not allowed') % kind,
                           hint=_("see 'hg help config.subrepos' for details"))
 
     if kind not in types:
         raise error.Abort(_('unknown subrepo type %s') % kind)
 
 def subrepo(ctx, path, allowwdir=False, allowcreate=True):
     """return instance of the right subrepo class for subrepo in path"""
     # subrepo inherently violates our import layering rules
     # because it wants to make repo objects from deep inside the stack
     # so we manually delay the circular imports to not break
     # scripts that don't use our demand-loading
     global hg
     from . import hg as h
     hg = h
 
     repo = ctx.repo()
     _auditsubrepopath(repo, path)
     state = ctx.substate[path]
     _checktype(repo.ui, state[2])
     if allowwdir:
         state = (state[0], ctx.subrev(path), state[2])
     return types[state[2]](ctx, path, state[:2], allowcreate)
 
 def nullsubrepo(ctx, path, pctx):
     """return an empty subrepo in pctx for the extant subrepo in ctx"""
     # subrepo inherently violates our import layering rules
     # because it wants to make repo objects from deep inside the stack
     # so we manually delay the circular imports to not break
     # scripts that don't use our demand-loading
     global hg
     from . import hg as h
     hg = h
 
     repo = ctx.repo()
     _auditsubrepopath(repo, path)
     state = ctx.substate[path]
     _checktype(repo.ui, state[2])
     subrev = ''
     if state[2] == 'hg':
         subrev = "0" * 40
     return types[state[2]](pctx, path, (state[0], subrev), True)
 
 # subrepo classes need to implement the following abstract class:
 
 class abstractsubrepo(object):
 
     def __init__(self, ctx, path):
         """Initialize abstractsubrepo part
 
         ``ctx`` is the context referring this subrepository in the
         parent repository.
 
         ``path`` is the path to this subrepository as seen from
         innermost repository.
         """
         self.ui = ctx.repo().ui
         self._ctx = ctx
         self._path = path
 
     def addwebdirpath(self, serverpath, webconf):
         """Add the hgwebdir entries for this subrepo, and any of its subrepos.
 
         ``serverpath`` is the path component of the URL for this repo.
 
         ``webconf`` is the dictionary of hgwebdir entries.
         """
         pass
 
     def storeclean(self, path):
         """
         returns true if the repository has not changed since it was last
         cloned from or pushed to a given repository.
         """
         return False
 
     def dirty(self, ignoreupdate=False, missing=False):
         """returns true if the dirstate of the subrepo is dirty or does not
         match current stored state. If ignoreupdate is true, only check
         whether the subrepo has uncommitted changes in its dirstate.  If missing
         is true, check for deleted files.
         """
         raise NotImplementedError
 
     def dirtyreason(self, ignoreupdate=False, missing=False):
         """return reason string if it is ``dirty()``
 
         Returned string should have enough information for the message
         of exception.
 
         This returns None, otherwise.
         """
         if self.dirty(ignoreupdate=ignoreupdate, missing=missing):
             return _('uncommitted changes in subrepository "%s"'
                      ) % subrelpath(self)
 
     def bailifchanged(self, ignoreupdate=False, hint=None):
         """raise Abort if subrepository is ``dirty()``
         """
         dirtyreason = self.dirtyreason(ignoreupdate=ignoreupdate,
                                        missing=True)
         if dirtyreason:
             raise error.Abort(dirtyreason, hint=hint)
 
     def basestate(self):
         """current working directory base state, disregarding .hgsubstate
         state and working directory modifications"""
         raise NotImplementedError
 
     def checknested(self, path):
         """check if path is a subrepository within this repository"""
         return False
 
     def commit(self, text, user, date):
         """commit the current changes to the subrepo with the given
         log message. Use given user and date if possible. Return the
         new state of the subrepo.
         """
         raise NotImplementedError
 
     def phase(self, state):
         """returns phase of specified state in the subrepository.
         """
         return phases.public
 
     def remove(self):
         """remove the subrepo
 
         (should verify the dirstate is not dirty first)
         """
         raise NotImplementedError
 
     def get(self, state, overwrite=False):
         """run whatever commands are needed to put the subrepo into
         this state
         """
         raise NotImplementedError
 
     def merge(self, state):
         """merge currently-saved state with the new state."""
         raise NotImplementedError
 
     def push(self, opts):
         """perform whatever action is analogous to 'hg push'
 
         This may be a no-op on some systems.
         """
         raise NotImplementedError
 
     def add(self, ui, match, prefix, explicitonly, **opts):
         return []
 
     def addremove(self, matcher, prefix, opts):
         self.ui.warn("%s: %s" % (prefix, _("addremove is not supported")))
         return 1
 
     def cat(self, match, fm, fntemplate, prefix, **opts):
         return 1
 
     def status(self, rev2, **opts):
         return scmutil.status([], [], [], [], [], [], [])
 
     def diff(self, ui, diffopts, node2, match, prefix, **opts):
         pass
 
     def outgoing(self, ui, dest, opts):
         return 1
 
     def incoming(self, ui, source, opts):
         return 1
 
     def files(self):
         """return filename iterator"""
         raise NotImplementedError
 
     def filedata(self, name, decode):
         """return file data, optionally passed through repo decoders"""
         raise NotImplementedError
 
     def fileflags(self, name):
         """return file flags"""
         return ''
 
     def matchfileset(self, expr, badfn=None):
         """Resolve the fileset expression for this repo"""
         return matchmod.nevermatcher(self.wvfs.base, '', badfn=badfn)
 
     def printfiles(self, ui, m, fm, fmt, subrepos):
         """handle the files command for this subrepo"""
         return 1
 
     def archive(self, archiver, prefix, match=None, decode=True):
         if match is not None:
             files = [f for f in self.files() if match(f)]
         else:
             files = self.files()
         total = len(files)
         relpath = subrelpath(self)
         progress = self.ui.makeprogress(_('archiving (%s)') % relpath,
                                         unit=_('files'), total=total)
         progress.update(0)
         for name in files:
             flags = self.fileflags(name)
             mode = 'x' in flags and 0o755 or 0o644
             symlink = 'l' in flags
             archiver.addfile(prefix + self._path + '/' + name,
                              mode, symlink, self.filedata(name, decode))
             progress.increment()
         progress.complete()
         return total
 
     def walk(self, match):
         '''
         walk recursively through the directory tree, finding all files
         matched by the match function
         '''
 
     def forget(self, match, prefix, dryrun, interactive):
         return ([], [])
 
     def removefiles(self, matcher, prefix, after, force, subrepos,
                     dryrun, warnings):
         """remove the matched files from the subrepository and the filesystem,
         possibly by force and/or after the file has been removed from the
         filesystem.  Return 0 on success, 1 on any warning.
         """
         warnings.append(_("warning: removefiles not implemented (%s)")
                         % self._path)
         return 1
 
     def revert(self, substate, *pats, **opts):
         self.ui.warn(_('%s: reverting %s subrepos is unsupported\n') \
             % (substate[0], substate[2]))
         return []
 
     def shortid(self, revid):
         return revid
 
     def unshare(self):
         '''
         convert this repository from shared to normal storage.
         '''
 
     def verify(self):
         '''verify the integrity of the repository.  Return 0 on success or
         warning, 1 on any error.
         '''
         return 0
 
