convert: enable deterministic conversion progress bar for svn

This produces slightly bad results when branches are in play, but
overall I think it's probably worthwhile. We might be able to do
better with branches somehow, but I haven't given it any thought.

hgext/convert/subversion.py

@@ -1,1310 +1,1313 b''
 # Subversion 1.4/1.5 Python API backend
 #
 # Copyright(C) 2007 Daniel Holth et al
 
 import os, re, sys, tempfile, urllib, urllib2
 import xml.dom.minidom
 import cPickle as pickle
 
 from mercurial import strutil, scmutil, util, encoding
 from mercurial.i18n import _
 
 propertycache = util.propertycache
 
 # Subversion stuff. Works best with very recent Python SVN bindings
 # e.g. SVN 1.5 or backports. Thanks to the bzr folks for enhancing
 # these bindings.
 
 from cStringIO import StringIO
 
 from common import NoRepo, MissingTool, commit, encodeargs, decodeargs
 from common import commandline, converter_source, converter_sink, mapfile
 from common import makedatetimestamp
 
 try:
     from svn.core import SubversionException, Pool
     import svn
     import svn.client
     import svn.core
     import svn.ra
     import svn.delta
     import transport
     import warnings
     warnings.filterwarnings('ignore',
             module='svn.core',
             category=DeprecationWarning)
 
 except ImportError:
     svn = None
 
 class SvnPathNotFound(Exception):
     pass
 
 def revsplit(rev):
     """Parse a revision string and return (uuid, path, revnum).
     >>> revsplit('svn:a2147622-4a9f-4db4-a8d3-13562ff547b2'
     ...          '/proj%20B/mytrunk/mytrunk@1')
     ('a2147622-4a9f-4db4-a8d3-13562ff547b2', '/proj%20B/mytrunk/mytrunk', 1)
     >>> revsplit('svn:8af66a51-67f5-4354-b62c-98d67cc7be1d@1')
     ('', '', 1)
     >>> revsplit('@7')
     ('', '', 7)
     >>> revsplit('7')
     ('', '', 0)
     >>> revsplit('bad')
     ('', '', 0)
     """
     parts = rev.rsplit('@', 1)
     revnum = 0
     if len(parts) > 1:
         revnum = int(parts[1])
     parts = parts[0].split('/', 1)
     uuid = ''
     mod = ''
     if len(parts) > 1 and parts[0].startswith('svn:'):
         uuid = parts[0][4:]
         mod = '/' + parts[1]
     return uuid, mod, revnum
 
 def quote(s):
     # As of svn 1.7, many svn calls expect "canonical" paths. In
     # theory, we should call svn.core.*canonicalize() on all paths
     # before passing them to the API.  Instead, we assume the base url
     # is canonical and copy the behaviour of svn URL encoding function
     # so we can extend it safely with new components. The "safe"
     # characters were taken from the "svn_uri__char_validity" table in
     # libsvn_subr/path.c.
     return urllib.quote(s, "!$&'()*+,-./:=@_~")
 
 def geturl(path):
     try:
         return svn.client.url_from_path(svn.core.svn_path_canonicalize(path))
     except SubversionException:
         # svn.client.url_from_path() fails with local repositories
         pass
     if os.path.isdir(path):
         path = os.path.normpath(os.path.abspath(path))
         if os.name == 'nt':
             path = '/' + util.normpath(path)
         # Module URL is later compared with the repository URL returned
         # by svn API, which is UTF-8.
         path = encoding.tolocal(path)
         path = 'file://%s' % quote(path)
     return svn.core.svn_path_canonicalize(path)
 
 def optrev(number):
     optrev = svn.core.svn_opt_revision_t()
     optrev.kind = svn.core.svn_opt_revision_number
     optrev.value.number = number
     return optrev
 
 class changedpath(object):
     def __init__(self, p):
         self.copyfrom_path = p.copyfrom_path
         self.copyfrom_rev = p.copyfrom_rev
         self.action = p.action
 
 def get_log_child(fp, url, paths, start, end, limit=0,
                   discover_changed_paths=True, strict_node_history=False):
     protocol = -1
     def receiver(orig_paths, revnum, author, date, message, pool):
         paths = {}
         if orig_paths is not None:
             for k, v in orig_paths.iteritems():
                 paths[k] = changedpath(v)
         pickle.dump((paths, revnum, author, date, message),
                     fp, protocol)
 
     try:
         # Use an ra of our own so that our parent can consume
         # our results without confusing the server.
         t = transport.SvnRaTransport(url=url)
         svn.ra.get_log(t.ra, paths, start, end, limit,
                        discover_changed_paths,
                        strict_node_history,
                        receiver)
     except IOError:
         # Caller may interrupt the iteration
         pickle.dump(None, fp, protocol)
     except Exception, inst:
         pickle.dump(str(inst), fp, protocol)
     else:
         pickle.dump(None, fp, protocol)
     fp.close()
     # With large history, cleanup process goes crazy and suddenly
     # consumes *huge* amount of memory. The output file being closed,
     # there is no need for clean termination.
     os._exit(0)
 
 def debugsvnlog(ui, **opts):
     """Fetch SVN log in a subprocess and channel them back to parent to
     avoid memory collection issues.
     """
     if svn is None:
         raise util.Abort(_('debugsvnlog could not load Subversion python '
                            'bindings'))
 
     util.setbinary(sys.stdin)
     util.setbinary(sys.stdout)
     args = decodeargs(sys.stdin.read())
     get_log_child(sys.stdout, *args)
 
 class logstream(object):
     """Interruptible revision log iterator."""
     def __init__(self, stdout):
         self._stdout = stdout
 
     def __iter__(self):
         while True:
             try:
                 entry = pickle.load(self._stdout)
             except EOFError:
                 raise util.Abort(_('Mercurial failed to run itself, check'
                                    ' hg executable is in PATH'))
             try:
                 orig_paths, revnum, author, date, message = entry
             except (TypeError, ValueError):
                 if entry is None:
                     break
                 raise util.Abort(_("log stream exception '%s'") % entry)
             yield entry
 
     def close(self):
         if self._stdout:
             self._stdout.close()
             self._stdout = None
 
 class directlogstream(list):
     """Direct revision log iterator.
     This can be used for debugging and development but it will probably leak
     memory and is not suitable for real conversions."""
     def __init__(self, url, paths, start, end, limit=0,
                   discover_changed_paths=True, strict_node_history=False):
 
         def receiver(orig_paths, revnum, author, date, message, pool):
             paths = {}
             if orig_paths is not None:
                 for k, v in orig_paths.iteritems():
                     paths[k] = changedpath(v)
             self.append((paths, revnum, author, date, message))
 
         # Use an ra of our own so that our parent can consume
         # our results without confusing the server.
         t = transport.SvnRaTransport(url=url)
         svn.ra.get_log(t.ra, paths, start, end, limit,
                        discover_changed_paths,
                        strict_node_history,
                        receiver)
 
     def close(self):
         pass
 
 # Check to see if the given path is a local Subversion repo. Verify this by
 # looking for several svn-specific files and directories in the given
 # directory.
 def filecheck(ui, path, proto):
     for x in ('locks', 'hooks', 'format', 'db'):
         if not os.path.exists(os.path.join(path, x)):
             return False
     return True
 
