Merge with crew.

hgext/convert/darcs.py

@@ -0,0 +1,137 b''
+# darcs support for the convert extension
+
+from common import NoRepo, commit, converter_source
+from mercurial.i18n import _
+from mercurial import util
+import os, shutil, tempfile
+
+# The naming drift of ElementTree is fun!
+
+try: from xml.etree.cElementTree import ElementTree
+except ImportError:
+    try: from xml.etree.ElementTree import ElementTree
+    except ImportError:
+        try: from elementtree.cElementTree import ElementTree
+        except ImportError:
+            try: from elementtree.ElementTree import ElementTree
+            except ImportError: ElementTree = None
+
+
+class darcs_source(converter_source):
+    def __init__(self, ui, path, rev=None):
+        super(darcs_source, self).__init__(ui, path, rev=rev)
+
+        if not os.path.exists(os.path.join(path, '_darcs', 'inventory')):
+            raise NoRepo("couldn't open darcs repo %s" % path)
+
+        if ElementTree is None:
+            raise util.Abort(_("Python ElementTree module is not available"))
+
+        self.path = os.path.realpath(path)
+
+        self.lastrev = None
+        self.changes = {}
+        self.parents = {}
+        self.tags = {}
+
+    def before(self):
+        self.tmppath = tempfile.mkdtemp(
+            prefix='convert-' + os.path.basename(self.path) + '-')
+        output, status = self.run('init', repodir=self.tmppath)
+        self.checkexit(status)
+
+        tree = self.xml('changes', '--xml-output', '--summary')
+        tagname = None
+        child = None
+        for elt in tree.findall('patch'):
+            node = elt.get('hash')
+            name = elt.findtext('name', '')
+            if name.startswith('TAG '):
+                tagname = name[4:].strip()
+            elif tagname is not None:
+                self.tags[tagname] = node
+                tagname = None
+            self.changes[node] = elt
+            self.parents[child] = [node]
+            child = node
+        self.parents[child] = []
+
+    def after(self):
+        self.ui.debug('cleaning up %s\n' % self.tmppath)
+        #shutil.rmtree(self.tmppath, ignore_errors=True)
+
+    def _run(self, cmd, *args, **kwargs):
+        cmdline = 'darcs %s --repodir=%r %s </dev/null' % (
+            cmd, kwargs.get('repodir', self.path), ' '.join(args))
+        self.ui.debug(cmdline, '\n')
+        return os.popen(cmdline, 'r')
+
+    def run(self, cmd, *args, **kwargs):
+        fp = self._run(cmd, *args, **kwargs)
+        output = fp.read()
+        return output, fp.close()
+
+    def checkexit(self, status, output=''):
+        if status:
+            if output:
+                ui.warn(_('darcs error:\n'))
+                ui.warn(output)
+            msg = util.explain_exit(status)[0]
+            raise util.Abort(_('darcs %s') % msg)
+        
+    def xml(self, cmd, *opts):
+        etree = ElementTree()
+        fp = self._run(cmd, *opts)
+        etree.parse(fp)
+        self.checkexit(fp.close())
+        return etree.getroot()
+
+    def getheads(self):
+        return self.parents[None]
+
+    def getcommit(self, rev):
+        elt = self.changes[rev]
+        date = util.strdate(elt.get('local_date'), '%a %b %d %H:%M:%S %Z %Y')
+        desc = elt.findtext('name') + '\n' + elt.findtext('comment', '')
+        return commit(author=elt.get('author'), date=util.datestr(date),
+                      desc=desc.strip(), parents=self.parents[rev])
+
+    def pull(self, rev):
+        output, status = self.run('pull %r --all --match="hash %s"' %
+                                  (self.path, rev),
+                                  '--no-test', '--no-posthook',
+                                  '--external-merge=/bin/false',
+                                  repodir=self.tmppath)
+        if status:
+            if output.find('We have conflicts in') == -1:
+                self.checkexit(status, output)
+            output, status = self.run('revert --all', repodir=self.tmppath)
+            self.checkexit(status, output)
+
+    def getchanges(self, rev):
+        self.pull(rev)
+        copies = {}
+        changes = []
+        for elt in self.changes[rev].find('summary').getchildren():
+            if elt.tag in ('add_directory', 'remove_directory'):
+                continue
+            if elt.tag == 'move':
+                changes.append((elt.get('from'), rev))
+                copies[elt.get('from')] = elt.get('to')
+            else:
+                changes.append((elt.text.strip(), rev))
+        changes.sort()
+        self.lastrev = rev
+        return changes, copies
+
+    def getfile(self, name, rev):
+        if rev != self.lastrev:
+            raise util.Abort(_('internal calling inconsistency'))
+        return open(os.path.join(self.tmppath, name), 'rb').read()
+
+    def getmode(self, name, rev):
+        mode = os.lstat(os.path.join(self.tmppath, name)).st_mode
+        return (mode & 0111) and 'x' or ''
+
+    def gettags(self):
+        return self.tags

contrib/darcs2hg.py

@@ -1,251 +1,255 b''
 #!/usr/bin/env python
 # Encoding: iso-8859-1
 # vim: tw=80 ts=4 sw=4 noet
 # -----------------------------------------------------------------------------
 # Project   : Basic Darcs to Mercurial conversion script
+#
+# *** DEPRECATED. Use the convert extension instead. This script will
+# *** be removed soon.
+#
 # -----------------------------------------------------------------------------
 # Authors   : Sebastien Pierre                           <[e-mail]>
 #             TK Soh                                      <[e-mail]>
 # -----------------------------------------------------------------------------
 # Creation  : 24-May-2006
-# Last mod  : 05-Jun-2006
 # -----------------------------------------------------------------------------
 
 import os, sys
 import tempfile
 import xml.dom.minidom as xml_dom
 from time import strptime, mktime
 import re
 
 DARCS_REPO = None
 HG_REPO    = None
 
 USAGE = """\
 %s DARCSREPO HGREPO [SKIP]
 
     Converts the given Darcs repository to a new Mercurial repository. The given
     HGREPO must not exist, as it will be created and filled up (this will avoid
     overwriting valuable data.
 
     In case an error occurs within the process, you can resume the process by
     giving the last successfuly applied change number.
 """ % (os.path.basename(sys.argv[0]))
 
 # ------------------------------------------------------------------------------
 #
 # Utilities
 #
 # ------------------------------------------------------------------------------
 
 def cmd(text, path=None, silent=False):
 	"""Executes a command, in the given directory (if any), and returns the
 	command result as a string."""
 	cwd = None
 	if path:
 		path = os.path.abspath(path)
 		cwd  = os.getcwd()
 		os.chdir(path)
 	if not silent: print "> ", text
 	res = os.popen(text).read()
 	if path:
 		os.chdir(cwd)
 	return res
 
 def writefile(path, data):
 	"""Writes the given data into the given file."""
 	f = file(path, "w") ; f.write(data)  ; f.close()
 
 def error( *args ):
 	sys.stderr.write("ERROR: ")
 	for a in args: sys.stderr.write(str(a))
 	sys.stderr.write("\n")
 	sys.stderr.write("You can make manual fixes if necessary and then resume by"
 	" giving the last changeset number")
 	sys.exit(-1)
 
 # ------------------------------------------------------------------------------
 #
 # Darcs interface
 #
 # ------------------------------------------------------------------------------
 
 def darcs_changes(darcsRepo):
 	"""Gets the changes list from the given darcs repository. This returns the
 	chronological list of changes as (change name, change summary)."""
 	changes    = cmd("darcs changes --reverse --xml-output", darcsRepo)
 	doc        = xml_dom.parseString(changes)
 	for patch_node in doc.childNodes[0].childNodes:
 		name = filter(lambda n: n.nodeName == "name", patch_node.childNodes)
 		comm = filter(lambda n: n.nodeName == "comment", patch_node.childNodes)
 		if not name:continue
 		else: name = name[0].childNodes[0].data
 		if not comm: comm = ""
 		else: comm = comm[0].childNodes[0].data
 		author = patch_node.getAttribute("author")
 		date   = patch_node.getAttribute("date")
 		chash  = os.path.splitext(patch_node.getAttribute("hash"))[0]
 		yield author, date, name, chash, comm
 
 def darcs_tip(darcs_repo):
 	changes = cmd("darcs changes",darcs_repo,silent=True)
 	changes = filter(lambda l: l.strip().startswith("* "), changes.split("\n"))
 	return len(changes)
 
 def darcs_pull(hg_repo, darcs_repo, chash):
 	old_tip = darcs_tip(darcs_repo)
 	res     = cmd("darcs pull \"%s\" --all --match=\"hash %s\"" % (darcs_repo, chash), hg_repo)
 	if re.search('^We have conflicts in the following files:$', res, re.MULTILINE):
 		print "Trying to revert files to work around conflict..."
 		rev_res = cmd ("darcs revert --all", hg_repo)
 		print rev_res
 	print res
 	new_tip = darcs_tip(darcs_repo)
 	if not new_tip != old_tip + 1:
 		error("Darcs pull did not work as expected: " + res)
 
 def darcs_changes_summary(darcs_repo, chash):
 	"""Gets the changes from the darcs summary. This returns the chronological
 	list of changes as (change_type, args). Eg. ('add_file', 'foo.txt') or
 	('move', ['foo.txt','bar.txt'])."""
 	change = cmd("darcs changes --summary --xml-output --match=\"hash %s\"" % (chash), darcs_repo)
 	doc = xml_dom.parseString(change)
 	for patch_node in doc.childNodes[0].childNodes:
 		summary_nodes = filter(lambda n: n.nodeName == "summary" and n.nodeType == n.ELEMENT_NODE, patch_node.childNodes)
 		for summary_node in summary_nodes:
 			change_nodes = filter(lambda n: n.nodeType == n.ELEMENT_NODE, summary_node.childNodes)
 			if len(change_nodes) == 0:
 				name = filter(lambda n: n.nodeName == "name", patch_node.childNodes)
 				if not name:
 					error("Darcs patch has an empty summary node and no name: " + patch_node.toxml())
 				name = name[0].childNodes[0].data.strip()
 				(tag, sub_count) = re.subn('^TAG ', '', name, 1)
 				if sub_count != 1:
 					error("Darcs patch has an empty summary node but doesn't look like a tag: " + patch_node.toxml());
 			for change_node in change_nodes:
 				change = change_node.nodeName
 				if change == 'modify_file':
 					yield change, change_node.childNodes[0].data.strip()
 				elif change == 'add_file':
 					yield change, change_node.childNodes[0].data.strip()
 				elif change == 'remove_file':
 					yield change, change_node.childNodes[0].data.strip()
 				elif change == 'add_directory':
 					yield change, change_node.childNodes[0].data.strip()
 				elif change == 'remove_directory':
 					yield change, change_node.childNodes[0].data.strip()
 				elif change == 'move':
 					yield change, (change_node.getAttribute('from'), change_node.getAttribute('to'))
 				else:
 					error('Problem parsing summary xml: Unexpected element: ' + change_node.toxml())
 
