merge with stable

mercurial/dispatch.py

@@ -1,729 +1,738 b''
 # dispatch.py - command dispatching for mercurial
 #
 # Copyright 2005-2007 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from i18n import _
 import os, sys, atexit, signal, pdb, socket, errno, shlex, time, traceback, re
 import util, commands, hg, fancyopts, extensions, hook, error
 import cmdutil, encoding
 import ui as uimod
 
 class request(object):
     def __init__(self, args, ui=None, repo=None, fin=None, fout=None, ferr=None):
         self.args = args
         self.ui = ui
         self.repo = repo
 
         # input/output/error streams
         self.fin = fin
         self.fout = fout
         self.ferr = ferr
 
 def run():
     "run the command in sys.argv"
     sys.exit(dispatch(request(sys.argv[1:])))
 
 def dispatch(req):
     "run the command specified in req.args"
     if req.ferr:
         ferr = req.ferr
     elif req.ui:
         ferr = req.ui.ferr
     else:
         ferr = sys.stderr
 
     try:
         if not req.ui:
             req.ui = uimod.ui()
         if '--traceback' in req.args:
             req.ui.setconfig('ui', 'traceback', 'on')
 
         # set ui streams from the request
         if req.fin:
             req.ui.fin = req.fin
         if req.fout:
             req.ui.fout = req.fout
         if req.ferr:
             req.ui.ferr = req.ferr
     except util.Abort, inst:
         ferr.write(_("abort: %s\n") % inst)
         if inst.hint:
             ferr.write(_("(%s)\n") % inst.hint)
         return -1
     except error.ParseError, inst:
         if len(inst.args) > 1:
             ferr.write(_("hg: parse error at %s: %s\n") %
                              (inst.args[1], inst.args[0]))
         else:
             ferr.write(_("hg: parse error: %s\n") % inst.args[0])
         return -1
 
     return _runcatch(req)
 
 def _runcatch(req):
     def catchterm(*args):
         raise error.SignalInterrupt
 
     ui = req.ui
     try:
         for name in 'SIGBREAK', 'SIGHUP', 'SIGTERM':
             num = getattr(signal, name, None)
             if num:
                 signal.signal(num, catchterm)
     except ValueError:
         pass # happens if called in a thread
 
     try:
         try:
             # enter the debugger before command execution
             if '--debugger' in req.args:
                 ui.warn(_("entering debugger - "
                         "type c to continue starting hg or h for help\n"))
                 pdb.set_trace()
             try:
                 return _dispatch(req)
             finally:
                 ui.flush()
         except:
             # enter the debugger when we hit an exception
             if '--debugger' in req.args:
                 traceback.print_exc()
                 pdb.post_mortem(sys.exc_info()[2])
             ui.traceback()
             raise
 
     # Global exception handling, alphabetically
     # Mercurial-specific first, followed by built-in and library exceptions
     except error.AmbiguousCommand, inst:
         ui.warn(_("hg: command '%s' is ambiguous:\n    %s\n") %
                 (inst.args[0], " ".join(inst.args[1])))
     except error.ParseError, inst:
         if len(inst.args) > 1:
             ui.warn(_("hg: parse error at %s: %s\n") %
                              (inst.args[1], inst.args[0]))
         else:
             ui.warn(_("hg: parse error: %s\n") % inst.args[0])
         return -1
     except error.LockHeld, inst:
         if inst.errno == errno.ETIMEDOUT:
             reason = _('timed out waiting for lock held by %s') % inst.locker
         else:
             reason = _('lock held by %s') % inst.locker
         ui.warn(_("abort: %s: %s\n") % (inst.desc or inst.filename, reason))
     except error.LockUnavailable, inst:
         ui.warn(_("abort: could not lock %s: %s\n") %
                (inst.desc or inst.filename, inst.strerror))
     except error.CommandError, inst:
         if inst.args[0]:
             ui.warn(_("hg %s: %s\n") % (inst.args[0], inst.args[1]))
             commands.help_(ui, inst.args[0], full=False, command=True)
         else:
             ui.warn(_("hg: %s\n") % inst.args[1])
             commands.help_(ui, 'shortlist')
     except error.OutOfBandError, inst:
         ui.warn("abort: remote error:\n")
         ui.warn(''.join(inst.args))
     except error.RepoError, inst:
         ui.warn(_("abort: %s!\n") % inst)
         if inst.hint:
             ui.warn(_("(%s)\n") % inst.hint)
     except error.ResponseError, inst:
         ui.warn(_("abort: %s") % inst.args[0])
         if not isinstance(inst.args[1], basestring):
             ui.warn(" %r\n" % (inst.args[1],))
         elif not inst.args[1]:
             ui.warn(_(" empty string\n"))
         else:
             ui.warn("\n%r\n" % util.ellipsis(inst.args[1]))
     except error.RevlogError, inst:
         ui.warn(_("abort: %s!\n") % inst)
     except error.SignalInterrupt:
         ui.warn(_("killed!\n"))
     except error.UnknownCommand, inst:
         ui.warn(_("hg: unknown command '%s'\n") % inst.args[0])
         try:
             # check if the command is in a disabled extension
             # (but don't check for extensions themselves)
             commands.help_(ui, inst.args[0], unknowncmd=True)
         except error.UnknownCommand:
             commands.help_(ui, 'shortlist')
     except util.Abort, inst:
         ui.warn(_("abort: %s\n") % inst)
         if inst.hint:
             ui.warn(_("(%s)\n") % inst.hint)
     except ImportError, inst:
         ui.warn(_("abort: %s!\n") % inst)
         m = str(inst).split()[-1]
         if m in "mpatch bdiff".split():
             ui.warn(_("(did you forget to compile extensions?)\n"))
         elif m in "zlib".split():
             ui.warn(_("(is your Python install correct?)\n"))
     except IOError, inst:
         if util.safehasattr(inst, "code"):
             ui.warn(_("abort: %s\n") % inst)
         elif util.safehasattr(inst, "reason"):
             try: # usually it is in the form (errno, strerror)
                 reason = inst.reason.args[1]
             except (AttributeError, IndexError):
                  # it might be anything, for example a string
                 reason = inst.reason
             ui.warn(_("abort: error: %s\n") % reason)
         elif util.safehasattr(inst, "args") and inst.args[0] == errno.EPIPE:
             if ui.debugflag:
                 ui.warn(_("broken pipe\n"))
         elif getattr(inst, "strerror", None):
             if getattr(inst, "filename", None):
                 ui.warn(_("abort: %s: %s\n") % (inst.strerror, inst.filename))
             else:
                 ui.warn(_("abort: %s\n") % inst.strerror)
         else:
             raise
     except OSError, inst:
         if getattr(inst, "filename", None):
             ui.warn(_("abort: %s: %s\n") % (inst.strerror, inst.filename))
         else:
             ui.warn(_("abort: %s\n") % inst.strerror)
     except KeyboardInterrupt:
         try:
             ui.warn(_("interrupted!\n"))
         except IOError, inst:
             if inst.errno == errno.EPIPE:
                 if ui.debugflag:
                     ui.warn(_("\nbroken pipe\n"))
             else:
                 raise
     except MemoryError:
         ui.warn(_("abort: out of memory\n"))
     except SystemExit, inst:
         # Commands shouldn't sys.exit directly, but give a return code.
         # Just in case catch this and and pass exit code to caller.
         return inst.code
     except socket.error, inst:
         ui.warn(_("abort: %s\n") % inst.args[-1])
     except:
         ui.warn(_("** unknown exception encountered,"
                   " please report by visiting\n"))
         ui.warn(_("**  http://mercurial.selenic.com/wiki/BugTracker\n"))
         ui.warn(_("** Python %s\n") % sys.version.replace('\n', ''))
         ui.warn(_("** Mercurial Distributed SCM (version %s)\n")
                % util.version())
         ui.warn(_("** Extensions loaded: %s\n")
                % ", ".join([x[0] for x in extensions.extensions()]))
         raise
 
     return -1
 
 def aliasargs(fn, givenargs):
     args = getattr(fn, 'args', [])
     if args and givenargs:
         cmd = ' '.join(map(util.shellquote, args))
 
         nums = []
         def replacer(m):
             num = int(m.group(1)) - 1
             nums.append(num)
             return givenargs[num]
         cmd = re.sub(r'\$(\d+|\$)', replacer, cmd)
         givenargs = [x for i, x in enumerate(givenargs)
                      if i not in nums]
         args = shlex.split(cmd)
     return args + givenargs
 
 class cmdalias(object):
     def __init__(self, name, definition, cmdtable):
         self.name = self.cmd = name
         self.cmdname = ''
         self.definition = definition
         self.args = []
         self.opts = []
         self.help = ''
         self.norepo = True
         self.badalias = False
 
         try:
             aliases, entry = cmdutil.findcmd(self.name, cmdtable)
             for alias, e in cmdtable.iteritems():
                 if e is entry:
                     self.cmd = alias
                     break
             self.shadows = True
         except error.UnknownCommand:
             self.shadows = False
 
         if not self.definition:
             def fn(ui, *args):
                 ui.warn(_("no definition for alias '%s'\n") % self.name)
                 return 1
             self.fn = fn
             self.badalias = True
 
             return
 
         if self.definition.startswith('!'):
             self.shell = True
             def fn(ui, *args):
                 env = {'HG_ARGS': ' '.join((self.name,) + args)}
                 def _checkvar(m):
                     if m.groups()[0] == '$':
                         return m.group()
                     elif int(m.groups()[0]) <= len(args):
                         return m.group()
                     else:
                         ui.debug("No argument found for substitution "
                                  "of %i variable in alias '%s' definition."
                                  % (int(m.groups()[0]), self.name))
                         return ''
                 cmd = re.sub(r'\$(\d+|\$)', _checkvar, self.definition[1:])
                 replace = dict((str(i + 1), arg) for i, arg in enumerate(args))
                 replace['0'] = self.name
                 replace['@'] = ' '.join(args)
                 cmd = util.interpolate(r'\$', replace, cmd, escape_prefix=True)
                 return util.system(cmd, environ=env, out=ui.fout)
             self.fn = fn
             return
 
         args = shlex.split(self.definition)
         self.cmdname = cmd = args.pop(0)
         args = map(util.expandpath, args)
 
         for invalidarg in ("--cwd", "-R", "--repository", "--repo"):
             if _earlygetopt([invalidarg], args):
                 def fn(ui, *args):
                     ui.warn(_("error in definition for alias '%s': %s may only "
                               "be given on the command line\n")
                             % (self.name, invalidarg))
                     return 1
 
                 self.fn = fn
                 self.badalias = True
                 return
 
         try:
             tableentry = cmdutil.findcmd(cmd, cmdtable, False)[1]
             if len(tableentry) > 2:
                 self.fn, self.opts, self.help = tableentry
             else:
                 self.fn, self.opts = tableentry
 
             self.args = aliasargs(self.fn, args)
             if cmd not in commands.norepo.split(' '):
                 self.norepo = False
             if self.help.startswith("hg " + cmd):
                 # drop prefix in old-style help lines so hg shows the alias
                 self.help = self.help[4 + len(cmd):]
             self.__doc__ = self.fn.__doc__
 
         except error.UnknownCommand:
             def fn(ui, *args):
                 ui.warn(_("alias '%s' resolves to unknown command '%s'\n") \
                             % (self.name, cmd))
                 try:
                     # check if the command is in a disabled extension
                     commands.help_(ui, cmd, unknowncmd=True)
                 except error.UnknownCommand:
                     pass
                 return 1
             self.fn = fn
             self.badalias = True
         except error.AmbiguousCommand:
             def fn(ui, *args):
                 ui.warn(_("alias '%s' resolves to ambiguous command '%s'\n") \
                             % (self.name, cmd))
                 return 1
             self.fn = fn
             self.badalias = True
 
     def __call__(self, ui, *args, **opts):
         if self.shadows:
             ui.debug("alias '%s' shadows command '%s'\n" %
                      (self.name, self.cmdname))
 
         if util.safehasattr(self, 'shell'):
             return self.fn(ui, *args, **opts)
         else:
             try:
                 util.checksignature(self.fn)(ui, *args, **opts)
             except error.SignatureError:
                 args = ' '.join([self.cmdname] + self.args)
                 ui.debug("alias '%s' expands to '%s'\n" % (self.name, args))
                 raise
 
 def addaliases(ui, cmdtable):
     # aliases are processed after extensions have been loaded, so they
     # may use extension commands. Aliases can also use other alias definitions,
     # but only if they have been defined prior to the current definition.
     for alias, definition in ui.configitems('alias'):
         aliasdef = cmdalias(alias, definition, cmdtable)
+
+        try:
+            olddef = cmdtable[aliasdef.cmd][0]
+            if olddef.definition == aliasdef.definition:
+                continue
+        except (KeyError, AttributeError):
+            # definition might not exist or it might not be a cmdalias
+            pass
+
         cmdtable[aliasdef.cmd] = (aliasdef, aliasdef.opts, aliasdef.help)
         if aliasdef.norepo:
             commands.norepo += ' %s' % alias
 
 def _parse(ui, args):
     options = {}
     cmdoptions = {}
 
     try:
         args = fancyopts.fancyopts(args, commands.globalopts, options)
     except fancyopts.getopt.GetoptError, inst:
         raise error.CommandError(None, inst)
 
     if args:
         cmd, args = args[0], args[1:]
         aliases, entry = cmdutil.findcmd(cmd, commands.table,
                                      ui.config("ui", "strict"))
         cmd = aliases[0]
         args = aliasargs(entry[0], args)
         defaults = ui.config("defaults", cmd)
         if defaults:
             args = map(util.expandpath, shlex.split(defaults)) + args
         c = list(entry[1])
     else:
         cmd = None
         c = []
 
     # combine global options into local
     for o in commands.globalopts:
         c.append((o[0], o[1], options[o[1]], o[3]))
 
     try:
         args = fancyopts.fancyopts(args, c, cmdoptions, True)
     except fancyopts.getopt.GetoptError, inst:
         raise error.CommandError(cmd, inst)
 
     # separate global options back out
     for o in commands.globalopts:
         n = o[1]
         options[n] = cmdoptions[n]
         del cmdoptions[n]
 
     return (cmd, cmd and entry[0] or None, args, options, cmdoptions)
 
 def _parseconfig(ui, config):
     """parse the --config options from the command line"""
     configs = []
 
     for cfg in config:
         try:
             name, value = cfg.split('=', 1)
             section, name = name.split('.', 1)
             if not section or not name:
                 raise IndexError
             ui.setconfig(section, name, value)
             configs.append((section, name, value))
         except (IndexError, ValueError):
             raise util.Abort(_('malformed --config option: %r '
                                '(use --config section.name=value)') % cfg)
 
     return configs
 
 def _earlygetopt(aliases, args):
     """Return list of values for an option (or aliases).
 