 # Check to see if a given path is the root of an svn repo over http. We verify
 # this by requesting a version-controlled URL we know can't exist and looking
 # for the svn-specific "not found" XML.
 def httpcheck(ui, path, proto):
     try:
         opener = urllib2.build_opener()
         rsp = opener.open('%s://%s/!svn/ver/0/.svn' % (proto, path))
         data = rsp.read()
     except urllib2.HTTPError, inst:
         if inst.code != 404:
             # Except for 404 we cannot know for sure this is not an svn repo
             ui.warn(_('svn: cannot probe remote repository, assume it could '
                       'be a subversion repository. Use --source-type if you '
                       'know better.\n'))
             return True
         data = inst.fp.read()
     except Exception:
         # Could be urllib2.URLError if the URL is invalid or anything else.
         return False
     return '<m:human-readable errcode="160013">' in data
 
 protomap = {'http': httpcheck,
             'https': httpcheck,
             'file': filecheck,
             }
 def issvnurl(ui, url):
     try:
         proto, path = url.split('://', 1)
         if proto == 'file':
             if (os.name == 'nt' and path[:1] == '/' and path[1:2].isalpha()
                 and path[2:6].lower() == '%3a/'):
                 path = path[:2] + ':/' + path[6:]
             path = urllib.url2pathname(path)
     except ValueError:
         proto = 'file'
         path = os.path.abspath(url)
     if proto == 'file':
         path = util.pconvert(path)
     check = protomap.get(proto, lambda *args: False)
     while '/' in path:
         if check(ui, path, proto):
             return True
         path = path.rsplit('/', 1)[0]
     return False
 
 # SVN conversion code stolen from bzr-svn and tailor
 #
 # Subversion looks like a versioned filesystem, branches structures
 # are defined by conventions and not enforced by the tool. First,
 # we define the potential branches (modules) as "trunk" and "branches"
 # children directories. Revisions are then identified by their
 # module and revision number (and a repository identifier).
 #
 # The revision graph is really a tree (or a forest). By default, a
 # revision parent is the previous revision in the same module. If the
 # module directory is copied/moved from another module then the
 # revision is the module root and its parent the source revision in
 # the parent module. A revision has at most one parent.
 #
 class svn_source(converter_source):
     def __init__(self, ui, url, rev=None):
         super(svn_source, self).__init__(ui, url, rev=rev)
 
         if not (url.startswith('svn://') or url.startswith('svn+ssh://') or
                 (os.path.exists(url) and
                  os.path.exists(os.path.join(url, '.svn'))) or
                 issvnurl(ui, url)):
             raise NoRepo(_("%s does not look like a Subversion repository")
                          % url)
         if svn is None:
             raise MissingTool(_('could not load Subversion python bindings'))
 
         try:
             version = svn.core.SVN_VER_MAJOR, svn.core.SVN_VER_MINOR
             if version < (1, 4):
                 raise MissingTool(_('Subversion python bindings %d.%d found, '
                                     '1.4 or later required') % version)
         except AttributeError:
             raise MissingTool(_('Subversion python bindings are too old, 1.4 '
                                 'or later required'))
 
         self.lastrevs = {}
 
         latest = None
         try:
             # Support file://path@rev syntax. Useful e.g. to convert
             # deleted branches.
             at = url.rfind('@')
             if at >= 0:
                 latest = int(url[at + 1:])
                 url = url[:at]
         except ValueError:
             pass
         self.url = geturl(url)
         self.encoding = 'UTF-8' # Subversion is always nominal UTF-8
         try:
             self.transport = transport.SvnRaTransport(url=self.url)
             self.ra = self.transport.ra
             self.ctx = self.transport.client
             self.baseurl = svn.ra.get_repos_root(self.ra)
             # Module is either empty or a repository path starting with
             # a slash and not ending with a slash.
             self.module = urllib.unquote(self.url[len(self.baseurl):])
             self.prevmodule = None
             self.rootmodule = self.module
             self.commits = {}
             self.paths = {}
             self.uuid = svn.ra.get_uuid(self.ra)
         except SubversionException:
             ui.traceback()
             raise NoRepo(_("%s does not look like a Subversion repository")
                          % self.url)
 
         if rev:
             try:
                 latest = int(rev)
             except ValueError:
                 raise util.Abort(_('svn: revision %s is not an integer') % rev)
 
         self.trunkname = self.ui.config('convert', 'svn.trunk',
                                         'trunk').strip('/')
         self.startrev = self.ui.config('convert', 'svn.startrev', default=0)
         try:
             self.startrev = int(self.startrev)
             if self.startrev < 0:
                 self.startrev = 0
         except ValueError:
             raise util.Abort(_('svn: start revision %s is not an integer')
                              % self.startrev)
 
         try:
             self.head = self.latest(self.module, latest)
         except SvnPathNotFound:
             self.head = None
         if not self.head:
             raise util.Abort(_('no revision found in module %s')
                              % self.module)
         self.last_changed = self.revnum(self.head)
 
         self._changescache = (None, None)
 
         if os.path.exists(os.path.join(url, '.svn/entries')):
             self.wc = url
         else:
             self.wc = None
         self.convertfp = None
 
     def setrevmap(self, revmap):
         lastrevs = {}
         for revid in revmap.iterkeys():
             uuid, module, revnum = revsplit(revid)
             lastrevnum = lastrevs.setdefault(module, revnum)
             if revnum > lastrevnum:
                 lastrevs[module] = revnum
         self.lastrevs = lastrevs
 
     def exists(self, path, optrev):
         try:
             svn.client.ls(self.url.rstrip('/') + '/' + quote(path),
                                  optrev, False, self.ctx)
             return True
         except SubversionException:
             return False
 
     def getheads(self):
 
         def isdir(path, revnum):
             kind = self._checkpath(path, revnum)
             return kind == svn.core.svn_node_dir
 
         def getcfgpath(name, rev):
             cfgpath = self.ui.config('convert', 'svn.' + name)
             if cfgpath is not None and cfgpath.strip() == '':
                 return None
             path = (cfgpath or name).strip('/')
             if not self.exists(path, rev):
                 if self.module.endswith(path) and name == 'trunk':
                     # we are converting from inside this directory
                     return None
                 if cfgpath:
                     raise util.Abort(_('expected %s to be at %r, but not found')
                                  % (name, path))
                 return None
             self.ui.note(_('found %s at %r\n') % (name, path))
             return path
 
         rev = optrev(self.last_changed)
         oldmodule = ''
         trunk = getcfgpath('trunk', rev)
         self.tags = getcfgpath('tags', rev)
         branches = getcfgpath('branches', rev)
 
         # If the project has a trunk or branches, we will extract heads
         # from them. We keep the project root otherwise.
         if trunk:
             oldmodule = self.module or ''
             self.module += '/' + trunk
             self.head = self.latest(self.module, self.last_changed)
             if not self.head:
                 raise util.Abort(_('no revision found in module %s')
                                  % self.module)
 
         # First head in the list is the module's head
         self.heads = [self.head]
         if self.tags is not None:
             self.tags = '%s/%s' % (oldmodule , (self.tags or 'tags'))
 
         # Check if branches bring a few more heads to the list
         if branches:
             rpath = self.url.strip('/')
             branchnames = svn.client.ls(rpath + '/' + quote(branches),
                                         rev, False, self.ctx)
             for branch in sorted(branchnames):
                 module = '%s/%s/%s' % (oldmodule, branches, branch)
                 if not isdir(module, self.last_changed):
                     continue
                 brevid = self.latest(module, self.last_changed)
                 if not brevid:
                     self.ui.note(_('ignoring empty branch %s\n') % branch)
                     continue
                 self.ui.note(_('found branch %s at %d\n') %
                              (branch, self.revnum(brevid)))
                 self.heads.append(brevid)
 
         if self.startrev and self.heads:
             if len(self.heads) > 1:
                 raise util.Abort(_('svn: start revision is not supported '
                                    'with more than one branch'))
             revnum = self.revnum(self.heads[0])
             if revnum < self.startrev:
                 raise util.Abort(
                     _('svn: no revision found after start revision %d')
                                  % self.startrev)
 
         return self.heads
 
     def _getchanges(self, rev, full):
         (paths, parents) = self.paths[rev]
         copies = {}
         if parents:
             files, self.removed, copies = self.expandpaths(rev, paths, parents)
         if full or not parents:
             # Perform a full checkout on roots
             uuid, module, revnum = revsplit(rev)
             entries = svn.client.ls(self.baseurl + quote(module),
                                     optrev(revnum), True, self.ctx)
             files = [n for n, e in entries.iteritems()
                      if e.kind == svn.core.svn_node_file]
             self.removed = set()
 
