pytype: convert type comment for inline variable too

Same logic as for the previous changeset, but for "type comment" annotating
variables, not function/method.

As for the previous changeset, we had to adjust for of the types to actually match what was happening.

mercurial/chgserver.py

@@ -1,755 +1,761 b''
 # chgserver.py - command server extension for cHg
 #
 # Copyright 2011 Yuya Nishihara <[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.
 
 """command server extension for cHg
 
 'S' channel (read/write)
     propagate ui.system() request to client
 
 'attachio' command
     attach client's stdio passed by sendmsg()
 
 'chdir' command
     change current directory
 
 'setenv' command
     replace os.environ completely
 
 'setumask' command (DEPRECATED)
 'setumask2' command
     set umask
 
 'validate' command
     reload the config and check if the server is up to date
 
 Config
 ------
 
 ::
 
   [chgserver]
   # how long (in seconds) should an idle chg server exit
   idletimeout = 3600
 
   # whether to skip config or env change checks
   skiphash = False
 """
 
 
 import inspect
 import os
 import re
 import socket
 import stat
 import struct
 import time
 
+from typing import (
+    Optional,
+)
+
 from .i18n import _
 from .node import hex
 
 from . import (
     commandserver,
     encoding,
     error,
     extensions,
     pycompat,
     util,
 )
 
 from .utils import (
     hashutil,
     procutil,
     stringutil,
 )
 
 
 def _hashlist(items):
     """return sha1 hexdigest for a list"""
     return hex(hashutil.sha1(stringutil.pprint(items)).digest())
 
 
 # sensitive config sections affecting confighash
 _configsections = [
     b'alias',  # affects global state commands.table
     b'diff-tools',  # affects whether gui or not in extdiff's uisetup
     b'eol',  # uses setconfig('eol', ...)
     b'extdiff',  # uisetup will register new commands
     b'extensions',
     b'fastannotate',  # affects annotate command and adds fastannonate cmd
     b'merge-tools',  # affects whether gui or not in extdiff's uisetup
     b'schemes',  # extsetup will update global hg.schemes
 ]
 
 _configsectionitems = [
     (b'commands', b'show.aliasprefix'),  # show.py reads it in extsetup
 ]
 
 # sensitive environment variables affecting confighash
 _envre = re.compile(
     br'''\A(?:
                     CHGHG
                     |HG(?:DEMANDIMPORT|EMITWARNINGS|MODULEPOLICY|PROF|RCPATH)?
                     |HG(?:ENCODING|PLAIN).*
                     |LANG(?:UAGE)?
                     |LC_.*
                     |LD_.*
                     |PATH
                     |PYTHON.*
                     |TERM(?:INFO)?
                     |TZ
                     )\Z''',
     re.X,
 )
 
 
 def _confighash(ui):
     """return a quick hash for detecting config/env changes
 
     confighash is the hash of sensitive config items and environment variables.
 
     for chgserver, it is designed that once confighash changes, the server is
     not qualified to serve its client and should redirect the client to a new
     server. different from mtimehash, confighash change will not mark the
     server outdated and exit since the user can have different configs at the
     same time.
     """
     sectionitems = []
     for section in _configsections:
         sectionitems.append(ui.configitems(section))
     for section, item in _configsectionitems:
         sectionitems.append(ui.config(section, item))
     sectionhash = _hashlist(sectionitems)
     # If $CHGHG is set, the change to $HG should not trigger a new chg server
     if b'CHGHG' in encoding.environ:
         ignored = {b'HG'}
     else:
         ignored = set()
     envitems = [
         (k, v)
         for k, v in encoding.environ.items()
         if _envre.match(k) and k not in ignored
     ]
     envhash = _hashlist(sorted(envitems))
     return sectionhash[:6] + envhash[:6]
 
 
 def _getmtimepaths(ui):
     """get a list of paths that should be checked to detect change
 
     The list will include:
     - extensions (will not cover all files for complex extensions)
     - mercurial/__version__.py
     - python binary
     """
     modules = [m for n, m in extensions.extensions(ui)]
     try:
         from . import __version__
 
         modules.append(__version__)
     except ImportError:
         pass
     files = []
     if pycompat.sysexecutable:
         files.append(pycompat.sysexecutable)
     for m in modules:
         try:
             files.append(pycompat.fsencode(inspect.getabsfile(m)))
         except TypeError:
             pass
     return sorted(set(files))
 
 
 def _mtimehash(paths):
     """return a quick hash for detecting file changes
 
     mtimehash calls stat on given paths and calculate a hash based on size and
     mtime of each file. mtimehash does not read file content because reading is
     expensive. therefore it's not 100% reliable for detecting content changes.
     it's possible to return different hashes for same file contents.
     it's also possible to return a same hash for different file contents for
     some carefully crafted situation.
 
     for chgserver, it is designed that once mtimehash changes, the server is
     considered outdated immediately and should no longer provide service.
 
     mtimehash is not included in confighash because we only know the paths of
     extensions after importing them (there is imp.find_module but that faces
     race conditions). We need to calculate confighash without importing.
     """
 
     def trystat(path):
         try:
             st = os.stat(path)
             return (st[stat.ST_MTIME], st.st_size)
         except OSError:
             # could be ENOENT, EPERM etc. not fatal in any case
             pass
 
     return _hashlist(pycompat.maplist(trystat, paths))[:12]
 
 
 class hashstate:
     """a structure storing confighash, mtimehash, paths used for mtimehash"""
 
     def __init__(self, confighash, mtimehash, mtimepaths):
         self.confighash = confighash
         self.mtimehash = mtimehash
         self.mtimepaths = mtimepaths
 
     @staticmethod
     def fromui(ui, mtimepaths=None):
         if mtimepaths is None:
             mtimepaths = _getmtimepaths(ui)
         confighash = _confighash(ui)
         mtimehash = _mtimehash(mtimepaths)
         ui.log(
             b'cmdserver',
             b'confighash = %s mtimehash = %s\n',
             confighash,
             mtimehash,
         )
         return hashstate(confighash, mtimehash, mtimepaths)
 
 
 def _newchgui(srcui, csystem, attachio):
     class chgui(srcui.__class__):
         def __init__(self, src=None):
             super(chgui, self).__init__(src)
             if src:
                 self._csystem = getattr(src, '_csystem', csystem)
             else:
                 self._csystem = csystem
 
         def _runsystem(self, cmd, environ, cwd, out):
             # fallback to the original system method if
             #  a. the output stream is not stdout (e.g. stderr, cStringIO),
             #  b. or stdout is redirected by protectfinout(),
             # because the chg client is not aware of these situations and
             # will behave differently (i.e. write to stdout).
             if (
                 out is not self.fout
                 or not hasattr(self.fout, 'fileno')
                 or self.fout.fileno() != procutil.stdout.fileno()
                 or self._finoutredirected
             ):
                 return procutil.system(cmd, environ=environ, cwd=cwd, out=out)
             self.flush()
             return self._csystem(cmd, procutil.shellenviron(environ), cwd)
 
         def _runpager(self, cmd, env=None):
             self._csystem(
                 cmd,
                 procutil.shellenviron(env),
                 type=b'pager',
                 cmdtable={b'attachio': attachio},
             )
             return True
 
     return chgui(srcui)
 
 
 def _loadnewui(srcui, args, cdebug):
     from . import dispatch  # avoid cycle
 
     newui = srcui.__class__.load()
     for a in ['fin', 'fout', 'ferr', 'environ']:
         setattr(newui, a, getattr(srcui, a))
     if hasattr(srcui, '_csystem'):
         newui._csystem = srcui._csystem
 
     # command line args
     options = dispatch._earlyparseopts(newui, args)
     dispatch._parseconfig(newui, options[b'config'])
 
     # stolen from tortoisehg.util.copydynamicconfig()
     for section, name, value in srcui.walkconfig():
         source = srcui.configsource(section, name)
         if b':' in source or source == b'--config' or source.startswith(b'$'):
             # path:line or command line, or environ
             continue
         newui.setconfig(section, name, value, source)
 
     # load wd and repo config, copied from dispatch.py
     cwd = options[b'cwd']
     cwd = cwd and os.path.realpath(cwd) or None
     rpath = options[b'repository']
     path, newlui = dispatch._getlocal(newui, rpath, wd=cwd)
 
     extensions.populateui(newui)
     commandserver.setuplogging(newui, fp=cdebug)
     if newui is not newlui:
         extensions.populateui(newlui)
         commandserver.setuplogging(newlui, fp=cdebug)
 
     return (newui, newlui)
 
 
 class channeledsystem:
     """Propagate ui.system() request in the following format:
 
     payload length (unsigned int),
     type, '\0',
     cmd, '\0',
     cwd, '\0',
     envkey, '=', val, '\0',
     ...
     envkey, '=', val
 
     if type == 'system', waits for:
 
     exitcode length (unsigned int),
     exitcode (int)
 
     if type == 'pager', repetitively waits for a command name ending with '\n'
     and executes it defined by cmdtable, or exits the loop if the command name
     is empty.
     """
 
     def __init__(self, in_, out, channel):
         self.in_ = in_
         self.out = out
         self.channel = channel
 
     def __call__(self, cmd, environ, cwd=None, type=b'system', cmdtable=None):
         args = [type, cmd, util.abspath(cwd or b'.')]
         args.extend(b'%s=%s' % (k, v) for k, v in environ.items())
         data = b'\0'.join(args)
         self.out.write(struct.pack(b'>cI', self.channel, len(data)))
         self.out.write(data)
         self.out.flush()
 
         if type == b'system':
             length = self.in_.read(4)
             (length,) = struct.unpack(b'>I', length)
             if length != 4:
                 raise error.Abort(_(b'invalid response'))
             (rc,) = struct.unpack(b'>i', self.in_.read(4))
             return rc
         elif type == b'pager':
             while True:
                 cmd = self.in_.readline()[:-1]
                 if not cmd:
                     break
                 if cmdtable and cmd in cmdtable:
                     cmdtable[cmd]()
                 else:
                     raise error.Abort(_(b'unexpected command: %s') % cmd)
         else:
             raise error.ProgrammingError(b'invalid S channel type: %s' % type)
 
 
 _iochannels = [
     # server.ch, ui.fp, mode
     ('cin', 'fin', 'rb'),
     ('cout', 'fout', 'wb'),
     ('cerr', 'ferr', 'wb'),
 ]
 
 
 class chgcmdserver(commandserver.server):
     def __init__(
         self, ui, repo, fin, fout, sock, prereposetups, hashstate, baseaddress
     ):
         super(chgcmdserver, self).__init__(
             _newchgui(ui, channeledsystem(fin, fout, b'S'), self.attachio),
             repo,
             fin,
             fout,
             prereposetups,
         )
         self.clientsock = sock
         self._ioattached = False
         self._oldios = []  # original (self.ch, ui.fp, fd) before "attachio"
         self.hashstate = hashstate
         self.baseaddress = baseaddress
         if hashstate is not None:
             self.capabilities = self.capabilities.copy()
             self.capabilities[b'validate'] = chgcmdserver.validate
 
     def cleanup(self):
         super(chgcmdserver, self).cleanup()
         # dispatch._runcatch() does not flush outputs if exception is not
         # handled by dispatch._dispatch()
         self.ui.flush()
         self._restoreio()
         self._ioattached = False
 
     def attachio(self):
         """Attach to client's stdio passed via unix domain socket; all
         channels except cresult will no longer be used
         """
         # tell client to sendmsg() with 1-byte payload, which makes it
         # distinctive from "attachio\n" command consumed by client.read()
         self.clientsock.sendall(struct.pack(b'>cI', b'I', 1))
 
         data, ancdata, msg_flags, address = self.clientsock.recvmsg(1, 256)
         assert len(ancdata) == 1
         cmsg_level, cmsg_type, cmsg_data = ancdata[0]
         assert cmsg_level == socket.SOL_SOCKET
         assert cmsg_type == socket.SCM_RIGHTS
         # memoryview.cast() was added in typeshed 61600d68772a, but pytype
         # still complains
         # pytype: disable=attribute-error
         clientfds = memoryview(cmsg_data).cast('i').tolist()
         # pytype: enable=attribute-error
         self.ui.log(b'chgserver', b'received fds: %r\n', clientfds)
 
         ui = self.ui
         ui.flush()
         self._saveio()
         for fd, (cn, fn, mode) in zip(clientfds, _iochannels):
             assert fd > 0
             fp = getattr(ui, fn)
             os.dup2(fd, fp.fileno())
             os.close(fd)
             if self._ioattached:
                 continue
             # reset buffering mode when client is first attached. as we want
             # to see output immediately on pager, the mode stays unchanged
             # when client re-attached. ferr is unchanged because it should
             # be unbuffered no matter if it is a tty or not.
             if fn == b'ferr':
                 newfp = fp
             else:
                 # On Python 3, the standard library doesn't offer line-buffered
                 # binary streams, so wrap/unwrap it.
                 if fp.isatty():
                     newfp = procutil.make_line_buffered(fp)
                 else:
                     newfp = procutil.unwrap_line_buffered(fp)
             if newfp is not fp:
                 setattr(ui, fn, newfp)
             setattr(self, cn, newfp)
 
         self._ioattached = True
         self.cresult.write(struct.pack(b'>i', len(clientfds)))
 
     def _saveio(self):
         if self._oldios:
             return
         ui = self.ui
         for cn, fn, _mode in _iochannels:
             ch = getattr(self, cn)
             fp = getattr(ui, fn)
             fd = os.dup(fp.fileno())
             self._oldios.append((ch, fp, fd))
 
     def _restoreio(self):
         if not self._oldios:
             return
         nullfd = os.open(os.devnull, os.O_WRONLY)
         ui = self.ui
         for (ch, fp, fd), (cn, fn, mode) in zip(self._oldios, _iochannels):
             try:
                 if 'w' in mode:
                     # Discard buffered data which couldn't be flushed because
                     # of EPIPE. The data should belong to the current session
                     # and should never persist.
                     os.dup2(nullfd, fp.fileno())
                     fp.flush()
                 os.dup2(fd, fp.fileno())
                 os.close(fd)
             except OSError as err:
                 # According to issue6330, running chg on heavy loaded systems
                 # can lead to EBUSY. [man dup2] indicates that, on Linux,
                 # EBUSY comes from a race condition between open() and dup2().
                 # However it's not clear why open() race occurred for
                 # newfd=stdin/out/err.
                 self.ui.log(
                     b'chgserver',
                     b'got %s while duplicating %s\n',
                     stringutil.forcebytestr(err),
                     fn,
                 )
             setattr(self, cn, ch)
             setattr(ui, fn, fp)
         os.close(nullfd)
         del self._oldios[:]
 
     def validate(self):
         """Reload the config and check if the server is up to date
 
