py3: catch specific OSError subclasses instead of checking errno

Contrary to the previous changesets in this series, this covers cases where
errno was checked for multiple values.

EACCES -> PermissionError
ENOENT -> FileNotFoundError
ENOTDIR -> NotADirectoryError
EISDIR -> IsADirectoryError

mercurial/dirstate.py

@@ -1,1470 +1,1467 b''
 # dirstate.py - working directory tracking for mercurial
 #
 # Copyright 2005-2007 Olivia 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.
 
 
 import collections
 import contextlib
-import errno
 import os
 import stat
 import uuid
 
 from .i18n import _
 from .pycompat import delattr
 
 from hgdemandimport import tracing
 
 from . import (
     dirstatemap,
     encoding,
     error,
     match as matchmod,
     node,
     pathutil,
     policy,
     pycompat,
     scmutil,
     sparse,
     util,
 )
 
 from .dirstateutils import (
     timestamp,
 )
 
 from .interfaces import (
     dirstate as intdirstate,
     util as interfaceutil,
 )
 
 parsers = policy.importmod('parsers')
 rustmod = policy.importrust('dirstate')
 
 HAS_FAST_DIRSTATE_V2 = rustmod is not None
 
 propertycache = util.propertycache
 filecache = scmutil.filecache
 _rangemask = dirstatemap.rangemask
 
 DirstateItem = dirstatemap.DirstateItem
 
 
 class repocache(filecache):
     """filecache for files in .hg/"""
 
     def join(self, obj, fname):
         return obj._opener.join(fname)
 
 
 class rootcache(filecache):
     """filecache for files in the repository root"""
 
     def join(self, obj, fname):
         return obj._join(fname)
 
 
 def requires_parents_change(func):
     def wrap(self, *args, **kwargs):
         if not self.pendingparentchange():
             msg = 'calling `%s` outside of a parentchange context'
             msg %= func.__name__
             raise error.ProgrammingError(msg)
         return func(self, *args, **kwargs)
 
     return wrap
 
 
 def requires_no_parents_change(func):
     def wrap(self, *args, **kwargs):
         if self.pendingparentchange():
             msg = 'calling `%s` inside of a parentchange context'
             msg %= func.__name__
             raise error.ProgrammingError(msg)
         return func(self, *args, **kwargs)
 
     return wrap
 
 
 @interfaceutil.implementer(intdirstate.idirstate)
 class dirstate:
     def __init__(
         self,
         opener,
         ui,
         root,
         validate,
         sparsematchfn,
         nodeconstants,
         use_dirstate_v2,
         use_tracked_hint=False,
     ):
         """Create a new dirstate object.
 
         opener is an open()-like callable that can be used to open the
         dirstate file; root is the root of the directory tracked by
         the dirstate.
         """
         self._use_dirstate_v2 = use_dirstate_v2
         self._use_tracked_hint = use_tracked_hint
         self._nodeconstants = nodeconstants
         self._opener = opener
         self._validate = validate
         self._root = root
         self._sparsematchfn = sparsematchfn
         # ntpath.join(root, '') of Python 2.7.9 does not add sep if root is
         # UNC path pointing to root share (issue4557)
         self._rootdir = pathutil.normasprefix(root)
         # True is any internal state may be different
         self._dirty = False
         # True if the set of tracked file may be different
         self._dirty_tracked_set = False
         self._ui = ui
         self._filecache = {}
         self._parentwriters = 0
         self._filename = b'dirstate'
         self._filename_th = b'dirstate-tracked-hint'
         self._pendingfilename = b'%s.pending' % self._filename
         self._plchangecallbacks = {}
         self._origpl = None
         self._mapcls = dirstatemap.dirstatemap
         # Access and cache cwd early, so we don't access it for the first time
         # after a working-copy update caused it to not exist (accessing it then
         # raises an exception).
         self._cwd
 
     def prefetch_parents(self):
         """make sure the parents are loaded
 
         Used to avoid a race condition.
         """
         self._pl
 
     @contextlib.contextmanager
     def parentchange(self):
         """Context manager for handling dirstate parents.
 
         If an exception occurs in the scope of the context manager,
         the incoherent dirstate won't be written when wlock is
         released.
         """
         self._parentwriters += 1
         yield
         # Typically we want the "undo" step of a context manager in a
         # finally block so it happens even when an exception
         # occurs. In this case, however, we only want to decrement
         # parentwriters if the code in the with statement exits
         # normally, so we don't have a try/finally here on purpose.
         self._parentwriters -= 1
 
     def pendingparentchange(self):
         """Returns true if the dirstate is in the middle of a set of changes
         that modify the dirstate parent.
         """
         return self._parentwriters > 0
 
     @propertycache
     def _map(self):
         """Return the dirstate contents (see documentation for dirstatemap)."""
         self._map = self._mapcls(
             self._ui,
             self._opener,
             self._root,
             self._nodeconstants,
             self._use_dirstate_v2,
         )
         return self._map
 
     @property
     def _sparsematcher(self):
         """The matcher for the sparse checkout.
 
         The working directory may not include every file from a manifest. The
         matcher obtained by this property will match a path if it is to be
         included in the working directory.
         """
         # TODO there is potential to cache this property. For now, the matcher
         # is resolved on every access. (But the called function does use a
         # cache to keep the lookup fast.)
         return self._sparsematchfn()
 
     @repocache(b'branch')
     def _branch(self):
         try:
             return self._opener.read(b"branch").strip() or b"default"
         except FileNotFoundError:
             return b"default"
 
     @property
     def _pl(self):
         return self._map.parents()
 
     def hasdir(self, d):
         return self._map.hastrackeddir(d)
 
     @rootcache(b'.hgignore')
     def _ignore(self):
         files = self._ignorefiles()
         if not files:
             return matchmod.never()
 
         pats = [b'include:%s' % f for f in files]
         return matchmod.match(self._root, b'', [], pats, warn=self._ui.warn)
 
     @propertycache
     def _slash(self):
         return self._ui.configbool(b'ui', b'slash') and pycompat.ossep != b'/'
 
     @propertycache
     def _checklink(self):
         return util.checklink(self._root)
 
     @propertycache
     def _checkexec(self):
         return bool(util.checkexec(self._root))
 
     @propertycache
     def _checkcase(self):
         return not util.fscasesensitive(self._join(b'.hg'))
 
     def _join(self, f):
         # much faster than os.path.join()
         # it's safe because f is always a relative path
         return self._rootdir + f
 
     def flagfunc(self, buildfallback):
         """build a callable that returns flags associated with a filename
 
         The information is extracted from three possible layers:
         1. the file system if it supports the information
         2. the "fallback" information stored in the dirstate if any
         3. a more expensive mechanism inferring the flags from the parents.
         """
 
         # small hack to cache the result of buildfallback()
         fallback_func = []
 
         def get_flags(x):
             entry = None
             fallback_value = None
             try:
                 st = os.lstat(self._join(x))
             except OSError:
                 return b''
 
             if self._checklink:
                 if util.statislink(st):
                     return b'l'
             else:
                 entry = self.get_entry(x)
                 if entry.has_fallback_symlink:
                     if entry.fallback_symlink:
                         return b'l'
                 else:
                     if not fallback_func:
                         fallback_func.append(buildfallback())
                     fallback_value = fallback_func[0](x)
                     if b'l' in fallback_value:
                         return b'l'
 
             if self._checkexec:
                 if util.statisexec(st):
                     return b'x'
             else:
                 if entry is None:
                     entry = self.get_entry(x)
                 if entry.has_fallback_exec:
                     if entry.fallback_exec:
                         return b'x'
                 else:
                     if fallback_value is None:
                         if not fallback_func:
                             fallback_func.append(buildfallback())
                         fallback_value = fallback_func[0](x)
                     if b'x' in fallback_value:
                         return b'x'
             return b''
 
         return get_flags
 
     @propertycache
     def _cwd(self):
         # internal config: ui.forcecwd
         forcecwd = self._ui.config(b'ui', b'forcecwd')
         if forcecwd:
             return forcecwd
         return encoding.getcwd()
 
     def getcwd(self):
         """Return the path from which a canonical path is calculated.
 
         This path should be used to resolve file patterns or to convert
         canonical paths back to file paths for display. It shouldn't be
         used to get real file paths. Use vfs functions instead.
         """
         cwd = self._cwd
         if cwd == self._root:
             return b''
         # self._root ends with a path separator if self._root is '/' or 'C:\'
         rootsep = self._root
         if not util.endswithsep(rootsep):
             rootsep += pycompat.ossep
         if cwd.startswith(rootsep):
             return cwd[len(rootsep) :]
         else:
             # we're outside the repo. return an absolute path.
             return cwd
 
     def pathto(self, f, cwd=None):
         if cwd is None:
             cwd = self.getcwd()
         path = util.pathto(self._root, cwd, f)
         if self._slash:
             return util.pconvert(path)
         return path
 
     def get_entry(self, path):
         """return a DirstateItem for the associated path"""
         entry = self._map.get(path)
         if entry is None:
             return DirstateItem()
         return entry
 
     def __contains__(self, key):
         return key in self._map
 
     def __iter__(self):
         return iter(sorted(self._map))
 
     def items(self):
         return self._map.items()
 
     iteritems = items
 
     def parents(self):
         return [self._validate(p) for p in self._pl]
 
     def p1(self):
         return self._validate(self._pl[0])
 
     def p2(self):
         return self._validate(self._pl[1])
 
     @property
     def in_merge(self):
         """True if a merge is in progress"""
         return self._pl[1] != self._nodeconstants.nullid
 
     def branch(self):
         return encoding.tolocal(self._branch)
 
     def setparents(self, p1, p2=None):
         """Set dirstate parents to p1 and p2.
 
         When moving from two parents to one, "merged" entries a
         adjusted to normal and previous copy records discarded and
         returned by the call.
 
         See localrepo.setparents()
         """
         if p2 is None:
             p2 = self._nodeconstants.nullid
         if self._parentwriters == 0:
             raise ValueError(
                 b"cannot set dirstate parent outside of "
                 b"dirstate.parentchange context manager"
             )
 
         self._dirty = True
         oldp2 = self._pl[1]
         if self._origpl is None:
             self._origpl = self._pl
         nullid = self._nodeconstants.nullid
         # True if we need to fold p2 related state back to a linear case
         fold_p2 = oldp2 != nullid and p2 == nullid
         return self._map.setparents(p1, p2, fold_p2=fold_p2)
 
     def setbranch(self, branch):
         self.__class__._branch.set(self, encoding.fromlocal(branch))
         f = self._opener(b'branch', b'w', atomictemp=True, checkambig=True)
         try:
             f.write(self._branch + b'\n')
             f.close()
 
             # make sure filecache has the correct stat info for _branch after
             # replacing the underlying file
             ce = self._filecache[b'_branch']
             if ce:
                 ce.refresh()
         except:  # re-raises
             f.discard()
             raise
 
     def invalidate(self):
         """Causes the next access to reread the dirstate.
 
         This is different from localrepo.invalidatedirstate() because it always
         rereads the dirstate. Use localrepo.invalidatedirstate() if you want to
         check whether the dirstate has changed before rereading it."""
 
         for a in ("_map", "_branch", "_ignore"):
             if a in self.__dict__:
                 delattr(self, a)
         self._dirty = False
         self._dirty_tracked_set = False
         self._parentwriters = 0
         self._origpl = None
 
     def copy(self, source, dest):
         """Mark dest as a copy of source. Unmark dest if source is None."""
         if source == dest:
             return
         self._dirty = True
         if source is not None:
             self._map.copymap[dest] = source
         else:
             self._map.copymap.pop(dest, None)
 
     def copied(self, file):
         return self._map.copymap.get(file, None)
 
     def copies(self):
         return self._map.copymap
 
     @requires_no_parents_change
     def set_tracked(self, filename, reset_copy=False):
         """a "public" method for generic code to mark a file as tracked
 
         This function is to be called outside of "update/merge" case. For
         example by a command like `hg add X`.
 
         if reset_copy is set, any existing copy information will be dropped.
 
         return True the file was previously untracked, False otherwise.
         """
         self._dirty = True
         entry = self._map.get(filename)
         if entry is None or not entry.tracked:
             self._check_new_tracked_filename(filename)
         pre_tracked = self._map.set_tracked(filename)
         if reset_copy:
             self._map.copymap.pop(filename, None)
         if pre_tracked:
             self._dirty_tracked_set = True
         return pre_tracked
 
     @requires_no_parents_change
     def set_untracked(self, filename):
         """a "public" method for generic code to mark a file as untracked
 
         This function is to be called outside of "update/merge" case. For
         example by a command like `hg remove X`.
 
         return True the file was previously tracked, False otherwise.
         """
         ret = self._map.set_untracked(filename)
         if ret:
             self._dirty = True
             self._dirty_tracked_set = True
         return ret
 
     @requires_no_parents_change
     def set_clean(self, filename, parentfiledata):
         """record that the current state of the file on disk is known to be clean"""
         self._dirty = True
         if not self._map[filename].tracked:
             self._check_new_tracked_filename(filename)
         (mode, size, mtime) = parentfiledata
         self._map.set_clean(filename, mode, size, mtime)
 
     @requires_no_parents_change
     def set_possibly_dirty(self, filename):
         """record that the current state of the file on disk is unknown"""
         self._dirty = True
         self._map.set_possibly_dirty(filename)
 
     @requires_parents_change
     def update_file_p1(
         self,
         filename,
         p1_tracked,
     ):
         """Set a file as tracked in the parent (or not)
 
         This is to be called when adjust the dirstate to a new parent after an history
         rewriting operation.
 
         It should not be called during a merge (p2 != nullid) and only within
         a `with dirstate.parentchange():` context.
         """
         if self.in_merge:
             msg = b'update_file_reference should not be called when merging'
             raise error.ProgrammingError(msg)
         entry = self._map.get(filename)
         if entry is None:
             wc_tracked = False
         else:
             wc_tracked = entry.tracked
         if not (p1_tracked or wc_tracked):
             # the file is no longer relevant to anyone
             if self._map.get(filename) is not None:
                 self._map.reset_state(filename)
                 self._dirty = True
         elif (not p1_tracked) and wc_tracked:
             if entry is not None and entry.added:
                 return  # avoid dropping copy information (maybe?)
 
         self._map.reset_state(
             filename,
             wc_tracked,
             p1_tracked,
             # the underlying reference might have changed, we will have to
             # check it.
             has_meaningful_mtime=False,
         )
 
     @requires_parents_change
     def update_file(
         self,
         filename,
         wc_tracked,
         p1_tracked,
         p2_info=False,
         possibly_dirty=False,
         parentfiledata=None,
     ):
         """update the information about a file in the dirstate
 
         This is to be called when the direstates parent changes to keep track
         of what is the file situation in regards to the working copy and its parent.
 
         This function must be called within a `dirstate.parentchange` context.
 
         note: the API is at an early stage and we might need to adjust it
         depending of what information ends up being relevant and useful to
         other processing.
         """
 
         # note: I do not think we need to double check name clash here since we
         # are in a update/merge case that should already have taken care of
         # this. The test agrees
 
         self._dirty = True
         old_entry = self._map.get(filename)
         if old_entry is None:
             prev_tracked = False
         else:
             prev_tracked = old_entry.tracked
         if prev_tracked != wc_tracked:
             self._dirty_tracked_set = True
 
         self._map.reset_state(
             filename,
             wc_tracked,
             p1_tracked,
             p2_info=p2_info,
             has_meaningful_mtime=not possibly_dirty,
             parentfiledata=parentfiledata,
         )
 
     def _check_new_tracked_filename(self, filename):
         scmutil.checkfilename(filename)
         if self._map.hastrackeddir(filename):
             msg = _(b'directory %r already in dirstate')
             msg %= pycompat.bytestr(filename)
             raise error.Abort(msg)
         # shadows
         for d in pathutil.finddirs(filename):
             if self._map.hastrackeddir(d):
                 break
             entry = self._map.get(d)
             if entry is not None and not entry.removed:
                 msg = _(b'file %r in dirstate clashes with %r')
                 msg %= (pycompat.bytestr(d), pycompat.bytestr(filename))
                 raise error.Abort(msg)
 
     def _discoverpath(self, path, normed, ignoremissing, exists, storemap):
         if exists is None:
             exists = os.path.lexists(os.path.join(self._root, path))
         if not exists:
             # Maybe a path component exists
             if not ignoremissing and b'/' in path:
                 d, f = path.rsplit(b'/', 1)
                 d = self._normalize(d, False, ignoremissing, None)
                 folded = d + b"/" + f
             else:
                 # No path components, preserve original case
                 folded = path
         else:
             # recursively normalize leading directory components
             # against dirstate
             if b'/' in normed:
                 d, f = normed.rsplit(b'/', 1)
                 d = self._normalize(d, False, ignoremissing, True)
                 r = self._root + b"/" + d
                 folded = d + b"/" + util.fspath(f, r)
             else:
                 folded = util.fspath(normed, self._root)
             storemap[normed] = folded
 
         return folded
 
     def _normalizefile(self, path, isknown, ignoremissing=False, exists=None):
         normed = util.normcase(path)
         folded = self._map.filefoldmap.get(normed, None)
         if folded is None:
             if isknown:
                 folded = path
             else:
                 folded = self._discoverpath(
                     path, normed, ignoremissing, exists, self._map.filefoldmap
                 )
         return folded
 
     def _normalize(self, path, isknown, ignoremissing=False, exists=None):
         normed = util.normcase(path)
         folded = self._map.filefoldmap.get(normed, None)
         if folded is None:
             folded = self._map.dirfoldmap.get(normed, None)
         if folded is None:
             if isknown:
                 folded = path
             else:
                 # store discovered result in dirfoldmap so that future
                 # normalizefile calls don't start matching directories
                 folded = self._discoverpath(
                     path, normed, ignoremissing, exists, self._map.dirfoldmap
                 )
         return folded
 
     def normalize(self, path, isknown=False, ignoremissing=False):
         """
         normalize the case of a pathname when on a casefolding filesystem
 
         isknown specifies whether the filename came from walking the
         disk, to avoid extra filesystem access.
 
         If ignoremissing is True, missing path are returned
         unchanged. Otherwise, we try harder to normalize possibly
         existing path components.
 