 # ------------------------------------------------------------------------------
 #
 # Mercurial interface
 #
 # ------------------------------------------------------------------------------
 
 def hg_commit( hg_repo, text, author, date ):
 	fd, tmpfile = tempfile.mkstemp(prefix="darcs2hg_")
 	writefile(tmpfile, text)
 	old_tip = hg_tip(hg_repo)
 	cmd("hg add -X _darcs", hg_repo)
 	cmd("hg remove -X _darcs --after", hg_repo)
 	res = cmd("hg commit -l %s -u \"%s\" -d \"%s 0\""  % (tmpfile, author, date), hg_repo)
 	os.close(fd)
 	os.unlink(tmpfile)
 	new_tip = hg_tip(hg_repo)
 	if not new_tip == old_tip + 1:
 		# Sometimes we may have empty commits, we simply skip them
 		if res.strip().lower().find("nothing changed") != -1:
 			pass
 		else:
 			error("Mercurial commit did not work as expected: " + res)
 
 def hg_tip( hg_repo ):
 	"""Returns the latest local revision number in the given repository."""
 	tip = cmd("hg tip", hg_repo, silent=True)
 	tip = tip.split("\n")[0].split(":")[1].strip()
 	return int(tip)
 
 def hg_rename( hg_repo, from_file, to_file ):
 	cmd("hg rename --after \"%s\" \"%s\"" % (from_file, to_file), hg_repo);
 	
 def hg_tag ( hg_repo, text, author, date ):
 	old_tip = hg_tip(hg_repo)
 	res = cmd("hg tag -u \"%s\" -d \"%s 0\" \"%s\""	 % (author, date, text), hg_repo)
 	new_tip = hg_tip(hg_repo)
 	if not new_tip == old_tip + 1:
 		error("Mercurial tag did not work as expected: " + res)
 
 def hg_handle_change( hg_repo, author, date, change, arg ):
 	"""Processes a change event as output by darcs_changes_summary. These
 	consist of file move/rename/add/delete commands."""
 	if change == 'modify_file':
 		pass
 	elif change == 'add_file':
 		pass
 	elif change =='remove_file':
 		pass
 	elif change == 'add_directory':
 		pass
 	elif change == 'remove_directory':
 		pass
 	elif change == 'move':
 		hg_rename(hg_repo, arg[0], arg[1])
 	elif change == 'tag':
 		hg_tag(hg_repo, arg, author, date)
 	else:
 		error('Unknown change type ' + change + ': ' + arg)
 
 # ------------------------------------------------------------------------------
 #
 # Main
 #
 # ------------------------------------------------------------------------------
 
 if __name__ == "__main__":
 	args = sys.argv[1:]
 	# We parse the arguments
 	if len(args)   == 2:
 		darcs_repo = os.path.abspath(args[0])
 		hg_repo    = os.path.abspath(args[1])
 		skip       = None
 	elif len(args) == 3:
 		darcs_repo = os.path.abspath(args[0])
 		hg_repo    = os.path.abspath(args[1])
 		skip       = int(args[2])
 	else:
 		print USAGE
 		sys.exit(-1)
+	print 'This command is deprecated.  Use the convert extension instead.'
 	# Initializes the target repo
 	if not os.path.isdir(darcs_repo + "/_darcs"):
 		print "No darcs directory found at: " + darcs_repo
 		sys.exit(-1)
 	if not os.path.isdir(hg_repo):
 		os.mkdir(hg_repo)
 	elif skip == None:
 		print "Given HG repository must not exist when no SKIP is specified."
 		sys.exit(-1)
 	if skip == None:
 		cmd("hg init \"%s\"" % (hg_repo))
 		cmd("darcs initialize", hg_repo)
 	# Get the changes from the Darcs repository
 	change_number = 0
 	for author, date, summary, chash, description in darcs_changes(darcs_repo):
 		print "== changeset", change_number,
 		if skip != None and change_number <= skip:
 			print "(skipping)"
 		else:
 			text = summary + "\n" + description
 			# The commit hash has a date like 20021020201112
 			# --------------------------------YYYYMMDDHHMMSS
 			date = chash.split("-")[0]
 			epoch = int(mktime(strptime(date, '%Y%m%d%H%M%S')))
 			darcs_pull(hg_repo, darcs_repo, chash)
 			for change, arg in darcs_changes_summary(darcs_repo, chash):
 				hg_handle_change(hg_repo, author, epoch, change, arg)
 			hg_commit(hg_repo, text, author, epoch)
 		change_number += 1
 	print "Darcs repository (_darcs) was not deleted. You can keep or remove it."
 
 # EOF

hgext/convert/__init__.py

@@ -1,483 +1,489 b''
 # convert.py Foreign SCM converter
 #
 # Copyright 2005-2007 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms
 # of the GNU General Public License, incorporated herein by reference.
 
 from common import NoRepo, converter_source, converter_sink
 from cvs import convert_cvs
+from darcs import darcs_source
 from git import convert_git
 from hg import mercurial_source, mercurial_sink
 from subversion import convert_svn, debugsvnlog
 
 import os, shlex, shutil
 from mercurial import hg, ui, util, commands
 from mercurial.i18n import _
 
 commands.norepo += " convert debugsvnlog"
 
 converters = [convert_cvs, convert_git, convert_svn, mercurial_source,
-              mercurial_sink]
+              mercurial_sink, darcs_source]
 
 def convertsource(ui, path, **opts):
     for c in converters:
         try:
             return c.getcommit and c(ui, path, **opts)
         except (AttributeError, NoRepo):
             pass
     raise util.Abort('%s: unknown repository type' % path)
 
 def convertsink(ui, path):
     if not os.path.isdir(path):
         raise util.Abort("%s: not a directory" % path)
     for c in converters:
         try:
             return c.putcommit and c(ui, path)
         except (AttributeError, NoRepo):
             pass
     raise util.Abort('%s: unknown repository type' % path)
 
 class converter(object):
     def __init__(self, ui, source, dest, revmapfile, filemapper, opts):
 
         self.source = source
         self.dest = dest
         self.ui = ui
         self.opts = opts
         self.commitcache = {}
         self.revmapfile = revmapfile
         self.revmapfilefd = None
         self.authors = {}
         self.authorfile = None
         self.mapfile = filemapper
 
         self.map = {}
         try:
             origrevmapfile = open(self.revmapfile, 'r')
             for l in origrevmapfile:
                 sv, dv = l[:-1].split()
                 self.map[sv] = dv
             origrevmapfile.close()
         except IOError:
             pass
 
         # Read first the dst author map if any
         authorfile = self.dest.authorfile()
         if authorfile and os.path.exists(authorfile):
             self.readauthormap(authorfile)
         # Extend/Override with new author map if necessary
         if opts.get('authors'):
             self.readauthormap(opts.get('authors'))
             self.authorfile = self.dest.authorfile()
 
     def walktree(self, heads):
         '''Return a mapping that identifies the uncommitted parents of every
         uncommitted changeset.'''
         visit = heads
         known = {}
         parents = {}
         while visit:
             n = visit.pop(0)
             if n in known or n in self.map: continue
             known[n] = 1
             commit = self.cachecommit(n)
             parents[n] = []
             for p in commit.parents:
                 parents[n].append(p)
                 visit.append(p)
 
         return parents
 
     def toposort(self, parents):
         '''Return an ordering such that every uncommitted changeset is
         preceeded by all its uncommitted ancestors.'''
         visit = parents.keys()
         seen = {}
         children = {}
 
         while visit:
             n = visit.pop(0)
             if n in seen: continue
             seen[n] = 1
             # Ensure that nodes without parents are present in the 'children'
             # mapping.
             children.setdefault(n, [])
             for p in parents[n]:
                 if not p in self.map:
                     visit.append(p)
                 children.setdefault(p, []).append(n)
 
         s = []
         removed = {}
         visit = children.keys()
         while visit:
             n = visit.pop(0)
             if n in removed: continue
             dep = 0
             if n in parents:
                 for p in parents[n]:
                     if p in self.map: continue
                     if p not in removed:
                         # we're still dependent
                         visit.append(n)
                         dep = 1
                         break
 
             if not dep:
                 # all n's parents are in the list
                 removed[n] = 1
                 if n not in self.map:
                     s.append(n)
                 if n in children:
                     for c in children[n]:
                         visit.insert(0, c)
 
         if self.opts.get('datesort'):
             depth = {}
             for n in s:
                 depth[n] = 0
                 pl = [p for p in self.commitcache[n].parents
                       if p not in self.map]
                 if pl:
                     depth[n] = max([depth[p] for p in pl]) + 1
 
             s = [(depth[n], self.commitcache[n].date, n) for n in s]
             s.sort()
             s = [e[2] for e in s]
 
         return s
 
     def mapentry(self, src, dst):
         if self.revmapfilefd is None:
             try:
                 self.revmapfilefd = open(self.revmapfile, "a")
             except IOError, (errno, strerror):
                 raise util.Abort("Could not open map file %s: %s, %s\n" % (self.revmapfile, errno, strerror))
         self.map[src] = dst
         self.revmapfilefd.write("%s %s\n" % (src, dst))
         self.revmapfilefd.flush()
 
     def writeauthormap(self):
         authorfile = self.authorfile
         if authorfile:
            self.ui.status('Writing author map file %s\n' % authorfile)
            ofile = open(authorfile, 'w+')
            for author in self.authors:
                ofile.write("%s=%s\n" % (author, self.authors[author]))
            ofile.close()
 
