cleanup: eliminate procutil.quotecommand()

After some compatibility code was removed, the function was the identity
function on all platforms.

hgext/convert/cvs.py

@@ -1,333 +1,332 b''
 # cvs.py: CVS conversion code inspired by hg-cvs-import and git-cvsimport
 #
 #  Copyright 2005-2009 Matt Mackall <[e-mail]> and others
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 from __future__ import absolute_import
 
 import errno
 import os
 import re
 import socket
 
 from mercurial.i18n import _
 from mercurial.pycompat import (
     getattr,
     open,
 )
 from mercurial import (
     encoding,
     error,
     pycompat,
     util,
 )
 from mercurial.utils import (
     dateutil,
     procutil,
 )
 
 from . import (
     common,
     cvsps,
 )
 
 stringio = util.stringio
 checktool = common.checktool
 commit = common.commit
 converter_source = common.converter_source
 makedatetimestamp = common.makedatetimestamp
 NoRepo = common.NoRepo
 
 
 class convert_cvs(converter_source):
     def __init__(self, ui, repotype, path, revs=None):
         super(convert_cvs, self).__init__(ui, repotype, path, revs=revs)
 
         cvs = os.path.join(path, b"CVS")
         if not os.path.exists(cvs):
             raise NoRepo(_(b"%s does not look like a CVS checkout") % path)
 
         checktool(b'cvs')
 
         self.changeset = None
         self.files = {}
         self.tags = {}
         self.lastbranch = {}
         self.socket = None
         self.cvsroot = open(os.path.join(cvs, b"Root"), b'rb').read()[:-1]
         self.cvsrepo = open(os.path.join(cvs, b"Repository"), b'rb').read()[:-1]
         self.encoding = encoding.encoding
 
         self._connect()
 
     def _parse(self):
         if self.changeset is not None:
             return
         self.changeset = {}
 
         maxrev = 0
         if self.revs:
             if len(self.revs) > 1:
                 raise error.Abort(
                     _(
                         b'cvs source does not support specifying '
                         b'multiple revs'
                     )
                 )
             # TODO: handle tags
             try:
                 # patchset number?
                 maxrev = int(self.revs[0])
             except ValueError:
                 raise error.Abort(
                     _(b'revision %s is not a patchset number') % self.revs[0]
                 )
 
         d = encoding.getcwd()
         try:
             os.chdir(self.path)
 
             cache = b'update'
             if not self.ui.configbool(b'convert', b'cvsps.cache'):
                 cache = None
             db = cvsps.createlog(self.ui, cache=cache)
             db = cvsps.createchangeset(
                 self.ui,
                 db,
                 fuzz=int(self.ui.config(b'convert', b'cvsps.fuzz')),
                 mergeto=self.ui.config(b'convert', b'cvsps.mergeto'),
                 mergefrom=self.ui.config(b'convert', b'cvsps.mergefrom'),
             )
 
             for cs in db:
                 if maxrev and cs.id > maxrev:
                     break
                 id = b"%d" % cs.id
                 cs.author = self.recode(cs.author)
                 self.lastbranch[cs.branch] = id
                 cs.comment = self.recode(cs.comment)
                 if self.ui.configbool(b'convert', b'localtimezone'):
                     cs.date = makedatetimestamp(cs.date[0])
                 date = dateutil.datestr(cs.date, b'%Y-%m-%d %H:%M:%S %1%2')
                 self.tags.update(dict.fromkeys(cs.tags, id))
 
                 files = {}
                 for f in cs.entries:
                     files[f.file] = b"%s%s" % (
                         b'.'.join([(b"%d" % x) for x in f.revision]),
                         [b'', b'(DEAD)'][f.dead],
                     )
 
                 # add current commit to set
                 c = commit(
                     author=cs.author,
                     date=date,
                     parents=[(b"%d" % p.id) for p in cs.parents],
                     desc=cs.comment,
                     branch=cs.branch or b'',
                 )
                 self.changeset[id] = c
                 self.files[id] = files
 
             self.heads = self.lastbranch.values()
         finally:
             os.chdir(d)
 
     def _connect(self):
         root = self.cvsroot
         conntype = None
         user, host = None, None
         cmd = [b'cvs', b'server']
 
         self.ui.status(_(b"connecting to %s\n") % root)
 
         if root.startswith(b":pserver:"):
             root = root[9:]
             m = re.match(r'(?:(.*?)(?::(.*?))?@)?([^:/]*)(?::(\d*))?(.*)', root)
             if m:
                 conntype = b"pserver"
                 user, passw, serv, port, root = m.groups()
                 if not user:
                     user = b"anonymous"
                 if not port:
                     port = 2401
                 else:
                     port = int(port)
                 format0 = b":pserver:%s@%s:%s" % (user, serv, root)
                 format1 = b":pserver:%s@%s:%d%s" % (user, serv, port, root)
 
                 if not passw:
                     passw = b"A"
                     cvspass = os.path.expanduser(b"~/.cvspass")
                     try:
                         pf = open(cvspass, b'rb')
                         for line in pf.read().splitlines():
                             part1, part2 = line.split(b' ', 1)
                             # /1 :pserver:[e-mail]:2401/cvsroot/foo
                             # Ah<Z
                             if part1 == b'/1':
                                 part1, part2 = part2.split(b' ', 1)
                                 format = format1
                             # :pserver:[e-mail]:/cvsroot/foo Ah<Z
                             else:
                                 format = format0
                             if part1 == format:
                                 passw = part2
                                 break
                         pf.close()
                     except IOError as inst:
                         if inst.errno != errno.ENOENT:
                             if not getattr(inst, 'filename', None):
                                 inst.filename = cvspass
                             raise
 
                 sck = socket.socket()
                 sck.connect((serv, port))
                 sck.send(
                     b"\n".join(
                         [
                             b"BEGIN AUTH REQUEST",
                             root,
                             user,
                             passw,
                             b"END AUTH REQUEST",
                             b"",
                         ]
                     )
                 )
                 if sck.recv(128) != b"I LOVE YOU\n":
                     raise error.Abort(_(b"CVS pserver authentication failed"))
 
                 self.writep = self.readp = sck.makefile(b'r+')
 
         if not conntype and root.startswith(b":local:"):
             conntype = b"local"
             root = root[7:]
 
         if not conntype:
             # :ext:user@host/home/user/path/to/cvsroot
             if root.startswith(b":ext:"):
                 root = root[5:]
             m = re.match(br'(?:([^@:/]+)@)?([^:/]+):?(.*)', root)
             # Do not take Windows path "c:\foo\bar" for a connection strings
             if os.path.isdir(root) or not m:
                 conntype = b"local"
             else:
                 conntype = b"rsh"
                 user, host, root = m.group(1), m.group(2), m.group(3)
 
         if conntype != b"pserver":
             if conntype == b"rsh":
                 rsh = encoding.environ.get(b"CVS_RSH") or b"ssh"
                 if user:
                     cmd = [rsh, b'-l', user, host] + cmd
                 else:
                     cmd = [rsh, host] + cmd
 
             # popen2 does not support argument lists under Windows
-            cmd = [procutil.shellquote(arg) for arg in cmd]
-            cmd = procutil.quotecommand(b' '.join(cmd))
+            cmd = b' '.join(procutil.shellquote(arg) for arg in cmd)
             self.writep, self.readp = procutil.popen2(cmd)
 
         self.realroot = root
 
         self.writep.write(b"Root %s\n" % root)
         self.writep.write(
             b"Valid-responses ok error Valid-requests Mode"
             b" M Mbinary E Checked-in Created Updated"
             b" Merged Removed\n"
         )
         self.writep.write(b"valid-requests\n")
         self.writep.flush()
         r = self.readp.readline()
         if not r.startswith(b"Valid-requests"):
             raise error.Abort(
                 _(
                     b'unexpected response from CVS server '
                     b'(expected "Valid-requests", but got %r)'
                 )
                 % r
             )
         if b"UseUnchanged" in r:
             self.writep.write(b"UseUnchanged\n")
             self.writep.flush()
             self.readp.readline()
 
     def getheads(self):
         self._parse()
         return self.heads
 
     def getfile(self, name, rev):
         def chunkedread(fp, count):
             # file-objects returned by socket.makefile() do not handle
             # large read() requests very well.
             chunksize = 65536
             output = stringio()
             while count > 0:
                 data = fp.read(min(count, chunksize))
                 if not data:
                     raise error.Abort(
                         _(b"%d bytes missing from remote file") % count
                     )
                 count -= len(data)
                 output.write(data)
             return output.getvalue()
 
         self._parse()
         if rev.endswith(b"(DEAD)"):
             return None, None
 
         args = (b"-N -P -kk -r %s --" % rev).split()
         args.append(self.cvsrepo + b'/' + name)
         for x in args:
             self.writep.write(b"Argument %s\n" % x)
         self.writep.write(b"Directory .\n%s\nco\n" % self.realroot)
         self.writep.flush()
 
         data = b""
         mode = None
         while True:
             line = self.readp.readline()
             if line.startswith(b"Created ") or line.startswith(b"Updated "):
                 self.readp.readline()  # path
                 self.readp.readline()  # entries
                 mode = self.readp.readline()[:-1]
                 count = int(self.readp.readline()[:-1])
                 data = chunkedread(self.readp, count)
             elif line.startswith(b" "):
                 data += line[1:]
             elif line.startswith(b"M "):
                 pass
             elif line.startswith(b"Mbinary "):
                 count = int(self.readp.readline()[:-1])
                 data = chunkedread(self.readp, count)
             else:
                 if line == b"ok\n":
                     if mode is None:
                         raise error.Abort(_(b'malformed response from CVS'))
                     return (data, b"x" in mode and b"x" or b"")
                 elif line.startswith(b"E "):
                     self.ui.warn(_(b"cvs server: %s\n") % line[2:])
                 elif line.startswith(b"Remove"):
                     self.readp.readline()
                 else:
                     raise error.Abort(_(b"unknown CVS response: %s") % line)
 
     def getchanges(self, rev, full):
         if full:
             raise error.Abort(_(b"convert from cvs does not support --full"))
         self._parse()
         return sorted(pycompat.iteritems(self.files[rev])), {}, set()
 
     def getcommit(self, rev):
         self._parse()
         return self.changeset[rev]
 
     def gettags(self):
         self._parse()
         return self.tags
 
     def getchangedfiles(self, rev, i):
         self._parse()
         return sorted(self.files[rev])

hgext/convert/gnuarch.py

@@ -1,378 +1,378 b''
 # gnuarch.py - GNU Arch support for the convert extension
 #
 #  Copyright 2008, 2009 Aleix Conchillo Flaque <[e-mail]>
 #  and others
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 from __future__ import absolute_import
 
 import os
 import shutil
 import stat
 import tempfile
 
 from mercurial.i18n import _
 from mercurial import (
     encoding,
     error,
     mail,
     pycompat,
     util,
 )
 from mercurial.utils import (
     dateutil,
     procutil,
 )
 from . import common
 
 
 class gnuarch_source(common.converter_source, common.commandline):
     class gnuarch_rev(object):
         def __init__(self, rev):
             self.rev = rev
             self.summary = b''
             self.date = None
             self.author = b''
             self.continuationof = None
             self.add_files = []
             self.mod_files = []
             self.del_files = []
             self.ren_files = {}
             self.ren_dirs = {}
 
     def __init__(self, ui, repotype, path, revs=None):
         super(gnuarch_source, self).__init__(ui, repotype, path, revs=revs)
 
         if not os.path.exists(os.path.join(path, b'{arch}')):
             raise common.NoRepo(
                 _(b"%s does not look like a GNU Arch repository") % path
             )
 
         # Could use checktool, but we want to check for baz or tla.
         self.execmd = None
         if procutil.findexe(b'baz'):
             self.execmd = b'baz'
         else:
             if procutil.findexe(b'tla'):
                 self.execmd = b'tla'
             else:
                 raise error.Abort(_(b'cannot find a GNU Arch tool'))
 
         common.commandline.__init__(self, ui, self.execmd)
 
         self.path = os.path.realpath(path)
         self.tmppath = None
 
         self.treeversion = None
         self.lastrev = None
         self.changes = {}
         self.parents = {}
         self.tags = {}
         self.encoding = encoding.encoding
         self.archives = []
 
     def before(self):
         # Get registered archives
         self.archives = [
             i.rstrip(b'\n') for i in self.runlines0(b'archives', b'-n')
         ]
 
         if self.execmd == b'tla':
             output = self.run0(b'tree-version', self.path)
         else:
             output = self.run0(b'tree-version', b'-d', self.path)
         self.treeversion = output.strip()
 
         # Get name of temporary directory
         version = self.treeversion.split(b'/')
         self.tmppath = os.path.join(
             pycompat.fsencode(tempfile.gettempdir()), b'hg-%s' % version[1]
         )
 
         # Generate parents dictionary
         self.parents[None] = []
         treeversion = self.treeversion
         child = None
         while treeversion:
             self.ui.status(_(b'analyzing tree version %s...\n') % treeversion)
 
             archive = treeversion.split(b'/')[0]
             if archive not in self.archives:
                 self.ui.status(
                     _(
                         b'tree analysis stopped because it points to '
                         b'an unregistered archive %s...\n'
                     )
                     % archive
                 )
                 break
 
             # Get the complete list of revisions for that tree version
             output, status = self.runlines(
                 b'revisions', b'-r', b'-f', treeversion
             )
             self.checkexit(
                 status, b'failed retrieving revisions for %s' % treeversion
             )
 
             # No new iteration unless a revision has a continuation-of header
             treeversion = None
 
             for l in output:
                 rev = l.strip()
                 self.changes[rev] = self.gnuarch_rev(rev)
                 self.parents[rev] = []
 
                 # Read author, date and summary
                 catlog, status = self.run(b'cat-log', b'-d', self.path, rev)
                 if status:
                     catlog = self.run0(b'cat-archive-log', rev)
                 self._parsecatlog(catlog, rev)
 
                 # Populate the parents map
                 self.parents[child].append(rev)
 
                 # Keep track of the current revision as the child of the next
                 # revision scanned
                 child = rev
 
                 # Check if we have to follow the usual incremental history
                 # or if we have to 'jump' to a different treeversion given
                 # by the continuation-of header.
                 if self.changes[rev].continuationof:
                     treeversion = b'--'.join(
                         self.changes[rev].continuationof.split(b'--')[:-1]
                     )
                     break
 
                 # If we reached a base-0 revision w/o any continuation-of
                 # header, it means the tree history ends here.
                 if rev[-6:] == b'base-0':
                     break
 
     def after(self):
         self.ui.debug(b'cleaning up %s\n' % self.tmppath)
         shutil.rmtree(self.tmppath, ignore_errors=True)
 
     def getheads(self):
         return self.parents[None]
 
     def getfile(self, name, rev):
         if rev != self.lastrev:
             raise error.Abort(_(b'internal calling inconsistency'))
 
         if not os.path.lexists(os.path.join(self.tmppath, name)):
             return None, None
 
         return self._getfile(name, rev)
 
     def getchanges(self, rev, full):
         if full:
             raise error.Abort(_(b"convert from arch does not support --full"))
         self._update(rev)
         changes = []
         copies = {}
 
         for f in self.changes[rev].add_files:
             changes.append((f, rev))
 
         for f in self.changes[rev].mod_files:
             changes.append((f, rev))
 
         for f in self.changes[rev].del_files:
             changes.append((f, rev))
 
         for src in self.changes[rev].ren_files:
             to = self.changes[rev].ren_files[src]
             changes.append((src, rev))
             changes.append((to, rev))
             copies[to] = src
 
         for src in self.changes[rev].ren_dirs:
             to = self.changes[rev].ren_dirs[src]
             chgs, cps = self._rendirchanges(src, to)
             changes += [(f, rev) for f in chgs]
             copies.update(cps)
 
         self.lastrev = rev
         return sorted(set(changes)), copies, set()
 
     def getcommit(self, rev):
         changes = self.changes[rev]
         return common.commit(
             author=changes.author,
             date=changes.date,
             desc=changes.summary,
             parents=self.parents[rev],
             rev=rev,
         )
 
     def gettags(self):
         return self.tags
 
     def _execute(self, cmd, *args, **kwargs):
         cmdline = [self.execmd, cmd]
         cmdline += args
         cmdline = [procutil.shellquote(arg) for arg in cmdline]
         bdevnull = pycompat.bytestr(os.devnull)
         cmdline += [b'>', bdevnull, b'2>', bdevnull]
-        cmdline = procutil.quotecommand(b' '.join(cmdline))
+        cmdline = b' '.join(cmdline)
         self.ui.debug(cmdline, b'\n')
         return os.system(pycompat.rapply(procutil.tonativestr, cmdline))
 
     def _update(self, rev):
         self.ui.debug(b'applying revision %s...\n' % rev)
         changeset, status = self.runlines(b'replay', b'-d', self.tmppath, rev)
         if status:
             # Something went wrong while merging (baz or tla
             # issue?), get latest revision and try from there
             shutil.rmtree(self.tmppath, ignore_errors=True)
             self._obtainrevision(rev)
         else:
             old_rev = self.parents[rev][0]
             self.ui.debug(
                 b'computing changeset between %s and %s...\n' % (old_rev, rev)
             )
             self._parsechangeset(changeset, rev)
 
     def _getfile(self, name, rev):
         mode = os.lstat(os.path.join(self.tmppath, name)).st_mode
         if stat.S_ISLNK(mode):
             data = util.readlink(os.path.join(self.tmppath, name))
             if mode:
                 mode = b'l'
             else:
                 mode = b''
         else:
             data = util.readfile(os.path.join(self.tmppath, name))
             mode = (mode & 0o111) and b'x' or b''
         return data, mode
 
     def _exclude(self, name):
         exclude = [b'{arch}', b'.arch-ids', b'.arch-inventory']
         for exc in exclude:
             if name.find(exc) != -1:
                 return True
         return False
 
     def _readcontents(self, path):
         files = []
         contents = os.listdir(path)
         while len(contents) > 0:
             c = contents.pop()
             p = os.path.join(path, c)
             # os.walk could be used, but here we avoid internal GNU
             # Arch files and directories, thus saving a lot time.
             if not self._exclude(p):
                 if os.path.isdir(p):
                     contents += [os.path.join(c, f) for f in os.listdir(p)]
                 else:
                     files.append(c)
         return files
 
     def _rendirchanges(self, src, dest):
         changes = []
         copies = {}
         files = self._readcontents(os.path.join(self.tmppath, dest))
         for f in files:
             s = os.path.join(src, f)
             d = os.path.join(dest, f)
             changes.append(s)
             changes.append(d)
             copies[d] = s
         return changes, copies
 
     def _obtainrevision(self, rev):
         self.ui.debug(b'obtaining revision %s...\n' % rev)
         output = self._execute(b'get', rev, self.tmppath)
         self.checkexit(output)
         self.ui.debug(b'analyzing revision %s...\n' % rev)
         files = self._readcontents(self.tmppath)
         self.changes[rev].add_files += files
 
     def _stripbasepath(self, path):
         if path.startswith(b'./'):
             return path[2:]
         return path
 
     def _parsecatlog(self, data, rev):
         try:
             catlog = mail.parsebytes(data)
 
             # Commit date
             self.changes[rev].date = dateutil.datestr(
                 dateutil.strdate(catlog['Standard-date'], b'%Y-%m-%d %H:%M:%S')
             )
 
             # Commit author
             self.changes[rev].author = self.recode(catlog['Creator'])
 
             # Commit description
             self.changes[rev].summary = b'\n\n'.join(
                 (
                     self.recode(catlog['Summary']),
                     self.recode(catlog.get_payload()),
                 )
             )
             self.changes[rev].summary = self.recode(self.changes[rev].summary)
 
             # Commit revision origin when dealing with a branch or tag
             if 'Continuation-of' in catlog:
                 self.changes[rev].continuationof = self.recode(
                     catlog['Continuation-of']
                 )
         except Exception:
             raise error.Abort(_(b'could not parse cat-log of %s') % rev)
 
     def _parsechangeset(self, data, rev):
         for l in data:
             l = l.strip()
             # Added file (ignore added directory)
             if l.startswith(b'A') and not l.startswith(b'A/'):
                 file = self._stripbasepath(l[1:].strip())
                 if not self._exclude(file):
                     self.changes[rev].add_files.append(file)
             # Deleted file (ignore deleted directory)
             elif l.startswith(b'D') and not l.startswith(b'D/'):
                 file = self._stripbasepath(l[1:].strip())
                 if not self._exclude(file):
                     self.changes[rev].del_files.append(file)
             # Modified binary file
             elif l.startswith(b'Mb'):
                 file = self._stripbasepath(l[2:].strip())
                 if not self._exclude(file):
                     self.changes[rev].mod_files.append(file)
             # Modified link
             elif l.startswith(b'M->'):
                 file = self._stripbasepath(l[3:].strip())
                 if not self._exclude(file):
                     self.changes[rev].mod_files.append(file)
             # Modified file
             elif l.startswith(b'M'):
                 file = self._stripbasepath(l[1:].strip())
                 if not self._exclude(file):
                     self.changes[rev].mod_files.append(file)
             # Renamed file (or link)
             elif l.startswith(b'=>'):
                 files = l[2:].strip().split(b' ')
                 if len(files) == 1:
                     files = l[2:].strip().split(b'\t')
                 src = self._stripbasepath(files[0])
                 dst = self._stripbasepath(files[1])
                 if not self._exclude(src) and not self._exclude(dst):
                     self.changes[rev].ren_files[src] = dst
             # Conversion from file to link or from link to file (modified)
             elif l.startswith(b'ch'):
                 file = self._stripbasepath(l[2:].strip())
                 if not self._exclude(file):
                     self.changes[rev].mod_files.append(file)
             # Renamed directory
             elif l.startswith(b'/>'):
                 dirs = l[2:].strip().split(b' ')
                 if len(dirs) == 1:
                     dirs = l[2:].strip().split(b'\t')
                 src = self._stripbasepath(dirs[0])
                 dst = self._stripbasepath(dirs[1])
                 if not self._exclude(src) and not self._exclude(dst):
                     self.changes[rev].ren_dirs[src] = dst

hgext/convert/subversion.py

@@ -1,1565 +1,1565 b''
 # Subversion 1.4/1.5 Python API backend
 #
 # Copyright(C) 2007 Daniel Holth et al
 from __future__ import absolute_import
 
 import os
 import re
 import xml.dom.minidom
 
 from mercurial.i18n import _
 from mercurial.pycompat import open
 from mercurial import (
     encoding,
     error,
     pycompat,
     util,
     vfs as vfsmod,
 )
 from mercurial.utils import (
     dateutil,
     procutil,
     stringutil,
 )
 
 from . import common
 
 pickle = util.pickle
 stringio = util.stringio
 propertycache = util.propertycache
 urlerr = util.urlerr
 urlreq = util.urlreq
 
 commandline = common.commandline
 commit = common.commit
 converter_sink = common.converter_sink
 converter_source = common.converter_source
 decodeargs = common.decodeargs
 encodeargs = common.encodeargs
 makedatetimestamp = common.makedatetimestamp
 mapfile = common.mapfile
 MissingTool = common.MissingTool
 NoRepo = common.NoRepo
 