         files.sort()
         files = zip(files, [rev] * len(files))
         return (files, copies)
 
     def getchanges(self, rev, full):
         # reuse cache from getchangedfiles
         if self._changescache[0] == rev and not full:
             (files, copies) = self._changescache[1]
         else:
             (files, copies) = self._getchanges(rev, full)
             # caller caches the result, so free it here to release memory
             del self.paths[rev]
         return (files, copies)
 
     def getchangedfiles(self, rev, i):
         # called from filemap - cache computed values for reuse in getchanges
         (files, copies) = self._getchanges(rev, False)
         self._changescache = (rev, (files, copies))
         return [f[0] for f in files]
 
     def getcommit(self, rev):
         if rev not in self.commits:
             uuid, module, revnum = revsplit(rev)
             self.module = module
             self.reparent(module)
             # We assume that:
             # - requests for revisions after "stop" come from the
             # revision graph backward traversal. Cache all of them
             # down to stop, they will be used eventually.
             # - requests for revisions before "stop" come to get
             # isolated branches parents. Just fetch what is needed.
             stop = self.lastrevs.get(module, 0)
             if revnum < stop:
                 stop = revnum + 1
             self._fetch_revisions(revnum, stop)
             if rev not in self.commits:
                 raise util.Abort(_('svn: revision %s not found') % revnum)
         revcommit = self.commits[rev]
         # caller caches the result, so free it here to release memory
         del self.commits[rev]
         return revcommit
 
     def checkrevformat(self, revstr, mapname='splicemap'):
         """ fails if revision format does not match the correct format"""
         if not re.match(r'svn:[0-9a-f]{8,8}-[0-9a-f]{4,4}-'
                               '[0-9a-f]{4,4}-[0-9a-f]{4,4}-[0-9a-f]'
                               '{12,12}(.*)\@[0-9]+$',revstr):
             raise util.Abort(_('%s entry %s is not a valid revision'
                                ' identifier') % (mapname, revstr))
 
+    def numcommits(self):
+        return int(self.head.rsplit('@', 1)[1]) - self.startrev
+
     def gettags(self):
         tags = {}
         if self.tags is None:
             return tags
 
         # svn tags are just a convention, project branches left in a
         # 'tags' directory. There is no other relationship than
         # ancestry, which is expensive to discover and makes them hard
         # to update incrementally.  Worse, past revisions may be
         # referenced by tags far away in the future, requiring a deep
         # history traversal on every calculation.  Current code
         # performs a single backward traversal, tracking moves within
         # the tags directory (tag renaming) and recording a new tag
         # everytime a project is copied from outside the tags
         # directory. It also lists deleted tags, this behaviour may
         # change in the future.
         pendings = []
         tagspath = self.tags
         start = svn.ra.get_latest_revnum(self.ra)
         stream = self._getlog([self.tags], start, self.startrev)
         try:
             for entry in stream:
                 origpaths, revnum, author, date, message = entry
                 if not origpaths:
                     origpaths = []
                 copies = [(e.copyfrom_path, e.copyfrom_rev, p) for p, e
                           in origpaths.iteritems() if e.copyfrom_path]
                 # Apply moves/copies from more specific to general
                 copies.sort(reverse=True)
 
                 srctagspath = tagspath
                 if copies and copies[-1][2] == tagspath:
                     # Track tags directory moves
                     srctagspath = copies.pop()[0]
 
                 for source, sourcerev, dest in copies:
                     if not dest.startswith(tagspath + '/'):
                         continue
                     for tag in pendings:
                         if tag[0].startswith(dest):
                             tagpath = source + tag[0][len(dest):]
                             tag[:2] = [tagpath, sourcerev]
                             break
                     else:
                         pendings.append([source, sourcerev, dest])
 
                 # Filter out tags with children coming from different
                 # parts of the repository like:
                 # /tags/tag.1 (from /trunk:10)
                 # /tags/tag.1/foo (from /branches/foo:12)
                 # Here/tags/tag.1 discarded as well as its children.
                 # It happens with tools like cvs2svn. Such tags cannot
                 # be represented in mercurial.
                 addeds = dict((p, e.copyfrom_path) for p, e
                               in origpaths.iteritems()
                               if e.action == 'A' and e.copyfrom_path)
                 badroots = set()
                 for destroot in addeds:
                     for source, sourcerev, dest in pendings:
                         if (not dest.startswith(destroot + '/')
                             or source.startswith(addeds[destroot] + '/')):
                             continue
                         badroots.add(destroot)
                         break
 
                 for badroot in badroots:
                     pendings = [p for p in pendings if p[2] != badroot
                                 and not p[2].startswith(badroot + '/')]
 
                 # Tell tag renamings from tag creations
                 renamings = []
                 for source, sourcerev, dest in pendings:
                     tagname = dest.split('/')[-1]
                     if source.startswith(srctagspath):
                         renamings.append([source, sourcerev, tagname])
                         continue
                     if tagname in tags:
                         # Keep the latest tag value
                         continue
                     # From revision may be fake, get one with changes
                     try:
                         tagid = self.latest(source, sourcerev)
                         if tagid and tagname not in tags:
                             tags[tagname] = tagid
                     except SvnPathNotFound:
                         # It happens when we are following directories
                         # we assumed were copied with their parents
                         # but were really created in the tag
                         # directory.
                         pass
                 pendings = renamings
                 tagspath = srctagspath
         finally:
             stream.close()
         return tags
 
     def converted(self, rev, destrev):
         if not self.wc:
             return
         if self.convertfp is None:
             self.convertfp = open(os.path.join(self.wc, '.svn', 'hg-shamap'),
                                   'a')
         self.convertfp.write('%s %d\n' % (destrev, self.revnum(rev)))
         self.convertfp.flush()
 
     def revid(self, revnum, module=None):
         return 'svn:%s%s@%s' % (self.uuid, module or self.module, revnum)
 
     def revnum(self, rev):
         return int(rev.split('@')[-1])
 
     def latest(self, path, stop=None):
         """Find the latest revid affecting path, up to stop revision
         number. If stop is None, default to repository latest
         revision. It may return a revision in a different module,
         since a branch may be moved without a change being
         reported. Return None if computed module does not belong to
         rootmodule subtree.
         """
         def findchanges(path, start, stop=None):
             stream = self._getlog([path], start, stop or 1)
             try:
                 for entry in stream:
                     paths, revnum, author, date, message = entry
                     if stop is None and paths:
                         # We do not know the latest changed revision,
                         # keep the first one with changed paths.
                         break
                     if revnum <= stop:
                         break
 
                     for p in paths:
                         if (not path.startswith(p) or
                             not paths[p].copyfrom_path):
                             continue
                         newpath = paths[p].copyfrom_path + path[len(p):]
                         self.ui.debug("branch renamed from %s to %s at %d\n" %
                                       (path, newpath, revnum))
                         path = newpath
                         break
                 if not paths:
                     revnum = None
                 return revnum, path
             finally:
                 stream.close()
 
         if not path.startswith(self.rootmodule):
             # Requests on foreign branches may be forbidden at server level
             self.ui.debug('ignoring foreign branch %r\n' % path)
             return None
 
         if stop is None:
             stop = svn.ra.get_latest_revnum(self.ra)
         try:
             prevmodule = self.reparent('')
             dirent = svn.ra.stat(self.ra, path.strip('/'), stop)
             self.reparent(prevmodule)
         except SubversionException:
             dirent = None
         if not dirent:
             raise SvnPathNotFound(_('%s not found up to revision %d')
                                   % (path, stop))
 
