removed trailing whitespace

hgext/convert/__init__.py

@@ -1,353 +1,353 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 git import convert_git
 from hg import convert_mercurial
 from subversion import convert_svn
 
 import os, shutil
 from mercurial import hg, ui, util, commands
 
 commands.norepo += " convert"
 
 converters = [convert_cvs, convert_git, convert_svn, convert_mercurial]
 
 def convertsource(ui, path, **opts):
     for c in converters:
         if not hasattr(c, 'getcommit'):
             continue
         try:
             return c(ui, path, **opts)
         except 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:
         if not hasattr(c, 'putcommit'):
             continue
         try:
             return c(ui, path)
         except NoRepo:
             pass
     raise util.Abort('%s: unknown repository type' % path)
 
 class convert(object):
     def __init__(self, ui, source, dest, mapfile, opts):
 
         self.source = source
         self.dest = dest
         self.ui = ui
         self.opts = opts
         self.commitcache = {}
         self.mapfile = mapfile
         self.mapfilefd = None
         self.authors = {}
         self.authorfile = None
 
         self.map = {}
         try:
             origmapfile = open(self.mapfile, 'r')
             for l in origmapfile:
                 sv, dv = l[:-1].split()
                 self.map[sv] = dv
             origmapfile.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
             self.commitcache[n] = self.source.getcommit(n)
             cp = self.commitcache[n].parents
             parents[n] = []
             for p in cp:
                 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.mapfilefd is None:
             try:
                 self.mapfilefd = open(self.mapfile, "a")
             except IOError, (errno, strerror):
                 raise util.Abort("Could not open map file %s: %s, %s\n" % (self.mapfile, errno, strerror))
         self.map[src] = dst
         self.mapfilefd.write("%s %s\n" % (src, dst))
         self.mapfilefd.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 copy(self, rev):
         c = self.commitcache[rev]
         files = self.source.getchanges(rev)
-        
+
         do_copies = (hasattr(c, 'copies') and hasattr(self.dest, 'copyfile'))
 
         for f, v in files:
             try:
                 data = self.source.getfile(f, v)
             except IOError, inst:
                 self.dest.delfile(f)
             else:
                 e = self.source.getmode(f, v)
                 self.dest.putfile(f, e, data)
                 if do_copies:
                     if f in c.copies:
                         # Merely marks that a copy happened.
                         self.dest.copyfile(c.copies[f], f)
 
 
         r = [self.map[v] for v in c.parents]
         f = [f for f, v in files]
         newnode = self.dest.putcommit(f, r, c)
         self.mapentry(rev, newnode)
 
     def convert(self):
         try:
             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]
                 author = self.commitcache[c].author
                 author = self.authors.get(author, author)
                 self.commitcache[c].author = author
                 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):
        if self.mapfilefd:
            self.mapfilefd.close()
 
 def _convert(ui, src, dest=None, mapfile=None, **opts):
     '''Convert a foreign SCM repository to a Mercurial one.
 
     Accepted source formats:
     - GIT
     - CVS
     - SVN
 
     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 <mapfile> isn\'t given, it will be put in a default location
     (<dest>/.hg/shamap by default).  The <mapfile> 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
     '''
 
     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 mapfile:
         try:
             mapfile = destc.mapfile()
         except:
             mapfile = os.path.join(destc, "map")
 
     c = convert(ui, srcc, destc, mapfile, opts)
     c.convert()
 
 cmdtable = {
     "convert":
         (_convert,
          [('A', 'authors', '', 'username mapping filename'),
           ('r', 'rev', '', 'import up to target revision REV'),
           ('', 'datesort', None, 'try to sort changesets by date')],
          'hg convert [OPTION]... SOURCE [DEST [MAPFILE]]'),
 }

hgext/convert/common.py

@@ -1,121 +1,121 b''
 # common code for the convert extension
 
 class NoRepo(Exception): pass
 
 class commit(object):
     def __init__(self, **parts):
         self.rev = None
         self.branch = None
 
         for x in "author date desc parents".split():
             if not x in parts:
                 raise util.Abort("commit missing field %s" % x)
         self.__dict__.update(parts)
         if not self.desc or self.desc.isspace():
             self.desc = '*** empty log message ***'
 
 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 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):
         """Return sorted list of (filename, id) tuples for all files changed in rev.
 
         id just tells us which revision to return in getfile(), e.g. in
         git it's an object hash."""
         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'
-            
+
         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 mapfile(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()

hgext/convert/hg.py

@@ -1,97 +1,97 b''
 # hg backend for convert extension
 
 import os, time
 from mercurial import hg
 
 from common import NoRepo, converter_sink
 
 class convert_mercurial(converter_sink):
     def __init__(self, ui, path):
         self.path = path
         self.ui = ui
         try:
             self.repo = hg.repository(self.ui, path)
         except:
             raise NoRepo("could open hg repo %s" % path)
 
     def mapfile(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 [ hg.hex(x) for x in h ]
 
     def putfile(self, f, e, data):
         self.repo.wwrite(f, data, e)
         if self.repo.dirstate.state(f) == '?':
             self.repo.dirstate.update([f], "a")
 
     def copyfile(self, source, dest):
         self.repo.copy(source, dest)
 
     def delfile(self, f):
         try:
             os.unlink(self.repo.wjoin(f))
             #self.repo.remove([f])
         except:
             pass
 
     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 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,
                                     hg.bin(p1), hg.bin(p2), extra=extra)
             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()))
             self.repo.rawcommit([".hgtags"], "update tags", "convert-repo",
                                 date, self.repo.changelog.tip(), hg.nullid)
             return hg.hex(self.repo.changelog.tip())

hgext/convert/subversion.py

@@ -1,669 +1,669 b''
 # Subversion 1.4/1.5 Python API backend
 #
 # Copyright(C) 2007 Daniel Holth et al
 #
 # Configuration options:
 #
 # convert.svn.trunk
 #   Relative path to the trunk (default: "trunk")
 # convert.svn.branches
 #   Relative path to tree of branches (default: "branches")
 #
 # Set these in a hgrc, or on the command line as follows:
 #
 #   hg convert --config convert.svn.trunk=wackoname [...]
 
 import pprint
 import locale
 import os
 import cPickle as pickle
 from mercurial import util
 
 # 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, commit, converter_source
 
 try:
     from svn.core import SubversionException, Pool
     import svn.core
     import svn.ra
     import svn.delta
     import svn
     import transport
 except ImportError:
     pass
 
 class CompatibilityException(Exception): pass
 
 class changedpath(object):
     def __init__(self, p):
         self.copyfrom_path = p.copyfrom_path
         self.copyfrom_rev = p.copyfrom_rev
         self.action = p.action
 