     def readauthormap(self, authorfile):
         afile = open(authorfile, 'r')
         for line in afile:
             try:
                 srcauthor = line.split('=')[0].strip()
                 dstauthor = line.split('=')[1].strip()
                 if srcauthor in self.authors and dstauthor != self.authors[srcauthor]:
                     self.ui.status(
                         'Overriding mapping for author %s, was %s, will be %s\n'
                         % (srcauthor, self.authors[srcauthor], dstauthor))
                 else:
                     self.ui.debug('Mapping author %s to %s\n'
                                   % (srcauthor, dstauthor))
                     self.authors[srcauthor] = dstauthor
             except IndexError:
                 self.ui.warn(
                     'Ignoring bad line in author file map %s: %s\n'
                     % (authorfile, line))
         afile.close()
 
     def cachecommit(self, rev):
         commit = self.source.getcommit(rev)
         commit.author = self.authors.get(commit.author, commit.author)
         self.commitcache[rev] = commit
         return commit
 
     def copy(self, rev):
         commit = self.commitcache[rev]
         do_copies = hasattr(self.dest, 'copyfile')
         filenames = []
 
         files, copies = self.source.getchanges(rev)
         parents = [self.map[r] for r in commit.parents]
         if commit.parents:
             prev = commit.parents[0]
             if prev not in self.commitcache:
                 self.cachecommit(prev)
             pbranch = self.commitcache[prev].branch
         else:
             pbranch = None
         self.dest.setbranch(commit.branch, pbranch, parents)
         for f, v in files:
             newf = self.mapfile(f)
             if not newf:
                 continue
             filenames.append(newf)
             try:
                 data = self.source.getfile(f, v)
             except IOError, inst:
                 self.dest.delfile(newf)
             else:
                 e = self.source.getmode(f, v)
                 self.dest.putfile(newf, e, data)
                 if do_copies:
                     if f in copies:
                         copyf = self.mapfile(copies[f])
                         if copyf:
                             # Merely marks that a copy happened.
                             self.dest.copyfile(copyf, newf)
 
         if not filenames and self.mapfile.active():
             newnode = parents[0]
         else:
             newnode = self.dest.putcommit(filenames, parents, commit)
         self.mapentry(rev, newnode)
 
     def convert(self):
         try:
+            self.source.before()
             self.dest.before()
             self.source.setrevmap(self.map)
             self.ui.status("scanning source...\n")
             heads = self.source.getheads()
             parents = self.walktree(heads)
             self.ui.status("sorting...\n")
             t = self.toposort(parents)
             num = len(t)
             c = None
 
             self.ui.status("converting...\n")
             for c in t:
                 num -= 1
                 desc = self.commitcache[c].desc
                 if "\n" in desc:
                     desc = desc.splitlines()[0]
                 self.ui.status("%d %s\n" % (num, desc))
                 self.copy(c)
 
             tags = self.source.gettags()
             ctags = {}
             for k in tags:
                 v = tags[k]
                 if v in self.map:
                     ctags[k] = self.map[v]
 
             if c and ctags:
                 nrev = self.dest.puttags(ctags)
                 # write another hash correspondence to override the previous
                 # one so we don't end up with extra tag heads
                 if nrev:
                     self.mapentry(c, nrev)
 
             self.writeauthormap()
         finally:
             self.cleanup()
 
     def cleanup(self):
-        self.dest.after()
+        try:
+            self.dest.after()
+        finally:
+            self.source.after()
         if self.revmapfilefd:
             self.revmapfilefd.close()
 
 def rpairs(name):
     e = len(name)
     while e != -1:
         yield name[:e], name[e+1:]
         e = name.rfind('/', 0, e)
 
 class filemapper(object):
     '''Map and filter filenames when importing.
     A name can be mapped to itself, a new name, or None (omit from new
     repository).'''
 
     def __init__(self, ui, path=None):
         self.ui = ui
         self.include = {}
         self.exclude = {}
         self.rename = {}
         if path:
             if self.parse(path):
                 raise util.Abort(_('errors in filemap'))
 
     def parse(self, path):
         errs = 0
         def check(name, mapping, listname):
             if name in mapping:
                 self.ui.warn(_('%s:%d: %r already in %s list\n') %
                              (lex.infile, lex.lineno, name, listname))
                 return 1
             return 0
         lex = shlex.shlex(open(path), path, True)
         lex.wordchars += '!@#$%^&*()-=+[]{}|;:,./<>?'
         cmd = lex.get_token()
         while cmd:
             if cmd == 'include':
                 name = lex.get_token()
                 errs += check(name, self.exclude, 'exclude')
                 self.include[name] = name
             elif cmd == 'exclude':
                 name = lex.get_token()
                 errs += check(name, self.include, 'include')
                 errs += check(name, self.rename, 'rename')
                 self.exclude[name] = name
             elif cmd == 'rename':
                 src = lex.get_token()
                 dest = lex.get_token()
                 errs += check(src, self.exclude, 'exclude')
                 self.rename[src] = dest
             elif cmd == 'source':
                 errs += self.parse(lex.get_token())
             else:
                 self.ui.warn(_('%s:%d: unknown directive %r\n') %
                              (lex.infile, lex.lineno, cmd))
                 errs += 1
             cmd = lex.get_token()
         return errs
 
     def lookup(self, name, mapping):
         for pre, suf in rpairs(name):
             try:
                 return mapping[pre], pre, suf
             except KeyError, err:
                 pass
         return '', name, ''
 
     def __call__(self, name):
         if self.include:
             inc = self.lookup(name, self.include)[0]
         else:
             inc = name
         if self.exclude:
             exc = self.lookup(name, self.exclude)[0]
         else:
             exc = ''
         if not inc or exc:
             return None
         newpre, pre, suf = self.lookup(name, self.rename)
         if newpre:
             if newpre == '.':
                 return suf
             if suf:
                 return newpre + '/' + suf
             return newpre
         return name
 
     def active(self):
         return bool(self.include or self.exclude or self.rename)
 
 def convert(ui, src, dest=None, revmapfile=None, **opts):
     """Convert a foreign SCM repository to a Mercurial one.
 
     Accepted source formats:
-    - GIT
     - CVS
-    - SVN
+    - Darcs
+    - git
+    - Subversion
 
     Accepted destination formats:
     - Mercurial
 
     If no revision is given, all revisions will be converted. Otherwise,
     convert will only import up to the named revision (given in a format
     understood by the source).
 
     If no destination directory name is specified, it defaults to the
     basename of the source with '-hg' appended.  If the destination
     repository doesn't exist, it will be created.
 
     If <revmapfile> isn't given, it will be put in a default location
     (<dest>/.hg/shamap by default).  The <revmapfile> is a simple text
     file that maps each source commit ID to the destination ID for
     that revision, like so:
     <source ID> <destination ID>
 
     If the file doesn't exist, it's automatically created.  It's updated
     on each commit copied, so convert-repo can be interrupted and can
     be run repeatedly to copy new commits.
 
     The [username mapping] file is a simple text file that maps each source
     commit author to a destination commit author. It is handy for source SCMs
     that use unix logins to identify authors (eg: CVS). One line per author
     mapping and the line format is:
     srcauthor=whatever string you want
 
     The filemap is a file that allows filtering and remapping of files
     and directories.  Comment lines start with '#'.  Each line can
     contain one of the following directives:
 
       include path/to/file
 
       exclude path/to/file
 
       rename from/file to/file
     
     The 'include' directive causes a file, or all files under a
     directory, to be included in the destination repository.  The
     'exclude' directive causes files or directories to be omitted.
     The 'rename' directive renames a file or directory.  To rename
     from a subdirectory into the root of the repository, use '.' as
     the path to rename to.
     """
 
     util._encoding = 'UTF-8'
 
     if not dest:
         dest = hg.defaultdest(src) + "-hg"
         ui.status("assuming destination %s\n" % dest)
 
     # Try to be smart and initalize things when required
     created = False
     if os.path.isdir(dest):
         if len(os.listdir(dest)) > 0:
             try:
                 hg.repository(ui, dest)
                 ui.status("destination %s is a Mercurial repository\n" % dest)
             except hg.RepoError:
                 raise util.Abort(
                     "destination directory %s is not empty.\n"
                     "Please specify an empty directory to be initialized\n"
                     "or an already initialized mercurial repository"
                     % dest)
         else:
             ui.status("initializing destination %s repository\n" % dest)
             hg.repository(ui, dest, create=True)
             created = True
     elif os.path.exists(dest):
         raise util.Abort("destination %s exists and is not a directory" % dest)
     else:
         ui.status("initializing destination %s repository\n" % dest)
         hg.repository(ui, dest, create=True)
         created = True
 
     destc = convertsink(ui, dest)
 
     try:
         srcc = convertsource(ui, src, rev=opts.get('rev'))
     except Exception:
         if created:
             shutil.rmtree(dest, True)
         raise
 
     if not revmapfile:
         try:
             revmapfile = destc.revmapfile()
         except:
             revmapfile = os.path.join(destc, "map")
 
 
     c = converter(ui, srcc, destc, revmapfile, filemapper(ui, opts['filemap']),
                   opts)
     c.convert()
 
 
 cmdtable = {
     "convert":
         (convert,
          [('A', 'authors', '', 'username mapping filename'),
           ('', 'filemap', '', 'remap file names using contents of file'),
           ('r', 'rev', '', 'import up to target revision REV'),
           ('', 'datesort', None, 'try to sort changesets by date')],
          'hg convert [OPTION]... SOURCE [DEST [MAPFILE]]'),
     "debugsvnlog":
         (debugsvnlog,
          [],
          'hg debugsvnlog'),
 }
 

hgext/convert/common.py

@@ -1,145 +1,151 b''
 # common code for the convert extension
 import base64
 import cPickle as pickle
 
 def encodeargs(args):
     def encodearg(s):
         lines = base64.encodestring(s)
         lines = [l.splitlines()[0] for l in lines]
         return ''.join(lines)
 
     s = pickle.dumps(args)
     return encodearg(s)
 
 def decodeargs(s):
     s = base64.decodestring(s)
     return pickle.loads(s)
 
 class NoRepo(Exception): pass
 
 class commit(object):
     def __init__(self, author, date, desc, parents, branch=None, rev=None):
         self.author = author
         self.date = date
         self.desc = desc
         self.parents = parents
         self.branch = branch
         self.rev = rev
 
 class converter_source(object):
     """Conversion source interface"""
 
     def __init__(self, ui, path, rev=None):
         """Initialize conversion source (or raise NoRepo("message")
         exception if path is not a valid repository)"""
         self.ui = ui
         self.path = path
         self.rev = rev
 
         self.encoding = 'utf-8'
 