         # stat() gives us the previous revision on this line of
         # development, but it might be in *another module*. Fetch the
         # log and detect renames down to the latest revision.
         revnum, realpath = findchanges(path, stop, dirent.created_rev)
         if revnum is None:
             # Tools like svnsync can create empty revision, when
             # synchronizing only a subtree for instance. These empty
             # revisions created_rev still have their original values
             # despite all changes having disappeared and can be
             # returned by ra.stat(), at least when stating the root
             # module. In that case, do not trust created_rev and scan
             # the whole history.
             revnum, realpath = findchanges(path, stop)
             if revnum is None:
                 self.ui.debug('ignoring empty branch %r\n' % realpath)
                 return None
 
         if not realpath.startswith(self.rootmodule):
             self.ui.debug('ignoring foreign branch %r\n' % realpath)
             return None
         return self.revid(revnum, realpath)
 
     def reparent(self, module):
         """Reparent the svn transport and return the previous parent."""
         if self.prevmodule == module:
             return module
         svnurl = self.baseurl + quote(module)
         prevmodule = self.prevmodule
         if prevmodule is None:
             prevmodule = ''
         self.ui.debug("reparent to %s\n" % svnurl)
         svn.ra.reparent(self.ra, svnurl)
         self.prevmodule = module
         return prevmodule
 
     def expandpaths(self, rev, paths, parents):
         changed, removed = set(), set()
         copies = {}
 
         new_module, revnum = revsplit(rev)[1:]
         if new_module != self.module:
             self.module = new_module
             self.reparent(self.module)
 
         for i, (path, ent) in enumerate(paths):
             self.ui.progress(_('scanning paths'), i, item=path,
                              total=len(paths))
             entrypath = self.getrelpath(path)
 
             kind = self._checkpath(entrypath, revnum)
             if kind == svn.core.svn_node_file:
                 changed.add(self.recode(entrypath))
                 if not ent.copyfrom_path or not parents:
                     continue
                 # Copy sources not in parent revisions cannot be
                 # represented, ignore their origin for now
                 pmodule, prevnum = revsplit(parents[0])[1:]
                 if ent.copyfrom_rev < prevnum:
                     continue
                 copyfrom_path = self.getrelpath(ent.copyfrom_path, pmodule)
                 if not copyfrom_path:
                     continue
                 self.ui.debug("copied to %s from %s@%s\n" %
                               (entrypath, copyfrom_path, ent.copyfrom_rev))
                 copies[self.recode(entrypath)] = self.recode(copyfrom_path)
             elif kind == 0: # gone, but had better be a deleted *file*
                 self.ui.debug("gone from %s\n" % ent.copyfrom_rev)
                 pmodule, prevnum = revsplit(parents[0])[1:]
                 parentpath = pmodule + "/" + entrypath
                 fromkind = self._checkpath(entrypath, prevnum, pmodule)
 
                 if fromkind == svn.core.svn_node_file:
                     removed.add(self.recode(entrypath))
                 elif fromkind == svn.core.svn_node_dir:
                     oroot = parentpath.strip('/')
                     nroot = path.strip('/')
                     children = self._iterfiles(oroot, prevnum)
                     for childpath in children:
                         childpath = childpath.replace(oroot, nroot)
                         childpath = self.getrelpath("/" + childpath, pmodule)
                         if childpath:
                             removed.add(self.recode(childpath))
                 else:
                     self.ui.debug('unknown path in revision %d: %s\n' % \
                                   (revnum, path))
             elif kind == svn.core.svn_node_dir:
                 if ent.action == 'M':
                     # If the directory just had a prop change,
                     # then we shouldn't need to look for its children.
                     continue
                 if ent.action == 'R' and parents:
                     # If a directory is replacing a file, mark the previous
                     # file as deleted
                     pmodule, prevnum = revsplit(parents[0])[1:]
                     pkind = self._checkpath(entrypath, prevnum, pmodule)
                     if pkind == svn.core.svn_node_file:
                         removed.add(self.recode(entrypath))
                     elif pkind == svn.core.svn_node_dir:
                         # We do not know what files were kept or removed,
                         # mark them all as changed.
                         for childpath in self._iterfiles(pmodule, prevnum):
                             childpath = self.getrelpath("/" + childpath)
                             if childpath:
                                 changed.add(self.recode(childpath))
 
                 for childpath in self._iterfiles(path, revnum):
                     childpath = self.getrelpath("/" + childpath)
                     if childpath:
                         changed.add(self.recode(childpath))
 
                 # Handle directory copies
                 if not ent.copyfrom_path or not parents:
                     continue
                 # Copy sources not in parent revisions cannot be
                 # represented, ignore their origin for now
                 pmodule, prevnum = revsplit(parents[0])[1:]
                 if ent.copyfrom_rev < prevnum:
                     continue
                 copyfrompath = self.getrelpath(ent.copyfrom_path, pmodule)
                 if not copyfrompath:
                     continue
                 self.ui.debug("mark %s came from %s:%d\n"
                               % (path, copyfrompath, ent.copyfrom_rev))
                 children = self._iterfiles(ent.copyfrom_path, ent.copyfrom_rev)
                 for childpath in children:
                     childpath = self.getrelpath("/" + childpath, pmodule)
                     if not childpath:
                         continue
                     copytopath = path + childpath[len(copyfrompath):]
                     copytopath = self.getrelpath(copytopath)
                     copies[self.recode(copytopath)] = self.recode(childpath)
 
         self.ui.progress(_('scanning paths'), None)
         changed.update(removed)
         return (list(changed), removed, copies)
 
     def _fetch_revisions(self, from_revnum, to_revnum):
         if from_revnum < to_revnum:
             from_revnum, to_revnum = to_revnum, from_revnum
 
         self.child_cset = None
 
         def parselogentry(orig_paths, revnum, author, date, message):
             """Return the parsed commit object or None, and True if
             the revision is a branch root.
             """
             self.ui.debug("parsing revision %d (%d changes)\n" %
                           (revnum, len(orig_paths)))
 
             branched = False
             rev = self.revid(revnum)
             # branch log might return entries for a parent we already have
 
             if rev in self.commits or revnum < to_revnum:
                 return None, branched
 
             parents = []
             # check whether this revision is the start of a branch or part
             # of a branch renaming
             orig_paths = sorted(orig_paths.iteritems())
             root_paths = [(p, e) for p, e in orig_paths
                           if self.module.startswith(p)]
             if root_paths:
                 path, ent = root_paths[-1]
                 if ent.copyfrom_path:
                     branched = True
                     newpath = ent.copyfrom_path + self.module[len(path):]
                     # ent.copyfrom_rev may not be the actual last revision
                     previd = self.latest(newpath, ent.copyfrom_rev)
                     if previd is not None:
                         prevmodule, prevnum = revsplit(previd)[1:]
                         if prevnum >= self.startrev:
                             parents = [previd]
                             self.ui.note(
                                 _('found parent of branch %s at %d: %s\n') %
                                 (self.module, prevnum, prevmodule))
                 else:
                     self.ui.debug("no copyfrom path, don't know what to do.\n")
 
             paths = []
             # filter out unrelated paths
             for path, ent in orig_paths:
                 if self.getrelpath(path) is None:
                     continue
                 paths.append((path, ent))
 