         The normalized case is determined based on the following precedence:
 
         - version of name already stored in the dirstate
         - version of name stored on disk
         - version provided via command arguments
         """
 
         if self._checkcase:
             return self._normalize(path, isknown, ignoremissing)
         return path
 
     def clear(self):
         self._map.clear()
         self._dirty = True
 
     def rebuild(self, parent, allfiles, changedfiles=None):
         if changedfiles is None:
             # Rebuild entire dirstate
             to_lookup = allfiles
             to_drop = []
             self.clear()
         elif len(changedfiles) < 10:
             # Avoid turning allfiles into a set, which can be expensive if it's
             # large.
             to_lookup = []
             to_drop = []
             for f in changedfiles:
                 if f in allfiles:
                     to_lookup.append(f)
                 else:
                     to_drop.append(f)
         else:
             changedfilesset = set(changedfiles)
             to_lookup = changedfilesset & set(allfiles)
             to_drop = changedfilesset - to_lookup
 
         if self._origpl is None:
             self._origpl = self._pl
         self._map.setparents(parent, self._nodeconstants.nullid)
 
         for f in to_lookup:
 
             if self.in_merge:
                 self.set_tracked(f)
             else:
                 self._map.reset_state(
                     f,
                     wc_tracked=True,
                     p1_tracked=True,
                 )
         for f in to_drop:
             self._map.reset_state(f)
 
         self._dirty = True
 
     def identity(self):
         """Return identity of dirstate itself to detect changing in storage
 
         If identity of previous dirstate is equal to this, writing
         changes based on the former dirstate out can keep consistency.
         """
         return self._map.identity
 
     def write(self, tr):
         if not self._dirty:
             return
 
         write_key = self._use_tracked_hint and self._dirty_tracked_set
         if tr:
             # delay writing in-memory changes out
             tr.addfilegenerator(
                 b'dirstate-1-main',
                 (self._filename,),
                 lambda f: self._writedirstate(tr, f),
                 location=b'plain',
                 post_finalize=True,
             )
             if write_key:
                 tr.addfilegenerator(
                     b'dirstate-2-key-post',
                     (self._filename_th,),
                     lambda f: self._write_tracked_hint(tr, f),
                     location=b'plain',
                     post_finalize=True,
                 )
             return
 
         file = lambda f: self._opener(f, b"w", atomictemp=True, checkambig=True)
         with file(self._filename) as f:
             self._writedirstate(tr, f)
         if write_key:
             # we update the key-file after writing to make sure reader have a
             # key that match the newly written content
             with file(self._filename_th) as f:
                 self._write_tracked_hint(tr, f)
 
     def delete_tracked_hint(self):
         """remove the tracked_hint file
 
         To be used by format downgrades operation"""
         self._opener.unlink(self._filename_th)
         self._use_tracked_hint = False
 
     def addparentchangecallback(self, category, callback):
         """add a callback to be called when the wd parents are changed
 
         Callback will be called with the following arguments:
             dirstate, (oldp1, oldp2), (newp1, newp2)
 
         Category is a unique identifier to allow overwriting an old callback
         with a newer callback.
         """
         self._plchangecallbacks[category] = callback
 
     def _writedirstate(self, tr, st):
         # notify callbacks about parents change
         if self._origpl is not None and self._origpl != self._pl:
             for c, callback in sorted(self._plchangecallbacks.items()):
                 callback(self, self._origpl, self._pl)
             self._origpl = None
         self._map.write(tr, st)
         self._dirty = False
         self._dirty_tracked_set = False
 
     def _write_tracked_hint(self, tr, f):
         key = node.hex(uuid.uuid4().bytes)
         f.write(b"1\n%s\n" % key)  # 1 is the format version
 
     def _dirignore(self, f):
         if self._ignore(f):
             return True
         for p in pathutil.finddirs(f):
             if self._ignore(p):
                 return True
         return False
 
     def _ignorefiles(self):
         files = []
         if os.path.exists(self._join(b'.hgignore')):
             files.append(self._join(b'.hgignore'))
         for name, path in self._ui.configitems(b"ui"):
             if name == b'ignore' or name.startswith(b'ignore.'):
                 # we need to use os.path.join here rather than self._join
                 # because path is arbitrary and user-specified
                 files.append(os.path.join(self._rootdir, util.expandpath(path)))
         return files
 
     def _ignorefileandline(self, f):
         files = collections.deque(self._ignorefiles())
         visited = set()
         while files:
             i = files.popleft()
             patterns = matchmod.readpatternfile(
                 i, self._ui.warn, sourceinfo=True
             )
             for pattern, lineno, line in patterns:
                 kind, p = matchmod._patsplit(pattern, b'glob')
                 if kind == b"subinclude":
                     if p not in visited:
                         files.append(p)
                     continue
                 m = matchmod.match(
                     self._root, b'', [], [pattern], warn=self._ui.warn
                 )
                 if m(f):
                     return (i, lineno, line)
             visited.add(i)
         return (None, -1, b"")
 
     def _walkexplicit(self, match, subrepos):
         """Get stat data about the files explicitly specified by match.
 
         Return a triple (results, dirsfound, dirsnotfound).
         - results is a mapping from filename to stat result. It also contains
           listings mapping subrepos and .hg to None.
         - dirsfound is a list of files found to be directories.
         - dirsnotfound is a list of files that the dirstate thinks are
           directories and that were not found."""
 
         def badtype(mode):
             kind = _(b'unknown')
             if stat.S_ISCHR(mode):
                 kind = _(b'character device')
             elif stat.S_ISBLK(mode):
                 kind = _(b'block device')
             elif stat.S_ISFIFO(mode):
                 kind = _(b'fifo')
             elif stat.S_ISSOCK(mode):
                 kind = _(b'socket')
             elif stat.S_ISDIR(mode):
                 kind = _(b'directory')
             return _(b'unsupported file type (type is %s)') % kind
 
         badfn = match.bad
         dmap = self._map
         lstat = os.lstat
         getkind = stat.S_IFMT
         dirkind = stat.S_IFDIR
         regkind = stat.S_IFREG
         lnkkind = stat.S_IFLNK
         join = self._join
         dirsfound = []
         foundadd = dirsfound.append
         dirsnotfound = []
         notfoundadd = dirsnotfound.append
 
         if not match.isexact() and self._checkcase:
             normalize = self._normalize
         else:
             normalize = None
 
         files = sorted(match.files())
         subrepos.sort()
         i, j = 0, 0
         while i < len(files) and j < len(subrepos):
             subpath = subrepos[j] + b"/"
             if files[i] < subpath:
                 i += 1
                 continue
             while i < len(files) and files[i].startswith(subpath):
                 del files[i]
             j += 1
 
         if not files or b'' in files:
             files = [b'']
             # constructing the foldmap is expensive, so don't do it for the
             # common case where files is ['']
             normalize = None
         results = dict.fromkeys(subrepos)
         results[b'.hg'] = None
 
         for ff in files:
             if normalize:
                 nf = normalize(ff, False, True)
             else:
                 nf = ff
             if nf in results:
                 continue
 
             try:
                 st = lstat(join(nf))
                 kind = getkind(st.st_mode)
                 if kind == dirkind:
                     if nf in dmap:
                         # file replaced by dir on disk but still in dirstate
                         results[nf] = None
                     foundadd((nf, ff))
                 elif kind == regkind or kind == lnkkind:
                     results[nf] = st
                 else:
                     badfn(ff, badtype(kind))
                     if nf in dmap:
                         results[nf] = None
             except OSError as inst:  # nf not found on disk - it is dirstate only
                 if nf in dmap:  # does it exactly match a missing file?
                     results[nf] = None
                 else:  # does it match a missing directory?
                     if self._map.hasdir(nf):
                         notfoundadd(nf)
                     else:
                         badfn(ff, encoding.strtolocal(inst.strerror))
 
         # match.files() may contain explicitly-specified paths that shouldn't
         # be taken; drop them from the list of files found. dirsfound/notfound
         # aren't filtered here because they will be tested later.
         if match.anypats():
             for f in list(results):
                 if f == b'.hg' or f in subrepos:
                     # keep sentinel to disable further out-of-repo walks
                     continue
                 if not match(f):
                     del results[f]
 
         # Case insensitive filesystems cannot rely on lstat() failing to detect
         # a case-only rename.  Prune the stat object for any file that does not
         # match the case in the filesystem, if there are multiple files that
         # normalize to the same path.
         if match.isexact() and self._checkcase:
             normed = {}
 
             for f, st in results.items():
                 if st is None:
                     continue
 
                 nc = util.normcase(f)
                 paths = normed.get(nc)
 
                 if paths is None:
                     paths = set()
                     normed[nc] = paths
 
                 paths.add(f)
 
             for norm, paths in normed.items():
                 if len(paths) > 1:
                     for path in paths:
                         folded = self._discoverpath(
                             path, norm, True, None, self._map.dirfoldmap
                         )
                         if path != folded:
                             results[path] = None
 
         return results, dirsfound, dirsnotfound
 
     def walk(self, match, subrepos, unknown, ignored, full=True):
         """
         Walk recursively through the directory tree, finding all files
         matched by match.
 
         If full is False, maybe skip some known-clean files.
 
         Return a dict mapping filename to stat-like object (either
         mercurial.osutil.stat instance or return value of os.stat()).
 
         """
         # full is a flag that extensions that hook into walk can use -- this
         # implementation doesn't use it at all. This satisfies the contract
         # because we only guarantee a "maybe".
 
         if ignored:
             ignore = util.never
             dirignore = util.never
         elif unknown:
             ignore = self._ignore
             dirignore = self._dirignore
         else:
             # if not unknown and not ignored, drop dir recursion and step 2
             ignore = util.always
             dirignore = util.always
 
         matchfn = match.matchfn
         matchalways = match.always()
         matchtdir = match.traversedir
         dmap = self._map
         listdir = util.listdir
         lstat = os.lstat
         dirkind = stat.S_IFDIR
         regkind = stat.S_IFREG
         lnkkind = stat.S_IFLNK
         join = self._join
 
         exact = skipstep3 = False
         if match.isexact():  # match.exact
             exact = True
             dirignore = util.always  # skip step 2
         elif match.prefix():  # match.match, no patterns
             skipstep3 = True
 
         if not exact and self._checkcase:
             normalize = self._normalize
             normalizefile = self._normalizefile
             skipstep3 = False
         else:
             normalize = self._normalize
             normalizefile = None
 
         # step 1: find all explicit files
         results, work, dirsnotfound = self._walkexplicit(match, subrepos)
         if matchtdir:
             for d in work:
                 matchtdir(d[0])
             for d in dirsnotfound:
                 matchtdir(d)
 
         skipstep3 = skipstep3 and not (work or dirsnotfound)
         work = [d for d in work if not dirignore(d[0])]
 
         # step 2: visit subdirectories
         def traverse(work, alreadynormed):
             wadd = work.append
             while work:
                 tracing.counter('dirstate.walk work', len(work))
                 nd = work.pop()
                 visitentries = match.visitchildrenset(nd)
                 if not visitentries:
                     continue
                 if visitentries == b'this' or visitentries == b'all':
                     visitentries = None
                 skip = None
                 if nd != b'':
                     skip = b'.hg'
                 try:
                     with tracing.log('dirstate.walk.traverse listdir %s', nd):
                         entries = listdir(join(nd), stat=True, skip=skip)
-                except OSError as inst:
-                    if inst.errno in (errno.EACCES, errno.ENOENT):
+                except (PermissionError, FileNotFoundError) as inst:
                         match.bad(
                             self.pathto(nd), encoding.strtolocal(inst.strerror)
                         )
                         continue
-                    raise
                 for f, kind, st in entries:
                     # Some matchers may return files in the visitentries set,
                     # instead of 'this', if the matcher explicitly mentions them
                     # and is not an exactmatcher. This is acceptable; we do not
                     # make any hard assumptions about file-or-directory below
                     # based on the presence of `f` in visitentries. If
                     # visitchildrenset returned a set, we can always skip the
                     # entries *not* in the set it provided regardless of whether
                     # they're actually a file or a directory.
                     if visitentries and f not in visitentries:
                         continue
                     if normalizefile:
                         # even though f might be a directory, we're only
                         # interested in comparing it to files currently in the
                         # dmap -- therefore normalizefile is enough
                         nf = normalizefile(
                             nd and (nd + b"/" + f) or f, True, True
                         )
                     else:
                         nf = nd and (nd + b"/" + f) or f
                     if nf not in results:
                         if kind == dirkind:
                             if not ignore(nf):
                                 if matchtdir:
                                     matchtdir(nf)
                                 wadd(nf)
                             if nf in dmap and (matchalways or matchfn(nf)):
                                 results[nf] = None
                         elif kind == regkind or kind == lnkkind:
                             if nf in dmap:
                                 if matchalways or matchfn(nf):
                                     results[nf] = st
                             elif (matchalways or matchfn(nf)) and not ignore(
                                 nf
                             ):
                                 # unknown file -- normalize if necessary
                                 if not alreadynormed:
                                     nf = normalize(nf, False, True)
                                 results[nf] = st
                         elif nf in dmap and (matchalways or matchfn(nf)):
                             results[nf] = None
 
         for nd, d in work:
             # alreadynormed means that processwork doesn't have to do any
             # expensive directory normalization
             alreadynormed = not normalize or nd == d
             traverse([d], alreadynormed)
 
         for s in subrepos:
             del results[s]
         del results[b'.hg']
 
         # step 3: visit remaining files from dmap
         if not skipstep3 and not exact:
             # If a dmap file is not in results yet, it was either
             # a) not matching matchfn b) ignored, c) missing, or d) under a
             # symlink directory.
             if not results and matchalways:
                 visit = [f for f in dmap]
             else:
                 visit = [f for f in dmap if f not in results and matchfn(f)]
             visit.sort()
 
             if unknown:
                 # unknown == True means we walked all dirs under the roots
                 # that wasn't ignored, and everything that matched was stat'ed
                 # and is already in results.
                 # The rest must thus be ignored or under a symlink.
                 audit_path = pathutil.pathauditor(self._root, cached=True)
 
                 for nf in iter(visit):
                     # If a stat for the same file was already added with a
                     # different case, don't add one for this, since that would
                     # make it appear as if the file exists under both names
                     # on disk.
                     if (
                         normalizefile
                         and normalizefile(nf, True, True) in results
                     ):
                         results[nf] = None
                     # Report ignored items in the dmap as long as they are not
                     # under a symlink directory.
                     elif audit_path.check(nf):
                         try:
                             results[nf] = lstat(join(nf))
                             # file was just ignored, no links, and exists
                         except OSError:
                             # file doesn't exist
                             results[nf] = None
                     else:
                         # It's either missing or under a symlink directory
                         # which we in this case report as missing
                         results[nf] = None
             else:
                 # We may not have walked the full directory tree above,
                 # so stat and check everything we missed.
                 iv = iter(visit)
                 for st in util.statfiles([join(i) for i in visit]):
                     results[next(iv)] = st
         return results
 
     def _rust_status(self, matcher, list_clean, list_ignored, list_unknown):
         # Force Rayon (Rust parallelism library) to respect the number of
         # workers. This is a temporary workaround until Rust code knows
         # how to read the config file.
         numcpus = self._ui.configint(b"worker", b"numcpus")
         if numcpus is not None:
             encoding.environ.setdefault(b'RAYON_NUM_THREADS', b'%d' % numcpus)
 
         workers_enabled = self._ui.configbool(b"worker", b"enabled", True)
         if not workers_enabled:
             encoding.environ[b"RAYON_NUM_THREADS"] = b"1"
 
         (
             lookup,
             modified,
             added,
             removed,
             deleted,
             clean,
             ignored,
             unknown,
             warnings,
             bad,
             traversed,
             dirty,
         ) = rustmod.status(
             self._map._map,
             matcher,
             self._rootdir,
             self._ignorefiles(),
             self._checkexec,
             bool(list_clean),
             bool(list_ignored),
             bool(list_unknown),
             bool(matcher.traversedir),
         )
 
         self._dirty |= dirty
 
         if matcher.traversedir:
             for dir in traversed:
                 matcher.traversedir(dir)
 
         if self._ui.warn:
             for item in warnings:
                 if isinstance(item, tuple):
                     file_path, syntax = item
                     msg = _(b"%s: ignoring invalid syntax '%s'\n") % (
                         file_path,
                         syntax,
                     )
                     self._ui.warn(msg)
                 else:
                     msg = _(b"skipping unreadable pattern file '%s': %s\n")
                     self._ui.warn(
                         msg
                         % (
                             pathutil.canonpath(
                                 self._rootdir, self._rootdir, item
                             ),
                             b"No such file or directory",
                         )
                     )
 
         for (fn, message) in bad:
             matcher.bad(fn, encoding.strtolocal(message))
 
         status = scmutil.status(
             modified=modified,
             added=added,
             removed=removed,
             deleted=deleted,
             unknown=unknown,
             ignored=ignored,
             clean=clean,
         )
         return (lookup, status)
 
     def status(self, match, subrepos, ignored, clean, unknown):
         """Determine the status of the working copy relative to the
         dirstate and return a pair of (unsure, status), where status is of type
         scmutil.status and:
 
           unsure:
             files that might have been modified since the dirstate was
             written, but need to be read to be sure (size is the same
             but mtime differs)
           status.modified:
             files that have definitely been modified since the dirstate
             was written (different size or mode)
           status.clean:
             files that have definitely not been modified since the
             dirstate was written
         """
         listignored, listclean, listunknown = ignored, clean, unknown
         lookup, modified, added, unknown, ignored = [], [], [], [], []
         removed, deleted, clean = [], [], []
 
         dmap = self._map
         dmap.preload()
 
         use_rust = True
 
         allowed_matchers = (
             matchmod.alwaysmatcher,
             matchmod.exactmatcher,
             matchmod.includematcher,
         )
 
         if rustmod is None:
             use_rust = False
         elif self._checkcase:
             # Case-insensitive filesystems are not handled yet
             use_rust = False
         elif subrepos:
             use_rust = False
         elif sparse.enabled:
             use_rust = False
         elif not isinstance(match, allowed_matchers):
             # Some matchers have yet to be implemented
             use_rust = False
 
         # Get the time from the filesystem so we can disambiguate files that
         # appear modified in the present or future.
         try:
             mtime_boundary = timestamp.get_fs_now(self._opener)
         except OSError:
             # In largefiles or readonly context
             mtime_boundary = None
 
         if use_rust:
             try:
                 res = self._rust_status(
                     match, listclean, listignored, listunknown
                 )
                 return res + (mtime_boundary,)
             except rustmod.FallbackError:
                 pass
 
         def noop(f):
             pass
 
         dcontains = dmap.__contains__
         dget = dmap.__getitem__
         ladd = lookup.append  # aka "unsure"
         madd = modified.append
         aadd = added.append
         uadd = unknown.append if listunknown else noop
         iadd = ignored.append if listignored else noop
         radd = removed.append
         dadd = deleted.append
         cadd = clean.append if listclean else noop
         mexact = match.exact
         dirignore = self._dirignore
         checkexec = self._checkexec
         checklink = self._checklink
         copymap = self._map.copymap
 