 # SVN conversion code stolen from bzr-svn and tailor
 class convert_svn(converter_source):
     def __init__(self, ui, url, rev=None):
         super(convert_svn, self).__init__(ui, url, rev=rev)
 
         try:
             SubversionException
         except NameError:
             msg = 'subversion python bindings could not be loaded\n'
             ui.warn(msg)
             raise NoRepo(msg)
 
         self.encoding = locale.getpreferredencoding()
         self.lastrevs = {}
 
         latest = None
         if rev:
             try:
                 latest = int(rev)
             except ValueError:
                 raise util.Abort('svn: revision %s is not an integer' % rev)
         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, e:
             pass
         self.url = url
         self.encoding = 'UTF-8' # Subversion is always nominal UTF-8
         try:
             self.transport = transport.SvnRaTransport(url=url)
             self.ra = self.transport.ra
             self.ctx = self.transport.client
             self.base = svn.ra.get_repos_root(self.ra)
             self.module = self.url[len(self.base):]
             self.modulemap = {} # revision, module
             self.commits = {}
             self.files = {}
             self.uuid = svn.ra.get_uuid(self.ra).decode(self.encoding)
         except SubversionException, e:
             raise NoRepo("couldn't open SVN repo %s" % url)
 
         try:
             self.get_blacklist()
         except IOError, e:
             pass
 
         self.last_changed = self.latest(self.module, latest)
 
         self.head = self.revid(self.last_changed)
 
     def setrevmap(self, revmap):
         lastrevs = {}
         for revid in revmap.keys():
             uuid, module, revnum = self.revsplit(revid)
             lastrevnum = lastrevs.setdefault(module, revnum)
             if revnum > lastrevnum:
                 lastrevs[module] = revnum
         self.lastrevs = lastrevs
 
     def exists(self, path, optrev):
         try:
             return svn.client.ls(self.url.rstrip('/') + '/' + path,
                                  optrev, False, self.ctx)
         except SubversionException, err:
             return []
 
     def getheads(self):
         # detect standard /branches, /tags, /trunk layout
         optrev = svn.core.svn_opt_revision_t()
         optrev.kind = svn.core.svn_opt_revision_number
         optrev.value.number = self.last_changed
         rpath = self.url.strip('/')
         cfgtrunk = self.ui.config('convert', 'svn.trunk')
         cfgbranches = self.ui.config('convert', 'svn.branches')
         trunk = (cfgtrunk or 'trunk').strip('/')
         branches = (cfgbranches or 'branches').strip('/')
         if self.exists(trunk, optrev) and self.exists(branches, optrev):
             self.ui.note('found trunk at %r and branches at %r\n' %
                          (trunk, branches))
             oldmodule = self.module
             self.module += '/' + trunk
             lt = self.latest(self.module, self.last_changed)
             self.head = self.revid(lt)
             self.heads = [self.head]
             branchnames = svn.client.ls(rpath + '/' + branches, optrev, False,
                                         self.ctx)
             for branch in branchnames.keys():
                 if oldmodule:
                     module = '/' + oldmodule + '/' + branches + '/' + branch
                 else:
                     module = '/' + branches + '/' + branch
                 brevnum = self.latest(module, self.last_changed)
                 brev = self.revid(brevnum, module)
                 self.ui.note('found branch %s at %d\n' % (branch, brevnum))
                 self.heads.append(brev)
         elif cfgtrunk or cfgbranches:
             raise util.Abort(_('trunk/branch layout expected, '
                                'but not found'))
         else:
             self.ui.note('working with one branch\n')
             self.heads = [self.head]
         return self.heads
 
     def getfile(self, file, rev):
         data, mode = self._getfile(file, rev)
         self.modecache[(file, rev)] = mode
         return data
 
-    def getmode(self, file, rev):        
+    def getmode(self, file, rev):
         return self.modecache[(file, rev)]
 
     def getchanges(self, rev):
         self.modecache = {}
         files = self.files[rev]
         cl = files
         cl.sort()
         # caller caches the result, so free it here to release memory
         del self.files[rev]
         return cl
 
     def getcommit(self, rev):
         if rev not in self.commits:
             uuid, module, revnum = self.revsplit(rev)
             self.module = module
             self.reparent(module)
             stop = self.lastrevs.get(module, 0)
             self._fetch_revisions(from_revnum=revnum, to_revnum=stop)
         commit = self.commits[rev]
         # caller caches the result, so free it here to release memory
         del self.commits[rev]
         return commit
 
     def get_log(self, paths, start, end, limit=0, discover_changed_paths=True,
                 strict_node_history=False):
         '''wrapper for svn.ra.get_log.
         on a large repository, svn.ra.get_log pins huge amounts of
         memory that cannot be recovered.  work around it by forking
         and writing results over a pipe.'''
 
         def child(fp):
             protocol = -1
             def receiver(orig_paths, revnum, author, date, message, pool):
                 if orig_paths is not None:
                     for k, v in orig_paths.iteritems():
                         orig_paths[k] = changedpath(v)
                 pickle.dump((orig_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=self.url)
                 svn.ra.get_log(t.ra, paths, start, end, limit,
                                discover_changed_paths,
                                strict_node_history,
                                receiver)
             except SubversionException, (_, num):
                 self.ui.print_exc()
                 pickle.dump(num, fp, protocol)
             else:
                 pickle.dump(None, fp, protocol)
             fp.close()
 
         def parent(fp):
             while True:
                 entry = pickle.load(fp)
                 try:
                     orig_paths, revnum, author, date, message = entry
                 except:
                     if entry is None:
                         break
                     raise SubversionException("child raised exception", entry)
                 yield entry
 
         rfd, wfd = os.pipe()
         pid = os.fork()
         if pid:
             os.close(wfd)
             for p in parent(os.fdopen(rfd, 'rb')):
                 yield p
             ret = os.waitpid(pid, 0)[1]
             if ret:
                 raise util.Abort(_('get_log %s') % util.explain_exit(ret))
         else:
             os.close(rfd)
             child(os.fdopen(wfd, 'wb'))
             os._exit(0)
 
     def gettags(self):
         tags = {}
         start = self.revnum(self.head)
         try:
             for entry in self.get_log(['/tags'], 0, start):
                 orig_paths, revnum, author, date, message = entry
                 for path in orig_paths:
                     if not path.startswith('/tags/'):
                         continue
                     ent = orig_paths[path]
                     source = ent.copyfrom_path
                     rev = ent.copyfrom_rev
                     tag = path.split('/', 2)[2]
                     tags[tag] = self.revid(rev, module=source)
         except SubversionException, (_, num):
             self.ui.note('no tags found at revision %d\n' % start)
         return tags
 
     # -- helper functions --
 
     def revid(self, revnum, module=None):
         if not module:
             module = self.module
         return (u"svn:%s%s@%s" % (self.uuid, module, revnum)).decode(self.encoding)
 
     def revnum(self, rev):
         return int(rev.split('@')[-1])
 
     def revsplit(self, rev):
         url, revnum = rev.encode(self.encoding).split('@', 1)
         revnum = int(revnum)
         parts = url.split('/', 1)
         uuid = parts.pop(0)[4:]
         mod = ''
         if parts:
             mod = '/' + parts[0]
         return uuid, mod, revnum
 
     def latest(self, path, stop=0):
         'find the latest revision affecting path, up to stop'
         if not stop:
             stop = svn.ra.get_latest_revnum(self.ra)
         try:
             self.reparent('')
             dirent = svn.ra.stat(self.ra, path.strip('/'), stop)
             self.reparent(self.module)
         except SubversionException:
             dirent = None
         if not dirent:
             print self.base, path
             raise util.Abort('%s not found up to revision %d' % (path, stop))
 