             # Example SVN datetime. Includes microseconds.
             # ISO-8601 conformant
             # '2007-01-04T17:35:00.902377Z'
             date = util.parsedate(date[:19] + " UTC", ["%Y-%m-%dT%H:%M:%S"])
             if self.ui.configbool('convert', 'localtimezone'):
                 date = makedatetimestamp(date[0])
 
             log = message and self.recode(message) or ''
             author = author and self.recode(author) or ''
             try:
                 branch = self.module.split("/")[-1]
                 if branch == self.trunkname:
                     branch = None
             except IndexError:
                 branch = None
 
             cset = commit(author=author,
                           date=util.datestr(date, '%Y-%m-%d %H:%M:%S %1%2'),
                           desc=log,
                           parents=parents,
                           branch=branch,
                           rev=rev)
 
             self.commits[rev] = cset
             # The parents list is *shared* among self.paths and the
             # commit object. Both will be updated below.
             self.paths[rev] = (paths, cset.parents)
             if self.child_cset and not self.child_cset.parents:
                 self.child_cset.parents[:] = [rev]
             self.child_cset = cset
             return cset, branched
 
         self.ui.note(_('fetching revision log for "%s" from %d to %d\n') %
                      (self.module, from_revnum, to_revnum))
 
         try:
             firstcset = None
             lastonbranch = False
             stream = self._getlog([self.module], from_revnum, to_revnum)
             try:
                 for entry in stream:
                     paths, revnum, author, date, message = entry
                     if revnum < self.startrev:
                         lastonbranch = True
                         break
                     if not paths:
                         self.ui.debug('revision %d has no entries\n' % revnum)
                         # If we ever leave the loop on an empty
                         # revision, do not try to get a parent branch
                         lastonbranch = lastonbranch or revnum == 0
                         continue
                     cset, lastonbranch = parselogentry(paths, revnum, author,
                                                        date, message)
                     if cset:
                         firstcset = cset
                     if lastonbranch:
                         break
             finally:
                 stream.close()
 
             if not lastonbranch and firstcset and not firstcset.parents:
                 # The first revision of the sequence (the last fetched one)
                 # has invalid parents if not a branch root. Find the parent
                 # revision now, if any.
                 try:
                     firstrevnum = self.revnum(firstcset.rev)
                     if firstrevnum > 1:
                         latest = self.latest(self.module, firstrevnum - 1)
                         if latest:
                             firstcset.parents.append(latest)
                 except SvnPathNotFound:
                     pass
         except SubversionException, (inst, num):
             if num == svn.core.SVN_ERR_FS_NO_SUCH_REVISION:
                 raise util.Abort(_('svn: branch has no revision %s')
                                  % to_revnum)
             raise
 
     def getfile(self, file, rev):
         # TODO: ra.get_file transmits the whole file instead of diffs.
         if file in self.removed:
             return None, None
         mode = ''
         try:
             new_module, revnum = revsplit(rev)[1:]
             if self.module != new_module:
                 self.module = new_module
                 self.reparent(self.module)
             io = StringIO()
             info = svn.ra.get_file(self.ra, file, revnum, io)
             data = io.getvalue()
             # ra.get_file() seems to keep a reference on the input buffer
             # preventing collection. Release it explicitly.
             io.close()
             if isinstance(info, list):
                 info = info[-1]
             mode = ("svn:executable" in info) and 'x' or ''
             mode = ("svn:special" in info) and 'l' or mode
         except SubversionException, e:
             notfound = (svn.core.SVN_ERR_FS_NOT_FOUND,
                 svn.core.SVN_ERR_RA_DAV_PATH_NOT_FOUND)
             if e.apr_err in notfound: # File not found
                 return None, None
             raise
         if mode == 'l':
             link_prefix = "link "
             if data.startswith(link_prefix):
                 data = data[len(link_prefix):]
         return data, mode
 
     def _iterfiles(self, path, revnum):
         """Enumerate all files in path at revnum, recursively."""
         path = path.strip('/')
         pool = Pool()
         rpath = '/'.join([self.baseurl, quote(path)]).strip('/')
         entries = svn.client.ls(rpath, optrev(revnum), True, self.ctx, pool)
         if path:
             path += '/'
         return ((path + p) for p, e in entries.iteritems()
                 if e.kind == svn.core.svn_node_file)
 
     def getrelpath(self, path, module=None):
         if module is None:
             module = self.module
         # Given the repository url of this wc, say
         #   "http://server/plone/CMFPlone/branches/Plone-2_0-branch"
         # extract the "entry" portion (a relative path) from what
         # svn log --xml says, i.e.
         #   "/CMFPlone/branches/Plone-2_0-branch/tests/PloneTestCase.py"
         # that is to say "tests/PloneTestCase.py"
         if path.startswith(module):
             relative = path.rstrip('/')[len(module):]
             if relative.startswith('/'):
                 return relative[1:]
             elif relative == '':
                 return relative
 
         # The path is outside our tracked tree...
         self.ui.debug('%r is not under %r, ignoring\n' % (path, module))
         return None
 
     def _checkpath(self, path, revnum, module=None):
         if module is not None:
             prevmodule = self.reparent('')
             path = module + '/' + path
         try:
             # ra.check_path does not like leading slashes very much, it leads
             # to PROPFIND subversion errors
             return svn.ra.check_path(self.ra, path.strip('/'), revnum)
         finally:
             if module is not None:
                 self.reparent(prevmodule)
 
     def _getlog(self, paths, start, end, limit=0, discover_changed_paths=True,
                 strict_node_history=False):
         # Normalize path names, svn >= 1.5 only wants paths relative to
         # supplied URL
         relpaths = []
         for p in paths:
             if not p.startswith('/'):
                 p = self.module + '/' + p
             relpaths.append(p.strip('/'))
         args = [self.baseurl, relpaths, start, end, limit,
                 discover_changed_paths, strict_node_history]
         # undocumented feature: debugsvnlog can be disabled
         if not self.ui.configbool('convert', 'svn.debugsvnlog', True):
             return directlogstream(*args)
         arg = encodeargs(args)
         hgexe = util.hgexecutable()
         cmd = '%s debugsvnlog' % util.shellquote(hgexe)
         stdin, stdout = util.popen2(util.quotecommand(cmd))
         stdin.write(arg)
         try:
             stdin.close()
         except IOError:
             raise util.Abort(_('Mercurial failed to run itself, check'
                                ' hg executable is in PATH'))
         return logstream(stdout)
 
 pre_revprop_change = '''#!/bin/sh
 
 REPOS="$1"
 REV="$2"
 USER="$3"
 PROPNAME="$4"
 ACTION="$5"
 
 if [ "$ACTION" = "M" -a "$PROPNAME" = "svn:log" ]; then exit 0; fi
 if [ "$ACTION" = "A" -a "$PROPNAME" = "hg:convert-branch" ]; then exit 0; fi
 if [ "$ACTION" = "A" -a "$PROPNAME" = "hg:convert-rev" ]; then exit 0; fi
 
 echo "Changing prohibited revision property" >&2
 exit 1
 '''
 
 class svn_sink(converter_sink, commandline):
     commit_re = re.compile(r'Committed revision (\d+).', re.M)
     uuid_re = re.compile(r'Repository UUID:\s*(\S+)', re.M)
 
     def prerun(self):
         if self.wc:
             os.chdir(self.wc)
 
     def postrun(self):
         if self.wc:
             os.chdir(self.cwd)
 
     def join(self, name):
         return os.path.join(self.wc, '.svn', name)
 
     def revmapfile(self):
         return self.join('hg-shamap')
 
     def authorfile(self):
         return self.join('hg-authormap')
 
     def __init__(self, ui, path):
 
         converter_sink.__init__(self, ui, path)
         commandline.__init__(self, ui, 'svn')
         self.delete = []
         self.setexec = []
         self.delexec = []
         self.copies = []
         self.wc = None
         self.cwd = os.getcwd()
 
         created = False
         if os.path.isfile(os.path.join(path, '.svn', 'entries')):
             self.wc = os.path.realpath(path)
             self.run0('update')
         else:
             if not re.search(r'^(file|http|https|svn|svn\+ssh)\://', path):
                 path = os.path.realpath(path)
                 if os.path.isdir(os.path.dirname(path)):
                     if not os.path.exists(os.path.join(path, 'db', 'fs-type')):
                         ui.status(_('initializing svn repository %r\n') %
                                   os.path.basename(path))
                         commandline(ui, 'svnadmin').run0('create', path)
                         created = path
                     path = util.normpath(path)
                     if not path.startswith('/'):
                         path = '/' + path
                     path = 'file://' + path
 
             wcpath = os.path.join(os.getcwd(), os.path.basename(path) + '-wc')
             ui.status(_('initializing svn working copy %r\n')
                       % os.path.basename(wcpath))
             self.run0('checkout', path, wcpath)
 
             self.wc = wcpath
         self.opener = scmutil.opener(self.wc)
         self.wopener = scmutil.opener(self.wc)
         self.childmap = mapfile(ui, self.join('hg-childmap'))
         self.is_exec = util.checkexec(self.wc) and util.isexec or None
 
         if created:
             hook = os.path.join(created, 'hooks', 'pre-revprop-change')
             fp = open(hook, 'w')
             fp.write(pre_revprop_change)
             fp.close()
             util.setflags(hook, False, True)
 
         output = self.run0('info')
         self.uuid = self.uuid_re.search(output).group(1).strip()
 
     def wjoin(self, *names):
         return os.path.join(self.wc, *names)
 