     @propertycache
     def wvfs(self):
         """return vfs to access the working directory of this subrepository
         """
         return vfsmod.vfs(self._ctx.repo().wvfs.join(self._path))
 
     @propertycache
     def _relpath(self):
         """return path to this subrepository as seen from outermost repository
         """
         return self.wvfs.reljoin(reporelpath(self._ctx.repo()), self._path)
 
 class hgsubrepo(abstractsubrepo):
     def __init__(self, ctx, path, state, allowcreate):
         super(hgsubrepo, self).__init__(ctx, path)
         self._state = state
         r = ctx.repo()
         root = r.wjoin(path)
         create = allowcreate and not r.wvfs.exists('%s/.hg' % path)
         # repository constructor does expand variables in path, which is
         # unsafe since subrepo path might come from untrusted source.
         if os.path.realpath(util.expandpath(root)) != root:
             raise error.Abort(_('subrepo path contains illegal component: %s')
                               % path)
         self._repo = hg.repository(r.baseui, root, create=create)
         if self._repo.root != root:
             raise error.ProgrammingError('failed to reject unsafe subrepo '
                                          'path: %s (expanded to %s)'
                                          % (root, self._repo.root))
 
         # Propagate the parent's --hidden option
         if r is r.unfiltered():
             self._repo = self._repo.unfiltered()
 
         self.ui = self._repo.ui
         for s, k in [('ui', 'commitsubrepos')]:
             v = r.ui.config(s, k)
             if v:
                 self.ui.setconfig(s, k, v, 'subrepo')
         # internal config: ui._usedassubrepo
         self.ui.setconfig('ui', '_usedassubrepo', 'True', 'subrepo')
         self._initrepo(r, state[0], create)
 
     @annotatesubrepoerror
     def addwebdirpath(self, serverpath, webconf):
         cmdutil.addwebdirpath(self._repo, subrelpath(self), webconf)
 
     def storeclean(self, path):
         with self._repo.lock():
             return self._storeclean(path)
 
     def _storeclean(self, path):
         clean = True
         itercache = self._calcstorehash(path)
         for filehash in self._readstorehashcache(path):
             if filehash != next(itercache, None):
                 clean = False
                 break
         if clean:
             # if not empty:
             # the cached and current pull states have a different size
             clean = next(itercache, None) is None
         return clean
 
     def _calcstorehash(self, remotepath):
         '''calculate a unique "store hash"
 
         This method is used to to detect when there are changes that may
         require a push to a given remote path.'''
         # sort the files that will be hashed in increasing (likely) file size
         filelist = ('bookmarks', 'store/phaseroots', 'store/00changelog.i')
         yield '# %s\n' % _expandedabspath(remotepath)
         vfs = self._repo.vfs
         for relname in filelist:
             filehash = node.hex(hashlib.sha1(vfs.tryread(relname)).digest())
             yield '%s = %s\n' % (relname, filehash)
 
     @propertycache
     def _cachestorehashvfs(self):
         return vfsmod.vfs(self._repo.vfs.join('cache/storehash'))
 
     def _readstorehashcache(self, remotepath):
         '''read the store hash cache for a given remote repository'''
         cachefile = _getstorehashcachename(remotepath)
         return self._cachestorehashvfs.tryreadlines(cachefile, 'r')
 
     def _cachestorehash(self, remotepath):
         '''cache the current store hash
 
         Each remote repo requires its own store hash cache, because a subrepo
         store may be "clean" versus a given remote repo, but not versus another
         '''
         cachefile = _getstorehashcachename(remotepath)
         with self._repo.lock():
             storehash = list(self._calcstorehash(remotepath))
             vfs = self._cachestorehashvfs
             vfs.writelines(cachefile, storehash, mode='wb', notindexed=True)
 
     def _getctx(self):
         '''fetch the context for this subrepo revision, possibly a workingctx
         '''
         if self._ctx.rev() is None:
             return self._repo[None]  # workingctx if parent is workingctx
         else:
             rev = self._state[1]
             return self._repo[rev]
 
     @annotatesubrepoerror
     def _initrepo(self, parentrepo, source, create):
         self._repo._subparent = parentrepo
         self._repo._subsource = source
 
         if create:
             lines = ['[paths]\n']
 
             def addpathconfig(key, value):
                 if value:
                     lines.append('%s = %s\n' % (key, value))
                     self.ui.setconfig('paths', key, value, 'subrepo')
 
             defpath = _abssource(self._repo, abort=False)
             defpushpath = _abssource(self._repo, True, abort=False)
             addpathconfig('default', defpath)
             if defpath != defpushpath:
                 addpathconfig('default-push', defpushpath)
 
             self._repo.vfs.write('hgrc', util.tonativeeol(''.join(lines)))
 
     @annotatesubrepoerror
     def add(self, ui, match, prefix, explicitonly, **opts):
         return cmdutil.add(ui, self._repo, match, prefix, explicitonly, **opts)
 
     @annotatesubrepoerror
     def addremove(self, m, prefix, opts):
         # In the same way as sub directories are processed, once in a subrepo,
         # always entry any of its subrepos.  Don't corrupt the options that will
         # be used to process sibling subrepos however.
         opts = copy.copy(opts)
         opts['subrepos'] = True
-        return scmutil.addremove(self._repo, m,
-                                 self.wvfs.reljoin(prefix, self._path), opts)
+        return scmutil.addremove(self._repo, m, prefix, opts)
 
     @annotatesubrepoerror
     def cat(self, match, fm, fntemplate, prefix, **opts):
         rev = self._state[1]
         ctx = self._repo[rev]
         return cmdutil.cat(self.ui, self._repo, ctx, match, fm, fntemplate,
                            prefix, **opts)
 
     @annotatesubrepoerror
     def status(self, rev2, **opts):
         try:
             rev1 = self._state[1]
             ctx1 = self._repo[rev1]
             ctx2 = self._repo[rev2]
             return self._repo.status(ctx1, ctx2, **opts)
         except error.RepoLookupError as inst:
             self.ui.warn(_('warning: error "%s" in subrepository "%s"\n')
                          % (inst, subrelpath(self)))
             return scmutil.status([], [], [], [], [], [], [])
 
     @annotatesubrepoerror
     def diff(self, ui, diffopts, node2, match, prefix, **opts):
         try:
             node1 = node.bin(self._state[1])
             # We currently expect node2 to come from substate and be
             # in hex format
             if node2 is not None:
                 node2 = node.bin(node2)
             logcmdutil.diffordiffstat(ui, self._repo, diffopts,
                                       node1, node2, match,
                                       prefix=posixpath.join(prefix, self._path),
                                       listsubrepos=True, **opts)
         except error.RepoLookupError as inst:
             self.ui.warn(_('warning: error "%s" in subrepository "%s"\n')
                           % (inst, subrelpath(self)))
 
     @annotatesubrepoerror
     def archive(self, archiver, prefix, match=None, decode=True):
         self._get(self._state + ('hg',))
         files = self.files()
         if match:
             files = [f for f in files if match(f)]
         rev = self._state[1]
         ctx = self._repo[rev]
         scmutil.prefetchfiles(self._repo, [ctx.rev()],
                               scmutil.matchfiles(self._repo, files))
         total = abstractsubrepo.archive(self, archiver, prefix, match)
         for subpath in ctx.substate:
             s = subrepo(ctx, subpath, True)
             submatch = matchmod.subdirmatcher(subpath, match)
             total += s.archive(archiver, prefix + self._path + '/', submatch,
                                decode)
         return total
 