 # Subversion stuff. Works best with very recent Python SVN bindings
 # e.g. SVN 1.5 or backports. Thanks to the bzr folks for enhancing
 # these bindings.
 
 try:
     import svn
     import svn.client
     import svn.core
     import svn.ra
     import svn.delta
     from . import transport
     import warnings
 
     warnings.filterwarnings(
         b'ignore', module=b'svn.core', category=DeprecationWarning
     )
     svn.core.SubversionException  # trigger import to catch error
 
 except ImportError:
     svn = None
 
 
 class SvnPathNotFound(Exception):
     pass
 
 
 def revsplit(rev):
     """Parse a revision string and return (uuid, path, revnum).
     >>> revsplit(b'svn:a2147622-4a9f-4db4-a8d3-13562ff547b2'
     ...          b'/proj%20B/mytrunk/mytrunk@1')
     ('a2147622-4a9f-4db4-a8d3-13562ff547b2', '/proj%20B/mytrunk/mytrunk', 1)
     >>> revsplit(b'svn:8af66a51-67f5-4354-b62c-98d67cc7be1d@1')
     ('', '', 1)
     >>> revsplit(b'@7')
     ('', '', 7)
     >>> revsplit(b'7')
     ('', '', 0)
     >>> revsplit(b'bad')
     ('', '', 0)
     """
     parts = rev.rsplit(b'@', 1)
     revnum = 0
     if len(parts) > 1:
         revnum = int(parts[1])
     parts = parts[0].split(b'/', 1)
     uuid = b''
     mod = b''
     if len(parts) > 1 and parts[0].startswith(b'svn:'):
         uuid = parts[0][4:]
         mod = b'/' + parts[1]
     return uuid, mod, revnum
 
 
 def quote(s):
     # As of svn 1.7, many svn calls expect "canonical" paths. In
     # theory, we should call svn.core.*canonicalize() on all paths
     # before passing them to the API.  Instead, we assume the base url
     # is canonical and copy the behaviour of svn URL encoding function
     # so we can extend it safely with new components. The "safe"
     # characters were taken from the "svn_uri__char_validity" table in
     # libsvn_subr/path.c.
     return urlreq.quote(s, b"!$&'()*+,-./:=@_~")
 
 
 def geturl(path):
     try:
         return svn.client.url_from_path(svn.core.svn_path_canonicalize(path))
     except svn.core.SubversionException:
         # svn.client.url_from_path() fails with local repositories
         pass
     if os.path.isdir(path):
         path = os.path.normpath(os.path.abspath(path))
         if pycompat.iswindows:
             path = b'/' + util.normpath(path)
         # Module URL is later compared with the repository URL returned
         # by svn API, which is UTF-8.
         path = encoding.tolocal(path)
         path = b'file://%s' % quote(path)
     return svn.core.svn_path_canonicalize(path)
 
 
 def optrev(number):
     optrev = svn.core.svn_opt_revision_t()
     optrev.kind = svn.core.svn_opt_revision_number
     optrev.value.number = number
     return optrev
 
 
 class changedpath(object):
     def __init__(self, p):
         self.copyfrom_path = p.copyfrom_path
         self.copyfrom_rev = p.copyfrom_rev
         self.action = p.action
 
 
 def get_log_child(
     fp,
     url,
     paths,
     start,
     end,
     limit=0,
     discover_changed_paths=True,
     strict_node_history=False,
 ):
     protocol = -1
 
     def receiver(orig_paths, revnum, author, date, message, pool):
         paths = {}
         if orig_paths is not None:
             for k, v in pycompat.iteritems(orig_paths):
                 paths[k] = changedpath(v)
         pickle.dump((paths, revnum, author, date, message), fp, protocol)
 
     try:
         # Use an ra of our own so that our parent can consume
         # our results without confusing the server.
         t = transport.SvnRaTransport(url=url)
         svn.ra.get_log(
             t.ra,
             paths,
             start,
             end,
             limit,
             discover_changed_paths,
             strict_node_history,
             receiver,
         )
     except IOError:
         # Caller may interrupt the iteration
         pickle.dump(None, fp, protocol)
     except Exception as inst:
         pickle.dump(stringutil.forcebytestr(inst), fp, protocol)
     else:
         pickle.dump(None, fp, protocol)
     fp.flush()
     # With large history, cleanup process goes crazy and suddenly
     # consumes *huge* amount of memory. The output file being closed,
     # there is no need for clean termination.
     os._exit(0)
 
 
 def debugsvnlog(ui, **opts):
     """Fetch SVN log in a subprocess and channel them back to parent to
     avoid memory collection issues.
     """
     if svn is None:
         raise error.Abort(
             _(b'debugsvnlog could not load Subversion python bindings')
         )
 
     args = decodeargs(ui.fin.read())
     get_log_child(ui.fout, *args)
 
 
 class logstream(object):
     """Interruptible revision log iterator."""
 
     def __init__(self, stdout):
         self._stdout = stdout
 
     def __iter__(self):
         while True:
             try:
                 entry = pickle.load(self._stdout)
             except EOFError:
                 raise error.Abort(
                     _(
                         b'Mercurial failed to run itself, check'
                         b' hg executable is in PATH'
                     )
                 )
             try:
                 orig_paths, revnum, author, date, message = entry
             except (TypeError, ValueError):
                 if entry is None:
                     break
                 raise error.Abort(_(b"log stream exception '%s'") % entry)
             yield entry
 
     def close(self):
         if self._stdout:
             self._stdout.close()
             self._stdout = None
 
 
 class directlogstream(list):
     """Direct revision log iterator.
     This can be used for debugging and development but it will probably leak
     memory and is not suitable for real conversions."""
 
     def __init__(
         self,
         url,
         paths,
         start,
         end,
         limit=0,
         discover_changed_paths=True,
         strict_node_history=False,
     ):
         def receiver(orig_paths, revnum, author, date, message, pool):
             paths = {}
             if orig_paths is not None:
                 for k, v in pycompat.iteritems(orig_paths):
                     paths[k] = changedpath(v)
             self.append((paths, revnum, author, date, message))
 
         # Use an ra of our own so that our parent can consume
         # our results without confusing the server.
         t = transport.SvnRaTransport(url=url)
         svn.ra.get_log(
             t.ra,
             paths,
             start,
             end,
             limit,
             discover_changed_paths,
             strict_node_history,
             receiver,
         )
 
     def close(self):
         pass
 
 
 # Check to see if the given path is a local Subversion repo. Verify this by
 # looking for several svn-specific files and directories in the given
 # directory.
 def filecheck(ui, path, proto):
     for x in (b'locks', b'hooks', b'format', b'db'):
         if not os.path.exists(os.path.join(path, x)):
             return False
     return True
 
 
 # Check to see if a given path is the root of an svn repo over http. We verify
 # this by requesting a version-controlled URL we know can't exist and looking
 # for the svn-specific "not found" XML.
 def httpcheck(ui, path, proto):
     try:
         opener = urlreq.buildopener()
         rsp = opener.open(b'%s://%s/!svn/ver/0/.svn' % (proto, path), b'rb')
         data = rsp.read()
     except urlerr.httperror as inst:
         if inst.code != 404:
             # Except for 404 we cannot know for sure this is not an svn repo
             ui.warn(
                 _(
                     b'svn: cannot probe remote repository, assume it could '
                     b'be a subversion repository. Use --source-type if you '
                     b'know better.\n'
                 )
             )
             return True
         data = inst.fp.read()
     except Exception:
         # Could be urlerr.urlerror if the URL is invalid or anything else.
         return False
     return b'<m:human-readable errcode="160013">' in data
 
 
 protomap = {
     b'http': httpcheck,
     b'https': httpcheck,
     b'file': filecheck,
 }
 
 
 def issvnurl(ui, url):
     try:
         proto, path = url.split(b'://', 1)
         if proto == b'file':
             if (
                 pycompat.iswindows
                 and path[:1] == b'/'
                 and path[1:2].isalpha()
                 and path[2:6].lower() == b'%3a/'
             ):
                 path = path[:2] + b':/' + path[6:]
             path = urlreq.url2pathname(path)
     except ValueError:
         proto = b'file'
         path = os.path.abspath(url)
     if proto == b'file':
         path = util.pconvert(path)
     check = protomap.get(proto, lambda *args: False)
     while b'/' in path:
         if check(ui, path, proto):
             return True
         path = path.rsplit(b'/', 1)[0]
     return False
 
 
 # SVN conversion code stolen from bzr-svn and tailor
 #
 # Subversion looks like a versioned filesystem, branches structures
 # are defined by conventions and not enforced by the tool. First,
 # we define the potential branches (modules) as "trunk" and "branches"
 # children directories. Revisions are then identified by their
 # module and revision number (and a repository identifier).
 #
 # The revision graph is really a tree (or a forest). By default, a
 # revision parent is the previous revision in the same module. If the
 # module directory is copied/moved from another module then the
 # revision is the module root and its parent the source revision in
 # the parent module. A revision has at most one parent.
 #
 class svn_source(converter_source):
     def __init__(self, ui, repotype, url, revs=None):
         super(svn_source, self).__init__(ui, repotype, url, revs=revs)
 
         if not (
             url.startswith(b'svn://')
             or url.startswith(b'svn+ssh://')
             or (
                 os.path.exists(url)
                 and os.path.exists(os.path.join(url, b'.svn'))
             )
             or issvnurl(ui, url)
         ):
             raise NoRepo(
                 _(b"%s does not look like a Subversion repository") % url
             )
         if svn is None:
             raise MissingTool(_(b'could not load Subversion python bindings'))
 
         try:
             version = svn.core.SVN_VER_MAJOR, svn.core.SVN_VER_MINOR
             if version < (1, 4):
                 raise MissingTool(
                     _(
                         b'Subversion python bindings %d.%d found, '
                         b'1.4 or later required'
                     )
                     % version
                 )
         except AttributeError:
             raise MissingTool(
                 _(
                     b'Subversion python bindings are too old, 1.4 '
                     b'or later required'
                 )
             )
 
         self.lastrevs = {}
 
         latest = None
         try:
             # Support file://path@rev syntax. Useful e.g. to convert
             # deleted branches.
             at = url.rfind(b'@')
             if at >= 0:
                 latest = int(url[at + 1 :])
                 url = url[:at]
         except ValueError:
             pass
         self.url = geturl(url)
         self.encoding = b'UTF-8'  # Subversion is always nominal UTF-8
         try:
             self.transport = transport.SvnRaTransport(url=self.url)
             self.ra = self.transport.ra
             self.ctx = self.transport.client
             self.baseurl = svn.ra.get_repos_root(self.ra)
             # Module is either empty or a repository path starting with
             # a slash and not ending with a slash.
             self.module = urlreq.unquote(self.url[len(self.baseurl) :])
             self.prevmodule = None
             self.rootmodule = self.module
             self.commits = {}
             self.paths = {}
             self.uuid = svn.ra.get_uuid(self.ra)
         except svn.core.SubversionException:
             ui.traceback()
             svnversion = b'%d.%d.%d' % (
                 svn.core.SVN_VER_MAJOR,
                 svn.core.SVN_VER_MINOR,
                 svn.core.SVN_VER_MICRO,
             )
             raise NoRepo(
                 _(
                     b"%s does not look like a Subversion repository "
                     b"to libsvn version %s"
                 )
                 % (self.url, svnversion)
             )
 
         if revs:
             if len(revs) > 1:
                 raise error.Abort(
                     _(
                         b'subversion source does not support '
                         b'specifying multiple revisions'
                     )
                 )
             try:
                 latest = int(revs[0])
             except ValueError:
                 raise error.Abort(
                     _(b'svn: revision %s is not an integer') % revs[0]
                 )
 
         trunkcfg = self.ui.config(b'convert', b'svn.trunk')
         if trunkcfg is None:
             trunkcfg = b'trunk'
         self.trunkname = trunkcfg.strip(b'/')
         self.startrev = self.ui.config(b'convert', b'svn.startrev')
         try:
             self.startrev = int(self.startrev)
             if self.startrev < 0:
                 self.startrev = 0
         except ValueError:
             raise error.Abort(
                 _(b'svn: start revision %s is not an integer') % self.startrev
             )
 
         try:
             self.head = self.latest(self.module, latest)
         except SvnPathNotFound:
             self.head = None
         if not self.head:
             raise error.Abort(
                 _(b'no revision found in module %s') % self.module
             )
         self.last_changed = self.revnum(self.head)
 
         self._changescache = (None, None)
 
         if os.path.exists(os.path.join(url, b'.svn/entries')):
             self.wc = url
         else:
             self.wc = None
         self.convertfp = None
 
     def setrevmap(self, revmap):
         lastrevs = {}
         for revid in revmap:
             uuid, module, revnum = revsplit(revid)
             lastrevnum = lastrevs.setdefault(module, revnum)
             if revnum > lastrevnum:
                 lastrevs[module] = revnum
         self.lastrevs = lastrevs
 
     def exists(self, path, optrev):
         try:
             svn.client.ls(
                 self.url.rstrip(b'/') + b'/' + quote(path),
                 optrev,
                 False,
                 self.ctx,
             )
             return True
         except svn.core.SubversionException:
             return False
 
     def getheads(self):
         def isdir(path, revnum):
             kind = self._checkpath(path, revnum)
             return kind == svn.core.svn_node_dir
 
         def getcfgpath(name, rev):
             cfgpath = self.ui.config(b'convert', b'svn.' + name)
             if cfgpath is not None and cfgpath.strip() == b'':
                 return None
             path = (cfgpath or name).strip(b'/')
             if not self.exists(path, rev):
                 if self.module.endswith(path) and name == b'trunk':
                     # we are converting from inside this directory
                     return None
                 if cfgpath:
                     raise error.Abort(
                         _(b'expected %s to be at %r, but not found')
                         % (name, path)
                     )
                 return None
             self.ui.note(_(b'found %s at %r\n') % (name, path))
             return path
 
         rev = optrev(self.last_changed)
         oldmodule = b''
         trunk = getcfgpath(b'trunk', rev)
         self.tags = getcfgpath(b'tags', rev)
         branches = getcfgpath(b'branches', rev)
 
         # If the project has a trunk or branches, we will extract heads
         # from them. We keep the project root otherwise.
         if trunk:
             oldmodule = self.module or b''
             self.module += b'/' + trunk
             self.head = self.latest(self.module, self.last_changed)
             if not self.head:
                 raise error.Abort(
                     _(b'no revision found in module %s') % self.module
                 )
 
         # First head in the list is the module's head
         self.heads = [self.head]
         if self.tags is not None:
             self.tags = b'%s/%s' % (oldmodule, (self.tags or b'tags'))
 
         # Check if branches bring a few more heads to the list
         if branches:
             rpath = self.url.strip(b'/')
             branchnames = svn.client.ls(
                 rpath + b'/' + quote(branches), rev, False, self.ctx
             )
             for branch in sorted(branchnames):
                 module = b'%s/%s/%s' % (oldmodule, branches, branch)
                 if not isdir(module, self.last_changed):
                     continue
                 brevid = self.latest(module, self.last_changed)
                 if not brevid:
                     self.ui.note(_(b'ignoring empty branch %s\n') % branch)
                     continue
                 self.ui.note(
                     _(b'found branch %s at %d\n')
                     % (branch, self.revnum(brevid))
                 )
                 self.heads.append(brevid)
 
         if self.startrev and self.heads:
             if len(self.heads) > 1:
                 raise error.Abort(
                     _(
                         b'svn: start revision is not supported '
                         b'with more than one branch'
                     )
                 )
             revnum = self.revnum(self.heads[0])
             if revnum < self.startrev:
                 raise error.Abort(
                     _(b'svn: no revision found after start revision %d')
                     % self.startrev
                 )
 
         return self.heads
 
     def _getchanges(self, rev, full):
         (paths, parents) = self.paths[rev]
         copies = {}
         if parents:
             files, self.removed, copies = self.expandpaths(rev, paths, parents)
         if full or not parents:
             # Perform a full checkout on roots
             uuid, module, revnum = revsplit(rev)
             entries = svn.client.ls(
                 self.baseurl + quote(module), optrev(revnum), True, self.ctx
             )
             files = [
                 n
                 for n, e in pycompat.iteritems(entries)
                 if e.kind == svn.core.svn_node_file
             ]
             self.removed = set()
 
         files.sort()
         files = zip(files, [rev] * len(files))
         return (files, copies)
 
     def getchanges(self, rev, full):
         # reuse cache from getchangedfiles
         if self._changescache[0] == rev and not full:
             (files, copies) = self._changescache[1]
         else:
             (files, copies) = self._getchanges(rev, full)
             # caller caches the result, so free it here to release memory
             del self.paths[rev]
         return (files, copies, set())
 
     def getchangedfiles(self, rev, i):
         # called from filemap - cache computed values for reuse in getchanges
         (files, copies) = self._getchanges(rev, False)
         self._changescache = (rev, (files, copies))
         return [f[0] for f in files]
 
     def getcommit(self, rev):
         if rev not in self.commits:
             uuid, module, revnum = revsplit(rev)
             self.module = module
             self.reparent(module)
             # We assume that:
             # - requests for revisions after "stop" come from the
             # revision graph backward traversal. Cache all of them
             # down to stop, they will be used eventually.
             # - requests for revisions before "stop" come to get
             # isolated branches parents. Just fetch what is needed.
             stop = self.lastrevs.get(module, 0)
             if revnum < stop:
                 stop = revnum + 1
             self._fetch_revisions(revnum, stop)
             if rev not in self.commits:
                 raise error.Abort(_(b'svn: revision %s not found') % revnum)
         revcommit = self.commits[rev]
         # caller caches the result, so free it here to release memory
         del self.commits[rev]
         return revcommit
 
     def checkrevformat(self, revstr, mapname=b'splicemap'):
         """ fails if revision format does not match the correct format"""
         if not re.match(
             r'svn:[0-9a-f]{8,8}-[0-9a-f]{4,4}-'
             r'[0-9a-f]{4,4}-[0-9a-f]{4,4}-[0-9a-f]'
             r'{12,12}(.*)@[0-9]+$',
             revstr,
         ):
             raise error.Abort(
                 _(b'%s entry %s is not a valid revision identifier')
                 % (mapname, revstr)
             )
 
     def numcommits(self):
         return int(self.head.rsplit(b'@', 1)[1]) - self.startrev
 
     def gettags(self):
         tags = {}
         if self.tags is None:
             return tags
 
         # svn tags are just a convention, project branches left in a
         # 'tags' directory. There is no other relationship than
         # ancestry, which is expensive to discover and makes them hard
         # to update incrementally.  Worse, past revisions may be
         # referenced by tags far away in the future, requiring a deep
         # history traversal on every calculation.  Current code
         # performs a single backward traversal, tracking moves within
         # the tags directory (tag renaming) and recording a new tag
         # everytime a project is copied from outside the tags
         # directory. It also lists deleted tags, this behaviour may
         # change in the future.
         pendings = []
         tagspath = self.tags
         start = svn.ra.get_latest_revnum(self.ra)
         stream = self._getlog([self.tags], start, self.startrev)
         try:
             for entry in stream:
                 origpaths, revnum, author, date, message = entry
                 if not origpaths:
                     origpaths = []
                 copies = [
                     (e.copyfrom_path, e.copyfrom_rev, p)
                     for p, e in pycompat.iteritems(origpaths)
                     if e.copyfrom_path
                 ]
                 # Apply moves/copies from more specific to general
                 copies.sort(reverse=True)
 
                 srctagspath = tagspath
                 if copies and copies[-1][2] == tagspath:
                     # Track tags directory moves
                     srctagspath = copies.pop()[0]
 
                 for source, sourcerev, dest in copies:
                     if not dest.startswith(tagspath + b'/'):
                         continue
                     for tag in pendings:
                         if tag[0].startswith(dest):
                             tagpath = source + tag[0][len(dest) :]
                             tag[:2] = [tagpath, sourcerev]
                             break
                     else:
                         pendings.append([source, sourcerev, dest])
 
                 # Filter out tags with children coming from different
                 # parts of the repository like:
                 # /tags/tag.1 (from /trunk:10)
                 # /tags/tag.1/foo (from /branches/foo:12)
                 # Here/tags/tag.1 discarded as well as its children.
                 # It happens with tools like cvs2svn. Such tags cannot
                 # be represented in mercurial.
                 addeds = {
                     p: e.copyfrom_path
                     for p, e in pycompat.iteritems(origpaths)
                     if e.action == b'A' and e.copyfrom_path
                 }
                 badroots = set()
                 for destroot in addeds:
                     for source, sourcerev, dest in pendings:
                         if not dest.startswith(
                             destroot + b'/'
                         ) or source.startswith(addeds[destroot] + b'/'):
                             continue
                         badroots.add(destroot)
                         break
 
                 for badroot in badroots:
                     pendings = [
                         p
                         for p in pendings
                         if p[2] != badroot
                         and not p[2].startswith(badroot + b'/')
                     ]
 