     @propertycache
     def manifest(self):
         # As of svn 1.7, the "add" command fails when receiving
         # already tracked entries, so we have to track and filter them
         # ourselves.
         m = set()
         output = self.run0('ls', recursive=True, xml=True)
         doc = xml.dom.minidom.parseString(output)
         for e in doc.getElementsByTagName('entry'):
             for n in e.childNodes:
                 if n.nodeType != n.ELEMENT_NODE or n.tagName != 'name':
                     continue
                 name = ''.join(c.data for c in n.childNodes
                                if c.nodeType == c.TEXT_NODE)
                 # Entries are compared with names coming from
                 # mercurial, so bytes with undefined encoding. Our
                 # best bet is to assume they are in local
                 # encoding. They will be passed to command line calls
                 # later anyway, so they better be.
                 m.add(encoding.tolocal(name.encode('utf-8')))
                 break
         return m
 
     def putfile(self, filename, flags, data):
         if 'l' in flags:
             self.wopener.symlink(data, filename)
         else:
             try:
                 if os.path.islink(self.wjoin(filename)):
                     os.unlink(filename)
             except OSError:
                 pass
             self.wopener.write(filename, data)
 
             if self.is_exec:
                 if self.is_exec(self.wjoin(filename)):
                     if 'x' not in flags:
                         self.delexec.append(filename)
                 else:
                     if 'x' in flags:
                         self.setexec.append(filename)
                 util.setflags(self.wjoin(filename), False, 'x' in flags)
 
     def _copyfile(self, source, dest):
         # SVN's copy command pukes if the destination file exists, but
         # our copyfile method expects to record a copy that has
         # already occurred.  Cross the semantic gap.
         wdest = self.wjoin(dest)
         exists = os.path.lexists(wdest)
         if exists:
             fd, tempname = tempfile.mkstemp(
                 prefix='hg-copy-', dir=os.path.dirname(wdest))
             os.close(fd)
             os.unlink(tempname)
             os.rename(wdest, tempname)
         try:
             self.run0('copy', source, dest)
         finally:
             self.manifest.add(dest)
             if exists:
                 try:
                     os.unlink(wdest)
                 except OSError:
                     pass
                 os.rename(tempname, wdest)
 
     def dirs_of(self, files):
         dirs = set()
         for f in files:
             if os.path.isdir(self.wjoin(f)):
                 dirs.add(f)
             for i in strutil.rfindall(f, '/'):
                 dirs.add(f[:i])
         return dirs
 
     def add_dirs(self, files):
         add_dirs = [d for d in sorted(self.dirs_of(files))
                     if d not in self.manifest]
         if add_dirs:
             self.manifest.update(add_dirs)
             self.xargs(add_dirs, 'add', non_recursive=True, quiet=True)
         return add_dirs
 
     def add_files(self, files):
         files = [f for f in files if f not in self.manifest]
         if files:
             self.manifest.update(files)
             self.xargs(files, 'add', quiet=True)
         return files
 
     def addchild(self, parent, child):
         self.childmap[parent] = child
 
     def revid(self, rev):
         return u"svn:%s@%s" % (self.uuid, rev)
 
     def putcommit(self, files, copies, parents, commit, source, revmap, full):
         for parent in parents:
             try:
                 return self.revid(self.childmap[parent])
             except KeyError:
                 pass
 
         # Apply changes to working copy
         for f, v in files:
             data, mode = source.getfile(f, v)
             if data is None:
                 self.delete.append(f)
             else:
                 self.putfile(f, mode, data)
                 if f in copies:
                     self.copies.append([copies[f], f])
         if full:
             self.delete.extend(sorted(self.manifest.difference(files)))
         files = [f[0] for f in files]
 
         entries = set(self.delete)
         files = frozenset(files)
         entries.update(self.add_dirs(files.difference(entries)))
         if self.copies:
             for s, d in self.copies:
                 self._copyfile(s, d)
             self.copies = []
         if self.delete:
             self.xargs(self.delete, 'delete')
             for f in self.delete:
                 self.manifest.remove(f)
             self.delete = []
         entries.update(self.add_files(files.difference(entries)))
         if self.delexec:
             self.xargs(self.delexec, 'propdel', 'svn:executable')
             self.delexec = []
         if self.setexec:
             self.xargs(self.setexec, 'propset', 'svn:executable', '*')
             self.setexec = []
 
         fd, messagefile = tempfile.mkstemp(prefix='hg-convert-')
         fp = os.fdopen(fd, 'w')
         fp.write(commit.desc)
         fp.close()
         try:
             output = self.run0('commit',
                                username=util.shortuser(commit.author),
                                file=messagefile,
                                encoding='utf-8')
             try:
                 rev = self.commit_re.search(output).group(1)
             except AttributeError:
                 if not files:
                     return parents[0]
                 self.ui.warn(_('unexpected svn output:\n'))
                 self.ui.warn(output)
                 raise util.Abort(_('unable to cope with svn output'))
             if commit.rev:
                 self.run('propset', 'hg:convert-rev', commit.rev,
                          revprop=True, revision=rev)
             if commit.branch and commit.branch != 'default':
                 self.run('propset', 'hg:convert-branch', commit.branch,
                          revprop=True, revision=rev)
             for parent in parents:
                 self.addchild(parent, rev)
             return self.revid(rev)
         finally:
             os.unlink(messagefile)
 
     def puttags(self, tags):
         self.ui.warn(_('writing Subversion tags is not yet implemented\n'))
         return None, None
 
     def hascommitfrommap(self, rev):
         # We trust that revisions referenced in a map still is present
         # TODO: implement something better if necessary and feasible
         return True
 
     def hascommitforsplicemap(self, rev):
         # This is not correct as one can convert to an existing subversion
         # repository and childmap would not list all revisions. Too bad.
         if rev in self.childmap:
             return True
         raise util.Abort(_('splice map revision %s not found in subversion '
                            'child map (revision lookups are not implemented)')
                          % rev)

tests/test-convert-svn-encoding.t

@@ -1,133 +1,133 b''
 #require svn svn-bindings
 
   $ cat >> $HGRCPATH <<EOF
   > [extensions]
   > convert =
   > EOF
 
   $ svnadmin create svn-repo
   $ svnadmin load -q svn-repo < "$TESTDIR/svn/encoding.svndump"
 