     @annotatesubrepoerror
     def dirty(self, ignoreupdate=False, missing=False):
         r = self._state[1]
         if r == '' and not ignoreupdate: # no state recorded
             return True
         w = self._repo[None]
         if r != w.p1().hex() and not ignoreupdate:
             # different version checked out
             return True
         return w.dirty(missing=missing) # working directory changed
 
     def basestate(self):
         return self._repo['.'].hex()
 
     def checknested(self, path):
         return self._repo._checknested(self._repo.wjoin(path))
 
     @annotatesubrepoerror
     def commit(self, text, user, date):
         # don't bother committing in the subrepo if it's only been
         # updated
         if not self.dirty(True):
             return self._repo['.'].hex()
         self.ui.debug("committing subrepo %s\n" % subrelpath(self))
         n = self._repo.commit(text, user, date)
         if not n:
             return self._repo['.'].hex() # different version checked out
         return node.hex(n)
 
     @annotatesubrepoerror
     def phase(self, state):
         return self._repo[state or '.'].phase()
 
     @annotatesubrepoerror
     def remove(self):
         # we can't fully delete the repository as it may contain
         # local-only history
         self.ui.note(_('removing subrepo %s\n') % subrelpath(self))
         hg.clean(self._repo, node.nullid, False)
 
     def _get(self, state):
         source, revision, kind = state
         parentrepo = self._repo._subparent
 
         if revision in self._repo.unfiltered():
             # Allow shared subrepos tracked at null to setup the sharedpath
             if len(self._repo) != 0 or not parentrepo.shared():
                 return True
         self._repo._subsource = source
         srcurl = _abssource(self._repo)
 
         # Defer creating the peer until after the status message is logged, in
         # case there are network problems.
         getpeer = lambda: hg.peer(self._repo, {}, srcurl)
 
         if len(self._repo) == 0:
             # use self._repo.vfs instead of self.wvfs to remove .hg only
             self._repo.vfs.rmtree()
 
             # A remote subrepo could be shared if there is a local copy
             # relative to the parent's share source.  But clone pooling doesn't
             # assemble the repos in a tree, so that can't be consistently done.
             # A simpler option is for the user to configure clone pooling, and
             # work with that.
             if parentrepo.shared() and hg.islocal(srcurl):
                 self.ui.status(_('sharing subrepo %s from %s\n')
                                % (subrelpath(self), srcurl))
                 shared = hg.share(self._repo._subparent.baseui,
                                   getpeer(), self._repo.root,
                                   update=False, bookmarks=False)
                 self._repo = shared.local()
             else:
                 # TODO: find a common place for this and this code in the
                 # share.py wrap of the clone command.
                 if parentrepo.shared():
                     pool = self.ui.config('share', 'pool')
                     if pool:
                         pool = util.expandpath(pool)
 
                     shareopts = {
                         'pool': pool,
                         'mode': self.ui.config('share', 'poolnaming'),
                     }
                 else:
                     shareopts = {}
 
                 self.ui.status(_('cloning subrepo %s from %s\n')
                                % (subrelpath(self), util.hidepassword(srcurl)))
                 other, cloned = hg.clone(self._repo._subparent.baseui, {},
                                          getpeer(), self._repo.root,
                                          update=False, shareopts=shareopts)
                 self._repo = cloned.local()
             self._initrepo(parentrepo, source, create=True)
             self._cachestorehash(srcurl)
         else:
             self.ui.status(_('pulling subrepo %s from %s\n')
                            % (subrelpath(self), util.hidepassword(srcurl)))
             cleansub = self.storeclean(srcurl)
             exchange.pull(self._repo, getpeer())
             if cleansub:
                 # keep the repo clean after pull
                 self._cachestorehash(srcurl)
         return False
 
     @annotatesubrepoerror
     def get(self, state, overwrite=False):
         inrepo = self._get(state)
         source, revision, kind = state
         repo = self._repo
         repo.ui.debug("getting subrepo %s\n" % self._path)
         if inrepo:
             urepo = repo.unfiltered()
             ctx = urepo[revision]
             if ctx.hidden():
                 urepo.ui.warn(
                     _('revision %s in subrepository "%s" is hidden\n') \
                     % (revision[0:12], self._path))
                 repo = urepo
         hg.updaterepo(repo, revision, overwrite)
 
     @annotatesubrepoerror
     def merge(self, state):
         self._get(state)
         cur = self._repo['.']
         dst = self._repo[state[1]]
         anc = dst.ancestor(cur)
 
         def mergefunc():
             if anc == cur and dst.branch() == cur.branch():
                 self.ui.debug('updating subrepository "%s"\n'
                               % subrelpath(self))
                 hg.update(self._repo, state[1])
             elif anc == dst:
                 self.ui.debug('skipping subrepository "%s"\n'
                               % subrelpath(self))
             else:
                 self.ui.debug('merging subrepository "%s"\n' % subrelpath(self))
                 hg.merge(self._repo, state[1], remind=False)
 
         wctx = self._repo[None]
         if self.dirty():
             if anc != dst:
                 if _updateprompt(self.ui, self, wctx.dirty(), cur, dst):
                     mergefunc()
             else:
                 mergefunc()
         else:
             mergefunc()
 
     @annotatesubrepoerror
     def push(self, opts):
         force = opts.get('force')
         newbranch = opts.get('new_branch')
         ssh = opts.get('ssh')
 
         # push subrepos depth-first for coherent ordering
         c = self._repo['.']
         subs = c.substate # only repos that are committed
         for s in sorted(subs):
             if c.sub(s).push(opts) == 0:
                 return False
 
         dsturl = _abssource(self._repo, True)
         if not force:
             if self.storeclean(dsturl):
                 self.ui.status(
                     _('no changes made to subrepo %s since last push to %s\n')
                     % (subrelpath(self), util.hidepassword(dsturl)))
                 return None
         self.ui.status(_('pushing subrepo %s to %s\n') %
             (subrelpath(self), util.hidepassword(dsturl)))
         other = hg.peer(self._repo, {'ssh': ssh}, dsturl)
         res = exchange.push(self._repo, other, force, newbranch=newbranch)
 
         # the repo is now clean
         self._cachestorehash(dsturl)
         return res.cgresult
 
     @annotatesubrepoerror
     def outgoing(self, ui, dest, opts):
         if 'rev' in opts or 'branch' in opts:
             opts = copy.copy(opts)
             opts.pop('rev', None)
             opts.pop('branch', None)
         return hg.outgoing(ui, self._repo, _abssource(self._repo, True), opts)
 
     @annotatesubrepoerror
     def incoming(self, ui, source, opts):
         if 'rev' in opts or 'branch' in opts:
             opts = copy.copy(opts)
             opts.pop('rev', None)
             opts.pop('branch', None)
         return hg.incoming(ui, self._repo, _abssource(self._repo, False), opts)
 
     @annotatesubrepoerror
     def files(self):
         rev = self._state[1]
         ctx = self._repo[rev]
         return ctx.manifest().keys()
 
     def filedata(self, name, decode):
         rev = self._state[1]
         data = self._repo[rev][name].data()
         if decode:
             data = self._repo.wwritedata(name, data)
         return data
 
     def fileflags(self, name):
         rev = self._state[1]
         ctx = self._repo[rev]
         return ctx.flags(name)
 