         Read a list of '\0' separated arguments.
         Write a non-empty list of '\0' separated instruction strings or '\0'
         if the list is empty.
         An instruction string could be either:
             - "unlink $path", the client should unlink the path to stop the
               outdated server.
             - "redirect $path", the client should attempt to connect to $path
               first. If it does not work, start a new server. It implies
               "reconnect".
             - "exit $n", the client should exit directly with code n.
               This may happen if we cannot parse the config.
             - "reconnect", the client should close the connection and
               reconnect.
         If neither "reconnect" nor "redirect" is included in the instruction
         list, the client can continue with this server after completing all
         the instructions.
         """
         args = self._readlist()
         errorraised = False
         detailed_exit_code = 255
         try:
             self.ui, lui = _loadnewui(self.ui, args, self.cdebug)
         except error.RepoError as inst:
             # RepoError can be raised while trying to read shared source
             # configuration
             self.ui.error(_(b"abort: %s\n") % stringutil.forcebytestr(inst))
             if inst.hint:
                 self.ui.error(_(b"(%s)\n") % inst.hint)
             errorraised = True
         except error.Error as inst:
             if inst.detailed_exit_code is not None:
                 detailed_exit_code = inst.detailed_exit_code
             self.ui.error(inst.format())
             errorraised = True
 
         if errorraised:
             self.ui.flush()
             exit_code = 255
             if self.ui.configbool(b'ui', b'detailed-exit-code'):
                 exit_code = detailed_exit_code
             self.cresult.write(b'exit %d' % exit_code)
             return
         newhash = hashstate.fromui(lui, self.hashstate.mtimepaths)
         insts = []
         if newhash.mtimehash != self.hashstate.mtimehash:
             addr = _hashaddress(self.baseaddress, self.hashstate.confighash)
             insts.append(b'unlink %s' % addr)
             # mtimehash is empty if one or more extensions fail to load.
             # to be compatible with hg, still serve the client this time.
             if self.hashstate.mtimehash:
                 insts.append(b'reconnect')
         if newhash.confighash != self.hashstate.confighash:
             addr = _hashaddress(self.baseaddress, newhash.confighash)
             insts.append(b'redirect %s' % addr)
         self.ui.log(b'chgserver', b'validate: %s\n', stringutil.pprint(insts))
         self.cresult.write(b'\0'.join(insts) or b'\0')
 
     def chdir(self):
         """Change current directory
 
         Note that the behavior of --cwd option is bit different from this.
         It does not affect --config parameter.
         """
         path = self._readstr()
         if not path:
             return
         self.ui.log(b'chgserver', b"chdir to '%s'\n", path)
         os.chdir(path)
 
     def setumask(self):
         """Change umask (DEPRECATED)"""
         # BUG: this does not follow the message frame structure, but kept for
         # backward compatibility with old chg clients for some time
         self._setumask(self._read(4))
 
     def setumask2(self):
         """Change umask"""
         data = self._readstr()
         if len(data) != 4:
             raise ValueError(b'invalid mask length in setumask2 request')
         self._setumask(data)
 
     def _setumask(self, data):
         mask = struct.unpack(b'>I', data)[0]
         self.ui.log(b'chgserver', b'setumask %r\n', mask)
         util.setumask(mask)
 
     def runcommand(self):
         # pager may be attached within the runcommand session, which should
         # be detached at the end of the session. otherwise the pager wouldn't
         # receive EOF.
         globaloldios = self._oldios
         self._oldios = []
         try:
             return super(chgcmdserver, self).runcommand()
         finally:
             self._restoreio()
             self._oldios = globaloldios
 
     def setenv(self):
         """Clear and update os.environ
 
         Note that not all variables can make an effect on the running process.
         """
         l = self._readlist()
         try:
             newenv = dict(s.split(b'=', 1) for s in l)
         except ValueError:
             raise ValueError(b'unexpected value in setenv request')
         self.ui.log(b'chgserver', b'setenv: %r\n', sorted(newenv.keys()))
 
         encoding.environ.clear()
         encoding.environ.update(newenv)
 
     capabilities = commandserver.server.capabilities.copy()
     capabilities.update(
         {
             b'attachio': attachio,
             b'chdir': chdir,
             b'runcommand': runcommand,
             b'setenv': setenv,
             b'setumask': setumask,
             b'setumask2': setumask2,
         }
     )
 
     if hasattr(procutil, 'setprocname'):
 
         def setprocname(self):
             """Change process title"""
             name = self._readstr()
             self.ui.log(b'chgserver', b'setprocname: %r\n', name)
             procutil.setprocname(name)
 
         capabilities[b'setprocname'] = setprocname
 
 
 def _tempaddress(address):
     return b'%s.%d.tmp' % (address, os.getpid())
 
 
 def _hashaddress(address, hashstr):
     # if the basename of address contains '.', use only the left part. this
     # makes it possible for the client to pass 'server.tmp$PID' and follow by
     # an atomic rename to avoid locking when spawning new servers.
     dirname, basename = os.path.split(address)
     basename = basename.split(b'.', 1)[0]
     return b'%s-%s' % (os.path.join(dirname, basename), hashstr)
 
 
 class chgunixservicehandler:
     """Set of operations for chg services"""
 
     pollinterval = 1  # [sec]
 
+    _hashstate: Optional[hashstate]
+    _baseaddress: Optional[bytes]
+    _realaddress: Optional[bytes]
+
     def __init__(self, ui):
         self.ui = ui
 
-        # TODO: use PEP 526 syntax (`_hashstate: hashstate` at the class level)
-        #  when 3.5 support is dropped.
-        self._hashstate = None  # type: hashstate
-        self._baseaddress = None  # type: bytes
-        self._realaddress = None  # type: bytes
+        self._hashstate = None
+        self._baseaddress = None
+        self._realaddress = None
 
         self._idletimeout = ui.configint(b'chgserver', b'idletimeout')
         self._lastactive = time.time()
 
     def bindsocket(self, sock, address):
         self._inithashstate(address)
         self._checkextensions()
         self._bind(sock)
         self._createsymlink()
         # no "listening at" message should be printed to simulate hg behavior
 
     def _inithashstate(self, address):
         self._baseaddress = address
         if self.ui.configbool(b'chgserver', b'skiphash'):
             self._hashstate = None
             self._realaddress = address
             return
         self._hashstate = hashstate.fromui(self.ui)
         self._realaddress = _hashaddress(address, self._hashstate.confighash)
 
     def _checkextensions(self):
         if not self._hashstate:
             return
         if extensions.notloaded():
             # one or more extensions failed to load. mtimehash becomes
             # meaningless because we do not know the paths of those extensions.
             # set mtimehash to an illegal hash value to invalidate the server.
             self._hashstate.mtimehash = b''
 
     def _bind(self, sock):
         # use a unique temp address so we can stat the file and do ownership
         # check later
         tempaddress = _tempaddress(self._realaddress)
         util.bindunixsocket(sock, tempaddress)
         self._socketstat = os.stat(tempaddress)
         sock.listen(socket.SOMAXCONN)
         # rename will replace the old socket file if exists atomically. the
         # old server will detect ownership change and exit.
         util.rename(tempaddress, self._realaddress)
 
     def _createsymlink(self):
         if self._baseaddress == self._realaddress:
             return
         tempaddress = _tempaddress(self._baseaddress)
         os.symlink(os.path.basename(self._realaddress), tempaddress)
         util.rename(tempaddress, self._baseaddress)
 
     def _issocketowner(self):
         try:
             st = os.stat(self._realaddress)
             return (
                 st.st_ino == self._socketstat.st_ino
                 and st[stat.ST_MTIME] == self._socketstat[stat.ST_MTIME]
             )
         except OSError:
             return False
 
     def unlinksocket(self, address):
         if not self._issocketowner():
             return
         # it is possible to have a race condition here that we may
         # remove another server's socket file. but that's okay
         # since that server will detect and exit automatically and
         # the client will start a new server on demand.
         util.tryunlink(self._realaddress)
 
     def shouldexit(self):
         if not self._issocketowner():
             self.ui.log(
                 b'chgserver', b'%s is not owned, exiting.\n', self._realaddress
             )
             return True
         if time.time() - self._lastactive > self._idletimeout:
             self.ui.log(b'chgserver', b'being idle too long. exiting.\n')
             return True
         return False
 
     def newconnection(self):
         self._lastactive = time.time()
 
     def createcmdserver(self, repo, conn, fin, fout, prereposetups):
         return chgcmdserver(
             self.ui,
             repo,
             fin,
             fout,
             conn,
             prereposetups,
             self._hashstate,
             self._baseaddress,
         )
 
 
 def chgunixservice(ui, repo, opts):
     # CHGINTERNALMARK is set by chg client. It is an indication of things are
     # started by chg so other code can do things accordingly, like disabling
     # demandimport or detecting chg client started by chg client. When executed
     # here, CHGINTERNALMARK is no longer useful and hence dropped to make
     # environ cleaner.
     if b'CHGINTERNALMARK' in encoding.environ:
         del encoding.environ[b'CHGINTERNALMARK']
     # Python3.7+ "coerces" the LC_CTYPE environment variable to a UTF-8 one if
     # it thinks the current value is "C". This breaks the hash computation and
     # causes chg to restart loop.
     if b'CHGORIG_LC_CTYPE' in encoding.environ:
         encoding.environ[b'LC_CTYPE'] = encoding.environ[b'CHGORIG_LC_CTYPE']
         del encoding.environ[b'CHGORIG_LC_CTYPE']
     elif b'CHG_CLEAR_LC_CTYPE' in encoding.environ:
         if b'LC_CTYPE' in encoding.environ:
             del encoding.environ[b'LC_CTYPE']
         del encoding.environ[b'CHG_CLEAR_LC_CTYPE']
 
     if repo:
         # one chgserver can serve multiple repos. drop repo information
         ui.setconfig(b'bundle', b'mainreporoot', b'', b'repo')
     h = chgunixservicehandler(ui)
     return commandserver.unixforkingservice(ui, repo=None, opts=opts, handler=h)

mercurial/error.py

@@ -1,704 +1,704 b''
 # error.py - Mercurial exceptions
 #
 # Copyright 2005-2008 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.
 
 """Mercurial exceptions.
 
 This allows us to catch exceptions at higher levels without forcing
 imports.
 """
 
 
 import difflib
 
 from typing import (
     Any,
     AnyStr,
     Iterable,
     List,
     Optional,
     Sequence,
     Union,
 )
 
 # Do not import anything but pycompat here, please
 from . import pycompat
 
 
 # keeps pyflakes happy
 assert [
     Any,
     AnyStr,
     Iterable,
     List,
     Optional,
     Sequence,
     Union,
 ]
 
 
 def _tobytes(exc) -> bytes:
     """Byte-stringify exception in the same way as BaseException_str()"""
     if not exc.args:
         return b''
     if len(exc.args) == 1:
         return pycompat.bytestr(exc.args[0])
     return b'(%s)' % b', '.join(b"'%s'" % pycompat.bytestr(a) for a in exc.args)
 
 
 class Hint:
     """Mix-in to provide a hint of an error
 
     This should come first in the inheritance list to consume a hint and
     pass remaining arguments to the exception class.
     """
 
     def __init__(self, *args, **kw):
-        self.hint = kw.pop('hint', None)  # type: Optional[bytes]
+        self.hint: Optional[bytes] = kw.pop('hint', None)
         super(Hint, self).__init__(*args, **kw)
 
 
 class Error(Hint, Exception):
     """Base class for Mercurial errors."""
 
     coarse_exit_code = None
     detailed_exit_code = None
 
     def __init__(self, message: bytes, hint: Optional[bytes] = None) -> None:
         self.message = message
         self.hint = hint
         # Pass the message into the Exception constructor to help extensions
         # that look for exc.args[0].
         Exception.__init__(self, message)
 
     def __bytes__(self):
         return self.message
 
     def __str__(self) -> str:
         # the output would be unreadable if the message was translated,
         # but do not replace it with encoding.strfromlocal(), which
         # may raise another exception.
         return pycompat.sysstr(self.__bytes__())
 
     def format(self) -> bytes:
         from .i18n import _
 
         message = _(b"abort: %s\n") % self.message
         if self.hint:
             message += _(b"(%s)\n") % self.hint
         return message
 
 
 class Abort(Error):
     """Raised if a command needs to print an error and exit."""
 
 
 class StorageError(Error):
     """Raised when an error occurs in a storage layer.
 
     Usually subclassed by a storage-specific exception.
     """
 
     detailed_exit_code = 50
 
 
 class RevlogError(StorageError):
     pass
 
 
 class SidedataHashError(RevlogError):
     def __init__(self, key: int, expected: bytes, got: bytes) -> None:
         self.hint = None
         self.sidedatakey = key
         self.expecteddigest = expected
         self.actualdigest = got
 
 
 class FilteredIndexError(IndexError):
     __bytes__ = _tobytes
 
 
 class LookupError(RevlogError, KeyError):
     def __init__(self, name: bytes, index: bytes, message: bytes) -> None:
         self.name = name
         self.index = index
         # this can't be called 'message' because at least some installs of
         # Python 2.6+ complain about the 'message' property being deprecated
         self.lookupmessage = message
         if isinstance(name, bytes) and len(name) == 20:
             from .node import hex
 
             name = hex(name)
         # if name is a binary node, it can be None
         RevlogError.__init__(
             self, b'%s@%s: %s' % (index, pycompat.bytestr(name), message)
         )
 
     def __bytes__(self):
         return RevlogError.__bytes__(self)
 
     def __str__(self):
         return RevlogError.__str__(self)
 
 
 class AmbiguousPrefixLookupError(LookupError):
     pass
 
 
 class FilteredLookupError(LookupError):
     pass
 
 
 class ManifestLookupError(LookupError):
     pass
 
 
 class CommandError(Exception):
     """Exception raised on errors in parsing the command line."""
 
     def __init__(self, command: Optional[bytes], message: bytes) -> None:
         self.command = command
         self.message = message
         super(CommandError, self).__init__()
 
     __bytes__ = _tobytes
 
 
 class UnknownCommand(Exception):
     """Exception raised if command is not in the command table."""
 
     def __init__(
         self,
         command: bytes,
         all_commands: Optional[List[bytes]] = None,
     ) -> None:
         self.command = command
         self.all_commands = all_commands
         super(UnknownCommand, self).__init__()
 
     __bytes__ = _tobytes
 
 
 class AmbiguousCommand(Exception):
     """Exception raised if command shortcut matches more than one command."""
 
     def __init__(self, prefix: bytes, matches: List[bytes]) -> None:
         self.prefix = prefix
         self.matches = matches
         super(AmbiguousCommand, self).__init__()
 
     __bytes__ = _tobytes
 
 
 class WorkerError(Exception):
     """Exception raised when a worker process dies."""
 
     def __init__(self, status_code: int) -> None:
         self.status_code = status_code
         # Pass status code to superclass just so it becomes part of __bytes__
         super(WorkerError, self).__init__(status_code)
 
     __bytes__ = _tobytes
 
 
 class InterventionRequired(Abort):
     """Exception raised when a command requires human intervention."""
 
     coarse_exit_code = 1
     detailed_exit_code = 240
 
     def format(self) -> bytes:
         from .i18n import _
 
         message = _(b"%s\n") % self.message
         if self.hint:
             message += _(b"(%s)\n") % self.hint
         return message
 
 
 class ConflictResolutionRequired(InterventionRequired):
     """Exception raised when a continuable command required merge conflict resolution."""
 
     def __init__(self, opname: bytes) -> None:
         from .i18n import _
 
         self.opname = opname
         InterventionRequired.__init__(
             self,
             _(
                 b"unresolved conflicts (see 'hg resolve', then 'hg %s --continue')"
             )
             % opname,
         )
 
 
 class InputError(Abort):
     """Indicates that the user made an error in their input.
 