+    def before(self):
+        pass
+
+    def after(self):
+        pass
+
     def setrevmap(self, revmap):
         """set the map of already-converted revisions"""
         pass
 
     def getheads(self):
         """Return a list of this repository's heads"""
         raise NotImplementedError()
 
     def getfile(self, name, rev):
         """Return file contents as a string"""
         raise NotImplementedError()
 
     def getmode(self, name, rev):
         """Return file mode, eg. '', 'x', or 'l'"""
         raise NotImplementedError()
 
     def getchanges(self, version):
         """Returns a tuple of (files, copies)
         Files is a sorted list of (filename, id) tuples for all files changed
         in version, where id is the source revision id of the file.
 
         copies is a dictionary of dest: source
         """
         raise NotImplementedError()
 
     def getcommit(self, version):
         """Return the commit object for version"""
         raise NotImplementedError()
 
     def gettags(self):
         """Return the tags as a dictionary of name: revision"""
         raise NotImplementedError()
 
     def recode(self, s, encoding=None):
         if not encoding:
             encoding = self.encoding or 'utf-8'
 
         if isinstance(s, unicode):
             return s.encode("utf-8")
         try:
             return s.decode(encoding).encode("utf-8")
         except:
             try:
                 return s.decode("latin-1").encode("utf-8")
             except:
                 return s.decode(encoding, "replace").encode("utf-8")
 
 class converter_sink(object):
     """Conversion sink (target) interface"""
 
     def __init__(self, ui, path):
         """Initialize conversion sink (or raise NoRepo("message")
         exception if path is not a valid repository)"""
         raise NotImplementedError()
 
     def getheads(self):
         """Return a list of this repository's heads"""
         raise NotImplementedError()
 
     def revmapfile(self):
         """Path to a file that will contain lines
         source_rev_id sink_rev_id
         mapping equivalent revision identifiers for each system."""
         raise NotImplementedError()
 
     def authorfile(self):
         """Path to a file that will contain lines
         srcauthor=dstauthor
         mapping equivalent authors identifiers for each system."""
         return None
 
     def putfile(self, f, e, data):
         """Put file for next putcommit().
         f: path to file
         e: '', 'x', or 'l' (regular file, executable, or symlink)
         data: file contents"""
         raise NotImplementedError()
 
     def delfile(self, f):
         """Delete file for next putcommit().
         f: path to file"""
         raise NotImplementedError()
 
     def putcommit(self, files, parents, commit):
         """Create a revision with all changed files listed in 'files'
         and having listed parents. 'commit' is a commit object containing
         at a minimum the author, date, and message for this changeset.
         Called after putfile() and delfile() calls. Note that the sink
         repository is not told to update itself to a particular revision
         (or even what that revision would be) before it receives the
         file data."""
         raise NotImplementedError()
 
     def puttags(self, tags):
         """Put tags into sink.
         tags: {tagname: sink_rev_id, ...}"""
         raise NotImplementedError()
 
     def setbranch(self, branch, pbranch, parents):
         """Set the current branch name. Called before the first putfile
         on the branch.
         branch: branch name for subsequent commits
         pbranch: branch name of parent commit
         parents: destination revisions of parent"""
         pass

hgext/convert/hg.py

@@ -1,210 +1,213 b''
 # hg backend for convert extension
 
 # Note for hg->hg conversion: Old versions of Mercurial didn't trim
 # the whitespace from the ends of commit messages, but new versions
 # do.  Changesets created by those older versions, then converted, may
 # thus have different hashes for changesets that are otherwise
 # identical.
 
 
 import os, time
 from mercurial.i18n import _
 from mercurial.node import *
 from mercurial import hg, lock, revlog, util
 
 from common import NoRepo, commit, converter_source, converter_sink
 
 class mercurial_sink(converter_sink):
     def __init__(self, ui, path):
         self.path = path
         self.ui = ui
         self.branchnames = ui.configbool('convert', 'hg.usebranchnames', True)
         self.clonebranches = ui.configbool('convert', 'hg.clonebranches', False)
         self.tagsbranch = ui.config('convert', 'hg.tagsbranch', 'default')
         self.lastbranch = None
         try:
             self.repo = hg.repository(self.ui, path)
         except:
             raise NoRepo("could not open hg repo %s as sink" % path)
         self.lock = None
         self.wlock = None
 
     def before(self):
         self.wlock = self.repo.wlock()
         self.lock = self.repo.lock()
         self.repo.dirstate.clear()
 
     def after(self):
         self.repo.dirstate.invalidate()
         self.lock = None
         self.wlock = None
 
     def revmapfile(self):
         return os.path.join(self.path, ".hg", "shamap")
 
     def authorfile(self):
         return os.path.join(self.path, ".hg", "authormap")
 
     def getheads(self):
         h = self.repo.changelog.heads()
         return [ hex(x) for x in h ]
 
     def putfile(self, f, e, data):
         self.repo.wwrite(f, data, e)
         if f not in self.repo.dirstate:
             self.repo.dirstate.normallookup(f)
 
     def copyfile(self, source, dest):
         self.repo.copy(source, dest)
 
     def delfile(self, f):
         try:
             util.unlink(self.repo.wjoin(f))
             #self.repo.remove([f])
         except OSError:
             pass
 
     def setbranch(self, branch, pbranch, parents):
         if (not self.clonebranches) or (branch == self.lastbranch):
             return
 
         self.lastbranch = branch
         self.after()
         if not branch:
             branch = 'default'
         if not pbranch:
             pbranch = 'default'
 
         branchpath = os.path.join(self.path, branch)
         try:
             self.repo = hg.repository(self.ui, branchpath)
         except:
             if not parents:
                 self.repo = hg.repository(self.ui, branchpath, create=True)
             else:
                 self.ui.note(_('cloning branch %s to %s\n') % (pbranch, branch))
                 hg.clone(self.ui, os.path.join(self.path, pbranch),
                          branchpath, rev=parents, update=False,
                          stream=True)
                 self.repo = hg.repository(self.ui, branchpath)
 
     def putcommit(self, files, parents, commit):
         seen = {}
         pl = []
         for p in parents:
             if p not in seen:
                 pl.append(p)
                 seen[p] = 1
         parents = pl
 
         if len(parents) < 2: parents.append("0" * 40)
         if len(parents) < 2: parents.append("0" * 40)
         p2 = parents.pop(0)
 
         text = commit.desc
         extra = {}
         if self.branchnames and commit.branch:
             extra['branch'] = commit.branch
         if commit.rev:
             extra['convert_revision'] = commit.rev
 
         while parents:
             p1 = p2
             p2 = parents.pop(0)
             a = self.repo.rawcommit(files, text, commit.author, commit.date,
                                     bin(p1), bin(p2), extra=extra)
             self.repo.dirstate.clear()
             text = "(octopus merge fixup)\n"
             p2 = hg.hex(self.repo.changelog.tip())
 
         return p2
 
     def puttags(self, tags):
         try:
             old = self.repo.wfile(".hgtags").read()
             oldlines = old.splitlines(1)
             oldlines.sort()
         except:
             oldlines = []
 
         k = tags.keys()
         k.sort()
         newlines = []
         for tag in k:
             newlines.append("%s %s\n" % (tags[tag], tag))
 
         newlines.sort()
 
         if newlines != oldlines:
             self.ui.status("updating tags\n")
             f = self.repo.wfile(".hgtags", "w")
             f.write("".join(newlines))
             f.close()
             if not oldlines: self.repo.add([".hgtags"])
             date = "%s 0" % int(time.mktime(time.gmtime()))
             extra = {}
             if self.tagsbranch != 'default':
                 extra['branch'] = self.tagsbranch
             try:
                 tagparent = self.repo.changectx(self.tagsbranch).node()
             except hg.RepoError, inst:
                 tagparent = nullid
             self.repo.rawcommit([".hgtags"], "update tags", "convert-repo",
                                 date, tagparent, nullid)
             return hex(self.repo.changelog.tip())
 
 class mercurial_source(converter_source):
     def __init__(self, ui, path, rev=None):
         converter_source.__init__(self, ui, path, rev)
-        self.repo = hg.repository(self.ui, path)
+        try:
+            self.repo = hg.repository(self.ui, path)
+        except:
+            raise NoRepo("could not open hg repo %s as source" % path)
         self.lastrev = None
         self.lastctx = None
 
     def changectx(self, rev):
         if self.lastrev != rev:
             self.lastctx = self.repo.changectx(rev)
             self.lastrev = rev
         return self.lastctx
 
     def getheads(self):
         if self.rev:
             return [hex(self.repo.changectx(self.rev).node())]
         else:
             return [hex(node) for node in self.repo.heads()]
 
     def getfile(self, name, rev):
         try:
             return self.changectx(rev).filectx(name).data()
         except revlog.LookupError, err:
             raise IOError(err)
 
     def getmode(self, name, rev):
         m = self.changectx(rev).manifest()
         return (m.execf(name) and 'x' or '') + (m.linkf(name) and 'l' or '')
 
     def getchanges(self, rev):
         ctx = self.changectx(rev)
         m, a, r = self.repo.status(ctx.parents()[0].node(), ctx.node())[:3]
         changes = [(name, rev) for name in m + a + r]
         changes.sort()
         return (changes, self.getcopies(ctx, m + a))
 
     def getcopies(self, ctx, files):
         copies = {}
         for name in files:
             try:
                 copies[name] = ctx.filectx(name).renamed()[0]
             except TypeError:
                 pass
         return copies
 
     def getcommit(self, rev):
         ctx = self.changectx(rev)
         parents = [hex(p.node()) for p in ctx.parents() if p.node() != nullid]
         return commit(author=ctx.user(), date=util.datestr(ctx.date()),
                       desc=ctx.description(), parents=parents,
                       branch=ctx.branch())
 
     def gettags(self):
         tags = [t for t in self.repo.tagslist() if t[0] != 'tip']
         return dict([(name, hex(node)) for name, node in tags])

mercurial/util.py

@@ -1,1696 +1,1696 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, incorporated herein by reference.
 