     @annotatesubrepoerror
     def printfiles(self, ui, m, fm, fmt, subrepos):
         # If the parent context is a workingctx, use the workingctx here for
         # consistency.
         if self._ctx.rev() is None:
             ctx = self._repo[None]
         else:
             rev = self._state[1]
             ctx = self._repo[rev]
         return cmdutil.files(ui, ctx, m, fm, fmt, subrepos)
 
     @annotatesubrepoerror
     def matchfileset(self, expr, badfn=None):
         repo = self._repo
         if self._ctx.rev() is None:
             ctx = repo[None]
         else:
             rev = self._state[1]
             ctx = repo[rev]
 
         matchers = [ctx.matchfileset(expr, badfn=badfn)]
 
         for subpath in ctx.substate:
             sub = ctx.sub(subpath)
 
             try:
                 sm = sub.matchfileset(expr, badfn=badfn)
                 pm = matchmod.prefixdirmatcher(repo.root, repo.getcwd(),
                                                subpath, sm, badfn=badfn)
                 matchers.append(pm)
             except error.LookupError:
                 self.ui.status(_("skipping missing subrepository: %s\n")
                                % self.wvfs.reljoin(reporelpath(self), subpath))
         if len(matchers) == 1:
             return matchers[0]
         return matchmod.unionmatcher(matchers)
 
     def walk(self, match):
         ctx = self._repo[None]
         return ctx.walk(match)
 
     @annotatesubrepoerror
     def forget(self, match, prefix, dryrun, interactive):
         return cmdutil.forget(self.ui, self._repo, match, prefix,
                               True, dryrun=dryrun, interactive=interactive)
 
     @annotatesubrepoerror
     def removefiles(self, matcher, prefix, after, force, subrepos,
                     dryrun, warnings):
         return cmdutil.remove(self.ui, self._repo, matcher, prefix,
                               after, force, subrepos, dryrun)
 
     @annotatesubrepoerror
     def revert(self, substate, *pats, **opts):
         # reverting a subrepo is a 2 step process:
         # 1. if the no_backup is not set, revert all modified
         #    files inside the subrepo
         # 2. update the subrepo to the revision specified in
         #    the corresponding substate dictionary
         self.ui.status(_('reverting subrepo %s\n') % substate[0])
         if not opts.get(r'no_backup'):
             # Revert all files on the subrepo, creating backups
             # Note that this will not recursively revert subrepos
             # We could do it if there was a set:subrepos() predicate
             opts = opts.copy()
             opts[r'date'] = None
             opts[r'rev'] = substate[1]
 
             self.filerevert(*pats, **opts)
 
         # Update the repo to the revision specified in the given substate
         if not opts.get(r'dry_run'):
             self.get(substate, overwrite=True)
 
     def filerevert(self, *pats, **opts):
         ctx = self._repo[opts[r'rev']]
         parents = self._repo.dirstate.parents()
         if opts.get(r'all'):
             pats = ['set:modified()']
         else:
             pats = []
         cmdutil.revert(self.ui, self._repo, ctx, parents, *pats, **opts)
 
     def shortid(self, revid):
         return revid[:12]
 
     @annotatesubrepoerror
     def unshare(self):
         # subrepo inherently violates our import layering rules
         # because it wants to make repo objects from deep inside the stack
         # so we manually delay the circular imports to not break
         # scripts that don't use our demand-loading
         global hg
         from . import hg as h
         hg = h
 
         # Nothing prevents a user from sharing in a repo, and then making that a
         # subrepo.  Alternately, the previous unshare attempt may have failed
         # part way through.  So recurse whether or not this layer is shared.
         if self._repo.shared():
             self.ui.status(_("unsharing subrepo '%s'\n") % self._relpath)
 
         hg.unshare(self.ui, self._repo)
 
     def verify(self):
         try:
             rev = self._state[1]
             ctx = self._repo.unfiltered()[rev]
             if ctx.hidden():
                 # Since hidden revisions aren't pushed/pulled, it seems worth an
                 # explicit warning.
                 ui = self._repo.ui
                 ui.warn(_("subrepo '%s' is hidden in revision %s\n") %
                         (self._relpath, node.short(self._ctx.node())))
             return 0
         except error.RepoLookupError:
             # A missing subrepo revision may be a case of needing to pull it, so
             # don't treat this as an error.
             self._repo.ui.warn(_("subrepo '%s' not found in revision %s\n") %
                                (self._relpath, node.short(self._ctx.node())))
             return 0
 
     @propertycache
     def wvfs(self):
         """return own wvfs for efficiency and consistency
         """
         return self._repo.wvfs
 
     @propertycache
     def _relpath(self):
         """return path to this subrepository as seen from outermost repository
         """
         # Keep consistent dir separators by avoiding vfs.join(self._path)
         return reporelpath(self._repo)
 
 class svnsubrepo(abstractsubrepo):
     def __init__(self, ctx, path, state, allowcreate):
         super(svnsubrepo, self).__init__(ctx, path)
         self._state = state
         self._exe = procutil.findexe('svn')
         if not self._exe:
             raise error.Abort(_("'svn' executable not found for subrepo '%s'")
                              % self._path)
 
     def _svncommand(self, commands, filename='', failok=False):
         cmd = [self._exe]
         extrakw = {}
         if not self.ui.interactive():
             # Making stdin be a pipe should prevent svn from behaving
             # interactively even if we can't pass --non-interactive.
             extrakw[r'stdin'] = subprocess.PIPE
             # Starting in svn 1.5 --non-interactive is a global flag
             # instead of being per-command, but we need to support 1.4 so
             # we have to be intelligent about what commands take
             # --non-interactive.
             if commands[0] in ('update', 'checkout', 'commit'):
                 cmd.append('--non-interactive')
         cmd.extend(commands)
         if filename is not None:
             path = self.wvfs.reljoin(self._ctx.repo().origroot,
                                      self._path, filename)
             cmd.append(path)
         env = dict(encoding.environ)
         # Avoid localized output, preserve current locale for everything else.
         lc_all = env.get('LC_ALL')
         if lc_all:
             env['LANG'] = lc_all
             del env['LC_ALL']
         env['LC_MESSAGES'] = 'C'
         p = subprocess.Popen(pycompat.rapply(procutil.tonativestr, cmd),
                              bufsize=-1, close_fds=procutil.closefds,
                              stdout=subprocess.PIPE, stderr=subprocess.PIPE,
                              env=procutil.tonativeenv(env), **extrakw)
         stdout, stderr = map(util.fromnativeeol, p.communicate())
         stderr = stderr.strip()
         if not failok:
             if p.returncode:
                 raise error.Abort(stderr or 'exited with code %d'
                                   % p.returncode)
             if stderr:
                 self.ui.warn(stderr + '\n')
         return stdout, stderr
 