     Examples: Invalid command, invalid flags, invalid revision.
     """
 
     detailed_exit_code = 10
 
 
 class StateError(Abort):
     """Indicates that the operation might work if retried in a different state.
 
     Examples: Unresolved merge conflicts, unfinished operations.
     """
 
     detailed_exit_code = 20
 
 
 class CanceledError(Abort):
     """Indicates that the user canceled the operation.
 
     Examples: Close commit editor with error status, quit chistedit.
     """
 
     detailed_exit_code = 250
 
 
 class SecurityError(Abort):
     """Indicates that some aspect of security failed.
 
     Examples: Bad server credentials, expired local credentials for network
     filesystem, mismatched GPG signature, DoS protection.
     """
 
     detailed_exit_code = 150
 
 
 class HookLoadError(Abort):
     """raised when loading a hook fails, aborting an operation
 
     Exists to allow more specialized catching."""
 
 
 class HookAbort(Abort):
     """raised when a validation hook fails, aborting an operation
 
     Exists to allow more specialized catching."""
 
     detailed_exit_code = 40
 
 
 class ConfigError(Abort):
     """Exception raised when parsing config files"""
 
     detailed_exit_code = 30
 
     def __init__(
         self,
         message: bytes,
         location: Optional[bytes] = None,
         hint: Optional[bytes] = None,
     ) -> None:
         super(ConfigError, self).__init__(message, hint=hint)
         self.location = location
 
     def format(self) -> bytes:
         from .i18n import _
 
         if self.location is not None:
             message = _(b"config error at %s: %s\n") % (
                 pycompat.bytestr(self.location),
                 self.message,
             )
         else:
             message = _(b"config error: %s\n") % self.message
         if self.hint:
             message += _(b"(%s)\n") % self.hint
         return message
 
 
 class UpdateAbort(Abort):
     """Raised when an update is aborted for destination issue"""
 
 
 class MergeDestAbort(Abort):
     """Raised when an update is aborted for destination issues"""
 
 
 class NoMergeDestAbort(MergeDestAbort):
     """Raised when an update is aborted because there is nothing to merge"""
 
 
 class ManyMergeDestAbort(MergeDestAbort):
     """Raised when an update is aborted because destination is ambiguous"""
 
 
 class ResponseExpected(Abort):
     """Raised when an EOF is received for a prompt"""
 
     def __init__(self):
         from .i18n import _
 
         Abort.__init__(self, _(b'response expected'))
 
 
 class RemoteError(Abort):
     """Exception raised when interacting with a remote repo fails"""
 
     detailed_exit_code = 100
 
 
 class OutOfBandError(RemoteError):
     """Exception raised when a remote repo reports failure"""
 
     def __init__(
         self,
         message: Optional[bytes] = None,
         hint: Optional[bytes] = None,
     ):
         from .i18n import _
 
         if message:
             # Abort.format() adds a trailing newline
             message = _(b"remote error:\n%s") % message.rstrip(b'\n')
         else:
             message = _(b"remote error")
         super(OutOfBandError, self).__init__(message, hint=hint)
 
 
 class ParseError(Abort):
     """Raised when parsing config files and {rev,file}sets (msg[, pos])"""
 
     detailed_exit_code = 10
 
     def __init__(
         self,
         message: bytes,
         location: Optional[Union[bytes, int]] = None,
         hint: Optional[bytes] = None,
     ):
         super(ParseError, self).__init__(message, hint=hint)
         self.location = location
 
     def format(self) -> bytes:
         from .i18n import _
 
         if self.location is not None:
             message = _(b"hg: parse error at %s: %s\n") % (
                 pycompat.bytestr(self.location),
                 self.message,
             )
         else:
             message = _(b"hg: parse error: %s\n") % self.message
         if self.hint:
             message += _(b"(%s)\n") % self.hint
         return message
 
 
 class PatchError(Exception):
     __bytes__ = _tobytes
 
 
 class PatchParseError(PatchError):
     __bytes__ = _tobytes
 
 
 class PatchApplicationError(PatchError):
     __bytes__ = _tobytes
 
 
 def getsimilar(symbols: Iterable[bytes], value: bytes) -> List[bytes]:
     sim = lambda x: difflib.SequenceMatcher(None, value, x).ratio()
     # The cutoff for similarity here is pretty arbitrary. It should
     # probably be investigated and tweaked.
     return [s for s in symbols if sim(s) > 0.6]
 
 
 def similarity_hint(similar: List[bytes]) -> Optional[bytes]:
     from .i18n import _
 
     if len(similar) == 1:
         return _(b"did you mean %s?") % similar[0]
     elif similar:
         ss = b", ".join(sorted(similar))
         return _(b"did you mean one of %s?") % ss
     else:
         return None
 
 
 class UnknownIdentifier(ParseError):
     """Exception raised when a {rev,file}set references an unknown identifier"""
 
     def __init__(self, function: bytes, symbols: Iterable[bytes]) -> None:
         from .i18n import _
 
         similar = getsimilar(symbols, function)
         hint = similarity_hint(similar)
 
         ParseError.__init__(
             self, _(b"unknown identifier: %s") % function, hint=hint
         )
 
 
 class RepoError(Hint, Exception):
     __bytes__ = _tobytes
 
 
 class RepoLookupError(RepoError):
     pass
 
 
 class FilteredRepoLookupError(RepoLookupError):
     pass
 
 
 class CapabilityError(RepoError):
     pass
 
 
 class RequirementError(RepoError):
     """Exception raised if .hg/requires has an unknown entry."""
 
 
 class StdioError(IOError):
     """Raised if I/O to stdout or stderr fails"""
 
     def __init__(self, err: IOError) -> None:
         IOError.__init__(self, err.errno, err.strerror)
 
     # no __bytes__() because error message is derived from the standard IOError
 
 
 class UnsupportedMergeRecords(Abort):
     def __init__(self, recordtypes: Iterable[bytes]) -> None:
         from .i18n import _
 
         self.recordtypes = sorted(recordtypes)
         s = b' '.join(self.recordtypes)
         Abort.__init__(
             self,
             _(b'unsupported merge state records: %s') % s,
             hint=_(
                 b'see https://mercurial-scm.org/wiki/MergeStateRecords for '
                 b'more information'
             ),
         )
 
 
 class UnknownVersion(Abort):
     """generic exception for aborting from an encounter with an unknown version"""
 
     def __init__(
         self,
         msg: bytes,
         hint: Optional[bytes] = None,
         version: Optional[bytes] = None,
     ) -> None:
         self.version = version
         super(UnknownVersion, self).__init__(msg, hint=hint)
 
 
 class LockError(IOError):
     def __init__(
         self,
         errno: int,
         strerror: str,
         filename: bytes,
         desc: Optional[bytes],
     ) -> None:
         IOError.__init__(self, errno, strerror, filename)
         self.desc = desc
 
     # no __bytes__() because error message is derived from the standard IOError
 
 
 class LockHeld(LockError):
     def __init__(
         self,
         errno: int,
         filename: bytes,
         desc: Optional[bytes],
         locker,
     ):
         LockError.__init__(self, errno, 'Lock held', filename, desc)
         self.filename: bytes = filename
         self.locker = locker
 
 
 class LockUnavailable(LockError):
     pass
 
 
 # LockError is for errors while acquiring the lock -- this is unrelated
 class LockInheritanceContractViolation(RuntimeError):
     __bytes__ = _tobytes
 
 
 class ResponseError(Exception):
     """Raised to print an error with part of output and exit."""
 
     __bytes__ = _tobytes
 
 
 # derived from KeyboardInterrupt to simplify some breakout code
 class SignalInterrupt(KeyboardInterrupt):
     """Exception raised on SIGTERM and SIGHUP."""
 
 
 class SignatureError(Exception):
     __bytes__ = _tobytes
 
 
 class PushRaced(RuntimeError):
     """An exception raised during unbundling that indicate a push race"""
 
     __bytes__ = _tobytes
 
 
 class ProgrammingError(Hint, RuntimeError):
     """Raised if a mercurial (core or extension) developer made a mistake"""
 
     def __init__(self, msg: AnyStr, *args, **kwargs):
         # On Python 3, turn the message back into a string since this is
         # an internal-only error that won't be printed except in a
         # stack traces.
         msg = pycompat.sysstr(msg)
         super(ProgrammingError, self).__init__(msg, *args, **kwargs)
 
     __bytes__ = _tobytes
 
 
 class WdirUnsupported(Exception):
     """An exception which is raised when 'wdir()' is not supported"""
 
     __bytes__ = _tobytes
 
 
 # bundle2 related errors
 class BundleValueError(ValueError):
     """error raised when bundle2 cannot be processed"""
 
     __bytes__ = _tobytes
 
 
 class BundleUnknownFeatureError(BundleValueError):
     def __init__(self, parttype=None, params=(), values=()):
         self.parttype = parttype
         self.params = params
         self.values = values
         if self.parttype is None:
             msg = b'Stream Parameter'
         else:
             msg = parttype
         entries = self.params
         if self.params and self.values:
             assert len(self.params) == len(self.values)
             entries = []
             for idx, par in enumerate(self.params):
                 val = self.values[idx]
                 if val is None:
                     entries.append(val)
                 else:
                     entries.append(b"%s=%r" % (par, pycompat.maybebytestr(val)))
         if entries:
             msg = b'%s - %s' % (msg, b', '.join(entries))
         ValueError.__init__(self, msg)  # TODO: convert to str?
 
 
 class ReadOnlyPartError(RuntimeError):
     """error raised when code tries to alter a part being generated"""
 
     __bytes__ = _tobytes
 
 
 class PushkeyFailed(Abort):
     """error raised when a pushkey part failed to update a value"""
 
     def __init__(
         self, partid, namespace=None, key=None, new=None, old=None, ret=None
     ):
         self.partid = partid
         self.namespace = namespace
         self.key = key
         self.new = new
         self.old = old
         self.ret = ret
         # no i18n expected to be processed into a better message
         Abort.__init__(
             self, b'failed to update value for "%s/%s"' % (namespace, key)
         )
 
 
 class CensoredNodeError(StorageError):
     """error raised when content verification fails on a censored node
 
     Also contains the tombstone data substituted for the uncensored data.
     """
 
     def __init__(self, filename: bytes, node: bytes, tombstone: bytes):
         from .node import short
 
         StorageError.__init__(self, b'%s:%s' % (filename, short(node)))
         self.tombstone = tombstone
 
 
 class CensoredBaseError(StorageError):
     """error raised when a delta is rejected because its base is censored
 
     A delta based on a censored revision must be formed as single patch
     operation which replaces the entire base with new content. This ensures
     the delta may be applied by clones which have not censored the base.
     """
 
 
 class InvalidBundleSpecification(Exception):
     """error raised when a bundle specification is invalid.
 
     This is used for syntax errors as opposed to support errors.
     """
 
     __bytes__ = _tobytes
 
 
 class UnsupportedBundleSpecification(Exception):
     """error raised when a bundle specification is not supported."""
 
     __bytes__ = _tobytes
 
 
 class CorruptedState(Exception):
     """error raised when a command is not able to read its state from file"""
 
     __bytes__ = _tobytes
 
 
 class CorruptedDirstate(Exception):
     """error raised the dirstate appears corrupted on-disk. It may be due to
     a dirstate version mismatch (i.e. expecting v2 and finding v1 on disk)."""
 
     __bytes__ = _tobytes
 
 
 class PeerTransportError(Abort):
     """Transport-level I/O error when communicating with a peer repo."""
 
 
 class InMemoryMergeConflictsError(Exception):
     """Exception raised when merge conflicts arose during an in-memory merge."""
 
     __bytes__ = _tobytes
 
 
 class WireprotoCommandError(Exception):
     """Represents an error during execution of a wire protocol command.
 
     Should only be thrown by wire protocol version 2 commands.
 
     The error is a formatter string and an optional iterable of arguments.
     """
 
     def __init__(
         self,
         message: bytes,
         args: Optional[Sequence[bytes]] = None,
     ) -> None:
         self.message = message
         self.messageargs = args

mercurial/i18n.py

@@ -1,131 +1,131 b''
 # i18n.py - internationalization support 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 gettext as gettextmod
 import locale
 import os
 import sys
 
 from typing import (
     Callable,
     List,
 )
 
 from .utils import resourceutil
 from . import (
     encoding,
     pycompat,
 )
 
 # keeps pyflakes happy
 assert [
     Callable,
     List,
 ]
 
 # modelled after templater.templatepath:
 if getattr(sys, 'frozen', None) is not None:
     module = pycompat.sysexecutable
 else:
     module = pycompat.fsencode(__file__)
 
 _languages = None
 if (
     pycompat.iswindows
     and b'LANGUAGE' not in encoding.environ
     and b'LC_ALL' not in encoding.environ
     and b'LC_MESSAGES' not in encoding.environ
     and b'LANG' not in encoding.environ
 ):
     # Try to detect UI language by "User Interface Language Management" API
     # if no locale variables are set. Note that locale.getdefaultlocale()
     # uses GetLocaleInfo(), which may be different from UI language.
     # (See http://msdn.microsoft.com/en-us/library/dd374098(v=VS.85).aspx )
     try:
         import ctypes
 
         # pytype: disable=module-attr
         langid = ctypes.windll.kernel32.GetUserDefaultUILanguage()
         # pytype: enable=module-attr
 
         _languages = [locale.windows_locale[langid]]
     except (ImportError, AttributeError, KeyError):
         # ctypes not found or unknown langid
         pass
 
 
 datapath = pycompat.fsdecode(resourceutil.datapath)
 localedir = os.path.join(datapath, 'locale')
 t = gettextmod.translation('hg', localedir, _languages, fallback=True)
 try:
     _ugettext = t.ugettext  # pytype: disable=attribute-error
 except AttributeError:
     _ugettext = t.gettext
 
 
 _msgcache = {}  # encoding: {message: translation}
 
 
 def gettext(message: bytes) -> bytes:
     """Translate message.
 
     The message is looked up in the catalog to get a Unicode string,
     which is encoded in the local encoding before being returned.
 
     Important: message is restricted to characters in the encoding
     given by sys.getdefaultencoding() which is most likely 'ascii'.
     """
     # If message is None, t.ugettext will return u'None' as the
     # translation whereas our callers expect us to return None.
     if message is None or not _ugettext:
         return message
 
     cache = _msgcache.setdefault(encoding.encoding, {})
     if message not in cache:
         if type(message) is str:
             # goofy unicode docstrings in test
-            paragraphs = message.split(u'\n\n')  # type: List[str]
+            paragraphs: List[str] = message.split(u'\n\n')
         else:
             # should be ascii, but we have unicode docstrings in test, which
             # are converted to utf-8 bytes on Python 3.
             paragraphs = [p.decode("utf-8") for p in message.split(b'\n\n')]
         # Be careful not to translate the empty string -- it holds the
         # meta data of the .po file.
         u = u'\n\n'.join([p and _ugettext(p) or u'' for p in paragraphs])
         try:
             # encoding.tolocal cannot be used since it will first try to
             # decode the Unicode string. Calling u.decode(enc) really
             # means u.encode(sys.getdefaultencoding()).decode(enc). Since
             # the Python encoding defaults to 'ascii', this fails if the
             # translated string use non-ASCII characters.
             encodingstr = pycompat.sysstr(encoding.encoding)
             cache[message] = u.encode(encodingstr, "replace")
         except LookupError:
             # An unknown encoding results in a LookupError.
             cache[message] = message
     return cache[message]
 
 
 def _plain():
     if (
         b'HGPLAIN' not in encoding.environ
         and b'HGPLAINEXCEPT' not in encoding.environ
     ):
         return False
     exceptions = encoding.environ.get(b'HGPLAINEXCEPT', b'').strip().split(b',')
     return b'i18n' not in exceptions
 
 
 if _plain():
 
     def _(message: bytes) -> bytes:
         return message
 
 
 else:
     _ = gettext

mercurial/pathutil.py

@@ -1,395 +1,395 b''
 import contextlib
 import errno
 import os
 import posixpath
 import stat
 
 from typing import (
     Any,
     Callable,
     Iterator,
     Optional,
 )
 
 from .i18n import _
 from . import (
     encoding,
     error,
     policy,
     pycompat,
     util,
 )
 
 rustdirs = policy.importrust('dirstate', 'Dirs')
 parsers = policy.importmod('parsers')
 
 # keeps pyflakes happy
 assert [
     Any,
     Callable,
     Iterator,
     Optional,
 ]
 
 
 def _lowerclean(s: bytes) -> bytes:
     return encoding.hfsignoreclean(s.lower())
 
 
 class pathauditor:
     """ensure that a filesystem path contains no banned components.
     the following properties of a path are checked:
 
     - ends with a directory separator
     - under top-level .hg
     - starts at the root of a windows drive
     - contains ".."
 