     The values are listed in the order they appear in args.
     The options and values are removed from args.
     """
     try:
         argcount = args.index("--")
     except ValueError:
         argcount = len(args)
     shortopts = [opt for opt in aliases if len(opt) == 2]
     values = []
     pos = 0
     while pos < argcount:
         if args[pos] in aliases:
             if pos + 1 >= argcount:
                 # ignore and let getopt report an error if there is no value
                 break
             del args[pos]
             values.append(args.pop(pos))
             argcount -= 2
         elif args[pos][:2] in shortopts:
             # short option can have no following space, e.g. hg log -Rfoo
             values.append(args.pop(pos)[2:])
             argcount -= 1
         else:
             pos += 1
     return values
 
 def runcommand(lui, repo, cmd, fullargs, ui, options, d, cmdpats, cmdoptions):
     # run pre-hook, and abort if it fails
     ret = hook.hook(lui, repo, "pre-%s" % cmd, False, args=" ".join(fullargs),
                     pats=cmdpats, opts=cmdoptions)
     if ret:
         return ret
     ret = _runcommand(ui, options, cmd, d)
     # run post-hook, passing command result
     hook.hook(lui, repo, "post-%s" % cmd, False, args=" ".join(fullargs),
               result=ret, pats=cmdpats, opts=cmdoptions)
     return ret
 
 def _getlocal(ui, rpath):
     """Return (path, local ui object) for the given target path.
 
     Takes paths in [cwd]/.hg/hgrc into account."
     """
     try:
         wd = os.getcwd()
     except OSError, e:
         raise util.Abort(_("error getting current working directory: %s") %
                          e.strerror)
     path = cmdutil.findrepo(wd) or ""
     if not path:
         lui = ui
     else:
         lui = ui.copy()
         lui.readconfig(os.path.join(path, ".hg", "hgrc"), path)
 
     if rpath and rpath[-1]:
         path = lui.expandpath(rpath[-1])
         lui = ui.copy()
         lui.readconfig(os.path.join(path, ".hg", "hgrc"), path)
 
     return path, lui
 
 def _checkshellalias(lui, ui, args):
     norepo = commands.norepo
     options = {}
 
     try:
         args = fancyopts.fancyopts(args, commands.globalopts, options)
     except fancyopts.getopt.GetoptError:
         return
 
     if not args:
         return
 
     cmdtable = commands.table.copy()
     addaliases(lui, cmdtable)
 
     cmd = args[0]
     try:
         aliases, entry = cmdutil.findcmd(cmd, cmdtable, lui.config("ui", "strict"))
     except (error.AmbiguousCommand, error.UnknownCommand):
         commands.norepo = norepo
         return
 
     cmd = aliases[0]
     fn = entry[0]
 
     if cmd and util.safehasattr(fn, 'shell'):
         d = lambda: fn(ui, *args[1:])
         return lambda: runcommand(lui, None, cmd, args[:1], ui, options, d, [], {})
 
     commands.norepo = norepo
 
 _loaded = set()
 def _dispatch(req):
     args = req.args
     ui = req.ui
 
     # read --config before doing anything else
     # (e.g. to change trust settings for reading .hg/hgrc)
     cfgs = _parseconfig(ui, _earlygetopt(['--config'], args))
 
     # check for cwd
     cwd = _earlygetopt(['--cwd'], args)
     if cwd:
         os.chdir(cwd[-1])
 
     rpath = _earlygetopt(["-R", "--repository", "--repo"], args)
     path, lui = _getlocal(ui, rpath)
 
     # Now that we're operating in the right directory/repository with
     # the right config settings, check for shell aliases
     shellaliasfn = _checkshellalias(lui, ui, args)
     if shellaliasfn:
         return shellaliasfn()
 
     # Configure extensions in phases: uisetup, extsetup, cmdtable, and
     # reposetup. Programs like TortoiseHg will call _dispatch several
     # times so we keep track of configured extensions in _loaded.
     extensions.loadall(lui)
     exts = [ext for ext in extensions.extensions() if ext[0] not in _loaded]
     # Propagate any changes to lui.__class__ by extensions
     ui.__class__ = lui.__class__
 
     # (uisetup and extsetup are handled in extensions.loadall)
 
     for name, module in exts:
         cmdtable = getattr(module, 'cmdtable', {})
         overrides = [cmd for cmd in cmdtable if cmd in commands.table]
         if overrides:
             ui.warn(_("extension '%s' overrides commands: %s\n")
                     % (name, " ".join(overrides)))
         commands.table.update(cmdtable)
         _loaded.add(name)
 
     # (reposetup is handled in hg.repository)
 
     addaliases(lui, commands.table)
 
     # check for fallback encoding
     fallback = lui.config('ui', 'fallbackencoding')
     if fallback:
         encoding.fallbackencoding = fallback
 
     fullargs = args
     cmd, func, args, options, cmdoptions = _parse(lui, args)
 
     if options["config"]:
         raise util.Abort(_("option --config may not be abbreviated!"))
     if options["cwd"]:
         raise util.Abort(_("option --cwd may not be abbreviated!"))
     if options["repository"]:
         raise util.Abort(_(
             "Option -R has to be separated from other options (e.g. not -qR) "
             "and --repository may only be abbreviated as --repo!"))
 
     if options["encoding"]:
         encoding.encoding = options["encoding"]
     if options["encodingmode"]:
         encoding.encodingmode = options["encodingmode"]
     if options["time"]:
         def get_times():
             t = os.times()
             if t[4] == 0.0: # Windows leaves this as zero, so use time.clock()
                 t = (t[0], t[1], t[2], t[3], time.clock())
             return t
         s = get_times()
         def print_time():
             t = get_times()
             ui.warn(_("Time: real %.3f secs (user %.3f+%.3f sys %.3f+%.3f)\n") %
                 (t[4]-s[4], t[0]-s[0], t[2]-s[2], t[1]-s[1], t[3]-s[3]))
         atexit.register(print_time)
 
     uis = set([ui, lui])
 
     if req.repo:
         uis.add(req.repo.ui)
 
         # copy configs that were passed on the cmdline (--config) to the repo ui
         for cfg in cfgs:
             req.repo.ui.setconfig(*cfg)
 
     if options['verbose'] or options['debug'] or options['quiet']:
         for opt in ('verbose', 'debug', 'quiet'):
             val = str(bool(options[opt]))
             for ui_ in uis:
                 ui_.setconfig('ui', opt, val)
 
     if options['traceback']:
         for ui_ in uis:
             ui_.setconfig('ui', 'traceback', 'on')
 
     if options['noninteractive']:
         for ui_ in uis:
             ui_.setconfig('ui', 'interactive', 'off')
 
     if cmdoptions.get('insecure', False):
         for ui_ in uis:
             ui_.setconfig('web', 'cacerts', '')
 
     if options['version']:
         return commands.version_(ui)
     if options['help']:
         return commands.help_(ui, cmd)
     elif not cmd:
         return commands.help_(ui, 'shortlist')
 
     repo = None
     cmdpats = args[:]
     if cmd not in commands.norepo.split():
         # use the repo from the request only if we don't have -R
         if not rpath and not cwd:
             repo = req.repo
 
         if repo:
             # set the descriptors of the repo ui to those of ui
             repo.ui.fin = ui.fin
             repo.ui.fout = ui.fout
             repo.ui.ferr = ui.ferr
         else:
             try:
                 repo = hg.repository(ui, path=path)
                 if not repo.local():
                     raise util.Abort(_("repository '%s' is not local") % path)
                 repo.ui.setconfig("bundle", "mainreporoot", repo.root)
             except error.RequirementError:
                 raise
             except error.RepoError:
                 if cmd not in commands.optionalrepo.split():
                     if args and not path: # try to infer -R from command args
                         repos = map(cmdutil.findrepo, args)
                         guess = repos[0]
                         if guess and repos.count(guess) == len(repos):
                             req.args = ['--repository', guess] + fullargs
                             return _dispatch(req)
                     if not path:
                         raise error.RepoError(_("no repository found in '%s'"
                                                 " (.hg not found)") % os.getcwd())
                     raise
         if repo:
             ui = repo.ui
         args.insert(0, repo)
     elif rpath:
         ui.warn(_("warning: --repository ignored\n"))
 
     msg = ' '.join(' ' in a and repr(a) or a for a in fullargs)
     ui.log("command", msg + "\n")
     d = lambda: util.checksignature(func)(ui, *args, **cmdoptions)
     try:
         return runcommand(lui, repo, cmd, fullargs, ui, options, d,
                           cmdpats, cmdoptions)
     finally:
         if repo and repo != req.repo:
             repo.close()
 
 def _runcommand(ui, options, cmd, cmdfunc):
     def checkargs():
         try:
             return cmdfunc()
         except error.SignatureError:
             raise error.CommandError(cmd, _("invalid arguments"))
 
     if options['profile']:
         format = ui.config('profiling', 'format', default='text')
 
         if not format in ['text', 'kcachegrind']:
             ui.warn(_("unrecognized profiling format '%s'"
                         " - Ignored\n") % format)
             format = 'text'
 
         output = ui.config('profiling', 'output')
 
         if output:
             path = ui.expandpath(output)
             ostream = open(path, 'wb')
         else:
             ostream = sys.stderr
 
         try:
             from mercurial import lsprof
         except ImportError:
             raise util.Abort(_(
                 'lsprof not available - install from '
                 'http://codespeak.net/svn/user/arigo/hack/misc/lsprof/'))
         p = lsprof.Profiler()
         p.enable(subcalls=True)
         try:
             return checkargs()
         finally:
             p.disable()
 
             if format == 'kcachegrind':
                 import lsprofcalltree
                 calltree = lsprofcalltree.KCacheGrind(p)
                 calltree.output(ostream)
             else:
                 # format == 'text'
                 stats = lsprof.Stats(p.getstats())
                 stats.sort()
                 stats.pprint(top=10, file=ostream, climit=5)
 
             if output:
                 ostream.close()
     else:
         return checkargs()

mercurial/httpconnection.py

@@ -1,282 +1,282 b''
 # httpconnection.py - urllib2 handler for new http support
 #
 # Copyright 2005, 2006, 2007, 2008 Matt Mackall <[e-mail]>
 # Copyright 2006, 2007 Alexis S. L. Carvalho <[e-mail]>
 # Copyright 2006 Vadim Gelfer <[e-mail]>
 # Copyright 2011 Google, Inc.
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 import logging
 import socket
 import urllib
 import urllib2
 import os
 
 from mercurial import httpclient
 from mercurial import sslutil
 from mercurial import util
 from mercurial.i18n import _
 
 # moved here from url.py to avoid a cycle
 class httpsendfile(object):
     """This is a wrapper around the objects returned by python's "open".
 
     Its purpose is to send file-like objects via HTTP and, to do so, it
     defines a __len__ attribute to feed the Content-Length header.
     """
 
     def __init__(self, ui, *args, **kwargs):
         # We can't just "self._data = open(*args, **kwargs)" here because there
         # is an "open" function defined in this module that shadows the global
         # one
         self.ui = ui
         self._data = open(*args, **kwargs)
         self.seek = self._data.seek
         self.close = self._data.close
         self.write = self._data.write
         self._len = os.fstat(self._data.fileno()).st_size
         self._pos = 0
         self._total = self._len / 1024 * 2
 
     def read(self, *args, **kwargs):
         try:
             ret = self._data.read(*args, **kwargs)
         except EOFError:
             self.ui.progress(_('sending'), None)
         self._pos += len(ret)
         # We pass double the max for total because we currently have
         # to send the bundle twice in the case of a server that
         # requires authentication. Since we can't know until we try
         # once whether authentication will be required, just lie to
         # the user and maybe the push succeeds suddenly at 50%.
         self.ui.progress(_('sending'), self._pos / 1024,
                          unit=_('kb'), total=self._total)
         return ret
 
     def __len__(self):
         return self._len
 
 # moved here from url.py to avoid a cycle
-def readauthforuri(ui, uri):
+def readauthforuri(ui, uri, user):
     # Read configuration
     config = dict()
     for key, val in ui.configitems('auth'):
         if '.' not in key:
             ui.warn(_("ignoring invalid [auth] key '%s'\n") % key)
             continue
         group, setting = key.rsplit('.', 1)
         gdict = config.setdefault(group, dict())
         if setting in ('username', 'cert', 'key'):
             val = util.expandpath(val)
         gdict[setting] = val
 
     # Find the best match
-    uri = util.url(uri)
-    user = uri.user
-    uri.user = uri.password = None
-    uri = str(uri)
     scheme, hostpath = uri.split('://', 1)
     bestuser = None
     bestlen = 0
     bestauth = None
     for group, auth in config.iteritems():
         if user and user != auth.get('username', user):
             # If a username was set in the URI, the entry username
             # must either match it or be unset
             continue
         prefix = auth.get('prefix')
         if not prefix:
             continue
         p = prefix.split('://', 1)
         if len(p) > 1:
             schemes, prefix = [p[0]], p[1]
         else:
             schemes = (auth.get('schemes') or 'https').split()
         if (prefix == '*' or hostpath.startswith(prefix)) and \
             (len(prefix) > bestlen or (len(prefix) == bestlen and \
                 not bestuser and 'username' in auth)) \
              and scheme in schemes:
             bestlen = len(prefix)
             bestauth = group, auth
             bestuser = auth.get('username')
             if user and not bestuser:
                 auth['username'] = user
     return bestauth
 
 # Mercurial (at least until we can remove the old codepath) requires
 # that the http response object be sufficiently file-like, so we
 # provide a close() method here.
 class HTTPResponse(httpclient.HTTPResponse):
     def close(self):
         pass
 
 class HTTPConnection(httpclient.HTTPConnection):
     response_class = HTTPResponse
     def request(self, method, uri, body=None, headers={}):
         if isinstance(body, httpsendfile):
             body.seek(0)
         httpclient.HTTPConnection.request(self, method, uri, body=body,
                                           headers=headers)
 
 
 _configuredlogging = False
 LOGFMT = '%(levelname)s:%(name)s:%(lineno)d:%(message)s'
 # Subclass BOTH of these because otherwise urllib2 "helpfully"
 # reinserts them since it notices we don't include any subclasses of
 # them.
 class http2handler(urllib2.HTTPHandler, urllib2.HTTPSHandler):
     def __init__(self, ui, pwmgr):
         global _configuredlogging
         urllib2.AbstractHTTPHandler.__init__(self)
         self.ui = ui
         self.pwmgr = pwmgr
         self._connections = {}
         loglevel = ui.config('ui', 'http2debuglevel', default=None)
         if loglevel and not _configuredlogging:
             _configuredlogging = True
             logger = logging.getLogger('mercurial.httpclient')
             logger.setLevel(getattr(logging, loglevel.upper()))
             handler = logging.StreamHandler()
             handler.setFormatter(logging.Formatter(LOGFMT))
             logger.addHandler(handler)
 
     def close_all(self):
         """Close and remove all connection objects being kept for reuse."""
         for openconns in self._connections.values():
             for conn in openconns:
                 conn.close()
         self._connections = {}
 
     # shamelessly borrowed from urllib2.AbstractHTTPHandler
     def do_open(self, http_class, req, use_ssl):
         """Return an addinfourl object for the request, using http_class.
 