                 # Tell tag renamings from tag creations
                 renamings = []
                 for source, sourcerev, dest in pendings:
                     tagname = dest.split(b'/')[-1]
                     if source.startswith(srctagspath):
                         renamings.append([source, sourcerev, tagname])
                         continue
                     if tagname in tags:
                         # Keep the latest tag value
                         continue
                     # From revision may be fake, get one with changes
                     try:
                         tagid = self.latest(source, sourcerev)
                         if tagid and tagname not in tags:
                             tags[tagname] = tagid
                     except SvnPathNotFound:
                         # It happens when we are following directories
                         # we assumed were copied with their parents
                         # but were really created in the tag
                         # directory.
                         pass
                 pendings = renamings
                 tagspath = srctagspath
         finally:
             stream.close()
         return tags
 
     def converted(self, rev, destrev):
         if not self.wc:
             return
         if self.convertfp is None:
             self.convertfp = open(
                 os.path.join(self.wc, b'.svn', b'hg-shamap'), b'ab'
             )
         self.convertfp.write(
             util.tonativeeol(b'%s %d\n' % (destrev, self.revnum(rev)))
         )
         self.convertfp.flush()
 
     def revid(self, revnum, module=None):
         return b'svn:%s%s@%s' % (self.uuid, module or self.module, revnum)
 
     def revnum(self, rev):
         return int(rev.split(b'@')[-1])
 
     def latest(self, path, stop=None):
         """Find the latest revid affecting path, up to stop revision
         number. If stop is None, default to repository latest
         revision. It may return a revision in a different module,
         since a branch may be moved without a change being
         reported. Return None if computed module does not belong to
         rootmodule subtree.
         """
 
         def findchanges(path, start, stop=None):
             stream = self._getlog([path], start, stop or 1)
             try:
                 for entry in stream:
                     paths, revnum, author, date, message = entry
                     if stop is None and paths:
                         # We do not know the latest changed revision,
                         # keep the first one with changed paths.
                         break
                     if revnum <= stop:
                         break
 
                     for p in paths:
                         if not path.startswith(p) or not paths[p].copyfrom_path:
                             continue
                         newpath = paths[p].copyfrom_path + path[len(p) :]
                         self.ui.debug(
                             b"branch renamed from %s to %s at %d\n"
                             % (path, newpath, revnum)
                         )
                         path = newpath
                         break
                 if not paths:
                     revnum = None
                 return revnum, path
             finally:
                 stream.close()
 
         if not path.startswith(self.rootmodule):
             # Requests on foreign branches may be forbidden at server level
             self.ui.debug(b'ignoring foreign branch %r\n' % path)
             return None
 
         if stop is None:
             stop = svn.ra.get_latest_revnum(self.ra)
         try:
             prevmodule = self.reparent(b'')
             dirent = svn.ra.stat(self.ra, path.strip(b'/'), stop)
             self.reparent(prevmodule)
         except svn.core.SubversionException:
             dirent = None
         if not dirent:
             raise SvnPathNotFound(
                 _(b'%s not found up to revision %d') % (path, stop)
             )
 
         # stat() gives us the previous revision on this line of
         # development, but it might be in *another module*. Fetch the
         # log and detect renames down to the latest revision.
         revnum, realpath = findchanges(path, stop, dirent.created_rev)
         if revnum is None:
             # Tools like svnsync can create empty revision, when
             # synchronizing only a subtree for instance. These empty
             # revisions created_rev still have their original values
             # despite all changes having disappeared and can be
             # returned by ra.stat(), at least when stating the root
             # module. In that case, do not trust created_rev and scan
             # the whole history.
             revnum, realpath = findchanges(path, stop)
             if revnum is None:
                 self.ui.debug(b'ignoring empty branch %r\n' % realpath)
                 return None
 
         if not realpath.startswith(self.rootmodule):
             self.ui.debug(b'ignoring foreign branch %r\n' % realpath)
             return None
         return self.revid(revnum, realpath)
 
     def reparent(self, module):
         """Reparent the svn transport and return the previous parent."""
         if self.prevmodule == module:
             return module
         svnurl = self.baseurl + quote(module)
         prevmodule = self.prevmodule
         if prevmodule is None:
             prevmodule = b''
         self.ui.debug(b"reparent to %s\n" % svnurl)
         svn.ra.reparent(self.ra, svnurl)
         self.prevmodule = module
         return prevmodule
 
     def expandpaths(self, rev, paths, parents):
         changed, removed = set(), set()
         copies = {}
 
         new_module, revnum = revsplit(rev)[1:]
         if new_module != self.module:
             self.module = new_module
             self.reparent(self.module)
 
         progress = self.ui.makeprogress(
             _(b'scanning paths'), unit=_(b'paths'), total=len(paths)
         )
         for i, (path, ent) in enumerate(paths):
             progress.update(i, item=path)
             entrypath = self.getrelpath(path)
 
             kind = self._checkpath(entrypath, revnum)
             if kind == svn.core.svn_node_file:
                 changed.add(self.recode(entrypath))
                 if not ent.copyfrom_path or not parents:
                     continue
                 # Copy sources not in parent revisions cannot be
                 # represented, ignore their origin for now
                 pmodule, prevnum = revsplit(parents[0])[1:]
                 if ent.copyfrom_rev < prevnum:
                     continue
                 copyfrom_path = self.getrelpath(ent.copyfrom_path, pmodule)
                 if not copyfrom_path:
                     continue
                 self.ui.debug(
                     b"copied to %s from %s@%s\n"
                     % (entrypath, copyfrom_path, ent.copyfrom_rev)
                 )
                 copies[self.recode(entrypath)] = self.recode(copyfrom_path)
             elif kind == 0:  # gone, but had better be a deleted *file*
                 self.ui.debug(b"gone from %s\n" % ent.copyfrom_rev)
                 pmodule, prevnum = revsplit(parents[0])[1:]
                 parentpath = pmodule + b"/" + entrypath
                 fromkind = self._checkpath(entrypath, prevnum, pmodule)
 
                 if fromkind == svn.core.svn_node_file:
                     removed.add(self.recode(entrypath))
                 elif fromkind == svn.core.svn_node_dir:
                     oroot = parentpath.strip(b'/')
                     nroot = path.strip(b'/')
                     children = self._iterfiles(oroot, prevnum)
                     for childpath in children:
                         childpath = childpath.replace(oroot, nroot)
                         childpath = self.getrelpath(b"/" + childpath, pmodule)
                         if childpath:
                             removed.add(self.recode(childpath))
                 else:
                     self.ui.debug(
                         b'unknown path in revision %d: %s\n' % (revnum, path)
                     )
             elif kind == svn.core.svn_node_dir:
                 if ent.action == b'M':
                     # If the directory just had a prop change,
                     # then we shouldn't need to look for its children.
                     continue
                 if ent.action == b'R' and parents:
                     # If a directory is replacing a file, mark the previous
                     # file as deleted
                     pmodule, prevnum = revsplit(parents[0])[1:]
                     pkind = self._checkpath(entrypath, prevnum, pmodule)
                     if pkind == svn.core.svn_node_file:
                         removed.add(self.recode(entrypath))
                     elif pkind == svn.core.svn_node_dir:
                         # We do not know what files were kept or removed,
                         # mark them all as changed.
                         for childpath in self._iterfiles(pmodule, prevnum):
                             childpath = self.getrelpath(b"/" + childpath)
                             if childpath:
                                 changed.add(self.recode(childpath))
 
                 for childpath in self._iterfiles(path, revnum):
                     childpath = self.getrelpath(b"/" + childpath)
                     if childpath:
                         changed.add(self.recode(childpath))
 
                 # Handle directory copies
                 if not ent.copyfrom_path or not parents:
                     continue
                 # Copy sources not in parent revisions cannot be
                 # represented, ignore their origin for now
                 pmodule, prevnum = revsplit(parents[0])[1:]
                 if ent.copyfrom_rev < prevnum:
                     continue
                 copyfrompath = self.getrelpath(ent.copyfrom_path, pmodule)
                 if not copyfrompath:
                     continue
                 self.ui.debug(
                     b"mark %s came from %s:%d\n"
                     % (path, copyfrompath, ent.copyfrom_rev)
                 )
                 children = self._iterfiles(ent.copyfrom_path, ent.copyfrom_rev)
                 for childpath in children:
                     childpath = self.getrelpath(b"/" + childpath, pmodule)
                     if not childpath:
                         continue
                     copytopath = path + childpath[len(copyfrompath) :]
                     copytopath = self.getrelpath(copytopath)
                     copies[self.recode(copytopath)] = self.recode(childpath)
 
         progress.complete()
         changed.update(removed)
         return (list(changed), removed, copies)
 
     def _fetch_revisions(self, from_revnum, to_revnum):
         if from_revnum < to_revnum:
             from_revnum, to_revnum = to_revnum, from_revnum
 
         self.child_cset = None
 
         def parselogentry(orig_paths, revnum, author, date, message):
             """Return the parsed commit object or None, and True if
             the revision is a branch root.
             """
             self.ui.debug(
                 b"parsing revision %d (%d changes)\n"
                 % (revnum, len(orig_paths))
             )
 
             branched = False
             rev = self.revid(revnum)
             # branch log might return entries for a parent we already have
 
             if rev in self.commits or revnum < to_revnum:
                 return None, branched
 
             parents = []
             # check whether this revision is the start of a branch or part
             # of a branch renaming
             orig_paths = sorted(pycompat.iteritems(orig_paths))
             root_paths = [
                 (p, e) for p, e in orig_paths if self.module.startswith(p)
             ]
             if root_paths:
                 path, ent = root_paths[-1]
                 if ent.copyfrom_path:
                     branched = True
                     newpath = ent.copyfrom_path + self.module[len(path) :]
                     # ent.copyfrom_rev may not be the actual last revision
                     previd = self.latest(newpath, ent.copyfrom_rev)
                     if previd is not None:
                         prevmodule, prevnum = revsplit(previd)[1:]
                         if prevnum >= self.startrev:
                             parents = [previd]
                             self.ui.note(
                                 _(b'found parent of branch %s at %d: %s\n')
                                 % (self.module, prevnum, prevmodule)
                             )
                 else:
                     self.ui.debug(b"no copyfrom path, don't know what to do.\n")
 
             paths = []
             # filter out unrelated paths
             for path, ent in orig_paths:
                 if self.getrelpath(path) is None:
                     continue
                 paths.append((path, ent))
 
             # Example SVN datetime. Includes microseconds.
             # ISO-8601 conformant
             # '2007-01-04T17:35:00.902377Z'
             date = dateutil.parsedate(
                 date[:19] + b" UTC", [b"%Y-%m-%dT%H:%M:%S"]
             )
             if self.ui.configbool(b'convert', b'localtimezone'):
                 date = makedatetimestamp(date[0])
 
             if message:
                 log = self.recode(message)
             else:
                 log = b''
 
             if author:
                 author = self.recode(author)
             else:
                 author = b''
 
             try:
                 branch = self.module.split(b"/")[-1]
                 if branch == self.trunkname:
                     branch = None
             except IndexError:
                 branch = None
 
             cset = commit(
                 author=author,
                 date=dateutil.datestr(date, b'%Y-%m-%d %H:%M:%S %1%2'),
                 desc=log,
                 parents=parents,
                 branch=branch,
                 rev=rev,
             )
 
             self.commits[rev] = cset
             # The parents list is *shared* among self.paths and the
             # commit object. Both will be updated below.
             self.paths[rev] = (paths, cset.parents)
             if self.child_cset and not self.child_cset.parents:
                 self.child_cset.parents[:] = [rev]
             self.child_cset = cset
             return cset, branched
 
         self.ui.note(
             _(b'fetching revision log for "%s" from %d to %d\n')
             % (self.module, from_revnum, to_revnum)
         )
 
         try:
             firstcset = None
             lastonbranch = False
             stream = self._getlog([self.module], from_revnum, to_revnum)
             try:
                 for entry in stream:
                     paths, revnum, author, date, message = entry
                     if revnum < self.startrev:
                         lastonbranch = True
                         break
                     if not paths:
                         self.ui.debug(b'revision %d has no entries\n' % revnum)
                         # If we ever leave the loop on an empty
                         # revision, do not try to get a parent branch
                         lastonbranch = lastonbranch or revnum == 0
                         continue
                     cset, lastonbranch = parselogentry(
                         paths, revnum, author, date, message
                     )
                     if cset:
                         firstcset = cset
                     if lastonbranch:
                         break
             finally:
                 stream.close()
 
             if not lastonbranch and firstcset and not firstcset.parents:
                 # The first revision of the sequence (the last fetched one)
                 # has invalid parents if not a branch root. Find the parent
                 # revision now, if any.
                 try:
                     firstrevnum = self.revnum(firstcset.rev)
                     if firstrevnum > 1:
                         latest = self.latest(self.module, firstrevnum - 1)
                         if latest:
                             firstcset.parents.append(latest)
                 except SvnPathNotFound:
                     pass
         except svn.core.SubversionException as xxx_todo_changeme:
             (inst, num) = xxx_todo_changeme.args
             if num == svn.core.SVN_ERR_FS_NO_SUCH_REVISION:
                 raise error.Abort(
                     _(b'svn: branch has no revision %s') % to_revnum
                 )
             raise
 
     def getfile(self, file, rev):
         # TODO: ra.get_file transmits the whole file instead of diffs.
         if file in self.removed:
             return None, None
         try:
             new_module, revnum = revsplit(rev)[1:]
             if self.module != new_module:
                 self.module = new_module
                 self.reparent(self.module)
             io = stringio()
             info = svn.ra.get_file(self.ra, file, revnum, io)
             data = io.getvalue()
             # ra.get_file() seems to keep a reference on the input buffer
             # preventing collection. Release it explicitly.
             io.close()
             if isinstance(info, list):
                 info = info[-1]
             mode = (b"svn:executable" in info) and b'x' or b''
             mode = (b"svn:special" in info) and b'l' or mode
         except svn.core.SubversionException as e:
             notfound = (
                 svn.core.SVN_ERR_FS_NOT_FOUND,
                 svn.core.SVN_ERR_RA_DAV_PATH_NOT_FOUND,
             )
             if e.apr_err in notfound:  # File not found
                 return None, None
             raise
         if mode == b'l':
             link_prefix = b"link "
             if data.startswith(link_prefix):
                 data = data[len(link_prefix) :]
         return data, mode
 
     def _iterfiles(self, path, revnum):
         """Enumerate all files in path at revnum, recursively."""
         path = path.strip(b'/')
         pool = svn.core.Pool()
         rpath = b'/'.join([self.baseurl, quote(path)]).strip(b'/')
         entries = svn.client.ls(rpath, optrev(revnum), True, self.ctx, pool)
         if path:
             path += b'/'
         return (
             (path + p)
             for p, e in pycompat.iteritems(entries)
             if e.kind == svn.core.svn_node_file
         )
 
     def getrelpath(self, path, module=None):
         if module is None:
             module = self.module
         # Given the repository url of this wc, say
         #   "http://server/plone/CMFPlone/branches/Plone-2_0-branch"
         # extract the "entry" portion (a relative path) from what
         # svn log --xml says, i.e.
         #   "/CMFPlone/branches/Plone-2_0-branch/tests/PloneTestCase.py"
         # that is to say "tests/PloneTestCase.py"
         if path.startswith(module):
             relative = path.rstrip(b'/')[len(module) :]
             if relative.startswith(b'/'):
                 return relative[1:]
             elif relative == b'':
                 return relative
 
         # The path is outside our tracked tree...
         self.ui.debug(b'%r is not under %r, ignoring\n' % (path, module))
         return None
 
     def _checkpath(self, path, revnum, module=None):
         if module is not None:
             prevmodule = self.reparent(b'')
             path = module + b'/' + path
         try:
             # ra.check_path does not like leading slashes very much, it leads
             # to PROPFIND subversion errors
             return svn.ra.check_path(self.ra, path.strip(b'/'), revnum)
         finally:
             if module is not None:
                 self.reparent(prevmodule)
 
     def _getlog(
         self,
         paths,
         start,
         end,
         limit=0,
         discover_changed_paths=True,
         strict_node_history=False,
     ):
         # Normalize path names, svn >= 1.5 only wants paths relative to
         # supplied URL
         relpaths = []
         for p in paths:
             if not p.startswith(b'/'):
                 p = self.module + b'/' + p
             relpaths.append(p.strip(b'/'))
         args = [
             self.baseurl,
             relpaths,
             start,
             end,
             limit,
             discover_changed_paths,
             strict_node_history,
         ]
         # developer config: convert.svn.debugsvnlog
         if not self.ui.configbool(b'convert', b'svn.debugsvnlog'):
             return directlogstream(*args)
         arg = encodeargs(args)
         hgexe = procutil.hgexecutable()
         cmd = b'%s debugsvnlog' % procutil.shellquote(hgexe)
-        stdin, stdout = procutil.popen2(procutil.quotecommand(cmd))
+        stdin, stdout = procutil.popen2(cmd)
         stdin.write(arg)
         try:
             stdin.close()
         except IOError:
             raise error.Abort(
                 _(
                     b'Mercurial failed to run itself, check'
                     b' hg executable is in PATH'
                 )
             )
         return logstream(stdout)
 
 
 pre_revprop_change = b'''#!/bin/sh
 
 REPOS="$1"
 REV="$2"
 USER="$3"
 PROPNAME="$4"
 ACTION="$5"
 
 if [ "$ACTION" = "M" -a "$PROPNAME" = "svn:log" ]; then exit 0; fi
 if [ "$ACTION" = "A" -a "$PROPNAME" = "hg:convert-branch" ]; then exit 0; fi
 if [ "$ACTION" = "A" -a "$PROPNAME" = "hg:convert-rev" ]; then exit 0; fi
 
 echo "Changing prohibited revision property" >&2
 exit 1
 '''
 
 
 class svn_sink(converter_sink, commandline):
     commit_re = re.compile(br'Committed revision (\d+).', re.M)
     uuid_re = re.compile(br'Repository UUID:\s*(\S+)', re.M)
 
     def prerun(self):
         if self.wc:
             os.chdir(self.wc)
 
     def postrun(self):
         if self.wc:
             os.chdir(self.cwd)
 
     def join(self, name):
         return os.path.join(self.wc, b'.svn', name)
 
     def revmapfile(self):
         return self.join(b'hg-shamap')
 
     def authorfile(self):
         return self.join(b'hg-authormap')
 
     def __init__(self, ui, repotype, path):
 
         converter_sink.__init__(self, ui, repotype, path)
         commandline.__init__(self, ui, b'svn')
         self.delete = []
         self.setexec = []
         self.delexec = []
         self.copies = []
         self.wc = None
         self.cwd = encoding.getcwd()
 
         created = False
         if os.path.isfile(os.path.join(path, b'.svn', b'entries')):
             self.wc = os.path.realpath(path)
             self.run0(b'update')
         else:
             if not re.search(br'^(file|http|https|svn|svn\+ssh)://', path):
                 path = os.path.realpath(path)
                 if os.path.isdir(os.path.dirname(path)):
                     if not os.path.exists(
                         os.path.join(path, b'db', b'fs-type')
                     ):
                         ui.status(
                             _(b"initializing svn repository '%s'\n")
                             % os.path.basename(path)
                         )
                         commandline(ui, b'svnadmin').run0(b'create', path)
                         created = path
                     path = util.normpath(path)
                     if not path.startswith(b'/'):
                         path = b'/' + path
                     path = b'file://' + path
 
             wcpath = os.path.join(
                 encoding.getcwd(), os.path.basename(path) + b'-wc'
             )
             ui.status(
                 _(b"initializing svn working copy '%s'\n")
                 % os.path.basename(wcpath)
             )
             self.run0(b'checkout', path, wcpath)
 
             self.wc = wcpath
         self.opener = vfsmod.vfs(self.wc)
         self.wopener = vfsmod.vfs(self.wc)
         self.childmap = mapfile(ui, self.join(b'hg-childmap'))
         if util.checkexec(self.wc):
             self.is_exec = util.isexec
         else:
             self.is_exec = None
 
         if created:
             hook = os.path.join(created, b'hooks', b'pre-revprop-change')
             fp = open(hook, b'wb')
             fp.write(pre_revprop_change)
             fp.close()
             util.setflags(hook, False, True)
 
         output = self.run0(b'info')
         self.uuid = self.uuid_re.search(output).group(1).strip()
 
     def wjoin(self, *names):
         return os.path.join(self.wc, *names)
 
     @propertycache
     def manifest(self):
         # As of svn 1.7, the "add" command fails when receiving
         # already tracked entries, so we have to track and filter them
         # ourselves.
         m = set()
         output = self.run0(b'ls', recursive=True, xml=True)
         doc = xml.dom.minidom.parseString(output)
         for e in doc.getElementsByTagName('entry'):
             for n in e.childNodes:
                 if n.nodeType != n.ELEMENT_NODE or n.tagName != 'name':
                     continue
                 name = ''.join(
                     c.data for c in n.childNodes if c.nodeType == c.TEXT_NODE
                 )
                 # Entries are compared with names coming from
                 # mercurial, so bytes with undefined encoding. Our
                 # best bet is to assume they are in local
                 # encoding. They will be passed to command line calls
                 # later anyway, so they better be.
                 m.add(encoding.unitolocal(name))
                 break
         return m
 
     def putfile(self, filename, flags, data):
         if b'l' in flags:
             self.wopener.symlink(data, filename)
         else:
             try:
                 if os.path.islink(self.wjoin(filename)):
                     os.unlink(filename)
             except OSError:
                 pass
 
             if self.is_exec:
                 # We need to check executability of the file before the change,
                 # because `vfs.write` is able to reset exec bit.
                 wasexec = False
                 if os.path.exists(self.wjoin(filename)):
                     wasexec = self.is_exec(self.wjoin(filename))
 
             self.wopener.write(filename, data)
 
             if self.is_exec:
                 if wasexec:
                     if b'x' not in flags:
                         self.delexec.append(filename)
                 else:
                     if b'x' in flags:
                         self.setexec.append(filename)
                 util.setflags(self.wjoin(filename), False, b'x' in flags)
 
     def _copyfile(self, source, dest):
         # SVN's copy command pukes if the destination file exists, but
         # our copyfile method expects to record a copy that has
         # already occurred.  Cross the semantic gap.
         wdest = self.wjoin(dest)
         exists = os.path.lexists(wdest)
         if exists:
             fd, tempname = pycompat.mkstemp(
                 prefix=b'hg-copy-', dir=os.path.dirname(wdest)
             )
             os.close(fd)
             os.unlink(tempname)
             os.rename(wdest, tempname)
         try:
             self.run0(b'copy', source, dest)
         finally:
             self.manifest.add(dest)
             if exists:
                 try:
                     os.unlink(wdest)
                 except OSError:
                     pass
                 os.rename(tempname, wdest)
 
     def dirs_of(self, files):
         dirs = set()
         for f in files:
             if os.path.isdir(self.wjoin(f)):
                 dirs.add(f)
             i = len(f)
             for i in iter(lambda: f.rfind(b'/', 0, i), -1):
                 dirs.add(f[:i])
         return dirs
 
     def add_dirs(self, files):
         add_dirs = [
             d for d in sorted(self.dirs_of(files)) if d not in self.manifest
         ]
         if add_dirs:
             self.manifest.update(add_dirs)
             self.xargs(add_dirs, b'add', non_recursive=True, quiet=True)
         return add_dirs
 
     def add_files(self, files):
         files = [f for f in files if f not in self.manifest]
         if files:
             self.manifest.update(files)
             self.xargs(files, b'add', quiet=True)
         return files
 
     def addchild(self, parent, child):
         self.childmap[parent] = child
 
     def revid(self, rev):
         return b"svn:%s@%s" % (self.uuid, rev)
 
     def putcommit(
         self, files, copies, parents, commit, source, revmap, full, cleanp2
     ):
         for parent in parents:
             try:
                 return self.revid(self.childmap[parent])
             except KeyError:
                 pass
 