     More check are also done about the file system states:
     - traverses a symlink (e.g. a/symlink_here/b)
     - inside a nested repository (a callback can be used to approve
       some nested repositories, e.g., subrepositories)
 
     The file system checks are only done when 'realfs' is set to True (the
     default). They should be disable then we are auditing path for operation on
     stored history.
 
     If 'cached' is set to True, audited paths and sub-directories are cached.
     Be careful to not keep the cache of unmanaged directories for long because
     audited paths may be replaced with symlinks.
     """
 
     def __init__(self, root, callback=None, realfs=True, cached=False):
         self.audited = set()
         self.auditeddir = dict()
         self.root = root
         self._realfs = realfs
         self._cached = cached
         self.callback = callback
         if os.path.lexists(root) and not util.fscasesensitive(root):
             self.normcase = util.normcase
         else:
             self.normcase = lambda x: x
 
     def __call__(self, path: bytes, mode: Optional[Any] = None) -> None:
         """Check the relative path.
         path may contain a pattern (e.g. foodir/**.txt)"""
 
         path = util.localpath(path)
         if path in self.audited:
             return
         # AIX ignores "/" at end of path, others raise EISDIR.
         if util.endswithsep(path):
             raise error.InputError(
                 _(b"path ends in directory separator: %s") % path
             )
         parts = util.splitpath(path)
         if (
             os.path.splitdrive(path)[0]
             or _lowerclean(parts[0]) in (b'.hg', b'.hg.', b'')
             or pycompat.ospardir in parts
         ):
             raise error.InputError(
                 _(b"path contains illegal component: %s") % path
             )
         # Windows shortname aliases
         if b"~" in path:
             for p in parts:
                 if b"~" in p:
                     first, last = p.split(b"~", 1)
                     if last.isdigit() and first.upper() in [b"HG", b"HG8B6C"]:
                         raise error.InputError(
                             _(b"path contains illegal component: %s") % path
                         )
         if b'.hg' in _lowerclean(path):
             lparts = [_lowerclean(p) for p in parts]
             for p in b'.hg', b'.hg.':
                 if p in lparts[1:]:
                     pos = lparts.index(p)
                     base = os.path.join(*parts[:pos])
                     raise error.InputError(
                         _(b"path '%s' is inside nested repo %r")
                         % (path, pycompat.bytestr(base))
                     )
 
         if self._realfs:
             # It's important that we check the path parts starting from the root.
             # We don't want to add "foo/bar/baz" to auditeddir before checking if
             # there's a "foo/.hg" directory. This also means we won't accidentally
             # traverse a symlink into some other filesystem (which is potentially
             # expensive to access).
             for prefix in finddirs_rev_noroot(path):
                 if prefix in self.auditeddir:
                     res = self.auditeddir[prefix]
                 else:
                     res = pathauditor._checkfs_exists(
                         self.root, prefix, path, self.callback
                     )
                     if self._cached:
                         self.auditeddir[prefix] = res
                 if not res:
                     break
 
         if self._cached:
             self.audited.add(path)
 
     @staticmethod
     def _checkfs_exists(
         root,
         prefix: bytes,
         path: bytes,
         callback: Optional[Callable[[bytes], bool]] = None,
     ):
         """raise exception if a file system backed check fails.
 
         Return a bool that indicates that the directory (or file) exists."""
         curpath = os.path.join(root, prefix)
         try:
             st = os.lstat(curpath)
         except OSError as err:
             if err.errno == errno.ENOENT:
                 return False
             # EINVAL can be raised as invalid path syntax under win32.
             # They must be ignored for patterns can be checked too.
             if err.errno not in (errno.ENOENT, errno.ENOTDIR, errno.EINVAL):
                 raise
         else:
             if stat.S_ISLNK(st.st_mode):
                 msg = _(b'path %r traverses symbolic link %r') % (
                     pycompat.bytestr(path),
                     pycompat.bytestr(prefix),
                 )
                 raise error.Abort(msg)
             elif stat.S_ISDIR(st.st_mode) and os.path.isdir(
                 os.path.join(curpath, b'.hg')
             ):
                 if not callback or not callback(curpath):
                     msg = _(b"path '%s' is inside nested repo %r")
                     raise error.Abort(msg % (path, pycompat.bytestr(prefix)))
         return True
 
     def check(self, path: bytes) -> bool:
         try:
             self(path)
             return True
         except (OSError, error.Abort):
             return False
 
     @contextlib.contextmanager
     def cached(self):
         if self._cached:
             yield
         else:
             try:
                 self._cached = True
                 yield
             finally:
                 self.audited.clear()
                 self.auditeddir.clear()
                 self._cached = False
 
 
 def canonpath(
     root: bytes,
     cwd: bytes,
     myname: bytes,
     auditor: Optional[pathauditor] = None,
 ) -> bytes:
     """return the canonical path of myname, given cwd and root
 
     >>> def check(root, cwd, myname):
     ...     a = pathauditor(root, realfs=False)
     ...     try:
     ...         return canonpath(root, cwd, myname, a)
     ...     except error.Abort:
     ...         return 'aborted'
     >>> def unixonly(root, cwd, myname, expected='aborted'):
     ...     if pycompat.iswindows:
     ...         return expected
     ...     return check(root, cwd, myname)
     >>> def winonly(root, cwd, myname, expected='aborted'):
     ...     if not pycompat.iswindows:
     ...         return expected
     ...     return check(root, cwd, myname)
     >>> winonly(b'd:\\\\repo', b'c:\\\\dir', b'filename')
     'aborted'
     >>> winonly(b'c:\\\\repo', b'c:\\\\dir', b'filename')
     'aborted'
     >>> winonly(b'c:\\\\repo', b'c:\\\\', b'filename')
     'aborted'
     >>> winonly(b'c:\\\\repo', b'c:\\\\', b'repo\\\\filename',
     ...         b'filename')
     'filename'
     >>> winonly(b'c:\\\\repo', b'c:\\\\repo', b'filename', b'filename')
     'filename'
     >>> winonly(b'c:\\\\repo', b'c:\\\\repo\\\\subdir', b'filename',
     ...         b'subdir/filename')
     'subdir/filename'
     >>> unixonly(b'/repo', b'/dir', b'filename')
     'aborted'
     >>> unixonly(b'/repo', b'/', b'filename')
     'aborted'
     >>> unixonly(b'/repo', b'/', b'repo/filename', b'filename')
     'filename'
     >>> unixonly(b'/repo', b'/repo', b'filename', b'filename')
     'filename'
     >>> unixonly(b'/repo', b'/repo/subdir', b'filename', b'subdir/filename')
     'subdir/filename'
     """
     if util.endswithsep(root):
         rootsep = root
     else:
         rootsep = root + pycompat.ossep
     name = myname
     if not os.path.isabs(name):
         name = os.path.join(root, cwd, name)
     name = os.path.normpath(name)
     if auditor is None:
         auditor = pathauditor(root)
     if name != rootsep and name.startswith(rootsep):
         name = name[len(rootsep) :]
         auditor(name)
         return util.pconvert(name)
     elif name == root:
         return b''
     else:
         # Determine whether `name' is in the hierarchy at or beneath `root',
         # by iterating name=dirname(name) until that causes no change (can't
         # check name == '/', because that doesn't work on windows). The list
         # `rel' holds the reversed list of components making up the relative
         # file name we want.
         rel = []
         while True:
             try:
                 s = util.samefile(name, root)
             except OSError:
                 s = False
             if s:
                 if not rel:
                     # name was actually the same as root (maybe a symlink)
                     return b''
                 rel.reverse()
                 name = os.path.join(*rel)
                 auditor(name)
                 return util.pconvert(name)
             dirname, basename = util.split(name)
             rel.append(basename)
             if dirname == name:
                 break
             name = dirname
 
         # A common mistake is to use -R, but specify a file relative to the repo
         # instead of cwd.  Detect that case, and provide a hint to the user.
         hint = None
         try:
             if cwd != root:
                 canonpath(root, root, myname, auditor)
                 relpath = util.pathto(root, cwd, b'')
                 if relpath.endswith(pycompat.ossep):
                     relpath = relpath[:-1]
                 hint = _(b"consider using '--cwd %s'") % relpath
         except error.Abort:
             pass
 
         raise error.Abort(
             _(b"%s not under root '%s'") % (myname, root), hint=hint
         )
 
 
 def normasprefix(path: bytes) -> bytes:
     """normalize the specified path as path prefix
 
     Returned value can be used safely for "p.startswith(prefix)",
     "p[len(prefix):]", and so on.
 
     For efficiency, this expects "path" argument to be already
     normalized by "os.path.normpath", "os.path.realpath", and so on.
 
     See also issue3033 for detail about need of this function.
 
     >>> normasprefix(b'/foo/bar').replace(pycompat.ossep, b'/')
     '/foo/bar/'
     >>> normasprefix(b'/').replace(pycompat.ossep, b'/')
     '/'
     """
     d, p = os.path.splitdrive(path)
     if len(p) != len(pycompat.ossep):
         return path + pycompat.ossep
     else:
         return path
 
 
 def finddirs(path: bytes) -> Iterator[bytes]:
     pos = path.rfind(b'/')
     while pos != -1:
         yield path[:pos]
         pos = path.rfind(b'/', 0, pos)
     yield b''
 
 
 def finddirs_rev_noroot(path: bytes) -> Iterator[bytes]:
     pos = path.find(pycompat.ossep)
     while pos != -1:
         yield path[:pos]
         pos = path.find(pycompat.ossep, pos + 1)
 
 
 class dirs:
     '''a multiset of directory names from a set of file paths'''
 
     def __init__(self, map, only_tracked=False):
         """
         a dict map indicates a dirstate while a list indicates a manifest
         """
         self._dirs = {}
         addpath = self.addpath
         if isinstance(map, dict) and only_tracked:
             for f, s in map.items():
                 if s.state != b'r':
                     addpath(f)
         elif only_tracked:
             msg = b"`only_tracked` is only supported with a dict source"
             raise error.ProgrammingError(msg)
         else:
             for f in map:
                 addpath(f)
 
     def addpath(self, path: bytes) -> None:
         dirs = self._dirs
         for base in finddirs(path):
             if base.endswith(b'/'):
                 raise ValueError(
                     "found invalid consecutive slashes in path: %r" % base
                 )
             if base in dirs:
                 dirs[base] += 1
                 return
             dirs[base] = 1
 
     def delpath(self, path: bytes) -> None:
         dirs = self._dirs
         for base in finddirs(path):
             if dirs[base] > 1:
                 dirs[base] -= 1
                 return
             del dirs[base]
 
     def __iter__(self):
         return iter(self._dirs)
 
     def __contains__(self, d: bytes) -> bool:
         return d in self._dirs
 
 
 if hasattr(parsers, 'dirs'):
     dirs = parsers.dirs
 
 if rustdirs is not None:
     dirs = rustdirs
 
 
 # forward two methods from posixpath that do what we need, but we'd
 # rather not let our internals know that we're thinking in posix terms
 # - instead we'll let them be oblivious.
 join = posixpath.join
-dirname = posixpath.dirname  # type: Callable[[bytes], bytes]
+dirname: Callable[[bytes], bytes] = posixpath.dirname

mercurial/phases.py

@@ -1,1004 +1,1006 b''
 """ Mercurial phases support code
 
     ---
 
     Copyright 2011 Pierre-Yves David <[e-mail]>
                    Logilab SA        <[e-mail]>
                    Augie Fackler     <[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.
 
     ---
 
 This module implements most phase logic in mercurial.
 
 
 Basic Concept
 =============
 
 A 'changeset phase' is an indicator that tells us how a changeset is
 manipulated and communicated. The details of each phase is described
 below, here we describe the properties they have in common.
 
 Like bookmarks, phases are not stored in history and thus are not
 permanent and leave no audit trail.
 
 First, no changeset can be in two phases at once. Phases are ordered,
 so they can be considered from lowest to highest. The default, lowest
 phase is 'public' - this is the normal phase of existing changesets. A
 child changeset can not be in a lower phase than its parents.
 
 These phases share a hierarchy of traits:
 
             immutable shared
     public:     X        X
     draft:               X
     secret:
 
 Local commits are draft by default.
 
 Phase Movement and Exchange
 ===========================
 
 Phase data is exchanged by pushkey on pull and push. Some servers have
 a publish option set, we call such a server a "publishing server".
 Pushing a draft changeset to a publishing server changes the phase to
 public.
 
 A small list of fact/rules define the exchange of phase:
 
 * old client never changes server states
 * pull never changes server states
 * publish and old server changesets are seen as public by client
 * any secret changeset seen in another repository is lowered to at
   least draft
 
 Here is the final table summing up the 49 possible use cases of phase
 exchange:
 
                            server
                   old     publish      non-publish
                  N   X    N   D   P    N   D   P
     old client
     pull
      N           -   X/X  -   X/D X/P  -   X/D X/P
      X           -   X/X  -   X/D X/P  -   X/D X/P
     push
      X           X/X X/X  X/P X/P X/P  X/D X/D X/P
     new client
     pull
      N           -   P/X  -   P/D P/P  -   D/D P/P
      D           -   P/X  -   P/D P/P  -   D/D P/P
      P           -   P/X  -   P/D P/P  -   P/D P/P
     push
      D           P/X P/X  P/P P/P P/P  D/D D/D P/P
      P           P/X P/X  P/P P/P P/P  P/P P/P P/P
 
 Legend:
 
     A/B = final state on client / state on server
 
     * N = new/not present,
     * P = public,
     * D = draft,
     * X = not tracked (i.e., the old client or server has no internal
           way of recording the phase.)
 
     passive = only pushes
 
 
     A cell here can be read like this:
 
     "When a new client pushes a draft changeset (D) to a publishing
     server where it's not present (N), it's marked public on both
     sides (P/P)."
 