         http_class must implement the HTTPConnection API from httplib.
         The addinfourl return value is a file-like object.  It also
         has methods and attributes including:
             - info(): return a mimetools.Message object for the headers
             - geturl(): return the original request URL
             - code: HTTP status code
         """
         # If using a proxy, the host returned by get_host() is
         # actually the proxy. On Python 2.6.1, the real destination
         # hostname is encoded in the URI in the urllib2 request
         # object. On Python 2.6.5, it's stored in the _tunnel_host
         # attribute which has no accessor.
         tunhost = getattr(req, '_tunnel_host', None)
         host = req.get_host()
         if tunhost:
             proxyhost = host
             host = tunhost
         elif req.has_proxy():
             proxyhost = req.get_host()
             host = req.get_selector().split('://', 1)[1].split('/', 1)[0]
         else:
             proxyhost = None
 
         if proxyhost:
             if ':' in proxyhost:
                 # Note: this means we'll explode if we try and use an
                 # IPv6 http proxy. This isn't a regression, so we
                 # won't worry about it for now.
                 proxyhost, proxyport = proxyhost.rsplit(':', 1)
             else:
                 proxyport = 3128 # squid default
             proxy = (proxyhost, proxyport)
         else:
             proxy = None
 
         if not host:
             raise urllib2.URLError('no host given')
 
         connkey = use_ssl, host, proxy
         allconns = self._connections.get(connkey, [])
         conns = [c for c in allconns if not c.busy()]
         if conns:
             h = conns[0]
         else:
             if allconns:
                 self.ui.debug('all connections for %s busy, making a new '
                               'one\n' % host)
             timeout = None
             if req.timeout is not socket._GLOBAL_DEFAULT_TIMEOUT:
                 timeout = req.timeout
             h = http_class(host, timeout=timeout, proxy_hostport=proxy)
             self._connections.setdefault(connkey, []).append(h)
 
         headers = dict(req.headers)
         headers.update(req.unredirected_hdrs)
         headers = dict(
             (name.title(), val) for name, val in headers.items())
         try:
             path = req.get_selector()
             if '://' in path:
                 path = path.split('://', 1)[1].split('/', 1)[1]
             if path[0] != '/':
                 path = '/' + path
             h.request(req.get_method(), path, req.data, headers)
             r = h.getresponse()
         except socket.error, err: # XXX what error?
             raise urllib2.URLError(err)
 
         # Pick apart the HTTPResponse object to get the addinfourl
         # object initialized properly.
         r.recv = r.read
 
         resp = urllib.addinfourl(r, r.headers, req.get_full_url())
         resp.code = r.status
         resp.msg = r.reason
         return resp
 
     # httplib always uses the given host/port as the socket connect
     # target, and then allows full URIs in the request path, which it
     # then observes and treats as a signal to do proxying instead.
     def http_open(self, req):
         if req.get_full_url().startswith('https'):
             return self.https_open(req)
         return self.do_open(HTTPConnection, req, False)
 
     def https_open(self, req):
-        res = readauthforuri(self.ui, req.get_full_url())
+        # req.get_full_url() does not contain credentials and we may
+        # need them to match the certificates.
+        url = req.get_full_url()
+        user, password = self.pwmgr.find_stored_password(url)
+        res = readauthforuri(self.ui, url, user)
         if res:
             group, auth = res
             self.auth = auth
             self.ui.debug("using auth.%s.* for authentication\n" % group)
         else:
             self.auth = None
         return self.do_open(self._makesslconnection, req, True)
 
     def _makesslconnection(self, host, port=443, *args, **kwargs):
         keyfile = None
         certfile = None
 
         if args: # key_file
             keyfile = args.pop(0)
         if args: # cert_file
             certfile = args.pop(0)
 
         # if the user has specified different key/cert files in
         # hgrc, we prefer these
         if self.auth and 'key' in self.auth and 'cert' in self.auth:
             keyfile = self.auth['key']
             certfile = self.auth['cert']
 
         # let host port take precedence
         if ':' in host and '[' not in host or ']:' in host:
             host, port = host.rsplit(':', 1)
             port = int(port)
             if '[' in host:
                 host = host[1:-1]
 
         if keyfile:
             kwargs['keyfile'] = keyfile
         if certfile:
             kwargs['certfile'] = certfile
 
         kwargs.update(sslutil.sslkwargs(self.ui, host))
 
         con = HTTPConnection(host, port, use_ssl=True,
                              ssl_validator=sslutil.validator(self.ui, host),
                              **kwargs)
         return con

mercurial/url.py

@@ -1,463 +1,471 b''
 # url.py - HTTP handling for mercurial
 #
 # Copyright 2005, 2006, 2007, 2008 Matt Mackall <[e-mail]>
 # Copyright 2006, 2007 Alexis S. L. Carvalho <[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.
 
 import urllib, urllib2, httplib, os, socket, cStringIO
 from i18n import _
 import keepalive, util, sslutil
 import httpconnection as httpconnectionmod
 
 class passwordmgr(urllib2.HTTPPasswordMgrWithDefaultRealm):
     def __init__(self, ui):
         urllib2.HTTPPasswordMgrWithDefaultRealm.__init__(self)
         self.ui = ui
 
     def find_user_password(self, realm, authuri):
         authinfo = urllib2.HTTPPasswordMgrWithDefaultRealm.find_user_password(
             self, realm, authuri)
         user, passwd = authinfo
         if user and passwd:
             self._writedebug(user, passwd)
             return (user, passwd)
 
         if not user or not passwd:
-            res = httpconnectionmod.readauthforuri(self.ui, authuri)
+            res = httpconnectionmod.readauthforuri(self.ui, authuri, user)
             if res:
                 group, auth = res
                 user, passwd = auth.get('username'), auth.get('password')
                 self.ui.debug("using auth.%s.* for authentication\n" % group)
         if not user or not passwd:
             if not self.ui.interactive():
                 raise util.Abort(_('http authorization required'))
 
             self.ui.write(_("http authorization required\n"))
             self.ui.write(_("realm: %s\n") % realm)
             if user:
                 self.ui.write(_("user: %s\n") % user)
             else:
                 user = self.ui.prompt(_("user:"), default=None)
 
             if not passwd:
                 passwd = self.ui.getpass()
 
         self.add_password(realm, authuri, user, passwd)
         self._writedebug(user, passwd)
         return (user, passwd)
 
     def _writedebug(self, user, passwd):
         msg = _('http auth: user %s, password %s\n')
         self.ui.debug(msg % (user, passwd and '*' * len(passwd) or 'not set'))
 
+    def find_stored_password(self, authuri):
+        return urllib2.HTTPPasswordMgrWithDefaultRealm.find_user_password(
+            self, None, authuri)
+
 class proxyhandler(urllib2.ProxyHandler):
     def __init__(self, ui):
         proxyurl = ui.config("http_proxy", "host") or os.getenv('http_proxy')
         # XXX proxyauthinfo = None
 
         if proxyurl:
             # proxy can be proper url or host[:port]
             if not (proxyurl.startswith('http:') or
                     proxyurl.startswith('https:')):
                 proxyurl = 'http://' + proxyurl + '/'
             proxy = util.url(proxyurl)
             if not proxy.user:
                 proxy.user = ui.config("http_proxy", "user")
                 proxy.passwd = ui.config("http_proxy", "passwd")
 
             # see if we should use a proxy for this url
             no_list = ["localhost", "127.0.0.1"]
             no_list.extend([p.lower() for
                             p in ui.configlist("http_proxy", "no")])
             no_list.extend([p.strip().lower() for
                             p in os.getenv("no_proxy", '').split(',')
                             if p.strip()])
             # "http_proxy.always" config is for running tests on localhost
             if ui.configbool("http_proxy", "always"):
                 self.no_list = []
             else:
                 self.no_list = no_list
 
             proxyurl = str(proxy)
             proxies = {'http': proxyurl, 'https': proxyurl}
             ui.debug('proxying through http://%s:%s\n' %
                       (proxy.host, proxy.port))
         else:
             proxies = {}
 
         # urllib2 takes proxy values from the environment and those
         # will take precedence if found, so drop them
         for env in ["HTTP_PROXY", "http_proxy", "no_proxy"]:
             try:
                 if env in os.environ:
                     del os.environ[env]
             except OSError:
                 pass
 
         urllib2.ProxyHandler.__init__(self, proxies)
         self.ui = ui
 
     def proxy_open(self, req, proxy, type_):
         host = req.get_host().split(':')[0]
         if host in self.no_list:
             return None
 
         # work around a bug in Python < 2.4.2
         # (it leaves a "\n" at the end of Proxy-authorization headers)
         baseclass = req.__class__
         class _request(baseclass):
             def add_header(self, key, val):
                 if key.lower() == 'proxy-authorization':
                     val = val.strip()
                 return baseclass.add_header(self, key, val)
         req.__class__ = _request
 
         return urllib2.ProxyHandler.proxy_open(self, req, proxy, type_)
 
 def _gen_sendfile(orgsend):
     def _sendfile(self, data):
         # send a file
         if isinstance(data, httpconnectionmod.httpsendfile):
             # if auth required, some data sent twice, so rewind here
             data.seek(0)
             for chunk in util.filechunkiter(data):
                 orgsend(self, chunk)
         else:
             orgsend(self, data)
     return _sendfile
 
 has_https = util.safehasattr(urllib2, 'HTTPSHandler')
 if has_https:
     try:
         _create_connection = socket.create_connection
     except AttributeError:
         _GLOBAL_DEFAULT_TIMEOUT = object()
 
         def _create_connection(address, timeout=_GLOBAL_DEFAULT_TIMEOUT,
                                source_address=None):
             # lifted from Python 2.6
 
             msg = "getaddrinfo returns an empty list"
             host, port = address
             for res in socket.getaddrinfo(host, port, 0, socket.SOCK_STREAM):
                 af, socktype, proto, canonname, sa = res
                 sock = None
                 try:
                     sock = socket.socket(af, socktype, proto)
                     if timeout is not _GLOBAL_DEFAULT_TIMEOUT:
                         sock.settimeout(timeout)
                     if source_address:
                         sock.bind(source_address)
                     sock.connect(sa)
                     return sock
 
                 except socket.error, msg:
                     if sock is not None:
                         sock.close()
 
             raise socket.error, msg
 
 class httpconnection(keepalive.HTTPConnection):
     # must be able to send big bundle as stream.
     send = _gen_sendfile(keepalive.HTTPConnection.send)
 
     def connect(self):
         if has_https and self.realhostport: # use CONNECT proxy
             self.sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
             self.sock.connect((self.host, self.port))
             if _generic_proxytunnel(self):
                 # we do not support client x509 certificates
                 self.sock = sslutil.ssl_wrap_socket(self.sock, None, None)
         else:
             keepalive.HTTPConnection.connect(self)
 
     def getresponse(self):
         proxyres = getattr(self, 'proxyres', None)
         if proxyres:
             if proxyres.will_close:
                 self.close()
             self.proxyres = None
             return proxyres
         return keepalive.HTTPConnection.getresponse(self)
 
 # general transaction handler to support different ways to handle
 # HTTPS proxying before and after Python 2.6.3.
 def _generic_start_transaction(handler, h, req):
     tunnel_host = getattr(req, '_tunnel_host', None)
     if tunnel_host:
         if tunnel_host[:7] not in ['http://', 'https:/']:
             tunnel_host = 'https://' + tunnel_host
         new_tunnel = True
     else:
         tunnel_host = req.get_selector()
         new_tunnel = False
 
     if new_tunnel or tunnel_host == req.get_full_url(): # has proxy
         u = util.url(tunnel_host)
         if new_tunnel or u.scheme == 'https': # only use CONNECT for HTTPS
             h.realhostport = ':'.join([u.host, (u.port or '443')])
             h.headers = req.headers.copy()
             h.headers.update(handler.parent.addheaders)
             return
 
     h.realhostport = None
     h.headers = None
 
 def _generic_proxytunnel(self):
     proxyheaders = dict(
             [(x, self.headers[x]) for x in self.headers
              if x.lower().startswith('proxy-')])
     self._set_hostport(self.host, self.port)
     self.send('CONNECT %s HTTP/1.0\r\n' % self.realhostport)
     for header in proxyheaders.iteritems():
         self.send('%s: %s\r\n' % header)
     self.send('\r\n')
 
     # majority of the following code is duplicated from
     # httplib.HTTPConnection as there are no adequate places to
     # override functions to provide the needed functionality
     res = self.response_class(self.sock,
                               strict=self.strict,
                               method=self._method)
 
     while True:
         version, status, reason = res._read_status()
         if status != httplib.CONTINUE:
             break
         while True:
             skip = res.fp.readline().strip()
             if not skip:
                 break
     res.status = status
     res.reason = reason.strip()
 
     if res.status == 200:
         while True:
             line = res.fp.readline()
             if line == '\r\n':
                 break
         return True
 
     if version == 'HTTP/1.0':
         res.version = 10
     elif version.startswith('HTTP/1.'):
         res.version = 11
     elif version == 'HTTP/0.9':
         res.version = 9
     else:
         raise httplib.UnknownProtocol(version)
 
     if res.version == 9:
         res.length = None
         res.chunked = 0
         res.will_close = 1
         res.msg = httplib.HTTPMessage(cStringIO.StringIO())
         return False
 
     res.msg = httplib.HTTPMessage(res.fp)
     res.msg.fp = None
 
     # are we using the chunked-style of transfer encoding?
     trenc = res.msg.getheader('transfer-encoding')
     if trenc and trenc.lower() == "chunked":
         res.chunked = 1
         res.chunk_left = None
     else:
         res.chunked = 0
 
     # will the connection close at the end of the response?
     res.will_close = res._check_close()
 
     # do we have a Content-Length?
     # NOTE: RFC 2616, S4.4, #3 says we ignore this if tr_enc is "chunked"
     length = res.msg.getheader('content-length')
     if length and not res.chunked:
         try:
             res.length = int(length)
         except ValueError:
             res.length = None
         else:
             if res.length < 0:  # ignore nonsensical negative lengths
                 res.length = None
     else:
         res.length = None
 
     # does the body have a fixed length? (of zero)
     if (status == httplib.NO_CONTENT or status == httplib.NOT_MODIFIED or
         100 <= status < 200 or # 1xx codes
         res._method == 'HEAD'):
         res.length = 0
 
     # if the connection remains open, and we aren't using chunked, and
     # a content-length was not provided, then assume that the connection
     # WILL close.
     if (not res.will_close and
        not res.chunked and
        res.length is None):
         res.will_close = 1
 
     self.proxyres = res
 
     return False
 
 class httphandler(keepalive.HTTPHandler):
     def http_open(self, req):
         return self.do_open(httpconnection, req)
 
     def _start_transaction(self, h, req):
         _generic_start_transaction(self, h, req)
         return keepalive.HTTPHandler._start_transaction(self, h, req)
 
 if has_https:
     class httpsconnection(httplib.HTTPSConnection):
         response_class = keepalive.HTTPResponse
         # must be able to send big bundle as stream.
         send = _gen_sendfile(keepalive.safesend)
         getresponse = keepalive.wrapgetresponse(httplib.HTTPSConnection)
 
         def connect(self):
             self.sock = _create_connection((self.host, self.port))
 
             host = self.host
             if self.realhostport: # use CONNECT proxy
                 _generic_proxytunnel(self)
                 host = self.realhostport.rsplit(':', 1)[0]
             self.sock = sslutil.ssl_wrap_socket(
                 self.sock, self.key_file, self.cert_file,
                 **sslutil.sslkwargs(self.ui, host))
             sslutil.validator(self.ui, host)(self.sock)
 
     class httpshandler(keepalive.KeepAliveHandler, urllib2.HTTPSHandler):
         def __init__(self, ui):
             keepalive.KeepAliveHandler.__init__(self)
             urllib2.HTTPSHandler.__init__(self)
             self.ui = ui
             self.pwmgr = passwordmgr(self.ui)
 
         def _start_transaction(self, h, req):
             _generic_start_transaction(self, h, req)
             return keepalive.KeepAliveHandler._start_transaction(self, h, req)
 
         def https_open(self, req):
-            res = httpconnectionmod.readauthforuri(self.ui, req.get_full_url())
+            # req.get_full_url() does not contain credentials and we may
+            # need them to match the certificates.
+            url = req.get_full_url()
+            user, password = self.pwmgr.find_stored_password(url)
+            res = httpconnectionmod.readauthforuri(self.ui, url, user)
             if res:
                 group, auth = res
                 self.auth = auth
                 self.ui.debug("using auth.%s.* for authentication\n" % group)
             else:
                 self.auth = None
             return self.do_open(self._makeconnection, req)
 
         def _makeconnection(self, host, port=None, *args, **kwargs):
             keyfile = None
             certfile = None
 
             if len(args) >= 1: # key_file
                 keyfile = args[0]
             if len(args) >= 2: # cert_file
                 certfile = args[1]
             args = args[2:]
 