 This contains helper routines that are independent of the SCM core and hide
 platform-specific details from the core.
 """
 
 from i18n import _
 import cStringIO, errno, getpass, popen2, re, shutil, sys, tempfile, strutil
 import os, stat, threading, time, calendar, ConfigParser, locale, glob
 
 try:
     set = set
     frozenset = frozenset
 except NameError:
     from sets import Set as set, ImmutableSet as frozenset
 
 try:
     _encoding = os.environ.get("HGENCODING")
     if sys.platform == 'darwin' and not _encoding:
         # On darwin, getpreferredencoding ignores the locale environment and
         # always returns mac-roman. We override this if the environment is
         # not C (has been customized by the user).
         locale.setlocale(locale.LC_CTYPE, '')
         _encoding = locale.getlocale()[1]
     if not _encoding:
         _encoding = locale.getpreferredencoding() or 'ascii'
 except locale.Error:
     _encoding = 'ascii'
 _encodingmode = os.environ.get("HGENCODINGMODE", "strict")
 _fallbackencoding = 'ISO-8859-1'
 
 def tolocal(s):
     """
     Convert a string from internal UTF-8 to local encoding
 
     All internal strings should be UTF-8 but some repos before the
     implementation of locale support may contain latin1 or possibly
     other character sets. We attempt to decode everything strictly
     using UTF-8, then Latin-1, and failing that, we use UTF-8 and
     replace unknown characters.
     """
     for e in ('UTF-8', _fallbackencoding):
         try:
             u = s.decode(e) # attempt strict decoding
             return u.encode(_encoding, "replace")
         except LookupError, k:
             raise Abort(_("%s, please check your locale settings") % k)
         except UnicodeDecodeError:
             pass
     u = s.decode("utf-8", "replace") # last ditch
     return u.encode(_encoding, "replace")
 
 def fromlocal(s):
     """
     Convert a string from the local character encoding to UTF-8
 
     We attempt to decode strings using the encoding mode set by
     HGENCODINGMODE, which defaults to 'strict'. In this mode, unknown
     characters will cause an error message. Other modes include
     'replace', which replaces unknown characters with a special
     Unicode character, and 'ignore', which drops the character.
     """
     try:
         return s.decode(_encoding, _encodingmode).encode("utf-8")
     except UnicodeDecodeError, inst:
         sub = s[max(0, inst.start-10):inst.start+10]
         raise Abort("decoding near '%s': %s!" % (sub, inst))
     except LookupError, k:
         raise Abort(_("%s, please check your locale settings") % k)
 
 def locallen(s):
     """Find the length in characters of a local string"""
     return len(s.decode(_encoding, "replace"))
 
 def localsub(s, a, b=None):
     try:
         u = s.decode(_encoding, _encodingmode)
         if b is not None:
             u = u[a:b]
         else:
             u = u[:a]
         return u.encode(_encoding, _encodingmode)
     except UnicodeDecodeError, inst:
         sub = s[max(0, inst.start-10), inst.start+10]
         raise Abort(_("decoding near '%s': %s!") % (sub, inst))
 
 # 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:%SP',
     '%H:%M',
     '%I:%M%p',
 )
 
 extendeddateformats = defaultdateformats + (
     "%Y",
     "%Y-%m",
     "%b",
     "%b %Y",
     )
 
 class SignalInterrupt(Exception):
     """Exception raised on SIGTERM and SIGHUP."""
 
 # differences from SafeConfigParser:
 # - case-sensitive keys
 # - allows values that are not strings (this means that you may not
 #   be able to save the configuration to a file)
 class configparser(ConfigParser.SafeConfigParser):
     def optionxform(self, optionstr):
         return optionstr
 
     def set(self, section, option, value):
         return ConfigParser.ConfigParser.set(self, section, option, value)
 
     def _interpolate(self, section, option, rawval, vars):
         if not isinstance(rawval, basestring):
             return rawval
         return ConfigParser.SafeConfigParser._interpolate(self, section,
                                                           option, rawval, vars)
 
 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 pipefilter(s, cmd):
     '''filter string S through command CMD, returning its output'''
     (pin, pout) = os.popen2(cmd, 'b')
     def writer():
         try:
             pin.write(s)
             pin.close()
         except IOError, inst:
             if inst.errno != errno.EPIPE:
                 raise
 
     # we should use select instead on UNIX, but this will work on most
     # systems, including Windows
     w = threading.Thread(target=writer)
     w.start()
     f = pout.read()
     pout.close()
     w.join()
     return f
 
 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, explain_exit(code)))
         return open(outname, 'rb').read()
     finally:
         try:
             if inname: os.unlink(inname)
         except: pass
         try:
             if outname: os.unlink(outname)
         except: 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 using diff's heuristic"""
     if s and '\0' in s[:4096]:
         return True
     return False
 
 def unique(g):
     """return the uniq elements of iterable g"""
     seen = {}
     l = []
     for f in g:
         if f not in seen:
             seen[f] = 1
             l.append(f)
     return l
 
 class Abort(Exception):
     """Raised if a command needs to print an error and exit."""
 
 class UnexpectedOutput(Abort):
     """Raised to print an error with part of output and exit."""
 
 def always(fn): return True
 def never(fn): return False
 
 def expand_glob(pats):
     '''On Windows, expand the implicit globs in a list of patterns'''
     if os.name != 'nt':
         return list(pats)
     ret = []
     for p in pats:
         kind, name = patkind(p, None)
         if kind is None:
             globbed = glob.glob(name)
             if globbed:
                 ret.extend(globbed)
                 continue
             # if we couldn't expand the glob, just keep it around
         ret.append(p)
     return ret
 
 def patkind(name, dflt_pat='glob'):
     """Split a string into an optional pattern kind prefix and the
     actual pattern."""
     for prefix in 're', 'glob', 'path', 'relglob', 'relpath', 'relre':
         if name.startswith(prefix + ':'): return name.split(':', 1)
     return dflt_pat, name
 
 def globre(pat, head='^', tail='$'):
     "convert a glob pattern into a regexp"
     i, n = 0, len(pat)
     res = ''
     group = False
     def peek(): return i < n and pat[i]
     while i < n:
         c = pat[i]
         i = i+1
         if c == '*':
             if peek() == '*':
                 i += 1
                 res += '.*'
             else:
                 res += '[^/]*'
         elif c == '?':
             res += '.'
         elif c == '[':
             j = i
             if j < n and pat[j] in '!]':
                 j += 1
             while j < n and pat[j] != ']':
                 j += 1
             if j >= n:
                 res += '\\['
             else:
                 stuff = pat[i:j].replace('\\','\\\\')
                 i = j + 1
                 if stuff[0] == '!':
                     stuff = '^' + stuff[1:]
                 elif stuff[0] == '^':
                     stuff = '\\' + stuff
                 res = '%s[%s]' % (res, stuff)
         elif c == '{':
             group = True
             res += '(?:'
         elif c == '}' and group:
             res += ')'
             group = False
         elif c == ',' and group:
             res += '|'
         elif c == '\\':
             p = peek()
             if p:
                 i += 1
                 res += re.escape(p)
             else:
                 res += re.escape(c)
         else:
             res += re.escape(c)
     return head + res + tail
 
 _globchars = {'[': 1, '{': 1, '*': 1, '?': 1}
 
 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 = n1.split(os.sep), 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)
 
 def canonpath(root, cwd, myname):
     """return the canonical path of myname, given cwd and root"""
     if root == os.sep:
         rootsep = os.sep
     elif root.endswith(os.sep):
         rootsep = root
     else:
         rootsep = root + os.sep
     name = myname
     if not os.path.isabs(name):
         name = os.path.join(root, cwd, name)
     name = os.path.normpath(name)
     audit_path = path_auditor(root)
     if name != rootsep and name.startswith(rootsep):
         name = name[len(rootsep):]
         audit_path(name)
         return pconvert(name)
     elif name == root:
         return ''
     else:
         # Determine whether `name' is in the hierarchy at or beneath `root',
         # by iterating name=dirname(name) until that causes no change (can't
         # check name == '/', because that doesn't work on windows).  For each
         # `name', compare dev/inode numbers.  If they match, the list `rel'
         # holds the reversed list of components making up the relative file
         # name we want.
         root_st = os.stat(root)
         rel = []
         while True:
             try:
                 name_st = os.stat(name)
             except OSError:
                 break
             if samestat(name_st, root_st):
                 if not rel:
                     # name was actually the same as root (maybe a symlink)
                     return ''
                 rel.reverse()
                 name = os.path.join(*rel)
                 audit_path(name)
                 return pconvert(name)
             dirname, basename = os.path.split(name)
             rel.append(basename)
             if dirname == name:
                 break
             name = dirname
 
         raise Abort('%s not under root' % myname)
 
 def matcher(canonroot, cwd='', names=[], inc=[], exc=[], src=None):
     return _matcher(canonroot, cwd, names, inc, exc, 'glob', src)
 
 def cmdmatcher(canonroot, cwd='', names=[], inc=[], exc=[], src=None,
                globbed=False, default=None):
     default = default or 'relpath'
     if default == 'relpath' and not globbed:
         names = expand_glob(names)
     return _matcher(canonroot, cwd, names, inc, exc, default, src)
 
 def _matcher(canonroot, cwd, names, inc, exc, dflt_pat, src):
     """build a function to match a set of file patterns
 
     arguments:
     canonroot - the canonical root of the tree you're matching against
     cwd - the current working directory, if relevant
     names - patterns to find
     inc - patterns to include
     exc - patterns to exclude
     dflt_pat - if a pattern in names has no explicit type, assume this one
     src - where these patterns came from (e.g. .hgignore)
 
     a pattern is one of:
     'glob:<glob>' - a glob relative to cwd
     're:<regexp>' - a regular expression
     'path:<path>' - a path relative to canonroot
     'relglob:<glob>' - an unrooted glob (*.c matches C files in all dirs)
     'relpath:<path>' - a path relative to cwd
     'relre:<regexp>' - a regexp that doesn't have to match the start of a name
     '<something>' - one of the cases above, selected by the dflt_pat argument
 
     returns:
     a 3-tuple containing
     - list of roots (places where one should start a recursive walk of the fs);
       this often matches the explicit non-pattern names passed in, but also
       includes the initial part of glob: patterns that has no glob characters
     - a bool match(filename) function
     - a bool indicating if any patterns were passed in
     """
 