     @propertycache
     def _svnversion(self):
         output, err = self._svncommand(['--version', '--quiet'], filename=None)
         m = re.search(br'^(\d+)\.(\d+)', output)
         if not m:
             raise error.Abort(_('cannot retrieve svn tool version'))
         return (int(m.group(1)), int(m.group(2)))
 
     def _svnmissing(self):
         return not self.wvfs.exists('.svn')
 
     def _wcrevs(self):
         # Get the working directory revision as well as the last
         # commit revision so we can compare the subrepo state with
         # both. We used to store the working directory one.
         output, err = self._svncommand(['info', '--xml'])
         doc = xml.dom.minidom.parseString(output)
         entries = doc.getElementsByTagName(r'entry')
         lastrev, rev = '0', '0'
         if entries:
             rev = pycompat.bytestr(entries[0].getAttribute(r'revision')) or '0'
             commits = entries[0].getElementsByTagName(r'commit')
             if commits:
                 lastrev = pycompat.bytestr(
                     commits[0].getAttribute(r'revision')) or '0'
         return (lastrev, rev)
 
     def _wcrev(self):
         return self._wcrevs()[0]
 
     def _wcchanged(self):
         """Return (changes, extchanges, missing) where changes is True
         if the working directory was changed, extchanges is
         True if any of these changes concern an external entry and missing
         is True if any change is a missing entry.
         """
         output, err = self._svncommand(['status', '--xml'])
         externals, changes, missing = [], [], []
         doc = xml.dom.minidom.parseString(output)
         for e in doc.getElementsByTagName(r'entry'):
             s = e.getElementsByTagName(r'wc-status')
             if not s:
                 continue
             item = s[0].getAttribute(r'item')
             props = s[0].getAttribute(r'props')
             path = e.getAttribute(r'path').encode('utf8')
             if item == r'external':
                 externals.append(path)
             elif item == r'missing':
                 missing.append(path)
             if (item not in (r'', r'normal', r'unversioned', r'external')
                 or props not in (r'', r'none', r'normal')):
                 changes.append(path)
         for path in changes:
             for ext in externals:
                 if path == ext or path.startswith(ext + pycompat.ossep):
                     return True, True, bool(missing)
         return bool(changes), False, bool(missing)
 
     @annotatesubrepoerror
     def dirty(self, ignoreupdate=False, missing=False):
         if self._svnmissing():
             return self._state[1] != ''
         wcchanged = self._wcchanged()
         changed = wcchanged[0] or (missing and wcchanged[2])
         if not changed:
             if self._state[1] in self._wcrevs() or ignoreupdate:
                 return False
         return True
 
     def basestate(self):
         lastrev, rev = self._wcrevs()
         if lastrev != rev:
             # Last committed rev is not the same than rev. We would
             # like to take lastrev but we do not know if the subrepo
             # URL exists at lastrev.  Test it and fallback to rev it
             # is not there.
             try:
                 self._svncommand(['list', '%s@%s' % (self._state[0], lastrev)])
                 return lastrev
             except error.Abort:
                 pass
         return rev
 
     @annotatesubrepoerror
     def commit(self, text, user, date):
         # user and date are out of our hands since svn is centralized
         changed, extchanged, missing = self._wcchanged()
         if not changed:
             return self.basestate()
         if extchanged:
             # Do not try to commit externals
             raise error.Abort(_('cannot commit svn externals'))
         if missing:
             # svn can commit with missing entries but aborting like hg
             # seems a better approach.
             raise error.Abort(_('cannot commit missing svn entries'))
         commitinfo, err = self._svncommand(['commit', '-m', text])
         self.ui.status(commitinfo)
         newrev = re.search('Committed revision ([0-9]+).', commitinfo)
         if not newrev:
             if not commitinfo.strip():
                 # Sometimes, our definition of "changed" differs from
                 # svn one. For instance, svn ignores missing files
                 # when committing. If there are only missing files, no
                 # commit is made, no output and no error code.
                 raise error.Abort(_('failed to commit svn changes'))
             raise error.Abort(commitinfo.splitlines()[-1])
         newrev = newrev.groups()[0]
         self.ui.status(self._svncommand(['update', '-r', newrev])[0])
         return newrev
 
     @annotatesubrepoerror
     def remove(self):
         if self.dirty():
             self.ui.warn(_('not removing repo %s because '
                            'it has changes.\n') % self._path)
             return
         self.ui.note(_('removing subrepo %s\n') % self._path)
 
         self.wvfs.rmtree(forcibly=True)
         try:
             pwvfs = self._ctx.repo().wvfs
             pwvfs.removedirs(pwvfs.dirname(self._path))
         except OSError:
             pass
 
     @annotatesubrepoerror
     def get(self, state, overwrite=False):
         if overwrite:
             self._svncommand(['revert', '--recursive'])
         args = ['checkout']
         if self._svnversion >= (1, 5):
             args.append('--force')
         # The revision must be specified at the end of the URL to properly
         # update to a directory which has since been deleted and recreated.
         args.append('%s@%s' % (state[0], state[1]))
 
         # SEC: check that the ssh url is safe
         util.checksafessh(state[0])
 
         status, err = self._svncommand(args, failok=True)
         _sanitize(self.ui, self.wvfs, '.svn')
         if not re.search('Checked out revision [0-9]+.', status):
             if ('is already a working copy for a different URL' in err
                 and (self._wcchanged()[:2] == (False, False))):
                 # obstructed but clean working copy, so just blow it away.
                 self.remove()
                 self.get(state, overwrite=False)
                 return
             raise error.Abort((status or err).splitlines()[-1])
         self.ui.status(status)
 
     @annotatesubrepoerror
     def merge(self, state):
         old = self._state[1]
         new = state[1]
         wcrev = self._wcrev()
         if new != wcrev:
             dirty = old == wcrev or self._wcchanged()[0]
             if _updateprompt(self.ui, self, dirty, wcrev, new):
                 self.get(state, False)
 
     def push(self, opts):
         # push is a no-op for SVN
         return True
 
     @annotatesubrepoerror
     def files(self):
         output = self._svncommand(['list', '--recursive', '--xml'])[0]
         doc = xml.dom.minidom.parseString(output)
         paths = []
         for e in doc.getElementsByTagName(r'entry'):
             kind = pycompat.bytestr(e.getAttribute(r'kind'))
             if kind != 'file':
                 continue
             name = r''.join(c.data for c
                             in e.getElementsByTagName(r'name')[0].childNodes
                             if c.nodeType == c.TEXT_NODE)
             paths.append(name.encode('utf8'))
         return paths
 
     def filedata(self, name, decode):
         return self._svncommand(['cat'], name)[0]
 
 
 class gitsubrepo(abstractsubrepo):
     def __init__(self, ctx, path, state, allowcreate):
         super(gitsubrepo, self).__init__(ctx, path)
         self._state = state
         self._abspath = ctx.repo().wjoin(path)
         self._subparent = ctx.repo()
         self._ensuregit()
 
     def _ensuregit(self):
         try:
             self._gitexecutable = 'git'
             out, err = self._gitnodir(['--version'])
         except OSError as e:
             genericerror = _("error executing git for subrepo '%s': %s")
             notfoundhint = _("check git is installed and in your PATH")
             if e.errno != errno.ENOENT:
                 raise error.Abort(genericerror % (
                     self._path, encoding.strtolocal(e.strerror)))
             elif pycompat.iswindows:
                 try:
                     self._gitexecutable = 'git.cmd'
                     out, err = self._gitnodir(['--version'])
                 except OSError as e2:
                     if e2.errno == errno.ENOENT:
                         raise error.Abort(_("couldn't find 'git' or 'git.cmd'"
                             " for subrepo '%s'") % self._path,
                             hint=notfoundhint)
                     else:
                         raise error.Abort(genericerror % (self._path,
                             encoding.strtolocal(e2.strerror)))
             else:
                 raise error.Abort(_("couldn't find git for subrepo '%s'")
                     % self._path, hint=notfoundhint)
         versionstatus = self._checkversion(out)
         if versionstatus == 'unknown':
             self.ui.warn(_('cannot retrieve git version\n'))
         elif versionstatus == 'abort':
             raise error.Abort(_('git subrepo requires at least 1.6.0 or later'))
         elif versionstatus == 'warning':
             self.ui.warn(_('git subrepo requires at least 1.6.0 or later\n'))
 