         # We need to do full walks when either
         # - we're listing all clean files, or
         # - match.traversedir does something, because match.traversedir should
         #   be called for every dir in the working dir
         full = listclean or match.traversedir is not None
         for fn, st in self.walk(
             match, subrepos, listunknown, listignored, full=full
         ).items():
             if not dcontains(fn):
                 if (listignored or mexact(fn)) and dirignore(fn):
                     if listignored:
                         iadd(fn)
                 else:
                     uadd(fn)
                 continue
 
             t = dget(fn)
             mode = t.mode
             size = t.size
 
             if not st and t.tracked:
                 dadd(fn)
             elif t.p2_info:
                 madd(fn)
             elif t.added:
                 aadd(fn)
             elif t.removed:
                 radd(fn)
             elif t.tracked:
                 if not checklink and t.has_fallback_symlink:
                     # If the file system does not support symlink, the mode
                     # might not be correctly stored in the dirstate, so do not
                     # trust it.
                     ladd(fn)
                 elif not checkexec and t.has_fallback_exec:
                     # If the file system does not support exec bits, the mode
                     # might not be correctly stored in the dirstate, so do not
                     # trust it.
                     ladd(fn)
                 elif (
                     size >= 0
                     and (
                         (size != st.st_size and size != st.st_size & _rangemask)
                         or ((mode ^ st.st_mode) & 0o100 and checkexec)
                     )
                     or fn in copymap
                 ):
                     if stat.S_ISLNK(st.st_mode) and size != st.st_size:
                         # issue6456: Size returned may be longer due to
                         # encryption on EXT-4 fscrypt, undecided.
                         ladd(fn)
                     else:
                         madd(fn)
                 elif not t.mtime_likely_equal_to(timestamp.mtime_of(st)):
                     # There might be a change in the future if for example the
                     # internal clock is off, but this is a case where the issues
                     # the user would face would be a lot worse and there is
                     # nothing we can really do.
                     ladd(fn)
                 elif listclean:
                     cadd(fn)
         status = scmutil.status(
             modified, added, removed, deleted, unknown, ignored, clean
         )
         return (lookup, status, mtime_boundary)
 
     def matches(self, match):
         """
         return files in the dirstate (in whatever state) filtered by match
         """
         dmap = self._map
         if rustmod is not None:
             dmap = self._map._map
 
         if match.always():
             return dmap.keys()
         files = match.files()
         if match.isexact():
             # fast path -- filter the other way around, since typically files is
             # much smaller than dmap
             return [f for f in files if f in dmap]
         if match.prefix() and all(fn in dmap for fn in files):
             # fast path -- all the values are known to be files, so just return
             # that
             return list(files)
         return [f for f in dmap if match(f)]
 
     def _actualfilename(self, tr):
         if tr:
             return self._pendingfilename
         else:
             return self._filename
 
     def savebackup(self, tr, backupname):
         '''Save current dirstate into backup file'''
         filename = self._actualfilename(tr)
         assert backupname != filename
 
         # use '_writedirstate' instead of 'write' to write changes certainly,
         # because the latter omits writing out if transaction is running.
         # output file will be used to create backup of dirstate at this point.
         if self._dirty or not self._opener.exists(filename):
             self._writedirstate(
                 tr,
                 self._opener(filename, b"w", atomictemp=True, checkambig=True),
             )
 
         if tr:
             # ensure that subsequent tr.writepending returns True for
             # changes written out above, even if dirstate is never
             # changed after this
             tr.addfilegenerator(
                 b'dirstate-1-main',
                 (self._filename,),
                 lambda f: self._writedirstate(tr, f),
                 location=b'plain',
                 post_finalize=True,
             )
 
             # ensure that pending file written above is unlinked at
             # failure, even if tr.writepending isn't invoked until the
             # end of this transaction
             tr.registertmp(filename, location=b'plain')
 
         self._opener.tryunlink(backupname)
         # hardlink backup is okay because _writedirstate is always called
         # with an "atomictemp=True" file.
         util.copyfile(
             self._opener.join(filename),
             self._opener.join(backupname),
             hardlink=True,
         )
 
     def restorebackup(self, tr, backupname):
         '''Restore dirstate by backup file'''
         # this "invalidate()" prevents "wlock.release()" from writing
         # changes of dirstate out after restoring from backup file
         self.invalidate()
         filename = self._actualfilename(tr)
         o = self._opener
         if util.samefile(o.join(backupname), o.join(filename)):
             o.unlink(backupname)
         else:
             o.rename(backupname, filename, checkambig=True)
 
     def clearbackup(self, tr, backupname):
         '''Clear backup file'''
         self._opener.unlink(backupname)
 
     def verify(self, m1, m2):
         """check the dirstate content again the parent manifest and yield errors"""
         missing_from_p1 = b"%s in state %s, but not in manifest1\n"
         unexpected_in_p1 = b"%s in state %s, but also in manifest1\n"
         missing_from_ps = b"%s in state %s, but not in either manifest\n"
         missing_from_ds = b"%s in manifest1, but listed as state %s\n"
         for f, entry in self.items():
             state = entry.state
             if state in b"nr" and f not in m1:
                 yield (missing_from_p1, f, state)
             if state in b"a" and f in m1:
                 yield (unexpected_in_p1, f, state)
             if state in b"m" and f not in m1 and f not in m2:
                 yield (missing_from_ps, f, state)
         for f in m1:
             state = self.get_entry(f).state
             if state not in b"nrm":
                 yield (missing_from_ds, f, state)

mercurial/fileset.py

@@ -1,621 +1,618 b''
 # fileset.py - file set queries for mercurial
 #
 # Copyright 2010 Olivia 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.
 
 
-import errno
 import re
 
 from .i18n import _
 from .pycompat import getattr
 from . import (
     error,
     filesetlang,
     match as matchmod,
     mergestate as mergestatemod,
     pycompat,
     registrar,
     scmutil,
     util,
 )
 from .utils import stringutil
 
 # common weight constants
 _WEIGHT_CHECK_FILENAME = filesetlang.WEIGHT_CHECK_FILENAME
 _WEIGHT_READ_CONTENTS = filesetlang.WEIGHT_READ_CONTENTS
 _WEIGHT_STATUS = filesetlang.WEIGHT_STATUS
 _WEIGHT_STATUS_THOROUGH = filesetlang.WEIGHT_STATUS_THOROUGH
 
 # helpers for processing parsed tree
 getsymbol = filesetlang.getsymbol
 getstring = filesetlang.getstring
 _getkindpat = filesetlang.getkindpat
 getpattern = filesetlang.getpattern
 getargs = filesetlang.getargs
 
 
 def getmatch(mctx, x):
     if not x:
         raise error.ParseError(_(b"missing argument"))
     return methods[x[0]](mctx, *x[1:])
 
 
 def getmatchwithstatus(mctx, x, hint):
     keys = set(getstring(hint, b'status hint must be a string').split())
     return getmatch(mctx.withstatus(keys), x)
 
 
 def stringmatch(mctx, x):
     return mctx.matcher([x])
 
 
 def kindpatmatch(mctx, x, y):
     return stringmatch(
         mctx,
         _getkindpat(
             x, y, matchmod.allpatternkinds, _(b"pattern must be a string")
         ),
     )
 
 
 def patternsmatch(mctx, *xs):
     allkinds = matchmod.allpatternkinds
     patterns = [
         getpattern(x, allkinds, _(b"pattern must be a string")) for x in xs
     ]
     return mctx.matcher(patterns)
 
 
 def andmatch(mctx, x, y):
     xm = getmatch(mctx, x)
     ym = getmatch(mctx.narrowed(xm), y)
     return matchmod.intersectmatchers(xm, ym)
 
 
 def ormatch(mctx, *xs):
     ms = [getmatch(mctx, x) for x in xs]
     return matchmod.unionmatcher(ms)
 
 
 def notmatch(mctx, x):
     m = getmatch(mctx, x)
     return mctx.predicate(lambda f: not m(f), predrepr=(b'<not %r>', m))
 
 
 def minusmatch(mctx, x, y):
     xm = getmatch(mctx, x)
     ym = getmatch(mctx.narrowed(xm), y)
     return matchmod.differencematcher(xm, ym)
 
 
 def listmatch(mctx, *xs):
     raise error.ParseError(
         _(b"can't use a list in this context"),
         hint=_(b'see \'hg help "filesets.x or y"\''),
     )
 
 
 def func(mctx, a, b):
     funcname = getsymbol(a)
     if funcname in symbols:
         return symbols[funcname](mctx, b)
 
     keep = lambda fn: getattr(fn, '__doc__', None) is not None
 
     syms = [s for (s, fn) in symbols.items() if keep(fn)]
     raise error.UnknownIdentifier(funcname, syms)
 
 
 # symbols are callable like:
 #  fun(mctx, x)
 # with:
 #  mctx - current matchctx instance
 #  x - argument in tree form
 symbols = filesetlang.symbols
 
 predicate = registrar.filesetpredicate(symbols)
 
 
 @predicate(b'modified()', callstatus=True, weight=_WEIGHT_STATUS)
 def modified(mctx, x):
     """File that is modified according to :hg:`status`."""
     # i18n: "modified" is a keyword
     getargs(x, 0, 0, _(b"modified takes no arguments"))
     s = set(mctx.status().modified)
     return mctx.predicate(s.__contains__, predrepr=b'modified')
 
 
 @predicate(b'added()', callstatus=True, weight=_WEIGHT_STATUS)
 def added(mctx, x):
     """File that is added according to :hg:`status`."""
     # i18n: "added" is a keyword
     getargs(x, 0, 0, _(b"added takes no arguments"))
     s = set(mctx.status().added)
     return mctx.predicate(s.__contains__, predrepr=b'added')
 
 
 @predicate(b'removed()', callstatus=True, weight=_WEIGHT_STATUS)
 def removed(mctx, x):
     """File that is removed according to :hg:`status`."""
     # i18n: "removed" is a keyword
     getargs(x, 0, 0, _(b"removed takes no arguments"))
     s = set(mctx.status().removed)
     return mctx.predicate(s.__contains__, predrepr=b'removed')
 
 
 @predicate(b'deleted()', callstatus=True, weight=_WEIGHT_STATUS)
 def deleted(mctx, x):
     """Alias for ``missing()``."""
     # i18n: "deleted" is a keyword
     getargs(x, 0, 0, _(b"deleted takes no arguments"))
     s = set(mctx.status().deleted)
     return mctx.predicate(s.__contains__, predrepr=b'deleted')
 
 
 @predicate(b'missing()', callstatus=True, weight=_WEIGHT_STATUS)
 def missing(mctx, x):
     """File that is missing according to :hg:`status`."""
     # i18n: "missing" is a keyword
     getargs(x, 0, 0, _(b"missing takes no arguments"))
     s = set(mctx.status().deleted)
     return mctx.predicate(s.__contains__, predrepr=b'deleted')
 
 
 @predicate(b'unknown()', callstatus=True, weight=_WEIGHT_STATUS_THOROUGH)
 def unknown(mctx, x):
     """File that is unknown according to :hg:`status`."""
     # i18n: "unknown" is a keyword
     getargs(x, 0, 0, _(b"unknown takes no arguments"))
     s = set(mctx.status().unknown)
     return mctx.predicate(s.__contains__, predrepr=b'unknown')
 
 
 @predicate(b'ignored()', callstatus=True, weight=_WEIGHT_STATUS_THOROUGH)
 def ignored(mctx, x):
     """File that is ignored according to :hg:`status`."""
     # i18n: "ignored" is a keyword
     getargs(x, 0, 0, _(b"ignored takes no arguments"))
     s = set(mctx.status().ignored)
     return mctx.predicate(s.__contains__, predrepr=b'ignored')
 
 
 @predicate(b'clean()', callstatus=True, weight=_WEIGHT_STATUS)
 def clean(mctx, x):
     """File that is clean according to :hg:`status`."""
     # i18n: "clean" is a keyword
     getargs(x, 0, 0, _(b"clean takes no arguments"))
     s = set(mctx.status().clean)
     return mctx.predicate(s.__contains__, predrepr=b'clean')
 
 
 @predicate(b'tracked()')
 def tracked(mctx, x):
     """File that is under Mercurial control."""
     # i18n: "tracked" is a keyword
     getargs(x, 0, 0, _(b"tracked takes no arguments"))
     return mctx.predicate(mctx.ctx.__contains__, predrepr=b'tracked')
 
 
 @predicate(b'binary()', weight=_WEIGHT_READ_CONTENTS)
 def binary(mctx, x):
     """File that appears to be binary (contains NUL bytes)."""
     # i18n: "binary" is a keyword
     getargs(x, 0, 0, _(b"binary takes no arguments"))
     return mctx.fpredicate(
         lambda fctx: fctx.isbinary(), predrepr=b'binary', cache=True
     )
 
 
 @predicate(b'exec()')
 def exec_(mctx, x):
     """File that is marked as executable."""
     # i18n: "exec" is a keyword
     getargs(x, 0, 0, _(b"exec takes no arguments"))
     ctx = mctx.ctx
     return mctx.predicate(lambda f: ctx.flags(f) == b'x', predrepr=b'exec')
 
 
 @predicate(b'symlink()')
 def symlink(mctx, x):
     """File that is marked as a symlink."""
     # i18n: "symlink" is a keyword
     getargs(x, 0, 0, _(b"symlink takes no arguments"))
     ctx = mctx.ctx
     return mctx.predicate(lambda f: ctx.flags(f) == b'l', predrepr=b'symlink')
 
 
 @predicate(b'resolved()', weight=_WEIGHT_STATUS)
 def resolved(mctx, x):
     """File that is marked resolved according to :hg:`resolve -l`."""
     # i18n: "resolved" is a keyword
     getargs(x, 0, 0, _(b"resolved takes no arguments"))
     if mctx.ctx.rev() is not None:
         return mctx.never()
     ms = mergestatemod.mergestate.read(mctx.ctx.repo())
     return mctx.predicate(
         lambda f: f in ms and ms[f] == b'r', predrepr=b'resolved'
     )
 
 
 @predicate(b'unresolved()', weight=_WEIGHT_STATUS)
 def unresolved(mctx, x):
     """File that is marked unresolved according to :hg:`resolve -l`."""
     # i18n: "unresolved" is a keyword
     getargs(x, 0, 0, _(b"unresolved takes no arguments"))
     if mctx.ctx.rev() is not None:
         return mctx.never()
     ms = mergestatemod.mergestate.read(mctx.ctx.repo())
     return mctx.predicate(
         lambda f: f in ms and ms[f] == b'u', predrepr=b'unresolved'
     )
 
 
 @predicate(b'hgignore()', weight=_WEIGHT_STATUS)
 def hgignore(mctx, x):
     """File that matches the active .hgignore pattern."""
     # i18n: "hgignore" is a keyword
     getargs(x, 0, 0, _(b"hgignore takes no arguments"))
     return mctx.ctx.repo().dirstate._ignore
 
 
 @predicate(b'portable()', weight=_WEIGHT_CHECK_FILENAME)
 def portable(mctx, x):
     """File that has a portable name. (This doesn't include filenames with case
     collisions.)
     """
     # i18n: "portable" is a keyword
     getargs(x, 0, 0, _(b"portable takes no arguments"))
     return mctx.predicate(
         lambda f: util.checkwinfilename(f) is None, predrepr=b'portable'
     )
 
 
 @predicate(b'grep(regex)', weight=_WEIGHT_READ_CONTENTS)
 def grep(mctx, x):
     """File contains the given regular expression."""
     try:
         # i18n: "grep" is a keyword
         r = re.compile(getstring(x, _(b"grep requires a pattern")))
     except re.error as e:
         raise error.ParseError(
             _(b'invalid match pattern: %s') % stringutil.forcebytestr(e)
         )
     return mctx.fpredicate(
         lambda fctx: r.search(fctx.data()),
         predrepr=(b'grep(%r)', r.pattern),
         cache=True,
     )
 
 
 def _sizetomax(s):
     try:
         s = s.strip().lower()
         for k, v in util._sizeunits:
             if s.endswith(k):
                 # max(4k) = 5k - 1, max(4.5k) = 4.6k - 1
                 n = s[: -len(k)]
                 inc = 1.0
                 if b"." in n:
                     inc /= 10 ** len(n.split(b".")[1])
                 return int((float(n) + inc) * v) - 1
         # no extension, this is a precise value
         return int(s)
     except ValueError:
         raise error.ParseError(_(b"couldn't parse size: %s") % s)
 
 
 def sizematcher(expr):
     """Return a function(size) -> bool from the ``size()`` expression"""
     expr = expr.strip()
     if b'-' in expr:  # do we have a range?
         a, b = expr.split(b'-', 1)
         a = util.sizetoint(a)
         b = util.sizetoint(b)
         return lambda x: x >= a and x <= b
     elif expr.startswith(b"<="):
         a = util.sizetoint(expr[2:])
         return lambda x: x <= a
     elif expr.startswith(b"<"):
         a = util.sizetoint(expr[1:])
         return lambda x: x < a
     elif expr.startswith(b">="):
         a = util.sizetoint(expr[2:])
         return lambda x: x >= a
     elif expr.startswith(b">"):
         a = util.sizetoint(expr[1:])
         return lambda x: x > a
     else:
         a = util.sizetoint(expr)
         b = _sizetomax(expr)
         return lambda x: x >= a and x <= b
 
 
 @predicate(b'size(expression)', weight=_WEIGHT_STATUS)
 def size(mctx, x):
     """File size matches the given expression. Examples:
 
     - size('1k') - files from 1024 to 2047 bytes
     - size('< 20k') - files less than 20480 bytes
     - size('>= .5MB') - files at least 524288 bytes
     - size('4k - 1MB') - files from 4096 bytes to 1048576 bytes
     """
     # i18n: "size" is a keyword
     expr = getstring(x, _(b"size requires an expression"))
     m = sizematcher(expr)
     return mctx.fpredicate(
         lambda fctx: m(fctx.size()), predrepr=(b'size(%r)', expr), cache=True
     )
 
 
 @predicate(b'encoding(name)', weight=_WEIGHT_READ_CONTENTS)
 def encoding(mctx, x):
     """File can be successfully decoded with the given character
     encoding. May not be useful for encodings other than ASCII and
     UTF-8.
     """
 
     # i18n: "encoding" is a keyword
     enc = getstring(x, _(b"encoding requires an encoding name"))
 
     def encp(fctx):
         d = fctx.data()
         try:
             d.decode(pycompat.sysstr(enc))
             return True
         except LookupError:
             raise error.Abort(_(b"unknown encoding '%s'") % enc)
         except UnicodeDecodeError:
             return False
 
     return mctx.fpredicate(encp, predrepr=(b'encoding(%r)', enc), cache=True)
 
 
 @predicate(b'eol(style)', weight=_WEIGHT_READ_CONTENTS)
 def eol(mctx, x):
     """File contains newlines of the given style (dos, unix, mac). Binary
     files are excluded, files with mixed line endings match multiple
     styles.
     """
 
     # i18n: "eol" is a keyword
     enc = getstring(x, _(b"eol requires a style name"))
 
     def eolp(fctx):
         if fctx.isbinary():
             return False
         d = fctx.data()
         if (enc == b'dos' or enc == b'win') and b'\r\n' in d:
             return True
         elif enc == b'unix' and re.search(b'(?<!\r)\n', d):
             return True
         elif enc == b'mac' and re.search(b'\r(?!\n)', d):
             return True
         return False
 
     return mctx.fpredicate(eolp, predrepr=(b'eol(%r)', enc), cache=True)
 
 
 @predicate(b'copied()')
 def copied(mctx, x):
     """File that is recorded as being copied."""
     # i18n: "copied" is a keyword
     getargs(x, 0, 0, _(b"copied takes no arguments"))
 
     def copiedp(fctx):
         p = fctx.parents()
         return p and p[0].path() != fctx.path()
 
     return mctx.fpredicate(copiedp, predrepr=b'copied', cache=True)
 