     # a common case: no patterns at all
     if not names and not inc and not exc:
         return [], always, False
 
     def contains_glob(name):
         for c in name:
             if c in _globchars: return True
         return False
 
     def regex(kind, name, tail):
         '''convert a pattern into a regular expression'''
         if not name:
             return ''
         if kind == 're':
             return name
         elif kind == 'path':
             return '^' + re.escape(name) + '(?:/|$)'
         elif kind == 'relglob':
             return globre(name, '(?:|.*/)', tail)
         elif kind == 'relpath':
             return re.escape(name) + '(?:/|$)'
         elif kind == 'relre':
             if name.startswith('^'):
                 return name
             return '.*' + name
         return globre(name, '', tail)
 
     def matchfn(pats, tail):
         """build a matching function from a set of patterns"""
         if not pats:
             return
         try:
             pat = '(?:%s)' % '|'.join([regex(k, p, tail) for (k, p) in pats])
             return re.compile(pat).match
         except OverflowError:
             # We're using a Python with a tiny regex engine and we
             # made it explode, so we'll divide the pattern list in two
             # until it works
             l = len(pats)
             if l < 2:
                 raise
             a, b = matchfn(pats[:l/2], tail), matchfn(pats[l/2:], tail)
             return lambda s: a(s) or b(s)
         except re.error:
             for k, p in pats:
                 try:
                     re.compile('(?:%s)' % regex(k, p, tail))
                 except re.error:
                     if src:
                         raise Abort("%s: invalid pattern (%s): %s" %
                                     (src, k, p))
                     else:
                         raise Abort("invalid pattern (%s): %s" % (k, p))
             raise Abort("invalid pattern")
 
     def globprefix(pat):
         '''return the non-glob prefix of a path, e.g. foo/* -> foo'''
         root = []
         for p in pat.split('/'):
             if contains_glob(p): break
             root.append(p)
         return '/'.join(root) or '.'
 
     def normalizepats(names, default):
         pats = []
         roots = []
         anypats = False
         for kind, name in [patkind(p, default) for p in names]:
             if kind in ('glob', 'relpath'):
                 name = canonpath(canonroot, cwd, name)
             elif kind in ('relglob', 'path'):
                 name = normpath(name)
 
             pats.append((kind, name))
 
             if kind in ('glob', 're', 'relglob', 'relre'):
                 anypats = True
 
             if kind == 'glob':
                 root = globprefix(name)
                 roots.append(root)
             elif kind in ('relpath', 'path'):
                 roots.append(name or '.')
             elif kind == 'relglob':
                 roots.append('.')
         return roots, pats, anypats
 
     roots, pats, anypats = normalizepats(names, dflt_pat)
 
     patmatch = matchfn(pats, '$') or always
     incmatch = always
     if inc:
         dummy, inckinds, dummy = normalizepats(inc, 'glob')
         incmatch = matchfn(inckinds, '(?:/|$)')
     excmatch = lambda fn: False
     if exc:
         dummy, exckinds, dummy = normalizepats(exc, 'glob')
         excmatch = matchfn(exckinds, '(?:/|$)')
 
     if not names and inc and not exc:
         # common case: hgignore patterns
         match = incmatch
     else:
         match = lambda fn: incmatch(fn) and not excmatch(fn) and patmatch(fn)
 
     return (roots, match, (inc or exc or anypats) and True)
 
 _hgexecutable = None
 
 def hgexecutable():
     """return location of the 'hg' executable.
 
     Defaults to $HG or 'hg' in the search path.
     """
     if _hgexecutable is None:
         set_hgexecutable(os.environ.get('HG') or find_exe('hg', 'hg'))
     return _hgexecutable
 
 def set_hgexecutable(path):
     """set location of the 'hg' executable"""
     global _hgexecutable
     _hgexecutable = path
 
 def system(cmd, environ={}, cwd=None, onerr=None, errprefix=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.'''
     def py2shell(val):
         'convert python object into string that is useful to shell'
         if val in (None, False):
             return '0'
         if val == True:
             return '1'
         return str(val)
     oldenv = {}
     for k in environ:
         oldenv[k] = os.environ.get(k)
     if cwd is not None:
         oldcwd = os.getcwd()
     origcmd = cmd
     if os.name == 'nt':
         cmd = '"%s"' % cmd
     try:
         for k, v in environ.iteritems():
             os.environ[k] = py2shell(v)
         os.environ['HG'] = hgexecutable()
         if cwd is not None and oldcwd != cwd:
             os.chdir(cwd)
         rc = os.system(cmd)
         if sys.platform == 'OpenVMS' and rc & 1:
             rc = 0
         if rc and onerr:
             errmsg = '%s %s' % (os.path.basename(origcmd.split(None, 1)[0]),
                                 explain_exit(rc)[0])
             if errprefix:
                 errmsg = '%s: %s' % (errprefix, errmsg)
             try:
                 onerr.warn(errmsg + '\n')
             except AttributeError:
                 raise onerr(errmsg)
         return rc
     finally:
         for k, v in oldenv.iteritems():
             if v is None:
                 del os.environ[k]
             else:
                 os.environ[k] = v
         if cwd is not None and oldcwd != cwd:
             os.chdir(oldcwd)
 
 # os.path.lexists is not available on python2.3
 def lexists(filename):
     "test whether a file with this name exists. does not follow symlinks"
     try:
         os.lstat(filename)
     except:
         return False
     return True
 
 def rename(src, dst):
     """forcibly rename a file"""
     try:
         os.rename(src, dst)
     except OSError, err: # FIXME: check err (EEXIST ?)
         # on windows, rename to existing file is not allowed, so we
         # must delete destination first. but if file is open, unlink
         # schedules it for delete but does not delete it. rename
         # happens immediately even for open files, so we create
         # temporary file, delete it, rename destination to that name,
         # then delete that. then rename is safe to do.
         fd, temp = tempfile.mkstemp(dir=os.path.dirname(dst) or '.')
         os.close(fd)
         os.unlink(temp)
         os.rename(dst, temp)
         os.unlink(temp)
         os.rename(src, dst)
 
 def unlink(f):
     """unlink and remove the directory if it is empty"""
     os.unlink(f)
     # try removing directories that might now be empty
     try:
         os.removedirs(os.path.dirname(f))
     except OSError:
         pass
 
 def copyfile(src, dest):
     "copy a file, preserving mode"
     if os.path.islink(src):
         try:
             os.unlink(dest)
         except:
             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)
 
     if os.path.isdir(src):
         os.mkdir(dst)
         for name in os.listdir(src):
             srcname = os.path.join(src, name)
             dstname = os.path.join(dst, name)
             copyfiles(srcname, dstname, hardlink)
     else:
         if hardlink:
             try:
                 os_link(src, dst)
             except (IOError, OSError):
                 hardlink = False
                 shutil.copy(src, dst)
         else:
             shutil.copy(src, dst)
 
 class path_auditor(object):
     '''ensure that a filesystem path contains no banned components.
     the following properties of a path are checked:
 
     - under top-level .hg
     - starts at the root of a windows drive
     - contains ".."
     - traverses a symlink (e.g. a/symlink_here/b)
     - inside a nested repository'''
 
     def __init__(self, root):
         self.audited = set()
         self.auditeddir = set()
         self.root = root
 
     def __call__(self, path):
         if path in self.audited:
             return
         normpath = os.path.normcase(path)
         parts = normpath.split(os.sep)
         if (os.path.splitdrive(path)[0] or parts[0] in ('.hg', '')
             or os.pardir in parts):
             raise Abort(_("path contains illegal component: %s") % path)
         def check(prefix):
             curpath = os.path.join(self.root, prefix)
             try:
                 st = os.lstat(curpath)
             except OSError, err:
                 # EINVAL can be raised as invalid path syntax under win32.
                 # They must be ignored for patterns can be checked too.
                 if err.errno not in (errno.ENOENT, errno.EINVAL):
                     raise
             else:
                 if stat.S_ISLNK(st.st_mode):
                     raise Abort(_('path %r traverses symbolic link %r') %
                                 (path, prefix))
                 elif (stat.S_ISDIR(st.st_mode) and
                       os.path.isdir(os.path.join(curpath, '.hg'))):
                     raise Abort(_('path %r is inside repo %r') %
                                 (path, prefix))
 
         prefixes = []
         for c in strutil.rfindall(normpath, os.sep):
             prefix = normpath[:c]
             if prefix in self.auditeddir:
                 break
             check(prefix)
             prefixes.append(prefix)
 
         self.audited.add(path)
         # only add prefixes to the cache after checking everything: we don't
         # want to add "foo/bar/baz" before checking if there's a "foo/.hg"
         self.auditeddir.update(prefixes)
 
 def _makelock_file(info, pathname):
     ld = os.open(pathname, os.O_CREAT | os.O_WRONLY | os.O_EXCL)
     os.write(ld, info)
     os.close(ld)
 
 def _readlock_file(pathname):
     return posixfile(pathname).read()
 
 def nlinks(pathname):
     """Return number of hardlinks for the given file."""
     return os.lstat(pathname).st_nlink
 
 if hasattr(os, 'link'):
     os_link = os.link
 else:
     def os_link(src, dst):
         raise OSError(0, _("Hardlinks not supported"))
 
 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)
 
 posixfile = file
 
 def is_win_9x():
     '''return true if run on windows 95, 98 or me.'''
     try:
         return sys.getwindowsversion()[3] == 1
     except AttributeError:
         return os.name == 'nt' and 'command' in os.environ.get('comspec', '')
 
 getuser_fallback = None
 
 def getuser():
     '''return name of current user'''
     try:
         return getpass.getuser()
     except ImportError:
         # import of pwd will fail on windows - try fallback
         if getuser_fallback:
             return getuser_fallback()
     # raised if win32api not available
     raise Abort(_('user name not available - set USERNAME '
                   'environment variable'))
 
 def username(uid=None):
     """Return the name of the user with the given uid.
 