     @staticmethod
     def _gitversion(out):
         m = re.search(br'^git version (\d+)\.(\d+)\.(\d+)', out)
         if m:
             return (int(m.group(1)), int(m.group(2)), int(m.group(3)))
 
         m = re.search(br'^git version (\d+)\.(\d+)', out)
         if m:
             return (int(m.group(1)), int(m.group(2)), 0)
 
         return -1
 
     @staticmethod
     def _checkversion(out):
         '''ensure git version is new enough
 
         >>> _checkversion = gitsubrepo._checkversion
         >>> _checkversion(b'git version 1.6.0')
         'ok'
         >>> _checkversion(b'git version 1.8.5')
         'ok'
         >>> _checkversion(b'git version 1.4.0')
         'abort'
         >>> _checkversion(b'git version 1.5.0')
         'warning'
         >>> _checkversion(b'git version 1.9-rc0')
         'ok'
         >>> _checkversion(b'git version 1.9.0.265.g81cdec2')
         'ok'
         >>> _checkversion(b'git version 1.9.0.GIT')
         'ok'
         >>> _checkversion(b'git version 12345')
         'unknown'
         >>> _checkversion(b'no')
         'unknown'
         '''
         version = gitsubrepo._gitversion(out)
         # git 1.4.0 can't work at all, but 1.5.X can in at least some cases,
         # despite the docstring comment.  For now, error on 1.4.0, warn on
         # 1.5.0 but attempt to continue.
         if version == -1:
             return 'unknown'
         if version < (1, 5, 0):
             return 'abort'
         elif version < (1, 6, 0):
             return 'warning'
         return 'ok'
 
     def _gitcommand(self, commands, env=None, stream=False):
         return self._gitdir(commands, env=env, stream=stream)[0]
 
     def _gitdir(self, commands, env=None, stream=False):
         return self._gitnodir(commands, env=env, stream=stream,
                               cwd=self._abspath)
 
     def _gitnodir(self, commands, env=None, stream=False, cwd=None):
         """Calls the git command
 
         The methods tries to call the git command. versions prior to 1.6.0
         are not supported and very probably fail.
         """
         self.ui.debug('%s: git %s\n' % (self._relpath, ' '.join(commands)))
         if env is None:
             env = encoding.environ.copy()
         # disable localization for Git output (issue5176)
         env['LC_ALL'] = 'C'
         # fix for Git CVE-2015-7545
         if 'GIT_ALLOW_PROTOCOL' not in env:
             env['GIT_ALLOW_PROTOCOL'] = 'file:git:http:https:ssh'
         # unless ui.quiet is set, print git's stderr,
         # which is mostly progress and useful info
         errpipe = None
         if self.ui.quiet:
             errpipe = open(os.devnull, 'w')
         if self.ui._colormode and len(commands) and commands[0] == "diff":
             # insert the argument in the front,
             # the end of git diff arguments is used for paths
             commands.insert(1, '--color')
         p = subprocess.Popen(pycompat.rapply(procutil.tonativestr,
                                              [self._gitexecutable] + commands),
                              bufsize=-1,
                              cwd=pycompat.rapply(procutil.tonativestr, cwd),
                              env=procutil.tonativeenv(env),
                              close_fds=procutil.closefds,
                              stdout=subprocess.PIPE, stderr=errpipe)
         if stream:
             return p.stdout, None
 
         retdata = p.stdout.read().strip()
         # wait for the child to exit to avoid race condition.
         p.wait()
 
         if p.returncode != 0 and p.returncode != 1:
             # there are certain error codes that are ok
             command = commands[0]
             if command in ('cat-file', 'symbolic-ref'):
                 return retdata, p.returncode
             # for all others, abort
             raise error.Abort(_('git %s error %d in %s') %
                              (command, p.returncode, self._relpath))
 
         return retdata, p.returncode
 
     def _gitmissing(self):
         return not self.wvfs.exists('.git')
 
     def _gitstate(self):
         return self._gitcommand(['rev-parse', 'HEAD'])
 
     def _gitcurrentbranch(self):
         current, err = self._gitdir(['symbolic-ref', 'HEAD', '--quiet'])
         if err:
             current = None
         return current
 
     def _gitremote(self, remote):
         out = self._gitcommand(['remote', 'show', '-n', remote])
         line = out.split('\n')[1]
         i = line.index('URL: ') + len('URL: ')
         return line[i:]
 
     def _githavelocally(self, revision):
         out, code = self._gitdir(['cat-file', '-e', revision])
         return code == 0
 
     def _gitisancestor(self, r1, r2):
         base = self._gitcommand(['merge-base', r1, r2])
         return base == r1
 
     def _gitisbare(self):
         return self._gitcommand(['config', '--bool', 'core.bare']) == 'true'
 
     def _gitupdatestat(self):
         """This must be run before git diff-index.
         diff-index only looks at changes to file stat;
         this command looks at file contents and updates the stat."""
         self._gitcommand(['update-index', '-q', '--refresh'])
 
     def _gitbranchmap(self):
         '''returns 2 things:
         a map from git branch to revision
         a map from revision to branches'''
         branch2rev = {}
         rev2branch = {}
 
         out = self._gitcommand(['for-each-ref', '--format',
                                 '%(objectname) %(refname)'])
         for line in out.split('\n'):
             revision, ref = line.split(' ')
             if (not ref.startswith('refs/heads/') and
                 not ref.startswith('refs/remotes/')):
                 continue
             if ref.startswith('refs/remotes/') and ref.endswith('/HEAD'):
                 continue # ignore remote/HEAD redirects
             branch2rev[ref] = revision
             rev2branch.setdefault(revision, []).append(ref)
         return branch2rev, rev2branch
 
     def _gittracking(self, branches):
         'return map of remote branch to local tracking branch'
         # assumes no more than one local tracking branch for each remote
         tracking = {}
         for b in branches:
             if b.startswith('refs/remotes/'):
                 continue
             bname = b.split('/', 2)[2]
             remote = self._gitcommand(['config', 'branch.%s.remote' % bname])
             if remote:
                 ref = self._gitcommand(['config', 'branch.%s.merge' % bname])
                 tracking['refs/remotes/%s/%s' %
                          (remote, ref.split('/', 2)[2])] = b
         return tracking
 
     def _abssource(self, source):
         if '://' not in source:
             # recognize the scp syntax as an absolute source
             colon = source.find(':')
             if colon != -1 and '/' not in source[:colon]:
                 return source
         self._subsource = source
         return _abssource(self)
 
     def _fetch(self, source, revision):
         if self._gitmissing():
             # SEC: check for safe ssh url
             util.checksafessh(source)
 
             source = self._abssource(source)
             self.ui.status(_('cloning subrepo %s from %s\n') %
                             (self._relpath, source))
             self._gitnodir(['clone', source, self._abspath])
         if self._githavelocally(revision):
             return
         self.ui.status(_('pulling subrepo %s from %s\n') %
                         (self._relpath, self._gitremote('origin')))
         # try only origin: the originally cloned repo
         self._gitcommand(['fetch'])
         if not self._githavelocally(revision):
             raise error.Abort(_('revision %s does not exist in subrepository '
                                 '"%s"\n') % (revision, self._relpath))
 