             # if the user has specified different key/cert files in
             # hgrc, we prefer these
             if self.auth and 'key' in self.auth and 'cert' in self.auth:
                 keyfile = self.auth['key']
                 certfile = self.auth['cert']
 
             conn = httpsconnection(host, port, keyfile, certfile, *args, **kwargs)
             conn.ui = self.ui
             return conn
 
 class httpdigestauthhandler(urllib2.HTTPDigestAuthHandler):
     def __init__(self, *args, **kwargs):
         urllib2.HTTPDigestAuthHandler.__init__(self, *args, **kwargs)
         self.retried_req = None
 
     def reset_retry_count(self):
         # Python 2.6.5 will call this on 401 or 407 errors and thus loop
         # forever. We disable reset_retry_count completely and reset in
         # http_error_auth_reqed instead.
         pass
 
     def http_error_auth_reqed(self, auth_header, host, req, headers):
         # Reset the retry counter once for each request.
         if req is not self.retried_req:
             self.retried_req = req
             self.retried = 0
         # In python < 2.5 AbstractDigestAuthHandler raises a ValueError if
         # it doesn't know about the auth type requested. This can happen if
         # somebody is using BasicAuth and types a bad password.
         try:
             return urllib2.HTTPDigestAuthHandler.http_error_auth_reqed(
                         self, auth_header, host, req, headers)
         except ValueError, inst:
             arg = inst.args[0]
             if arg.startswith("AbstractDigestAuthHandler doesn't know "):
                 return
             raise
 
 class httpbasicauthhandler(urllib2.HTTPBasicAuthHandler):
     def __init__(self, *args, **kwargs):
         urllib2.HTTPBasicAuthHandler.__init__(self, *args, **kwargs)
         self.retried_req = None
 
     def reset_retry_count(self):
         # Python 2.6.5 will call this on 401 or 407 errors and thus loop
         # forever. We disable reset_retry_count completely and reset in
         # http_error_auth_reqed instead.
         pass
 
     def http_error_auth_reqed(self, auth_header, host, req, headers):
         # Reset the retry counter once for each request.
         if req is not self.retried_req:
             self.retried_req = req
             self.retried = 0
         return urllib2.HTTPBasicAuthHandler.http_error_auth_reqed(
                         self, auth_header, host, req, headers)
 
 handlerfuncs = []
 
 def opener(ui, authinfo=None):
     '''
     construct an opener suitable for urllib2
     authinfo will be added to the password manager
     '''
     if ui.configbool('ui', 'usehttp2', False):
         handlers = [httpconnectionmod.http2handler(ui, passwordmgr(ui))]
     else:
         handlers = [httphandler()]
         if has_https:
             handlers.append(httpshandler(ui))
 
     handlers.append(proxyhandler(ui))
 
     passmgr = passwordmgr(ui)
     if authinfo is not None:
         passmgr.add_password(*authinfo)
         user, passwd = authinfo[2:4]
         ui.debug('http auth: user %s, password %s\n' %
                  (user, passwd and '*' * len(passwd) or 'not set'))
 
     handlers.extend((httpbasicauthhandler(passmgr),
                      httpdigestauthhandler(passmgr)))
     handlers.extend([h(ui, passmgr) for h in handlerfuncs])
     opener = urllib2.build_opener(*handlers)
 
     # 1.0 here is the _protocol_ version
     opener.addheaders = [('User-agent', 'mercurial/proto-1.0')]
     opener.addheaders.append(('Accept', 'application/mercurial-0.1'))
     return opener
 
 def open(ui, url_, data=None):
     u = util.url(url_)
     if u.scheme:
         u.scheme = u.scheme.lower()
         url_, authinfo = u.authinfo()
     else:
         path = util.normpath(os.path.abspath(url_))
         url_ = 'file://' + urllib.pathname2url(path)
         authinfo = None
     return opener(ui, authinfo).open(url_, data)

mercurial/util.py

@@ -1,1647 +1,1646 b''
 # util.py - Mercurial utility functions and platform specfic implementations
 #
 #  Copyright 2005 K. Thananchayan <[e-mail]>
 #  Copyright 2005-2007 Matt Mackall <[e-mail]>
 #  Copyright 2006 Vadim Gelfer <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 """Mercurial utility functions and platform specfic implementations.
 
 This contains helper routines that are independent of the SCM core and
 hide platform-specific details from the core.
 """
 
 from i18n import _
 import error, osutil, encoding
 import errno, re, shutil, sys, tempfile, traceback
 import os, time, calendar, textwrap, unicodedata, signal
 import imp, socket, urllib
 
 if os.name == 'nt':
     import windows as platform
 else:
     import posix as platform
 
 cachestat = platform.cachestat
 checkexec = platform.checkexec
 checklink = platform.checklink
 copymode = platform.copymode
 executablepath = platform.executablepath
 expandglobs = platform.expandglobs
 explainexit = platform.explainexit
 findexe = platform.findexe
 gethgcmd = platform.gethgcmd
 getuser = platform.getuser
 groupmembers = platform.groupmembers
 groupname = platform.groupname
 hidewindow = platform.hidewindow
 isexec = platform.isexec
 isowner = platform.isowner
 localpath = platform.localpath
 lookupreg = platform.lookupreg
 makedir = platform.makedir
 nlinks = platform.nlinks
 normpath = platform.normpath
 nulldev = platform.nulldev
 openhardlinks = platform.openhardlinks
 oslink = platform.oslink
 parsepatchoutput = platform.parsepatchoutput
 pconvert = platform.pconvert
 popen = platform.popen
 posixfile = platform.posixfile
 quotecommand = platform.quotecommand
 realpath = platform.realpath
 rename = platform.rename
 samedevice = platform.samedevice
 samefile = platform.samefile
 samestat = platform.samestat
 setbinary = platform.setbinary
 setflags = platform.setflags
 setsignalhandler = platform.setsignalhandler
 shellquote = platform.shellquote
 spawndetached = platform.spawndetached
 sshargs = platform.sshargs
 statfiles = platform.statfiles
 termwidth = platform.termwidth
 testpid = platform.testpid
 umask = platform.umask
 unlink = platform.unlink
 unlinkpath = platform.unlinkpath
 username = platform.username
 
 # Python compatibility
 
 def sha1(s):
     return _fastsha1(s)
 
 _notset = object()
 def safehasattr(thing, attr):
     return getattr(thing, attr, _notset) is not _notset
 
 def _fastsha1(s):
     # This function will import sha1 from hashlib or sha (whichever is
     # available) and overwrite itself with it on the first call.
     # Subsequent calls will go directly to the imported function.
     if sys.version_info >= (2, 5):
         from hashlib import sha1 as _sha1
     else:
         from sha import sha as _sha1
     global _fastsha1, sha1
     _fastsha1 = sha1 = _sha1
     return _sha1(s)
 
 import __builtin__
 
 if sys.version_info[0] < 3:
     def fakebuffer(sliceable, offset=0):
         return sliceable[offset:]
 else:
     def fakebuffer(sliceable, offset=0):
         return memoryview(sliceable)[offset:]
 try:
     buffer
 except NameError:
     __builtin__.buffer = fakebuffer
 
 import subprocess
 closefds = os.name == 'posix'
 
 def popen2(cmd, env=None, newlines=False):
     # 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(cmd, shell=True, bufsize=-1,
                          close_fds=closefds,
                          stdin=subprocess.PIPE, stdout=subprocess.PIPE,
                          universal_newlines=newlines,
                          env=env)
     return p.stdin, p.stdout
 
 def popen3(cmd, env=None, newlines=False):
     p = subprocess.Popen(cmd, shell=True, bufsize=-1,
                          close_fds=closefds,
                          stdin=subprocess.PIPE, stdout=subprocess.PIPE,
                          stderr=subprocess.PIPE,
                          universal_newlines=newlines,
                          env=env)
     return p.stdin, p.stdout, p.stderr
 
 def version():
     """Return version information if available."""
     try:
         import __version__
         return __version__.version
     except ImportError:
         return 'unknown'
 
 # used by parsedate
 defaultdateformats = (
     '%Y-%m-%d %H:%M:%S',
     '%Y-%m-%d %I:%M:%S%p',
     '%Y-%m-%d %H:%M',
     '%Y-%m-%d %I:%M%p',
     '%Y-%m-%d',
     '%m-%d',
     '%m/%d',
     '%m/%d/%y',
     '%m/%d/%Y',
     '%a %b %d %H:%M:%S %Y',
     '%a %b %d %I:%M:%S%p %Y',
     '%a, %d %b %Y %H:%M:%S',        #  GNU coreutils "/bin/date --rfc-2822"
     '%b %d %H:%M:%S %Y',
     '%b %d %I:%M:%S%p %Y',
     '%b %d %H:%M:%S',
     '%b %d %I:%M:%S%p',
     '%b %d %H:%M',
     '%b %d %I:%M%p',
     '%b %d %Y',
     '%b %d',
     '%H:%M:%S',
     '%I:%M:%S%p',
     '%H:%M',
     '%I:%M%p',
 )
 
 extendeddateformats = defaultdateformats + (
     "%Y",
     "%Y-%m",
     "%b",
     "%b %Y",
     )
 
 def cachefunc(func):
     '''cache the result of function calls'''
     # XXX doesn't handle keywords args
     cache = {}
     if func.func_code.co_argcount == 1:
         # we gain a small amount of time because
         # we don't need to pack/unpack the list
         def f(arg):
             if arg not in cache:
                 cache[arg] = func(arg)
             return cache[arg]
     else:
         def f(*args):
             if args not in cache:
                 cache[args] = func(*args)
             return cache[args]
 
     return f
 
 def lrucachefunc(func):
     '''cache most recent results of function calls'''
     cache = {}
     order = []
     if func.func_code.co_argcount == 1:
         def f(arg):
             if arg not in cache:
                 if len(cache) > 20:
                     del cache[order.pop(0)]
                 cache[arg] = func(arg)
             else:
                 order.remove(arg)
             order.append(arg)
             return cache[arg]
     else:
         def f(*args):
             if args not in cache:
                 if len(cache) > 20:
                     del cache[order.pop(0)]
                 cache[args] = func(*args)
             else:
                 order.remove(args)
             order.append(args)
             return cache[args]
 
     return f
 
 class propertycache(object):
     def __init__(self, func):
         self.func = func
         self.name = func.__name__
     def __get__(self, obj, type=None):
         result = self.func(obj)
         setattr(obj, self.name, result)
         return result
 
 def pipefilter(s, cmd):
     '''filter string S through command CMD, returning its output'''
     p = subprocess.Popen(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 = tempfile.mkstemp(prefix='hg-filter-in-')
         fp = os.fdopen(infd, 'wb')
         fp.write(s)
         fp.close()
         outfd, outname = tempfile.mkstemp(prefix='hg-filter-out-')
         os.close(outfd)
         cmd = cmd.replace('INFILE', inname)
         cmd = cmd.replace('OUTFILE', outname)
         code = os.system(cmd)
         if sys.platform == 'OpenVMS' and code & 1:
             code = 0
         if code:
             raise Abort(_("command '%s' failed: %s") %
                         (cmd, explainexit(code)))
         fp = open(outname, 'rb')
         r = fp.read()
         fp.close()
         return r
     finally:
         try:
             if inname:
                 os.unlink(inname)
         except OSError:
             pass
         try:
             if outname:
                 os.unlink(outname)
         except OSError:
             pass
 
 filtertable = {
     'tempfile:': tempfilter,
     'pipe:': pipefilter,
     }
 
 def filter(s, cmd):
     "filter a string through a command that transforms its input to its output"
     for name, fn in filtertable.iteritems():
         if cmd.startswith(name):
             return fn(s, cmd[len(name):].lstrip())
     return pipefilter(s, cmd)
 
 def binary(s):
     """return true if a string is binary data"""
     return bool(s and '\0' in s)
 
 def increasingchunks(source, min=1024, max=65536):
     '''return no less than min bytes per chunk while data remains,
     doubling min after each chunk until it reaches max'''
     def log2(x):
         if not x:
             return 0
         i = 0
         while x:
             x >>= 1
             i += 1
         return i - 1
 
     buf = []
     blen = 0
     for chunk in source:
         buf.append(chunk)
         blen += len(chunk)
         if blen >= min:
             if min < max:
                 min = min << 1
                 nmin = 1 << log2(blen)
                 if nmin > min:
                     min = nmin
                 if min > max:
                     min = max
             yield ''.join(buf)
             blen = 0
             buf = []
     if buf:
         yield ''.join(buf)
 
 Abort = error.Abort
 
 def always(fn):
     return True
 
 def never(fn):
     return False
 
 def pathto(root, n1, n2):
     '''return the relative path from one place to another.
     root should use os.sep to separate directories
     n1 should use os.sep to separate directories
     n2 should use "/" to separate directories
     returns an os.sep-separated path.
 
     If n1 is a relative path, it's assumed it's
     relative to root.
     n2 should always be relative to root.
     '''
     if not n1:
         return localpath(n2)
     if os.path.isabs(n1):
         if os.path.splitdrive(root)[0] != os.path.splitdrive(n1)[0]:
             return os.path.join(root, localpath(n2))
         n2 = '/'.join((pconvert(root), n2))
     a, b = splitpath(n1), n2.split('/')
     a.reverse()
     b.reverse()
     while a and b and a[-1] == b[-1]:
         a.pop()
         b.pop()
     b.reverse()
     return os.sep.join((['..'] * len(a)) + b) or '.'
 
 _hgexecutable = None
 
 def mainfrozen():
     """return True if we are a frozen executable.
 
     The code supports py2exe (most common, Windows only) and tools/freeze
     (portable, not much used).
     """
     return (safehasattr(sys, "frozen") or # new py2exe
             safehasattr(sys, "importers") or # old py2exe
             imp.is_frozen("__main__")) # tools/freeze
 
 def hgexecutable():
     """return location of the 'hg' executable.
 
     Defaults to $HG or 'hg' in the search path.
     """
     if _hgexecutable is None:
         hg = os.environ.get('HG')
         if hg:
             _sethgexecutable(hg)
         elif mainfrozen():
             _sethgexecutable(sys.executable)
         else:
             exe = findexe('hg') or os.path.basename(sys.argv[0])
             _sethgexecutable(exe)
     return _hgexecutable
 
 def _sethgexecutable(path):
     """set location of the 'hg' executable"""
     global _hgexecutable
     _hgexecutable = path
 
 def system(cmd, environ={}, cwd=None, onerr=None, errprefix=None, out=None):
     '''enhanced shell command execution.
     run with environment maybe modified, maybe in different dir.
 
     if command fails and onerr is None, return status.  if ui object,
     print error message and return status, else raise onerr object as
     exception.
 