         return dirent.created_rev
 
     def get_blacklist(self):
         """Avoid certain revision numbers.
         It is not uncommon for two nearby revisions to cancel each other
         out, e.g. 'I copied trunk into a subdirectory of itself instead
         of making a branch'. The converted repository is significantly
         smaller if we ignore such revisions."""
         self.blacklist = set()
         blacklist = self.blacklist
         for line in file("blacklist.txt", "r"):
             if not line.startswith("#"):
                 try:
                     svn_rev = int(line.strip())
                     blacklist.add(svn_rev)
                 except ValueError, e:
                     pass # not an integer or a comment
 
     def is_blacklisted(self, svn_rev):
         return svn_rev in self.blacklist
 
     def reparent(self, module):
         svn_url = self.base + module
         self.ui.debug("reparent to %s\n" % svn_url.encode(self.encoding))
         svn.ra.reparent(self.ra, svn_url.encode(self.encoding))
 
     def _fetch_revisions(self, from_revnum = 0, to_revnum = 347):
         def get_entry_from_path(path, 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, ie
             #   "/CMFPlone/branches/Plone-2_0-branch/tests/PloneTestCase.py"
             # that is to say "tests/PloneTestCase.py"
 
             if path.startswith(module):
                 relative = path[len(module):]
                 if relative.startswith('/'):
                     return relative[1:]
                 else:
                     return relative
 
             # The path is outside our tracked tree...
             self.ui.debug('Ignoring %r since it is not under %r\n' % (path, module))
             return None
 
         self.child_cset = None
         def parselogentry(orig_paths, revnum, author, date, message):
             self.ui.debug("parsing revision %d (%d changes)\n" %
                           (revnum, len(orig_paths)))
 
             if revnum in self.modulemap:
                 new_module = self.modulemap[revnum]
                 if new_module != self.module:
                     self.module = new_module
                     self.reparent(self.module)
 
             copyfrom = {} # Map of entrypath, revision for finding source of deleted revisions.
             copies = {}
             entries = []
             rev = self.revid(revnum)
             parents = []
 
             # branch log might return entries for a parent we already have
             if (rev in self.commits or
                 (revnum < self.lastrevs.get(self.module, 0))):
                 return
 
             try:
                 branch = self.module.split("/")[-1]
                 if branch == 'trunk':
                     branch = ''
             except IndexError:
                 branch = None
 
             orig_paths = orig_paths.items()
             orig_paths.sort()
             for path, ent in orig_paths:
                 # self.ui.write("path %s\n" % path)
                 if path == self.module: # Follow branching back in history
                     if ent:
                         if ent.copyfrom_path:
                             # ent.copyfrom_rev may not be the actual last revision
                             prev = self.latest(ent.copyfrom_path, ent.copyfrom_rev)
                             self.modulemap[prev] = ent.copyfrom_path
                             parents = [self.revid(prev, ent.copyfrom_path)]
                             self.ui.note('found parent of branch %s at %d: %s\n' % \
                                          (self.module, prev, ent.copyfrom_path))
                         else:
                             self.ui.debug("No copyfrom path, don't know what to do.\n")
                             # Maybe it was added and there is no more history.
                 entrypath = get_entry_from_path(path, module=self.module)
                 # self.ui.write("entrypath %s\n" % entrypath)
                 if entrypath is None:
                     # Outside our area of interest
                     self.ui.debug("boring@%s: %s\n" % (revnum, path))
                     continue
                 entry = entrypath.decode(self.encoding)
 
                 kind = svn.ra.check_path(self.ra, entrypath, revnum)
                 if kind == svn.core.svn_node_file:
                     if ent.copyfrom_path:
                         copyfrom_path = get_entry_from_path(ent.copyfrom_path)
                         if copyfrom_path:
                             self.ui.debug("Copied to %s from %s@%s\n" % (entry, copyfrom_path, ent.copyfrom_rev))
                             # It's probably important for hg that the source
                             # exists in the revision's parent, not just the
                             # ent.copyfrom_rev
                             fromkind = svn.ra.check_path(self.ra, copyfrom_path, ent.copyfrom_rev)
                             if fromkind != 0:
                                 copies[self.recode(entry)] = self.recode(copyfrom_path)
                     entries.append(self.recode(entry))
                 elif kind == 0: # gone, but had better be a deleted *file*
                     self.ui.debug("gone from %s\n" % ent.copyfrom_rev)
 
                     # if a branch is created but entries are removed in the same
                     # changeset, get the right fromrev
                     if parents:
                         uuid, old_module, fromrev = self.revsplit(parents[0])
                     else:
                         fromrev = revnum - 1
                         # might always need to be revnum - 1 in these 3 lines?
                         old_module = self.modulemap.get(fromrev, self.module)
 
                     basepath = old_module + "/" + get_entry_from_path(path, module=self.module)
                     entrypath = old_module + "/" + get_entry_from_path(path, module=self.module)
 
                     def lookup_parts(p):
                         rc = None
                         parts = p.split("/")
                         for i in range(len(parts)):
                             part = "/".join(parts[:i])
                             info = part, copyfrom.get(part, None)
                             if info[1] is not None:
                                 self.ui.debug("Found parent directory %s\n" % info[1])
                                 rc = info
                         return rc
 
                     self.ui.debug("base, entry %s %s\n" % (basepath, entrypath))
 
                     frompath, froment = lookup_parts(entrypath) or (None, revnum - 1)
 
                     # need to remove fragment from lookup_parts and replace with copyfrom_path
                     if frompath is not None:
                         self.ui.debug("munge-o-matic\n")
                         self.ui.debug(entrypath + '\n')
                         self.ui.debug(entrypath[len(frompath):] + '\n')
                         entrypath = froment.copyfrom_path + entrypath[len(frompath):]
                         fromrev = froment.copyfrom_rev
                         self.ui.debug("Info: %s %s %s %s\n" % (frompath, froment, ent, entrypath))
 
                     fromkind = svn.ra.check_path(self.ra, entrypath, fromrev)
                     if fromkind == svn.core.svn_node_file:   # a deleted file
                         entries.append(self.recode(entry))
                     elif fromkind == svn.core.svn_node_dir:
                         # print "Deleted/moved non-file:", revnum, path, ent
                         # children = self._find_children(path, revnum - 1)
                         # print "find children %s@%d from %d action %s" % (path, revnum, ent.copyfrom_rev, ent.action)
                         # Sometimes this is tricky. For example: in
                         # The Subversion Repository revision 6940 a dir
-                        # was copied and one of its files was deleted 
+                        # was copied and one of its files was deleted
                         # from the new location in the same commit. This
                         # code can't deal with that yet.
                         if ent.action == 'C':
                             children = self._find_children(path, fromrev)
                         else:
                             oroot = entrypath.strip('/')
                             nroot = path.strip('/')
                             children = self._find_children(oroot, fromrev)
                             children = [s.replace(oroot,nroot) for s in children]
                         # Mark all [files, not directories] as deleted.
                         for child in children:
                             # Can we move a child directory and its
                             # parent in the same commit? (probably can). Could
-                            # cause problems if instead of revnum -1, 
+                            # cause problems if instead of revnum -1,
                             # we have to look in (copyfrom_path, revnum - 1)
                             entrypath = get_entry_from_path("/" + child, module=old_module)
                             if entrypath:
                                 entry = self.recode(entrypath.decode(self.encoding))
                                 if entry in copies:
                                     # deleted file within a copy
                                     del copies[entry]
                                 else:
                                     entries.append(entry)
                     else:
                         self.ui.debug('unknown path in revision %d: %s\n' % \
                                       (revnum, path))
                 elif kind == svn.core.svn_node_dir:
                     # Should probably synthesize normal file entries
                     # and handle as above to clean up copy/rename handling.
 