 Note: old client behave as a publishing server with draft only content
 - other people see it as public
 - content is pushed as draft
 
 """
 
 
 import struct
 import typing
 
 from typing import (
     Any,
     Callable,
     Dict,
     Iterable,
     List,
     Optional,
     Set,
     Tuple,
 )
 
 from .i18n import _
 from .node import (
     bin,
     hex,
     nullrev,
     short,
     wdirrev,
 )
 from . import (
     error,
     pycompat,
     requirements,
     smartset,
     txnutil,
     util,
 )
 
 # keeps pyflakes happy
 assert [
     Any,
     Callable,
     Dict,
     Iterable,
     List,
     Optional,
     Set,
     Tuple,
 ]
 
 Phaseroots = Dict[int, Set[bytes]]
 
 if typing.TYPE_CHECKING:
     from . import (
         localrepo,
         ui as uimod,
     )
 
     # keeps pyflakes happy
     assert [uimod]
 
     Phasedefaults = List[
         Callable[[localrepo.localrepository, Phaseroots], Phaseroots]
     ]
 
 
 _fphasesentry = struct.Struct(b'>i20s')
 
 # record phase index
-public, draft, secret = range(3)  # type: int
+public: int = 0
+draft: int = 1
+secret: int = 2
 archived = 32  # non-continuous for compatibility
 internal = 96  # non-continuous for compatibility
 allphases = (public, draft, secret, archived, internal)
 trackedphases = (draft, secret, archived, internal)
 not_public_phases = trackedphases
 # record phase names
 cmdphasenames = [b'public', b'draft', b'secret']  # known to `hg phase` command
 phasenames = dict(enumerate(cmdphasenames))
 phasenames[archived] = b'archived'
 phasenames[internal] = b'internal'
 # map phase name to phase number
 phasenumber = {name: phase for phase, name in phasenames.items()}
 # like phasenumber, but also include maps for the numeric and binary
 # phase number to the phase number
 phasenumber2 = phasenumber.copy()
 phasenumber2.update({phase: phase for phase in phasenames})
 phasenumber2.update({b'%i' % phase: phase for phase in phasenames})
 # record phase property
 mutablephases = (draft, secret, archived, internal)
 relevant_mutable_phases = (draft, secret)  # could be obsolete or unstable
 remotehiddenphases = (secret, archived, internal)
 localhiddenphases = (internal, archived)
 
 all_internal_phases = tuple(p for p in allphases if p & internal)
 # We do not want any internal content to exit the repository, ever.
 no_bundle_phases = all_internal_phases
 
 
 def supportinternal(repo: "localrepo.localrepository") -> bool:
     """True if the internal phase can be used on a repository"""
     return requirements.INTERNAL_PHASE_REQUIREMENT in repo.requirements
 
 
 def supportarchived(repo: "localrepo.localrepository") -> bool:
     """True if the archived phase can be used on a repository"""
     return requirements.ARCHIVED_PHASE_REQUIREMENT in repo.requirements
 
 
 def _readroots(
     repo: "localrepo.localrepository",
     phasedefaults: Optional["Phasedefaults"] = None,
 ) -> Tuple[Phaseroots, bool]:
     """Read phase roots from disk
 
     phasedefaults is a list of fn(repo, roots) callable, which are
     executed if the phase roots file does not exist. When phases are
     being initialized on an existing repository, this could be used to
     set selected changesets phase to something else than public.
 
     Return (roots, dirty) where dirty is true if roots differ from
     what is being stored.
     """
     repo = repo.unfiltered()
     dirty = False
     roots = {i: set() for i in allphases}
     try:
         f, pending = txnutil.trypending(repo.root, repo.svfs, b'phaseroots')
         try:
             for line in f:
                 phase, nh = line.split()
                 roots[int(phase)].add(bin(nh))
         finally:
             f.close()
     except FileNotFoundError:
         if phasedefaults:
             for f in phasedefaults:
                 roots = f(repo, roots)
         dirty = True
     return roots, dirty
 
 
 def binaryencode(phasemapping: Dict[int, List[bytes]]) -> bytes:
     """encode a 'phase -> nodes' mapping into a binary stream
 
     The revision lists are encoded as (phase, root) pairs.
     """
     binarydata = []
     for phase, nodes in phasemapping.items():
         for head in nodes:
             binarydata.append(_fphasesentry.pack(phase, head))
     return b''.join(binarydata)
 
 
 def binarydecode(stream) -> Dict[int, List[bytes]]:
     """decode a binary stream into a 'phase -> nodes' mapping
 
     The (phase, root) pairs are turned back into a dictionary with
     the phase as index and the aggregated roots of that phase as value."""
     headsbyphase = {i: [] for i in allphases}
     entrysize = _fphasesentry.size
     while True:
         entry = stream.read(entrysize)
         if len(entry) < entrysize:
             if entry:
                 raise error.Abort(_(b'bad phase-heads stream'))
             break
         phase, node = _fphasesentry.unpack(entry)
         headsbyphase[phase].append(node)
     return headsbyphase
 
 
 def _sortedrange_insert(data, idx, rev, t):
     merge_before = False
     if idx:
         r1, t1 = data[idx - 1]
         merge_before = r1[-1] + 1 == rev and t1 == t
     merge_after = False
     if idx < len(data):
         r2, t2 = data[idx]
         merge_after = r2[0] == rev + 1 and t2 == t
 
     if merge_before and merge_after:
         data[idx - 1] = (range(r1[0], r2[-1] + 1), t)
         data.pop(idx)
     elif merge_before:
         data[idx - 1] = (range(r1[0], rev + 1), t)
     elif merge_after:
         data[idx] = (range(rev, r2[-1] + 1), t)
     else:
         data.insert(idx, (range(rev, rev + 1), t))
 
 
 def _sortedrange_split(data, idx, rev, t):
     r1, t1 = data[idx]
     if t == t1:
         return
     t = (t1[0], t[1])
     if len(r1) == 1:
         data.pop(idx)
         _sortedrange_insert(data, idx, rev, t)
     elif r1[0] == rev:
         data[idx] = (range(rev + 1, r1[-1] + 1), t1)
         _sortedrange_insert(data, idx, rev, t)
     elif r1[-1] == rev:
         data[idx] = (range(r1[0], rev), t1)
         _sortedrange_insert(data, idx + 1, rev, t)
     else:
         data[idx : idx + 1] = [
             (range(r1[0], rev), t1),
             (range(rev, rev + 1), t),
             (range(rev + 1, r1[-1] + 1), t1),
         ]
 
 
 def _trackphasechange(data, rev, old, new):
     """add a phase move to the <data> list of ranges
 
     If data is None, nothing happens.
     """
     if data is None:
         return
 
     # If data is empty, create a one-revision range and done
     if not data:
         data.insert(0, (range(rev, rev + 1), (old, new)))
         return
 
     low = 0
     high = len(data)
     t = (old, new)
     while low < high:
         mid = (low + high) // 2
         revs = data[mid][0]
         revs_low = revs[0]
         revs_high = revs[-1]
 
         if rev >= revs_low and rev <= revs_high:
             _sortedrange_split(data, mid, rev, t)
             return
 
         if revs_low == rev + 1:
             if mid and data[mid - 1][0][-1] == rev:
                 _sortedrange_split(data, mid - 1, rev, t)
             else:
                 _sortedrange_insert(data, mid, rev, t)
             return
 
         if revs_high == rev - 1:
             if mid + 1 < len(data) and data[mid + 1][0][0] == rev:
                 _sortedrange_split(data, mid + 1, rev, t)
             else:
                 _sortedrange_insert(data, mid + 1, rev, t)
             return
 
         if revs_low > rev:
             high = mid
         else:
             low = mid + 1
 
     if low == len(data):
         data.append((range(rev, rev + 1), t))
         return
 
     r1, t1 = data[low]
     if r1[0] > rev:
         data.insert(low, (range(rev, rev + 1), t))
     else:
         data.insert(low + 1, (range(rev, rev + 1), t))
 
 
 class phasecache:
     def __init__(
         self,
         repo: "localrepo.localrepository",
         phasedefaults: Optional["Phasedefaults"],
         _load: bool = True,
     ):
         if _load:
             # Cheap trick to allow shallow-copy without copy module
             self.phaseroots, self.dirty = _readroots(repo, phasedefaults)
             self._loadedrevslen = 0
             self._phasesets = None
             self.filterunknown(repo)
             self.opener = repo.svfs
 
     def hasnonpublicphases(self, repo: "localrepo.localrepository") -> bool:
         """detect if there are revisions with non-public phase"""
         repo = repo.unfiltered()
         cl = repo.changelog
         if len(cl) >= self._loadedrevslen:
             self.invalidate()
             self.loadphaserevs(repo)
         return any(
             revs for phase, revs in self.phaseroots.items() if phase != public
         )
 
     def nonpublicphaseroots(
         self, repo: "localrepo.localrepository"
     ) -> Set[bytes]:
         """returns the roots of all non-public phases
 
         The roots are not minimized, so if the secret revisions are
         descendants of draft revisions, their roots will still be present.
         """
         repo = repo.unfiltered()
         cl = repo.changelog
         if len(cl) >= self._loadedrevslen:
             self.invalidate()
             self.loadphaserevs(repo)
         return set().union(
             *[
                 revs
                 for phase, revs in self.phaseroots.items()
                 if phase != public
             ]
         )
 
     def getrevset(
         self,
         repo: "localrepo.localrepository",
         phases: Iterable[int],
         subset: Optional[Any] = None,
     ) -> Any:
         # TODO: finish typing this
         """return a smartset for the given phases"""
         self.loadphaserevs(repo)  # ensure phase's sets are loaded
         phases = set(phases)
         publicphase = public in phases
 
         if publicphase:
             # In this case, phases keeps all the *other* phases.
             phases = set(allphases).difference(phases)
             if not phases:
                 return smartset.fullreposet(repo)
 
         # fast path: _phasesets contains the interesting sets,
         # might only need a union and post-filtering.
         revsneedscopy = False
         if len(phases) == 1:
             [p] = phases
             revs = self._phasesets[p]
             revsneedscopy = True  # Don't modify _phasesets
         else:
             # revs has the revisions in all *other* phases.
             revs = set.union(*[self._phasesets[p] for p in phases])
 
         def _addwdir(wdirsubset, wdirrevs):
             if wdirrev in wdirsubset and repo[None].phase() in phases:
                 if revsneedscopy:
                     wdirrevs = wdirrevs.copy()
                 # The working dir would never be in the # cache, but it was in
                 # the subset being filtered for its phase (or filtered out,
                 # depending on publicphase), so add it to the output to be
                 # included (or filtered out).
                 wdirrevs.add(wdirrev)
             return wdirrevs
 
         if not publicphase:
             if repo.changelog.filteredrevs:
                 revs = revs - repo.changelog.filteredrevs
 
             if subset is None:
                 return smartset.baseset(revs)
             else:
                 revs = _addwdir(subset, revs)
                 return subset & smartset.baseset(revs)
         else:
             if subset is None:
                 subset = smartset.fullreposet(repo)
 
             revs = _addwdir(subset, revs)
 
             if not revs:
                 return subset
             return subset.filter(lambda r: r not in revs)
 
     def copy(self):
         # Shallow copy meant to ensure isolation in
         # advance/retractboundary(), nothing more.
         ph = self.__class__(None, None, _load=False)
         ph.phaseroots = self.phaseroots.copy()
         ph.dirty = self.dirty
         ph.opener = self.opener
         ph._loadedrevslen = self._loadedrevslen
         ph._phasesets = self._phasesets
         return ph
 
     def replace(self, phcache):
         """replace all values in 'self' with content of phcache"""
         for a in (
             'phaseroots',
             'dirty',
             'opener',
             '_loadedrevslen',
             '_phasesets',
         ):
             setattr(self, a, getattr(phcache, a))
 
     def _getphaserevsnative(self, repo):
         repo = repo.unfiltered()
         return repo.changelog.computephases(self.phaseroots)
 
     def _computephaserevspure(self, repo):
         repo = repo.unfiltered()
         cl = repo.changelog
         self._phasesets = {phase: set() for phase in allphases}
         lowerroots = set()
         for phase in reversed(trackedphases):
             roots = pycompat.maplist(cl.rev, self.phaseroots[phase])
             if roots:
                 ps = set(cl.descendants(roots))
                 for root in roots:
                     ps.add(root)
                 ps.difference_update(lowerroots)
                 lowerroots.update(ps)
                 self._phasesets[phase] = ps
         self._loadedrevslen = len(cl)
 
     def loadphaserevs(self, repo: "localrepo.localrepository") -> None:
         """ensure phase information is loaded in the object"""
         if self._phasesets is None:
             try:
                 res = self._getphaserevsnative(repo)
                 self._loadedrevslen, self._phasesets = res
             except AttributeError:
                 self._computephaserevspure(repo)
 
     def invalidate(self):
         self._loadedrevslen = 0
         self._phasesets = None
 
     def phase(self, repo: "localrepo.localrepository", rev: int) -> int:
         # We need a repo argument here to be able to build _phasesets
         # if necessary. The repository instance is not stored in
         # phasecache to avoid reference cycles. The changelog instance
         # is not stored because it is a filecache() property and can
         # be replaced without us being notified.
         if rev == nullrev:
             return public
         if rev < nullrev:
             raise ValueError(_(b'cannot lookup negative revision'))
         if rev >= self._loadedrevslen:
             self.invalidate()
             self.loadphaserevs(repo)
         for phase in trackedphases:
             if rev in self._phasesets[phase]:
                 return phase
         return public
 
     def write(self):
         if not self.dirty:
             return
         f = self.opener(b'phaseroots', b'w', atomictemp=True, checkambig=True)
         try:
             self._write(f)
         finally:
             f.close()
 
     def _write(self, fp):
         for phase, roots in self.phaseroots.items():
             for h in sorted(roots):
                 fp.write(b'%i %s\n' % (phase, hex(h)))
         self.dirty = False
 
     def _updateroots(self, phase, newroots, tr):
         self.phaseroots[phase] = newroots
         self.invalidate()
         self.dirty = True
 
         tr.addfilegenerator(b'phase', (b'phaseroots',), self._write)
         tr.hookargs[b'phases_moved'] = b'1'
 
     def registernew(self, repo, tr, targetphase, revs):
         repo = repo.unfiltered()
         self._retractboundary(repo, tr, targetphase, [], revs=revs)
         if tr is not None and b'phases' in tr.changes:
             phasetracking = tr.changes[b'phases']
             phase = self.phase
             for rev in sorted(revs):
                 revphase = phase(repo, rev)
                 _trackphasechange(phasetracking, rev, None, revphase)
         repo.invalidatevolatilesets()
 
     def advanceboundary(
         self, repo, tr, targetphase, nodes, revs=None, dryrun=None
     ):
         """Set all 'nodes' to phase 'targetphase'
 
         Nodes with a phase lower than 'targetphase' are not affected.
 