 
 @predicate(b'revs(revs, pattern)', weight=_WEIGHT_STATUS)
 def revs(mctx, x):
     """Evaluate set in the specified revisions. If the revset match multiple
     revs, this will return file matching pattern in any of the revision.
     """
     # i18n: "revs" is a keyword
     r, x = getargs(x, 2, 2, _(b"revs takes two arguments"))
     # i18n: "revs" is a keyword
     revspec = getstring(r, _(b"first argument to revs must be a revision"))
     repo = mctx.ctx.repo()
     revs = scmutil.revrange(repo, [revspec])
 
     matchers = []
     for r in revs:
         ctx = repo[r]
         mc = mctx.switch(ctx.p1(), ctx)
         matchers.append(getmatch(mc, x))
     if not matchers:
         return mctx.never()
     if len(matchers) == 1:
         return matchers[0]
     return matchmod.unionmatcher(matchers)
 
 
 @predicate(b'status(base, rev, pattern)', weight=_WEIGHT_STATUS)
 def status(mctx, x):
     """Evaluate predicate using status change between ``base`` and
     ``rev``. Examples:
 
     - ``status(3, 7, added())`` - matches files added from "3" to "7"
     """
     repo = mctx.ctx.repo()
     # i18n: "status" is a keyword
     b, r, x = getargs(x, 3, 3, _(b"status takes three arguments"))
     # i18n: "status" is a keyword
     baseerr = _(b"first argument to status must be a revision")
     baserevspec = getstring(b, baseerr)
     if not baserevspec:
         raise error.ParseError(baseerr)
     reverr = _(b"second argument to status must be a revision")
     revspec = getstring(r, reverr)
     if not revspec:
         raise error.ParseError(reverr)
     basectx, ctx = scmutil.revpair(repo, [baserevspec, revspec])
     mc = mctx.switch(basectx, ctx)
     return getmatch(mc, x)
 
 
 @predicate(b'subrepo([pattern])')
 def subrepo(mctx, x):
     """Subrepositories whose paths match the given pattern."""
     # i18n: "subrepo" is a keyword
     getargs(x, 0, 1, _(b"subrepo takes at most one argument"))
     ctx = mctx.ctx
     sstate = ctx.substate
     if x:
         pat = getpattern(
             x,
             matchmod.allpatternkinds,
             # i18n: "subrepo" is a keyword
             _(b"subrepo requires a pattern or no arguments"),
         )
         fast = not matchmod.patkind(pat)
         if fast:
 
             def m(s):
                 return s == pat
 
         else:
             m = matchmod.match(ctx.repo().root, b'', [pat], ctx=ctx)
         return mctx.predicate(
             lambda f: f in sstate and m(f), predrepr=(b'subrepo(%r)', pat)
         )
     else:
         return mctx.predicate(sstate.__contains__, predrepr=b'subrepo')
 
 
 methods = {
     b'withstatus': getmatchwithstatus,
     b'string': stringmatch,
     b'symbol': stringmatch,
     b'kindpat': kindpatmatch,
     b'patterns': patternsmatch,
     b'and': andmatch,
     b'or': ormatch,
     b'minus': minusmatch,
     b'list': listmatch,
     b'not': notmatch,
     b'func': func,
 }
 
 
 class matchctx:
     def __init__(self, basectx, ctx, cwd, badfn=None):
         self._basectx = basectx
         self.ctx = ctx
         self._badfn = badfn
         self._match = None
         self._status = None
         self.cwd = cwd
 
     def narrowed(self, match):
         """Create matchctx for a sub-tree narrowed by the given matcher"""
         mctx = matchctx(self._basectx, self.ctx, self.cwd, self._badfn)
         mctx._match = match
         # leave wider status which we don't have to care
         mctx._status = self._status
         return mctx
 
     def switch(self, basectx, ctx):
         mctx = matchctx(basectx, ctx, self.cwd, self._badfn)
         mctx._match = self._match
         return mctx
 
     def withstatus(self, keys):
         """Create matchctx which has precomputed status specified by the keys"""
         mctx = matchctx(self._basectx, self.ctx, self.cwd, self._badfn)
         mctx._match = self._match
         mctx._buildstatus(keys)
         return mctx
 
     def _buildstatus(self, keys):
         self._status = self._basectx.status(
             self.ctx,
             self._match,
             listignored=b'ignored' in keys,
             listclean=b'clean' in keys,
             listunknown=b'unknown' in keys,
         )
 
     def status(self):
         return self._status
 
     def matcher(self, patterns):
         return self.ctx.match(patterns, badfn=self._badfn, cwd=self.cwd)
 
     def predicate(self, predfn, predrepr=None, cache=False):
         """Create a matcher to select files by predfn(filename)"""
         if cache:
             predfn = util.cachefunc(predfn)
         return matchmod.predicatematcher(
             predfn, predrepr=predrepr, badfn=self._badfn
         )
 
     def fpredicate(self, predfn, predrepr=None, cache=False):
         """Create a matcher to select files by predfn(fctx) at the current
         revision
 
         Missing files are ignored.
         """
         ctx = self.ctx
         if ctx.rev() is None:
 
             def fctxpredfn(f):
                 try:
                     fctx = ctx[f]
                 except error.LookupError:
                     return False
                 try:
                     fctx.audit()
                 except error.Abort:
                     return False
                 try:
                     return predfn(fctx)
-                except (IOError, OSError) as e:
-                    # open()-ing a directory fails with EACCES on Windows
-                    if e.errno in (
-                        errno.ENOENT,
-                        errno.EACCES,
-                        errno.ENOTDIR,
-                        errno.EISDIR,
+                # open()-ing a directory fails with PermissionError on Windows
+                except (
+                    FileNotFoundError,
+                    PermissionError,
+                    NotADirectoryError,
+                    IsADirectoryError,
                     ):
                         return False
-                    raise
 
         else:
 
             def fctxpredfn(f):
                 try:
                     fctx = ctx[f]
                 except error.LookupError:
                     return False
                 return predfn(fctx)
 
         return self.predicate(fctxpredfn, predrepr=predrepr, cache=cache)
 
     def never(self):
         """Create a matcher to select nothing"""
         return matchmod.never(badfn=self._badfn)
 
 
 def match(ctx, cwd, expr, badfn=None):
     """Create a matcher for a single fileset expression"""
     tree = filesetlang.parse(expr)
     tree = filesetlang.analyze(tree)
     tree = filesetlang.optimize(tree)
     mctx = matchctx(ctx.p1(), ctx, cwd, badfn=badfn)
     return getmatch(mctx, tree)
 
 
 def loadpredicate(ui, extname, registrarobj):
     """Load fileset predicates from specified registrarobj"""
     for name, func in registrarobj._table.items():
         symbols[name] = func
 
 
 # tell hggettext to extract docstrings from these functions:
 i18nfunctions = symbols.values()

mercurial/lock.py

@@ -1,394 +1,393 b''
 # lock.py - simple advisory locking scheme for mercurial
 #
 # Copyright 2005, 2006 Olivia 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.
 
 
 import contextlib
 import errno
 import os
 import signal
 import socket
 import time
 import warnings
 
 from .i18n import _
 from .pycompat import getattr
 
 from . import (
     encoding,
     error,
     pycompat,
     util,
 )
 
 from .utils import procutil
 
 
 def _getlockprefix():
     """Return a string which is used to differentiate pid namespaces
 
     It's useful to detect "dead" processes and remove stale locks with
     confidence. Typically it's just hostname. On modern linux, we include an
     extra Linux-specific pid namespace identifier.
     """
     result = encoding.strtolocal(socket.gethostname())
     if pycompat.sysplatform.startswith(b'linux'):
         try:
             result += b'/%x' % os.stat(b'/proc/self/ns/pid').st_ino
-        except OSError as ex:
-            if ex.errno not in (errno.ENOENT, errno.EACCES, errno.ENOTDIR):
-                raise
+        except (FileNotFoundError, PermissionError, NotADirectoryError):
+            pass
     return result
 
 
 @contextlib.contextmanager
 def _delayedinterrupt():
     """Block signal interrupt while doing something critical
 
     This makes sure that the code block wrapped by this context manager won't
     be interrupted.
 
     For Windows developers: It appears not possible to guard time.sleep()
     from CTRL_C_EVENT, so please don't use time.sleep() to test if this is
     working.
     """
     assertedsigs = []
     blocked = False
     orighandlers = {}
 
     def raiseinterrupt(num):
         if num == getattr(signal, 'SIGINT', None) or num == getattr(
             signal, 'CTRL_C_EVENT', None
         ):
             raise KeyboardInterrupt
         else:
             raise error.SignalInterrupt
 
     def catchterm(num, frame):
         if blocked:
             assertedsigs.append(num)
         else:
             raiseinterrupt(num)
 
     try:
         # save handlers first so they can be restored even if a setup is
         # interrupted between signal.signal() and orighandlers[] =.
         for name in [
             b'CTRL_C_EVENT',
             b'SIGINT',
             b'SIGBREAK',
             b'SIGHUP',
             b'SIGTERM',
         ]:
             num = getattr(signal, name, None)
             if num and num not in orighandlers:
                 orighandlers[num] = signal.getsignal(num)
         try:
             for num in orighandlers:
                 signal.signal(num, catchterm)
         except ValueError:
             pass  # in a thread? no luck
 
         blocked = True
         yield
     finally:
         # no simple way to reliably restore all signal handlers because
         # any loops, recursive function calls, except blocks, etc. can be
         # interrupted. so instead, make catchterm() raise interrupt.
         blocked = False
         try:
             for num, handler in orighandlers.items():
                 signal.signal(num, handler)
         except ValueError:
             pass  # in a thread?
 
     # re-raise interrupt exception if any, which may be shadowed by a new
     # interrupt occurred while re-raising the first one
     if assertedsigs:
         raiseinterrupt(assertedsigs[0])
 
 
 def trylock(ui, vfs, lockname, timeout, warntimeout, *args, **kwargs):
     """return an acquired lock or raise an a LockHeld exception
 
     This function is responsible to issue warnings and or debug messages about
     the held lock while trying to acquires it."""
 
     def printwarning(printer, locker):
         """issue the usual "waiting on lock" message through any channel"""
         # show more details for new-style locks
         if b':' in locker:
             host, pid = locker.split(b":", 1)
             msg = _(
                 b"waiting for lock on %s held by process %r on host %r\n"
             ) % (
                 pycompat.bytestr(l.desc),
                 pycompat.bytestr(pid),
                 pycompat.bytestr(host),
             )
         else:
             msg = _(b"waiting for lock on %s held by %r\n") % (
                 l.desc,
                 pycompat.bytestr(locker),
             )
         printer(msg)
 
     l = lock(vfs, lockname, 0, *args, dolock=False, **kwargs)
 
     debugidx = 0 if (warntimeout and timeout) else -1
     warningidx = 0
     if not timeout:
         warningidx = -1
     elif warntimeout:
         warningidx = warntimeout
 
     delay = 0
     while True:
         try:
             l._trylock()
             break
         except error.LockHeld as inst:
             if delay == debugidx:
                 printwarning(ui.debug, inst.locker)
             if delay == warningidx:
                 printwarning(ui.warn, inst.locker)
             if timeout <= delay:
                 raise error.LockHeld(
                     errno.ETIMEDOUT, inst.filename, l.desc, inst.locker
                 )
             time.sleep(1)
             delay += 1
 
     l.delay = delay
     if l.delay:
         if 0 <= warningidx <= l.delay:
             ui.warn(_(b"got lock after %d seconds\n") % l.delay)
         else:
             ui.debug(b"got lock after %d seconds\n" % l.delay)
     if l.acquirefn:
         l.acquirefn()
     return l
 
 
 class lock:
     """An advisory lock held by one process to control access to a set
     of files.  Non-cooperating processes or incorrectly written scripts
     can ignore Mercurial's locking scheme and stomp all over the
     repository, so don't do that.
 
     Typically used via localrepository.lock() to lock the repository
     store (.hg/store/) or localrepository.wlock() to lock everything
     else under .hg/."""
 
     # lock is symlink on platforms that support it, file on others.
 
     # symlink is used because create of directory entry and contents
     # are atomic even over nfs.
 
     # old-style lock: symlink to pid
     # new-style lock: symlink to hostname:pid
 
     _host = None
 
     def __init__(
         self,
         vfs,
         fname,
         timeout=-1,
         releasefn=None,
         acquirefn=None,
         desc=None,
         signalsafe=True,
         dolock=True,
     ):
         self.vfs = vfs
         self.f = fname
         self.held = 0
         self.timeout = timeout
         self.releasefn = releasefn
         self.acquirefn = acquirefn
         self.desc = desc
         if signalsafe:
             self._maybedelayedinterrupt = _delayedinterrupt
         else:
             self._maybedelayedinterrupt = util.nullcontextmanager
         self.postrelease = []
         self.pid = self._getpid()
         if dolock:
             self.delay = self.lock()
             if self.acquirefn:
                 self.acquirefn()
 
     def __enter__(self):
         return self
 
     def __exit__(self, exc_type, exc_value, exc_tb):
         success = all(a is None for a in (exc_type, exc_value, exc_tb))
         self.release(success=success)
 
     def __del__(self):
         if self.held:
             warnings.warn(
                 "use lock.release instead of del lock",
                 category=DeprecationWarning,
                 stacklevel=2,
             )
 
             # ensure the lock will be removed
             # even if recursive locking did occur
             self.held = 1
 
         self.release()
 
     def _getpid(self):
         # wrapper around procutil.getpid() to make testing easier
         return procutil.getpid()
 
     def lock(self):
         timeout = self.timeout
         while True:
             try:
                 self._trylock()
                 return self.timeout - timeout
             except error.LockHeld as inst:
                 if timeout != 0:
                     time.sleep(1)
                     if timeout > 0:
                         timeout -= 1
                     continue
                 raise error.LockHeld(
                     errno.ETIMEDOUT, inst.filename, self.desc, inst.locker
                 )
 
     def _trylock(self):
         if self.held:
             self.held += 1
             return
         if lock._host is None:
             lock._host = _getlockprefix()
         lockname = b'%s:%d' % (lock._host, self.pid)
         retry = 5
         while not self.held and retry:
             retry -= 1
             try:
                 with self._maybedelayedinterrupt():
                     self.vfs.makelock(lockname, self.f)
                     self.held = 1
             except (OSError, IOError) as why:
                 if why.errno == errno.EEXIST:
                     locker = self._readlock()
                     if locker is None:
                         continue
 
                     locker = self._testlock(locker)
                     if locker is not None:
                         raise error.LockHeld(
                             errno.EAGAIN,
                             self.vfs.join(self.f),
                             self.desc,
                             locker,
                         )
                 else:
                     raise error.LockUnavailable(
                         why.errno, why.strerror, why.filename, self.desc
                     )
 
         if not self.held:
             # use empty locker to mean "busy for frequent lock/unlock
             # by many processes"
             raise error.LockHeld(
                 errno.EAGAIN, self.vfs.join(self.f), self.desc, b""
             )
 
     def _readlock(self):
         """read lock and return its value
 
         Returns None if no lock exists, pid for old-style locks, and host:pid
         for new-style locks.
         """
         try:
             return self.vfs.readlock(self.f)
         except FileNotFoundError:
             return None
 
     def _lockshouldbebroken(self, locker):
         if locker is None:
             return False
         try:
             host, pid = locker.split(b":", 1)
         except ValueError:
             return False
         if host != lock._host:
             return False
         try:
             pid = int(pid)
         except ValueError:
             return False
         if procutil.testpid(pid):
             return False
         return True
 
     def _testlock(self, locker):
         if not self._lockshouldbebroken(locker):
             return locker
 
         # if locker dead, break lock.  must do this with another lock
         # held, or can race and break valid lock.
         try:
             with lock(self.vfs, self.f + b'.break', timeout=0):
                 locker = self._readlock()
                 if not self._lockshouldbebroken(locker):
                     return locker
                 self.vfs.unlink(self.f)
         except error.LockError:
             return locker
 
     def testlock(self):
         """return id of locker if lock is valid, else None.
 
         If old-style lock, we cannot tell what machine locker is on.
         with new-style lock, if locker is on this machine, we can
         see if locker is alive.  If locker is on this machine but
         not alive, we can safely break lock.
 
         The lock file is only deleted when None is returned.
 
         """
         locker = self._readlock()
         return self._testlock(locker)
 
     def release(self, success=True):
         """release the lock and execute callback function if any
 
         If the lock has been acquired multiple times, the actual release is
         delayed to the last release call."""
         if self.held > 1:
             self.held -= 1
         elif self.held == 1:
             self.held = 0
             if self._getpid() != self.pid:
                 # we forked, and are not the parent
                 return
             try:
                 if self.releasefn:
                     self.releasefn()
             finally:
                 try:
                     self.vfs.unlink(self.f)
                 except OSError:
                     pass
             # The postrelease functions typically assume the lock is not held
             # at all.
             for callback in self.postrelease:
                 callback(success)
             # Prevent double usage and help clear cycles.
             self.postrelease = None
 
 
 def release(*locks):
     for lock in locks:
         if lock is not None:
             lock.release()

mercurial/posix.py

@@ -1,755 +1,753 b''
 # posix.py - Posix utility function implementations for Mercurial
 #
 #  Copyright 2005-2009 Olivia Mackall <[e-mail]> and others
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 
 import errno
 import fcntl
 import getpass
 import grp
 import os
 import pwd
 import re
 import select
 import stat
 import sys
 import tempfile
 import unicodedata
 
 from .i18n import _
 from .pycompat import (
     getattr,
     open,
 )
 from . import (
     encoding,
     error,
     policy,
     pycompat,
 )
 
 osutil = policy.importmod('osutil')
 
 normpath = os.path.normpath
 samestat = os.path.samestat
 abspath = os.path.abspath  # re-exports
 
 try:
     oslink = os.link
 except AttributeError:
     # Some platforms build Python without os.link on systems that are
     # vaguely unix-like but don't have hardlink support. For those
     # poor souls, just say we tried and that it failed so we fall back
     # to copies.
     def oslink(src, dst):
         raise OSError(
             errno.EINVAL, b'hardlinks not supported: %s to %s' % (src, dst)
         )
 
 
 readlink = os.readlink
 unlink = os.unlink
 rename = os.rename
 removedirs = os.removedirs
 expandglobs = False
 
 umask = os.umask(0)
 os.umask(umask)
 
 posixfile = open
 
 
 def split(p):
     """Same as posixpath.split, but faster
 
     >>> import posixpath
     >>> for f in [b'/absolute/path/to/file',
     ...           b'relative/path/to/file',
     ...           b'file_alone',
     ...           b'path/to/directory/',
     ...           b'/multiple/path//separators',
     ...           b'/file_at_root',
     ...           b'///multiple_leading_separators_at_root',
     ...           b'']:
     ...     assert split(f) == posixpath.split(f), f
     """
     ht = p.rsplit(b'/', 1)
     if len(ht) == 1:
         return b'', p
     nh = ht[0].rstrip(b'/')
     if nh:
         return nh, ht[1]
     return ht[0] + b'/', ht[1]
 
 
 def openhardlinks():
     '''return true if it is safe to hold open file handles to hardlinks'''
     return True
 
 
 def nlinks(name):
     '''return number of hardlinks for the given file'''
     return os.lstat(name).st_nlink
 
 
 def parsepatchoutput(output_line):
     """parses the output produced by patch and returns the filename"""
     pf = output_line[14:]
     if pycompat.sysplatform == b'OpenVMS':
         if pf[0] == b'`':
             pf = pf[1:-1]  # Remove the quotes
     else:
         if pf.startswith(b"'") and pf.endswith(b"'") and b" " in pf:
             pf = pf[1:-1]  # Remove the quotes
     return pf
 
 
 def sshargs(sshcmd, host, user, port):
     '''Build argument list for ssh'''
     args = user and (b"%s@%s" % (user, host)) or host
     if b'-' in args[:1]:
         raise error.Abort(
             _(b'illegal ssh hostname or username starting with -: %s') % args
         )
     args = shellquote(args)
     if port:
         args = b'-p %s %s' % (shellquote(port), args)
     return args
 
 
 def isexec(f):
     """check whether a file is executable"""
     return os.lstat(f).st_mode & 0o100 != 0
 
 
 def setflags(f, l, x):
     st = os.lstat(f)
     s = st.st_mode
     if l:
         if not stat.S_ISLNK(s):
             # switch file to link
             with open(f, b'rb') as fp:
                 data = fp.read()
             unlink(f)
             try:
                 os.symlink(data, f)
             except OSError:
                 # failed to make a link, rewrite file
                 with open(f, b"wb") as fp:
                     fp.write(data)
 