     @annotatesubrepoerror
     def dirty(self, ignoreupdate=False, missing=False):
         if self._gitmissing():
             return self._state[1] != ''
         if self._gitisbare():
             return True
         if not ignoreupdate and self._state[1] != self._gitstate():
             # different version checked out
             return True
         # check for staged changes or modified files; ignore untracked files
         self._gitupdatestat()
         out, code = self._gitdir(['diff-index', '--quiet', 'HEAD'])
         return code == 1
 
     def basestate(self):
         return self._gitstate()
 
     @annotatesubrepoerror
     def get(self, state, overwrite=False):
         source, revision, kind = state
         if not revision:
             self.remove()
             return
         self._fetch(source, revision)
         # if the repo was set to be bare, unbare it
         if self._gitisbare():
             self._gitcommand(['config', 'core.bare', 'false'])
             if self._gitstate() == revision:
                 self._gitcommand(['reset', '--hard', 'HEAD'])
                 return
         elif self._gitstate() == revision:
             if overwrite:
                 # first reset the index to unmark new files for commit, because
                 # reset --hard will otherwise throw away files added for commit,
                 # not just unmark them.
                 self._gitcommand(['reset', 'HEAD'])
                 self._gitcommand(['reset', '--hard', 'HEAD'])
             return
         branch2rev, rev2branch = self._gitbranchmap()
 
         def checkout(args):
             cmd = ['checkout']
             if overwrite:
                 # first reset the index to unmark new files for commit, because
                 # the -f option will otherwise throw away files added for
                 # commit, not just unmark them.
                 self._gitcommand(['reset', 'HEAD'])
                 cmd.append('-f')
             self._gitcommand(cmd + args)
             _sanitize(self.ui, self.wvfs, '.git')
 
         def rawcheckout():
             # no branch to checkout, check it out with no branch
             self.ui.warn(_('checking out detached HEAD in '
                            'subrepository "%s"\n') % self._relpath)
             self.ui.warn(_('check out a git branch if you intend '
                             'to make changes\n'))
             checkout(['-q', revision])
 
         if revision not in rev2branch:
             rawcheckout()
             return
         branches = rev2branch[revision]
         firstlocalbranch = None
         for b in branches:
             if b == 'refs/heads/master':
                 # master trumps all other branches
                 checkout(['refs/heads/master'])
                 return
             if not firstlocalbranch and not b.startswith('refs/remotes/'):
                 firstlocalbranch = b
         if firstlocalbranch:
             checkout([firstlocalbranch])
             return
 
         tracking = self._gittracking(branch2rev.keys())
         # choose a remote branch already tracked if possible
         remote = branches[0]
         if remote not in tracking:
             for b in branches:
                 if b in tracking:
                     remote = b
                     break
 
         if remote not in tracking:
             # create a new local tracking branch
             local = remote.split('/', 3)[3]
             checkout(['-b', local, remote])
         elif self._gitisancestor(branch2rev[tracking[remote]], remote):
             # When updating to a tracked remote branch,
             # if the local tracking branch is downstream of it,
             # a normal `git pull` would have performed a "fast-forward merge"
             # which is equivalent to updating the local branch to the remote.
             # Since we are only looking at branching at update, we need to
             # detect this situation and perform this action lazily.
             if tracking[remote] != self._gitcurrentbranch():
                 checkout([tracking[remote]])
             self._gitcommand(['merge', '--ff', remote])
             _sanitize(self.ui, self.wvfs, '.git')
         else:
             # a real merge would be required, just checkout the revision
             rawcheckout()
 
     @annotatesubrepoerror
     def commit(self, text, user, date):
         if self._gitmissing():
             raise error.Abort(_("subrepo %s is missing") % self._relpath)
         cmd = ['commit', '-a', '-m', text]
         env = encoding.environ.copy()
         if user:
             cmd += ['--author', user]
         if date:
             # git's date parser silently ignores when seconds < 1e9
             # convert to ISO8601
             env['GIT_AUTHOR_DATE'] = dateutil.datestr(date,
                                                   '%Y-%m-%dT%H:%M:%S %1%2')
         self._gitcommand(cmd, env=env)
         # make sure commit works otherwise HEAD might not exist under certain
         # circumstances
         return self._gitstate()
 
     @annotatesubrepoerror
     def merge(self, state):
         source, revision, kind = state
         self._fetch(source, revision)
         base = self._gitcommand(['merge-base', revision, self._state[1]])
         self._gitupdatestat()
         out, code = self._gitdir(['diff-index', '--quiet', 'HEAD'])
 
         def mergefunc():
             if base == revision:
                 self.get(state) # fast forward merge
             elif base != self._state[1]:
                 self._gitcommand(['merge', '--no-commit', revision])
             _sanitize(self.ui, self.wvfs, '.git')
 
         if self.dirty():
             if self._gitstate() != revision:
                 dirty = self._gitstate() == self._state[1] or code != 0
                 if _updateprompt(self.ui, self, dirty,
                                  self._state[1][:7], revision[:7]):
                     mergefunc()
         else:
             mergefunc()
 
     @annotatesubrepoerror
     def push(self, opts):
         force = opts.get('force')
 
         if not self._state[1]:
             return True
         if self._gitmissing():
             raise error.Abort(_("subrepo %s is missing") % self._relpath)
         # if a branch in origin contains the revision, nothing to do
         branch2rev, rev2branch = self._gitbranchmap()
         if self._state[1] in rev2branch:
             for b in rev2branch[self._state[1]]:
                 if b.startswith('refs/remotes/origin/'):
                     return True
         for b, revision in branch2rev.iteritems():
             if b.startswith('refs/remotes/origin/'):
                 if self._gitisancestor(self._state[1], revision):
                     return True
         # otherwise, try to push the currently checked out branch
         cmd = ['push']
         if force:
             cmd.append('--force')
 
         current = self._gitcurrentbranch()
         if current:
             # determine if the current branch is even useful
             if not self._gitisancestor(self._state[1], current):
                 self.ui.warn(_('unrelated git branch checked out '
                                'in subrepository "%s"\n') % self._relpath)
                 return False
             self.ui.status(_('pushing branch %s of subrepository "%s"\n') %
                            (current.split('/', 2)[2], self._relpath))
             ret = self._gitdir(cmd + ['origin', current])
             return ret[1] == 0
         else:
             self.ui.warn(_('no branch checked out in subrepository "%s"\n'
                            'cannot push revision %s\n') %
                           (self._relpath, self._state[1]))
             return False
 
     @annotatesubrepoerror
     def add(self, ui, match, prefix, explicitonly, **opts):
         if self._gitmissing():
             return []
 
         s = self.status(None, unknown=True, clean=True)
 
         tracked = set()
         # dirstates 'amn' warn, 'r' is added again
         for l in (s.modified, s.added, s.deleted, s.clean):
             tracked.update(l)
 