     If uid is None, return the name of the current user."""
     try:
         import pwd
         if uid is None:
             uid = os.getuid()
         try:
             return pwd.getpwuid(uid)[0]
         except KeyError:
             return str(uid)
     except ImportError:
         return None
 
 def groupname(gid=None):
     """Return the name of the group with the given gid.
 
     If gid is None, return the name of the current group."""
     try:
         import grp
         if gid is None:
             gid = os.getgid()
         try:
             return grp.getgrgid(gid)[0]
         except KeyError:
             return str(gid)
     except ImportError:
         return None
 
 # File system features
 
 def checkfolding(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:
         return True
 
 def checkexec(path):
     """
     Check whether the given path is on a filesystem with UNIX-like exec flags
 
     Requires a directory (like /foo/.hg)
     """
     try:
         fh, fn = tempfile.mkstemp("", "", path)
         os.close(fh)
         m = os.stat(fn).st_mode
         os.chmod(fn, m ^ 0111)
         r = (os.stat(fn).st_mode != m)
         os.unlink(fn)
     except (IOError,OSError):
         # we don't care, the user probably won't be able to commit anyway
         return False
     return r
 
 def execfunc(path, fallback):
     '''return an is_exec() function with default to fallback'''
     if checkexec(path):
         return lambda x: is_exec(os.path.join(path, x))
     return fallback
 
 def checklink(path):
     """check whether the given path is on a symlink-capable filesystem"""
     # mktemp is not racy because symlink creation will fail if the
     # file already exists
     name = tempfile.mktemp(dir=path)
     try:
         os.symlink(".", name)
         os.unlink(name)
         return True
     except (OSError, AttributeError):
         return False
 
 def linkfunc(path, fallback):
     '''return an is_link() function with default to fallback'''
     if checklink(path):
         return lambda x: os.path.islink(os.path.join(path, x))
     return fallback
 
 _umask = os.umask(0)
 os.umask(_umask)
 
 def needbinarypatch():
     """return True if patches should be applied in binary mode by default."""
     return os.name == 'nt'
 
 # Platform specific variants
 if os.name == 'nt':
     import msvcrt
     nulldev = 'NUL:'
 
     class winstdout:
         '''stdout on windows misbehaves if sent through a pipe'''
 
         def __init__(self, fp):
             self.fp = fp
 
         def __getattr__(self, key):
             return getattr(self.fp, key)
 
         def close(self):
             try:
                 self.fp.close()
             except: pass
 
         def write(self, s):
             try:
                 return self.fp.write(s)
             except IOError, inst:
                 if inst.errno != 0: raise
                 self.close()
                 raise IOError(errno.EPIPE, 'Broken pipe')
 
         def flush(self):
             try:
                 return self.fp.flush()
             except IOError, inst:
                 if inst.errno != errno.EINVAL: raise
                 self.close()
                 raise IOError(errno.EPIPE, 'Broken pipe')
 
     sys.stdout = winstdout(sys.stdout)
 
     def system_rcpath():
         try:
             return system_rcpath_win32()
         except:
             return [r'c:\mercurial\mercurial.ini']
 
     def user_rcpath():
         '''return os-specific hgrc search path to the user dir'''
         try:
             userrc = user_rcpath_win32()
         except:
             userrc = os.path.join(os.path.expanduser('~'), 'mercurial.ini')
         path = [userrc]
         userprofile = os.environ.get('USERPROFILE')
         if userprofile:
             path.append(os.path.join(userprofile, 'mercurial.ini'))
         return path
 
     def parse_patch_output(output_line):
         """parses the output produced by patch and returns the file name"""
         pf = output_line[14:]
         if pf[0] == '`':
             pf = pf[1:-1] # Remove the quotes
         return pf
 
     def testpid(pid):
         '''return False if pid dead, True if running or not known'''
         return True
 
     def set_exec(f, mode):
         pass
 
     def set_link(f, mode):
         pass
 
     def set_binary(fd):
         msvcrt.setmode(fd.fileno(), os.O_BINARY)
 
     def pconvert(path):
         return path.replace("\\", "/")
 
     def localpath(path):
         return path.replace('/', '\\')
 
     def normpath(path):
         return pconvert(os.path.normpath(path))
 
     makelock = _makelock_file
     readlock = _readlock_file
 
     def samestat(s1, s2):
         return False
 
     # A sequence of backslashes is special iff it precedes a double quote:
     # - if there's an even number of backslashes, the double quote is not
     #   quoted (i.e. it ends the quoted region)
     # - if there's an odd number of backslashes, the double quote is quoted
     # - in both cases, every pair of backslashes is unquoted into a single
     #   backslash
     # (See http://msdn2.microsoft.com/en-us/library/a1y7w461.aspx )
     # So, to quote a string, we must surround it in double quotes, double
     # the number of backslashes that preceed double quotes and add another
     # backslash before every double quote (being careful with the double
     # quote we've appended to the end)
     _quotere = None
     def shellquote(s):
         global _quotere
         if _quotere is None:
             _quotere = re.compile(r'(\\*)("|\\$)')
         return '"%s"' % _quotere.sub(r'\1\1\\\2', s)
 
     def quotecommand(cmd):
         """Build a command string suitable for os.popen* calls."""
         # The extra quotes are needed because popen* runs the command
         # through the current COMSPEC. cmd.exe suppress enclosing quotes.
         return '"' + cmd + '"'
 
     def explain_exit(code):
         return _("exited with status %d") % code, code
 
     # if you change this stub into a real check, please try to implement the
     # username and groupname functions above, too.
     def isowner(fp, st=None):
         return True
 
     def find_in_path(name, path, default=None):
         '''find name in search path. path can be string (will be split
         with os.pathsep), or iterable thing that returns strings.  if name
         found, return path to name. else return default. name is looked up
         using cmd.exe rules, using PATHEXT.'''
         if isinstance(path, str):
             path = path.split(os.pathsep)
 
         pathext = os.environ.get('PATHEXT', '.COM;.EXE;.BAT;.CMD')
         pathext = pathext.lower().split(os.pathsep)
         isexec = os.path.splitext(name)[1].lower() in pathext
 
         for p in path:
             p_name = os.path.join(p, name)
 
             if isexec and os.path.exists(p_name):
                 return p_name
 
             for ext in pathext:
                 p_name_ext = p_name + ext
                 if os.path.exists(p_name_ext):
                     return p_name_ext
         return default
 
     def set_signal_handler():
         try:
             set_signal_handler_win32()
         except NameError:
             pass
 
     try:
         # override functions with win32 versions if possible
         from util_win32 import *
         if not is_win_9x():
             posixfile = posixfile_nt
     except ImportError:
         pass
 
 else:
     nulldev = '/dev/null'
 
     def rcfiles(path):
         rcs = [os.path.join(path, 'hgrc')]
         rcdir = os.path.join(path, 'hgrc.d')
         try:
             rcs.extend([os.path.join(rcdir, f) for f in os.listdir(rcdir)
                         if f.endswith(".rc")])
         except OSError:
             pass
         return rcs
 
     def system_rcpath():
         path = []
         # old mod_python does not set sys.argv
         if len(getattr(sys, 'argv', [])) > 0:
             path.extend(rcfiles(os.path.dirname(sys.argv[0]) +
                                   '/../etc/mercurial'))
         path.extend(rcfiles('/etc/mercurial'))
         return path
 
     def user_rcpath():
         return [os.path.expanduser('~/.hgrc')]
 
     def parse_patch_output(output_line):
         """parses the output produced by patch and returns the file name"""
         pf = output_line[14:]
         if os.sys.platform == 'OpenVMS':
             if pf[0] == '`':
                 pf = pf[1:-1] # Remove the quotes
         else:
            if pf.startswith("'") and pf.endswith("'") and " " in pf:
                 pf = pf[1:-1] # Remove the quotes
         return pf
 
     def is_exec(f):
         """check whether a file is executable"""
         return (os.lstat(f).st_mode & 0100 != 0)
 
     def set_exec(f, mode):
         s = os.lstat(f).st_mode
         if stat.S_ISLNK(s) or (s & 0100 != 0) == mode:
             return
         if mode:
             # Turn on +x for every +r bit when making a file executable
             # and obey umask.
             os.chmod(f, s | (s & 0444) >> 2 & ~_umask)
         else:
             os.chmod(f, s & 0666)
 
     def set_link(f, mode):
         """make a file a symbolic link/regular file
 
         if a file is changed to a link, its contents become the link data
         if a link is changed to a file, its link data become its contents
         """
 
         m = os.path.islink(f)
         if m == bool(mode):
             return
 
         if mode: # switch file to link
             data = file(f).read()
             os.unlink(f)
             os.symlink(data, f)
         else:
             data = os.readlink(f)
             os.unlink(f)
             file(f, "w").write(data)
 
     def set_binary(fd):
         pass
 
     def pconvert(path):
         return path
 
     def localpath(path):
         return path
 
     normpath = os.path.normpath
     samestat = os.path.samestat
 
     def makelock(info, pathname):
         try:
             os.symlink(info, pathname)
         except OSError, why:
             if why.errno == errno.EEXIST:
                 raise
             else:
                 _makelock_file(info, pathname)
 
     def readlock(pathname):
         try:
             return os.readlink(pathname)
         except OSError, why:
             if why.errno in (errno.EINVAL, errno.ENOSYS):
                 return _readlock_file(pathname)
             else:
                 raise
 
     def shellquote(s):
         if os.sys.platform == 'OpenVMS':
             return '"%s"' % s
         else:
             return "'%s'" % s.replace("'", "'\\''")
 
     def quotecommand(cmd):
         return cmd
 
     def testpid(pid):
         '''return False if pid dead, True if running or not sure'''
         if os.sys.platform == 'OpenVMS':
             return True
         try:
             os.kill(pid, 0)
             return True
         except OSError, inst:
             return inst.errno != errno.ESRCH
 
     def explain_exit(code):
         """return a 2-tuple (desc, code) describing a process's status"""
         if os.WIFEXITED(code):
             val = os.WEXITSTATUS(code)
             return _("exited with status %d") % val, val
         elif os.WIFSIGNALED(code):
             val = os.WTERMSIG(code)
             return _("killed by signal %d") % val, val
         elif os.WIFSTOPPED(code):
             val = os.WSTOPSIG(code)
             return _("stopped by signal %d") % val, val
         raise ValueError(_("invalid exit code"))
 
     def isowner(fp, st=None):
         """Return True if the file object f belongs to the current user.
 