         # no chmod needed at this point
         return
     if stat.S_ISLNK(s):
         # switch link to file
         data = os.readlink(f)
         unlink(f)
         with open(f, b"wb") as fp:
             fp.write(data)
         s = 0o666 & ~umask  # avoid restatting for chmod
 
     sx = s & 0o100
     if st.st_nlink > 1 and bool(x) != bool(sx):
         # the file is a hardlink, break it
         with open(f, b"rb") as fp:
             data = fp.read()
         unlink(f)
         with open(f, b"wb") as fp:
             fp.write(data)
 
     if x and not sx:
         # Turn on +x for every +r bit when making a file executable
         # and obey umask.
         os.chmod(f, s | (s & 0o444) >> 2 & ~umask)
     elif not x and sx:
         # Turn off all +x bits
         os.chmod(f, s & 0o666)
 
 
 def copymode(src, dst, mode=None, enforcewritable=False):
     """Copy the file mode from the file at path src to dst.
     If src doesn't exist, we're using mode instead. If mode is None, we're
     using umask."""
     try:
         st_mode = os.lstat(src).st_mode & 0o777
     except FileNotFoundError:
         st_mode = mode
         if st_mode is None:
             st_mode = ~umask
         st_mode &= 0o666
 
     new_mode = st_mode
 
     if enforcewritable:
         new_mode |= stat.S_IWUSR
 
     os.chmod(dst, new_mode)
 
 
 def checkexec(path):
     """
     Check whether the given path is on a filesystem with UNIX-like exec flags
 
     Requires a directory (like /foo/.hg)
     """
 
     # VFAT on some Linux versions can flip mode but it doesn't persist
     # a FS remount. Frequently we can detect it if files are created
     # with exec bit on.
 
     try:
         EXECFLAGS = stat.S_IXUSR | stat.S_IXGRP | stat.S_IXOTH
         basedir = os.path.join(path, b'.hg')
         cachedir = os.path.join(basedir, b'wcache')
         storedir = os.path.join(basedir, b'store')
         if not os.path.exists(cachedir):
             try:
                 # we want to create the 'cache' directory, not the '.hg' one.
                 # Automatically creating '.hg' directory could silently spawn
                 # invalid Mercurial repositories. That seems like a bad idea.
                 os.mkdir(cachedir)
                 if os.path.exists(storedir):
                     copymode(storedir, cachedir)
                 else:
                     copymode(basedir, cachedir)
             except (IOError, OSError):
                 # we other fallback logic triggers
                 pass
         if os.path.isdir(cachedir):
             checkisexec = os.path.join(cachedir, b'checkisexec')
             checknoexec = os.path.join(cachedir, b'checknoexec')
 
             try:
                 m = os.stat(checkisexec).st_mode
             except FileNotFoundError:
                 # checkisexec does not exist - fall through ...
                 pass
             else:
                 # checkisexec exists, check if it actually is exec
                 if m & EXECFLAGS != 0:
                     # ensure checkisexec exists, check it isn't exec
                     try:
                         m = os.stat(checknoexec).st_mode
                     except FileNotFoundError:
                         open(checknoexec, b'w').close()  # might fail
                         m = os.stat(checknoexec).st_mode
                     if m & EXECFLAGS == 0:
                         # check-exec is exec and check-no-exec is not exec
                         return True
                     # checknoexec exists but is exec - delete it
                     unlink(checknoexec)
                 # checkisexec exists but is not exec - delete it
                 unlink(checkisexec)
 
             # check using one file, leave it as checkisexec
             checkdir = cachedir
         else:
             # check directly in path and don't leave checkisexec behind
             checkdir = path
             checkisexec = None
         fh, fn = pycompat.mkstemp(dir=checkdir, prefix=b'hg-checkexec-')
         try:
             os.close(fh)
             m = os.stat(fn).st_mode
             if m & EXECFLAGS == 0:
                 os.chmod(fn, m & 0o777 | EXECFLAGS)
                 if os.stat(fn).st_mode & EXECFLAGS != 0:
                     if checkisexec is not None:
                         os.rename(fn, checkisexec)
                         fn = None
                     return True
         finally:
             if fn is not None:
                 unlink(fn)
     except (IOError, OSError):
         # we don't care, the user probably won't be able to commit anyway
         return False
 
 
 def checklink(path):
     """check whether the given path is on a symlink-capable filesystem"""
     # mktemp is not racy because symlink creation will fail if the
     # file already exists
     while True:
         cachedir = os.path.join(path, b'.hg', b'wcache')
         checklink = os.path.join(cachedir, b'checklink')
         # try fast path, read only
         if os.path.islink(checklink):
             return True
         if os.path.isdir(cachedir):
             checkdir = cachedir
         else:
             checkdir = path
             cachedir = None
         name = tempfile.mktemp(
             dir=pycompat.fsdecode(checkdir), prefix=r'checklink-'
         )
         name = pycompat.fsencode(name)
         try:
             fd = None
             if cachedir is None:
                 fd = pycompat.namedtempfile(
                     dir=checkdir, prefix=b'hg-checklink-'
                 )
                 target = os.path.basename(fd.name)
             else:
                 # create a fixed file to link to; doesn't matter if it
                 # already exists.
                 target = b'checklink-target'
                 try:
                     fullpath = os.path.join(cachedir, target)
                     open(fullpath, b'w').close()
                 except PermissionError:
                     # If we can't write to cachedir, just pretend
                     # that the fs is readonly and by association
                     # that the fs won't support symlinks. This
                     # seems like the least dangerous way to avoid
                     # data loss.
                     return False
             try:
                 os.symlink(target, name)
                 if cachedir is None:
                     unlink(name)
                 else:
                     try:
                         os.rename(name, checklink)
                     except OSError:
                         unlink(name)
                 return True
             except FileExistsError:
                 # link creation might race, try again
                 continue
             finally:
                 if fd is not None:
                     fd.close()
         except AttributeError:
             return False
         except OSError as inst:
             # sshfs might report failure while successfully creating the link
             if inst.errno == errno.EIO and os.path.exists(name):
                 unlink(name)
             return False
 
 
 def checkosfilename(path):
     """Check that the base-relative path is a valid filename on this platform.
     Returns None if the path is ok, or a UI string describing the problem."""
     return None  # on posix platforms, every path is ok
 
 
 def getfsmountpoint(dirpath):
     """Get the filesystem mount point from a directory (best-effort)
 
     Returns None if we are unsure. Raises OSError on ENOENT, EPERM, etc.
     """
     return getattr(osutil, 'getfsmountpoint', lambda x: None)(dirpath)
 
 
 def getfstype(dirpath):
     """Get the filesystem type name from a directory (best-effort)
 
     Returns None if we are unsure. Raises OSError on ENOENT, EPERM, etc.
     """
     return getattr(osutil, 'getfstype', lambda x: None)(dirpath)
 
 
 def get_password():
     return encoding.strtolocal(getpass.getpass(''))
 
 
 def setbinary(fd):
     pass
 
 
 def pconvert(path):
     return path
 
 
 def localpath(path):
     return path
 
 
 def samefile(fpath1, fpath2):
     """Returns whether path1 and path2 refer to the same file. This is only
     guaranteed to work for files, not directories."""
     return os.path.samefile(fpath1, fpath2)
 
 
 def samedevice(fpath1, fpath2):
     """Returns whether fpath1 and fpath2 are on the same device. This is only
     guaranteed to work for files, not directories."""
     st1 = os.lstat(fpath1)
     st2 = os.lstat(fpath2)
     return st1.st_dev == st2.st_dev
 
 
 # os.path.normcase is a no-op, which doesn't help us on non-native filesystems
 def normcase(path):
     return path.lower()
 
 
 # what normcase does to ASCII strings
 normcasespec = encoding.normcasespecs.lower
 # fallback normcase function for non-ASCII strings
 normcasefallback = normcase
 
 if pycompat.isdarwin:
 
     def normcase(path):
         """
         Normalize a filename for OS X-compatible comparison:
         - escape-encode invalid characters
         - decompose to NFD
         - lowercase
         - omit ignored characters [200c-200f, 202a-202e, 206a-206f,feff]
 
         >>> normcase(b'UPPER')
         'upper'
         >>> normcase(b'Caf\\xc3\\xa9')
         'cafe\\xcc\\x81'
         >>> normcase(b'\\xc3\\x89')
         'e\\xcc\\x81'
         >>> normcase(b'\\xb8\\xca\\xc3\\xca\\xbe\\xc8.JPG') # issue3918
         '%b8%ca%c3\\xca\\xbe%c8.jpg'
         """
 
         try:
             return encoding.asciilower(path)  # exception for non-ASCII
         except UnicodeDecodeError:
             return normcasefallback(path)
 
     normcasespec = encoding.normcasespecs.lower
 
     def normcasefallback(path):
         try:
             u = path.decode('utf-8')
         except UnicodeDecodeError:
             # OS X percent-encodes any bytes that aren't valid utf-8
             s = b''
             pos = 0
             l = len(path)
             while pos < l:
                 try:
                     c = encoding.getutf8char(path, pos)
                     pos += len(c)
                 except ValueError:
                     c = b'%%%02X' % ord(path[pos : pos + 1])
                     pos += 1
                 s += c
 
             u = s.decode('utf-8')
 
         # Decompose then lowercase (HFS+ technote specifies lower)
         enc = unicodedata.normalize('NFD', u).lower().encode('utf-8')
         # drop HFS+ ignored characters
         return encoding.hfsignoreclean(enc)
 
 
 if pycompat.sysplatform == b'cygwin':
     # workaround for cygwin, in which mount point part of path is
     # treated as case sensitive, even though underlying NTFS is case
     # insensitive.
 
     # default mount points
     cygwinmountpoints = sorted(
         [
             b"/usr/bin",
             b"/usr/lib",
             b"/cygdrive",
         ],
         reverse=True,
     )
 
     # use upper-ing as normcase as same as NTFS workaround
     def normcase(path):
         pathlen = len(path)
         if (pathlen == 0) or (path[0] != pycompat.ossep):
             # treat as relative
             return encoding.upper(path)
 
         # to preserve case of mountpoint part
         for mp in cygwinmountpoints:
             if not path.startswith(mp):
                 continue
 
             mplen = len(mp)
             if mplen == pathlen:  # mount point itself
                 return mp
             if path[mplen] == pycompat.ossep:
                 return mp + encoding.upper(path[mplen:])
 
         return encoding.upper(path)
 
     normcasespec = encoding.normcasespecs.other
     normcasefallback = normcase
 
     # Cygwin translates native ACLs to POSIX permissions,
     # but these translations are not supported by native
     # tools, so the exec bit tends to be set erroneously.
     # Therefore, disable executable bit access on Cygwin.
     def checkexec(path):
         return False
 
     # Similarly, Cygwin's symlink emulation is likely to create
     # problems when Mercurial is used from both Cygwin and native
     # Windows, with other native tools, or on shared volumes
     def checklink(path):
         return False
 
 
 _needsshellquote = None
 
 
 def shellquote(s):
     if pycompat.sysplatform == b'OpenVMS':
         return b'"%s"' % s
     global _needsshellquote
     if _needsshellquote is None:
         _needsshellquote = re.compile(br'[^a-zA-Z0-9._/+-]').search
     if s and not _needsshellquote(s):
         # "s" shouldn't have to be quoted
         return s
     else:
         return b"'%s'" % s.replace(b"'", b"'\\''")
 
 
 def shellsplit(s):
     """Parse a command string in POSIX shell way (best-effort)"""
     return pycompat.shlexsplit(s, posix=True)
 
 
 def testpid(pid):
     '''return False if pid dead, True if running or not sure'''
     if pycompat.sysplatform == b'OpenVMS':
         return True
     try:
         os.kill(pid, 0)
         return True
     except OSError as inst:
         return inst.errno != errno.ESRCH
 
 
 def isowner(st):
     """Return True if the stat object st is from the current user."""
     return st.st_uid == os.getuid()
 
 
 def findexe(command):
     """Find executable for command searching like which does.
     If command is a basename then PATH is searched for command.
     PATH isn't searched if command is an absolute or relative path.
     If command isn't found None is returned."""
     if pycompat.sysplatform == b'OpenVMS':
         return command
 
     def findexisting(executable):
         b'Will return executable if existing file'
         if os.path.isfile(executable) and os.access(executable, os.X_OK):
             return executable
         return None
 
     if pycompat.ossep in command:
         return findexisting(command)
 
     if pycompat.sysplatform == b'plan9':
         return findexisting(os.path.join(b'/bin', command))
 
     for path in encoding.environ.get(b'PATH', b'').split(pycompat.ospathsep):
         executable = findexisting(os.path.join(path, command))
         if executable is not None:
             return executable
     return None
 
 
 def setsignalhandler():
     pass
 
 
 _wantedkinds = {stat.S_IFREG, stat.S_IFLNK}
 
 
 def statfiles(files):
     """Stat each file in files. Yield each stat, or None if a file does not
     exist or has a type we don't care about."""
     lstat = os.lstat
     getkind = stat.S_IFMT
     for nf in files:
         try:
             st = lstat(nf)
             if getkind(st.st_mode) not in _wantedkinds:
                 st = None
-        except OSError as err:
-            if err.errno not in (errno.ENOENT, errno.ENOTDIR):
-                raise
+        except (FileNotFoundError, NotADirectoryError):
             st = None
         yield st
 
 
 def getuser():
     '''return name of current user'''
     return pycompat.fsencode(getpass.getuser())
 
 
 def username(uid=None):
     """Return the name of the user with the given uid.
 
     If uid is None, return the name of the current user."""
 
     if uid is None:
         uid = os.getuid()
     try:
         return pycompat.fsencode(pwd.getpwuid(uid)[0])
     except KeyError:
         return b'%d' % uid
 
 
 def groupname(gid=None):
     """Return the name of the group with the given gid.
 
     If gid is None, return the name of the current group."""
 
     if gid is None:
         gid = os.getgid()
     try:
         return pycompat.fsencode(grp.getgrgid(gid)[0])
     except KeyError:
         return pycompat.bytestr(gid)
 
 
 def groupmembers(name):
     """Return the list of members of the group with the given
     name, KeyError if the group does not exist.
     """
     name = pycompat.fsdecode(name)
     return pycompat.rapply(pycompat.fsencode, list(grp.getgrnam(name).gr_mem))
 
 
 def spawndetached(args):
     return os.spawnvp(os.P_NOWAIT | getattr(os, 'P_DETACH', 0), args[0], args)
 
 
 def gethgcmd():
     return sys.argv[:1]
 
 
 def makedir(path, notindexed):
     os.mkdir(path)
 
 
 def lookupreg(key, name=None, scope=None):
     return None
 
 
 def hidewindow():
     """Hide current shell window.
 
     Used to hide the window opened when starting asynchronous
     child process under Windows, unneeded on other systems.
     """
     pass
 
 
 class cachestat:
     def __init__(self, path):
         self.stat = os.stat(path)
 
     def cacheable(self):
         return bool(self.stat.st_ino)
 
     __hash__ = object.__hash__
 
     def __eq__(self, other):
         try:
             # Only dev, ino, size, mtime and atime are likely to change. Out
             # of these, we shouldn't compare atime but should compare the
             # rest. However, one of the other fields changing indicates
             # something fishy going on, so return False if anything but atime
             # changes.
             return (
                 self.stat.st_mode == other.stat.st_mode
                 and self.stat.st_ino == other.stat.st_ino
                 and self.stat.st_dev == other.stat.st_dev
                 and self.stat.st_nlink == other.stat.st_nlink
                 and self.stat.st_uid == other.stat.st_uid
                 and self.stat.st_gid == other.stat.st_gid
                 and self.stat.st_size == other.stat.st_size
                 and self.stat[stat.ST_MTIME] == other.stat[stat.ST_MTIME]
                 and self.stat[stat.ST_CTIME] == other.stat[stat.ST_CTIME]
             )
         except AttributeError:
             return False
 
     def __ne__(self, other):
         return not self == other
 
 
 def statislink(st):
     '''check whether a stat result is a symlink'''
     return st and stat.S_ISLNK(st.st_mode)
 
 
 def statisexec(st):
     '''check whether a stat result is an executable file'''
     return st and (st.st_mode & 0o100 != 0)
 
 
 def poll(fds):
     """block until something happens on any file descriptor
 
     This is a generic helper that will check for any activity
     (read, write.  exception) and return the list of touched files.
 