         # Apply changes to working copy
         for f, v in files:
             data, mode = source.getfile(f, v)
             if data is None:
                 self.delete.append(f)
             else:
                 self.putfile(f, mode, data)
                 if f in copies:
                     self.copies.append([copies[f], f])
         if full:
             self.delete.extend(sorted(self.manifest.difference(files)))
         files = [f[0] for f in files]
 
         entries = set(self.delete)
         files = frozenset(files)
         entries.update(self.add_dirs(files.difference(entries)))
         if self.copies:
             for s, d in self.copies:
                 self._copyfile(s, d)
             self.copies = []
         if self.delete:
             self.xargs(self.delete, b'delete')
             for f in self.delete:
                 self.manifest.remove(f)
             self.delete = []
         entries.update(self.add_files(files.difference(entries)))
         if self.delexec:
             self.xargs(self.delexec, b'propdel', b'svn:executable')
             self.delexec = []
         if self.setexec:
             self.xargs(self.setexec, b'propset', b'svn:executable', b'*')
             self.setexec = []
 
         fd, messagefile = pycompat.mkstemp(prefix=b'hg-convert-')
         fp = os.fdopen(fd, 'wb')
         fp.write(util.tonativeeol(commit.desc))
         fp.close()
         try:
             output = self.run0(
                 b'commit',
                 username=stringutil.shortuser(commit.author),
                 file=messagefile,
                 encoding=b'utf-8',
             )
             try:
                 rev = self.commit_re.search(output).group(1)
             except AttributeError:
                 if not files:
                     return parents[0] if parents else b'None'
                 self.ui.warn(_(b'unexpected svn output:\n'))
                 self.ui.warn(output)
                 raise error.Abort(_(b'unable to cope with svn output'))
             if commit.rev:
                 self.run(
                     b'propset',
                     b'hg:convert-rev',
                     commit.rev,
                     revprop=True,
                     revision=rev,
                 )
             if commit.branch and commit.branch != b'default':
                 self.run(
                     b'propset',
                     b'hg:convert-branch',
                     commit.branch,
                     revprop=True,
                     revision=rev,
                 )
             for parent in parents:
                 self.addchild(parent, rev)
             return self.revid(rev)
         finally:
             os.unlink(messagefile)
 
     def puttags(self, tags):
         self.ui.warn(_(b'writing Subversion tags is not yet implemented\n'))
         return None, None
 
     def hascommitfrommap(self, rev):
         # We trust that revisions referenced in a map still is present
         # TODO: implement something better if necessary and feasible
         return True
 
     def hascommitforsplicemap(self, rev):
         # This is not correct as one can convert to an existing subversion
         # repository and childmap would not list all revisions. Too bad.
         if rev in self.childmap:
             return True
         raise error.Abort(
             _(
                 b'splice map revision %s not found in subversion '
                 b'child map (revision lookups are not implemented)'
             )
             % rev
         )

hgext/extdiff.py

@@ -1,720 +1,719 b''
 # extdiff.py - external diff program support for mercurial
 #
 # Copyright 2006 Vadim Gelfer <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 '''command to allow external programs to compare revisions
 
 The extdiff Mercurial extension allows you to use external programs
 to compare revisions, or revision with working directory. The external
 diff programs are called with a configurable set of options and two
 non-option arguments: paths to directories containing snapshots of
 files to compare.
 
 If there is more than one file being compared and the "child" revision
 is the working directory, any modifications made in the external diff
 program will be copied back to the working directory from the temporary
 directory.
 
 The extdiff extension also allows you to configure new diff commands, so
 you do not need to type :hg:`extdiff -p kdiff3` always. ::
 
   [extdiff]
   # add new command that runs GNU diff(1) in 'context diff' mode
   cdiff = gdiff -Nprc5
   ## or the old way:
   #cmd.cdiff = gdiff
   #opts.cdiff = -Nprc5
 
   # add new command called meld, runs meld (no need to name twice).  If
   # the meld executable is not available, the meld tool in [merge-tools]
   # will be used, if available
   meld =
 
   # add new command called vimdiff, runs gvimdiff with DirDiff plugin
   # (see http://www.vim.org/scripts/script.php?script_id=102) Non
   # English user, be sure to put "let g:DirDiffDynamicDiffText = 1" in
   # your .vimrc
   vimdiff = gvim -f "+next" \\
             "+execute 'DirDiff' fnameescape(argv(0)) fnameescape(argv(1))"
 
 Tool arguments can include variables that are expanded at runtime::
 
   $parent1, $plabel1 - filename, descriptive label of first parent
   $child,   $clabel  - filename, descriptive label of child revision
   $parent2, $plabel2 - filename, descriptive label of second parent
   $root              - repository root
   $parent is an alias for $parent1.
 
 The extdiff extension will look in your [diff-tools] and [merge-tools]
 sections for diff tool arguments, when none are specified in [extdiff].
 
 ::
 
   [extdiff]
   kdiff3 =
 
   [diff-tools]
   kdiff3.diffargs=--L1 '$plabel1' --L2 '$clabel' $parent $child
 
 If a program has a graphical interface, it might be interesting to tell
 Mercurial about it. It will prevent the program from being mistakenly
 used in a terminal-only environment (such as an SSH terminal session),
 and will make :hg:`extdiff --per-file` open multiple file diffs at once
 instead of one by one (if you still want to open file diffs one by one,
 you can use the --confirm option).
 
 Declaring that a tool has a graphical interface can be done with the
 ``gui`` flag next to where ``diffargs`` are specified:
 
 ::
 
   [diff-tools]
   kdiff3.diffargs=--L1 '$plabel1' --L2 '$clabel' $parent $child
   kdiff3.gui = true
 
 You can use -I/-X and list of file or directory names like normal
 :hg:`diff` command. The extdiff extension makes snapshots of only
 needed files, so running the external diff program will actually be
 pretty fast (at least faster than having to compare the entire tree).
 '''
 
 from __future__ import absolute_import
 
 import os
 import re
 import shutil
 import stat
 import subprocess
 
 from mercurial.i18n import _
 from mercurial.node import (
     nullid,
     short,
 )
 from mercurial import (
     archival,
     cmdutil,
     encoding,
     error,
     filemerge,
     formatter,
     pycompat,
     registrar,
     scmutil,
     util,
 )
 from mercurial.utils import (
     procutil,
     stringutil,
 )
 
 cmdtable = {}
 command = registrar.command(cmdtable)
 
 configtable = {}
 configitem = registrar.configitem(configtable)
 
 configitem(
     b'extdiff', br'opts\..*', default=b'', generic=True,
 )
 
 configitem(
     b'extdiff', br'gui\..*', generic=True,
 )
 
 configitem(
     b'diff-tools', br'.*\.diffargs$', default=None, generic=True,
 )
 
 configitem(
     b'diff-tools', br'.*\.gui$', generic=True,
 )
 
 # Note for extension authors: ONLY specify testedwith = 'ships-with-hg-core' for
 # extensions which SHIP WITH MERCURIAL. Non-mainline extensions should
 # be specifying the version(s) of Mercurial they are tested with, or
 # leave the attribute unspecified.
 testedwith = b'ships-with-hg-core'
 
 
 def snapshot(ui, repo, files, node, tmproot, listsubrepos):
     '''snapshot files as of some revision
     if not using snapshot, -I/-X does not work and recursive diff
     in tools like kdiff3 and meld displays too many files.'''
     dirname = os.path.basename(repo.root)
     if dirname == b"":
         dirname = b"root"
     if node is not None:
         dirname = b'%s.%s' % (dirname, short(node))
     base = os.path.join(tmproot, dirname)
     os.mkdir(base)
     fnsandstat = []
 
     if node is not None:
         ui.note(
             _(b'making snapshot of %d files from rev %s\n')
             % (len(files), short(node))
         )
     else:
         ui.note(
             _(b'making snapshot of %d files from working directory\n')
             % (len(files))
         )
 
     if files:
         repo.ui.setconfig(b"ui", b"archivemeta", False)
 
         archival.archive(
             repo,
             base,
             node,
             b'files',
             match=scmutil.matchfiles(repo, files),
             subrepos=listsubrepos,
         )
 
         for fn in sorted(files):
             wfn = util.pconvert(fn)
             ui.note(b'  %s\n' % wfn)
 
             if node is None:
                 dest = os.path.join(base, wfn)
 
                 fnsandstat.append((dest, repo.wjoin(fn), os.lstat(dest)))
     return dirname, fnsandstat
 
 
 def formatcmdline(
     cmdline,
     repo_root,
     do3way,
     parent1,
     plabel1,
     parent2,
     plabel2,
     child,
     clabel,
 ):
     # Function to quote file/dir names in the argument string.
     # When not operating in 3-way mode, an empty string is
     # returned for parent2
     replace = {
         b'parent': parent1,
         b'parent1': parent1,
         b'parent2': parent2,
         b'plabel1': plabel1,
         b'plabel2': plabel2,
         b'child': child,
         b'clabel': clabel,
         b'root': repo_root,
     }
 
     def quote(match):
         pre = match.group(2)
         key = match.group(3)
         if not do3way and key == b'parent2':
             return pre
         return pre + procutil.shellquote(replace[key])
 
     # Match parent2 first, so 'parent1?' will match both parent1 and parent
     regex = (
         br'''(['"]?)([^\s'"$]*)'''
         br'\$(parent2|parent1?|child|plabel1|plabel2|clabel|root)\1'
     )
     if not do3way and not re.search(regex, cmdline):
         cmdline += b' $parent1 $child'
     return re.sub(regex, quote, cmdline)
 
 
 def _systembackground(cmd, environ=None, cwd=None):
     ''' like 'procutil.system', but returns the Popen object directly
         so we don't have to wait on it.
     '''
-    cmd = procutil.quotecommand(cmd)
     env = procutil.shellenviron(environ)
     proc = subprocess.Popen(
         procutil.tonativestr(cmd),
         shell=True,
         close_fds=procutil.closefds,
         env=procutil.tonativeenv(env),
         cwd=pycompat.rapply(procutil.tonativestr, cwd),
     )
     return proc
 
 
 def _runperfilediff(
     cmdline,
     repo_root,
     ui,
     guitool,
     do3way,
     confirm,
     commonfiles,
     tmproot,
     dir1a,
     dir1b,
     dir2root,
     dir2,
     rev1a,
     rev1b,
     rev2,
 ):
     # Note that we need to sort the list of files because it was
     # built in an "unstable" way and it's annoying to get files in a
     # random order, especially when "confirm" mode is enabled.
     waitprocs = []
     totalfiles = len(commonfiles)
     for idx, commonfile in enumerate(sorted(commonfiles)):
         path1a = os.path.join(tmproot, dir1a, commonfile)
         label1a = commonfile + rev1a
         if not os.path.isfile(path1a):
             path1a = pycompat.osdevnull
 
         path1b = b''
         label1b = b''
         if do3way:
             path1b = os.path.join(tmproot, dir1b, commonfile)
             label1b = commonfile + rev1b
             if not os.path.isfile(path1b):
                 path1b = pycompat.osdevnull
 
         path2 = os.path.join(dir2root, dir2, commonfile)
         label2 = commonfile + rev2
 
         if confirm:
             # Prompt before showing this diff
             difffiles = _(b'diff %s (%d of %d)') % (
                 commonfile,
                 idx + 1,
                 totalfiles,
             )
             responses = _(
                 b'[Yns?]'
                 b'$$ &Yes, show diff'
                 b'$$ &No, skip this diff'
                 b'$$ &Skip remaining diffs'
                 b'$$ &? (display help)'
             )
             r = ui.promptchoice(b'%s %s' % (difffiles, responses))
             if r == 3:  # ?
                 while r == 3:
                     for c, t in ui.extractchoices(responses)[1]:
                         ui.write(b'%s - %s\n' % (c, encoding.lower(t)))
                     r = ui.promptchoice(b'%s %s' % (difffiles, responses))
             if r == 0:  # yes
                 pass
             elif r == 1:  # no
                 continue
             elif r == 2:  # skip
                 break
 
         curcmdline = formatcmdline(
             cmdline,
             repo_root,
             do3way=do3way,
             parent1=path1a,
             plabel1=label1a,
             parent2=path1b,
             plabel2=label1b,
             child=path2,
             clabel=label2,
         )
 
         if confirm or not guitool:
             # Run the comparison program and wait for it to exit
             # before we show the next file.
             # This is because either we need to wait for confirmation
             # from the user between each invocation, or because, as far
             # as we know, the tool doesn't have a GUI, in which case
             # we can't run multiple CLI programs at the same time.
             ui.debug(
                 b'running %r in %s\n' % (pycompat.bytestr(curcmdline), tmproot)
             )
             ui.system(curcmdline, cwd=tmproot, blockedtag=b'extdiff')
         else:
             # Run the comparison program but don't wait, as we're
             # going to rapid-fire each file diff and then wait on
             # the whole group.
             ui.debug(
                 b'running %r in %s (backgrounded)\n'
                 % (pycompat.bytestr(curcmdline), tmproot)
             )
             proc = _systembackground(curcmdline, cwd=tmproot)
             waitprocs.append(proc)
 
     if waitprocs:
         with ui.timeblockedsection(b'extdiff'):
             for proc in waitprocs:
                 proc.wait()
 
 
 def dodiff(ui, repo, cmdline, pats, opts, guitool=False):
     '''Do the actual diff:
 
     - copy to a temp structure if diffing 2 internal revisions
     - copy to a temp structure if diffing working revision with
       another one and more than 1 file is changed
     - just invoke the diff for a single file in the working dir
     '''
 
     cmdutil.check_at_most_one_arg(opts, b'rev', b'change')
     revs = opts.get(b'rev')
     change = opts.get(b'change')
     do3way = b'$parent2' in cmdline
 
     if change:
         ctx2 = scmutil.revsingle(repo, change, None)
         ctx1a, ctx1b = ctx2.p1(), ctx2.p2()
     else:
         ctx1a, ctx2 = scmutil.revpair(repo, revs)
         if not revs:
             ctx1b = repo[None].p2()
         else:
             ctx1b = repo[nullid]
 
     perfile = opts.get(b'per_file')
     confirm = opts.get(b'confirm')
 
     node1a = ctx1a.node()
     node1b = ctx1b.node()
     node2 = ctx2.node()
 
     # Disable 3-way merge if there is only one parent
     if do3way:
         if node1b == nullid:
             do3way = False
 
     subrepos = opts.get(b'subrepos')
 
     matcher = scmutil.match(repo[node2], pats, opts)
 
     if opts.get(b'patch'):
         if subrepos:
             raise error.Abort(_(b'--patch cannot be used with --subrepos'))
         if perfile:
             raise error.Abort(_(b'--patch cannot be used with --per-file'))
         if node2 is None:
             raise error.Abort(_(b'--patch requires two revisions'))
     else:
         st = repo.status(node1a, node2, matcher, listsubrepos=subrepos)
         mod_a, add_a, rem_a = set(st.modified), set(st.added), set(st.removed)
         if do3way:
             stb = repo.status(node1b, node2, matcher, listsubrepos=subrepos)
             mod_b, add_b, rem_b = (
                 set(stb.modified),
                 set(stb.added),
                 set(stb.removed),
             )
         else:
             mod_b, add_b, rem_b = set(), set(), set()
         modadd = mod_a | add_a | mod_b | add_b
         common = modadd | rem_a | rem_b
         if not common:
             return 0
 
     tmproot = pycompat.mkdtemp(prefix=b'extdiff.')
     try:
         if not opts.get(b'patch'):
             # Always make a copy of node1a (and node1b, if applicable)
             dir1a_files = mod_a | rem_a | ((mod_b | add_b) - add_a)
             dir1a = snapshot(ui, repo, dir1a_files, node1a, tmproot, subrepos)[
                 0
             ]
             rev1a = b'@%d' % repo[node1a].rev()
             if do3way:
                 dir1b_files = mod_b | rem_b | ((mod_a | add_a) - add_b)
                 dir1b = snapshot(
                     ui, repo, dir1b_files, node1b, tmproot, subrepos
                 )[0]
                 rev1b = b'@%d' % repo[node1b].rev()
             else:
                 dir1b = None
                 rev1b = b''
 
             fnsandstat = []
 
             # If node2 in not the wc or there is >1 change, copy it
             dir2root = b''
             rev2 = b''
             if node2:
                 dir2 = snapshot(ui, repo, modadd, node2, tmproot, subrepos)[0]
                 rev2 = b'@%d' % repo[node2].rev()
             elif len(common) > 1:
                 # we only actually need to get the files to copy back to
                 # the working dir in this case (because the other cases
                 # are: diffing 2 revisions or single file -- in which case
                 # the file is already directly passed to the diff tool).
                 dir2, fnsandstat = snapshot(
                     ui, repo, modadd, None, tmproot, subrepos
                 )
             else:
                 # This lets the diff tool open the changed file directly
                 dir2 = b''
                 dir2root = repo.root
 
             label1a = rev1a
             label1b = rev1b
             label2 = rev2
 
             # If only one change, diff the files instead of the directories
             # Handle bogus modifies correctly by checking if the files exist
             if len(common) == 1:
                 common_file = util.localpath(common.pop())
                 dir1a = os.path.join(tmproot, dir1a, common_file)
                 label1a = common_file + rev1a
                 if not os.path.isfile(dir1a):
                     dir1a = pycompat.osdevnull
                 if do3way:
                     dir1b = os.path.join(tmproot, dir1b, common_file)
                     label1b = common_file + rev1b
                     if not os.path.isfile(dir1b):
                         dir1b = pycompat.osdevnull
                 dir2 = os.path.join(dir2root, dir2, common_file)
                 label2 = common_file + rev2
         else:
             template = b'hg-%h.patch'
             with formatter.nullformatter(ui, b'extdiff', {}) as fm:
                 cmdutil.export(
                     repo,
                     [repo[node1a].rev(), repo[node2].rev()],
                     fm,
                     fntemplate=repo.vfs.reljoin(tmproot, template),
                     match=matcher,
                 )
             label1a = cmdutil.makefilename(repo[node1a], template)
             label2 = cmdutil.makefilename(repo[node2], template)
             dir1a = repo.vfs.reljoin(tmproot, label1a)
             dir2 = repo.vfs.reljoin(tmproot, label2)
             dir1b = None
             label1b = None
             fnsandstat = []
 
         if not perfile:
             # Run the external tool on the 2 temp directories or the patches
             cmdline = formatcmdline(
                 cmdline,
                 repo.root,
                 do3way=do3way,
                 parent1=dir1a,
                 plabel1=label1a,
                 parent2=dir1b,
                 plabel2=label1b,
                 child=dir2,
                 clabel=label2,
             )
             ui.debug(
                 b'running %r in %s\n' % (pycompat.bytestr(cmdline), tmproot)
             )
             ui.system(cmdline, cwd=tmproot, blockedtag=b'extdiff')
         else:
             # Run the external tool once for each pair of files
             _runperfilediff(
                 cmdline,
                 repo.root,
                 ui,
                 guitool=guitool,
                 do3way=do3way,
                 confirm=confirm,
                 commonfiles=common,
                 tmproot=tmproot,
                 dir1a=dir1a,
                 dir1b=dir1b,
                 dir2root=dir2root,
                 dir2=dir2,
                 rev1a=rev1a,
                 rev1b=rev1b,
                 rev2=rev2,
             )
 
         for copy_fn, working_fn, st in fnsandstat:
             cpstat = os.lstat(copy_fn)
             # Some tools copy the file and attributes, so mtime may not detect
             # all changes.  A size check will detect more cases, but not all.
             # The only certain way to detect every case is to diff all files,
             # which could be expensive.
             # copyfile() carries over the permission, so the mode check could
             # be in an 'elif' branch, but for the case where the file has
             # changed without affecting mtime or size.
             if (
                 cpstat[stat.ST_MTIME] != st[stat.ST_MTIME]
                 or cpstat.st_size != st.st_size
                 or (cpstat.st_mode & 0o100) != (st.st_mode & 0o100)
             ):
                 ui.debug(
                     b'file changed while diffing. '
                     b'Overwriting: %s (src: %s)\n' % (working_fn, copy_fn)
                 )
                 util.copyfile(copy_fn, working_fn)
 
         return 1
     finally:
         ui.note(_(b'cleaning up temp directory\n'))
         shutil.rmtree(tmproot)
 
 
 extdiffopts = (
     [
         (
             b'o',
             b'option',
             [],
             _(b'pass option to comparison program'),
             _(b'OPT'),
         ),
         (b'r', b'rev', [], _(b'revision'), _(b'REV')),
         (b'c', b'change', b'', _(b'change made by revision'), _(b'REV')),
         (
             b'',
             b'per-file',
             False,
             _(b'compare each file instead of revision snapshots'),
         ),
         (
             b'',
             b'confirm',
             False,
             _(b'prompt user before each external program invocation'),
         ),
         (b'', b'patch', None, _(b'compare patches for two revisions')),
     ]
     + cmdutil.walkopts
     + cmdutil.subrepoopts
 )
 
 
 @command(
     b'extdiff',
     [(b'p', b'program', b'', _(b'comparison program to run'), _(b'CMD')),]
     + extdiffopts,
     _(b'hg extdiff [OPT]... [FILE]...'),
     helpcategory=command.CATEGORY_FILE_CONTENTS,
     inferrepo=True,
 )
 def extdiff(ui, repo, *pats, **opts):
     '''use external program to diff repository (or selected files)
 
     Show differences between revisions for the specified files, using
     an external program. The default program used is diff, with
     default options "-Npru".
 
     To select a different program, use the -p/--program option. The
     program will be passed the names of two directories to compare,
     unless the --per-file option is specified (see below). To pass
     additional options to the program, use -o/--option. These will be
     passed before the names of the directories or files to compare.
 
     When two revision arguments are given, then changes are shown
     between those revisions. If only one revision is specified then
     that revision is compared to the working directory, and, when no
     revisions are specified, the working directory files are compared
     to its parent.
 