         If dryrun is True, no actions will be performed
 
         Returns a set of revs whose phase is changed or should be changed
         """
         # Be careful to preserve shallow-copied values: do not update
         # phaseroots values, replace them.
         if revs is None:
             revs = []
         if tr is None:
             phasetracking = None
         else:
             phasetracking = tr.changes.get(b'phases')
 
         repo = repo.unfiltered()
         revs = [repo[n].rev() for n in nodes] + [r for r in revs]
 
         changes = set()  # set of revisions to be changed
         delroots = []  # set of root deleted by this path
         for phase in (phase for phase in allphases if phase > targetphase):
             # filter nodes that are not in a compatible phase already
             revs = [rev for rev in revs if self.phase(repo, rev) >= phase]
             if not revs:
                 break  # no roots to move anymore
 
             olds = self.phaseroots[phase]
 
             affected = repo.revs(b'%ln::%ld', olds, revs)
             changes.update(affected)
             if dryrun:
                 continue
             for r in affected:
                 _trackphasechange(
                     phasetracking, r, self.phase(repo, r), targetphase
                 )
 
             roots = {
                 ctx.node()
                 for ctx in repo.set(b'roots((%ln::) - %ld)', olds, affected)
             }
             if olds != roots:
                 self._updateroots(phase, roots, tr)
                 # some roots may need to be declared for lower phases
                 delroots.extend(olds - roots)
         if not dryrun:
             # declare deleted root in the target phase
             if targetphase != 0:
                 self._retractboundary(repo, tr, targetphase, delroots)
             repo.invalidatevolatilesets()
         return changes
 
     def retractboundary(self, repo, tr, targetphase, nodes):
         oldroots = {
             phase: revs
             for phase, revs in self.phaseroots.items()
             if phase <= targetphase
         }
         if tr is None:
             phasetracking = None
         else:
             phasetracking = tr.changes.get(b'phases')
         repo = repo.unfiltered()
         if (
             self._retractboundary(repo, tr, targetphase, nodes)
             and phasetracking is not None
         ):
 
             # find the affected revisions
             new = self.phaseroots[targetphase]
             old = oldroots[targetphase]
             affected = set(repo.revs(b'(%ln::) - (%ln::)', new, old))
 
             # find the phase of the affected revision
             for phase in range(targetphase, -1, -1):
                 if phase:
                     roots = oldroots.get(phase, [])
                     revs = set(repo.revs(b'%ln::%ld', roots, affected))
                     affected -= revs
                 else:  # public phase
                     revs = affected
                 for r in sorted(revs):
                     _trackphasechange(phasetracking, r, phase, targetphase)
         repo.invalidatevolatilesets()
 
     def _retractboundary(self, repo, tr, targetphase, nodes, revs=None):
         # Be careful to preserve shallow-copied values: do not update
         # phaseroots values, replace them.
         if revs is None:
             revs = []
         if (
             targetphase == internal
             and not supportinternal(repo)
             or targetphase == archived
             and not supportarchived(repo)
         ):
             name = phasenames[targetphase]
             msg = b'this repository does not support the %s phase' % name
             raise error.ProgrammingError(msg)
 
         repo = repo.unfiltered()
         torev = repo.changelog.rev
         tonode = repo.changelog.node
         currentroots = {torev(node) for node in self.phaseroots[targetphase]}
         finalroots = oldroots = set(currentroots)
         newroots = [torev(node) for node in nodes] + [r for r in revs]
         newroots = [
             rev for rev in newroots if self.phase(repo, rev) < targetphase
         ]
 
         if newroots:
             if nullrev in newroots:
                 raise error.Abort(_(b'cannot change null revision phase'))
             currentroots.update(newroots)
 
             # Only compute new roots for revs above the roots that are being
             # retracted.
             minnewroot = min(newroots)
             aboveroots = [rev for rev in currentroots if rev >= minnewroot]
             updatedroots = repo.revs(b'roots(%ld::)', aboveroots)
 
             finalroots = {rev for rev in currentroots if rev < minnewroot}
             finalroots.update(updatedroots)
         if finalroots != oldroots:
             self._updateroots(
                 targetphase, {tonode(rev) for rev in finalroots}, tr
             )
             return True
         return False
 
     def filterunknown(self, repo: "localrepo.localrepository") -> None:
         """remove unknown nodes from the phase boundary
 
         Nothing is lost as unknown nodes only hold data for their descendants.
         """
         filtered = False
         has_node = repo.changelog.index.has_node  # to filter unknown nodes
         for phase, nodes in self.phaseroots.items():
             missing = sorted(node for node in nodes if not has_node(node))
             if missing:
                 for mnode in missing:
                     repo.ui.debug(
                         b'removing unknown node %s from %i-phase boundary\n'
                         % (short(mnode), phase)
                     )
                 nodes.symmetric_difference_update(missing)
                 filtered = True
         if filtered:
             self.dirty = True
         # filterunknown is called by repo.destroyed, we may have no changes in
         # root but _phasesets contents is certainly invalid (or at least we
         # have not proper way to check that). related to issue 3858.
         #
         # The other caller is __init__ that have no _phasesets initialized
         # anyway. If this change we should consider adding a dedicated
         # "destroyed" function to phasecache or a proper cache key mechanism
         # (see branchmap one)
         self.invalidate()
 
 
 def advanceboundary(repo, tr, targetphase, nodes, revs=None, dryrun=None):
     """Add nodes to a phase changing other nodes phases if necessary.
 
     This function move boundary *forward* this means that all nodes
     are set in the target phase or kept in a *lower* phase.
 
     Simplify boundary to contains phase roots only.
 
     If dryrun is True, no actions will be performed
 
     Returns a set of revs whose phase is changed or should be changed
     """
     if revs is None:
         revs = []
     phcache = repo._phasecache.copy()
     changes = phcache.advanceboundary(
         repo, tr, targetphase, nodes, revs=revs, dryrun=dryrun
     )
     if not dryrun:
         repo._phasecache.replace(phcache)
     return changes
 
 
 def retractboundary(repo, tr, targetphase, nodes):
     """Set nodes back to a phase changing other nodes phases if
     necessary.
 
     This function move boundary *backward* this means that all nodes
     are set in the target phase or kept in a *higher* phase.
 
     Simplify boundary to contains phase roots only."""
     phcache = repo._phasecache.copy()
     phcache.retractboundary(repo, tr, targetphase, nodes)
     repo._phasecache.replace(phcache)
 
 
 def registernew(repo, tr, targetphase, revs):
     """register a new revision and its phase
 
     Code adding revisions to the repository should use this function to
     set new changeset in their target phase (or higher).
     """
     phcache = repo._phasecache.copy()
     phcache.registernew(repo, tr, targetphase, revs)
     repo._phasecache.replace(phcache)
 
 
 def listphases(repo: "localrepo.localrepository") -> Dict[bytes, bytes]:
     """List phases root for serialization over pushkey"""
     # Use ordered dictionary so behavior is deterministic.
     keys = util.sortdict()
     value = b'%i' % draft
     cl = repo.unfiltered().changelog
     for root in repo._phasecache.phaseroots[draft]:
         if repo._phasecache.phase(repo, cl.rev(root)) <= draft:
             keys[hex(root)] = value
 
     if repo.publishing():
         # Add an extra data to let remote know we are a publishing
         # repo. Publishing repo can't just pretend they are old repo.
         # When pushing to a publishing repo, the client still need to
         # push phase boundary
         #
         # Push do not only push changeset. It also push phase data.
         # New phase data may apply to common changeset which won't be
         # push (as they are common). Here is a very simple example:
         #
         # 1) repo A push changeset X as draft to repo B
         # 2) repo B make changeset X public
         # 3) repo B push to repo A. X is not pushed but the data that
         #    X as now public should
         #
         # The server can't handle it on it's own as it has no idea of
         # client phase data.
         keys[b'publishing'] = b'True'
     return keys
 
 
 def pushphase(
     repo: "localrepo.localrepository",
     nhex: bytes,
     oldphasestr: bytes,
     newphasestr: bytes,
 ) -> bool:
     """List phases root for serialization over pushkey"""
     repo = repo.unfiltered()
     with repo.lock():
         currentphase = repo[nhex].phase()
         newphase = abs(int(newphasestr))  # let's avoid negative index surprise
         oldphase = abs(int(oldphasestr))  # let's avoid negative index surprise
         if currentphase == oldphase and newphase < oldphase:
             with repo.transaction(b'pushkey-phase') as tr:
                 advanceboundary(repo, tr, newphase, [bin(nhex)])
             return True
         elif currentphase == newphase:
             # raced, but got correct result
             return True
         else:
             return False
 
 
 def subsetphaseheads(repo, subset):
     """Finds the phase heads for a subset of a history
 
     Returns a list indexed by phase number where each item is a list of phase
     head nodes.
     """
     cl = repo.changelog
 
     headsbyphase = {i: [] for i in allphases}
     for phase in allphases:
         revset = b"heads(%%ln & _phase(%d))" % phase
         headsbyphase[phase] = [cl.node(r) for r in repo.revs(revset, subset)]
     return headsbyphase
 
 
 def updatephases(repo, trgetter, headsbyphase):
     """Updates the repo with the given phase heads"""
     # Now advance phase boundaries of all phases
     #
     # run the update (and fetch transaction) only if there are actually things
     # to update. This avoid creating empty transaction during no-op operation.
 
     for phase in allphases:
         revset = b'%ln - _phase(%s)'
         heads = [c.node() for c in repo.set(revset, headsbyphase[phase], phase)]
         if heads:
             advanceboundary(repo, trgetter(), phase, heads)
 
 
 def analyzeremotephases(repo, subset, roots):
     """Compute phases heads and root in a subset of node from root dict
 
     * subset is heads of the subset
     * roots is {<nodeid> => phase} mapping. key and value are string.
 
     Accept unknown element input
     """
     repo = repo.unfiltered()
     # build list from dictionary
     draftroots = []
     has_node = repo.changelog.index.has_node  # to filter unknown nodes
     for nhex, phase in roots.items():
         if nhex == b'publishing':  # ignore data related to publish option
             continue
         node = bin(nhex)
         phase = int(phase)
         if phase == public:
             if node != repo.nullid:
                 repo.ui.warn(
                     _(
                         b'ignoring inconsistent public root'
                         b' from remote: %s\n'
                     )
                     % nhex
                 )
         elif phase == draft:
             if has_node(node):
                 draftroots.append(node)
         else:
             repo.ui.warn(
                 _(b'ignoring unexpected root from remote: %i %s\n')
                 % (phase, nhex)
             )
     # compute heads
     publicheads = newheads(repo, subset, draftroots)
     return publicheads, draftroots
 
 
 class remotephasessummary:
     """summarize phase information on the remote side
 
     :publishing: True is the remote is publishing
     :publicheads: list of remote public phase heads (nodes)
     :draftheads: list of remote draft phase heads (nodes)
     :draftroots: list of remote draft phase root (nodes)
     """
 
     def __init__(self, repo, remotesubset, remoteroots):
         unfi = repo.unfiltered()
         self._allremoteroots = remoteroots
 
         self.publishing = remoteroots.get(b'publishing', False)
 
         ana = analyzeremotephases(repo, remotesubset, remoteroots)
         self.publicheads, self.draftroots = ana
         # Get the list of all "heads" revs draft on remote
         dheads = unfi.set(b'heads(%ln::%ln)', self.draftroots, remotesubset)
         self.draftheads = [c.node() for c in dheads]
 
 
 def newheads(repo, heads, roots):
     """compute new head of a subset minus another
 
     * `heads`: define the first subset
     * `roots`: define the second we subtract from the first"""
     # prevent an import cycle
     # phases > dagop > patch > copies > scmutil > obsolete > obsutil > phases
     from . import dagop
 
     repo = repo.unfiltered()
     cl = repo.changelog
     rev = cl.index.get_rev
     if not roots:
         return heads
     if not heads or heads == [repo.nullid]:
         return []
     # The logic operated on revisions, convert arguments early for convenience
     new_heads = {rev(n) for n in heads if n != repo.nullid}
     roots = [rev(n) for n in roots]
     # compute the area we need to remove
     affected_zone = repo.revs(b"(%ld::%ld)", roots, new_heads)
     # heads in the area are no longer heads
     new_heads.difference_update(affected_zone)
     # revisions in the area have children outside of it,
     # They might be new heads
     candidates = repo.revs(
         b"parents(%ld + (%ld and merge())) and not null", roots, affected_zone
     )
     candidates -= affected_zone
     if new_heads or candidates:
         # remove candidate that are ancestors of other heads
         new_heads.update(candidates)
         prunestart = repo.revs(b"parents(%ld) and not null", new_heads)
         pruned = dagop.reachableroots(repo, candidates, prunestart)
         new_heads.difference_update(pruned)
 
     return pycompat.maplist(cl.node, sorted(new_heads))
 
 
 def newcommitphase(ui: "uimod.ui") -> int:
     """helper to get the target phase of new commit
 
     Handle all possible values for the phases.new-commit options.
 
     """
     v = ui.config(b'phases', b'new-commit')
     try:
         return phasenumber2[v]
     except KeyError:
         raise error.ConfigError(
             _(b"phases.new-commit: not a valid phase name ('%s')") % v
         )
 
 
 def hassecret(repo: "localrepo.localrepository") -> bool:
     """utility function that check if a repo have any secret changeset."""
     return bool(repo._phasecache.phaseroots[secret])
 
 
 def preparehookargs(
     node: bytes,
     old: Optional[int],
     new: Optional[int],
 ) -> Dict[bytes, bytes]:
     if old is None:
         old = b''
     else:
         old = phasenames[old]
     return {b'node': node, b'oldphase': old, b'phase': phasenames[new]}

mercurial/upgrade_utils/actions.py

@@ -1,1133 +1,1133 b''
 # upgrade.py - functions for in place upgrade of Mercurial repository
 #
 # Copyright (c) 2016-present, Gregory Szorc
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 import random
 
 from typing import (
     List,
     Type,
 )
 
 from ..i18n import _
 from .. import (
     error,
     localrepo,
     requirements,
     revlog,
     util,
 )
 
 from ..utils import compression
 
 # keeps pyflakes happy
 assert [
     List,
     Type,
 ]
 
 # list of requirements that request a clone of all revlog if added/removed
 RECLONES_REQUIREMENTS = {
     requirements.GENERALDELTA_REQUIREMENT,
     requirements.SPARSEREVLOG_REQUIREMENT,
     requirements.REVLOGV2_REQUIREMENT,
     requirements.CHANGELOGV2_REQUIREMENT,
 }
 
 
 def preservedrequirements(repo):
     preserved = {
         requirements.SHARED_REQUIREMENT,
         requirements.NARROW_REQUIREMENT,
     }
     return preserved & repo.requirements
 
 
 FORMAT_VARIANT = b'deficiency'
 OPTIMISATION = b'optimization'
 
 
 class improvement:
     """Represents an improvement that can be made as part of an upgrade."""
 