     In unsupported cases, it will raise a NotImplementedError"""
     try:
         res = select.select(fds, fds, fds)
     except ValueError:  # out of range file descriptor
         raise NotImplementedError()
     return sorted(list(set(sum(res, []))))
 
 
 def readpipe(pipe):
     """Read all available data from a pipe."""
     # We can't fstat() a pipe because Linux will always report 0.
     # So, we set the pipe to non-blocking mode and read everything
     # that's available.
     flags = fcntl.fcntl(pipe, fcntl.F_GETFL)
     flags |= os.O_NONBLOCK
     oldflags = fcntl.fcntl(pipe, fcntl.F_SETFL, flags)
 
     try:
         chunks = []
         while True:
             try:
                 s = pipe.read()
                 if not s:
                     break
                 chunks.append(s)
             except IOError:
                 break
 
         return b''.join(chunks)
     finally:
         fcntl.fcntl(pipe, fcntl.F_SETFL, oldflags)
 
 
 def bindunixsocket(sock, path):
     """Bind the UNIX domain socket to the specified path"""
     # use relative path instead of full path at bind() if possible, since
     # AF_UNIX path has very small length limit (107 chars) on common
     # platforms (see sys/un.h)
     dirname, basename = os.path.split(path)
     bakwdfd = None
 
     try:
         if dirname:
             bakwdfd = os.open(b'.', os.O_DIRECTORY)
             os.chdir(dirname)
         sock.bind(basename)
         if bakwdfd:
             os.fchdir(bakwdfd)
     finally:
         if bakwdfd:
             os.close(bakwdfd)

setup.py

@@ -1,1781 +1,1777 b''
 #
 # This is the mercurial setup script.
 #
 # 'python setup.py install', or
 # 'python setup.py --help' for more options
 import os
 
 # Mercurial can't work on 3.6.0 or 3.6.1 due to a bug in % formatting
 # in bytestrings.
 supportedpy = ','.join(
     [
         '>=3.6.2',
     ]
 )
 
 import sys, platform
 import sysconfig
 
 
 def sysstr(s):
     return s.decode('latin-1')
 
 
 import ssl
 
 # ssl.HAS_TLSv1* are preferred to check support but they were added in Python
 # 3.7. Prior to CPython commit 6e8cda91d92da72800d891b2fc2073ecbc134d98
 # (backported to the 3.7 branch), ssl.PROTOCOL_TLSv1_1 / ssl.PROTOCOL_TLSv1_2
 # were defined only if compiled against a OpenSSL version with TLS 1.1 / 1.2
 # support. At the mentioned commit, they were unconditionally defined.
 _notset = object()
 has_tlsv1_1 = getattr(ssl, 'HAS_TLSv1_1', _notset)
 if has_tlsv1_1 is _notset:
     has_tlsv1_1 = getattr(ssl, 'PROTOCOL_TLSv1_1', _notset) is not _notset
 has_tlsv1_2 = getattr(ssl, 'HAS_TLSv1_2', _notset)
 if has_tlsv1_2 is _notset:
     has_tlsv1_2 = getattr(ssl, 'PROTOCOL_TLSv1_2', _notset) is not _notset
 if not (has_tlsv1_1 or has_tlsv1_2):
     error = """
 The `ssl` module does not advertise support for TLS 1.1 or TLS 1.2.
 Please make sure that your Python installation was compiled against an OpenSSL
 version enabling these features (likely this requires the OpenSSL version to
 be at least 1.0.1).
 """
     print(error, file=sys.stderr)
     sys.exit(1)
 
 DYLIB_SUFFIX = sysconfig.get_config_vars()['EXT_SUFFIX']
 
 # Solaris Python packaging brain damage
 try:
     import hashlib
 
     sha = hashlib.sha1()
 except ImportError:
     try:
         import sha
 
         sha.sha  # silence unused import warning
     except ImportError:
         raise SystemExit(
             "Couldn't import standard hashlib (incomplete Python install)."
         )
 
 try:
     import zlib
 
     zlib.compressobj  # silence unused import warning
 except ImportError:
     raise SystemExit(
         "Couldn't import standard zlib (incomplete Python install)."
     )
 
 # The base IronPython distribution (as of 2.7.1) doesn't support bz2
 isironpython = False
 try:
     isironpython = (
         platform.python_implementation().lower().find("ironpython") != -1
     )
 except AttributeError:
     pass
 
 if isironpython:
     sys.stderr.write("warning: IronPython detected (no bz2 support)\n")
 else:
     try:
         import bz2
 
         bz2.BZ2Compressor  # silence unused import warning
     except ImportError:
         raise SystemExit(
             "Couldn't import standard bz2 (incomplete Python install)."
         )
 
 ispypy = "PyPy" in sys.version
 
 import ctypes
-import errno
 import stat, subprocess, time
 import re
 import shutil
 import tempfile
 
 # We have issues with setuptools on some platforms and builders. Until
 # those are resolved, setuptools is opt-in except for platforms where
 # we don't have issues.
 issetuptools = os.name == 'nt' or 'FORCE_SETUPTOOLS' in os.environ
 if issetuptools:
     from setuptools import setup
 else:
     from distutils.core import setup
 from distutils.ccompiler import new_compiler
 from distutils.core import Command, Extension
 from distutils.dist import Distribution
 from distutils.command.build import build
 from distutils.command.build_ext import build_ext
 from distutils.command.build_py import build_py
 from distutils.command.build_scripts import build_scripts
 from distutils.command.install import install
 from distutils.command.install_lib import install_lib
 from distutils.command.install_scripts import install_scripts
 from distutils import log
 from distutils.spawn import spawn, find_executable
 from distutils import file_util
 from distutils.errors import (
     CCompilerError,
     DistutilsError,
     DistutilsExecError,
 )
 from distutils.sysconfig import get_python_inc, get_config_var
 from distutils.version import StrictVersion
 
 # Explain to distutils.StrictVersion how our release candidates are versioned
 StrictVersion.version_re = re.compile(r'^(\d+)\.(\d+)(\.(\d+))?-?(rc(\d+))?$')
 
 
 def write_if_changed(path, content):
     """Write content to a file iff the content hasn't changed."""
     if os.path.exists(path):
         with open(path, 'rb') as fh:
             current = fh.read()
     else:
         current = b''
 
     if current != content:
         with open(path, 'wb') as fh:
             fh.write(content)
 
 
 scripts = ['hg']
 if os.name == 'nt':
     # We remove hg.bat if we are able to build hg.exe.
     scripts.append('contrib/win32/hg.bat')
 
 
 def cancompile(cc, code):
     tmpdir = tempfile.mkdtemp(prefix='hg-install-')
     devnull = oldstderr = None
     try:
         fname = os.path.join(tmpdir, 'testcomp.c')
         f = open(fname, 'w')
         f.write(code)
         f.close()
         # Redirect stderr to /dev/null to hide any error messages
         # from the compiler.
         # This will have to be changed if we ever have to check
         # for a function on Windows.
         devnull = open('/dev/null', 'w')
         oldstderr = os.dup(sys.stderr.fileno())
         os.dup2(devnull.fileno(), sys.stderr.fileno())
         objects = cc.compile([fname], output_dir=tmpdir)
         cc.link_executable(objects, os.path.join(tmpdir, "a.out"))
         return True
     except Exception:
         return False
     finally:
         if oldstderr is not None:
             os.dup2(oldstderr, sys.stderr.fileno())
         if devnull is not None:
             devnull.close()
         shutil.rmtree(tmpdir)
 
 
 # simplified version of distutils.ccompiler.CCompiler.has_function
 # that actually removes its temporary files.
 def hasfunction(cc, funcname):
     code = 'int main(void) { %s(); }\n' % funcname
     return cancompile(cc, code)
 
 
 def hasheader(cc, headername):
     code = '#include <%s>\nint main(void) { return 0; }\n' % headername
     return cancompile(cc, code)
 
 
 # py2exe needs to be installed to work
 try:
     import py2exe
 
     py2exe.patch_distutils()
     py2exeloaded = True
     # import py2exe's patched Distribution class
     from distutils.core import Distribution
 except ImportError:
     py2exeloaded = False
 
 
 def runcmd(cmd, env, cwd=None):
     p = subprocess.Popen(
         cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, env=env, cwd=cwd
     )
     out, err = p.communicate()
     return p.returncode, out, err
 
 
 class hgcommand:
     def __init__(self, cmd, env):
         self.cmd = cmd
         self.env = env
 
     def run(self, args):
         cmd = self.cmd + args
         returncode, out, err = runcmd(cmd, self.env)
         err = filterhgerr(err)
         if err or returncode != 0:
             print("stderr from '%s':" % (' '.join(cmd)), file=sys.stderr)
             print(err, file=sys.stderr)
             return b''
         return out
 
 
 def filterhgerr(err):
     # If root is executing setup.py, but the repository is owned by
     # another user (as in "sudo python setup.py install") we will get
     # trust warnings since the .hg/hgrc file is untrusted. That is
     # fine, we don't want to load it anyway.  Python may warn about
     # a missing __init__.py in mercurial/locale, we also ignore that.
     err = [
         e
         for e in err.splitlines()
         if (
             not e.startswith(b'not trusting file')
             and not e.startswith(b'warning: Not importing')
             and not e.startswith(b'obsolete feature not enabled')
             and not e.startswith(b'*** failed to import extension')
             and not e.startswith(b'devel-warn:')
             and not (
                 e.startswith(b'(third party extension')
                 and e.endswith(b'or newer of Mercurial; disabling)')
             )
         )
     ]
     return b'\n'.join(b'  ' + e for e in err)
 
 
 def findhg():
     """Try to figure out how we should invoke hg for examining the local
     repository contents.
 
     Returns an hgcommand object."""
     # By default, prefer the "hg" command in the user's path.  This was
     # presumably the hg command that the user used to create this repository.
     #
     # This repository may require extensions or other settings that would not
     # be enabled by running the hg script directly from this local repository.
     hgenv = os.environ.copy()
     # Use HGPLAIN to disable hgrc settings that would change output formatting,
     # and disable localization for the same reasons.
     hgenv['HGPLAIN'] = '1'
     hgenv['LANGUAGE'] = 'C'
     hgcmd = ['hg']
     # Run a simple "hg log" command just to see if using hg from the user's
     # path works and can successfully interact with this repository.  Windows
     # gives precedence to hg.exe in the current directory, so fall back to the
     # python invocation of local hg, where pythonXY.dll can always be found.
     check_cmd = ['log', '-r.', '-Ttest']
     if os.name != 'nt' or not os.path.exists("hg.exe"):
         try:
             retcode, out, err = runcmd(hgcmd + check_cmd, hgenv)
         except EnvironmentError:
             retcode = -1
         if retcode == 0 and not filterhgerr(err):
             return hgcommand(hgcmd, hgenv)
 
     # Fall back to trying the local hg installation.
     hgenv = localhgenv()
     hgcmd = [sys.executable, 'hg']
     try:
         retcode, out, err = runcmd(hgcmd + check_cmd, hgenv)
     except EnvironmentError:
         retcode = -1
     if retcode == 0 and not filterhgerr(err):
         return hgcommand(hgcmd, hgenv)
 
     raise SystemExit(
         'Unable to find a working hg binary to extract the '
         'version from the repository tags'
     )
 
 
 def localhgenv():
     """Get an environment dictionary to use for invoking or importing
     mercurial from the local repository."""
     # Execute hg out of this directory with a custom environment which takes
     # care to not use any hgrc files and do no localization.
     env = {
         'HGMODULEPOLICY': 'py',
         'HGRCPATH': '',
         'LANGUAGE': 'C',
         'PATH': '',
     }  # make pypi modules that use os.environ['PATH'] happy
     if 'LD_LIBRARY_PATH' in os.environ:
         env['LD_LIBRARY_PATH'] = os.environ['LD_LIBRARY_PATH']
     if 'SystemRoot' in os.environ:
         # SystemRoot is required by Windows to load various DLLs.  See:
         # https://bugs.python.org/issue13524#msg148850
         env['SystemRoot'] = os.environ['SystemRoot']
     return env
 
 
 version = ''
 
 if os.path.isdir('.hg'):
     hg = findhg()
     cmd = ['log', '-r', '.', '--template', '{tags}\n']
     numerictags = [t for t in sysstr(hg.run(cmd)).split() if t[0:1].isdigit()]
     hgid = sysstr(hg.run(['id', '-i'])).strip()
     if not hgid:
         # Bail out if hg is having problems interacting with this repository,
         # rather than falling through and producing a bogus version number.
         # Continuing with an invalid version number will break extensions
         # that define minimumhgversion.
         raise SystemExit('Unable to determine hg version from local repository')
     if numerictags:  # tag(s) found
         version = numerictags[-1]
         if hgid.endswith('+'):  # propagate the dirty status to the tag
             version += '+'
     else:  # no tag found
         ltagcmd = ['parents', '--template', '{latesttag}']
         ltag = sysstr(hg.run(ltagcmd))
         changessincecmd = ['log', '-T', 'x\n', '-r', "only(.,'%s')" % ltag]
         changessince = len(hg.run(changessincecmd).splitlines())
         version = '%s+hg%s.%s' % (ltag, changessince, hgid)
     if version.endswith('+'):
         version = version[:-1] + 'local' + time.strftime('%Y%m%d')
 elif os.path.exists('.hg_archival.txt'):
     kw = dict(
         [[t.strip() for t in l.split(':', 1)] for l in open('.hg_archival.txt')]
     )
     if 'tag' in kw:
         version = kw['tag']
     elif 'latesttag' in kw:
         if 'changessincelatesttag' in kw:
             version = (
                 '%(latesttag)s+hg%(changessincelatesttag)s.%(node).12s' % kw
             )
         else:
             version = '%(latesttag)s+hg%(latesttagdistance)s.%(node).12s' % kw
     else:
         version = '0+hg' + kw.get('node', '')[:12]
 elif os.path.exists('mercurial/__version__.py'):
     with open('mercurial/__version__.py') as f:
         data = f.read()
     version = re.search('version = b"(.*)"', data).group(1)
 
 if version:
     versionb = version
     if not isinstance(versionb, bytes):
         versionb = versionb.encode('ascii')
 
     write_if_changed(
         'mercurial/__version__.py',
         b''.join(
             [
                 b'# this file is autogenerated by setup.py\n'
                 b'version = b"%s"\n' % versionb,
             ]
         ),
     )
 
 
 class hgbuild(build):
     # Insert hgbuildmo first so that files in mercurial/locale/ are found
     # when build_py is run next.
     sub_commands = [('build_mo', None)] + build.sub_commands
 
 
 class hgbuildmo(build):
 
     description = "build translations (.mo files)"
 
     def run(self):
         if not find_executable('msgfmt'):
             self.warn(
                 "could not find msgfmt executable, no translations "
                 "will be built"
             )
             return
 
         podir = 'i18n'
         if not os.path.isdir(podir):
             self.warn("could not find %s/ directory" % podir)
             return
 
         join = os.path.join
         for po in os.listdir(podir):
             if not po.endswith('.po'):
                 continue
             pofile = join(podir, po)
             modir = join('locale', po[:-3], 'LC_MESSAGES')
             mofile = join(modir, 'hg.mo')
             mobuildfile = join('mercurial', mofile)
             cmd = ['msgfmt', '-v', '-o', mobuildfile, pofile]
             if sys.platform != 'sunos5':
                 # msgfmt on Solaris does not know about -c
                 cmd.append('-c')
             self.mkpath(join('mercurial', modir))
             self.make_file([pofile], mobuildfile, spawn, (cmd,))
 
 
 class hgdist(Distribution):
     pure = False
     rust = False
     no_rust = False
     cffi = ispypy
 
     global_options = Distribution.global_options + [
         ('pure', None, "use pure (slow) Python code instead of C extensions"),
         ('rust', None, "use Rust extensions additionally to C extensions"),
         (
             'no-rust',
             None,
             "do not use Rust extensions additionally to C extensions",
         ),
     ]
 
     negative_opt = Distribution.negative_opt.copy()
     boolean_options = ['pure', 'rust', 'no-rust']
     negative_opt['no-rust'] = 'rust'
 
     def _set_command_options(self, command_obj, option_dict=None):
         # Not all distutils versions in the wild have boolean_options.
         # This should be cleaned up when we're Python 3 only.
         command_obj.boolean_options = (
             getattr(command_obj, 'boolean_options', []) + self.boolean_options
         )
         return Distribution._set_command_options(
             self, command_obj, option_dict=option_dict
         )
 
     def parse_command_line(self):
         ret = Distribution.parse_command_line(self)
         if not (self.rust or self.no_rust):
             hgrustext = os.environ.get('HGWITHRUSTEXT')
             # TODO record it for proper rebuild upon changes
             # (see mercurial/__modulepolicy__.py)
             if hgrustext != 'cpython' and hgrustext is not None:
                 if hgrustext:
                     msg = 'unknown HGWITHRUSTEXT value: %s' % hgrustext
                     print(msg, file=sys.stderr)
                 hgrustext = None
             self.rust = hgrustext is not None
             self.no_rust = not self.rust
         return ret
 
     def has_ext_modules(self):
         # self.ext_modules is emptied in hgbuildpy.finalize_options which is
         # too late for some cases
         return not self.pure and Distribution.has_ext_modules(self)
 
 
 # This is ugly as a one-liner. So use a variable.
 buildextnegops = dict(getattr(build_ext, 'negative_options', {}))
 buildextnegops['no-zstd'] = 'zstd'
 buildextnegops['no-rust'] = 'rust'
 
 
 class hgbuildext(build_ext):
     user_options = build_ext.user_options + [
         ('zstd', None, 'compile zstd bindings [default]'),
         ('no-zstd', None, 'do not compile zstd bindings'),
         (
             'rust',
             None,
             'compile Rust extensions if they are in use '
             '(requires Cargo) [default]',
         ),
         ('no-rust', None, 'do not compile Rust extensions'),
     ]
 
     boolean_options = build_ext.boolean_options + ['zstd', 'rust']
     negative_opt = buildextnegops
 
     def initialize_options(self):
         self.zstd = True
         self.rust = True
 
         return build_ext.initialize_options(self)
 
     def finalize_options(self):
         # Unless overridden by the end user, build extensions in parallel.
         # Only influences behavior on Python 3.5+.
         if getattr(self, 'parallel', None) is None:
             self.parallel = True
 
         return build_ext.finalize_options(self)
 
     def build_extensions(self):
         ruststandalones = [
             e for e in self.extensions if isinstance(e, RustStandaloneExtension)
         ]
         self.extensions = [
             e for e in self.extensions if e not in ruststandalones
         ]
         # Filter out zstd if disabled via argument.
         if not self.zstd:
             self.extensions = [
                 e for e in self.extensions if e.name != 'mercurial.zstd'
             ]
 
         # Build Rust standalone extensions if it'll be used
         # and its build is not explicitly disabled (for external build
         # as Linux distributions would do)
         if self.distribution.rust and self.rust:
             if not sys.platform.startswith('linux'):
                 self.warn(
                     "rust extensions have only been tested on Linux "
                     "and may not behave correctly on other platforms"
                 )
 
             for rustext in ruststandalones:
                 rustext.build('' if self.inplace else self.build_lib)
 
         return build_ext.build_extensions(self)
 
     def build_extension(self, ext):
         if (
             self.distribution.rust
             and self.rust
             and isinstance(ext, RustExtension)
         ):
             ext.rustbuild()
         try:
             build_ext.build_extension(self, ext)
         except CCompilerError:
             if not getattr(ext, 'optional', False):
                 raise
             log.warn(
                 "Failed to build optional extension '%s' (skipping)", ext.name
             )
 
 
 class hgbuildscripts(build_scripts):
     def run(self):
         if os.name != 'nt' or self.distribution.pure:
             return build_scripts.run(self)
 
         exebuilt = False
         try:
             self.run_command('build_hgexe')
             exebuilt = True
         except (DistutilsError, CCompilerError):
             log.warn('failed to build optional hg.exe')
 
         if exebuilt:
             # Copying hg.exe to the scripts build directory ensures it is
             # installed by the install_scripts command.
             hgexecommand = self.get_finalized_command('build_hgexe')
             dest = os.path.join(self.build_dir, 'hg.exe')
             self.mkpath(self.build_dir)
             self.copy_file(hgexecommand.hgexepath, dest)
 