 Convert while testing all possible outputs
 
   $ hg --debug convert svn-repo A-hg
   initializing destination A-hg repository
   reparent to file://*/svn-repo (glob)
   run hg sink pre-conversion action
   scanning source...
   found trunk at 'trunk'
   found tags at 'tags'
   found branches at 'branches'
   found branch branch\xc3\xa9 at 5 (esc)
   found branch branch\xc3\xa9e at 6 (esc)
-  scanning: 1 revisions
+  scanning: 1/4 revisions (25.00%)
   reparent to file://*/svn-repo/trunk (glob)
   fetching revision log for "/trunk" from 4 to 0
   parsing revision 4 (2 changes)
   parsing revision 3 (4 changes)
   parsing revision 2 (3 changes)
   parsing revision 1 (3 changes)
   no copyfrom path, don't know what to do.
   '/branches' is not under '/trunk', ignoring
   '/tags' is not under '/trunk', ignoring
-  scanning: 2 revisions
+  scanning: 2/4 revisions (50.00%)
   reparent to file://*/svn-repo/branches/branch%C3%A9 (glob)
   fetching revision log for "/branches/branch\xc3\xa9" from 5 to 0 (esc)
   parsing revision 5 (1 changes)
   reparent to file://*/svn-repo (glob)
   reparent to file://*/svn-repo/branches/branch%C3%A9 (glob)
   found parent of branch /branches/branch\xc3\xa9 at 4: /trunk (esc)
-  scanning: 3 revisions
+  scanning: 3/4 revisions (75.00%)
   reparent to file://*/svn-repo/branches/branch%C3%A9e (glob)
   fetching revision log for "/branches/branch\xc3\xa9e" from 6 to 0 (esc)
   parsing revision 6 (1 changes)
   reparent to file://*/svn-repo (glob)
   reparent to file://*/svn-repo/branches/branch%C3%A9e (glob)
   found parent of branch /branches/branch\xc3\xa9e at 5: /branches/branch\xc3\xa9 (esc)
-  scanning: 4 revisions
-  scanning: 5 revisions
-  scanning: 6 revisions
+  scanning: 4/4 revisions (100.00%)
+  scanning: 5/4 revisions (125.00%)
+  scanning: 6/4 revisions (150.00%)
   sorting...
   converting...
   5 init projA
   source: svn:afeb9c47-92ff-4c0c-9f72-e1f6eb8ac9af/trunk@1
   converting: 0/6 revisions (0.00%)
   4 hello
   source: svn:afeb9c47-92ff-4c0c-9f72-e1f6eb8ac9af/trunk@2
   converting: 1/6 revisions (16.67%)
   reparent to file://*/svn-repo/trunk (glob)
   scanning paths: /trunk/\xc3\xa0 0/3 (0.00%) (esc)
   scanning paths: /trunk/\xc3\xa0/e\xcc\x81 1/3 (33.33%) (esc)
   scanning paths: /trunk/\xc3\xa9 2/3 (66.67%) (esc)
   \xc3\xa0/e\xcc\x81 (esc)
   getting files: \xc3\xa0/e\xcc\x81 1/2 (50.00%) (esc)
   \xc3\xa9 (esc)
   getting files: \xc3\xa9 2/2 (100.00%) (esc)
   3 copy files
   source: svn:afeb9c47-92ff-4c0c-9f72-e1f6eb8ac9af/trunk@3
   converting: 2/6 revisions (33.33%)
   scanning paths: /trunk/\xc3\xa0 0/4 (0.00%) (esc)
   gone from -1
   reparent to file://*/svn-repo (glob)
   reparent to file://*/svn-repo/trunk (glob)
   scanning paths: /trunk/\xc3\xa8 1/4 (25.00%) (esc)
   copied to \xc3\xa8 from \xc3\xa9@2 (esc)
   scanning paths: /trunk/\xc3\xa9 2/4 (50.00%) (esc)
   gone from -1
   reparent to file://*/svn-repo (glob)
   reparent to file://*/svn-repo/trunk (glob)
   scanning paths: /trunk/\xc3\xb9 3/4 (75.00%) (esc)
   mark /trunk/\xc3\xb9 came from \xc3\xa0:2 (esc)
   \xc3\xa0/e\xcc\x81 (esc)
   getting files: \xc3\xa0/e\xcc\x81 1/4 (25.00%) (esc)
   \xc3\xa8 (esc)
   getting files: \xc3\xa8 2/4 (50.00%) (esc)
    \xc3\xa8: copy \xc3\xa9:6b67ccefd5ce6de77e7ead4f5292843a0255329f (esc)
   \xc3\xa9 (esc)
   getting files: \xc3\xa9 3/4 (75.00%) (esc)
   \xc3\xb9/e\xcc\x81 (esc)
   getting files: \xc3\xb9/e\xcc\x81 4/4 (100.00%) (esc)
    \xc3\xb9/e\xcc\x81: copy \xc3\xa0/e\xcc\x81:a9092a3d84a37b9993b5c73576f6de29b7ea50f6 (esc)
   2 remove files
   source: svn:afeb9c47-92ff-4c0c-9f72-e1f6eb8ac9af/trunk@4
   converting: 3/6 revisions (50.00%)
   scanning paths: /trunk/\xc3\xa8 0/2 (0.00%) (esc)
   gone from -1
   reparent to file://*/svn-repo (glob)
   reparent to file://*/svn-repo/trunk (glob)
   scanning paths: /trunk/\xc3\xb9 1/2 (50.00%) (esc)
   gone from -1
   reparent to file://*/svn-repo (glob)
   reparent to file://*/svn-repo/trunk (glob)
   \xc3\xa8 (esc)
   getting files: \xc3\xa8 1/2 (50.00%) (esc)
   \xc3\xb9/e\xcc\x81 (esc)
   getting files: \xc3\xb9/e\xcc\x81 2/2 (100.00%) (esc)
   1 branch to branch?
   source: svn:afeb9c47-92ff-4c0c-9f72-e1f6eb8ac9af/branches/branch?@5
   converting: 4/6 revisions (66.67%)
   reparent to file://*/svn-repo/branches/branch%C3%A9 (glob)
   scanning paths: /branches/branch\xc3\xa9 0/1 (0.00%) (esc)
   0 branch to branch?e
   source: svn:afeb9c47-92ff-4c0c-9f72-e1f6eb8ac9af/branches/branch?e@6
   converting: 5/6 revisions (83.33%)
   reparent to file://*/svn-repo/branches/branch%C3%A9e (glob)
   scanning paths: /branches/branch\xc3\xa9e 0/1 (0.00%) (esc)
   reparent to file://*/svn-repo (glob)
   reparent to file://*/svn-repo/branches/branch%C3%A9e (glob)
   reparent to file://*/svn-repo (glob)
   reparent to file://*/svn-repo/branches/branch%C3%A9e (glob)
   updating tags
   .hgtags
   run hg sink post-conversion action
   $ cd A-hg
   $ hg up
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
 
 Check tags are in UTF-8
 
   $ cat .hgtags
   e94e4422020e715add80525e8f0f46c9968689f1 branch\xc3\xa9e (esc)
   f7e66f98380ed1e53a797c5c7a7a2616a7ab377d branch\xc3\xa9 (esc)
 
   $ cd ..

tests/test-convert-svn-move.t

@@ -1,247 +1,247 b''
 #require svn svn-bindings
 
   $ cat >> $HGRCPATH <<EOF
   > [extensions]
   > convert =
   > EOF
 
   $ svnadmin create svn-repo
   $ svnadmin load -q svn-repo < "$TESTDIR/svn/move.svndump"
   $ SVNREPOPATH=`pwd`/svn-repo
 #if windows
   $ SVNREPOURL=file:///`python -c "import urllib, sys; sys.stdout.write(urllib.quote(sys.argv[1]))" "$SVNREPOPATH"`
 #else
   $ SVNREPOURL=file://`python -c "import urllib, sys; sys.stdout.write(urllib.quote(sys.argv[1]))" "$SVNREPOPATH"`
 #endif
 
 Convert trunk and branches
 
   $ hg convert --datesort "$SVNREPOURL"/subproject A-hg
   initializing destination A-hg repository
   scanning source...
   sorting...
   converting...
   13 createtrunk
   12 moved1
   11 moved1
   10 moved2
   9 changeb and rm d2
   8 changeb and rm d2
   7 moved1again
   6 moved1again
   5 copyfilefrompast
   4 copydirfrompast
   3 add d3
   2 copy dir and remove subdir
   1 add d4old
   0 rename d4old into d4new
 
   $ cd A-hg
   $ hg log -G --template '{rev} {desc|firstline} files: {files}\n'
   o  13 rename d4old into d4new files: d4new/g d4old/g
   |
   o  12 add d4old files: d4old/g
   |
   o  11 copy dir and remove subdir files: d3/d31/e d4/d31/e d4/f
   |
   o  10 add d3 files: d3/d31/e d3/f
   |
   o  9 copydirfrompast files: d2/d
   |
   o  8 copyfilefrompast files: d
   |
   o  7 moved1again files: d1/b d1/c
   |
   | o  6 moved1again files:
   | |
   o |  5 changeb and rm d2 files: d1/b d2/d
   | |
   | o  4 changeb and rm d2 files: b
   | |
   o |  3 moved2 files: d2/d
   | |
   o |  2 moved1 files: d1/b d1/c
   | |
   | o  1 moved1 files: b c
   |
   o  0 createtrunk files:
   
 
 Check move copy records
 
   $ hg st --rev 12:13 --copies
   A d4new/g
     d4old/g
   R d4old/g
 
 Check branches
 
   $ hg branches
   default                       13:* (glob)
   d1                             6:* (glob)
   $ cd ..
 