         The return value of a util.fstat(f) may be passed as the st argument.
         """
         if st is None:
             st = fstat(fp)
         return st.st_uid == os.getuid()
 
     def find_in_path(name, path, default=None):
         '''find name in search path. path can be string (will be split
         with os.pathsep), or iterable thing that returns strings.  if name
         found, return path to name. else return default.'''
         if isinstance(path, str):
             path = path.split(os.pathsep)
         for p in path:
             p_name = os.path.join(p, name)
             if os.path.exists(p_name):
                 return p_name
         return default
 
     def set_signal_handler():
         pass
 
 def find_exe(name, default=None):
     '''find path of an executable.
     if name contains a path component, return it as is.  otherwise,
     use normal executable search path.'''
 
     if os.sep in name or sys.platform == 'OpenVMS':
         # don't check the executable bit.  if the file isn't
         # executable, whoever tries to actually run it will give a
         # much more useful error message.
         return name
     return find_in_path(name, os.environ.get('PATH', ''), default=default)
 
 def _buildencodefun():
     e = '_'
     win_reserved = [ord(x) for x in '\\:*?"<>|']
     cmap = dict([ (chr(x), chr(x)) for x in xrange(127) ])
     for x in (range(32) + range(126, 256) + win_reserved):
         cmap[chr(x)] = "~%02x" % x
     for x in range(ord("A"), ord("Z")+1) + [ord(e)]:
         cmap[chr(x)] = e + chr(x).lower()
     dmap = {}
     for k, v in cmap.iteritems():
         dmap[v] = k
     def decode(s):
         i = 0
         while i < len(s):
             for l in xrange(1, 4):
                 try:
                     yield dmap[s[i:i+l]]
                     i += l
                     break
                 except KeyError:
                     pass
             else:
                 raise KeyError
     return (lambda s: "".join([cmap[c] for c in s]),
             lambda s: "".join(list(decode(s))))
 
 encodefilename, decodefilename = _buildencodefun()
 
 def encodedopener(openerfn, fn):
     def o(path, *args, **kw):
         return openerfn(fn(path), *args, **kw)
     return o
 
 def mktempcopy(name, emptyok=False):
     """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.
     try:
         st_mode = os.lstat(name).st_mode
     except OSError, inst:
         if inst.errno != errno.ENOENT:
             raise
         st_mode = 0666 & ~_umask
     os.chmod(temp, st_mode)
     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(posixfile):
     """file-like object that atomically updates a file
 
     All writes will be redirected to a temporary copy of the original
     file.  When rename is called, the copy is renamed to the original
     name, making the changes visible.
     """
     def __init__(self, name, mode):
         self.__name = name
         self.temp = mktempcopy(name, emptyok=('w' in mode))
         posixfile.__init__(self, self.temp, mode)
 
     def rename(self):
         if not self.closed:
             posixfile.close(self)
             rename(self.temp, localpath(self.__name))
 
     def __del__(self):
         if not self.closed:
             try:
                 os.unlink(self.temp)
             except: pass
             posixfile.close(self)
 
 class opener(object):
     """Open files relative to a base directory
 
     This class is used to hide the details of COW semantics and
     remote file access from higher level code.
     """
     def __init__(self, base, audit=True):
         self.base = base
         if audit:
             self.audit_path = path_auditor(base)
         else:
             self.audit_path = always
 
     def __getattr__(self, name):
         if name == '_can_symlink':
             self._can_symlink = checklink(self.base)
             return self._can_symlink
         raise AttributeError(name)
 
     def __call__(self, path, mode="r", text=False, atomictemp=False):
         self.audit_path(path)
         f = os.path.join(self.base, path)
 
         if not text and "b" not in mode:
             mode += "b" # for that other OS
 
         if mode[0] != "r":
             try:
                 nlink = nlinks(f)
             except OSError:
                 nlink = 0
                 d = os.path.dirname(f)
                 if not os.path.isdir(d):
                     os.makedirs(d)
             if atomictemp:
                 return atomictempfile(f, mode)
             if nlink > 1:
                 rename(mktempcopy(f), f)
         return posixfile(f, mode)
 
     def symlink(self, src, dst):
         self.audit_path(dst)
         linkname = os.path.join(self.base, dst)
         try:
             os.unlink(linkname)
         except OSError:
             pass
 
         dirname = os.path.dirname(linkname)
         if not os.path.exists(dirname):
             os.makedirs(dirname)
 
         if self._can_symlink:
             try:
                 os.symlink(src, linkname)
             except OSError, err:
                 raise OSError(err.errno, _('could not symlink to %r: %s') %
                               (src, err.strerror), linkname)
         else:
             f = self(dst, "w")
             f.write(src)
             f.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, targetsize = 2**16):
         """in_iter is the iterator that's iterating over the input chunks.
         targetsize is how big a buffer to try to maintain."""
         self.in_iter = iter(in_iter)
         self.buf = ''
         self.targetsize = int(targetsize)
         if self.targetsize <= 0:
             raise ValueError(_("targetsize must be greater than 0, was %d") %
                              targetsize)
         self.iterempty = False
 
     def fillbuf(self):
         """Ignore target size; read every chunk from iterator until empty."""
         if not self.iterempty:
             collector = cStringIO.StringIO()
             collector.write(self.buf)
             for ch in self.in_iter:
                 collector.write(ch)
             self.buf = collector.getvalue()
             self.iterempty = True
 
     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."""
         if l > len(self.buf) and not self.iterempty:
             # Clamp to a multiple of self.targetsize
             targetsize = self.targetsize * ((l // self.targetsize) + 1)
             collector = cStringIO.StringIO()
             collector.write(self.buf)
             collected = len(self.buf)
             for chunk in self.in_iter:
                 collector.write(chunk)
                 collected += len(chunk)
                 if collected >= targetsize:
                     break
             if collected < targetsize:
                 self.iterempty = True
             self.buf = collector.getvalue()
         s, self.buf = self.buf[:l], buffer(self.buf, l)
         return s
 
 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
     return time.mktime(lt), tz
 
 def datestr(date=None, format='%a %b %d %H:%M:%S %Y', timezone=True, timezone_format=" %+03d%02d"):
     """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()
     s = time.strftime(format, time.gmtime(float(t) - tz))
     if timezone:
         s += timezone_format % (-tz / 3600, ((-tz % 3600) / 60))
     return s
 
-def strdate(string, format, defaults):
+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():
             tz = int(tz)
             offset = - 3600 * (tz / 100) - 60 * (tz % 100)
             return offset
         if tz == "GMT" or tz == "UTC":
             return 0
         return None
 
     # NOTE: unixtime = localunixtime + offset
     offset, date = timezone(string), string
     if offset != None:
         date = " ".join(string.split()[:-1])
 
     # add missing elements from defaults
     for part in defaults:
         found = [True for p in part if ("%"+p) in format]
         if not found:
             date += "@" + defaults[part]
             format += "@%" + part[0]
 
     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(string, formats=None, defaults=None):
     """parse a localized time string and return a (unixtime, offset) tuple.
     The date may be a "unixtime offset" string or in one of the specified
     formats."""
     if not string:
         return 0, 0
     if not formats:
         formats = defaultdateformats
     string = string.strip()
     try:
         when, offset = map(int, string.split(' '))
     except ValueError:
         # fill out defaults
         if not defaults:
             defaults = {}
         now = makedate()
         for part in "d mb yY HI M S".split():
             if part not in defaults:
                 if part[0] in "HMS":
                     defaults[part] = "00"
                 elif part[0] in "dm":
                     defaults[part] = "1"
                 else:
                     defaults[part] = datestr(now, "%" + part[0], False)
 
         for format in formats:
             try:
                 when, offset = strdate(string, format, defaults)
             except ValueError:
                 pass
             else:
                 break
         else:
             raise Abort(_('invalid date: %r ') % string)
     # 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 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
 
     """
 
     def lower(date):
         return parsedate(date, extendeddateformats)[0]
 
     def upper(date):
         d = dict(mb="12", HI="23", M="59", S="59")
         for days in "31 30 29".split():
             try:
                 d["d"] = days
                 return parsedate(date, extendeddateformats, d)[0]
             except:
                 pass
         d["d"] = "28"
         return parsedate(date, extendeddateformats, d)[0]
 
     if date[0] == "<":
         when = upper(date[1:])
         return lambda x: x <= when
     elif date[0] == ">":
         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:])
         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 ellipsis(text, maxlength=400):
     """Trim string to at most maxlength (default: 400) characters."""
     if len(text) <= maxlength:
         return text
     else:
         return "%s..." % (text[:maxlength-3])
 
 def walkrepos(path):
     '''yield every hg repository under path, recursively.'''
     def errhandler(err):
         if err.filename == path:
             raise err
 
     for root, dirs, files in os.walk(path, onerror=errhandler):
         for d in dirs:
             if d == '.hg':
                 yield root
                 dirs[:] = []
                 break
 
 _rcpath = None
 
 def os_rcpath():
     '''return default os-specific hgrc search path'''
     path = system_rcpath()
     path.extend(user_rcpath())
     path = [os.path.normpath(f) for f in path]
     return path
 
 def rcpath():
     '''return hgrc search path. if env var HGRCPATH is set, use it.
     for each item in path, if directory, use files ending in .rc,
     else use item.
     make HGRCPATH empty to only look in .hg/hgrc of current repo.
     if no HGRCPATH, use default os-specific path.'''
     global _rcpath
     if _rcpath is None:
         if 'HGRCPATH' in os.environ:
             _rcpath = []
             for p in os.environ['HGRCPATH'].split(os.pathsep):
                 if not p: continue
                 if os.path.isdir(p):
                     for f in os.listdir(p):
                         if f.endswith('.rc'):
                             _rcpath.append(os.path.join(p, f))
                 else:
                     _rcpath.append(p)
         else:
             _rcpath = os_rcpath()
     return _rcpath
 
 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 drop_scheme(scheme, path):
     sc = scheme + ':'
     if path.startswith(sc):
         path = path[len(sc):]
         if path.startswith('//'):
             path = path[2:]
     return path
 
 def uirepr(s):
     # Avoid double backslash in Windows path repr()
     return repr(s).replace('\\\\', '\\')