     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:
         sys.stdout.flush()
     except Exception:
         pass
     def py2shell(val):
         'convert python object into string that is useful to shell'
         if val is None or val is False:
             return '0'
         if val is True:
             return '1'
         return str(val)
     origcmd = cmd
     cmd = quotecommand(cmd)
     env = dict(os.environ)
     env.update((k, py2shell(v)) for k, v in environ.iteritems())
     env['HG'] = hgexecutable()
     if out is None or out == sys.__stdout__:
         rc = subprocess.call(cmd, shell=True, close_fds=closefds,
                              env=env, cwd=cwd)
     else:
         proc = subprocess.Popen(cmd, shell=True, close_fds=closefds,
                                 env=env, cwd=cwd, stdout=subprocess.PIPE,
                                 stderr=subprocess.STDOUT)
         for line in proc.stdout:
             out.write(line)
         proc.wait()
         rc = proc.returncode
     if sys.platform == 'OpenVMS' and rc & 1:
         rc = 0
     if rc and onerr:
         errmsg = '%s %s' % (os.path.basename(origcmd.split(None, 1)[0]),
                             explainexit(rc)[0])
         if errprefix:
             errmsg = '%s: %s' % (errprefix, errmsg)
         try:
             onerr.warn(errmsg + '\n')
         except AttributeError:
             raise onerr(errmsg)
     return rc
 
 def checksignature(func):
     '''wrap a function with code to check for calling errors'''
     def check(*args, **kwargs):
         try:
             return func(*args, **kwargs)
         except TypeError:
             if len(traceback.extract_tb(sys.exc_info()[2])) == 1:
                 raise error.SignatureError
             raise
 
     return check
 
 def copyfile(src, dest):
     "copy a file, preserving mode and atime/mtime"
     if os.path.islink(src):
         try:
             os.unlink(dest)
         except OSError:
             pass
         os.symlink(os.readlink(src), dest)
     else:
         try:
             shutil.copyfile(src, dest)
             shutil.copymode(src, dest)
         except shutil.Error, inst:
             raise Abort(str(inst))
 
 def copyfiles(src, dst, hardlink=None):
     """Copy a directory tree using hardlinks if possible"""
 
     if hardlink is None:
         hardlink = (os.stat(src).st_dev ==
                     os.stat(os.path.dirname(dst)).st_dev)
 
     num = 0
     if os.path.isdir(src):
         os.mkdir(dst)
         for name, kind in osutil.listdir(src):
             srcname = os.path.join(src, name)
             dstname = os.path.join(dst, name)
             hardlink, n = copyfiles(srcname, dstname, hardlink)
             num += n
     else:
         if hardlink:
             try:
                 oslink(src, dst)
             except (IOError, OSError):
                 hardlink = False
                 shutil.copy(src, dst)
         else:
             shutil.copy(src, dst)
         num += 1
 
     return hardlink, num
 
 _winreservednames = '''con prn aux nul
     com1 com2 com3 com4 com5 com6 com7 com8 com9
     lpt1 lpt2 lpt3 lpt4 lpt5 lpt6 lpt7 lpt8 lpt9'''.split()
 _winreservedchars = ':*?"<>|'
 def checkwinfilename(path):
     '''Check that the base-relative path is a valid filename on Windows.
     Returns None if the path is ok, or a UI string describing the problem.
 
     >>> checkwinfilename("just/a/normal/path")
     >>> checkwinfilename("foo/bar/con.xml")
     "filename contains 'con', which is reserved on Windows"
     >>> checkwinfilename("foo/con.xml/bar")
     "filename contains 'con', which is reserved on Windows"
     >>> checkwinfilename("foo/bar/xml.con")
     >>> checkwinfilename("foo/bar/AUX/bla.txt")
     "filename contains 'AUX', which is reserved on Windows"
     >>> checkwinfilename("foo/bar/bla:.txt")
     "filename contains ':', which is reserved on Windows"
     >>> checkwinfilename("foo/bar/b\07la.txt")
     "filename contains '\\\\x07', which is invalid on Windows"
     >>> checkwinfilename("foo/bar/bla ")
     "filename ends with ' ', which is not allowed on Windows"
     '''
     for n in path.replace('\\', '/').split('/'):
         if not n:
             continue
         for c in n:
             if c in _winreservedchars:
                 return _("filename contains '%s', which is reserved "
                          "on Windows") % c
             if ord(c) <= 31:
                 return _("filename contains %r, which is invalid "
                          "on Windows") % c
         base = n.split('.')[0]
         if base and base.lower() in _winreservednames:
             return _("filename contains '%s', which is reserved "
                      "on Windows") % base
         t = n[-1]
         if t in '. ':
             return _("filename ends with '%s', which is not allowed "
                      "on Windows") % t
 
 if os.name == 'nt':
     checkosfilename = checkwinfilename
 else:
     checkosfilename = platform.checkosfilename
 
 def makelock(info, pathname):
     try:
         return os.symlink(info, pathname)
     except OSError, why:
         if why.errno == errno.EEXIST:
             raise
     except AttributeError: # no symlink in os
         pass
 
     ld = os.open(pathname, os.O_CREAT | os.O_WRONLY | os.O_EXCL)
     os.write(ld, info)
     os.close(ld)
 
 def readlock(pathname):
     try:
         return os.readlink(pathname)
     except OSError, why:
         if why.errno not in (errno.EINVAL, errno.ENOSYS):
             raise
     except AttributeError: # no symlink in os
         pass
     fp = posixfile(pathname)
     r = fp.read()
     fp.close()
     return r
 
 def fstat(fp):
     '''stat file object that may not have fileno method.'''
     try:
         return os.fstat(fp.fileno())
     except AttributeError:
         return os.stat(fp.name)
 
 # File system features
 
 def checkcase(path):
     """
     Check whether the given path is on a case-sensitive filesystem
 
     Requires a path (like /foo/.hg) ending with a foldable final
     directory component.
     """
     s1 = os.stat(path)
     d, b = os.path.split(path)
     p2 = os.path.join(d, b.upper())
     if path == p2:
         p2 = os.path.join(d, b.lower())
     try:
         s2 = os.stat(p2)
         if s2 == s1:
             return False
         return True
     except OSError:
         return True
 
 _fspathcache = {}
 def fspath(name, root):
     '''Get name in the case stored in the filesystem
 
     The name is either relative to root, or it is an absolute path starting
     with root. Note that this function is unnecessary, and should not be
     called, for case-sensitive filesystems (simply because it's expensive).
     '''
     # If name is absolute, make it relative
     if name.lower().startswith(root.lower()):
         l = len(root)
         if name[l] == os.sep or name[l] == os.altsep:
             l = l + 1
         name = name[l:]
 
     if not os.path.lexists(os.path.join(root, name)):
         return None
 
     seps = os.sep
     if os.altsep:
         seps = seps + os.altsep
     # Protect backslashes. This gets silly very quickly.
     seps.replace('\\','\\\\')
     pattern = re.compile(r'([^%s]+)|([%s]+)' % (seps, seps))
     dir = os.path.normcase(os.path.normpath(root))
     result = []
     for part, sep in pattern.findall(name):
         if sep:
             result.append(sep)
             continue
 
         if dir not in _fspathcache:
             _fspathcache[dir] = os.listdir(dir)
         contents = _fspathcache[dir]
 
         lpart = part.lower()
         lenp = len(part)
         for n in contents:
             if lenp == len(n) and n.lower() == lpart:
                 result.append(n)
                 break
         else:
             # Cannot happen, as the file exists!
             result.append(part)
         dir = os.path.join(dir, lpart)
 
     return ''.join(result)
 
 def checknlink(testfile):
     '''check whether hardlink count reporting works properly'''
 
     # testfile may be open, so we need a separate file for checking to
     # work around issue2543 (or testfile may get lost on Samba shares)
     f1 = testfile + ".hgtmp1"
     if os.path.lexists(f1):
         return False
     try:
         posixfile(f1, 'w').close()
     except IOError:
         return False
 
     f2 = testfile + ".hgtmp2"
     fd = None
     try:
         try:
             oslink(f1, f2)
         except OSError:
             return False
 
         # nlinks() may behave differently for files on Windows shares if
         # the file is open.
         fd = posixfile(f2)
         return nlinks(f2) > 1
     finally:
         if fd is not None:
             fd.close()
         for f in (f1, f2):
             try:
                 os.unlink(f)
             except OSError:
                 pass
 
     return False
 
 def endswithsep(path):
     '''Check path ends with os.sep or os.altsep.'''
     return path.endswith(os.sep) or os.altsep and path.endswith(os.altsep)
 
 def splitpath(path):
     '''Split path by os.sep.
     Note that this function does not use os.altsep because this is
     an alternative of simple "xxx.split(os.sep)".
     It is recommended to use os.path.normpath() before using this
     function if need.'''
     return path.split(os.sep)
 
 def gui():
     '''Are we running in a GUI?'''
     if sys.platform == 'darwin':
         if 'SSH_CONNECTION' in os.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 os.name == "nt" or os.environ.get("DISPLAY")
 
 def mktempcopy(name, emptyok=False, createmode=None):
     """Create a temporary file with the same contents from name
 
     The permission bits are copied from the original file.
 
     If the temporary file is going to be truncated immediately, you
     can use emptyok=True as an optimization.
 
     Returns the name of the temporary file.
     """
     d, fn = os.path.split(name)
     fd, temp = tempfile.mkstemp(prefix='.%s-' % fn, dir=d)
     os.close(fd)
     # Temporary files are created with mode 0600, which is usually not
     # what we want.  If the original file already exists, just copy
     # its mode.  Otherwise, manually obey umask.
     copymode(name, temp, createmode)
     if emptyok:
         return temp
     try:
         try:
             ifp = posixfile(name, "rb")
         except IOError, inst:
             if inst.errno == errno.ENOENT:
                 return temp
             if not getattr(inst, 'filename', None):
                 inst.filename = name
             raise
         ofp = posixfile(temp, "wb")
         for chunk in filechunkiter(ifp):
             ofp.write(chunk)
         ifp.close()
         ofp.close()
     except:
         try: os.unlink(temp)
         except: pass
         raise
     return temp
 
 class atomictempfile(object):
     '''writeable file object that atomically updates a file
 
     All writes will go to a temporary copy of the original file. Call
     rename() when you are done writing, and atomictempfile will rename
     the temporary copy to the original name, making the changes visible.
 
     Unlike other file-like objects, close() discards your writes by
     simply deleting the temporary file.
     '''
     def __init__(self, name, mode='w+b', createmode=None):
         self.__name = name      # permanent name
         self._tempname = mktempcopy(name, emptyok=('w' in mode),
                                     createmode=createmode)
         self._fp = posixfile(self._tempname, mode)
 
         # delegated methods
         self.write = self._fp.write
         self.fileno = self._fp.fileno
 
     def rename(self):
         if not self._fp.closed:
             self._fp.close()
             rename(self._tempname, localpath(self.__name))
 
     def close(self):
         if not self._fp.closed:
             try:
                 os.unlink(self._tempname)
             except OSError:
                 pass
             self._fp.close()
 
     def __del__(self):
         if safehasattr(self, '_fp'): # constructor actually did something
             self.close()
 
 def makedirs(name, mode=None):
     """recursive directory creation with parent mode inheritance"""
     parent = os.path.abspath(os.path.dirname(name))
     try:
         os.mkdir(name)
         if mode is not None:
             os.chmod(name, mode)
         return
     except OSError, err:
         if err.errno == errno.EEXIST:
             return
         if not name or parent == name or err.errno != errno.ENOENT:
             raise
     makedirs(parent, mode)
     makedirs(name, mode)
 
 def readfile(path):
     fp = open(path, 'rb')
     try:
         return fp.read()
     finally:
         fp.close()
 
 def writefile(path, text):
     fp = open(path, 'wb')
     try:
         fp.write(text)
     finally:
         fp.close()
 
 def appendfile(path, text):
     fp = open(path, 'ab')
     try:
         fp.write(text)
     finally:
         fp.close()
 
 class chunkbuffer(object):
     """Allow arbitrary sized chunks of data to be efficiently read from an
     iterator over chunks of arbitrary size."""
 
     def __init__(self, in_iter):
         """in_iter is the iterator that's iterating over the input chunks.
         targetsize is how big a buffer to try to maintain."""
         def splitbig(chunks):
             for chunk in chunks:
                 if len(chunk) > 2**20:
                     pos = 0
                     while pos < len(chunk):
                         end = pos + 2 ** 18
                         yield chunk[pos:end]
                         pos = end
                 else:
                     yield chunk
         self.iter = splitbig(in_iter)
         self._queue = []
 
     def read(self, l):
         """Read L bytes of data from the iterator of chunks of data.
         Returns less than L bytes if the iterator runs dry."""
         left = l
         buf = ''
         queue = self._queue
         while left > 0:
             # refill the queue
             if not queue:
                 target = 2**18
                 for chunk in self.iter:
                     queue.append(chunk)
                     target -= len(chunk)
                     if target <= 0:
                         break
                 if not queue:
                     break
 
             chunk = queue.pop(0)
             left -= len(chunk)
             if left < 0:
                 queue.insert(0, chunk[left:])
                 buf += chunk[:left]
             else:
                 buf += chunk
 
         return buf
 
 def filechunkiter(f, size=65536, limit=None):
     """Create a generator that produces the data in the file size
     (default 65536) bytes at a time, up to optional limit (default is
     to read all data).  Chunks may be less than size bytes if the
     chunk is the last chunk in the file, or the file is a socket or
     some other type of file that sometimes reads less data than is
     requested."""
     assert size >= 0
     assert limit is None or limit >= 0
     while True:
         if limit is None:
             nbytes = size
         else:
             nbytes = min(limit, size)
         s = nbytes and f.read(nbytes)
         if not s:
             break
         if limit:
             limit -= len(s)
         yield s
 
 def makedate():
     lt = time.localtime()
     if lt[8] == 1 and time.daylight:
         tz = time.altzone
     else:
         tz = time.timezone
     t = time.mktime(lt)
     if t < 0:
         hint = _("check your clock")
         raise Abort(_("negative timestamp: %d") % t, hint=hint)
     return t, tz
 
 def datestr(date=None, format='%a %b %d %H:%M:%S %Y %1%2'):
     """represent a (unixtime, offset) tuple as a localized time.
     unixtime is seconds since the epoch, and offset is the time zone's
     number of seconds away from UTC. if timezone is false, do not
     append time zone to string."""
     t, tz = date or makedate()
     if t < 0:
         t = 0   # time.gmtime(lt) fails on Windows for lt < -43200
         tz = 0
     if "%1" in format or "%2" in format:
         sign = (tz > 0) and "-" or "+"
         minutes = abs(tz) // 60
         format = format.replace("%1", "%c%02d" % (sign, minutes // 60))
         format = format.replace("%2", "%02d" % (minutes % 60))
     s = time.strftime(format, time.gmtime(float(t) - tz))
     return s
 
 def shortdate(date=None):
     """turn (timestamp, tzoff) tuple into iso 8631 date."""
     return datestr(date, format='%Y-%m-%d')
 
 def strdate(string, format, defaults=[]):
     """parse a localized time string and return a (unixtime, offset) tuple.
     if the string cannot be parsed, ValueError is raised."""
     def timezone(string):
         tz = string.split()[-1]
         if tz[0] in "+-" and len(tz) == 5 and tz[1:].isdigit():
             sign = (tz[0] == "+") and 1 or -1
             hours = int(tz[1:3])
             minutes = int(tz[3:5])
             return -sign * (hours * 60 + minutes) * 60
         if tz == "GMT" or tz == "UTC":
             return 0
         return None
 
     # NOTE: unixtime = localunixtime + offset
     offset, date = timezone(string), string
     if offset is not None:
         date = " ".join(string.split()[:-1])
 
     # add missing elements from defaults
     usenow = False # default to using biased defaults
     for part in ("S", "M", "HI", "d", "mb", "yY"): # decreasing specificity
         found = [True for p in part if ("%"+p) in format]
         if not found:
             date += "@" + defaults[part][usenow]
             format += "@%" + part[0]
         else:
             # We've found a specific time element, less specific time
             # elements are relative to today
             usenow = True
 
     timetuple = time.strptime(date, format)
     localunixtime = int(calendar.timegm(timetuple))
     if offset is None:
         # local timezone
         unixtime = int(time.mktime(timetuple))
         offset = unixtime - localunixtime
     else:
         unixtime = localunixtime + offset
     return unixtime, offset
 
 def parsedate(date, formats=None, bias={}):
     """parse a localized date/time and return a (unixtime, offset) tuple.
 