                     # If the directory just had a prop change,
                     # then we shouldn't need to look for its children.
                     # Also this could create duplicate entries. Not sure
                     # whether this will matter. Maybe should make entries a set.
                     # print "Changed directory", revnum, path, ent.action, ent.copyfrom_path, ent.copyfrom_rev
                     # This will fail if a directory was copied
                     # from another branch and then some of its files
                     # were deleted in the same transaction.
                     children = self._find_children(path, revnum)
                     children.sort()
                     for child in children:
                         # Can we move a child directory and its
                         # parent in the same commit? (probably can). Could
-                        # cause problems if instead of revnum -1, 
+                        # cause problems if instead of revnum -1,
                         # we have to look in (copyfrom_path, revnum - 1)
                         entrypath = get_entry_from_path("/" + child, module=self.module)
                         # print child, self.module, entrypath
                         if entrypath:
                             # Need to filter out directories here...
                             kind = svn.ra.check_path(self.ra, entrypath, revnum)
                             if kind != svn.core.svn_node_dir:
                                 entries.append(self.recode(entrypath))
 
                     # Copies here (must copy all from source)
                     # Probably not a real problem for us if
                     # source does not exist
 
                     # Can do this with the copy command "hg copy"
                     # if ent.copyfrom_path:
                     #     copyfrom_entry = get_entry_from_path(ent.copyfrom_path.decode(self.encoding),
                     #             module=self.module)
                     #     copyto_entry = entrypath
                     #
                     #     print "copy directory", copyfrom_entry, 'to', copyto_entry
                     #
                     #     copies.append((copyfrom_entry, copyto_entry))
 
                     if ent.copyfrom_path:
                         copyfrom_path = ent.copyfrom_path.decode(self.encoding)
                         copyfrom_entry = get_entry_from_path(copyfrom_path, module=self.module)
                         if copyfrom_entry:
                             copyfrom[path] = ent
                             self.ui.debug("mark %s came from %s\n" % (path, copyfrom[path]))
 
                             # Good, /probably/ a regular copy. Really should check
                             # to see whether the parent revision actually contains
                             # the directory in question.
                             children = self._find_children(self.recode(copyfrom_path), ent.copyfrom_rev)
                             children.sort()
                             for child in children:
                                 entrypath = get_entry_from_path("/" + child, module=self.module)
                                 if entrypath:
                                     entry = entrypath.decode(self.encoding)
                                     # print "COPY COPY From", copyfrom_entry, entry
                                     copyto_path = path + entry[len(copyfrom_entry):]
                                     copyto_entry =  get_entry_from_path(copyto_path, module=self.module)
                                     # print "COPY", entry, "COPY To", copyto_entry
                                     copies[self.recode(copyto_entry)] = self.recode(entry)
                                     # copy from quux splort/quuxfile
 
             self.modulemap[revnum] = self.module # track backwards in time
             # a list of (filename, id) where id lets us retrieve the file.
-            # eg in git, id is the object hash. for svn it'll be the 
+            # eg in git, id is the object hash. for svn it'll be the
             self.files[rev] = zip(entries, [rev] * len(entries))
             if not entries:
                 return
 
             # Example SVN datetime. Includes microseconds.
             # ISO-8601 conformant
             # '2007-01-04T17:35:00.902377Z'
             date = util.parsedate(date[:18] + " UTC", ["%Y-%m-%dT%H:%M:%S"])
 
             log = message and self.recode(message)
             author = author and self.recode(author) or ''
 
             cset = commit(author=author,
-                          date=util.datestr(date), 
-                          desc=log, 
+                          date=util.datestr(date),
+                          desc=log,
                           parents=parents,
                           copies=copies,
                           branch=branch,
                           rev=rev.encode('utf-8'))
 
             self.commits[rev] = cset
             if self.child_cset and not self.child_cset.parents:
                 self.child_cset.parents = [rev]
             self.child_cset = cset
 
         self.ui.note('fetching revision log for "%s" from %d to %d\n' %
                      (self.module, from_revnum, to_revnum))
 
         try:
             discover_changed_paths = True
             strict_node_history = False
             for entry in self.get_log([self.module], from_revnum, to_revnum):
                 orig_paths, revnum, author, date, message = entry
                 if self.is_blacklisted(revnum):
                     self.ui.note('skipping blacklisted revision %d\n' % revnum)
                     continue
                 if orig_paths is None:
                     self.ui.debug('revision %d has no entries\n' % revnum)
                     continue
                 parselogentry(orig_paths, revnum, author, date, message)
         except SubversionException, (_, num):
             if num == svn.core.SVN_ERR_FS_NO_SUCH_REVISION:
-                raise NoSuchRevision(branch=self, 
+                raise NoSuchRevision(branch=self,
                     revision="Revision number %d" % to_revnum)
             raise
 
     def _getfile(self, file, rev):
         io = StringIO()
         # TODO: ra.get_file transmits the whole file instead of diffs.
         mode = ''
         try:
             revnum = self.revnum(rev)
             if self.module != self.modulemap[revnum]:
                 self.module = self.modulemap[revnum]
                 self.reparent(self.module)
             info = svn.ra.get_file(self.ra, file, revnum, io)
             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
                 raise IOError()
             raise
         data = io.getvalue()
         if mode == 'l':
             link_prefix = "link "
             if data.startswith(link_prefix):
                 data = data[len(link_prefix):]
         return data, mode
 
     def _find_children(self, path, revnum):
         path = path.strip("/")
 
         def _find_children_fallback(path, revnum):
             # SWIG python bindings for getdir are broken up to at least 1.4.3
             pool = Pool()
             optrev = svn.core.svn_opt_revision_t()
             optrev.kind = svn.core.svn_opt_revision_number
             optrev.value.number = revnum
             rpath = '/'.join([self.base, path]).strip('/')
             return ['%s/%s' % (path, x) for x in svn.client.ls(rpath, optrev, True, self.ctx, pool).keys()]
 
         if hasattr(self, '_find_children_fallback'):
             return _find_children_fallback(path, revnum)
 
         self.reparent("/" + path)
         pool = Pool()
 
         children = []
         def find_children_inner(children, path, revnum = revnum):
             if hasattr(svn.ra, 'get_dir2'): # Since SVN 1.4
                 fields = 0xffffffff # Binding does not provide SVN_DIRENT_ALL
                 getdir = svn.ra.get_dir2(self.ra, path, revnum, fields, pool)
             else:
                 getdir = svn.ra.get_dir(self.ra, path, revnum, pool)
             if type(getdir) == dict:
                 # python binding for getdir is broken up to at least 1.4.3
                 raise CompatibilityException()
             dirents = getdir[0]
             if type(dirents) == int:
                 # got here once due to infinite recursion bug
                 # pprint.pprint(getdir)
                 return
             c = dirents.keys()
             c.sort()
             for child in c:
                 dirent = dirents[child]
                 if dirent.kind == svn.core.svn_node_dir:
                     find_children_inner(children, (path + "/" + child).strip("/"))
                 else:
                     children.append((path + "/" + child).strip("/"))
 
         try:
             find_children_inner(children, "")
         except CompatibilityException:
             self._find_children_fallback = True
             self.reparent(self.module)
             return _find_children_fallback(path, revnum)
 
         self.reparent(self.module)
         return [path + "/" + c for c in children]

hgext/convert/transport.py

@@ -1,125 +1,125 b''
 # -*- coding: utf-8 -*-
 
 # Copyright (C) 2007 Daniel Holth <[e-mail]>
 # This is a stripped-down version of the original bzr-svn transport.py,
 # Copyright (C) 2006 Jelmer Vernooij <[e-mail]>
 
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
 # the Free Software Foundation; either version 2 of the License, or
 # (at your option) any later version.
 