     The --per-file option runs the external program repeatedly on each
     file to diff, instead of once on two directories. By default,
     this happens one by one, where the next file diff is open in the
     external program only once the previous external program (for the
     previous file diff) has exited. If the external program has a
     graphical interface, it can open all the file diffs at once instead
     of one by one. See :hg:`help -e extdiff` for information about how
     to tell Mercurial that a given program has a graphical interface.
 
     The --confirm option will prompt the user before each invocation of
     the external program. It is ignored if --per-file isn't specified.
     '''
     opts = pycompat.byteskwargs(opts)
     program = opts.get(b'program')
     option = opts.get(b'option')
     if not program:
         program = b'diff'
         option = option or [b'-Npru']
     cmdline = b' '.join(map(procutil.shellquote, [program] + option))
     return dodiff(ui, repo, cmdline, pats, opts)
 
 
 class savedcmd(object):
     """use external program to diff repository (or selected files)
 
     Show differences between revisions for the specified files, using
     the following program::
 
         %(path)s
 
     When two revision arguments are given, then changes are shown
     between those revisions. If only one revision is specified then
     that revision is compared to the working directory, and, when no
     revisions are specified, the working directory files are compared
     to its parent.
     """
 
     def __init__(self, path, cmdline, isgui):
         # We can't pass non-ASCII through docstrings (and path is
         # in an unknown encoding anyway), but avoid double separators on
         # Windows
         docpath = stringutil.escapestr(path).replace(b'\\\\', b'\\')
         self.__doc__ %= {'path': pycompat.sysstr(stringutil.uirepr(docpath))}
         self._cmdline = cmdline
         self._isgui = isgui
 
     def __call__(self, ui, repo, *pats, **opts):
         opts = pycompat.byteskwargs(opts)
         options = b' '.join(map(procutil.shellquote, opts[b'option']))
         if options:
             options = b' ' + options
         return dodiff(
             ui, repo, self._cmdline + options, pats, opts, guitool=self._isgui
         )
 
 
 def uisetup(ui):
     for cmd, path in ui.configitems(b'extdiff'):
         path = util.expandpath(path)
         if cmd.startswith(b'cmd.'):
             cmd = cmd[4:]
             if not path:
                 path = procutil.findexe(cmd)
                 if path is None:
                     path = filemerge.findexternaltool(ui, cmd) or cmd
             diffopts = ui.config(b'extdiff', b'opts.' + cmd)
             cmdline = procutil.shellquote(path)
             if diffopts:
                 cmdline += b' ' + diffopts
             isgui = ui.configbool(b'extdiff', b'gui.' + cmd)
         elif cmd.startswith(b'opts.') or cmd.startswith(b'gui.'):
             continue
         else:
             if path:
                 # case "cmd = path opts"
                 cmdline = path
                 diffopts = len(pycompat.shlexsplit(cmdline)) > 1
             else:
                 # case "cmd ="
                 path = procutil.findexe(cmd)
                 if path is None:
                     path = filemerge.findexternaltool(ui, cmd) or cmd
                 cmdline = procutil.shellquote(path)
                 diffopts = False
             isgui = ui.configbool(b'extdiff', b'gui.' + cmd)
         # look for diff arguments in [diff-tools] then [merge-tools]
         if not diffopts:
             key = cmd + b'.diffargs'
             for section in (b'diff-tools', b'merge-tools'):
                 args = ui.config(section, key)
                 if args:
                     cmdline += b' ' + args
                     if isgui is None:
                         isgui = ui.configbool(section, cmd + b'.gui') or False
                     break
         command(
             cmd,
             extdiffopts[:],
             _(b'hg %s [OPTION]... [FILE]...') % cmd,
             helpcategory=command.CATEGORY_FILE_CONTENTS,
             inferrepo=True,
         )(savedcmd(path, cmdline, isgui))
 
 
 # tell hggettext to extract docstrings from these functions:
 i18nfunctions = [savedcmd]

mercurial/chgserver.py

@@ -1,718 +1,718 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
 """
 
 from __future__ import absolute_import
 
 import inspect
 import os
 import re
 import socket
 import stat
 import struct
 import time
 
 from .i18n import _
 from .pycompat import (
     getattr,
     setattr,
 )
 
 from . import (
     commandserver,
     encoding,
     error,
     extensions,
     node,
     pycompat,
     util,
 )
 
 from .utils import (
     hashutil,
     procutil,
     stringutil,
 )
 
 
 def _hashlist(items):
     """return sha1 hexdigest for a list"""
     return node.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 pycompat.iteritems(encoding.environ)
         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(object):
     """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 util.safehasattr(self.fout, b'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 [b'fin', b'fout', b'ferr', b'environ']:
         setattr(newui, a, getattr(srcui, a))
     if util.safehasattr(srcui, b'_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(object):
     """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, procutil.quotecommand(cmd), os.path.abspath(cwd or b'.')]
+        args = [type, cmd, os.path.abspath(cwd or b'.')]
         args.extend(b'%s=%s' % (k, v) for k, v in pycompat.iteritems(environ))
         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
     (b'cin', b'fin', 'rb'),
     (b'cout', b'fout', 'wb'),
     (b'cerr', b'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))
         clientfds = util.recvfds(self.clientsock.fileno())
         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:
                 # make it line buffered explicitly because the default is
                 # decided on first write(), where fout could be a pager.
                 if fp.isatty():
                     bufsize = 1  # line buffered
                 else:
                     bufsize = -1  # system default
                 newfp = os.fdopen(fp.fileno(), mode, bufsize)
                 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):
         ui = self.ui
         for (ch, fp, fd), (cn, fn, _mode) in zip(self._oldios, _iochannels):
             newfp = getattr(ui, fn)
             # close newfp while it's associated with client; otherwise it
             # would be closed when newfp is deleted
             if newfp is not fp:
                 newfp.close()
             # restore original fd: fp is open again
             os.dup2(fd, fp.fileno())
             os.close(fd)
             setattr(self, cn, ch)
             setattr(ui, fn, fp)
         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.
         """
         from . import dispatch  # avoid cycle
 
         args = self._readlist()
         try:
             self.ui, lui = _loadnewui(self.ui, args, self.cdebug)
         except error.ParseError as inst:
             dispatch._formatparse(self.ui.warn, inst)
             self.ui.flush()
             self.cresult.write(b'exit 255')
             return
         except error.Abort as inst:
             self.ui.error(_(b"abort: %s\n") % inst)
             if inst.hint:
                 self.ui.error(_(b"(%s)\n") % inst.hint)
             self.ui.flush()
             self.cresult.write(b'exit 255')
             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 util.safehasattr(procutil, b'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(object):
     """Set of operations for chg services"""
 
     pollinterval = 1  # [sec]
 
     def __init__(self, ui):
         self.ui = ui
         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/posix.py

@@ -1,779 +1,775 b''
 # posix.py - Posix utility function implementations for Mercurial
 #
 #  Copyright 2005-2009 Matt Mackall <[e-mail]> and others
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import errno
 import fcntl
 import getpass
 import grp
 import os
 import pwd
 import re
 import select
 import stat
 import sys
 import tempfile
 import unicodedata
 
 from .i18n import _
 from .pycompat import (
     getattr,
     open,
 )
 from . import (
     encoding,
     error,
     policy,
     pycompat,
 )
 
 osutil = policy.importmod('osutil')
 
 normpath = os.path.normpath
 samestat = os.path.samestat
 try:
     oslink = os.link
 except AttributeError:
     # Some platforms build Python without os.link on systems that are
     # vaguely unix-like but don't have hardlink support. For those
     # poor souls, just say we tried and that it failed so we fall back
     # to copies.
     def oslink(src, dst):
         raise OSError(
             errno.EINVAL, b'hardlinks not supported: %s to %s' % (src, dst)
         )
 
 
 readlink = os.readlink
 unlink = os.unlink
 rename = os.rename
 removedirs = os.removedirs
 expandglobs = False
 
 umask = os.umask(0)
 os.umask(umask)
 
 if not pycompat.ispy3:
 
     def posixfile(name, mode='r', buffering=-1):
         fp = open(name, mode=mode, buffering=buffering)
         # The position when opening in append mode is implementation defined, so
         # make it consistent by always seeking to the end.
         if 'a' in mode:
             fp.seek(0, os.SEEK_END)
         return fp
 
 
 else:
     # The underlying file object seeks as required in Python 3:
     # https://github.com/python/cpython/blob/v3.7.3/Modules/_io/fileio.c#L474
     posixfile = open
 
 
 def split(p):
     '''Same as posixpath.split, but faster
 
     >>> import posixpath
     >>> for f in [b'/absolute/path/to/file',
     ...           b'relative/path/to/file',
     ...           b'file_alone',
     ...           b'path/to/directory/',
     ...           b'/multiple/path//separators',
     ...           b'/file_at_root',
     ...           b'///multiple_leading_separators_at_root',
     ...           b'']:
     ...     assert split(f) == posixpath.split(f), f
     '''
     ht = p.rsplit(b'/', 1)
     if len(ht) == 1:
         return b'', p
     nh = ht[0].rstrip(b'/')
     if nh:
         return nh, ht[1]
     return ht[0] + b'/', ht[1]
 
 
 def openhardlinks():
     '''return true if it is safe to hold open file handles to hardlinks'''
     return True
 
 
 def nlinks(name):
     '''return number of hardlinks for the given file'''
     return os.lstat(name).st_nlink
 
 
 def parsepatchoutput(output_line):
     """parses the output produced by patch and returns the filename"""
     pf = output_line[14:]
     if pycompat.sysplatform == b'OpenVMS':
         if pf[0] == b'`':
             pf = pf[1:-1]  # Remove the quotes
     else:
         if pf.startswith(b"'") and pf.endswith(b"'") and b" " in pf:
             pf = pf[1:-1]  # Remove the quotes
     return pf
 
 
 def sshargs(sshcmd, host, user, port):
     '''Build argument list for ssh'''
     args = user and (b"%s@%s" % (user, host)) or host
     if b'-' in args[:1]:
         raise error.Abort(
             _(b'illegal ssh hostname or username starting with -: %s') % args
         )
     args = shellquote(args)
     if port:
         args = b'-p %s %s' % (shellquote(port), args)
     return args
 
 
 def isexec(f):
     """check whether a file is executable"""
     return os.lstat(f).st_mode & 0o100 != 0
 
 
 def setflags(f, l, x):
     st = os.lstat(f)
     s = st.st_mode
     if l:
         if not stat.S_ISLNK(s):
             # switch file to link
             fp = open(f, b'rb')
             data = fp.read()
             fp.close()
             unlink(f)
             try:
                 os.symlink(data, f)
             except OSError:
                 # failed to make a link, rewrite file
                 fp = open(f, b"wb")
                 fp.write(data)
                 fp.close()
         # no chmod needed at this point
         return
     if stat.S_ISLNK(s):
         # switch link to file
         data = os.readlink(f)
         unlink(f)
         fp = open(f, b"wb")
         fp.write(data)
         fp.close()
         s = 0o666 & ~umask  # avoid restatting for chmod
 
     sx = s & 0o100
     if st.st_nlink > 1 and bool(x) != bool(sx):
         # the file is a hardlink, break it
         with open(f, b"rb") as fp:
             data = fp.read()
         unlink(f)
         with open(f, b"wb") as fp:
             fp.write(data)
 
     if x and not sx:
         # Turn on +x for every +r bit when making a file executable
         # and obey umask.
         os.chmod(f, s | (s & 0o444) >> 2 & ~umask)
     elif not x and sx:
         # Turn off all +x bits
         os.chmod(f, s & 0o666)
 
 
 def copymode(src, dst, mode=None, enforcewritable=False):
     '''Copy the file mode from the file at path src to dst.
     If src doesn't exist, we're using mode instead. If mode is None, we're
     using umask.'''
     try:
         st_mode = os.lstat(src).st_mode & 0o777
     except OSError as inst:
         if inst.errno != errno.ENOENT:
             raise
         st_mode = mode
         if st_mode is None:
             st_mode = ~umask
         st_mode &= 0o666
 
     new_mode = st_mode
 
     if enforcewritable:
         new_mode |= stat.S_IWUSR
 
     os.chmod(dst, new_mode)
 
 
 def checkexec(path):
     """
     Check whether the given path is on a filesystem with UNIX-like exec flags
 
     Requires a directory (like /foo/.hg)
     """
 
     # VFAT on some Linux versions can flip mode but it doesn't persist
     # a FS remount. Frequently we can detect it if files are created
     # with exec bit on.
 
     try:
         EXECFLAGS = stat.S_IXUSR | stat.S_IXGRP | stat.S_IXOTH
         basedir = os.path.join(path, b'.hg')
         cachedir = os.path.join(basedir, b'wcache')
         storedir = os.path.join(basedir, b'store')
         if not os.path.exists(cachedir):
             try:
                 # we want to create the 'cache' directory, not the '.hg' one.
                 # Automatically creating '.hg' directory could silently spawn
                 # invalid Mercurial repositories. That seems like a bad idea.
                 os.mkdir(cachedir)
                 if os.path.exists(storedir):
                     copymode(storedir, cachedir)
                 else:
                     copymode(basedir, cachedir)
             except (IOError, OSError):
                 # we other fallback logic triggers
                 pass
         if os.path.isdir(cachedir):
             checkisexec = os.path.join(cachedir, b'checkisexec')
             checknoexec = os.path.join(cachedir, b'checknoexec')
 
             try:
                 m = os.stat(checkisexec).st_mode
             except OSError as e:
                 if e.errno != errno.ENOENT:
                     raise
                 # checkisexec does not exist - fall through ...
             else:
                 # checkisexec exists, check if it actually is exec
                 if m & EXECFLAGS != 0:
                     # ensure checkisexec exists, check it isn't exec
                     try:
                         m = os.stat(checknoexec).st_mode
                     except OSError as e:
                         if e.errno != errno.ENOENT:
                             raise
                         open(checknoexec, b'w').close()  # might fail
                         m = os.stat(checknoexec).st_mode
                     if m & EXECFLAGS == 0:
                         # check-exec is exec and check-no-exec is not exec
                         return True
                     # checknoexec exists but is exec - delete it
                     unlink(checknoexec)
                 # checkisexec exists but is not exec - delete it
                 unlink(checkisexec)
 
             # check using one file, leave it as checkisexec
             checkdir = cachedir
         else:
             # check directly in path and don't leave checkisexec behind
             checkdir = path
             checkisexec = None
         fh, fn = pycompat.mkstemp(dir=checkdir, prefix=b'hg-checkexec-')
         try:
             os.close(fh)
             m = os.stat(fn).st_mode
             if m & EXECFLAGS == 0:
                 os.chmod(fn, m & 0o777 | EXECFLAGS)
                 if os.stat(fn).st_mode & EXECFLAGS != 0:
                     if checkisexec is not None:
                         os.rename(fn, checkisexec)
                         fn = None
                     return True
         finally:
             if fn is not None:
                 unlink(fn)
     except (IOError, OSError):
         # we don't care, the user probably won't be able to commit anyway
         return False
 
 
 def checklink(path):
     """check whether the given path is on a symlink-capable filesystem"""
     # mktemp is not racy because symlink creation will fail if the
     # file already exists
     while True:
         cachedir = os.path.join(path, b'.hg', b'wcache')
         checklink = os.path.join(cachedir, b'checklink')
         # try fast path, read only
         if os.path.islink(checklink):
             return True
         if os.path.isdir(cachedir):
             checkdir = cachedir
         else:
             checkdir = path
             cachedir = None
         name = tempfile.mktemp(
             dir=pycompat.fsdecode(checkdir), prefix=r'checklink-'
         )
         name = pycompat.fsencode(name)
         try:
             fd = None
             if cachedir is None:
                 fd = pycompat.namedtempfile(
                     dir=checkdir, prefix=b'hg-checklink-'
                 )
                 target = os.path.basename(fd.name)
             else:
                 # create a fixed file to link to; doesn't matter if it
                 # already exists.
                 target = b'checklink-target'
                 try:
                     fullpath = os.path.join(cachedir, target)
                     open(fullpath, b'w').close()
                 except IOError as inst:
                     if (
                         inst[0] == errno.EACCES
                     ):  # pytype: disable=unsupported-operands
                         # If we can't write to cachedir, just pretend
                         # that the fs is readonly and by association
                         # that the fs won't support symlinks. This
                         # seems like the least dangerous way to avoid
                         # data loss.
                         return False
                     raise
             try:
                 os.symlink(target, name)
                 if cachedir is None:
                     unlink(name)
                 else:
                     try:
                         os.rename(name, checklink)
                     except OSError:
                         unlink(name)
                 return True
             except OSError as inst:
                 # link creation might race, try again
                 if inst.errno == errno.EEXIST:
                     continue
                 raise
             finally:
                 if fd is not None:
                     fd.close()
         except AttributeError:
             return False
         except OSError as inst:
             # sshfs might report failure while successfully creating the link
             if inst.errno == errno.EIO and os.path.exists(name):
                 unlink(name)
             return False
 
 
 def checkosfilename(path):
     '''Check that the base-relative path is a valid filename on this platform.
     Returns None if the path is ok, or a UI string describing the problem.'''
     return None  # on posix platforms, every path is ok
 
 
 def getfsmountpoint(dirpath):
     '''Get the filesystem mount point from a directory (best-effort)
 
     Returns None if we are unsure. Raises OSError on ENOENT, EPERM, etc.
     '''
     return getattr(osutil, 'getfsmountpoint', lambda x: None)(dirpath)
 
 
 def getfstype(dirpath):
     '''Get the filesystem type name from a directory (best-effort)
 
     Returns None if we are unsure. Raises OSError on ENOENT, EPERM, etc.
     '''
     return getattr(osutil, 'getfstype', lambda x: None)(dirpath)
 
 
 def setbinary(fd):
     pass
 
 
 def pconvert(path):
     return path
 
 
 def localpath(path):
     return path
 
 
 def samefile(fpath1, fpath2):
     """Returns whether path1 and path2 refer to the same file. This is only
     guaranteed to work for files, not directories."""
     return os.path.samefile(fpath1, fpath2)
 
 
 def samedevice(fpath1, fpath2):
     """Returns whether fpath1 and fpath2 are on the same device. This is only
     guaranteed to work for files, not directories."""
     st1 = os.lstat(fpath1)
     st2 = os.lstat(fpath2)
     return st1.st_dev == st2.st_dev
 
 
 # os.path.normcase is a no-op, which doesn't help us on non-native filesystems
 def normcase(path):
     return path.lower()
 
 
 # what normcase does to ASCII strings
 normcasespec = encoding.normcasespecs.lower
 # fallback normcase function for non-ASCII strings
 normcasefallback = normcase
 
 if pycompat.isdarwin:
 
     def normcase(path):
         '''
         Normalize a filename for OS X-compatible comparison:
         - escape-encode invalid characters
         - decompose to NFD
         - lowercase
         - omit ignored characters [200c-200f, 202a-202e, 206a-206f,feff]
 
         >>> normcase(b'UPPER')
         'upper'
         >>> normcase(b'Caf\\xc3\\xa9')
         'cafe\\xcc\\x81'
         >>> normcase(b'\\xc3\\x89')
         'e\\xcc\\x81'
         >>> normcase(b'\\xb8\\xca\\xc3\\xca\\xbe\\xc8.JPG') # issue3918
         '%b8%ca%c3\\xca\\xbe%c8.jpg'
         '''
 
         try:
             return encoding.asciilower(path)  # exception for non-ASCII
         except UnicodeDecodeError:
             return normcasefallback(path)
 
     normcasespec = encoding.normcasespecs.lower
 
     def normcasefallback(path):
         try:
             u = path.decode('utf-8')
         except UnicodeDecodeError:
             # OS X percent-encodes any bytes that aren't valid utf-8
             s = b''
             pos = 0
             l = len(path)
             while pos < l:
                 try:
                     c = encoding.getutf8char(path, pos)
                     pos += len(c)
                 except ValueError:
                     c = b'%%%02X' % ord(path[pos : pos + 1])
                     pos += 1
                 s += c
 
             u = s.decode('utf-8')
 
         # Decompose then lowercase (HFS+ technote specifies lower)
         enc = unicodedata.normalize('NFD', u).lower().encode('utf-8')
         # drop HFS+ ignored characters
         return encoding.hfsignoreclean(enc)
 
 
 if pycompat.sysplatform == b'cygwin':
     # workaround for cygwin, in which mount point part of path is
     # treated as case sensitive, even though underlying NTFS is case
     # insensitive.
 