     The date may be a "unixtime offset" string or in one of the specified
     formats. If the date already is a (unixtime, offset) tuple, it is returned.
     """
     if not date:
         return 0, 0
     if isinstance(date, tuple) and len(date) == 2:
         return date
     if not formats:
         formats = defaultdateformats
     date = date.strip()
     try:
         when, offset = map(int, date.split(' '))
     except ValueError:
         # fill out defaults
         now = makedate()
         defaults = {}
         for part in ("d", "mb", "yY", "HI", "M", "S"):
             # this piece is for rounding the specific end of unknowns
             b = bias.get(part)
             if b is None:
                 if part[0] in "HMS":
                     b = "00"
                 else:
                     b = "0"
 
             # this piece is for matching the generic end to today's date
             n = datestr(now, "%" + part[0])
 
             defaults[part] = (b, n)
 
         for format in formats:
             try:
                 when, offset = strdate(date, format, defaults)
             except (ValueError, OverflowError):
                 pass
             else:
                 break
         else:
             raise Abort(_('invalid date: %r') % date)
     # validate explicit (probably user-specified) date and
     # time zone offset. values must fit in signed 32 bits for
     # current 32-bit linux runtimes. timezones go from UTC-12
     # to UTC+14
     if abs(when) > 0x7fffffff:
         raise Abort(_('date exceeds 32 bits: %d') % when)
     if when < 0:
         raise Abort(_('negative date value: %d') % when)
     if offset < -50400 or offset > 43200:
         raise Abort(_('impossible time zone offset: %d') % offset)
     return when, offset
 
 def matchdate(date):
     """Return a function that matches a given date match specifier
 
     Formats include:
 
     '{date}' match a given date to the accuracy provided
 
     '<{date}' on or before a given date
 
     '>{date}' on or after a given date
 
     >>> p1 = parsedate("10:29:59")
     >>> p2 = parsedate("10:30:00")
     >>> p3 = parsedate("10:30:59")
     >>> p4 = parsedate("10:31:00")
     >>> p5 = parsedate("Sep 15 10:30:00 1999")
     >>> f = matchdate("10:30")
     >>> f(p1[0])
     False
     >>> f(p2[0])
     True
     >>> f(p3[0])
     True
     >>> f(p4[0])
     False
     >>> f(p5[0])
     False
     """
 
     def lower(date):
         d = dict(mb="1", d="1")
         return parsedate(date, extendeddateformats, d)[0]
 
     def upper(date):
         d = dict(mb="12", HI="23", M="59", S="59")
         for days in ("31", "30", "29"):
             try:
                 d["d"] = days
                 return parsedate(date, extendeddateformats, d)[0]
             except:
                 pass
         d["d"] = "28"
         return parsedate(date, extendeddateformats, d)[0]
 
     date = date.strip()
 
     if not date:
         raise Abort(_("dates cannot consist entirely of whitespace"))
     elif date[0] == "<":
         if not date[1:]:
             raise Abort(_("invalid day spec, use '<DATE'"))
         when = upper(date[1:])
         return lambda x: x <= when
     elif date[0] == ">":
         if not date[1:]:
             raise Abort(_("invalid day spec, use '>DATE'"))
         when = lower(date[1:])
         return lambda x: x >= when
     elif date[0] == "-":
         try:
             days = int(date[1:])
         except ValueError:
             raise Abort(_("invalid day spec: %s") % date[1:])
         if days < 0:
             raise Abort(_("%s must be nonnegative (see 'hg help dates')")
                 % date[1:])
         when = makedate()[0] - days * 3600 * 24
         return lambda x: x >= when
     elif " to " in date:
         a, b = date.split(" to ")
         start, stop = lower(a), upper(b)
         return lambda x: x >= start and x <= stop
     else:
         start, stop = lower(date), upper(date)
         return lambda x: x >= start and x <= stop
 
 def shortuser(user):
     """Return a short representation of a user name or email address."""
     f = user.find('@')
     if f >= 0:
         user = user[:f]
     f = user.find('<')
     if f >= 0:
         user = user[f + 1:]
     f = user.find(' ')
     if f >= 0:
         user = user[:f]
     f = user.find('.')
     if f >= 0:
         user = user[:f]
     return user
 
 def email(author):
     '''get email of author.'''
     r = author.find('>')
     if r == -1:
         r = None
     return author[author.find('<') + 1:r]
 
 def _ellipsis(text, maxlength):
     if len(text) <= maxlength:
         return text, False
     else:
         return "%s..." % (text[:maxlength - 3]), True
 
 def ellipsis(text, maxlength=400):
     """Trim string to at most maxlength (default: 400) characters."""
     try:
         # use unicode not to split at intermediate multi-byte sequence
         utext, truncated = _ellipsis(text.decode(encoding.encoding),
                                      maxlength)
         if not truncated:
             return text
         return utext.encode(encoding.encoding)
     except (UnicodeDecodeError, UnicodeEncodeError):
         return _ellipsis(text, maxlength)[0]
 
 def bytecount(nbytes):
     '''return byte count formatted as readable string, with units'''
 
     units = (
         (100, 1 << 30, _('%.0f GB')),
         (10, 1 << 30, _('%.1f GB')),
         (1, 1 << 30, _('%.2f GB')),
         (100, 1 << 20, _('%.0f MB')),
         (10, 1 << 20, _('%.1f MB')),
         (1, 1 << 20, _('%.2f MB')),
         (100, 1 << 10, _('%.0f KB')),
         (10, 1 << 10, _('%.1f KB')),
         (1, 1 << 10, _('%.2f KB')),
         (1, 1, _('%.0f bytes')),
         )
 
     for multiplier, divisor, format in units:
         if nbytes >= divisor * multiplier:
             return format % (nbytes / float(divisor))
     return units[-1][2] % nbytes
 
 def uirepr(s):
     # Avoid double backslash in Windows path repr()
     return repr(s).replace('\\\\', '\\')
 
 # delay import of textwrap
 def MBTextWrapper(**kwargs):
     class tw(textwrap.TextWrapper):
         """
         Extend TextWrapper for double-width characters.
 
         Some Asian characters use two terminal columns instead of one.
         A good example of this behavior can be seen with u'\u65e5\u672c',
         the two Japanese characters for "Japan":
         len() returns 2, but when printed to a terminal, they eat 4 columns.
 
         (Note that this has nothing to do whatsoever with unicode
         representation, or encoding of the underlying string)
         """
         def __init__(self, **kwargs):
             textwrap.TextWrapper.__init__(self, **kwargs)
 
         def _cutdown(self, str, space_left):
             l = 0
             ucstr = unicode(str, encoding.encoding)
             colwidth = unicodedata.east_asian_width
             for i in xrange(len(ucstr)):
                 l += colwidth(ucstr[i]) in 'WFA' and 2 or 1
                 if space_left < l:
                     return (ucstr[:i].encode(encoding.encoding),
                             ucstr[i:].encode(encoding.encoding))
             return str, ''
 
         # overriding of base class
         def _handle_long_word(self, reversed_chunks, cur_line, cur_len, width):
             space_left = max(width - cur_len, 1)
 
             if self.break_long_words:
                 cut, res = self._cutdown(reversed_chunks[-1], space_left)
                 cur_line.append(cut)
                 reversed_chunks[-1] = res
             elif not cur_line:
                 cur_line.append(reversed_chunks.pop())
 
     global MBTextWrapper
     MBTextWrapper = tw
     return tw(**kwargs)
 
 def wrap(line, width, initindent='', hangindent=''):
     maxindent = max(len(hangindent), len(initindent))
     if width <= maxindent:
         # adjust for weird terminal size
         width = max(78, maxindent + 1)
     wrapper = MBTextWrapper(width=width,
                             initial_indent=initindent,
                             subsequent_indent=hangindent)
     return wrapper.fill(line)
 
 def iterlines(iterator):
     for chunk in iterator:
         for line in chunk.splitlines():
             yield line
 
 def expandpath(path):
     return os.path.expanduser(os.path.expandvars(path))
 
 def 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 mainfrozen():
         return [sys.executable]
     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)
 
 try:
     any, all = any, all
 except NameError:
     def any(iterable):
         for i in iterable:
             if i:
                 return True
         return False
 
     def all(iterable):
         for i in iterable:
             if not i:
                 return False
         return True
 
 def interpolate(prefix, mapping, s, fn=None, escape_prefix=False):
     """Return the result of interpolating items in the mapping into string s.
 
     prefix is a single character string, or a two character string with
     a backslash as the first character if the prefix needs to be escaped in
     a regular expression.
 
     fn is an optional function that will be applied to the replacement text
     just before replacement.
 
     escape_prefix is an optional flag that allows using doubled prefix for
     its escaping.
     """
     fn = fn or (lambda s: s)
     patterns = '|'.join(mapping.keys())
     if escape_prefix:
         patterns += '|' + prefix
         if len(prefix) > 1:
             prefix_char = prefix[1:]
         else:
             prefix_char = prefix
         mapping[prefix_char] = prefix_char
     r = re.compile(r'%s(%s)' % (prefix, patterns))
     return r.sub(lambda x: fn(mapping[x.group()[1:]]), s)
 
 def getport(port):
     """Return the port for a given network service.
 
     If port is an integer, it's returned as is. If it's a string, it's
     looked up using socket.getservbyname(). If there's no matching
     service, util.Abort is raised.
     """
     try:
         return int(port)
     except ValueError:
         pass
 
     try:
         return socket.getservbyname(port)
     except socket.error:
         raise Abort(_("no port number associated with service '%s'") % port)
 
 _booleans = {'1': True, 'yes': True, 'true': True, 'on': True, 'always': True,
              '0': False, 'no': False, 'false': False, 'off': False,
              'never': False}
 
 def parsebool(s):
     """Parse s into a boolean.
 
     If s is not a valid boolean, returns None.
     """
     return _booleans.get(s.lower(), None)
 
 _hexdig = '0123456789ABCDEFabcdef'
 _hextochr = dict((a + b, chr(int(a + b, 16)))
                  for a in _hexdig for b in _hexdig)
 
 def _urlunquote(s):
     """unquote('abc%20def') -> 'abc def'."""
     res = s.split('%')
     # fastpath
     if len(res) == 1:
         return s
     s = res[0]
     for item in res[1:]:
         try:
             s += _hextochr[item[:2]] + item[2:]
         except KeyError:
             s += '%' + item
         except UnicodeDecodeError:
             s += unichr(int(item[:2], 16)) + item[2:]
     return s
 
 class url(object):
     r"""Reliable URL parser.
 
     This parses URLs and provides attributes for the following
     components:
 
     <scheme>://<user>:<passwd>@<host>:<port>/<path>?<query>#<fragment>
 
     Missing components are set to None. The only exception is
     fragment, which is set to '' if present but empty.
 
     If parsefragment is False, fragment is included in query. If
     parsequery is False, query is included in path. If both are
     False, both fragment and query are included in path.
 
     See http://www.ietf.org/rfc/rfc2396.txt for more information.
 
     Note that for backward compatibility reasons, bundle URLs do not
     take host names. That means 'bundle://../' has a path of '../'.
 
     Examples:
 
     >>> url('http://www.ietf.org/rfc/rfc2396.txt')
     <url scheme: 'http', host: 'www.ietf.org', path: 'rfc/rfc2396.txt'>
     >>> url('ssh://[::1]:2200//home/joe/repo')
     <url scheme: 'ssh', host: '[::1]', port: '2200', path: '/home/joe/repo'>
     >>> url('file:///home/joe/repo')
     <url scheme: 'file', path: '/home/joe/repo'>
     >>> url('file:///c:/temp/foo/')
     <url scheme: 'file', path: 'c:/temp/foo/'>
     >>> url('bundle:foo')
     <url scheme: 'bundle', path: 'foo'>
     >>> url('bundle://../foo')
     <url scheme: 'bundle', path: '../foo'>
     >>> url(r'c:\foo\bar')
     <url path: 'c:\\foo\\bar'>
     >>> url(r'\\blah\blah\blah')
     <url path: '\\\\blah\\blah\\blah'>
 
     Authentication credentials:
 
     >>> url('ssh://joe:xyz@x/repo')
     <url scheme: 'ssh', user: 'joe', passwd: 'xyz', host: 'x', path: 'repo'>
     >>> url('ssh://joe@x/repo')
     <url scheme: 'ssh', user: 'joe', host: 'x', path: 'repo'>
 
     Query strings and fragments:
 
     >>> url('http://host/a?b#c')
     <url scheme: 'http', host: 'host', path: 'a', query: 'b', fragment: 'c'>
     >>> url('http://host/a?b#c', parsequery=False, parsefragment=False)
     <url scheme: 'http', host: 'host', path: 'a?b#c'>
     """
 
     _safechars = "!~*'()+"
     _safepchars = "/!~*'()+"
     _matchscheme = re.compile(r'^[a-zA-Z0-9+.\-]+:').match
 
     def __init__(self, path, parsequery=True, parsefragment=True):
         # We slowly chomp away at path until we have only the path left
         self.scheme = self.user = self.passwd = self.host = None
         self.port = self.path = self.query = self.fragment = None
         self._localpath = True
         self._hostport = ''
         self._origpath = path
 
         # special case for Windows drive letters and UNC paths
         if hasdriveletter(path) or path.startswith(r'\\'):
             self.path = path
             return
 
         # For compatibility reasons, we can't handle bundle paths as
         # normal URLS
         if path.startswith('bundle:'):
             self.scheme = 'bundle'
             path = path[7:]
             if path.startswith('//'):
                 path = path[2:]
             self.path = path
             return
 
         if self._matchscheme(path):
             parts = path.split(':', 1)
             if parts[0]:
                 self.scheme, path = parts
                 self._localpath = False
 
         if not path:
             path = None
             if self._localpath:
                 self.path = ''
                 return
         else:
             if parsefragment and '#' in path:
                 path, self.fragment = path.split('#', 1)
                 if not path:
                     path = None
             if self._localpath:
                 self.path = path
                 return
 
             if parsequery and '?' in path:
                 path, self.query = path.split('?', 1)
                 if not path:
                     path = None
                 if not self.query:
                     self.query = None
 
             # // is required to specify a host/authority
             if path and path.startswith('//'):
                 parts = path[2:].split('/', 1)
                 if len(parts) > 1:
                     self.host, path = parts
                     path = path
                 else:
                     self.host = parts[0]
                     path = None
                 if not self.host:
                     self.host = None
+                    # path of file:///d is /d
+                    # path of file:///d:/ is d:/, not /d:/
                     if path and not hasdriveletter(path):
                         path = '/' + path
 
             if self.host and '@' in self.host:
                 self.user, self.host = self.host.rsplit('@', 1)
                 if ':' in self.user:
                     self.user, self.passwd = self.user.split(':', 1)
                 if not self.host:
                     self.host = None
 
             # Don't split on colons in IPv6 addresses without ports
             if (self.host and ':' in self.host and
                 not (self.host.startswith('[') and self.host.endswith(']'))):
                 self._hostport = self.host
                 self.host, self.port = self.host.rsplit(':', 1)
                 if not self.host:
                     self.host = None
 
             if (self.host and self.scheme == 'file' and
                 self.host not in ('localhost', '127.0.0.1', '[::1]')):
                 raise Abort(_('file:// URLs can only refer to localhost'))
 
         self.path = path
 
         # leave the query string escaped
         for a in ('user', 'passwd', 'host', 'port',
                   'path', 'fragment'):
             v = getattr(self, a)
             if v is not None:
                 setattr(self, a, _urlunquote(v))
 
     def __repr__(self):
         attrs = []
         for a in ('scheme', 'user', 'passwd', 'host', 'port', 'path',
                   'query', 'fragment'):
             v = getattr(self, a)
             if v is not None:
                 attrs.append('%s: %r' % (a, v))
         return '<url %s>' % ', '.join(attrs)
 
     def __str__(self):
         r"""Join the URL's components back into a URL string.
 