         # Unknown files not of interest will be rejected by the matcher
         files = s.unknown
         files.extend(match.files())
 
         rejected = []
 
         files = [f for f in sorted(set(files)) if match(f)]
         for f in files:
             exact = match.exact(f)
             command = ["add"]
             if exact:
                 command.append("-f") #should be added, even if ignored
             if ui.verbose or not exact:
                 ui.status(_('adding %s\n') % match.rel(f))
 
             if f in tracked:  # hg prints 'adding' even if already tracked
                 if exact:
                     rejected.append(f)
                 continue
             if not opts.get(r'dry_run'):
                 self._gitcommand(command + [f])
 
         for f in rejected:
             ui.warn(_("%s already tracked!\n") % match.abs(f))
 
         return rejected
 
     @annotatesubrepoerror
     def remove(self):
         if self._gitmissing():
             return
         if self.dirty():
             self.ui.warn(_('not removing repo %s because '
                            'it has changes.\n') % self._relpath)
             return
         # we can't fully delete the repository as it may contain
         # local-only history
         self.ui.note(_('removing subrepo %s\n') % self._relpath)
         self._gitcommand(['config', 'core.bare', 'true'])
         for f, kind in self.wvfs.readdir():
             if f == '.git':
                 continue
             if kind == stat.S_IFDIR:
                 self.wvfs.rmtree(f)
             else:
                 self.wvfs.unlink(f)
 
     def archive(self, archiver, prefix, match=None, decode=True):
         total = 0
         source, revision = self._state
         if not revision:
             return total
         self._fetch(source, revision)
 
         # Parse git's native archive command.
         # This should be much faster than manually traversing the trees
         # and objects with many subprocess calls.
         tarstream = self._gitcommand(['archive', revision], stream=True)
         tar = tarfile.open(fileobj=tarstream, mode=r'r|')
         relpath = subrelpath(self)
         progress = self.ui.makeprogress(_('archiving (%s)') % relpath,
                                         unit=_('files'))
         progress.update(0)
         for info in tar:
             if info.isdir():
                 continue
             bname = pycompat.fsencode(info.name)
             if match and not match(bname):
                 continue
             if info.issym():
                 data = info.linkname
             else:
                 data = tar.extractfile(info).read()
             archiver.addfile(prefix + self._path + '/' + bname,
                              info.mode, info.issym(), data)
             total += 1
             progress.increment()
         progress.complete()
         return total
 
 
     @annotatesubrepoerror
     def cat(self, match, fm, fntemplate, prefix, **opts):
         rev = self._state[1]
         if match.anypats():
             return 1 #No support for include/exclude yet
 
         if not match.files():
             return 1
 
         # TODO: add support for non-plain formatter (see cmdutil.cat())
         for f in match.files():
             output = self._gitcommand(["show", "%s:%s" % (rev, f)])
             fp = cmdutil.makefileobj(self._ctx, fntemplate,
                                      pathname=self.wvfs.reljoin(prefix, f))
             fp.write(output)
             fp.close()
         return 0
 
 
     @annotatesubrepoerror
     def status(self, rev2, **opts):
         rev1 = self._state[1]
         if self._gitmissing() or not rev1:
             # if the repo is missing, return no results
             return scmutil.status([], [], [], [], [], [], [])
         modified, added, removed = [], [], []
         self._gitupdatestat()
         if rev2:
             command = ['diff-tree', '--no-renames', '-r', rev1, rev2]
         else:
             command = ['diff-index', '--no-renames', rev1]
         out = self._gitcommand(command)
         for line in out.split('\n'):
             tab = line.find('\t')
             if tab == -1:
                 continue
             status, f = line[tab - 1:tab], line[tab + 1:]
             if status == 'M':
                 modified.append(f)
             elif status == 'A':
                 added.append(f)
             elif status == 'D':
                 removed.append(f)
 
         deleted, unknown, ignored, clean = [], [], [], []
 
         command = ['status', '--porcelain', '-z']
         if opts.get(r'unknown'):
             command += ['--untracked-files=all']
         if opts.get(r'ignored'):
             command += ['--ignored']
         out = self._gitcommand(command)
 
         changedfiles = set()
         changedfiles.update(modified)
         changedfiles.update(added)
         changedfiles.update(removed)
         for line in out.split('\0'):
             if not line:
                 continue
             st = line[0:2]
             #moves and copies show 2 files on one line
             if line.find('\0') >= 0:
                 filename1, filename2 = line[3:].split('\0')
             else:
                 filename1 = line[3:]
                 filename2 = None
 
             changedfiles.add(filename1)
             if filename2:
                 changedfiles.add(filename2)
 
             if st == '??':
                 unknown.append(filename1)
             elif st == '!!':
                 ignored.append(filename1)
 
         if opts.get(r'clean'):
             out = self._gitcommand(['ls-files'])
             for f in out.split('\n'):
                 if not f in changedfiles:
                     clean.append(f)
 
         return scmutil.status(modified, added, removed, deleted,
                               unknown, ignored, clean)
 
     @annotatesubrepoerror
     def diff(self, ui, diffopts, node2, match, prefix, **opts):
         node1 = self._state[1]
         cmd = ['diff', '--no-renames']
         if opts[r'stat']:
             cmd.append('--stat')
         else:
             # for Git, this also implies '-p'
             cmd.append('-U%d' % diffopts.context)
 
         gitprefix = self.wvfs.reljoin(prefix, self._path)
 
         if diffopts.noprefix:
             cmd.extend(['--src-prefix=%s/' % gitprefix,
                         '--dst-prefix=%s/' % gitprefix])
         else:
             cmd.extend(['--src-prefix=a/%s/' % gitprefix,
                         '--dst-prefix=b/%s/' % gitprefix])
 
         if diffopts.ignorews:
             cmd.append('--ignore-all-space')
         if diffopts.ignorewsamount:
             cmd.append('--ignore-space-change')
         if self._gitversion(self._gitcommand(['--version'])) >= (1, 8, 4) \
                 and diffopts.ignoreblanklines:
             cmd.append('--ignore-blank-lines')
 
         cmd.append(node1)
         if node2:
             cmd.append(node2)
 
         output = ""
         if match.always():
             output += self._gitcommand(cmd) + '\n'
         else:
             st = self.status(node2)[:3]
             files = [f for sublist in st for f in sublist]
             for f in files:
                 if match(f):
                     output += self._gitcommand(cmd + ['--', f]) + '\n'
 
         if output.strip():
             ui.write(output)
 
     @annotatesubrepoerror
     def revert(self, substate, *pats, **opts):
         self.ui.status(_('reverting subrepo %s\n') % substate[0])
         if not opts.get(r'no_backup'):
             status = self.status(None)
             names = status.modified
             for name in names:
                 # backuppath() expects a path relative to the parent repo (the
                 # repo that ui.origbackuppath is relative to)
                 parentname = os.path.join(self._path, name)
                 bakname = scmutil.backuppath(self.ui, self._subparent,
                                              parentname)
                 self.ui.note(_('saving current version of %s as %s\n') %
                         (name, os.path.relpath(bakname)))
                 util.rename(self.wvfs.join(name), bakname)
 
         if not opts.get(r'dry_run'):
             self.get(substate, overwrite=True)
         return []
 
     def shortid(self, revid):
         return revid[:7]
 
 types = {
     'hg': hgsubrepo,
     'svn': svnsubrepo,
     'git': gitsubrepo,
     }