 # This program is distributed in the hope that it will be useful,
 # but WITHOUT ANY WARRANTY; without even the implied warranty of
 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 # GNU General Public License for more details.
 
 # You should have received a copy of the GNU General Public License
 # along with this program; if not, write to the Free Software
 # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
 
 from cStringIO import StringIO
 import os
 from tempfile import mktemp
 
 from svn.core import SubversionException, Pool
 import svn.ra
 import svn.client
 import svn.core
 
-# Some older versions of the Python bindings need to be 
+# Some older versions of the Python bindings need to be
 # explicitly initialized. But what we want to do probably
 # won't work worth a darn against those libraries anyway!
 svn.ra.initialize()
 
 svn_config = svn.core.svn_config_get_config(None)
 
 
 def _create_auth_baton(pool):
     """Create a Subversion authentication baton. """
     import svn.client
     # Give the client context baton a suite of authentication
     # providers.h
     providers = [
         svn.client.get_simple_provider(pool),
         svn.client.get_username_provider(pool),
         svn.client.get_ssl_client_cert_file_provider(pool),
         svn.client.get_ssl_client_cert_pw_file_provider(pool),
         svn.client.get_ssl_server_trust_file_provider(pool),
         ]
     return svn.core.svn_auth_open(providers, pool)
 
 class NotBranchError(SubversionException):
     pass
 
 class SvnRaTransport(object):
     """
     Open an ra connection to a Subversion repository.
     """
     def __init__(self, url="", ra=None):
         self.pool = Pool()
         self.svn_url = url
         self.username = ''
         self.password = ''
 
         # Only Subversion 1.4 has reparent()
         if ra is None or not hasattr(svn.ra, 'reparent'):
             self.client = svn.client.create_context(self.pool)
             ab = _create_auth_baton(self.pool)
             if False:
                 svn.core.svn_auth_set_parameter(
                     ab, svn.core.SVN_AUTH_PARAM_DEFAULT_USERNAME, self.username)
                 svn.core.svn_auth_set_parameter(
                     ab, svn.core.SVN_AUTH_PARAM_DEFAULT_PASSWORD, self.password)
             self.client.auth_baton = ab
             self.client.config = svn_config
             try:
                 self.ra = svn.client.open_ra_session(
-                    self.svn_url.encode('utf8'), 
+                    self.svn_url.encode('utf8'),
                     self.client, self.pool)
             except SubversionException, (_, num):
                 if num in (svn.core.SVN_ERR_RA_ILLEGAL_URL,
                            svn.core.SVN_ERR_RA_LOCAL_REPOS_OPEN_FAILED,
                            svn.core.SVN_ERR_BAD_URL):
                     raise NotBranchError(url)
                 raise
         else:
             self.ra = ra
             svn.ra.reparent(self.ra, self.svn_url.encode('utf8'))
 
     class Reporter:
         def __init__(self, (reporter, report_baton)):
             self._reporter = reporter
             self._baton = report_baton
 
         def set_path(self, path, revnum, start_empty, lock_token, pool=None):
             svn.ra.reporter2_invoke_set_path(self._reporter, self._baton,
                         path, revnum, start_empty, lock_token, pool)
 
         def delete_path(self, path, pool=None):
             svn.ra.reporter2_invoke_delete_path(self._reporter, self._baton,
                     path, pool)
 
         def link_path(self, path, url, revision, start_empty, lock_token,
                       pool=None):
             svn.ra.reporter2_invoke_link_path(self._reporter, self._baton,
                     path, url, revision, start_empty, lock_token,
                     pool)
 
         def finish_report(self, pool=None):
             svn.ra.reporter2_invoke_finish_report(self._reporter,
                     self._baton, pool)
 
         def abort_report(self, pool=None):
             svn.ra.reporter2_invoke_abort_report(self._reporter,
                     self._baton, pool)
 
     def do_update(self, revnum, path, *args, **kwargs):
         return self.Reporter(svn.ra.do_update(self.ra, revnum, path, *args, **kwargs))
 
     def clone(self, offset=None):
         """See Transport.clone()."""
         if offset is None:
             return self.__class__(self.base)
 
         return SvnRaTransport(urlutils.join(self.base, offset), ra=self.ra)

hgext/purge.py

@@ -1,168 +1,168 b''
 # Copyright (C) 2006 - Marco Barisione <[e-mail]>
 #
 # This is a small extension for Mercurial (http://www.selenic.com/mercurial)
 # that removes files not known to mercurial
 #
 # This program was inspired by the "cvspurge" script contained in CVS utilities
 # (http://www.red-bean.com/cvsutils/).
 #
 # To enable the "purge" extension put these lines in your ~/.hgrc:
 #  [extensions]
 #  hgext.purge =
 #
 # For help on the usage of "hg purge" use:
 #  hg help purge
 #
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License as published by
 # the Free Software Foundation; either version 2 of the License, or
 # (at your option) any later version.
 #
 # This program is distributed in the hope that it will be useful,
 # but WITHOUT ANY WARRANTY; without even the implied warranty of
 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 # GNU General Public License for more details.
 #
 # You should have received a copy of the GNU General Public License
 # along with this program; if not, write to the Free Software
 # Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
 
 from mercurial import hg, util
 from mercurial.i18n import _
 import os
 
-def dopurge(ui, repo, dirs=None, act=True, ignored=False, 
+def dopurge(ui, repo, dirs=None, act=True, ignored=False,
             abort_on_err=False, eol='\n',
             force=False, include=None, exclude=None):
     def error(msg):
         if abort_on_err:
             raise util.Abort(msg)
         else:
             ui.warn(_('warning: %s\n') % msg)
 
     def remove(remove_func, name):
         if act:
             try:
                 remove_func(os.path.join(repo.root, name))
             except OSError, e:
                 error(_('%s cannot be removed') % name)
         else:
             ui.write('%s%s' % (name, eol))
 
     directories = []
     files = []
     missing = []
     roots, match, anypats = util.cmdmatcher(repo.root, repo.getcwd(), dirs,
                                             include, exclude)
     for src, f, st in repo.dirstate.statwalk(files=roots, match=match,
                                              ignored=ignored, directories=True):
         if src == 'd':
             directories.append(f)
         elif src == 'm':
             missing.append(f)
         elif src == 'f' and f not in repo.dirstate:
             files.append(f)
 
     _check_missing(ui, repo, missing, force)
 
     directories.sort()
 
     for f in files:
         if f not in repo.dirstate:
             ui.note(_('Removing file %s\n') % f)
             remove(os.remove, f)
 
     for f in directories[::-1]:
         if match(f) and not os.listdir(repo.wjoin(f)):
             ui.note(_('Removing directory %s\n') % f)
             remove(os.rmdir, f)
 
 def _check_missing(ui, repo, missing, force=False):
     """Abort if there is the chance of having problems with name-mangling fs
 
     In a name mangling filesystem (e.g. a case insensitive one)
     dirstate.walk() can yield filenames different from the ones
     stored in the dirstate. This already confuses the status and
     add commands, but with purge this may cause data loss.
 