             # Remove hg.bat because it is redundant with hg.exe.
             self.scripts.remove('contrib/win32/hg.bat')
 
         return build_scripts.run(self)
 
 
 class hgbuildpy(build_py):
     def finalize_options(self):
         build_py.finalize_options(self)
 
         if self.distribution.pure:
             self.distribution.ext_modules = []
         elif self.distribution.cffi:
             from mercurial.cffi import (
                 bdiffbuild,
                 mpatchbuild,
             )
 
             exts = [
                 mpatchbuild.ffi.distutils_extension(),
                 bdiffbuild.ffi.distutils_extension(),
             ]
             # cffi modules go here
             if sys.platform == 'darwin':
                 from mercurial.cffi import osutilbuild
 
                 exts.append(osutilbuild.ffi.distutils_extension())
             self.distribution.ext_modules = exts
         else:
             h = os.path.join(get_python_inc(), 'Python.h')
             if not os.path.exists(h):
                 raise SystemExit(
                     'Python headers are required to build '
                     'Mercurial but weren\'t found in %s' % h
                 )
 
     def run(self):
         basepath = os.path.join(self.build_lib, 'mercurial')
         self.mkpath(basepath)
 
         rust = self.distribution.rust
         if self.distribution.pure:
             modulepolicy = 'py'
         elif self.build_lib == '.':
             # in-place build should run without rebuilding and Rust extensions
             modulepolicy = 'rust+c-allow' if rust else 'allow'
         else:
             modulepolicy = 'rust+c' if rust else 'c'
 
         content = b''.join(
             [
                 b'# this file is autogenerated by setup.py\n',
                 b'modulepolicy = b"%s"\n' % modulepolicy.encode('ascii'),
             ]
         )
         write_if_changed(os.path.join(basepath, '__modulepolicy__.py'), content)
 
         build_py.run(self)
 
 
 class buildhgextindex(Command):
     description = 'generate prebuilt index of hgext (for frozen package)'
     user_options = []
     _indexfilename = 'hgext/__index__.py'
 
     def initialize_options(self):
         pass
 
     def finalize_options(self):
         pass
 
     def run(self):
         if os.path.exists(self._indexfilename):
             with open(self._indexfilename, 'w') as f:
                 f.write('# empty\n')
 
         # here no extension enabled, disabled() lists up everything
         code = (
             'import pprint; from mercurial import extensions; '
             'ext = extensions.disabled();'
             'ext.pop("__index__", None);'
             'pprint.pprint(ext)'
         )
         returncode, out, err = runcmd(
             [sys.executable, '-c', code], localhgenv()
         )
         if err or returncode != 0:
             raise DistutilsExecError(err)
 
         with open(self._indexfilename, 'wb') as f:
             f.write(b'# this file is autogenerated by setup.py\n')
             f.write(b'docs = ')
             f.write(out)
 
 
 class buildhgexe(build_ext):
     description = 'compile hg.exe from mercurial/exewrapper.c'
     user_options = build_ext.user_options + [
         (
             'long-paths-support',
             None,
             'enable support for long paths on '
             'Windows (off by default and '
             'experimental)',
         ),
     ]
 
     LONG_PATHS_MANIFEST = """
     <?xml version="1.0" encoding="UTF-8" standalone="yes"?>
     <assembly xmlns="urn:schemas-microsoft-com:asm.v1" manifestVersion="1.0">
         <application>
             <windowsSettings
             xmlns:ws2="http://schemas.microsoft.com/SMI/2016/WindowsSettings">
                 <ws2:longPathAware>true</ws2:longPathAware>
             </windowsSettings>
         </application>
     </assembly>"""
 
     def initialize_options(self):
         build_ext.initialize_options(self)
         self.long_paths_support = False
 
     def build_extensions(self):
         if os.name != 'nt':
             return
         if isinstance(self.compiler, HackedMingw32CCompiler):
             self.compiler.compiler_so = self.compiler.compiler  # no -mdll
             self.compiler.dll_libraries = []  # no -lmsrvc90
 
         pythonlib = None
 
         dir = os.path.dirname(self.get_ext_fullpath('dummy'))
         self.hgtarget = os.path.join(dir, 'hg')
 
         if getattr(sys, 'dllhandle', None):
             # Different Python installs can have different Python library
             # names. e.g. the official CPython distribution uses pythonXY.dll
             # and MinGW uses libpythonX.Y.dll.
             _kernel32 = ctypes.windll.kernel32
             _kernel32.GetModuleFileNameA.argtypes = [
                 ctypes.c_void_p,
                 ctypes.c_void_p,
                 ctypes.c_ulong,
             ]
             _kernel32.GetModuleFileNameA.restype = ctypes.c_ulong
             size = 1000
             buf = ctypes.create_string_buffer(size + 1)
             filelen = _kernel32.GetModuleFileNameA(
                 sys.dllhandle, ctypes.byref(buf), size
             )
 
             if filelen > 0 and filelen != size:
                 dllbasename = os.path.basename(buf.value)
                 if not dllbasename.lower().endswith(b'.dll'):
                     raise SystemExit(
                         'Python DLL does not end with .dll: %s' % dllbasename
                     )
                 pythonlib = dllbasename[:-4]
 
                 # Copy the pythonXY.dll next to the binary so that it runs
                 # without tampering with PATH.
                 dest = os.path.join(
                     os.path.dirname(self.hgtarget),
                     os.fsdecode(dllbasename),
                 )
 
                 if not os.path.exists(dest):
                     shutil.copy(buf.value, dest)
 
                 # Also overwrite python3.dll so that hgext.git is usable.
                 # TODO: also handle the MSYS flavor
                 python_x = os.path.join(
                     os.path.dirname(os.fsdecode(buf.value)),
                     "python3.dll",
                 )
 
                 if os.path.exists(python_x):
                     dest = os.path.join(
                         os.path.dirname(self.hgtarget),
                         os.path.basename(python_x),
                     )
 
                     shutil.copy(python_x, dest)
 
         if not pythonlib:
             log.warn(
                 'could not determine Python DLL filename; assuming pythonXY'
             )
 
             hv = sys.hexversion
             pythonlib = b'python%d%d' % (hv >> 24, (hv >> 16) & 0xFF)
 
         log.info('using %s as Python library name' % pythonlib)
         with open('mercurial/hgpythonlib.h', 'wb') as f:
             f.write(b'/* this file is autogenerated by setup.py */\n')
             f.write(b'#define HGPYTHONLIB "%s"\n' % pythonlib)
 
         objects = self.compiler.compile(
             ['mercurial/exewrapper.c'],
             output_dir=self.build_temp,
             macros=[('_UNICODE', None), ('UNICODE', None)],
         )
         self.compiler.link_executable(
             objects, self.hgtarget, libraries=[], output_dir=self.build_temp
         )
         if self.long_paths_support:
             self.addlongpathsmanifest()
 
     def addlongpathsmanifest(self):
         r"""Add manifest pieces so that hg.exe understands long paths
 
         This is an EXPERIMENTAL feature, use with care.
         To enable long paths support, one needs to do two things:
         - build Mercurial with --long-paths-support option
         - change HKLM\SYSTEM\CurrentControlSet\Control\FileSystem\
                  LongPathsEnabled to have value 1.
 
         Please ignore 'warning 81010002: Unrecognized Element "longPathAware"';
         it happens because Mercurial uses mt.exe circa 2008, which is not
         yet aware of long paths support in the manifest (I think so at least).
         This does not stop mt.exe from embedding/merging the XML properly.
 
         Why resource #1 should be used for .exe manifests? I don't know and
         wasn't able to find an explanation for mortals. But it seems to work.
         """
         exefname = self.compiler.executable_filename(self.hgtarget)
         fdauto, manfname = tempfile.mkstemp(suffix='.hg.exe.manifest')
         os.close(fdauto)
         with open(manfname, 'w') as f:
             f.write(self.LONG_PATHS_MANIFEST)
         log.info("long paths manifest is written to '%s'" % manfname)
         inputresource = '-inputresource:%s;#1' % exefname
         outputresource = '-outputresource:%s;#1' % exefname
         log.info("running mt.exe to update hg.exe's manifest in-place")
         # supplying both -manifest and -inputresource to mt.exe makes
         # it merge the embedded and supplied manifests in the -outputresource
         self.spawn(
             [
                 'mt.exe',
                 '-nologo',
                 '-manifest',
                 manfname,
                 inputresource,
                 outputresource,
             ]
         )
         log.info("done updating hg.exe's manifest")
         os.remove(manfname)
 
     @property
     def hgexepath(self):
         dir = os.path.dirname(self.get_ext_fullpath('dummy'))
         return os.path.join(self.build_temp, dir, 'hg.exe')
 
 
 class hgbuilddoc(Command):
     description = 'build documentation'
     user_options = [
         ('man', None, 'generate man pages'),
         ('html', None, 'generate html pages'),
     ]
 
     def initialize_options(self):
         self.man = None
         self.html = None
 
     def finalize_options(self):
         # If --man or --html are set, only generate what we're told to.
         # Otherwise generate everything.
         have_subset = self.man is not None or self.html is not None
 
         if have_subset:
             self.man = True if self.man else False
             self.html = True if self.html else False
         else:
             self.man = True
             self.html = True
 
     def run(self):
         def normalizecrlf(p):
             with open(p, 'rb') as fh:
                 orig = fh.read()
 
             if b'\r\n' not in orig:
                 return
 
             log.info('normalizing %s to LF line endings' % p)
             with open(p, 'wb') as fh:
                 fh.write(orig.replace(b'\r\n', b'\n'))
 
         def gentxt(root):
             txt = 'doc/%s.txt' % root
             log.info('generating %s' % txt)
             res, out, err = runcmd(
                 [sys.executable, 'gendoc.py', root], os.environ, cwd='doc'
             )
             if res:
                 raise SystemExit(
                     'error running gendoc.py: %s'
                     % '\n'.join([sysstr(out), sysstr(err)])
                 )
 
             with open(txt, 'wb') as fh:
                 fh.write(out)
 
         def gengendoc(root):
             gendoc = 'doc/%s.gendoc.txt' % root
 
             log.info('generating %s' % gendoc)
             res, out, err = runcmd(
                 [sys.executable, 'gendoc.py', '%s.gendoc' % root],
                 os.environ,
                 cwd='doc',
             )
             if res:
                 raise SystemExit(
                     'error running gendoc: %s'
                     % '\n'.join([sysstr(out), sysstr(err)])
                 )
 
             with open(gendoc, 'wb') as fh:
                 fh.write(out)
 
         def genman(root):
             log.info('generating doc/%s' % root)
             res, out, err = runcmd(
                 [
                     sys.executable,
                     'runrst',
                     'hgmanpage',
                     '--halt',
                     'warning',
                     '--strip-elements-with-class',
                     'htmlonly',
                     '%s.txt' % root,
                     root,
                 ],
                 os.environ,
                 cwd='doc',
             )
             if res:
                 raise SystemExit(
                     'error running runrst: %s'
                     % '\n'.join([sysstr(out), sysstr(err)])
                 )
 
             normalizecrlf('doc/%s' % root)
 
         def genhtml(root):
             log.info('generating doc/%s.html' % root)
             res, out, err = runcmd(
                 [
                     sys.executable,
                     'runrst',
                     'html',
                     '--halt',
                     'warning',
                     '--link-stylesheet',
                     '--stylesheet-path',
                     'style.css',
                     '%s.txt' % root,
                     '%s.html' % root,
                 ],
                 os.environ,
                 cwd='doc',
             )
             if res:
                 raise SystemExit(
                     'error running runrst: %s'
                     % '\n'.join([sysstr(out), sysstr(err)])
                 )
 
             normalizecrlf('doc/%s.html' % root)
 
         # This logic is duplicated in doc/Makefile.
         sources = {
             f
             for f in os.listdir('mercurial/helptext')
             if re.search(r'[0-9]\.txt$', f)
         }
 
         # common.txt is a one-off.
         gentxt('common')
 
         for source in sorted(sources):
             assert source[-4:] == '.txt'
             root = source[:-4]
 
             gentxt(root)
             gengendoc(root)
 
             if self.man:
                 genman(root)
             if self.html:
                 genhtml(root)
 
 
 class hginstall(install):
 
     user_options = install.user_options + [
         (
             'old-and-unmanageable',
             None,
             'noop, present for eggless setuptools compat',
         ),
         (
             'single-version-externally-managed',
             None,
             'noop, present for eggless setuptools compat',
         ),
     ]
 
     sub_commands = install.sub_commands + [
         ('install_completion', lambda self: True)
     ]
 
     # Also helps setuptools not be sad while we refuse to create eggs.
     single_version_externally_managed = True
 
     def get_sub_commands(self):
         # Screen out egg related commands to prevent egg generation.  But allow
         # mercurial.egg-info generation, since that is part of modern
         # packaging.
         excl = {'bdist_egg'}
         return filter(lambda x: x not in excl, install.get_sub_commands(self))
 
 
 class hginstalllib(install_lib):
     """
     This is a specialization of install_lib that replaces the copy_file used
     there so that it supports setting the mode of files after copying them,
     instead of just preserving the mode that the files originally had.  If your
     system has a umask of something like 027, preserving the permissions when
     copying will lead to a broken install.
 
     Note that just passing keep_permissions=False to copy_file would be
     insufficient, as it might still be applying a umask.
     """
 
     def run(self):
         realcopyfile = file_util.copy_file
 
         def copyfileandsetmode(*args, **kwargs):
             src, dst = args[0], args[1]
             dst, copied = realcopyfile(*args, **kwargs)
             if copied:
                 st = os.stat(src)
                 # Persist executable bit (apply it to group and other if user
                 # has it)
                 if st[stat.ST_MODE] & stat.S_IXUSR:
                     setmode = int('0755', 8)
                 else:
                     setmode = int('0644', 8)
                 m = stat.S_IMODE(st[stat.ST_MODE])
                 m = (m & ~int('0777', 8)) | setmode
                 os.chmod(dst, m)
 
         file_util.copy_file = copyfileandsetmode
         try:
             install_lib.run(self)
         finally:
             file_util.copy_file = realcopyfile
 
 
 class hginstallscripts(install_scripts):
     """
     This is a specialization of install_scripts that replaces the @LIBDIR@ with
     the configured directory for modules. If possible, the path is made relative
     to the directory for scripts.
     """
 
     def initialize_options(self):
         install_scripts.initialize_options(self)
 
         self.install_lib = None
 
     def finalize_options(self):
         install_scripts.finalize_options(self)
         self.set_undefined_options('install', ('install_lib', 'install_lib'))
 
     def run(self):
         install_scripts.run(self)
 
         # It only makes sense to replace @LIBDIR@ with the install path if
         # the install path is known. For wheels, the logic below calculates
         # the libdir to be "../..". This is because the internal layout of a
         # wheel archive looks like:
         #
         #   mercurial-3.6.1.data/scripts/hg
         #   mercurial/__init__.py
         #
         # When installing wheels, the subdirectories of the "<pkg>.data"
         # directory are translated to system local paths and files therein
         # are copied in place. The mercurial/* files are installed into the
         # site-packages directory. However, the site-packages directory
         # isn't known until wheel install time. This means we have no clue
         # at wheel generation time what the installed site-packages directory
         # will be. And, wheels don't appear to provide the ability to register
         # custom code to run during wheel installation. This all means that
         # we can't reliably set the libdir in wheels: the default behavior
         # of looking in sys.path must do.
 
         if (
             os.path.splitdrive(self.install_dir)[0]
             != os.path.splitdrive(self.install_lib)[0]
         ):
             # can't make relative paths from one drive to another, so use an
             # absolute path instead
             libdir = self.install_lib
         else:
             libdir = os.path.relpath(self.install_lib, self.install_dir)
 
         for outfile in self.outfiles:
             with open(outfile, 'rb') as fp:
                 data = fp.read()
 
             # skip binary files
             if b'\0' in data:
                 continue
 
             # During local installs, the shebang will be rewritten to the final
             # install path. During wheel packaging, the shebang has a special
             # value.
             if data.startswith(b'#!python'):
                 log.info(
                     'not rewriting @LIBDIR@ in %s because install path '
                     'not known' % outfile
                 )
                 continue
 
             data = data.replace(b'@LIBDIR@', libdir.encode('unicode_escape'))
             with open(outfile, 'wb') as fp:
                 fp.write(data)
 
 
 class hginstallcompletion(Command):
     description = 'Install shell completion'
 
     def initialize_options(self):
         self.install_dir = None
         self.outputs = []
 
     def finalize_options(self):
         self.set_undefined_options(
             'install_data', ('install_dir', 'install_dir')
         )
 
     def get_outputs(self):
         return self.outputs
 
     def run(self):
         for src, dir_path, dest in (
             (
                 'bash_completion',
                 ('share', 'bash-completion', 'completions'),
                 'hg',
             ),
             ('zsh_completion', ('share', 'zsh', 'site-functions'), '_hg'),
         ):
             dir = os.path.join(self.install_dir, *dir_path)
             self.mkpath(dir)
 
             dest = os.path.join(dir, dest)
             self.outputs.append(dest)
             self.copy_file(os.path.join('contrib', src), dest)
 
 
 # virtualenv installs custom distutils/__init__.py and
 # distutils/distutils.cfg files which essentially proxy back to the
 # "real" distutils in the main Python install. The presence of this
 # directory causes py2exe to pick up the "hacked" distutils package
 # from the virtualenv and "import distutils" will fail from the py2exe
 # build because the "real" distutils files can't be located.
 #
 # We work around this by monkeypatching the py2exe code finding Python
 # modules to replace the found virtualenv distutils modules with the
 # original versions via filesystem scanning. This is a bit hacky. But
 # it allows us to use virtualenvs for py2exe packaging, which is more
 # deterministic and reproducible.
 #
 # It's worth noting that the common StackOverflow suggestions for this
 # problem involve copying the original distutils files into the
 # virtualenv or into the staging directory after setup() is invoked.
 # The former is very brittle and can easily break setup(). Our hacking
 # of the found modules routine has a similar result as copying the files
 # manually. But it makes fewer assumptions about how py2exe works and
 # is less brittle.
 