         Examples:
 
         >>> str(url('http://user:pw@host:80/?foo#bar'))
         'http://user:pw@host:80/?foo#bar'
         >>> str(url('http://user:pw@host:80/?foo=bar&baz=42'))
         'http://user:pw@host:80/?foo=bar&baz=42'
         >>> str(url('http://user:pw@host:80/?foo=bar%3dbaz'))
         'http://user:pw@host:80/?foo=bar%3dbaz'
         >>> str(url('ssh://user:pw@[::1]:2200//home/joe#'))
         'ssh://user:pw@[::1]:2200//home/joe#'
         >>> str(url('http://localhost:80//'))
         'http://localhost:80//'
         >>> str(url('http://localhost:80/'))
         'http://localhost:80/'
         >>> str(url('http://localhost:80'))
         'http://localhost:80/'
         >>> str(url('bundle:foo'))
         'bundle:foo'
         >>> str(url('bundle://../foo'))
         'bundle:../foo'
         >>> str(url('path'))
         'path'
         >>> str(url('file:///tmp/foo/bar'))
         'file:///tmp/foo/bar'
         >>> print url(r'bundle:foo\bar')
         bundle:foo\bar
         """
         if self._localpath:
             s = self.path
             if self.scheme == 'bundle':
                 s = 'bundle:' + s
             if self.fragment:
                 s += '#' + self.fragment
             return s
 
         s = self.scheme + ':'
         if self.user or self.passwd or self.host:
             s += '//'
         elif self.scheme and (not self.path or self.path.startswith('/')):
             s += '//'
         if self.user:
             s += urllib.quote(self.user, safe=self._safechars)
         if self.passwd:
             s += ':' + urllib.quote(self.passwd, safe=self._safechars)
         if self.user or self.passwd:
             s += '@'
         if self.host:
             if not (self.host.startswith('[') and self.host.endswith(']')):
                 s += urllib.quote(self.host)
             else:
                 s += self.host
         if self.port:
             s += ':' + urllib.quote(self.port)
         if self.host:
             s += '/'
         if self.path:
             # TODO: similar to the query string, we should not unescape the
             # path when we store it, the path might contain '%2f' = '/',
             # which we should *not* escape.
             s += urllib.quote(self.path, safe=self._safepchars)
         if self.query:
             # we store the query in escaped form.
             s += '?' + self.query
         if self.fragment is not None:
             s += '#' + urllib.quote(self.fragment, safe=self._safepchars)
         return s
 
     def authinfo(self):
         user, passwd = self.user, self.passwd
         try:
             self.user, self.passwd = None, None
             s = str(self)
         finally:
             self.user, self.passwd = user, passwd
         if not self.user:
             return (s, None)
-        return (s, (None, (str(self), self.host),
+        # authinfo[1] is passed to urllib2 password manager, and its URIs
+        # must not contain credentials.
+        return (s, (None, (s, self.host),
                     self.user, self.passwd or ''))
 
     def isabs(self):
         if self.scheme and self.scheme != 'file':
             return True # remote URL
         if hasdriveletter(self.path):
             return True # absolute for our purposes - can't be joined()
         if self.path.startswith(r'\\'):
             return True # Windows UNC path
         if self.path.startswith('/'):
             return True # POSIX-style
         return False
 
     def localpath(self):
         if self.scheme == 'file' or self.scheme == 'bundle':
             path = self.path or '/'
             # For Windows, we need to promote hosts containing drive
             # letters to paths with drive letters.
             if hasdriveletter(self._hostport):
                 path = self._hostport + '/' + self.path
             elif self.host is not None and self.path:
                 path = '/' + path
-            # We also need to handle the case of file:///C:/, which
-            # should return C:/, not /C:/.
-            elif hasdriveletter(path):
-                # Strip leading slash from paths with drive names
-                return path[1:]
             return path
         return self._origpath
 
 def hasscheme(path):
     return bool(url(path).scheme)
 
 def hasdriveletter(path):
     return path[1:2] == ':' and path[0:1].isalpha()
 
 def urllocalpath(path):
     return url(path, parsequery=False, parsefragment=False).localpath()
 
 def hidepassword(u):
     '''hide user credential in a url string'''
     u = url(u)
     if u.passwd:
         u.passwd = '***'
     return str(u)
 
 def removeauth(u):
     '''remove all authentication information from a url string'''
     u = url(u)
     u.user = u.passwd = None
     return str(u)
 
 def isatty(fd):
     try:
         return fd.isatty()
     except AttributeError:
         return False

tests/test-hgweb-auth.py

@@ -1,97 +1,107 b''
 from mercurial import demandimport; demandimport.enable()
+import urllib2
 from mercurial import ui, util
 from mercurial import url
 from mercurial.error import Abort
 
 class myui(ui.ui):
     def interactive(self):
         return False
 
 origui = myui()
 
 def writeauth(items):
     ui = origui.copy()
     for name, value in items.iteritems():
         ui.setconfig('auth', name, value)
     return ui
 
 def dumpdict(dict):
     return '{' + ', '.join(['%s: %s' % (k, dict[k])
                             for k in sorted(dict.iterkeys())]) + '}'
 
 def test(auth, urls=None):
     print 'CFG:', dumpdict(auth)
     prefixes = set()
     for k in auth:
         prefixes.add(k.split('.', 1)[0])
     for p in prefixes:
         for name in ('.username', '.password'):
             if (p + name) not in auth:
                 auth[p + name] = p
     auth = dict((k, v) for k, v in auth.iteritems() if v is not None)
 
     ui = writeauth(auth)
 
     def _test(uri):
         print 'URI:', uri
         try:
             pm = url.passwordmgr(ui)
-            authinfo = util.url(uri).authinfo()[1]
+            u, authinfo = util.url(uri).authinfo()
             if authinfo is not None:
                 pm.add_password(*authinfo)
-            print '    ', pm.find_user_password('test', uri)
+            print '    ', pm.find_user_password('test', u)
         except Abort, e:
             print 'abort'
 
     if not urls:
         urls = [
             'http://example.org/foo',
             'http://example.org/foo/bar',
             'http://example.org/bar',
             'https://example.org/foo',
             'https://example.org/foo/bar',
             'https://example.org/bar',
             'https://[e-mail]/bar',
             'https://[e-mail]/bar',
             ]
     for u in urls:
         _test(u)
 
 
 print '\n*** Test in-uri schemes\n'
 test({'x.prefix': 'http://example.org'})
 test({'x.prefix': 'https://example.org'})
 test({'x.prefix': 'http://example.org', 'x.schemes': 'https'})
 test({'x.prefix': 'https://example.org', 'x.schemes': 'http'})
 
 print '\n*** Test separately configured schemes\n'
 test({'x.prefix': 'example.org', 'x.schemes': 'http'})
 test({'x.prefix': 'example.org', 'x.schemes': 'https'})
 test({'x.prefix': 'example.org', 'x.schemes': 'http https'})
 
 print '\n*** Test prefix matching\n'
 test({'x.prefix': 'http://example.org/foo',
       'y.prefix': 'http://example.org/bar'})
 test({'x.prefix': 'http://example.org/foo',
       'y.prefix': 'http://example.org/foo/bar'})
 test({'x.prefix': '*', 'y.prefix': 'https://example.org/bar'})
 
 print '\n*** Test user matching\n'
 test({'x.prefix': 'http://example.org/foo',
       'x.username': None,
       'x.password': 'xpassword'},
      urls=['http://[e-mail]/foo'])
 test({'x.prefix': 'http://example.org/foo',
       'x.username': None,
       'x.password': 'xpassword',
       'y.prefix': 'http://example.org/foo',
       'y.username': 'y',
       'y.password': 'ypassword'},
      urls=['http://[e-mail]/foo'])
 test({'x.prefix': 'http://example.org/foo/bar',
       'x.username': None,
       'x.password': 'xpassword',
       'y.prefix': 'http://example.org/foo',
       'y.username': 'y',
       'y.password': 'ypassword'},
      urls=['http://[e-mail]/foo/bar'])
+
+def testauthinfo(fullurl, authurl):
+    print 'URIs:', fullurl, authurl
+    pm = urllib2.HTTPPasswordMgrWithDefaultRealm()
+    pm.add_password(*util.url(fullurl).authinfo()[1])
+    print pm.find_user_password('test', authurl)
+
+print '\n*** Test urllib2 and util.url\n'
+testauthinfo('http://[e-mail]:8080/foo', 'http://example.com:8080/foo')

tests/test-hgweb-auth.py.out

@@ -1,191 +1,196 b''
 
 *** Test in-uri schemes
 
 CFG: {x.prefix: http://example.org}
 URI: http://example.org/foo
      ('x', 'x')
 URI: http://example.org/foo/bar
      ('x', 'x')
 URI: http://example.org/bar
      ('x', 'x')
 URI: https://example.org/foo
      abort
 URI: https://example.org/foo/bar
      abort
 URI: https://example.org/bar
      abort
 URI: https://[e-mail]/bar
      abort
 URI: https://[e-mail]/bar
      abort
 CFG: {x.prefix: https://example.org}
 URI: http://example.org/foo
      abort
 URI: http://example.org/foo/bar
      abort
 URI: http://example.org/bar
      abort
 URI: https://example.org/foo
      ('x', 'x')
 URI: https://example.org/foo/bar
      ('x', 'x')
 URI: https://example.org/bar
      ('x', 'x')
 URI: https://[e-mail]/bar
      ('x', 'x')
 URI: https://[e-mail]/bar
      abort
 CFG: {x.prefix: http://example.org, x.schemes: https}
 URI: http://example.org/foo
      ('x', 'x')
 URI: http://example.org/foo/bar
      ('x', 'x')
 URI: http://example.org/bar
      ('x', 'x')
 URI: https://example.org/foo
      abort
 URI: https://example.org/foo/bar
      abort
 URI: https://example.org/bar
      abort
 URI: https://[e-mail]/bar
      abort
 URI: https://[e-mail]/bar
      abort
 CFG: {x.prefix: https://example.org, x.schemes: http}
 URI: http://example.org/foo
      abort
 URI: http://example.org/foo/bar
      abort
 URI: http://example.org/bar
      abort
 URI: https://example.org/foo
      ('x', 'x')
 URI: https://example.org/foo/bar
      ('x', 'x')
 URI: https://example.org/bar
      ('x', 'x')
 URI: https://[e-mail]/bar
      ('x', 'x')
 URI: https://[e-mail]/bar
      abort
 
 *** Test separately configured schemes
 
 CFG: {x.prefix: example.org, x.schemes: http}
 URI: http://example.org/foo
      ('x', 'x')
 URI: http://example.org/foo/bar
      ('x', 'x')
 URI: http://example.org/bar
      ('x', 'x')
 URI: https://example.org/foo
      abort
 URI: https://example.org/foo/bar
      abort
 URI: https://example.org/bar
      abort
 URI: https://[e-mail]/bar
      abort
 URI: https://[e-mail]/bar
      abort
 CFG: {x.prefix: example.org, x.schemes: https}
 URI: http://example.org/foo
      abort
 URI: http://example.org/foo/bar
      abort
 URI: http://example.org/bar
      abort
 URI: https://example.org/foo
      ('x', 'x')
 URI: https://example.org/foo/bar
      ('x', 'x')
 URI: https://example.org/bar
      ('x', 'x')
 URI: https://[e-mail]/bar
      ('x', 'x')
 URI: https://[e-mail]/bar
      abort
 CFG: {x.prefix: example.org, x.schemes: http https}
 URI: http://example.org/foo
      ('x', 'x')
 URI: http://example.org/foo/bar
      ('x', 'x')
 URI: http://example.org/bar
      ('x', 'x')
 URI: https://example.org/foo
      ('x', 'x')
 URI: https://example.org/foo/bar
      ('x', 'x')
 URI: https://example.org/bar
      ('x', 'x')
 URI: https://[e-mail]/bar
      ('x', 'x')
 URI: https://[e-mail]/bar
      abort
 
 *** Test prefix matching
 
 CFG: {x.prefix: http://example.org/foo, y.prefix: http://example.org/bar}
 URI: http://example.org/foo
      ('x', 'x')
 URI: http://example.org/foo/bar
      ('x', 'x')
 URI: http://example.org/bar
      ('y', 'y')
 URI: https://example.org/foo
      abort
 URI: https://example.org/foo/bar
      abort
 URI: https://example.org/bar
      abort
 URI: https://[e-mail]/bar
      abort
 URI: https://[e-mail]/bar
      abort
 CFG: {x.prefix: http://example.org/foo, y.prefix: http://example.org/foo/bar}
 URI: http://example.org/foo
      ('x', 'x')
 URI: http://example.org/foo/bar
      ('y', 'y')
 URI: http://example.org/bar
      abort
 URI: https://example.org/foo
      abort
 URI: https://example.org/foo/bar
      abort
 URI: https://example.org/bar
      abort
 URI: https://[e-mail]/bar
      abort
 URI: https://[e-mail]/bar
      abort
 CFG: {x.prefix: *, y.prefix: https://example.org/bar}
 URI: http://example.org/foo
      abort
 URI: http://example.org/foo/bar
      abort
 URI: http://example.org/bar
      abort
 URI: https://example.org/foo
      ('x', 'x')
 URI: https://example.org/foo/bar
      ('x', 'x')
 URI: https://example.org/bar
      ('y', 'y')
 URI: https://[e-mail]/bar
      ('x', 'x')
 URI: https://[e-mail]/bar
      ('y', 'y')
 