     To prevent this, _check_missing will abort if there are missing
     files. The force option will let the user skip the check if he
     knows it is safe.
 
     Even with the force option this function will check if any of the
     missing files is still available in the working dir: if so there
     may be some problem with the underlying filesystem, so it
     aborts unconditionally."""
 
     found = [f for f in missing if util.lexists(repo.wjoin(f))]
 
     if found:
         if not ui.quiet:
             ui.warn(_("The following tracked files weren't listed by the "
                       "filesystem, but could still be found:\n"))
             for f in found:
                 ui.warn("%s\n" % f)
             if util.checkfolding(repo.path):
                 ui.warn(_("This is probably due to a case-insensitive "
                           "filesystem\n"))
         raise util.Abort(_("purging on name mangling filesystems is not "
                            "yet fully supported"))
 
     if missing and not force:
         raise util.Abort(_("there are missing files in the working dir and "
                            "purge still has problems with them due to name "
                            "mangling filesystems. "
                            "Use --force if you know what you are doing"))
 
 
 def purge(ui, repo, *dirs, **opts):
     '''removes files not tracked by mercurial
 
     Delete files not known to mercurial, this is useful to test local and
     uncommitted changes in the otherwise clean source tree.
 
     This means that purge will delete:
      - Unknown files: files marked with "?" by "hg status"
      - Ignored files: files usually ignored by Mercurial because they match
        a pattern in a ".hgignore" file
      - Empty directories: in fact Mercurial ignores directories unless they
        contain files under source control managment
     But it will leave untouched:
      - Unmodified tracked files
      - Modified tracked files
      - New files added to the repository (with "hg add")
 
     If directories are given on the command line, only files in these
     directories are considered.
 
     Be careful with purge, you could irreversibly delete some files you
     forgot to add to the repository. If you only want to print the list of
     files that this program would delete use the --print option.
     '''
     act = not opts['print']
     ignored = bool(opts['all'])
     abort_on_err = bool(opts['abort_on_err'])
     eol = opts['print0'] and '\0' or '\n'
     if eol == '\0':
         # --print0 implies --print
         act = False
     force = bool(opts['force'])
     include = opts['include']
     exclude = opts['exclude']
     dopurge(ui, repo, dirs, act, ignored, abort_on_err,
             eol, force, include, exclude)
 
 
 cmdtable = {
     'purge|clean':
         (purge,
          [('a', 'abort-on-err', None, _('abort if an error occurs')),
           ('',  'all', None, _('purge ignored files too')),
           ('f', 'force', None, _('purge even when missing files are detected')),
           ('p', 'print', None, _('print the file names instead of deleting them')),
           ('0', 'print0', None, _('end filenames with NUL, for use with xargs'
                                   ' (implies -p)')),
           ('I', 'include', [], _('include names matching the given patterns')),
           ('X', 'exclude', [], _('exclude names matching the given patterns'))],
          _('hg purge [OPTION]... [DIR]...'))
 }

mercurial/hgweb/hgwebdir_mod.py

@@ -1,258 +1,258 b''
 # hgweb/hgwebdir_mod.py - Web interface for a directory of repositories.
 #
 # Copyright 21 May 2005 - (c) 2005 Jake Edge <[e-mail]>
 # Copyright 2005, 2006 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 mercurial import demandimport; demandimport.enable()
 import os, mimetools, cStringIO
 from mercurial.i18n import gettext as _
 from mercurial import ui, hg, util, templater
 from common import get_mtime, staticfile, style_map, paritygen
 from hgweb_mod import hgweb
 
 # This is a stopgap
 class hgwebdir(object):
     def __init__(self, config, parentui=None):
         def cleannames(items):
             return [(name.strip(os.sep), path) for name, path in items]
 
         self.parentui = parentui
         self.motd = None
         self.style = None
         self.stripecount = None
         self.repos_sorted = ('name', False)
         if isinstance(config, (list, tuple)):
             self.repos = cleannames(config)
             self.repos_sorted = ('', False)
         elif isinstance(config, dict):
             self.repos = cleannames(config.items())
             self.repos.sort()
         else:
             if isinstance(config, util.configparser):
                 cp = config
             else:
                 cp = util.configparser()
                 cp.read(config)
             self.repos = []
             if cp.has_section('web'):
                 if cp.has_option('web', 'motd'):
                     self.motd = cp.get('web', 'motd')
                 if cp.has_option('web', 'style'):
                     self.style = cp.get('web', 'style')
                 if cp.has_option('web', 'stripes'):
                     self.stripecount = int(cp.get('web', 'stripes'))
             if cp.has_section('paths'):
                 self.repos.extend(cleannames(cp.items('paths')))
             if cp.has_section('collections'):
                 for prefix, root in cp.items('collections'):
                     for path in util.walkrepos(root):
                         repo = os.path.normpath(path)
                         name = repo
                         if name.startswith(prefix):
                             name = name[len(prefix):]
                         self.repos.append((name.lstrip(os.sep), repo))
             self.repos.sort()
 
     def run(self):
         if not os.environ.get('GATEWAY_INTERFACE', '').startswith("CGI/1."):
             raise RuntimeError("This function is only intended to be called while running as a CGI script.")
         import mercurial.hgweb.wsgicgi as wsgicgi
         from request import wsgiapplication
         def make_web_app():
             return self
         wsgicgi.launch(wsgiapplication(make_web_app))
 
     def run_wsgi(self, req):
         def header(**map):
             header_file = cStringIO.StringIO(
                 ''.join(tmpl("header", encoding=util._encoding, **map)))
             msg = mimetools.Message(header_file, 0)
             req.header(msg.items())
             yield header_file.read()
 
         def footer(**map):
             yield tmpl("footer", **map)
 
         def motd(**map):
             if self.motd is not None:
                 yield self.motd
             else:
                 yield config('web', 'motd', '')
 
         parentui = self.parentui or ui.ui(report_untrusted=False)
 
         def config(section, name, default=None, untrusted=True):
             return parentui.config(section, name, default, untrusted)
 
         url = req.env['REQUEST_URI'].split('?')[0]
         if not url.endswith('/'):
             url += '/'
         pathinfo = req.env.get('PATH_INFO', '').strip('/') + '/'
         base = url[:len(url) - len(pathinfo)]
         if not base.endswith('/'):
             base += '/'
 
         staticurl = config('web', 'staticurl') or base + 'static/'
         if not staticurl.endswith('/'):
             staticurl += '/'
 
         style = self.style
         if style is None:
             style = config('web', 'style', '')
         if req.form.has_key('style'):
             style = req.form['style'][0]
         if self.stripecount is None:
             self.stripecount = int(config('web', 'stripes', 1))
         mapfile = style_map(templater.templatepath(), style)
         tmpl = templater.templater(mapfile, templater.common_filters,
                                    defaults={"header": header,
                                              "footer": footer,
                                              "motd": motd,
                                              "url": url,
                                              "staticurl": staticurl})
 