 # This only catches virtualenvs made with virtualenv (as opposed to
 # venv, which is likely what Python 3 uses).
 py2exehacked = py2exeloaded and getattr(sys, 'real_prefix', None) is not None
 
 if py2exehacked:
     from distutils.command.py2exe import py2exe as buildpy2exe
     from py2exe.mf import Module as py2exemodule
 
     class hgbuildpy2exe(buildpy2exe):
         def find_needed_modules(self, mf, files, modules):
             res = buildpy2exe.find_needed_modules(self, mf, files, modules)
 
             # Replace virtualenv's distutils modules with the real ones.
             modules = {}
             for k, v in res.modules.items():
                 if k != 'distutils' and not k.startswith('distutils.'):
                     modules[k] = v
 
             res.modules = modules
 
             import opcode
 
             distutilsreal = os.path.join(
                 os.path.dirname(opcode.__file__), 'distutils'
             )
 
             for root, dirs, files in os.walk(distutilsreal):
                 for f in sorted(files):
                     if not f.endswith('.py'):
                         continue
 
                     full = os.path.join(root, f)
 
                     parents = ['distutils']
 
                     if root != distutilsreal:
                         rel = os.path.relpath(root, distutilsreal)
                         parents.extend(p for p in rel.split(os.sep))
 
                     modname = '%s.%s' % ('.'.join(parents), f[:-3])
 
                     if modname.startswith('distutils.tests.'):
                         continue
 
                     if modname.endswith('.__init__'):
                         modname = modname[: -len('.__init__')]
                         path = os.path.dirname(full)
                     else:
                         path = None
 
                     res.modules[modname] = py2exemodule(
                         modname, full, path=path
                     )
 
             if 'distutils' not in res.modules:
                 raise SystemExit('could not find distutils modules')
 
             return res
 
 
 cmdclass = {
     'build': hgbuild,
     'build_doc': hgbuilddoc,
     'build_mo': hgbuildmo,
     'build_ext': hgbuildext,
     'build_py': hgbuildpy,
     'build_scripts': hgbuildscripts,
     'build_hgextindex': buildhgextindex,
     'install': hginstall,
     'install_completion': hginstallcompletion,
     'install_lib': hginstalllib,
     'install_scripts': hginstallscripts,
     'build_hgexe': buildhgexe,
 }
 
 if py2exehacked:
     cmdclass['py2exe'] = hgbuildpy2exe
 
 packages = [
     'mercurial',
     'mercurial.cext',
     'mercurial.cffi',
     'mercurial.defaultrc',
     'mercurial.dirstateutils',
     'mercurial.helptext',
     'mercurial.helptext.internals',
     'mercurial.hgweb',
     'mercurial.interfaces',
     'mercurial.pure',
     'mercurial.templates',
     'mercurial.thirdparty',
     'mercurial.thirdparty.attr',
     'mercurial.thirdparty.zope',
     'mercurial.thirdparty.zope.interface',
     'mercurial.upgrade_utils',
     'mercurial.utils',
     'mercurial.revlogutils',
     'mercurial.testing',
     'hgext',
     'hgext.convert',
     'hgext.fsmonitor',
     'hgext.fastannotate',
     'hgext.fsmonitor.pywatchman',
     'hgext.git',
     'hgext.highlight',
     'hgext.hooklib',
     'hgext.infinitepush',
     'hgext.largefiles',
     'hgext.lfs',
     'hgext.narrow',
     'hgext.remotefilelog',
     'hgext.zeroconf',
     'hgext3rd',
     'hgdemandimport',
 ]
 
 for name in os.listdir(os.path.join('mercurial', 'templates')):
     if name != '__pycache__' and os.path.isdir(
         os.path.join('mercurial', 'templates', name)
     ):
         packages.append('mercurial.templates.%s' % name)
 
 if 'HG_PY2EXE_EXTRA_INSTALL_PACKAGES' in os.environ:
     # py2exe can't cope with namespace packages very well, so we have to
     # install any hgext3rd.* extensions that we want in the final py2exe
     # image here. This is gross, but you gotta do what you gotta do.
     packages.extend(os.environ['HG_PY2EXE_EXTRA_INSTALL_PACKAGES'].split(' '))
 
 common_depends = [
     'mercurial/bitmanipulation.h',
     'mercurial/compat.h',
     'mercurial/cext/util.h',
 ]
 common_include_dirs = ['mercurial']
 
 common_cflags = []
 
 # MSVC 2008 still needs declarations at the top of the scope, but Python 3.9
 # makes declarations not at the top of a scope in the headers.
 if os.name != 'nt' and sys.version_info[1] < 9:
     common_cflags = ['-Werror=declaration-after-statement']
 
 osutil_cflags = []
 osutil_ldflags = []
 
 # platform specific macros
 for plat, func in [('bsd', 'setproctitle')]:
     if re.search(plat, sys.platform) and hasfunction(new_compiler(), func):
         osutil_cflags.append('-DHAVE_%s' % func.upper())
 
 for plat, macro, code in [
     (
         'bsd|darwin',
         'BSD_STATFS',
         '''
      #include <sys/param.h>
      #include <sys/mount.h>
      int main() { struct statfs s; return sizeof(s.f_fstypename); }
      ''',
     ),
     (
         'linux',
         'LINUX_STATFS',
         '''
      #include <linux/magic.h>
      #include <sys/vfs.h>
      int main() { struct statfs s; return sizeof(s.f_type); }
      ''',
     ),
 ]:
     if re.search(plat, sys.platform) and cancompile(new_compiler(), code):
         osutil_cflags.append('-DHAVE_%s' % macro)
 
 if sys.platform == 'darwin':
     osutil_ldflags += ['-framework', 'ApplicationServices']
 
 if sys.platform == 'sunos5':
     osutil_ldflags += ['-lsocket']
 
 xdiff_srcs = [
     'mercurial/thirdparty/xdiff/xdiffi.c',
     'mercurial/thirdparty/xdiff/xprepare.c',
     'mercurial/thirdparty/xdiff/xutils.c',
 ]
 
 xdiff_headers = [
     'mercurial/thirdparty/xdiff/xdiff.h',
     'mercurial/thirdparty/xdiff/xdiffi.h',
     'mercurial/thirdparty/xdiff/xinclude.h',
     'mercurial/thirdparty/xdiff/xmacros.h',
     'mercurial/thirdparty/xdiff/xprepare.h',
     'mercurial/thirdparty/xdiff/xtypes.h',
     'mercurial/thirdparty/xdiff/xutils.h',
 ]
 
 
 class RustCompilationError(CCompilerError):
     """Exception class for Rust compilation errors."""
 
 
 class RustExtension(Extension):
     """Base classes for concrete Rust Extension classes."""
 
     rusttargetdir = os.path.join('rust', 'target', 'release')
 
     def __init__(self, mpath, sources, rustlibname, subcrate, **kw):
         Extension.__init__(self, mpath, sources, **kw)
         srcdir = self.rustsrcdir = os.path.join('rust', subcrate)
 
         # adding Rust source and control files to depends so that the extension
         # gets rebuilt if they've changed
         self.depends.append(os.path.join(srcdir, 'Cargo.toml'))
         cargo_lock = os.path.join(srcdir, 'Cargo.lock')
         if os.path.exists(cargo_lock):
             self.depends.append(cargo_lock)
         for dirpath, subdir, fnames in os.walk(os.path.join(srcdir, 'src')):
             self.depends.extend(
                 os.path.join(dirpath, fname)
                 for fname in fnames
                 if os.path.splitext(fname)[1] == '.rs'
             )
 
     @staticmethod
     def rustdylibsuffix():
         """Return the suffix for shared libraries produced by rustc.
 
         See also: https://doc.rust-lang.org/reference/linkage.html
         """
         if sys.platform == 'darwin':
             return '.dylib'
         elif os.name == 'nt':
             return '.dll'
         else:
             return '.so'
 
     def rustbuild(self):
         env = os.environ.copy()
         if 'HGTEST_RESTOREENV' in env:
             # Mercurial tests change HOME to a temporary directory,
             # but, if installed with rustup, the Rust toolchain needs
             # HOME to be correct (otherwise the 'no default toolchain'
             # error message is issued and the build fails).
             # This happens currently with test-hghave.t, which does
             # invoke this build.
 
             # Unix only fix (os.path.expanduser not really reliable if
             # HOME is shadowed like this)
             import pwd
 
             env['HOME'] = pwd.getpwuid(os.getuid()).pw_dir
 
         cargocmd = ['cargo', 'rustc', '--release']
 
         rust_features = env.get("HG_RUST_FEATURES")
         if rust_features:
             cargocmd.extend(('--features', rust_features))
 
         cargocmd.append('--')
         if sys.platform == 'darwin':
             cargocmd.extend(
                 ("-C", "link-arg=-undefined", "-C", "link-arg=dynamic_lookup")
             )
         try:
             subprocess.check_call(cargocmd, env=env, cwd=self.rustsrcdir)
-        except OSError as exc:
-            if exc.errno == errno.ENOENT:
+        except FileNotFoundError:
                 raise RustCompilationError("Cargo not found")
-            elif exc.errno == errno.EACCES:
+        except PermissionError:
                 raise RustCompilationError(
                     "Cargo found, but permission to execute it is denied"
                 )
-            else:
-                raise
         except subprocess.CalledProcessError:
             raise RustCompilationError(
                 "Cargo failed. Working directory: %r, "
                 "command: %r, environment: %r"
                 % (self.rustsrcdir, cargocmd, env)
             )
 
 
 class RustStandaloneExtension(RustExtension):
     def __init__(self, pydottedname, rustcrate, dylibname, **kw):
         RustExtension.__init__(
             self, pydottedname, [], dylibname, rustcrate, **kw
         )
         self.dylibname = dylibname
 
     def build(self, target_dir):
         self.rustbuild()
         target = [target_dir]
         target.extend(self.name.split('.'))
         target[-1] += DYLIB_SUFFIX
         shutil.copy2(
             os.path.join(
                 self.rusttargetdir, self.dylibname + self.rustdylibsuffix()
             ),
             os.path.join(*target),
         )
 
 
 extmodules = [
     Extension(
         'mercurial.cext.base85',
         ['mercurial/cext/base85.c'],
         include_dirs=common_include_dirs,
         extra_compile_args=common_cflags,
         depends=common_depends,
     ),
     Extension(
         'mercurial.cext.bdiff',
         ['mercurial/bdiff.c', 'mercurial/cext/bdiff.c'] + xdiff_srcs,
         include_dirs=common_include_dirs,
         extra_compile_args=common_cflags,
         depends=common_depends + ['mercurial/bdiff.h'] + xdiff_headers,
     ),
     Extension(
         'mercurial.cext.mpatch',
         ['mercurial/mpatch.c', 'mercurial/cext/mpatch.c'],
         include_dirs=common_include_dirs,
         extra_compile_args=common_cflags,
         depends=common_depends,
     ),
     Extension(
         'mercurial.cext.parsers',
         [
             'mercurial/cext/charencode.c',
             'mercurial/cext/dirs.c',
             'mercurial/cext/manifest.c',
             'mercurial/cext/parsers.c',
             'mercurial/cext/pathencode.c',
             'mercurial/cext/revlog.c',
         ],
         include_dirs=common_include_dirs,
         extra_compile_args=common_cflags,
         depends=common_depends
         + [
             'mercurial/cext/charencode.h',
             'mercurial/cext/revlog.h',
         ],
     ),
     Extension(
         'mercurial.cext.osutil',
         ['mercurial/cext/osutil.c'],
         include_dirs=common_include_dirs,
         extra_compile_args=common_cflags + osutil_cflags,
         extra_link_args=osutil_ldflags,
         depends=common_depends,
     ),
     Extension(
         'mercurial.thirdparty.zope.interface._zope_interface_coptimizations',
         [
             'mercurial/thirdparty/zope/interface/_zope_interface_coptimizations.c',
         ],
         extra_compile_args=common_cflags,
     ),
     Extension(
         'mercurial.thirdparty.sha1dc',
         [
             'mercurial/thirdparty/sha1dc/cext.c',
             'mercurial/thirdparty/sha1dc/lib/sha1.c',
             'mercurial/thirdparty/sha1dc/lib/ubc_check.c',
         ],
         extra_compile_args=common_cflags,
     ),
     Extension(
         'hgext.fsmonitor.pywatchman.bser',
         ['hgext/fsmonitor/pywatchman/bser.c'],
         extra_compile_args=common_cflags,
     ),
     RustStandaloneExtension(
         'mercurial.rustext',
         'hg-cpython',
         'librusthg',
     ),
 ]
 
 
 sys.path.insert(0, 'contrib/python-zstandard')
 import setup_zstd
 
 zstd = setup_zstd.get_c_extension(
     name='mercurial.zstd', root=os.path.abspath(os.path.dirname(__file__))
 )
 zstd.extra_compile_args += common_cflags
 extmodules.append(zstd)
 
 try:
     from distutils import cygwinccompiler
 
     # the -mno-cygwin option has been deprecated for years
     mingw32compilerclass = cygwinccompiler.Mingw32CCompiler
 
     class HackedMingw32CCompiler(cygwinccompiler.Mingw32CCompiler):
         def __init__(self, *args, **kwargs):
             mingw32compilerclass.__init__(self, *args, **kwargs)
             for i in 'compiler compiler_so linker_exe linker_so'.split():
                 try:
                     getattr(self, i).remove('-mno-cygwin')
                 except ValueError:
                     pass
 
     cygwinccompiler.Mingw32CCompiler = HackedMingw32CCompiler
 except ImportError:
     # the cygwinccompiler package is not available on some Python
     # distributions like the ones from the optware project for Synology
     # DiskStation boxes
     class HackedMingw32CCompiler:
         pass
 
 
 if os.name == 'nt':
     # Allow compiler/linker flags to be added to Visual Studio builds.  Passing
     # extra_link_args to distutils.extensions.Extension() doesn't have any
     # effect.
     from distutils import msvccompiler
 
     msvccompilerclass = msvccompiler.MSVCCompiler
 
     class HackedMSVCCompiler(msvccompiler.MSVCCompiler):
         def initialize(self):
             msvccompilerclass.initialize(self)
             # "warning LNK4197: export 'func' specified multiple times"
             self.ldflags_shared.append('/ignore:4197')
             self.ldflags_shared_debug.append('/ignore:4197')
 
     msvccompiler.MSVCCompiler = HackedMSVCCompiler
 
 packagedata = {
     'mercurial': [
         'locale/*/LC_MESSAGES/hg.mo',
         'dummycert.pem',
     ],
     'mercurial.defaultrc': [
         '*.rc',
     ],
     'mercurial.helptext': [
         '*.txt',
     ],
     'mercurial.helptext.internals': [
         '*.txt',
     ],
 }
 
 
 def ordinarypath(p):
     return p and p[0] != '.' and p[-1] != '~'
 
 
 for root in ('templates',):
     for curdir, dirs, files in os.walk(os.path.join('mercurial', root)):
         packagename = curdir.replace(os.sep, '.')
         packagedata[packagename] = list(filter(ordinarypath, files))
 
 datafiles = []
 
 # distutils expects version to be str/unicode. Converting it to
 # unicode on Python 2 still works because it won't contain any
 # non-ascii bytes and will be implicitly converted back to bytes
 # when operated on.
 assert isinstance(version, str)
 setupversion = version
 
 extra = {}
 
 py2exepackages = [
     'hgdemandimport',
     'hgext3rd',
     'hgext',
     'email',
     # implicitly imported per module policy
     # (cffi wouldn't be used as a frozen exe)
     'mercurial.cext',
     #'mercurial.cffi',
     'mercurial.pure',
 ]
 
 py2exe_includes = []
 
 py2exeexcludes = []
 py2exedllexcludes = ['crypt32.dll']
 
 if issetuptools:
     extra['python_requires'] = supportedpy
 
 if py2exeloaded:
     extra['console'] = [
         {
             'script': 'hg',
             'copyright': 'Copyright (C) 2005-2022 Olivia Mackall and others',
             'product_version': version,
         }
     ]
     # Sub command of 'build' because 'py2exe' does not handle sub_commands.
     # Need to override hgbuild because it has a private copy of
     # build.sub_commands.
     hgbuild.sub_commands.insert(0, ('build_hgextindex', None))
     # put dlls in sub directory so that they won't pollute PATH
     extra['zipfile'] = 'lib/library.zip'
 
     # We allow some configuration to be supplemented via environment
     # variables. This is better than setup.cfg files because it allows
     # supplementing configs instead of replacing them.
     extrapackages = os.environ.get('HG_PY2EXE_EXTRA_PACKAGES')
     if extrapackages:
         py2exepackages.extend(extrapackages.split(' '))
 
     extra_includes = os.environ.get('HG_PY2EXE_EXTRA_INCLUDES')
     if extra_includes:
         py2exe_includes.extend(extra_includes.split(' '))
 
     excludes = os.environ.get('HG_PY2EXE_EXTRA_EXCLUDES')
     if excludes:
         py2exeexcludes.extend(excludes.split(' '))
 
     dllexcludes = os.environ.get('HG_PY2EXE_EXTRA_DLL_EXCLUDES')
     if dllexcludes:
         py2exedllexcludes.extend(dllexcludes.split(' '))
 
 if os.environ.get('PYOXIDIZER'):
     hgbuild.sub_commands.insert(0, ('build_hgextindex', None))
 
 if os.name == 'nt':
     # Windows binary file versions for exe/dll files must have the
     # form W.X.Y.Z, where W,X,Y,Z are numbers in the range 0..65535
     setupversion = setupversion.split(r'+', 1)[0]
 
 if sys.platform == 'darwin' and os.path.exists('/usr/bin/xcodebuild'):
     version = runcmd(['/usr/bin/xcodebuild', '-version'], {})[1].splitlines()
     if version:
         version = version[0].decode('utf-8')
         xcode4 = version.startswith('Xcode') and StrictVersion(
             version.split()[1]
         ) >= StrictVersion('4.0')
         xcode51 = re.match(r'^Xcode\s+5\.1', version) is not None
     else:
         # xcodebuild returns empty on OS X Lion with XCode 4.3 not
         # installed, but instead with only command-line tools. Assume
         # that only happens on >= Lion, thus no PPC support.
         xcode4 = True
         xcode51 = False
 
     # XCode 4.0 dropped support for ppc architecture, which is hardcoded in
     # distutils.sysconfig
     if xcode4:
         os.environ['ARCHFLAGS'] = ''
 
     # XCode 5.1 changes clang such that it now fails to compile if the
     # -mno-fused-madd flag is passed, but the version of Python shipped with
     # OS X 10.9 Mavericks includes this flag. This causes problems in all
     # C extension modules, and a bug has been filed upstream at
     # http://bugs.python.org/issue21244. We also need to patch this here
     # so Mercurial can continue to compile in the meantime.
     if xcode51:
         cflags = get_config_var('CFLAGS')
         if cflags and re.search(r'-mno-fused-madd\b', cflags) is not None:
             os.environ['CFLAGS'] = (
                 os.environ.get('CFLAGS', '') + ' -Qunused-arguments'
             )
 
 setup(
     name='mercurial',
     version=setupversion,
     author='Olivia Mackall and many others',
     author_email='[e-mail]',
     url='https://mercurial-scm.org/',
     download_url='https://mercurial-scm.org/release/',
     description=(
         'Fast scalable distributed SCM (revision control, version '
         'control) system'
     ),
     long_description=(
         'Mercurial is a distributed SCM tool written in Python.'
         ' It is used by a number of large projects that require'
         ' fast, reliable distributed revision control, such as '
         'Mozilla.'
     ),
     license='GNU GPLv2 or any later version',
     classifiers=[
         'Development Status :: 6 - Mature',
         'Environment :: Console',
         'Intended Audience :: Developers',
         'Intended Audience :: System Administrators',
         'License :: OSI Approved :: GNU General Public License (GPL)',
         'Natural Language :: Danish',
         'Natural Language :: English',
         'Natural Language :: German',
         'Natural Language :: Italian',
         'Natural Language :: Japanese',
         'Natural Language :: Portuguese (Brazilian)',
         'Operating System :: Microsoft :: Windows',
         'Operating System :: OS Independent',
         'Operating System :: POSIX',
         'Programming Language :: C',
         'Programming Language :: Python',
         'Topic :: Software Development :: Version Control',
     ],
     scripts=scripts,
     packages=packages,
     ext_modules=extmodules,
     data_files=datafiles,
     package_data=packagedata,
     cmdclass=cmdclass,
     distclass=hgdist,
     options={
         'py2exe': {
             'bundle_files': 3,
             'dll_excludes': py2exedllexcludes,
             'includes': py2exe_includes,
             'excludes': py2exeexcludes,
             'packages': py2exepackages,
         },
         'bdist_mpkg': {
             'zipdist': False,
             'license': 'COPYING',
             'readme': 'contrib/packaging/macosx/Readme.html',
             'welcome': 'contrib/packaging/macosx/Welcome.html',
         },
     },
     **extra
 )