     # default mount points
     cygwinmountpoints = sorted(
         [b"/usr/bin", b"/usr/lib", b"/cygdrive",], reverse=True
     )
 
     # use upper-ing as normcase as same as NTFS workaround
     def normcase(path):
         pathlen = len(path)
         if (pathlen == 0) or (path[0] != pycompat.ossep):
             # treat as relative
             return encoding.upper(path)
 
         # to preserve case of mountpoint part
         for mp in cygwinmountpoints:
             if not path.startswith(mp):
                 continue
 
             mplen = len(mp)
             if mplen == pathlen:  # mount point itself
                 return mp
             if path[mplen] == pycompat.ossep:
                 return mp + encoding.upper(path[mplen:])
 
         return encoding.upper(path)
 
     normcasespec = encoding.normcasespecs.other
     normcasefallback = normcase
 
     # Cygwin translates native ACLs to POSIX permissions,
     # but these translations are not supported by native
     # tools, so the exec bit tends to be set erroneously.
     # Therefore, disable executable bit access on Cygwin.
     def checkexec(path):
         return False
 
     # Similarly, Cygwin's symlink emulation is likely to create
     # problems when Mercurial is used from both Cygwin and native
     # Windows, with other native tools, or on shared volumes
     def checklink(path):
         return False
 
 
 _needsshellquote = None
 
 
 def shellquote(s):
     if pycompat.sysplatform == b'OpenVMS':
         return b'"%s"' % s
     global _needsshellquote
     if _needsshellquote is None:
         _needsshellquote = re.compile(br'[^a-zA-Z0-9._/+-]').search
     if s and not _needsshellquote(s):
         # "s" shouldn't have to be quoted
         return s
     else:
         return b"'%s'" % s.replace(b"'", b"'\\''")
 
 
 def shellsplit(s):
     """Parse a command string in POSIX shell way (best-effort)"""
     return pycompat.shlexsplit(s, posix=True)
 
 
-def quotecommand(cmd):
-    return cmd
-
-
 def testpid(pid):
     '''return False if pid dead, True if running or not sure'''
     if pycompat.sysplatform == b'OpenVMS':
         return True
     try:
         os.kill(pid, 0)
         return True
     except OSError as inst:
         return inst.errno != errno.ESRCH
 
 
 def isowner(st):
     """Return True if the stat object st is from the current user."""
     return st.st_uid == os.getuid()
 
 
 def findexe(command):
     '''Find executable for command searching like which does.
     If command is a basename then PATH is searched for command.
     PATH isn't searched if command is an absolute or relative path.
     If command isn't found None is returned.'''
     if pycompat.sysplatform == b'OpenVMS':
         return command
 
     def findexisting(executable):
         b'Will return executable if existing file'
         if os.path.isfile(executable) and os.access(executable, os.X_OK):
             return executable
         return None
 
     if pycompat.ossep in command:
         return findexisting(command)
 
     if pycompat.sysplatform == b'plan9':
         return findexisting(os.path.join(b'/bin', command))
 
     for path in encoding.environ.get(b'PATH', b'').split(pycompat.ospathsep):
         executable = findexisting(os.path.join(path, command))
         if executable is not None:
             return executable
     return None
 
 
 def setsignalhandler():
     pass
 
 
 _wantedkinds = {stat.S_IFREG, stat.S_IFLNK}
 
 
 def statfiles(files):
     '''Stat each file in files. Yield each stat, or None if a file does not
     exist or has a type we don't care about.'''
     lstat = os.lstat
     getkind = stat.S_IFMT
     for nf in files:
         try:
             st = lstat(nf)
             if getkind(st.st_mode) not in _wantedkinds:
                 st = None
         except OSError as err:
             if err.errno not in (errno.ENOENT, errno.ENOTDIR):
                 raise
             st = None
         yield st
 
 
 def getuser():
     '''return name of current user'''
     return pycompat.fsencode(getpass.getuser())
 
 
 def username(uid=None):
     """Return the name of the user with the given uid.
 
     If uid is None, return the name of the current user."""
 
     if uid is None:
         uid = os.getuid()
     try:
         return pycompat.fsencode(pwd.getpwuid(uid)[0])
     except KeyError:
         return b'%d' % uid
 
 
 def groupname(gid=None):
     """Return the name of the group with the given gid.
 
     If gid is None, return the name of the current group."""
 
     if gid is None:
         gid = os.getgid()
     try:
         return pycompat.fsencode(grp.getgrgid(gid)[0])
     except KeyError:
         return pycompat.bytestr(gid)
 
 
 def groupmembers(name):
     """Return the list of members of the group with the given
     name, KeyError if the group does not exist.
     """
     name = pycompat.fsdecode(name)
     return pycompat.rapply(pycompat.fsencode, list(grp.getgrnam(name).gr_mem))
 
 
 def spawndetached(args):
     return os.spawnvp(os.P_NOWAIT | getattr(os, 'P_DETACH', 0), args[0], args)
 
 
 def gethgcmd():
     return sys.argv[:1]
 
 
 def makedir(path, notindexed):
     os.mkdir(path)
 
 
 def lookupreg(key, name=None, scope=None):
     return None
 
 
 def hidewindow():
     """Hide current shell window.
 
     Used to hide the window opened when starting asynchronous
     child process under Windows, unneeded on other systems.
     """
     pass
 
 
 class cachestat(object):
     def __init__(self, path):
         self.stat = os.stat(path)
 
     def cacheable(self):
         return bool(self.stat.st_ino)
 
     __hash__ = object.__hash__
 
     def __eq__(self, other):
         try:
             # Only dev, ino, size, mtime and atime are likely to change. Out
             # of these, we shouldn't compare atime but should compare the
             # rest. However, one of the other fields changing indicates
             # something fishy going on, so return False if anything but atime
             # changes.
             return (
                 self.stat.st_mode == other.stat.st_mode
                 and self.stat.st_ino == other.stat.st_ino
                 and self.stat.st_dev == other.stat.st_dev
                 and self.stat.st_nlink == other.stat.st_nlink
                 and self.stat.st_uid == other.stat.st_uid
                 and self.stat.st_gid == other.stat.st_gid
                 and self.stat.st_size == other.stat.st_size
                 and self.stat[stat.ST_MTIME] == other.stat[stat.ST_MTIME]
                 and self.stat[stat.ST_CTIME] == other.stat[stat.ST_CTIME]
             )
         except AttributeError:
             return False
 
     def __ne__(self, other):
         return not self == other
 
 
 def statislink(st):
     '''check whether a stat result is a symlink'''
     return st and stat.S_ISLNK(st.st_mode)
 
 
 def statisexec(st):
     '''check whether a stat result is an executable file'''
     return st and (st.st_mode & 0o100 != 0)
 
 
 def poll(fds):
     """block until something happens on any file descriptor
 
     This is a generic helper that will check for any activity
     (read, write.  exception) and return the list of touched files.
 
     In unsupported cases, it will raise a NotImplementedError"""
     try:
         while True:
             try:
                 res = select.select(fds, fds, fds)
                 break
             except select.error as inst:
                 if inst.args[0] == errno.EINTR:
                     continue
                 raise
     except ValueError:  # out of range file descriptor
         raise NotImplementedError()
     return sorted(list(set(sum(res, []))))
 
 
 def readpipe(pipe):
     """Read all available data from a pipe."""
     # We can't fstat() a pipe because Linux will always report 0.
     # So, we set the pipe to non-blocking mode and read everything
     # that's available.
     flags = fcntl.fcntl(pipe, fcntl.F_GETFL)
     flags |= os.O_NONBLOCK
     oldflags = fcntl.fcntl(pipe, fcntl.F_SETFL, flags)
 
     try:
         chunks = []
         while True:
             try:
                 s = pipe.read()
                 if not s:
                     break
                 chunks.append(s)
             except IOError:
                 break
 
         return b''.join(chunks)
     finally:
         fcntl.fcntl(pipe, fcntl.F_SETFL, oldflags)
 
 
 def bindunixsocket(sock, path):
     """Bind the UNIX domain socket to the specified path"""
     # use relative path instead of full path at bind() if possible, since
     # AF_UNIX path has very small length limit (107 chars) on common
     # platforms (see sys/un.h)
     dirname, basename = os.path.split(path)
     bakwdfd = None
     if dirname:
         bakwdfd = os.open(b'.', os.O_DIRECTORY)
         os.chdir(dirname)
     sock.bind(basename)
     if bakwdfd:
         os.fchdir(bakwdfd)
         os.close(bakwdfd)

mercurial/sshpeer.py

@@ -1,709 +1,708 b''
 # sshpeer.py - ssh repository proxy class for mercurial
 #
 # Copyright 2005, 2006 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import re
 import uuid
 
 from .i18n import _
 from .pycompat import getattr
 from . import (
     error,
     pycompat,
     util,
     wireprotoserver,
     wireprototypes,
     wireprotov1peer,
     wireprotov1server,
 )
 from .utils import (
     procutil,
     stringutil,
 )
 
 
 def _serverquote(s):
     """quote a string for the remote shell ... which we assume is sh"""
     if not s:
         return s
     if re.match(b'[a-zA-Z0-9@%_+=:,./-]*$', s):
         return s
     return b"'%s'" % s.replace(b"'", b"'\\''")
 
 
 def _forwardoutput(ui, pipe, warn=False):
     """display all data currently available on pipe as remote output.
 
     This is non blocking."""
     if pipe:
         s = procutil.readpipe(pipe)
         if s:
             display = ui.warn if warn else ui.status
             for l in s.splitlines():
                 display(_(b"remote: "), l, b'\n')
 
 
 class doublepipe(object):
     """Operate a side-channel pipe in addition of a main one
 
     The side-channel pipe contains server output to be forwarded to the user
     input. The double pipe will behave as the "main" pipe, but will ensure the
     content of the "side" pipe is properly processed while we wait for blocking
     call on the "main" pipe.
 
     If large amounts of data are read from "main", the forward will cease after
     the first bytes start to appear. This simplifies the implementation
     without affecting actual output of sshpeer too much as we rarely issue
     large read for data not yet emitted by the server.
 
     The main pipe is expected to be a 'bufferedinputpipe' from the util module
     that handle all the os specific bits. This class lives in this module
     because it focus on behavior specific to the ssh protocol."""
 
     def __init__(self, ui, main, side):
         self._ui = ui
         self._main = main
         self._side = side
 
     def _wait(self):
         """wait until some data are available on main or side
 
         return a pair of boolean (ismainready, issideready)
 
         (This will only wait for data if the setup is supported by `util.poll`)
         """
         if (
             isinstance(self._main, util.bufferedinputpipe)
             and self._main.hasbuffer
         ):
             # Main has data. Assume side is worth poking at.
             return True, True
 
         fds = [self._main.fileno(), self._side.fileno()]
         try:
             act = util.poll(fds)
         except NotImplementedError:
             # non supported yet case, assume all have data.
             act = fds
         return (self._main.fileno() in act, self._side.fileno() in act)
 
     def write(self, data):
         return self._call(b'write', data)
 
     def read(self, size):
         r = self._call(b'read', size)
         if size != 0 and not r:
             # We've observed a condition that indicates the
             # stdout closed unexpectedly. Check stderr one
             # more time and snag anything that's there before
             # letting anyone know the main part of the pipe
             # closed prematurely.
             _forwardoutput(self._ui, self._side)
         return r
 
     def unbufferedread(self, size):
         r = self._call(b'unbufferedread', size)
         if size != 0 and not r:
             # We've observed a condition that indicates the
             # stdout closed unexpectedly. Check stderr one
             # more time and snag anything that's there before
             # letting anyone know the main part of the pipe
             # closed prematurely.
             _forwardoutput(self._ui, self._side)
         return r
 
     def readline(self):
         return self._call(b'readline')
 
     def _call(self, methname, data=None):
         """call <methname> on "main", forward output of "side" while blocking
         """
         # data can be '' or 0
         if (data is not None and not data) or self._main.closed:
             _forwardoutput(self._ui, self._side)
             return b''
         while True:
             mainready, sideready = self._wait()
             if sideready:
                 _forwardoutput(self._ui, self._side)
             if mainready:
                 meth = getattr(self._main, methname)
                 if data is None:
                     return meth()
                 else:
                     return meth(data)
 
     def close(self):
         return self._main.close()
 
     def flush(self):
         return self._main.flush()
 
 
 def _cleanuppipes(ui, pipei, pipeo, pipee):
     """Clean up pipes used by an SSH connection."""
     if pipeo:
         pipeo.close()
     if pipei:
         pipei.close()
 
     if pipee:
         # Try to read from the err descriptor until EOF.
         try:
             for l in pipee:
                 ui.status(_(b'remote: '), l)
         except (IOError, ValueError):
             pass
 
         pipee.close()
 
 
 def _makeconnection(ui, sshcmd, args, remotecmd, path, sshenv=None):
     """Create an SSH connection to a server.
 
     Returns a tuple of (process, stdin, stdout, stderr) for the
     spawned process.
     """
     cmd = b'%s %s %s' % (
         sshcmd,
         args,
         procutil.shellquote(
             b'%s -R %s serve --stdio'
             % (_serverquote(remotecmd), _serverquote(path))
         ),
     )
 
     ui.debug(b'running %s\n' % cmd)
-    cmd = procutil.quotecommand(cmd)
 
     # no buffer allow the use of 'select'
     # feel free to remove buffering and select usage when we ultimately
     # move to threading.
     stdin, stdout, stderr, proc = procutil.popen4(cmd, bufsize=0, env=sshenv)
 
     return proc, stdin, stdout, stderr
 
 
 def _clientcapabilities():
     """Return list of capabilities of this client.
 
     Returns a list of capabilities that are supported by this client.
     """
     protoparams = {b'partial-pull'}
     comps = [
         e.wireprotosupport().name
         for e in util.compengines.supportedwireengines(util.CLIENTROLE)
     ]
     protoparams.add(b'comp=%s' % b','.join(comps))
     return protoparams
 
 
 def _performhandshake(ui, stdin, stdout, stderr):
     def badresponse():
         # Flush any output on stderr. In general, the stderr contains errors
         # from the remote (ssh errors, some hg errors), and status indications
         # (like "adding changes"), with no current way to tell them apart.
         # Here we failed so early that it's almost certainly only errors, so
         # use warn=True so -q doesn't hide them.
         _forwardoutput(ui, stderr, warn=True)
 
         msg = _(b'no suitable response from remote hg')
         hint = ui.config(b'ui', b'ssherrorhint')
         raise error.RepoError(msg, hint=hint)
 
     # The handshake consists of sending wire protocol commands in reverse
     # order of protocol implementation and then sniffing for a response
     # to one of them.
     #
     # Those commands (from oldest to newest) are:
     #
     # ``between``
     #   Asks for the set of revisions between a pair of revisions. Command
     #   present in all Mercurial server implementations.
     #
     # ``hello``
     #   Instructs the server to advertise its capabilities. Introduced in
     #   Mercurial 0.9.1.
     #
     # ``upgrade``
     #   Requests upgrade from default transport protocol version 1 to
     #   a newer version. Introduced in Mercurial 4.6 as an experimental
     #   feature.
     #
     # The ``between`` command is issued with a request for the null
     # range. If the remote is a Mercurial server, this request will
     # generate a specific response: ``1\n\n``. This represents the
     # wire protocol encoded value for ``\n``. We look for ``1\n\n``
     # in the output stream and know this is the response to ``between``
     # and we're at the end of our handshake reply.
     #
     # The response to the ``hello`` command will be a line with the
     # length of the value returned by that command followed by that
     # value. If the server doesn't support ``hello`` (which should be
     # rare), that line will be ``0\n``. Otherwise, the value will contain
     # RFC 822 like lines. Of these, the ``capabilities:`` line contains
     # the capabilities of the server.
     #
     # The ``upgrade`` command isn't really a command in the traditional
     # sense of version 1 of the transport because it isn't using the
     # proper mechanism for formatting insteads: instead, it just encodes
     # arguments on the line, delimited by spaces.
     #
     # The ``upgrade`` line looks like ``upgrade <token> <capabilities>``.
     # If the server doesn't support protocol upgrades, it will reply to
     # this line with ``0\n``. Otherwise, it emits an
     # ``upgraded <token> <protocol>`` line to both stdout and stderr.
     # Content immediately following this line describes additional
     # protocol and server state.
     #
     # In addition to the responses to our command requests, the server
     # may emit "banner" output on stdout. SSH servers are allowed to
     # print messages to stdout on login. Issuing commands on connection
     # allows us to flush this banner output from the server by scanning
     # for output to our well-known ``between`` command. Of course, if
     # the banner contains ``1\n\n``, this will throw off our detection.
 
     requestlog = ui.configbool(b'devel', b'debug.peer-request')
 
     # Generate a random token to help identify responses to version 2
     # upgrade request.
     token = pycompat.sysbytes(str(uuid.uuid4()))
     upgradecaps = [
         (b'proto', wireprotoserver.SSHV2),
     ]
     upgradecaps = util.urlreq.urlencode(upgradecaps)
 
     try:
         pairsarg = b'%s-%s' % (b'0' * 40, b'0' * 40)
         handshake = [
             b'hello\n',
             b'between\n',
             b'pairs %d\n' % len(pairsarg),
             pairsarg,
         ]
 
         # Request upgrade to version 2 if configured.
         if ui.configbool(b'experimental', b'sshpeer.advertise-v2'):
             ui.debug(b'sending upgrade request: %s %s\n' % (token, upgradecaps))
             handshake.insert(0, b'upgrade %s %s\n' % (token, upgradecaps))
 
         if requestlog:
             ui.debug(b'devel-peer-request: hello+between\n')
             ui.debug(b'devel-peer-request:   pairs: %d bytes\n' % len(pairsarg))
         ui.debug(b'sending hello command\n')
         ui.debug(b'sending between command\n')
 
         stdin.write(b''.join(handshake))
         stdin.flush()
     except IOError:
         badresponse()
 
     # Assume version 1 of wire protocol by default.
     protoname = wireprototypes.SSHV1
     reupgraded = re.compile(b'^upgraded %s (.*)$' % stringutil.reescape(token))
 
     lines = [b'', b'dummy']
     max_noise = 500
     while lines[-1] and max_noise:
         try:
             l = stdout.readline()
             _forwardoutput(ui, stderr, warn=True)
 
             # Look for reply to protocol upgrade request. It has a token
             # in it, so there should be no false positives.
             m = reupgraded.match(l)
             if m:
                 protoname = m.group(1)
                 ui.debug(b'protocol upgraded to %s\n' % protoname)
                 # If an upgrade was handled, the ``hello`` and ``between``
                 # requests are ignored. The next output belongs to the
                 # protocol, so stop scanning lines.
                 break
 
             # Otherwise it could be a banner, ``0\n`` response if server
             # doesn't support upgrade.
 
             if lines[-1] == b'1\n' and l == b'\n':
                 break
             if l:
                 ui.debug(b'remote: ', l)
             lines.append(l)
             max_noise -= 1
         except IOError:
             badresponse()
     else:
         badresponse()
 
     caps = set()
 
     # For version 1, we should see a ``capabilities`` line in response to the
     # ``hello`` command.
     if protoname == wireprototypes.SSHV1:
         for l in reversed(lines):
             # Look for response to ``hello`` command. Scan from the back so
             # we don't misinterpret banner output as the command reply.
             if l.startswith(b'capabilities:'):
                 caps.update(l[:-1].split(b':')[1].split())
                 break
     elif protoname == wireprotoserver.SSHV2:
         # We see a line with number of bytes to follow and then a value
         # looking like ``capabilities: *``.
         line = stdout.readline()
         try:
             valuelen = int(line)
         except ValueError:
             badresponse()
 
         capsline = stdout.read(valuelen)
         if not capsline.startswith(b'capabilities: '):
             badresponse()
 
         ui.debug(b'remote: %s\n' % capsline)
 
         caps.update(capsline.split(b':')[1].split())
         # Trailing newline.
         stdout.read(1)
 
     # Error if we couldn't find capabilities, this means:
     #
     # 1. Remote isn't a Mercurial server
     # 2. Remote is a <0.9.1 Mercurial server
     # 3. Remote is a future Mercurial server that dropped ``hello``
     #    and other attempted handshake mechanisms.
     if not caps:
         badresponse()
 
     # Flush any output on stderr before proceeding.
     _forwardoutput(ui, stderr, warn=True)
 
     return protoname, caps
 
 
 class sshv1peer(wireprotov1peer.wirepeer):
     def __init__(
         self, ui, url, proc, stdin, stdout, stderr, caps, autoreadstderr=True
     ):
         """Create a peer from an existing SSH connection.
 
         ``proc`` is a handle on the underlying SSH process.
         ``stdin``, ``stdout``, and ``stderr`` are handles on the stdio
         pipes for that process.
         ``caps`` is a set of capabilities supported by the remote.
         ``autoreadstderr`` denotes whether to automatically read from
         stderr and to forward its output.
         """
         self._url = url
         self.ui = ui
         # self._subprocess is unused. Keeping a handle on the process
         # holds a reference and prevents it from being garbage collected.
         self._subprocess = proc
 
         # And we hook up our "doublepipe" wrapper to allow querying
         # stderr any time we perform I/O.
         if autoreadstderr:
             stdout = doublepipe(ui, util.bufferedinputpipe(stdout), stderr)
             stdin = doublepipe(ui, stdin, stderr)
 
         self._pipeo = stdin
         self._pipei = stdout
         self._pipee = stderr
         self._caps = caps
         self._autoreadstderr = autoreadstderr
 
     # Commands that have a "framed" response where the first line of the
     # response contains the length of that response.
     _FRAMED_COMMANDS = {
         b'batch',
     }
 
     # Begin of ipeerconnection interface.
 
     def url(self):
         return self._url
 
     def local(self):
         return None
 
     def peer(self):
         return self
 
     def canpush(self):
         return True
 
     def close(self):
         pass
 
     # End of ipeerconnection interface.
 
     # Begin of ipeercommands interface.
 
     def capabilities(self):
         return self._caps
 
     # End of ipeercommands interface.
 
     def _readerr(self):
         _forwardoutput(self.ui, self._pipee)
 
     def _abort(self, exception):
         self._cleanup()
         raise exception
 
     def _cleanup(self):
         _cleanuppipes(self.ui, self._pipei, self._pipeo, self._pipee)
 
     __del__ = _cleanup
 
     def _sendrequest(self, cmd, args, framed=False):
         if self.ui.debugflag and self.ui.configbool(
             b'devel', b'debug.peer-request'
         ):
             dbg = self.ui.debug
             line = b'devel-peer-request: %s\n'
             dbg(line % cmd)
             for key, value in sorted(args.items()):
                 if not isinstance(value, dict):
                     dbg(line % b'  %s: %d bytes' % (key, len(value)))
                 else:
                     for dk, dv in sorted(value.items()):
                         dbg(line % b'  %s-%s: %d' % (key, dk, len(dv)))
         self.ui.debug(b"sending %s command\n" % cmd)
         self._pipeo.write(b"%s\n" % cmd)
         _func, names = wireprotov1server.commands[cmd]
         keys = names.split()
         wireargs = {}
         for k in keys:
             if k == b'*':
                 wireargs[b'*'] = args
                 break
             else:
                 wireargs[k] = args[k]
                 del args[k]
         for k, v in sorted(pycompat.iteritems(wireargs)):
             self._pipeo.write(b"%s %d\n" % (k, len(v)))
             if isinstance(v, dict):
                 for dk, dv in pycompat.iteritems(v):
                     self._pipeo.write(b"%s %d\n" % (dk, len(dv)))
                     self._pipeo.write(dv)
             else:
                 self._pipeo.write(v)
         self._pipeo.flush()
 
         # We know exactly how many bytes are in the response. So return a proxy
         # around the raw output stream that allows reading exactly this many
         # bytes. Callers then can read() without fear of overrunning the
         # response.
         if framed:
             amount = self._getamount()
             return util.cappedreader(self._pipei, amount)
 
         return self._pipei
 
     def _callstream(self, cmd, **args):
         args = pycompat.byteskwargs(args)
         return self._sendrequest(cmd, args, framed=cmd in self._FRAMED_COMMANDS)
 
     def _callcompressable(self, cmd, **args):
         args = pycompat.byteskwargs(args)
         return self._sendrequest(cmd, args, framed=cmd in self._FRAMED_COMMANDS)
 
     def _call(self, cmd, **args):
         args = pycompat.byteskwargs(args)
         return self._sendrequest(cmd, args, framed=True).read()
 
     def _callpush(self, cmd, fp, **args):
         # The server responds with an empty frame if the client should
         # continue submitting the payload.
         r = self._call(cmd, **args)
         if r:
             return b'', r
 
         # The payload consists of frames with content followed by an empty
         # frame.
         for d in iter(lambda: fp.read(4096), b''):
             self._writeframed(d)
         self._writeframed(b"", flush=True)
 
         # In case of success, there is an empty frame and a frame containing
         # the integer result (as a string).
         # In case of error, there is a non-empty frame containing the error.
         r = self._readframed()
         if r:
             return b'', r
         return self._readframed(), b''
 
     def _calltwowaystream(self, cmd, fp, **args):
         # The server responds with an empty frame if the client should
         # continue submitting the payload.
         r = self._call(cmd, **args)
         if r:
             # XXX needs to be made better
             raise error.Abort(_(b'unexpected remote reply: %s') % r)
 
         # The payload consists of frames with content followed by an empty
         # frame.
         for d in iter(lambda: fp.read(4096), b''):
             self._writeframed(d)
         self._writeframed(b"", flush=True)
 
         return self._pipei
 
     def _getamount(self):
         l = self._pipei.readline()
         if l == b'\n':
             if self._autoreadstderr:
                 self._readerr()
             msg = _(b'check previous remote output')
             self._abort(error.OutOfBandError(hint=msg))
         if self._autoreadstderr:
             self._readerr()
         try:
             return int(l)
         except ValueError:
             self._abort(error.ResponseError(_(b"unexpected response:"), l))
 
     def _readframed(self):
         size = self._getamount()
         if not size:
             return b''
 
         return self._pipei.read(size)
 
     def _writeframed(self, data, flush=False):
         self._pipeo.write(b"%d\n" % len(data))
         if data:
             self._pipeo.write(data)
         if flush:
             self._pipeo.flush()
         if self._autoreadstderr:
             self._readerr()
 
 
 class sshv2peer(sshv1peer):
     """A peer that speakers version 2 of the transport protocol."""
 