         def archivelist(ui, nodeid, url):
             allowed = ui.configlist("web", "allow_archive", untrusted=True)
             for i in [('zip', '.zip'), ('gz', '.tar.gz'), ('bz2', '.tar.bz2')]:
                 if i[0] in allowed or ui.configbool("web", "allow" + i[0],
                                                     untrusted=True):
                     yield {"type" : i[0], "extension": i[1],
                            "node": nodeid, "url": url}
 
         def entries(sortcolumn="", descending=False, subdir="", **map):
             def sessionvars(**map):
                 fields = []
                 if req.form.has_key('style'):
                     style = req.form['style'][0]
                     if style != get('web', 'style', ''):
                         fields.append(('style', style))
 
                 separator = url[-1] == '?' and ';' or '?'
                 for name, value in fields:
                     yield dict(name=name, value=value, separator=separator)
                     separator = ';'
 
             rows = []
             parity = paritygen(self.stripecount)
             for name, path in self.repos:
                 if not name.startswith(subdir):
                     continue
                 name = name[len(subdir):]
 
                 u = ui.ui(parentui=parentui)
                 try:
                     u.readconfig(os.path.join(path, '.hg', 'hgrc'))
                 except IOError:
                     pass
                 def get(section, name, default=None):
                     return u.config(section, name, default, untrusted=True)
 
                 if u.configbool("web", "hidden", untrusted=True):
                     continue
 
                 url = ('/'.join([req.env["REQUEST_URI"].split('?')[0], name])
                        .replace("//", "/")) + '/'
 
                 # update time with local timezone
                 try:
                     d = (get_mtime(path), util.makedate()[1])
                 except OSError:
                     continue
 
                 contact = (get("ui", "username") or # preferred
                            get("web", "contact") or # deprecated
                            get("web", "author", "")) # also
                 description = get("web", "description", "")
                 name = get("web", "name", name)
                 row = dict(contact=contact or "unknown",
                            contact_sort=contact.upper() or "unknown",
                            name=name,
                            name_sort=name,
                            url=url,
                            description=description or "unknown",
                            description_sort=description.upper() or "unknown",
                            lastchange=d,
                            lastchange_sort=d[1]-d[0],
                            sessionvars=sessionvars,
                            archives=archivelist(u, "tip", url))
                 if (not sortcolumn
                     or (sortcolumn, descending) == self.repos_sorted):
                     # fast path for unsorted output
                     row['parity'] = parity.next()
                     yield row
                 else:
                     rows.append((row["%s_sort" % sortcolumn], row))
             if rows:
                 rows.sort()
                 if descending:
                     rows.reverse()
                 for key, row in rows:
                     row['parity'] = parity.next()
                     yield row
 
         def makeindex(req, subdir=""):
             sortable = ["name", "description", "contact", "lastchange"]
             sortcolumn, descending = self.repos_sorted
             if req.form.has_key('sort'):
                 sortcolumn = req.form['sort'][0]
                 descending = sortcolumn.startswith('-')
                 if descending:
                     sortcolumn = sortcolumn[1:]
                 if sortcolumn not in sortable:
                     sortcolumn = ""
 
             sort = [("sort_%s" % column,
                      "%s%s" % ((not descending and column == sortcolumn)
                                and "-" or "", column))
                     for column in sortable]
             req.write(tmpl("index", entries=entries, subdir=subdir,
                            sortcolumn=sortcolumn, descending=descending,
                            **dict(sort)))
 
         try:
             virtual = req.env.get("PATH_INFO", "").strip('/')
             if virtual.startswith('static/'):
                 static = os.path.join(templater.templatepath(), 'static')
                 fname = virtual[7:]
                 req.write(staticfile(static, fname, req) or
                           tmpl('error', error='%r not found' % fname))
             elif virtual:
                 repos = dict(self.repos)
                 while virtual:
                     real = repos.get(virtual)
                     if real:
                         req.env['REPO_NAME'] = virtual
                         try:
                             repo = hg.repository(parentui, real)
                             hgweb(repo).run_wsgi(req)
                         except IOError, inst:
                             req.write(tmpl("error", error=inst.strerror))
                         except hg.RepoError, inst:
                             req.write(tmpl("error", error=str(inst)))
                         return
 
                     # browse subdirectories
                     subdir = virtual + '/'
                     if [r for r in repos if r.startswith(subdir)]:
                         makeindex(req, subdir)
                         return
 
                     up = virtual.rfind('/')
                     if up < 0:
                         break
                     virtual = virtual[:up]
-                
+
                 req.write(tmpl("notfound", repo=virtual))
             else:
                 if req.form.has_key('static'):
                     static = os.path.join(templater.templatepath(), "static")
                     fname = req.form['static'][0]
                     req.write(staticfile(static, fname, req)
                               or tmpl("error", error="%r not found" % fname))
                 else:
                     makeindex(req)
         finally:
             tmpl = None

mercurial/hgweb/server.py

@@ -1,289 +1,289 b''
 # hgweb/server.py - The standalone hg web server.
 #
 # Copyright 21 May 2005 - (c) 2005 Jake Edge <[e-mail]>
 # 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.
 
 import os, sys, errno, urllib, BaseHTTPServer, socket, SocketServer, traceback
 from mercurial import ui, hg, util, templater
 from hgweb_mod import hgweb
 from hgwebdir_mod import hgwebdir
 from request import wsgiapplication
 from mercurial.i18n import gettext as _
 
 def _splitURI(uri):
     """ Return path and query splited from uri
 
     Just like CGI environment, the path is unquoted, the query is
     not.
     """
     if '?' in uri:
         path, query = uri.split('?', 1)
     else:
         path, query = uri, ''
     return urllib.unquote(path), query
 
 class _error_logger(object):
     def __init__(self, handler):
         self.handler = handler
     def flush(self):
         pass
     def write(self, str):
         self.writelines(str.split('\n'))
     def writelines(self, seq):
         for msg in seq:
             self.handler.log_error("HG error:  %s", msg)
 
 class _hgwebhandler(object, BaseHTTPServer.BaseHTTPRequestHandler):
 
     url_scheme = 'http'
-    
+
     def __init__(self, *args, **kargs):
         self.protocol_version = 'HTTP/1.1'
         BaseHTTPServer.BaseHTTPRequestHandler.__init__(self, *args, **kargs)
 
     def log_error(self, format, *args):
         errorlog = self.server.errorlog
         errorlog.write("%s - - [%s] %s\n" % (self.client_address[0],
                                              self.log_date_time_string(),
                                              format % args))
 
     def log_message(self, format, *args):
         accesslog = self.server.accesslog
         accesslog.write("%s - - [%s] %s\n" % (self.client_address[0],
                                               self.log_date_time_string(),
                                               format % args))
 
     def do_write(self):
         try:
             self.do_hgweb()
         except socket.error, inst:
             if inst[0] != errno.EPIPE:
                 raise
 
     def do_POST(self):
         try:
             self.do_write()
         except StandardError, inst:
             self._start_response("500 Internal Server Error", [])
             self._write("Internal Server Error")
             tb = "".join(traceback.format_exception(*sys.exc_info()))
             self.log_error("Exception happened during processing request '%s':\n%s",
                            self.path, tb)
 
     def do_GET(self):
         self.do_POST()
 
     def do_hgweb(self):
         path_info, query = _splitURI(self.path)
 