     ### The following attributes should be defined for each subclass:
 
     # Either ``FORMAT_VARIANT`` or ``OPTIMISATION``.
     # A format variant is where we change the storage format. Not all format
     # variant changes are an obvious problem.
     # An optimization is an action (sometimes optional) that
     # can be taken to further improve the state of the repository.
     type = None
 
     # machine-readable string uniquely identifying this improvement. it will be
     # mapped to an action later in the upgrade process.
     name = None
 
     # message intended for humans explaining the improvement in more detail,
     # including the implications of it ``FORMAT_VARIANT`` types, should be
     # worded
     # in the present tense.
     description = None
 
     # message intended for humans explaining what an upgrade addressing this
     # issue will do. should be worded in the future tense.
     upgrademessage = None
 
     # value of current Mercurial default for new repository
     default = None
 
     # Message intended for humans which will be shown post an upgrade
     # operation when the improvement will be added
     postupgrademessage = None
 
     # Message intended for humans which will be shown post an upgrade
     # operation in which this improvement was removed
     postdowngrademessage = None
 
     # By default we assume that every improvement touches requirements and all revlogs
 
     # Whether this improvement touches filelogs
     touches_filelogs = True
 
     # Whether this improvement touches manifests
     touches_manifests = True
 
     # Whether this improvement touches changelog
     touches_changelog = True
 
     # Whether this improvement changes repository requirements
     touches_requirements = True
 
     # Whether this improvement touches the dirstate
     touches_dirstate = False
 
     # Can this action be run on a share instead of its mains repository
     compatible_with_share = False
 
 
-allformatvariant = []  # type: List[Type['formatvariant']]
+allformatvariant: List[Type['formatvariant']] = []
 
 
 def registerformatvariant(cls):
     allformatvariant.append(cls)
     return cls
 
 
 class formatvariant(improvement):
     """an improvement subclass dedicated to repository format"""
 
     type = FORMAT_VARIANT
 
     @staticmethod
     def fromrepo(repo):
         """current value of the variant in the repository"""
         raise NotImplementedError()
 
     @staticmethod
     def fromconfig(repo):
         """current value of the variant in the configuration"""
         raise NotImplementedError()
 
 
 class requirementformatvariant(formatvariant):
     """formatvariant based on a 'requirement' name.
 
     Many format variant are controlled by a 'requirement'. We define a small
     subclass to factor the code.
     """
 
     # the requirement that control this format variant
     _requirement = None
 
     @staticmethod
     def _newreporequirements(ui):
         return localrepo.newreporequirements(
             ui, localrepo.defaultcreateopts(ui)
         )
 
     @classmethod
     def fromrepo(cls, repo):
         assert cls._requirement is not None
         return cls._requirement in repo.requirements
 
     @classmethod
     def fromconfig(cls, repo):
         assert cls._requirement is not None
         return cls._requirement in cls._newreporequirements(repo.ui)
 
 
 @registerformatvariant
 class fncache(requirementformatvariant):
     name = b'fncache'
 
     _requirement = requirements.FNCACHE_REQUIREMENT
 
     default = True
 
     description = _(
         b'long and reserved filenames may not work correctly; '
         b'repository performance is sub-optimal'
     )
 
     upgrademessage = _(
         b'repository will be more resilient to storing '
         b'certain paths and performance of certain '
         b'operations should be improved'
     )
 
 
 @registerformatvariant
 class dirstatev2(requirementformatvariant):
     name = b'dirstate-v2'
     _requirement = requirements.DIRSTATE_V2_REQUIREMENT
 
     default = False
 
     description = _(
         b'version 1 of the dirstate file format requires '
         b'reading and parsing it all at once.\n'
         b'Version 2 has a better structure,'
         b'better information and lighter update mechanism'
     )
 
     upgrademessage = _(b'"hg status" will be faster')
 
     touches_filelogs = False
     touches_manifests = False
     touches_changelog = False
     touches_requirements = True
     touches_dirstate = True
     compatible_with_share = True
 
 
 @registerformatvariant
 class dirstatetrackedkey(requirementformatvariant):
     name = b'tracked-hint'
     _requirement = requirements.DIRSTATE_TRACKED_HINT_V1
 
     default = False
 
     description = _(
         b'Add a small file to help external tooling that watch the tracked set'
     )
 
     upgrademessage = _(
         b'external tools will be informated of potential change in the tracked set'
     )
 
     touches_filelogs = False
     touches_manifests = False
     touches_changelog = False
     touches_requirements = True
     touches_dirstate = True
     compatible_with_share = True
 
 
 @registerformatvariant
 class dotencode(requirementformatvariant):
     name = b'dotencode'
 
     _requirement = requirements.DOTENCODE_REQUIREMENT
 
     default = True
 
     description = _(
         b'storage of filenames beginning with a period or '
         b'space may not work correctly'
     )
 
     upgrademessage = _(
         b'repository will be better able to store files '
         b'beginning with a space or period'
     )
 
 
 @registerformatvariant
 class generaldelta(requirementformatvariant):
     name = b'generaldelta'
 
     _requirement = requirements.GENERALDELTA_REQUIREMENT
 
     default = True
 
     description = _(
         b'deltas within internal storage are unable to '
         b'choose optimal revisions; repository is larger and '
         b'slower than it could be; interaction with other '
         b'repositories may require extra network and CPU '
         b'resources, making "hg push" and "hg pull" slower'
     )
 
     upgrademessage = _(
         b'repository storage will be able to create '
         b'optimal deltas; new repository data will be '
         b'smaller and read times should decrease; '
         b'interacting with other repositories using this '
         b'storage model should require less network and '
         b'CPU resources, making "hg push" and "hg pull" '
         b'faster'
     )
 
 
 @registerformatvariant
 class sharesafe(requirementformatvariant):
     name = b'share-safe'
     _requirement = requirements.SHARESAFE_REQUIREMENT
 
     default = True
 
     description = _(
         b'old shared repositories do not share source repository '
         b'requirements and config. This leads to various problems '
         b'when the source repository format is upgraded or some new '
         b'extensions are enabled.'
     )
 
     upgrademessage = _(
         b'Upgrades a repository to share-safe format so that future '
         b'shares of this repository share its requirements and configs.'
     )
 
     postdowngrademessage = _(
         b'repository downgraded to not use share safe mode, '
         b'existing shares will not work and need to be reshared.'
     )
 
     postupgrademessage = _(
         b'repository upgraded to share safe mode, existing'
         b' shares will still work in old non-safe mode. '
         b'Re-share existing shares to use them in safe mode'
         b' New shares will be created in safe mode.'
     )
 
     # upgrade only needs to change the requirements
     touches_filelogs = False
     touches_manifests = False
     touches_changelog = False
     touches_requirements = True
 
 
 @registerformatvariant
 class sparserevlog(requirementformatvariant):
     name = b'sparserevlog'
 
     _requirement = requirements.SPARSEREVLOG_REQUIREMENT
 
     default = True
 
     description = _(
         b'in order to limit disk reading and memory usage on older '
         b'version, the span of a delta chain from its root to its '
         b'end is limited, whatever the relevant data in this span. '
         b'This can severly limit Mercurial ability to build good '
         b'chain of delta resulting is much more storage space being '
         b'taken and limit reusability of on disk delta during '
         b'exchange.'
     )
 
     upgrademessage = _(
         b'Revlog supports delta chain with more unused data '
         b'between payload. These gaps will be skipped at read '
         b'time. This allows for better delta chains, making a '
         b'better compression and faster exchange with server.'
     )
 
 
 @registerformatvariant
 class persistentnodemap(requirementformatvariant):
     name = b'persistent-nodemap'
 
     _requirement = requirements.NODEMAP_REQUIREMENT
 
     default = False
 
     description = _(
         b'persist the node -> rev mapping on disk to speedup lookup'
     )
 
     upgrademessage = _(b'Speedup revision lookup by node id.')
 
 
 @registerformatvariant
 class copiessdc(requirementformatvariant):
     name = b'copies-sdc'
 
     _requirement = requirements.COPIESSDC_REQUIREMENT
 
     default = False
 
     description = _(b'Stores copies information alongside changesets.')
 
     upgrademessage = _(
         b'Allows to use more efficient algorithm to deal with copy tracing.'
     )
 
     touches_filelogs = False
     touches_manifests = False
 
 
 @registerformatvariant
 class revlogv2(requirementformatvariant):
     name = b'revlog-v2'
     _requirement = requirements.REVLOGV2_REQUIREMENT
     default = False
     description = _(b'Version 2 of the revlog.')
     upgrademessage = _(b'very experimental')
 
 
 @registerformatvariant
 class changelogv2(requirementformatvariant):
     name = b'changelog-v2'
     _requirement = requirements.CHANGELOGV2_REQUIREMENT
     default = False
     description = _(b'An iteration of the revlog focussed on changelog needs.')
     upgrademessage = _(b'quite experimental')
 
     touches_filelogs = False
     touches_manifests = False
 
 
 @registerformatvariant
 class removecldeltachain(formatvariant):
     name = b'plain-cl-delta'
 
     default = True
 
     description = _(
         b'changelog storage is using deltas instead of '
         b'raw entries; changelog reading and any '
         b'operation relying on changelog data are slower '
         b'than they could be'
     )
 
     upgrademessage = _(
         b'changelog storage will be reformated to '
         b'store raw entries; changelog reading will be '
         b'faster; changelog size may be reduced'
     )
 
     @staticmethod
     def fromrepo(repo):
         # Mercurial 4.0 changed changelogs to not use delta chains. Search for
         # changelogs with deltas.
         cl = repo.unfiltered().changelog
         if len(cl) <= 1000:
             some_rev = list(cl)
         else:
             # do a random sampling to speeds things up Scanning the whole
             # repository can get really slow on bigger repo.
             some_rev = sorted(
                 {random.randint(0, len(cl) - 1) for x in range(1000)}
             )
         chainbase = cl.chainbase
         return all(rev == chainbase(rev) for rev in some_rev)
 
     @staticmethod
     def fromconfig(repo):
         return True
 
 
 _has_zstd = (
     b'zstd' in util.compengines
     and util.compengines[b'zstd'].available()
     and util.compengines[b'zstd'].revlogheader()
 )
 
 
 @registerformatvariant
 class compressionengine(formatvariant):
     name = b'compression'
 
     if _has_zstd:
         default = b'zstd'
     else:
         default = b'zlib'
 
     description = _(
         b'Compresion algorithm used to compress data. '
         b'Some engine are faster than other'
     )
 
     upgrademessage = _(
         b'revlog content will be recompressed with the new algorithm.'
     )
 
     @classmethod
     def fromrepo(cls, repo):
         # we allow multiple compression engine requirement to co-exist because
         # strickly speaking, revlog seems to support mixed compression style.
         #
         # The compression used for new entries will be "the last one"
         compression = b'zlib'
         for req in repo.requirements:
             prefix = req.startswith
             if prefix(b'revlog-compression-') or prefix(b'exp-compression-'):
                 compression = req.split(b'-', 2)[2]
         return compression
 
     @classmethod
     def fromconfig(cls, repo):
         compengines = repo.ui.configlist(b'format', b'revlog-compression')
         # return the first valid value as the selection code would do
         for comp in compengines:
             if comp in util.compengines:
                 e = util.compengines[comp]
                 if e.available() and e.revlogheader():
                     return comp
 
         # no valide compression found lets display it all for clarity
         return b','.join(compengines)
 
 
 @registerformatvariant
 class compressionlevel(formatvariant):
     name = b'compression-level'
     default = b'default'
 
     description = _(b'compression level')
 
     upgrademessage = _(b'revlog content will be recompressed')
 
     @classmethod
     def fromrepo(cls, repo):
         comp = compressionengine.fromrepo(repo)
         level = None
         if comp == b'zlib':
             level = repo.ui.configint(b'storage', b'revlog.zlib.level')
         elif comp == b'zstd':
             level = repo.ui.configint(b'storage', b'revlog.zstd.level')
         if level is None:
             return b'default'
         return bytes(level)
 
     @classmethod
     def fromconfig(cls, repo):
         comp = compressionengine.fromconfig(repo)
         level = None
         if comp == b'zlib':
             level = repo.ui.configint(b'storage', b'revlog.zlib.level')
         elif comp == b'zstd':
             level = repo.ui.configint(b'storage', b'revlog.zstd.level')
         if level is None:
             return b'default'
         return bytes(level)
 
 
 def find_format_upgrades(repo):
     """returns a list of format upgrades which can be perform on the repo"""
     upgrades = []
 
     # We could detect lack of revlogv1 and store here, but they were added
     # in 0.9.2 and we don't support upgrading repos without these
     # requirements, so let's not bother.
 
     for fv in allformatvariant:
         if not fv.fromrepo(repo):
             upgrades.append(fv)
 
     return upgrades
 
 
 def find_format_downgrades(repo):
     """returns a list of format downgrades which will be performed on the repo
     because of disabled config option for them"""
 
     downgrades = []
 
     for fv in allformatvariant:
         if fv.name == b'compression':
             # If there is a compression change between repository
             # and config, destination repository compression will change
             # and current compression will be removed.
             if fv.fromrepo(repo) != fv.fromconfig(repo):
                 downgrades.append(fv)
             continue
         # format variant exist in repo but does not exist in new repository
         # config
         if fv.fromrepo(repo) and not fv.fromconfig(repo):
             downgrades.append(fv)
 
     return downgrades
 
 
 ALL_OPTIMISATIONS = []
 
 
 def register_optimization(obj):
     ALL_OPTIMISATIONS.append(obj)
     return obj
 
 
 class optimization(improvement):
     """an improvement subclass dedicated to optimizations"""
 
     type = OPTIMISATION
 
 
 @register_optimization
 class redeltaparents(optimization):
     name = b're-delta-parent'
 
     type = OPTIMISATION
 
     description = _(
         b'deltas within internal storage will be recalculated to '
         b'choose an optimal base revision where this was not '
         b'already done; the size of the repository may shrink and '
         b'various operations may become faster; the first time '
         b'this optimization is performed could slow down upgrade '
         b'execution considerably; subsequent invocations should '
         b'not run noticeably slower'
     )
 
     upgrademessage = _(
         b'deltas within internal storage will choose a new '
         b'base revision if needed'
     )
 
 
 @register_optimization
 class redeltamultibase(optimization):
     name = b're-delta-multibase'
 
     type = OPTIMISATION
 
     description = _(
         b'deltas within internal storage will be recalculated '
         b'against multiple base revision and the smallest '
         b'difference will be used; the size of the repository may '
         b'shrink significantly when there are many merges; this '
         b'optimization will slow down execution in proportion to '
         b'the number of merges in the repository and the amount '
         b'of files in the repository; this slow down should not '
         b'be significant unless there are tens of thousands of '
         b'files and thousands of merges'
     )
 
     upgrademessage = _(
         b'deltas within internal storage will choose an '
         b'optimal delta by computing deltas against multiple '
         b'parents; may slow down execution time '
         b'significantly'
     )
 
 
 @register_optimization
 class redeltaall(optimization):
     name = b're-delta-all'
 
     type = OPTIMISATION
 
     description = _(
         b'deltas within internal storage will always be '
         b'recalculated without reusing prior deltas; this will '
         b'likely make execution run several times slower; this '
         b'optimization is typically not needed'
     )
 
     upgrademessage = _(
         b'deltas within internal storage will be fully '
         b'recomputed; this will likely drastically slow down '
         b'execution time'
     )
 
 
 @register_optimization
 class redeltafulladd(optimization):
     name = b're-delta-fulladd'
 
     type = OPTIMISATION
 
     description = _(
         b'every revision will be re-added as if it was new '
         b'content. It will go through the full storage '
         b'mechanism giving extensions a chance to process it '
         b'(eg. lfs). This is similar to "re-delta-all" but even '
         b'slower since more logic is involved.'
     )
 
     upgrademessage = _(
         b'each revision will be added as new content to the '
         b'internal storage; this will likely drastically slow '
         b'down execution time, but some extensions might need '
         b'it'
     )
 
 
 def findoptimizations(repo):
     """Determine optimisation that could be used during upgrade"""
     # These are unconditionally added. There is logic later that figures out
     # which ones to apply.
     return list(ALL_OPTIMISATIONS)
 
 
 def determine_upgrade_actions(
     repo, format_upgrades, optimizations, sourcereqs, destreqs
 ):
     """Determine upgrade actions that will be performed.
 