     # Currently version 2 is identical to version 1 post handshake.
     # And handshake is performed before the peer is instantiated. So
     # we need no custom code.
 
 
 def makepeer(ui, path, proc, stdin, stdout, stderr, autoreadstderr=True):
     """Make a peer instance from existing pipes.
 
     ``path`` and ``proc`` are stored on the eventual peer instance and may
     not be used for anything meaningful.
 
     ``stdin``, ``stdout``, and ``stderr`` are the pipes connected to the
     SSH server's stdio handles.
 
     This function is factored out to allow creating peers that don't
     actually spawn a new process. It is useful for starting SSH protocol
     servers and clients via non-standard means, which can be useful for
     testing.
     """
     try:
         protoname, caps = _performhandshake(ui, stdin, stdout, stderr)
     except Exception:
         _cleanuppipes(ui, stdout, stdin, stderr)
         raise
 
     if protoname == wireprototypes.SSHV1:
         return sshv1peer(
             ui,
             path,
             proc,
             stdin,
             stdout,
             stderr,
             caps,
             autoreadstderr=autoreadstderr,
         )
     elif protoname == wireprototypes.SSHV2:
         return sshv2peer(
             ui,
             path,
             proc,
             stdin,
             stdout,
             stderr,
             caps,
             autoreadstderr=autoreadstderr,
         )
     else:
         _cleanuppipes(ui, stdout, stdin, stderr)
         raise error.RepoError(
             _(b'unknown version of SSH protocol: %s') % protoname
         )
 
 
 def instance(ui, path, create, intents=None, createopts=None):
     """Create an SSH peer.
 
     The returned object conforms to the ``wireprotov1peer.wirepeer`` interface.
     """
     u = util.url(path, parsequery=False, parsefragment=False)
     if u.scheme != b'ssh' or not u.host or u.path is None:
         raise error.RepoError(_(b"couldn't parse location %s") % path)
 
     util.checksafessh(path)
 
     if u.passwd is not None:
         raise error.RepoError(_(b'password in URL not supported'))
 
     sshcmd = ui.config(b'ui', b'ssh')
     remotecmd = ui.config(b'ui', b'remotecmd')
     sshaddenv = dict(ui.configitems(b'sshenv'))
     sshenv = procutil.shellenviron(sshaddenv)
     remotepath = u.path or b'.'
 
     args = procutil.sshargs(sshcmd, u.host, u.user, u.port)
 
     if create:
         # We /could/ do this, but only if the remote init command knows how to
         # handle them. We don't yet make any assumptions about that. And without
         # querying the remote, there's no way of knowing if the remote even
         # supports said requested feature.
         if createopts:
             raise error.RepoError(
                 _(
                     b'cannot create remote SSH repositories '
                     b'with extra options'
                 )
             )
 
         cmd = b'%s %s %s' % (
             sshcmd,
             args,
             procutil.shellquote(
                 b'%s init %s'
                 % (_serverquote(remotecmd), _serverquote(remotepath))
             ),
         )
         ui.debug(b'running %s\n' % cmd)
         res = ui.system(cmd, blockedtag=b'sshpeer', environ=sshenv)
         if res != 0:
             raise error.RepoError(_(b'could not create remote repo'))
 
     proc, stdin, stdout, stderr = _makeconnection(
         ui, sshcmd, args, remotecmd, remotepath, sshenv
     )
 
     peer = makepeer(ui, path, proc, stdin, stdout, stderr)
 
     # Finally, if supported by the server, notify it about our own
     # capabilities.
     if b'protocaps' in peer.capabilities():
         try:
             peer._call(
                 b"protocaps", caps=b' '.join(sorted(_clientcapabilities()))
             )
         except IOError:
             peer._cleanup()
             raise error.RepoError(_(b'capability exchange failed'))
 
     return peer

mercurial/utils/procutil.py

@@ -1,679 +1,677 b''
 # procutil.py - utility for managing processes and executable environment
 #
 #  Copyright 2005 K. Thananchayan <[e-mail]>
 #  Copyright 2005-2007 Matt Mackall <[e-mail]>
 #  Copyright 2006 Vadim Gelfer <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import contextlib
 import errno
 import io
 import os
 import signal
 import subprocess
 import sys
 import threading
 import time
 
 from ..i18n import _
 from ..pycompat import (
     getattr,
     open,
 )
 
 from .. import (
     encoding,
     error,
     policy,
     pycompat,
 )
 
 # Import like this to keep import-checker happy
 from ..utils import resourceutil
 
 osutil = policy.importmod('osutil')
 
 stderr = pycompat.stderr
 stdin = pycompat.stdin
 stdout = pycompat.stdout
 
 
 def isatty(fp):
     try:
         return fp.isatty()
     except AttributeError:
         return False
 
 
 # glibc determines buffering on first write to stdout - if we replace a TTY
 # destined stdout with a pipe destined stdout (e.g. pager), we want line
 # buffering (or unbuffered, on Windows)
 if isatty(stdout):
     if pycompat.iswindows:
         # Windows doesn't support line buffering
         stdout = os.fdopen(stdout.fileno(), 'wb', 0)
     elif not pycompat.ispy3:
         # on Python 3, stdout (sys.stdout.buffer) is already line buffered and
         # buffering=1 is not handled in binary mode
         stdout = os.fdopen(stdout.fileno(), 'wb', 1)
 
 if pycompat.iswindows:
     from .. import windows as platform
 
     stdout = platform.winstdout(stdout)
 else:
     from .. import posix as platform
 
 findexe = platform.findexe
 _gethgcmd = platform.gethgcmd
 getuser = platform.getuser
 getpid = os.getpid
 hidewindow = platform.hidewindow
-quotecommand = platform.quotecommand
 readpipe = platform.readpipe
 setbinary = platform.setbinary
 setsignalhandler = platform.setsignalhandler
 shellquote = platform.shellquote
 shellsplit = platform.shellsplit
 spawndetached = platform.spawndetached
 sshargs = platform.sshargs
 testpid = platform.testpid
 
 try:
     setprocname = osutil.setprocname
 except AttributeError:
     pass
 try:
     unblocksignal = osutil.unblocksignal
 except AttributeError:
     pass
 
 closefds = pycompat.isposix
 
 
 def explainexit(code):
     """return a message describing a subprocess status
     (codes from kill are negative - not os.system/wait encoding)"""
     if code >= 0:
         return _(b"exited with status %d") % code
     return _(b"killed by signal %d") % -code
 
 
 class _pfile(object):
     """File-like wrapper for a stream opened by subprocess.Popen()"""
 
     def __init__(self, proc, fp):
         self._proc = proc
         self._fp = fp
 
     def close(self):
         # unlike os.popen(), this returns an integer in subprocess coding
         self._fp.close()
         return self._proc.wait()
 
     def __iter__(self):
         return iter(self._fp)
 
     def __getattr__(self, attr):
         return getattr(self._fp, attr)
 
     def __enter__(self):
         return self
 
     def __exit__(self, exc_type, exc_value, exc_tb):
         self.close()
 
 
 def popen(cmd, mode=b'rb', bufsize=-1):
     if mode == b'rb':
         return _popenreader(cmd, bufsize)
     elif mode == b'wb':
         return _popenwriter(cmd, bufsize)
     raise error.ProgrammingError(b'unsupported mode: %r' % mode)
 
 
 def _popenreader(cmd, bufsize):
     p = subprocess.Popen(
-        tonativestr(quotecommand(cmd)),
+        tonativestr(cmd),
         shell=True,
         bufsize=bufsize,
         close_fds=closefds,
         stdout=subprocess.PIPE,
     )
     return _pfile(p, p.stdout)
 
 
 def _popenwriter(cmd, bufsize):
     p = subprocess.Popen(
-        tonativestr(quotecommand(cmd)),
+        tonativestr(cmd),
         shell=True,
         bufsize=bufsize,
         close_fds=closefds,
         stdin=subprocess.PIPE,
     )
     return _pfile(p, p.stdin)
 
 
 def popen2(cmd, env=None):
     # Setting bufsize to -1 lets the system decide the buffer size.
     # The default for bufsize is 0, meaning unbuffered. This leads to
     # poor performance on Mac OS X: http://bugs.python.org/issue4194
     p = subprocess.Popen(
         tonativestr(cmd),
         shell=True,
         bufsize=-1,
         close_fds=closefds,
         stdin=subprocess.PIPE,
         stdout=subprocess.PIPE,
         env=tonativeenv(env),
     )
     return p.stdin, p.stdout
 
 
 def popen3(cmd, env=None):
     stdin, stdout, stderr, p = popen4(cmd, env)
     return stdin, stdout, stderr
 
 
 def popen4(cmd, env=None, bufsize=-1):
     p = subprocess.Popen(
         tonativestr(cmd),
         shell=True,
         bufsize=bufsize,
         close_fds=closefds,
         stdin=subprocess.PIPE,
         stdout=subprocess.PIPE,
         stderr=subprocess.PIPE,
         env=tonativeenv(env),
     )
     return p.stdin, p.stdout, p.stderr, p
 
 
 def pipefilter(s, cmd):
     '''filter string S through command CMD, returning its output'''
     p = subprocess.Popen(
         tonativestr(cmd),
         shell=True,
         close_fds=closefds,
         stdin=subprocess.PIPE,
         stdout=subprocess.PIPE,
     )
     pout, perr = p.communicate(s)
     return pout
 
 
 def tempfilter(s, cmd):
     '''filter string S through a pair of temporary files with CMD.
     CMD is used as a template to create the real command to be run,
     with the strings INFILE and OUTFILE replaced by the real names of
     the temporary files generated.'''
     inname, outname = None, None
     try:
         infd, inname = pycompat.mkstemp(prefix=b'hg-filter-in-')
         fp = os.fdopen(infd, 'wb')
         fp.write(s)
         fp.close()
         outfd, outname = pycompat.mkstemp(prefix=b'hg-filter-out-')
         os.close(outfd)
         cmd = cmd.replace(b'INFILE', inname)
         cmd = cmd.replace(b'OUTFILE', outname)
         code = system(cmd)
         if pycompat.sysplatform == b'OpenVMS' and code & 1:
             code = 0
         if code:
             raise error.Abort(
                 _(b"command '%s' failed: %s") % (cmd, explainexit(code))
             )
         with open(outname, b'rb') as fp:
             return fp.read()
     finally:
         try:
             if inname:
                 os.unlink(inname)
         except OSError:
             pass
         try:
             if outname:
                 os.unlink(outname)
         except OSError:
             pass
 
 
 _filtertable = {
     b'tempfile:': tempfilter,
     b'pipe:': pipefilter,
 }
 
 
 def filter(s, cmd):
     """filter a string through a command that transforms its input to its
     output"""
     for name, fn in pycompat.iteritems(_filtertable):
         if cmd.startswith(name):
             return fn(s, cmd[len(name) :].lstrip())
     return pipefilter(s, cmd)
 
 
 _hgexecutable = None
 
 
 def hgexecutable():
     """return location of the 'hg' executable.
 
     Defaults to $HG or 'hg' in the search path.
     """
     if _hgexecutable is None:
         hg = encoding.environ.get(b'HG')
         mainmod = sys.modules['__main__']
         if hg:
             _sethgexecutable(hg)
         elif resourceutil.mainfrozen():
             if getattr(sys, 'frozen', None) == 'macosx_app':
                 # Env variable set by py2app
                 _sethgexecutable(encoding.environ[b'EXECUTABLEPATH'])
             else:
                 _sethgexecutable(pycompat.sysexecutable)
         elif (
             not pycompat.iswindows
             and os.path.basename(getattr(mainmod, '__file__', '')) == 'hg'
         ):
             _sethgexecutable(pycompat.fsencode(mainmod.__file__))
         else:
             _sethgexecutable(
                 findexe(b'hg') or os.path.basename(pycompat.sysargv[0])
             )
     return _hgexecutable
 
 
 def _sethgexecutable(path):
     """set location of the 'hg' executable"""
     global _hgexecutable
     _hgexecutable = path
 
 
 def _testfileno(f, stdf):
     fileno = getattr(f, 'fileno', None)
     try:
         return fileno and fileno() == stdf.fileno()
     except io.UnsupportedOperation:
         return False  # fileno() raised UnsupportedOperation
 
 
 def isstdin(f):
     return _testfileno(f, sys.__stdin__)
 
 
 def isstdout(f):
     return _testfileno(f, sys.__stdout__)
 
 
 def protectstdio(uin, uout):
     """Duplicate streams and redirect original if (uin, uout) are stdio
 
     If uin is stdin, it's redirected to /dev/null. If uout is stdout, it's
     redirected to stderr so the output is still readable.
 
     Returns (fin, fout) which point to the original (uin, uout) fds, but
     may be copy of (uin, uout). The returned streams can be considered
     "owned" in that print(), exec(), etc. never reach to them.
     """
     uout.flush()
     fin, fout = uin, uout
     if _testfileno(uin, stdin):
         newfd = os.dup(uin.fileno())
         nullfd = os.open(os.devnull, os.O_RDONLY)
         os.dup2(nullfd, uin.fileno())
         os.close(nullfd)
         fin = os.fdopen(newfd, 'rb')
     if _testfileno(uout, stdout):
         newfd = os.dup(uout.fileno())
         os.dup2(stderr.fileno(), uout.fileno())
         fout = os.fdopen(newfd, 'wb')
     return fin, fout
 
 
 def restorestdio(uin, uout, fin, fout):
     """Restore (uin, uout) streams from possibly duplicated (fin, fout)"""
     uout.flush()
     for f, uif in [(fin, uin), (fout, uout)]:
         if f is not uif:
             os.dup2(f.fileno(), uif.fileno())
             f.close()
 
 
 def shellenviron(environ=None):
     """return environ with optional override, useful for shelling out"""
 
     def py2shell(val):
         """convert python object into string that is useful to shell"""
         if val is None or val is False:
             return b'0'
         if val is True:
             return b'1'
         return pycompat.bytestr(val)
 
     env = dict(encoding.environ)
     if environ:
         env.update((k, py2shell(v)) for k, v in pycompat.iteritems(environ))
     env[b'HG'] = hgexecutable()
     return env
 
 
 if pycompat.iswindows:
 
     def shelltonative(cmd, env):
         return platform.shelltocmdexe(  # pytype: disable=module-attr
             cmd, shellenviron(env)
         )
 
     tonativestr = encoding.strfromlocal
 else:
 
     def shelltonative(cmd, env):
         return cmd
 
     tonativestr = pycompat.identity
 
 
 def tonativeenv(env):
     '''convert the environment from bytes to strings suitable for Popen(), etc.
     '''
     return pycompat.rapply(tonativestr, env)
 
 
 def system(cmd, environ=None, cwd=None, out=None):
     '''enhanced shell command execution.
     run with environment maybe modified, maybe in different dir.
 
     if out is specified, it is assumed to be a file-like object that has a
     write() method. stdout and stderr will be redirected to out.'''
     try:
         stdout.flush()
     except Exception:
         pass
-    cmd = quotecommand(cmd)
     env = shellenviron(environ)
     if out is None or isstdout(out):
         rc = subprocess.call(
             tonativestr(cmd),
             shell=True,
             close_fds=closefds,
             env=tonativeenv(env),
             cwd=pycompat.rapply(tonativestr, cwd),
         )
     else:
         proc = subprocess.Popen(
             tonativestr(cmd),
             shell=True,
             close_fds=closefds,
             env=tonativeenv(env),
             cwd=pycompat.rapply(tonativestr, cwd),
             stdout=subprocess.PIPE,
             stderr=subprocess.STDOUT,
         )
         for line in iter(proc.stdout.readline, b''):
             out.write(line)
         proc.wait()
         rc = proc.returncode
     if pycompat.sysplatform == b'OpenVMS' and rc & 1:
         rc = 0
     return rc
 
 
 _is_gui = None
 
 
 def _gui():
     '''Are we running in a GUI?'''
     if pycompat.isdarwin:
         if b'SSH_CONNECTION' in encoding.environ:
             # handle SSH access to a box where the user is logged in
             return False
         elif getattr(osutil, 'isgui', None):
             # check if a CoreGraphics session is available
             return osutil.isgui()
         else:
             # pure build; use a safe default
             return True
     else:
         return pycompat.iswindows or encoding.environ.get(b"DISPLAY")
 
 
 def gui():
     global _is_gui
     if _is_gui is None:
         _is_gui = _gui()
     return _is_gui
 
 
 def hgcmd():
     """Return the command used to execute current hg
 
     This is different from hgexecutable() because on Windows we want
     to avoid things opening new shell windows like batch files, so we
     get either the python call or current executable.
     """
     if resourceutil.mainfrozen():
         if getattr(sys, 'frozen', None) == 'macosx_app':
             # Env variable set by py2app
             return [encoding.environ[b'EXECUTABLEPATH']]
         else:
             return [pycompat.sysexecutable]
     return _gethgcmd()
 
 
 def rundetached(args, condfn):
     """Execute the argument list in a detached process.
 
     condfn is a callable which is called repeatedly and should return
     True once the child process is known to have started successfully.
     At this point, the child process PID is returned. If the child
     process fails to start or finishes before condfn() evaluates to
     True, return -1.
     """
     # Windows case is easier because the child process is either
     # successfully starting and validating the condition or exiting
     # on failure. We just poll on its PID. On Unix, if the child
     # process fails to start, it will be left in a zombie state until
     # the parent wait on it, which we cannot do since we expect a long
     # running process on success. Instead we listen for SIGCHLD telling
     # us our child process terminated.
     terminated = set()
 
     def handler(signum, frame):
         terminated.add(os.wait())
 
     prevhandler = None
     SIGCHLD = getattr(signal, 'SIGCHLD', None)
     if SIGCHLD is not None:
         prevhandler = signal.signal(SIGCHLD, handler)
     try:
         pid = spawndetached(args)
         while not condfn():
             if (pid in terminated or not testpid(pid)) and not condfn():
                 return -1
             time.sleep(0.1)
         return pid
     finally:
         if prevhandler is not None:
             signal.signal(signal.SIGCHLD, prevhandler)
 
 
 @contextlib.contextmanager
 def uninterruptible(warn):
     """Inhibit SIGINT handling on a region of code.
 
     Note that if this is called in a non-main thread, it turns into a no-op.
 
     Args:
       warn: A callable which takes no arguments, and returns True if the
             previous signal handling should be restored.
     """
 
     oldsiginthandler = [signal.getsignal(signal.SIGINT)]
     shouldbail = []
 
     def disabledsiginthandler(*args):
         if warn():
             signal.signal(signal.SIGINT, oldsiginthandler[0])
             del oldsiginthandler[0]
         shouldbail.append(True)
 
     try:
         try:
             signal.signal(signal.SIGINT, disabledsiginthandler)
         except ValueError:
             # wrong thread, oh well, we tried
             del oldsiginthandler[0]
         yield
     finally:
         if oldsiginthandler:
             signal.signal(signal.SIGINT, oldsiginthandler[0])
         if shouldbail:
             raise KeyboardInterrupt
 
 
 if pycompat.iswindows:
     # no fork on Windows, but we can create a detached process
     # https://msdn.microsoft.com/en-us/library/windows/desktop/ms684863.aspx
     # No stdlib constant exists for this value
     DETACHED_PROCESS = 0x00000008
     # Following creation flags might create a console GUI window.
     # Using subprocess.CREATE_NEW_CONSOLE might helps.
     # See https://phab.mercurial-scm.org/D1701 for discussion
     _creationflags = (
         DETACHED_PROCESS
         | subprocess.CREATE_NEW_PROCESS_GROUP  # pytype: disable=module-attr
     )
 
     def runbgcommand(
         script,
         env,
         shell=False,
         stdout=None,
         stderr=None,
         ensurestart=True,
         record_wait=None,
     ):
         '''Spawn a command without waiting for it to finish.'''
         # we can't use close_fds *and* redirect stdin. I'm not sure that we
         # need to because the detached process has no console connection.
         p = subprocess.Popen(
             tonativestr(script),
             shell=shell,
             env=tonativeenv(env),
             close_fds=True,
             creationflags=_creationflags,
             stdout=stdout,
             stderr=stderr,
         )
         if record_wait is not None:
             record_wait(p.wait)
 
 
 else:
 
     def runbgcommand(
         cmd,
         env,
         shell=False,
         stdout=None,
         stderr=None,
         ensurestart=True,
         record_wait=None,
     ):
         '''Spawn a command without waiting for it to finish.
 