         env = {}
         env['GATEWAY_INTERFACE'] = 'CGI/1.1'
         env['REQUEST_METHOD'] = self.command
         env['SERVER_NAME'] = self.server.server_name
         env['SERVER_PORT'] = str(self.server.server_port)
         env['REQUEST_URI'] = self.path
         env['PATH_INFO'] = path_info
         env['REMOTE_HOST'] = self.client_address[0]
         env['REMOTE_ADDR'] = self.client_address[0]
         if query:
             env['QUERY_STRING'] = query
 
         if self.headers.typeheader is None:
             env['CONTENT_TYPE'] = self.headers.type
         else:
             env['CONTENT_TYPE'] = self.headers.typeheader
         length = self.headers.getheader('content-length')
         if length:
             env['CONTENT_LENGTH'] = length
         for header in [h for h in self.headers.keys()
                        if h not in ('content-type', 'content-length')]:
             hkey = 'HTTP_' + header.replace('-', '_').upper()
             hval = self.headers.getheader(header)
             hval = hval.replace('\n', '').strip()
             if hval:
                 env[hkey] = hval
         env['SERVER_PROTOCOL'] = self.request_version
         env['wsgi.version'] = (1, 0)
         env['wsgi.url_scheme'] = self.url_scheme
         env['wsgi.input'] = self.rfile
         env['wsgi.errors'] = _error_logger(self)
         env['wsgi.multithread'] = isinstance(self.server,
                                              SocketServer.ThreadingMixIn)
         env['wsgi.multiprocess'] = isinstance(self.server,
                                               SocketServer.ForkingMixIn)
         env['wsgi.run_once'] = 0
 
         self.close_connection = True
         self.saved_status = None
         self.saved_headers = []
         self.sent_headers = False
         self.length = None
         req = self.server.reqmaker(env, self._start_response)
         for data in req:
             if data:
                 self._write(data)
 
     def send_headers(self):
         if not self.saved_status:
             raise AssertionError("Sending headers before start_response() called")
         saved_status = self.saved_status.split(None, 1)
         saved_status[0] = int(saved_status[0])
         self.send_response(*saved_status)
         should_close = True
         for h in self.saved_headers:
             self.send_header(*h)
             if h[0].lower() == 'content-length':
                 should_close = False
                 self.length = int(h[1])
         # The value of the Connection header is a list of case-insensitive
         # tokens separated by commas and optional whitespace.
         if 'close' in [token.strip().lower() for token in
                        self.headers.get('connection', '').split(',')]:
             should_close = True
         if should_close:
             self.send_header('Connection', 'close')
         self.close_connection = should_close
         self.end_headers()
         self.sent_headers = True
 
     def _start_response(self, http_status, headers, exc_info=None):
         code, msg = http_status.split(None, 1)
         code = int(code)
         self.saved_status = http_status
         bad_headers = ('connection', 'transfer-encoding')
         self.saved_headers = [h for h in headers
                               if h[0].lower() not in bad_headers]
         return self._write
 
     def _write(self, data):
         if not self.saved_status:
             raise AssertionError("data written before start_response() called")
         elif not self.sent_headers:
             self.send_headers()
         if self.length is not None:
             if len(data) > self.length:
                 raise AssertionError("Content-length header sent, but more bytes than specified are being written.")
             self.length = self.length - len(data)
         self.wfile.write(data)
         self.wfile.flush()
 
 class _shgwebhandler(_hgwebhandler):
 
     url_scheme = 'https'
-    
+
     def setup(self):
         self.connection = self.request
         self.rfile = socket._fileobject(self.request, "rb", self.rbufsize)
         self.wfile = socket._fileobject(self.request, "wb", self.wbufsize)
 
     def do_write(self):
         from OpenSSL.SSL import SysCallError
         try:
             super(_shgwebhandler, self).do_write()
         except SysCallError, inst:
             if inst.args[0] != errno.EPIPE:
                 raise
 
     def handle_one_request(self):
         from OpenSSL.SSL import SysCallError, ZeroReturnError
         try:
             super(_shgwebhandler, self).handle_one_request()
         except (SysCallError, ZeroReturnError):
             self.close_connection = True
             pass
 
 def create_server(ui, repo):
     use_threads = True
 
     def openlog(opt, default):
         if opt and opt != '-':
             return open(opt, 'w')
         return default
 
     address = repo.ui.config("web", "address", "")
     port = int(repo.ui.config("web", "port", 8000))
     use_ipv6 = repo.ui.configbool("web", "ipv6")
     webdir_conf = repo.ui.config("web", "webdir_conf")
     ssl_cert = repo.ui.config("web", "certificate")
     accesslog = openlog(repo.ui.config("web", "accesslog", "-"), sys.stdout)
     errorlog = openlog(repo.ui.config("web", "errorlog", "-"), sys.stderr)
 
     if use_threads:
         try:
             from threading import activeCount
         except ImportError:
             use_threads = False
 
     if use_threads:
         _mixin = SocketServer.ThreadingMixIn
     else:
         if hasattr(os, "fork"):
             _mixin = SocketServer.ForkingMixIn
         else:
             class _mixin:
                 pass
 
     class MercurialHTTPServer(object, _mixin, BaseHTTPServer.HTTPServer):
 
         # SO_REUSEADDR has broken semantics on windows
         if os.name == 'nt':
             allow_reuse_address = 0
 
         def __init__(self, *args, **kargs):
             BaseHTTPServer.HTTPServer.__init__(self, *args, **kargs)
             self.accesslog = accesslog
             self.errorlog = errorlog
             self.daemon_threads = True
             def make_handler():
                 if webdir_conf:
                     hgwebobj = hgwebdir(webdir_conf, ui)
                 elif repo is not None:
                     hgwebobj = hgweb(hg.repository(repo.ui, repo.root))
                 else:
                     raise hg.RepoError(_("There is no Mercurial repository here"
                                          " (.hg not found)"))
                 return hgwebobj
             self.reqmaker = wsgiapplication(make_handler)
 
             addr = address
             if addr in ('', '::'):
                 addr = socket.gethostname()
 
             self.addr, self.port = addr, port
 
             if ssl_cert:
                 try:
                     from OpenSSL import SSL
                     ctx = SSL.Context(SSL.SSLv23_METHOD)
                 except ImportError:
                     raise util.Abort("SSL support is unavailable")
                 ctx.use_privatekey_file(ssl_cert)
                 ctx.use_certificate_file(ssl_cert)
                 sock = socket.socket(self.address_family, self.socket_type)
                 self.socket = SSL.Connection(ctx, sock)
                 self.server_bind()
                 self.server_activate()
 
     class IPv6HTTPServer(MercurialHTTPServer):
         address_family = getattr(socket, 'AF_INET6', None)
 
         def __init__(self, *args, **kwargs):
             if self.address_family is None:
                 raise hg.RepoError(_('IPv6 not available on this system'))
             super(IPv6HTTPServer, self).__init__(*args, **kwargs)
 
     if ssl_cert:
         handler = _shgwebhandler
     else:
         handler = _hgwebhandler
 
     try:
         if use_ipv6:
             return IPv6HTTPServer((address, port), handler)
         else:
             return MercurialHTTPServer((address, port), handler)
     except socket.error, inst:
         raise util.Abort(_('cannot start server: %s') % inst.args[1])