   $ mkdir test-replace
   $ cd test-replace
   $ svnadmin create svn-repo
   $ svnadmin load -q svn-repo < "$TESTDIR/svn/replace.svndump"
 
 Convert files being replaced by directories
 
   $ hg convert svn-repo hg-repo
   initializing destination hg-repo repository
   scanning source...
   sorting...
   converting...
   6 initial
   5 clobber symlink
   4 clobber1
   3 clobber2
   2 adddb
   1 clobberdir
   0 branch
 
   $ cd hg-repo
 
 Manifest before
 
   $ hg -v manifest -r 1
   644   a
   644   d/b
   644   d2/a
   644 @ dlink
   644 @ dlink2
   644   dlink3
 
 Manifest after clobber1
 
   $ hg -v manifest -r 2
   644   a/b
   644   d/b
   644   d2/a
   644   dlink/b
   644 @ dlink2
   644   dlink3
 
 Manifest after clobber2
 
   $ hg -v manifest -r 3
   644   a/b
   644   d/b
   644   d2/a
   644   dlink/b
   644 @ dlink2
   644 @ dlink3
 
 Manifest after clobberdir
 
   $ hg -v manifest -r 6
   644   a/b
   644   d/b
   644   d2/a
   644   d2/c
   644   dlink/b
   644 @ dlink2
   644 @ dlink3
 
 Try updating
 
   $ hg up -qC default
   $ cd ..
 
 Test convert progress bar
 
   $ cat >> $HGRCPATH <<EOF
   > [extensions]
   > progress =
   > [progress]
   > assume-tty = 1
   > delay = 0
   > changedelay = 0
   > format = topic bar number
   > refresh = 0
   > width = 60
   > EOF
 
   $ hg convert svn-repo hg-progress
   \r (no-eol) (esc)
-  scanning [ <=>                                          ] 1\r (no-eol) (esc)
-  scanning [  <=>                                         ] 2\r (no-eol) (esc)
-  scanning [   <=>                                        ] 3\r (no-eol) (esc)
-  scanning [    <=>                                       ] 4\r (no-eol) (esc)
-  scanning [     <=>                                      ] 5\r (no-eol) (esc)
-  scanning [      <=>                                     ] 6\r (no-eol) (esc)
-  scanning [       <=>                                    ] 7\r (no-eol) (esc)
+  scanning [=====>                                      ] 1/7\r (no-eol) (esc)
+  scanning [===========>                                ] 2/7\r (no-eol) (esc)
+  scanning [=================>                          ] 3/7\r (no-eol) (esc)
+  scanning [========================>                   ] 4/7\r (no-eol) (esc)
+  scanning [==============================>             ] 5/7\r (no-eol) (esc)
+  scanning [====================================>       ] 6/7\r (no-eol) (esc)
+  scanning [===========================================>] 7/7\r (no-eol) (esc)
                                                               \r (no-eol) (esc)
   \r (no-eol) (esc)
   converting [                                          ] 0/7\r (no-eol) (esc)
   getting files [=====>                                 ] 1/6\r (no-eol) (esc)
   getting files [============>                          ] 2/6\r (no-eol) (esc)
   getting files [==================>                    ] 3/6\r (no-eol) (esc)
   getting files [=========================>             ] 4/6\r (no-eol) (esc)
   getting files [===============================>       ] 5/6\r (no-eol) (esc)
   getting files [======================================>] 6/6\r (no-eol) (esc)
                                                               \r (no-eol) (esc)
   \r (no-eol) (esc)
   converting [=====>                                    ] 1/7\r (no-eol) (esc)
   scanning paths [                                      ] 0/1\r (no-eol) (esc)
   getting files [======================================>] 1/1\r (no-eol) (esc)
                                                               \r (no-eol) (esc)
   \r (no-eol) (esc)
   converting [===========>                              ] 2/7\r (no-eol) (esc)
   scanning paths [                                      ] 0/2\r (no-eol) (esc)
   scanning paths [==================>                   ] 1/2\r (no-eol) (esc)
   getting files [========>                              ] 1/4\r (no-eol) (esc)
   getting files [==================>                    ] 2/4\r (no-eol) (esc)
   getting files [============================>          ] 3/4\r (no-eol) (esc)
   getting files [======================================>] 4/4\r (no-eol) (esc)
                                                               \r (no-eol) (esc)
   \r (no-eol) (esc)
   converting [=================>                        ] 3/7\r (no-eol) (esc)
   scanning paths [                                      ] 0/1\r (no-eol) (esc)
   getting files [======================================>] 1/1\r (no-eol) (esc)
                                                               \r (no-eol) (esc)
   \r (no-eol) (esc)
   converting [=======================>                  ] 4/7\r (no-eol) (esc)
   scanning paths [                                      ] 0/1\r (no-eol) (esc)
   getting files [======================================>] 1/1\r (no-eol) (esc)
                                                               \r (no-eol) (esc)
   \r (no-eol) (esc)
   converting [=============================>            ] 5/7\r (no-eol) (esc)
   scanning paths [                                      ] 0/1\r (no-eol) (esc)
   getting files [===>                                   ] 1/8\r (no-eol) (esc)
   getting files [========>                              ] 2/8\r (no-eol) (esc)
   getting files [=============>                         ] 3/8\r (no-eol) (esc)
   getting files [==================>                    ] 4/8\r (no-eol) (esc)
   getting files [=======================>               ] 5/8\r (no-eol) (esc)
   getting files [============================>          ] 6/8\r (no-eol) (esc)
   getting files [=================================>     ] 7/8\r (no-eol) (esc)
   getting files [======================================>] 8/8\r (no-eol) (esc)
                                                               \r (no-eol) (esc)
   \r (no-eol) (esc)
   converting [===================================>      ] 6/7\r (no-eol) (esc)
   scanning paths [                                      ] 0/3\r (no-eol) (esc)
   scanning paths [===========>                          ] 1/3\r (no-eol) (esc)
   scanning paths [========================>             ] 2/3\r (no-eol) (esc)
   getting files [===>                                   ] 1/8\r (no-eol) (esc)
   getting files [========>                              ] 2/8\r (no-eol) (esc)
   getting files [=============>                         ] 3/8\r (no-eol) (esc)
   getting files [==================>                    ] 4/8\r (no-eol) (esc)
   getting files [=======================>               ] 5/8\r (no-eol) (esc)
   getting files [============================>          ] 6/8\r (no-eol) (esc)
   getting files [=================================>     ] 7/8\r (no-eol) (esc)
   getting files [======================================>] 8/8\r (no-eol) (esc)
                                                               \r (no-eol) (esc)
   initializing destination hg-progress repository
   scanning source...
   sorting...
   converting...
   6 initial
   5 clobber symlink
   4 clobber1
   3 clobber2
   2 adddb
   1 clobberdir
   0 branch
 
   $ cd ..