 
         When `record_wait` is not None, the spawned process will not be fully
         detached and the `record_wait` argument will be called with a the
         `Subprocess.wait` function for the spawned process.  This is mostly
         useful for developers that need to make sure the spawned process
         finished before a certain point. (eg: writing test)'''
         if pycompat.isdarwin:
             # avoid crash in CoreFoundation in case another thread
             # calls gui() while we're calling fork().
             gui()
 
         # double-fork to completely detach from the parent process
         # based on http://code.activestate.com/recipes/278731
         if record_wait is None:
             pid = os.fork()
             if pid:
                 if not ensurestart:
                     # Even though we're not waiting on the child process,
                     # we still must call waitpid() on it at some point so
                     # it's not a zombie/defunct. This is especially relevant for
                     # chg since the parent process won't die anytime soon.
                     # We use a thread to make the overhead tiny.
                     def _do_wait():
                         os.waitpid(pid, 0)
 
                     threading.Thread(target=_do_wait, daemon=True).start()
                     return
                 # Parent process
                 (_pid, status) = os.waitpid(pid, 0)
                 if os.WIFEXITED(status):
                     returncode = os.WEXITSTATUS(status)
                 else:
                     returncode = -(os.WTERMSIG(status))
                 if returncode != 0:
                     # The child process's return code is 0 on success, an errno
                     # value on failure, or 255 if we don't have a valid errno
                     # value.
                     #
                     # (It would be slightly nicer to return the full exception info
                     # over a pipe as the subprocess module does.  For now it
                     # doesn't seem worth adding that complexity here, though.)
                     if returncode == 255:
                         returncode = errno.EINVAL
                     raise OSError(
                         returncode,
                         b'error running %r: %s'
                         % (cmd, os.strerror(returncode)),
                     )
                 return
 
         returncode = 255
         try:
             if record_wait is None:
                 # Start a new session
                 os.setsid()
 
             stdin = open(os.devnull, b'r')
             if stdout is None:
                 stdout = open(os.devnull, b'w')
             if stderr is None:
                 stderr = open(os.devnull, b'w')
 
             # connect stdin to devnull to make sure the subprocess can't
             # muck up that stream for mercurial.
             p = subprocess.Popen(
                 cmd,
                 shell=shell,
                 env=env,
                 close_fds=True,
                 stdin=stdin,
                 stdout=stdout,
                 stderr=stderr,
             )
             if record_wait is not None:
                 record_wait(p.wait)
             returncode = 0
         except EnvironmentError as ex:
             returncode = ex.errno & 0xFF
             if returncode == 0:
                 # This shouldn't happen, but just in case make sure the
                 # return code is never 0 here.
                 returncode = 255
         except Exception:
             returncode = 255
         finally:
             # mission accomplished, this child needs to exit and not
             # continue the hg process here.
             if record_wait is None:
                 os._exit(returncode)

mercurial/windows.py

@@ -1,680 +1,675 b''
 # windows.py - Windows utility function implementations for Mercurial
 #
 #  Copyright 2005-2009 Matt Mackall <[e-mail]> and others
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import errno
 import getpass
 import msvcrt
 import os
 import re
 import stat
 import string
 import sys
 
 from .i18n import _
 from .pycompat import getattr
 from . import (
     encoding,
     error,
     policy,
     pycompat,
     win32,
 )
 
 try:
     import _winreg as winreg  # pytype: disable=import-error
 
     winreg.CloseKey
 except ImportError:
     # py2 only
     import winreg  # pytype: disable=import-error
 
 osutil = policy.importmod('osutil')
 
 getfsmountpoint = win32.getvolumename
 getfstype = win32.getfstype
 getuser = win32.getuser
 hidewindow = win32.hidewindow
 makedir = win32.makedir
 nlinks = win32.nlinks
 oslink = win32.oslink
 samedevice = win32.samedevice
 samefile = win32.samefile
 setsignalhandler = win32.setsignalhandler
 spawndetached = win32.spawndetached
 split = os.path.split
 testpid = win32.testpid
 unlink = win32.unlink
 
 umask = 0o022
 
 
 class mixedfilemodewrapper(object):
     """Wraps a file handle when it is opened in read/write mode.
 
     fopen() and fdopen() on Windows have a specific-to-Windows requirement
     that files opened with mode r+, w+, or a+ make a call to a file positioning
     function when switching between reads and writes. Without this extra call,
     Python will raise a not very intuitive "IOError: [Errno 0] Error."
 
     This class wraps posixfile instances when the file is opened in read/write
     mode and automatically adds checks or inserts appropriate file positioning
     calls when necessary.
     """
 
     OPNONE = 0
     OPREAD = 1
     OPWRITE = 2
 
     def __init__(self, fp):
         object.__setattr__(self, '_fp', fp)
         object.__setattr__(self, '_lastop', 0)
 
     def __enter__(self):
         self._fp.__enter__()
         return self
 
     def __exit__(self, exc_type, exc_val, exc_tb):
         self._fp.__exit__(exc_type, exc_val, exc_tb)
 
     def __getattr__(self, name):
         return getattr(self._fp, name)
 
     def __setattr__(self, name, value):
         return self._fp.__setattr__(name, value)
 
     def _noopseek(self):
         self._fp.seek(0, os.SEEK_CUR)
 
     def seek(self, *args, **kwargs):
         object.__setattr__(self, '_lastop', self.OPNONE)
         return self._fp.seek(*args, **kwargs)
 
     def write(self, d):
         if self._lastop == self.OPREAD:
             self._noopseek()
 
         object.__setattr__(self, '_lastop', self.OPWRITE)
         return self._fp.write(d)
 
     def writelines(self, *args, **kwargs):
         if self._lastop == self.OPREAD:
             self._noopeseek()
 
         object.__setattr__(self, '_lastop', self.OPWRITE)
         return self._fp.writelines(*args, **kwargs)
 
     def read(self, *args, **kwargs):
         if self._lastop == self.OPWRITE:
             self._noopseek()
 
         object.__setattr__(self, '_lastop', self.OPREAD)
         return self._fp.read(*args, **kwargs)
 
     def readline(self, *args, **kwargs):
         if self._lastop == self.OPWRITE:
             self._noopseek()
 
         object.__setattr__(self, '_lastop', self.OPREAD)
         return self._fp.readline(*args, **kwargs)
 
     def readlines(self, *args, **kwargs):
         if self._lastop == self.OPWRITE:
             self._noopseek()
 
         object.__setattr__(self, '_lastop', self.OPREAD)
         return self._fp.readlines(*args, **kwargs)
 
 
 class fdproxy(object):
     """Wraps osutil.posixfile() to override the name attribute to reflect the
     underlying file name.
     """
 
     def __init__(self, name, fp):
         self.name = name
         self._fp = fp
 
     def __enter__(self):
         self._fp.__enter__()
         # Return this wrapper for the context manager so that the name is
         # still available.
         return self
 
     def __exit__(self, exc_type, exc_value, traceback):
         self._fp.__exit__(exc_type, exc_value, traceback)
 
     def __iter__(self):
         return iter(self._fp)
 
     def __getattr__(self, name):
         return getattr(self._fp, name)
 
 
 def posixfile(name, mode=b'r', buffering=-1):
     '''Open a file with even more POSIX-like semantics'''
     try:
         fp = osutil.posixfile(name, mode, buffering)  # may raise WindowsError
 
         # PyFile_FromFd() ignores the name, and seems to report fp.name as the
         # underlying file descriptor.
         if pycompat.ispy3:
             fp = fdproxy(name, fp)
 
         # The position when opening in append mode is implementation defined, so
         # make it consistent with other platforms, which position at EOF.
         if b'a' in mode:
             fp.seek(0, os.SEEK_END)
 
         if b'+' in mode:
             return mixedfilemodewrapper(fp)
 
         return fp
     except WindowsError as err:
         # convert to a friendlier exception
         raise IOError(
             err.errno, '%s: %s' % (encoding.strfromlocal(name), err.strerror)
         )
 
 
 # may be wrapped by win32mbcs extension
 listdir = osutil.listdir
 
 
 class winstdout(object):
     '''stdout on windows misbehaves if sent through a pipe'''
 
     def __init__(self, fp):
         self.fp = fp
 
     def __getattr__(self, key):
         return getattr(self.fp, key)
 
     def close(self):
         try:
             self.fp.close()
         except IOError:
             pass
 
     def write(self, s):
         try:
             # This is workaround for "Not enough space" error on
             # writing large size of data to console.
             limit = 16000
             l = len(s)
             start = 0
             self.softspace = 0
             while start < l:
                 end = start + limit
                 self.fp.write(s[start:end])
                 start = end
         except IOError as inst:
             if inst.errno != 0 and not win32.lasterrorwaspipeerror(inst):
                 raise
             self.close()
             raise IOError(errno.EPIPE, 'Broken pipe')
 
     def flush(self):
         try:
             return self.fp.flush()
         except IOError as inst:
             if not win32.lasterrorwaspipeerror(inst):
                 raise
             raise IOError(errno.EPIPE, 'Broken pipe')
 
 
 def openhardlinks():
     return True
 
 
 def parsepatchoutput(output_line):
     """parses the output produced by patch and returns the filename"""
     pf = output_line[14:]
     if pf[0] == b'`':
         pf = pf[1:-1]  # Remove the quotes
     return pf
 
 
 def sshargs(sshcmd, host, user, port):
     '''Build argument list for ssh or Plink'''
     pflag = b'plink' in sshcmd.lower() and b'-P' or b'-p'
     args = user and (b"%s@%s" % (user, host)) or host
     if args.startswith(b'-') or args.startswith(b'/'):
         raise error.Abort(
             _(b'illegal ssh hostname or username starting with - or /: %s')
             % args
         )
     args = shellquote(args)
     if port:
         args = b'%s %s %s' % (pflag, shellquote(port), args)
     return args
 
 
 def setflags(f, l, x):
     pass
 
 
 def copymode(src, dst, mode=None, enforcewritable=False):
     pass
 
 
 def checkexec(path):
     return False
 
 
 def checklink(path):
     return False
 
 
 def setbinary(fd):
     # When run without console, pipes may expose invalid
     # fileno(), usually set to -1.
     fno = getattr(fd, 'fileno', None)
     if fno is not None and fno() >= 0:
         msvcrt.setmode(fno(), os.O_BINARY)  # pytype: disable=module-attr
 
 
 def pconvert(path):
     return path.replace(pycompat.ossep, b'/')
 
 
 def localpath(path):
     return path.replace(b'/', b'\\')
 
 
 def normpath(path):
     return pconvert(os.path.normpath(path))
 
 
 def normcase(path):
     return encoding.upper(path)  # NTFS compares via upper()
 
 
 # see posix.py for definitions
 normcasespec = encoding.normcasespecs.upper
 normcasefallback = encoding.upperfallback
 
 
 def samestat(s1, s2):
     return False
 
 
 def shelltocmdexe(path, env):
     r"""Convert shell variables in the form $var and ${var} inside ``path``
     to %var% form.  Existing Windows style variables are left unchanged.
 
     The variables are limited to the given environment.  Unknown variables are
     left unchanged.
 
     >>> e = {b'var1': b'v1', b'var2': b'v2', b'var3': b'v3'}
     >>> # Only valid values are expanded
     >>> shelltocmdexe(b'cmd $var1 ${var2} %var3% $missing ${missing} %missing%',
     ...               e)
     'cmd %var1% %var2% %var3% $missing ${missing} %missing%'
     >>> # Single quote prevents expansion, as does \$ escaping
     >>> shelltocmdexe(b"cmd '$var1 ${var2} %var3%' \$var1 \${var2} \\", e)
     'cmd "$var1 ${var2} %var3%" $var1 ${var2} \\'
     >>> # $$ is not special. %% is not special either, but can be the end and
     >>> # start of consecutive variables
     >>> shelltocmdexe(b"cmd $$ %% %var1%%var2%", e)
     'cmd $$ %% %var1%%var2%'
     >>> # No double substitution
     >>> shelltocmdexe(b"$var1 %var1%", {b'var1': b'%var2%', b'var2': b'boom'})
     '%var1% %var1%'
     >>> # Tilde expansion
     >>> shelltocmdexe(b"~/dir ~\dir2 ~tmpfile \~/", {})
     '%USERPROFILE%/dir %USERPROFILE%\\dir2 ~tmpfile ~/'
     """
     if not any(c in path for c in b"$'~"):
         return path
 
     varchars = pycompat.sysbytes(string.ascii_letters + string.digits) + b'_-'
 
     res = b''
     index = 0
     pathlen = len(path)
     while index < pathlen:
         c = path[index : index + 1]
         if c == b'\'':  # no expansion within single quotes
             path = path[index + 1 :]
             pathlen = len(path)
             try:
                 index = path.index(b'\'')
                 res += b'"' + path[:index] + b'"'
             except ValueError:
                 res += c + path
                 index = pathlen - 1
         elif c == b'%':  # variable
             path = path[index + 1 :]
             pathlen = len(path)
             try:
                 index = path.index(b'%')
             except ValueError:
                 res += b'%' + path
                 index = pathlen - 1
             else:
                 var = path[:index]
                 res += b'%' + var + b'%'
         elif c == b'$':  # variable
             if path[index + 1 : index + 2] == b'{':
                 path = path[index + 2 :]
                 pathlen = len(path)
                 try:
                     index = path.index(b'}')
                     var = path[:index]
 
                     # See below for why empty variables are handled specially
                     if env.get(var, b'') != b'':
                         res += b'%' + var + b'%'
                     else:
                         res += b'${' + var + b'}'
                 except ValueError:
                     res += b'${' + path
                     index = pathlen - 1
             else:
                 var = b''
                 index += 1
                 c = path[index : index + 1]
                 while c != b'' and c in varchars:
                     var += c
                     index += 1
                     c = path[index : index + 1]
                 # Some variables (like HG_OLDNODE) may be defined, but have an
                 # empty value.  Those need to be skipped because when spawning
                 # cmd.exe to run the hook, it doesn't replace %VAR% for an empty
                 # VAR, and that really confuses things like revset expressions.
                 # OTOH, if it's left in Unix format and the hook runs sh.exe, it
                 # will substitute to an empty string, and everything is happy.
                 if env.get(var, b'') != b'':
                     res += b'%' + var + b'%'
                 else:
                     res += b'$' + var
 
                 if c != b'':
                     index -= 1
         elif (
             c == b'~'
             and index + 1 < pathlen
             and path[index + 1 : index + 2] in (b'\\', b'/')
         ):
             res += b"%USERPROFILE%"
         elif (
             c == b'\\'
             and index + 1 < pathlen
             and path[index + 1 : index + 2] in (b'$', b'~')
         ):
             # Skip '\', but only if it is escaping $ or ~
             res += path[index + 1 : index + 2]
             index += 1
         else:
             res += c
 
         index += 1
     return res
 
 
 # A sequence of backslashes is special iff it precedes a double quote:
 # - if there's an even number of backslashes, the double quote is not
 #   quoted (i.e. it ends the quoted region)
 # - if there's an odd number of backslashes, the double quote is quoted
 # - in both cases, every pair of backslashes is unquoted into a single
 #   backslash
 # (See http://msdn2.microsoft.com/en-us/library/a1y7w461.aspx )
 # So, to quote a string, we must surround it in double quotes, double
 # the number of backslashes that precede double quotes and add another
 # backslash before every double quote (being careful with the double
 # quote we've appended to the end)
 _quotere = None
 _needsshellquote = None
 
 
 def shellquote(s):
     r"""
     >>> shellquote(br'C:\Users\xyz')
     '"C:\\Users\\xyz"'
     >>> shellquote(br'C:\Users\xyz/mixed')
     '"C:\\Users\\xyz/mixed"'
     >>> # Would be safe not to quote too, since it is all double backslashes
     >>> shellquote(br'C:\\Users\\xyz')
     '"C:\\\\Users\\\\xyz"'
     >>> # But this must be quoted
     >>> shellquote(br'C:\\Users\\xyz/abc')
     '"C:\\\\Users\\\\xyz/abc"'
     """
     global _quotere
     if _quotere is None:
         _quotere = re.compile(br'(\\*)("|\\$)')
     global _needsshellquote
     if _needsshellquote is None:
         # ":" is also treated as "safe character", because it is used as a part
         # of path name on Windows.  "\" is also part of a path name, but isn't
         # safe because shlex.split() (kind of) treats it as an escape char and
         # drops it.  It will leave the next character, even if it is another
         # "\".
         _needsshellquote = re.compile(br'[^a-zA-Z0-9._:/-]').search
     if s and not _needsshellquote(s) and not _quotere.search(s):
         # "s" shouldn't have to be quoted
         return s
     return b'"%s"' % _quotere.sub(br'\1\1\\\2', s)
 
 
 def _unquote(s):
     if s.startswith(b'"') and s.endswith(b'"'):
         return s[1:-1]
     return s
 
 
 def shellsplit(s):
     """Parse a command string in cmd.exe way (best-effort)"""
     return pycompat.maplist(_unquote, pycompat.shlexsplit(s, posix=False))
 
 
-def quotecommand(cmd):
-    """Build a command string suitable for os.popen* calls."""
-    return cmd
-
-
 # if you change this stub into a real check, please try to implement the
 # username and groupname functions above, too.
 def isowner(st):
     return True
 
 
 def findexe(command):
     '''Find executable for command searching like cmd.exe does.
     If command is a basename then PATH is searched for command.
     PATH isn't searched if command is an absolute or relative path.
     An extension from PATHEXT is found and added if not present.
     If command isn't found None is returned.'''
     pathext = encoding.environ.get(b'PATHEXT', b'.COM;.EXE;.BAT;.CMD')
     pathexts = [ext for ext in pathext.lower().split(pycompat.ospathsep)]
     if os.path.splitext(command)[1].lower() in pathexts:
         pathexts = [b'']
 
     def findexisting(pathcommand):
         """Will append extension (if needed) and return existing file"""
         for ext in pathexts:
             executable = pathcommand + ext
             if os.path.exists(executable):
                 return executable
         return None
 
     if pycompat.ossep in command:
         return findexisting(command)
 
     for path in encoding.environ.get(b'PATH', b'').split(pycompat.ospathsep):
         executable = findexisting(os.path.join(path, command))
         if executable is not None:
             return executable
     return findexisting(os.path.expanduser(os.path.expandvars(command)))
 
 
 _wantedkinds = {stat.S_IFREG, stat.S_IFLNK}
 
 
 def statfiles(files):
     '''Stat each file in files. Yield each stat, or None if a file
     does not exist or has a type we don't care about.
 
     Cluster and cache stat per directory to minimize number of OS stat calls.'''
     dircache = {}  # dirname -> filename -> status | None if file does not exist
     getkind = stat.S_IFMT
     for nf in files:
         nf = normcase(nf)
         dir, base = os.path.split(nf)
         if not dir:
             dir = b'.'
         cache = dircache.get(dir, None)
         if cache is None:
             try:
                 dmap = {
                     normcase(n): s
                     for n, k, s in listdir(dir, True)
                     if getkind(s.st_mode) in _wantedkinds
                 }
             except OSError as err:
                 # Python >= 2.5 returns ENOENT and adds winerror field
                 # EINVAL is raised if dir is not a directory.
                 if err.errno not in (errno.ENOENT, errno.EINVAL, errno.ENOTDIR):
                     raise
                 dmap = {}
             cache = dircache.setdefault(dir, dmap)
         yield cache.get(base, None)
 
 
 def username(uid=None):
     """Return the name of the user with the given uid.
 
     If uid is None, return the name of the current user."""
     if not uid:
         return pycompat.fsencode(getpass.getuser())
     return None
 
 
 def groupname(gid=None):
     """Return the name of the group with the given gid.
 
     If gid is None, return the name of the current group."""
     return None
 
 
 def readlink(pathname):
     return pycompat.fsencode(os.readlink(pycompat.fsdecode(pathname)))
 
 
 def removedirs(name):
     """special version of os.removedirs that does not remove symlinked
     directories or junction points if they actually contain files"""
     if listdir(name):
         return
     os.rmdir(name)
     head, tail = os.path.split(name)
     if not tail:
         head, tail = os.path.split(head)
     while head and tail:
         try:
             if listdir(head):
                 return
             os.rmdir(head)
         except (ValueError, OSError):
             break
         head, tail = os.path.split(head)
 
 
 def rename(src, dst):
     '''atomically rename file src to dst, replacing dst if it exists'''
     try:
         os.rename(src, dst)
     except OSError as e:
         if e.errno != errno.EEXIST:
             raise
         unlink(dst)
         os.rename(src, dst)
 
 
 def gethgcmd():
     return [encoding.strtolocal(arg) for arg in [sys.executable] + sys.argv[:1]]
 
 
 def groupmembers(name):
     # Don't support groups on Windows for now
     raise KeyError
 
 
 def isexec(f):
     return False
 
 
 class cachestat(object):
     def __init__(self, path):
         pass
 
     def cacheable(self):
         return False
 
 
 def lookupreg(key, valname=None, scope=None):
     ''' Look up a key/value name in the Windows registry.
 
     valname: value name. If unspecified, the default value for the key
     is used.
     scope: optionally specify scope for registry lookup, this can be
     a sequence of scopes to look up in order. Default (CURRENT_USER,
     LOCAL_MACHINE).
     '''
     if scope is None:
         scope = (winreg.HKEY_CURRENT_USER, winreg.HKEY_LOCAL_MACHINE)
     elif not isinstance(scope, (list, tuple)):
         scope = (scope,)
     for s in scope:
         try:
             with winreg.OpenKey(s, encoding.strfromlocal(key)) as hkey:
                 name = valname and encoding.strfromlocal(valname) or valname
                 val = winreg.QueryValueEx(hkey, name)[0]
                 # never let a Unicode string escape into the wild
                 return encoding.unitolocal(val)
         except EnvironmentError:
             pass
 
 
 expandglobs = True
 
 
 def statislink(st):
     '''check whether a stat result is a symlink'''
     return False
 
 
 def statisexec(st):
     '''check whether a stat result is an executable file'''
     return False
 
 
 def poll(fds):
     # see posix.py for description
     raise NotImplementedError()
 
 
 def readpipe(pipe):
     """Read all available data from a pipe."""
     chunks = []
     while True:
         size = win32.peekpipe(pipe)
         if not size:
             break
 
         s = pipe.read(size)
         if not s:
             break
         chunks.append(s)
 
     return b''.join(chunks)
 
 
 def bindunixsocket(sock, path):
     raise NotImplementedError('unsupported platform')