 *** Test user matching
 
 CFG: {x.password: xpassword, x.prefix: http://example.org/foo, x.username: None}
 URI: http://[e-mail]/foo
      ('y', 'xpassword')
 CFG: {x.password: xpassword, x.prefix: http://example.org/foo, x.username: None, y.password: ypassword, y.prefix: http://example.org/foo, y.username: y}
 URI: http://[e-mail]/foo
      ('y', 'ypassword')
 CFG: {x.password: xpassword, x.prefix: http://example.org/foo/bar, x.username: None, y.password: ypassword, y.prefix: http://example.org/foo, y.username: y}
 URI: http://[e-mail]/foo/bar
      ('y', 'xpassword')
+
+*** Test urllib2 and util.url
+
+URIs: http://[e-mail]:8080/foo http://example.com:8080/foo
+('user', '')

tests/test-http.t

@@ -1,115 +1,164 b''
 
   $ hg init test
   $ cd test
   $ echo foo>foo
   $ mkdir foo.d foo.d/bAr.hg.d foo.d/baR.d.hg
   $ echo foo>foo.d/foo
   $ echo bar>foo.d/bAr.hg.d/BaR
   $ echo bar>foo.d/baR.d.hg/bAR
   $ hg commit -A -m 1
   adding foo
   adding foo.d/bAr.hg.d/BaR
   adding foo.d/baR.d.hg/bAR
   adding foo.d/foo
   $ hg serve -p $HGPORT -d --pid-file=../hg1.pid -E ../error.log
   $ hg --config server.uncompressed=False serve -p $HGPORT1 -d --pid-file=../hg2.pid
 
 Test server address cannot be reused
 
   $ hg serve -p $HGPORT1 2>&1
   abort: cannot start server at ':$HGPORT1': Address already in use
   [255]
   $ cd ..
   $ cat hg1.pid hg2.pid >> $DAEMON_PIDS
 
 clone via stream
 
   $ hg clone --uncompressed http://localhost:$HGPORT/ copy 2>&1
   streaming all changes
   6 files to transfer, 606 bytes of data
   transferred * bytes in * seconds (*/sec) (glob)
   updating to branch default
   4 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg verify -R copy
   checking changesets
   checking manifests
   crosschecking files in changesets and manifests
   checking files
   4 files, 1 changesets, 4 total revisions
 
 try to clone via stream, should use pull instead
 
   $ hg clone --uncompressed http://localhost:$HGPORT1/ copy2
   requesting all changes
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 4 changes to 4 files
   updating to branch default
   4 files updated, 0 files merged, 0 files removed, 0 files unresolved
 
 clone via pull
 
   $ hg clone http://localhost:$HGPORT1/ copy-pull
   requesting all changes
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 4 changes to 4 files
   updating to branch default
   4 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg verify -R copy-pull
   checking changesets
   checking manifests
   crosschecking files in changesets and manifests
   checking files
   4 files, 1 changesets, 4 total revisions
   $ cd test
   $ echo bar > bar
   $ hg commit -A -d '1 0' -m 2
   adding bar
   $ cd ..
 
 incoming via HTTP
 
   $ hg clone http://localhost:$HGPORT1/ --rev 0 partial
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 4 changes to 4 files
   updating to branch default
   4 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ cd partial
   $ touch LOCAL
   $ hg ci -qAm LOCAL
   $ hg incoming http://localhost:$HGPORT1/ --template '{desc}\n'
   comparing with http://localhost:$HGPORT1/
   searching for changes
   2
   $ cd ..
 
 pull
 
   $ cd copy-pull
   $ echo '[hooks]' >> .hg/hgrc
   $ echo 'changegroup = python "$TESTDIR"/printenv.py changegroup' >> .hg/hgrc
   $ hg pull
   pulling from http://localhost:$HGPORT1/
   searching for changes
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 1 changes to 1 files
   changegroup hook: HG_NODE=5fed3813f7f5e1824344fdc9cf8f63bb662c292d HG_SOURCE=pull HG_URL=http://localhost:$HGPORT1/ 
   (run 'hg update' to get a working copy)
   $ cd ..
 
 clone from invalid URL
 
   $ hg clone http://localhost:$HGPORT/bad
   abort: HTTP Error 404: Not Found
   [255]
 
+test http authentication
+
+  $ cd test
+  $ cat << EOT > userpass.py
+  > import base64
+  > from mercurial.hgweb import common
+  > def perform_authentication(hgweb, req, op):
+  >     auth = req.env.get('HTTP_AUTHORIZATION')
+  >     if not auth:
+  >         raise common.ErrorResponse(common.HTTP_UNAUTHORIZED, 'who',
+  >                 [('WWW-Authenticate', 'Basic Realm="mercurial"')])
+  >     if base64.b64decode(auth.split()[1]).split(':', 1) != ['user', 'pass']: 
+  >         raise common.ErrorResponse(common.HTTP_FORBIDDEN, 'no')
+  > def extsetup():
+  >     common.permhooks.insert(0, perform_authentication)
+  > EOT
+  $ hg --config extensions.x=userpass.py serve -p $HGPORT2 -d --pid-file=pid
+  $ cat pid >> $DAEMON_PIDS
+
+  $ hg id http://localhost:$HGPORT2/  
+  abort: http authorization required
+  [255]
+  $ hg id http://user@localhost:$HGPORT2/ 
+  abort: http authorization required
+  [255]
+  $ hg id http://user:pass@localhost:$HGPORT2/
+  5fed3813f7f5
+  $ echo '[auth]' >> .hg/hgrc 
+  $ echo 'l.schemes=http' >> .hg/hgrc
+  $ echo 'l.prefix=lo' >> .hg/hgrc
+  $ echo 'l.username=user' >> .hg/hgrc
+  $ echo 'l.password=pass' >> .hg/hgrc
+  $ hg id http://localhost:$HGPORT2/ 
+  5fed3813f7f5
+  $ hg id http://localhost:$HGPORT2/  
+  5fed3813f7f5
+  $ hg id http://user@localhost:$HGPORT2/ 
+  5fed3813f7f5
+  $ hg id http://user:pass@localhost:$HGPORT2/
+  5fed3813f7f5
+  $ hg id http://user2@localhost:$HGPORT2/ 
+  abort: http authorization required
+  [255]
+  $ hg id http://user:pass2@localhost:$HGPORT2/
+  abort: HTTP Error 403: no
+  [255]
+
+  $ cd ..
+
 check error log
 
   $ cat error.log

tests/test-url.py

@@ -1,221 +1,235 b''
 import sys
 
 def check(a, b):
     if a != b:
         print (a, b)
 
 def cert(cn):
     return dict(subject=((('commonName', cn),),))
 
 from mercurial.sslutil import _verifycert
 
 # Test non-wildcard certificates
 check(_verifycert(cert('example.com'), 'example.com'),
       None)
 check(_verifycert(cert('example.com'), 'www.example.com'),
       'certificate is for example.com')
 check(_verifycert(cert('www.example.com'), 'example.com'),
       'certificate is for www.example.com')
 
 # Test wildcard certificates
 check(_verifycert(cert('*.example.com'), 'www.example.com'),
       None)
 check(_verifycert(cert('*.example.com'), 'example.com'),
       'certificate is for *.example.com')
 check(_verifycert(cert('*.example.com'), 'w.w.example.com'),
       'certificate is for *.example.com')
 
 # Test subjectAltName
 san_cert = {'subject': ((('commonName', 'example.com'),),),
             'subjectAltName': (('DNS', '*.example.net'),
                                ('DNS', 'example.net'))}
 check(_verifycert(san_cert, 'example.net'),
       None)
 check(_verifycert(san_cert, 'foo.example.net'),
       None)
 # no fallback to subject commonName when subjectAltName has DNS
 check(_verifycert(san_cert, 'example.com'),
       'certificate is for *.example.net, example.net')
 # fallback to subject commonName when no DNS in subjectAltName
 san_cert = {'subject': ((('commonName', 'example.com'),),),
             'subjectAltName': (('IP Address', '8.8.8.8'),)}
 check(_verifycert(san_cert, 'example.com'), None)
 
 # Avoid some pitfalls
 check(_verifycert(cert('*.foo'), 'foo'),
       'certificate is for *.foo')
 check(_verifycert(cert('*o'), 'foo'),
       'certificate is for *o')
 
 check(_verifycert({'subject': ()},
                   'example.com'),
       'no commonName or subjectAltName found in certificate')
 check(_verifycert(None, 'example.com'),
       'no certificate received')
 
 # Unicode (IDN) certname isn't supported
 check(_verifycert(cert(u'\u4f8b.jp'), 'example.jp'),
       'IDN in certificate not supported')
 
 import doctest
 
 def test_url():
     """
     >>> from mercurial.util import url
 
     This tests for edge cases in url.URL's parsing algorithm. Most of
     these aren't useful for documentation purposes, so they aren't
     part of the class's doc tests.
 
     Query strings and fragments:
 
     >>> url('http://host/a?b#c')
     <url scheme: 'http', host: 'host', path: 'a', query: 'b', fragment: 'c'>
     >>> url('http://host/a?')
     <url scheme: 'http', host: 'host', path: 'a'>
     >>> url('http://host/a#b#c')
     <url scheme: 'http', host: 'host', path: 'a', fragment: 'b#c'>
     >>> url('http://host/a#b?c')
     <url scheme: 'http', host: 'host', path: 'a', fragment: 'b?c'>
     >>> url('http://host/?a#b')
     <url scheme: 'http', host: 'host', path: '', query: 'a', fragment: 'b'>
     >>> url('http://host/?a#b', parsequery=False)
     <url scheme: 'http', host: 'host', path: '?a', fragment: 'b'>
     >>> url('http://host/?a#b', parsefragment=False)
     <url scheme: 'http', host: 'host', path: '', query: 'a#b'>
     >>> url('http://host/?a#b', parsequery=False, parsefragment=False)
     <url scheme: 'http', host: 'host', path: '?a#b'>
 
     IPv6 addresses:
 
     >>> url('ldap://[2001:db8::7]/c=GB?objectClass?one')
     <url scheme: 'ldap', host: '[2001:db8::7]', path: 'c=GB',
          query: 'objectClass?one'>
     >>> url('ldap://joe:xxx@[2001:db8::7]:80/c=GB?objectClass?one')
     <url scheme: 'ldap', user: 'joe', passwd: 'xxx', host: '[2001:db8::7]',
          port: '80', path: 'c=GB', query: 'objectClass?one'>
 
     Missing scheme, host, etc.:
 
     >>> url('://192.0.2.16:80/')
     <url path: '://192.0.2.16:80/'>
     >>> url('http://mercurial.selenic.com')
     <url scheme: 'http', host: 'mercurial.selenic.com'>
     >>> url('/foo')
     <url path: '/foo'>
     >>> url('bundle:/foo')
     <url scheme: 'bundle', path: '/foo'>
     >>> url('a?b#c')
     <url path: 'a?b', fragment: 'c'>
     >>> url('http://x.com?arg=/foo')
     <url scheme: 'http', host: 'x.com', query: 'arg=/foo'>
     >>> url('http://joe:xxx@/foo')
     <url scheme: 'http', user: 'joe', passwd: 'xxx', path: 'foo'>
 
     Just a scheme and a path:
 
     >>> url('mailto:[e-mail]')
     <url scheme: 'mailto', path: '[e-mail]'>
     >>> url('a:b:c:d')
     <url path: 'a:b:c:d'>
     >>> url('aa:bb:cc:dd')
     <url scheme: 'aa', path: 'bb:cc:dd'>
 
     SSH examples:
 
     >>> url('ssh://joe@host//home/joe')
     <url scheme: 'ssh', user: 'joe', host: 'host', path: '/home/joe'>
     >>> url('ssh://joe:xxx@host/src')
     <url scheme: 'ssh', user: 'joe', passwd: 'xxx', host: 'host', path: 'src'>
     >>> url('ssh://joe:xxx@host')
     <url scheme: 'ssh', user: 'joe', passwd: 'xxx', host: 'host'>
     >>> url('ssh://joe@host')
     <url scheme: 'ssh', user: 'joe', host: 'host'>
     >>> url('ssh://host')
     <url scheme: 'ssh', host: 'host'>
     >>> url('ssh://')
     <url scheme: 'ssh'>
     >>> url('ssh:')
     <url scheme: 'ssh'>
 
     Non-numeric port:
 
     >>> url('http://example.com:dd')
     <url scheme: 'http', host: 'example.com', port: 'dd'>
     >>> url('ssh://joe:xxx@host:ssh/foo')
     <url scheme: 'ssh', user: 'joe', passwd: 'xxx', host: 'host', port: 'ssh',
          path: 'foo'>
 
     Bad authentication credentials:
 
     >>> url('http://joe@joeville:123@4:@host/a?b#c')
     <url scheme: 'http', user: 'joe@joeville', passwd: '123@4:',
          host: 'host', path: 'a', query: 'b', fragment: 'c'>
     >>> url('http://!*#?/@!*#?/:@host/a?b#c')
     <url scheme: 'http', host: '!*', fragment: '?/@!*#?/:@host/a?b#c'>
     >>> url('http://!*#?@!*#?:@host/a?b#c')
     <url scheme: 'http', host: '!*', fragment: '?@!*#?:@host/a?b#c'>
     >>> url('http://!*@:!*@@host/a?b#c')
     <url scheme: 'http', user: '!*@', passwd: '!*@', host: 'host',
          path: 'a', query: 'b', fragment: 'c'>
 
     File paths:
 
     >>> url('a/b/c/d.g.f')
     <url path: 'a/b/c/d.g.f'>
     >>> url('/x///z/y/')
     <url path: '/x///z/y/'>
     >>> url('/foo:bar')
     <url path: '/foo:bar'>
     >>> url('\\\\foo:bar')
     <url path: '\\\\foo:bar'>
     >>> url('./foo:bar')
     <url path: './foo:bar'>
 
     Non-localhost file URL:
 
     >>> u = url('file://mercurial.selenic.com/foo')
     Traceback (most recent call last):
       File "<stdin>", line 1, in ?
     Abort: file:// URLs can only refer to localhost
 
     Empty URL:
 
     >>> u = url('')
     >>> u
     <url path: ''>
     >>> str(u)
     ''
 
     Empty path with query string:
 
     >>> str(url('http://foo/?bar'))
     'http://foo/?bar'
 
     Invalid path:
 
     >>> u = url('http://foo/bar')
     >>> u.path = 'bar'
     >>> str(u)
     'http://foo/bar'
 
     >>> u = url('file:/foo/bar/baz')
     >>> u
     <url scheme: 'file', path: '/foo/bar/baz'>
     >>> str(u)
     'file:///foo/bar/baz'
+    >>> u.localpath()
+    '/foo/bar/baz'
 
     >>> u = url('file:///foo/bar/baz')
     >>> u
     <url scheme: 'file', path: '/foo/bar/baz'>
     >>> str(u)
     'file:///foo/bar/baz'
+    >>> u.localpath()
+    '/foo/bar/baz'
+
+    >>> u = url('file:///f:oo/bar/baz')
+    >>> u
+    <url scheme: 'file', path: 'f:oo/bar/baz'>
+    >>> str(u)
+    'file:f%3Aoo/bar/baz'
+    >>> u.localpath()
+    'f:oo/bar/baz'
 
     >>> u = url('file:foo/bar/baz')
     >>> u
     <url scheme: 'file', path: 'foo/bar/baz'>
     >>> str(u)
     'file:foo/bar/baz'
+    >>> u.localpath()
+    'foo/bar/baz'
     """
 
 doctest.testmod(optionflags=doctest.NORMALIZE_WHITESPACE)