     Given a list of improvements as returned by ``find_format_upgrades`` and
     ``findoptimizations``, determine the list of upgrade actions that
     will be performed.
 
     The role of this function is to filter improvements if needed, apply
     recommended optimizations from the improvements list that make sense,
     etc.
 
     Returns a list of action names.
     """
     newactions = []
 
     for d in format_upgrades:
         if hasattr(d, '_requirement'):
             name = d._requirement
         else:
             name = None
 
         # If the action is a requirement that doesn't show up in the
         # destination requirements, prune the action.
         if name is not None and name not in destreqs:
             continue
 
         newactions.append(d)
 
     newactions.extend(
         o
         for o in sorted(optimizations, key=(lambda x: x.name))
         if o not in newactions
     )
 
     # FUTURE consider adding some optimizations here for certain transitions.
     # e.g. adding generaldelta could schedule parent redeltas.
 
     return newactions
 
 
 class BaseOperation:
     """base class that contains the minimum for an upgrade to work
 
     (this might need to be extended as the usage for subclass alternative to
     UpgradeOperation extends)
     """
 
     def __init__(
         self,
         new_requirements,
         backup_store,
     ):
         self.new_requirements = new_requirements
         # should this operation create a backup of the store
         self.backup_store = backup_store
 
 
 class UpgradeOperation(BaseOperation):
     """represent the work to be done during an upgrade"""
 
     def __init__(
         self,
         ui,
         new_requirements,
         current_requirements,
         upgrade_actions,
         removed_actions,
         revlogs_to_process,
         backup_store,
     ):
         super().__init__(
             new_requirements,
             backup_store,
         )
         self.ui = ui
         self.current_requirements = current_requirements
         # list of upgrade actions the operation will perform
         self.upgrade_actions = upgrade_actions
         self.removed_actions = removed_actions
         self.revlogs_to_process = revlogs_to_process
         # requirements which will be added by the operation
         self._added_requirements = (
             self.new_requirements - self.current_requirements
         )
         # requirements which will be removed by the operation
         self._removed_requirements = (
             self.current_requirements - self.new_requirements
         )
         # requirements which will be preserved by the operation
         self._preserved_requirements = (
             self.current_requirements & self.new_requirements
         )
         # optimizations which are not used and it's recommended that they
         # should use them
         all_optimizations = findoptimizations(None)
         self.unused_optimizations = [
             i for i in all_optimizations if i not in self.upgrade_actions
         ]
 
         # delta reuse mode of this upgrade operation
         upgrade_actions_names = self.upgrade_actions_names
         self.delta_reuse_mode = revlog.revlog.DELTAREUSEALWAYS
         if b're-delta-all' in upgrade_actions_names:
             self.delta_reuse_mode = revlog.revlog.DELTAREUSENEVER
         elif b're-delta-parent' in upgrade_actions_names:
             self.delta_reuse_mode = revlog.revlog.DELTAREUSESAMEREVS
         elif b're-delta-multibase' in upgrade_actions_names:
             self.delta_reuse_mode = revlog.revlog.DELTAREUSESAMEREVS
         elif b're-delta-fulladd' in upgrade_actions_names:
             self.delta_reuse_mode = revlog.revlog.DELTAREUSEFULLADD
 
         # should this operation force re-delta of both parents
         self.force_re_delta_both_parents = (
             b're-delta-multibase' in upgrade_actions_names
         )
 
     @property
     def upgrade_actions_names(self):
         return set([a.name for a in self.upgrade_actions])
 
     @property
     def requirements_only(self):
         # does the operation only touches repository requirement
         return (
             self.touches_requirements
             and not self.touches_filelogs
             and not self.touches_manifests
             and not self.touches_changelog
             and not self.touches_dirstate
         )
 
     @property
     def touches_filelogs(self):
         for a in self.upgrade_actions:
             # in optimisations, we re-process the revlogs again
             if a.type == OPTIMISATION:
                 return True
             elif a.touches_filelogs:
                 return True
         for a in self.removed_actions:
             if a.touches_filelogs:
                 return True
         return False
 
     @property
     def touches_manifests(self):
         for a in self.upgrade_actions:
             # in optimisations, we re-process the revlogs again
             if a.type == OPTIMISATION:
                 return True
             elif a.touches_manifests:
                 return True
         for a in self.removed_actions:
             if a.touches_manifests:
                 return True
         return False
 
     @property
     def touches_changelog(self):
         for a in self.upgrade_actions:
             # in optimisations, we re-process the revlogs again
             if a.type == OPTIMISATION:
                 return True
             elif a.touches_changelog:
                 return True
         for a in self.removed_actions:
             if a.touches_changelog:
                 return True
         return False
 
     @property
     def touches_requirements(self):
         for a in self.upgrade_actions:
             # optimisations are used to re-process revlogs and does not result
             # in a requirement being added or removed
             if a.type == OPTIMISATION:
                 pass
             elif a.touches_requirements:
                 return True
         for a in self.removed_actions:
             if a.touches_requirements:
                 return True
 
     @property
     def touches_dirstate(self):
         for a in self.upgrade_actions:
             # revlog optimisations do not affect the dirstate
             if a.type == OPTIMISATION:
                 pass
             elif a.touches_dirstate:
                 return True
         for a in self.removed_actions:
             if a.touches_dirstate:
                 return True
 
         return False
 
     def _write_labeled(self, l, label: bytes):
         """
         Utility function to aid writing of a list under one label
         """
         first = True
         for r in sorted(l):
             if not first:
                 self.ui.write(b', ')
             self.ui.write(r, label=label)
             first = False
 
     def print_requirements(self):
         self.ui.write(_(b'requirements\n'))
         self.ui.write(_(b'   preserved: '))
         self._write_labeled(
             self._preserved_requirements, b"upgrade-repo.requirement.preserved"
         )
         self.ui.write((b'\n'))
         if self._removed_requirements:
             self.ui.write(_(b'   removed: '))
             self._write_labeled(
                 self._removed_requirements, b"upgrade-repo.requirement.removed"
             )
             self.ui.write((b'\n'))
         if self._added_requirements:
             self.ui.write(_(b'   added: '))
             self._write_labeled(
                 self._added_requirements, b"upgrade-repo.requirement.added"
             )
             self.ui.write((b'\n'))
         self.ui.write(b'\n')
 
     def print_optimisations(self):
         optimisations = [
             a for a in self.upgrade_actions if a.type == OPTIMISATION
         ]
         optimisations.sort(key=lambda a: a.name)
         if optimisations:
             self.ui.write(_(b'optimisations: '))
             self._write_labeled(
                 [a.name for a in optimisations],
                 b"upgrade-repo.optimisation.performed",
             )
             self.ui.write(b'\n\n')
 
     def print_upgrade_actions(self):
         for a in self.upgrade_actions:
             self.ui.status(b'%s\n   %s\n\n' % (a.name, a.upgrademessage))
 
     def print_affected_revlogs(self):
         if not self.revlogs_to_process:
             self.ui.write((b'no revlogs to process\n'))
         else:
             self.ui.write((b'processed revlogs:\n'))
             for r in sorted(self.revlogs_to_process):
                 self.ui.write((b'  - %s\n' % r))
         self.ui.write((b'\n'))
 
     def print_unused_optimizations(self):
         for i in self.unused_optimizations:
             self.ui.status(_(b'%s\n   %s\n\n') % (i.name, i.description))
 
     def has_upgrade_action(self, name):
         """Check whether the upgrade operation will perform this action"""
         return name in self._upgrade_actions_names
 
     def print_post_op_messages(self):
         """print post upgrade operation warning messages"""
         for a in self.upgrade_actions:
             if a.postupgrademessage is not None:
                 self.ui.warn(b'%s\n' % a.postupgrademessage)
         for a in self.removed_actions:
             if a.postdowngrademessage is not None:
                 self.ui.warn(b'%s\n' % a.postdowngrademessage)
 
 
 ###  Code checking if a repository can got through the upgrade process at all. #
 
 
 def requiredsourcerequirements(repo):
     """Obtain requirements required to be present to upgrade a repo.
 
     An upgrade will not be allowed if the repository doesn't have the
     requirements returned by this function.
     """
     return {
         # Introduced in Mercurial 0.9.2.
         requirements.STORE_REQUIREMENT,
     }
 
 
 def blocksourcerequirements(repo):
     """Obtain requirements that will prevent an upgrade from occurring.
 
     An upgrade cannot be performed if the source repository contains a
     requirements in the returned set.
     """
     return {
         # This was a precursor to generaldelta and was never enabled by default.
         # It should (hopefully) not exist in the wild.
         b'parentdelta',
     }
 
 
 def check_revlog_version(reqs):
     """Check that the requirements contain at least one Revlog version"""
     all_revlogs = {
         requirements.REVLOGV1_REQUIREMENT,
         requirements.REVLOGV2_REQUIREMENT,
     }
     if not all_revlogs.intersection(reqs):
         msg = _(b'cannot upgrade repository; missing a revlog version')
         raise error.Abort(msg)
 
 
 def check_source_requirements(repo):
     """Ensure that no existing requirements prevent the repository upgrade"""
 
     check_revlog_version(repo.requirements)
     required = requiredsourcerequirements(repo)
     missingreqs = required - repo.requirements
     if missingreqs:
         msg = _(b'cannot upgrade repository; requirement missing: %s')
         missingreqs = b', '.join(sorted(missingreqs))
         raise error.Abort(msg % missingreqs)
 
     blocking = blocksourcerequirements(repo)
     blockingreqs = blocking & repo.requirements
     if blockingreqs:
         m = _(b'cannot upgrade repository; unsupported source requirement: %s')
         blockingreqs = b', '.join(sorted(blockingreqs))
         raise error.Abort(m % blockingreqs)
     # Upgrade should operate on the actual store, not the shared link.
 
     bad_share = (
         requirements.SHARED_REQUIREMENT in repo.requirements
         and requirements.SHARESAFE_REQUIREMENT not in repo.requirements
     )
     if bad_share:
         m = _(b'cannot upgrade repository; share repository without share-safe')
         h = _(b'check :hg:`help config.format.use-share-safe`')
         raise error.Abort(m, hint=h)
 
 
 ### Verify the validity of the planned requirement changes ####################
 
 
 def supportremovedrequirements(repo):
     """Obtain requirements that can be removed during an upgrade.
 
     If an upgrade were to create a repository that dropped a requirement,
     the dropped requirement must appear in the returned set for the upgrade
     to be allowed.
     """
     supported = {
         requirements.SPARSEREVLOG_REQUIREMENT,
         requirements.COPIESSDC_REQUIREMENT,
         requirements.NODEMAP_REQUIREMENT,
         requirements.SHARESAFE_REQUIREMENT,
         requirements.REVLOGV2_REQUIREMENT,
         requirements.CHANGELOGV2_REQUIREMENT,
         requirements.REVLOGV1_REQUIREMENT,
         requirements.DIRSTATE_TRACKED_HINT_V1,
         requirements.DIRSTATE_V2_REQUIREMENT,
     }
     for name in compression.compengines:
         engine = compression.compengines[name]
         if engine.available() and engine.revlogheader():
             supported.add(b'exp-compression-%s' % name)
             if engine.name() == b'zstd':
                 supported.add(b'revlog-compression-zstd')
     return supported
 
 
 def supporteddestrequirements(repo):
     """Obtain requirements that upgrade supports in the destination.
 
     If the result of the upgrade would have requirements not in this set,
     the upgrade is disallowed.
 
     Extensions should monkeypatch this to add their custom requirements.
     """
     supported = {
         requirements.CHANGELOGV2_REQUIREMENT,
         requirements.COPIESSDC_REQUIREMENT,
         requirements.DIRSTATE_TRACKED_HINT_V1,
         requirements.DIRSTATE_V2_REQUIREMENT,
         requirements.DOTENCODE_REQUIREMENT,
         requirements.FNCACHE_REQUIREMENT,
         requirements.GENERALDELTA_REQUIREMENT,
         requirements.NODEMAP_REQUIREMENT,
         requirements.REVLOGV1_REQUIREMENT,  # allowed in case of downgrade
         requirements.REVLOGV2_REQUIREMENT,
         requirements.SHARED_REQUIREMENT,
         requirements.SHARESAFE_REQUIREMENT,
         requirements.SPARSEREVLOG_REQUIREMENT,
         requirements.STORE_REQUIREMENT,
         requirements.TREEMANIFEST_REQUIREMENT,
         requirements.NARROW_REQUIREMENT,
     }
     for name in compression.compengines:
         engine = compression.compengines[name]
         if engine.available() and engine.revlogheader():
             supported.add(b'exp-compression-%s' % name)
             if engine.name() == b'zstd':
                 supported.add(b'revlog-compression-zstd')
     return supported
 
 
 def allowednewrequirements(repo):
     """Obtain requirements that can be added to a repository during upgrade.
 
     This is used to disallow proposed requirements from being added when
     they weren't present before.
 
     We use a list of allowed requirement additions instead of a list of known
     bad additions because the whitelist approach is safer and will prevent
     future, unknown requirements from accidentally being added.
     """
     supported = {
         requirements.DOTENCODE_REQUIREMENT,
         requirements.FNCACHE_REQUIREMENT,
         requirements.GENERALDELTA_REQUIREMENT,
         requirements.SPARSEREVLOG_REQUIREMENT,
         requirements.COPIESSDC_REQUIREMENT,
         requirements.NODEMAP_REQUIREMENT,
         requirements.SHARESAFE_REQUIREMENT,
         requirements.REVLOGV1_REQUIREMENT,
         requirements.REVLOGV2_REQUIREMENT,
         requirements.CHANGELOGV2_REQUIREMENT,
         requirements.DIRSTATE_TRACKED_HINT_V1,
         requirements.DIRSTATE_V2_REQUIREMENT,
     }
     for name in compression.compengines:
         engine = compression.compengines[name]
         if engine.available() and engine.revlogheader():
             supported.add(b'exp-compression-%s' % name)
             if engine.name() == b'zstd':
                 supported.add(b'revlog-compression-zstd')
     return supported
 
 
 def check_requirements_changes(repo, new_reqs):
     old_reqs = repo.requirements
     check_revlog_version(repo.requirements)
     support_removal = supportremovedrequirements(repo)
     no_remove_reqs = old_reqs - new_reqs - support_removal
     if no_remove_reqs:
         msg = _(b'cannot upgrade repository; requirement would be removed: %s')
         no_remove_reqs = b', '.join(sorted(no_remove_reqs))
         raise error.Abort(msg % no_remove_reqs)
 
     support_addition = allowednewrequirements(repo)
     no_add_reqs = new_reqs - old_reqs - support_addition
     if no_add_reqs:
         m = _(b'cannot upgrade repository; do not support adding requirement: ')
         no_add_reqs = b', '.join(sorted(no_add_reqs))
         raise error.Abort(m + no_add_reqs)
 
     supported = supporteddestrequirements(repo)
     unsupported_reqs = new_reqs - supported
     if unsupported_reqs:
         msg = _(
             b'cannot upgrade repository; do not support destination '
             b'requirement: %s'
         )
         unsupported_reqs = b', '.join(sorted(unsupported_reqs))
         raise error.Abort(msg % unsupported_reqs)