changelog: rely on transaction for finalization

Instead of calling 'cl.finalize()' by hand (possibly at a bogus time) we
register it in the transaction during 'delayupdate' and rely on 'tr.close()' to
call it at the right time.

mercurial/changegroup.py

@@ -1,832 +1,830
 # changegroup.py - Mercurial changegroup manipulation functions
 #
 #  Copyright 2006 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 import weakref
 from i18n import _
 from node import nullrev, nullid, hex, short
 import mdiff, util, dagutil
 import struct, os, bz2, zlib, tempfile
 import discovery, error, phases, branchmap
 
 _CHANGEGROUPV1_DELTA_HEADER = "20s20s20s20s"
 _CHANGEGROUPV2_DELTA_HEADER = "20s20s20s20s20s"
 
 def readexactly(stream, n):
     '''read n bytes from stream.read and abort if less was available'''
     s = stream.read(n)
     if len(s) < n:
         raise util.Abort(_("stream ended unexpectedly"
                            " (got %d bytes, expected %d)")
                           % (len(s), n))
     return s
 
 def getchunk(stream):
     """return the next chunk from stream as a string"""
     d = readexactly(stream, 4)
     l = struct.unpack(">l", d)[0]
     if l <= 4:
         if l:
             raise util.Abort(_("invalid chunk length %d") % l)
         return ""
     return readexactly(stream, l - 4)
 
 def chunkheader(length):
     """return a changegroup chunk header (string)"""
     return struct.pack(">l", length + 4)
 
 def closechunk():
     """return a changegroup chunk header (string) for a zero-length chunk"""
     return struct.pack(">l", 0)
 
 class nocompress(object):
     def compress(self, x):
         return x
     def flush(self):
         return ""
 
 bundletypes = {
     "": ("", nocompress), # only when using unbundle on ssh and old http servers
                           # since the unification ssh accepts a header but there
                           # is no capability signaling it.
     "HG10UN": ("HG10UN", nocompress),
     "HG10BZ": ("HG10", lambda: bz2.BZ2Compressor()),
     "HG10GZ": ("HG10GZ", lambda: zlib.compressobj()),
 }
 
 # hgweb uses this list to communicate its preferred type
 bundlepriority = ['HG10GZ', 'HG10BZ', 'HG10UN']
 
 def writebundle(cg, filename, bundletype, vfs=None):
     """Write a bundle file and return its filename.
 
     Existing files will not be overwritten.
     If no filename is specified, a temporary file is created.
     bz2 compression can be turned off.
     The bundle file will be deleted in case of errors.
     """
 
     fh = None
     cleanup = None
     try:
         if filename:
             if vfs:
                 fh = vfs.open(filename, "wb")
             else:
                 fh = open(filename, "wb")
         else:
             fd, filename = tempfile.mkstemp(prefix="hg-bundle-", suffix=".hg")
             fh = os.fdopen(fd, "wb")
         cleanup = filename
 
         header, compressor = bundletypes[bundletype]
         fh.write(header)
         z = compressor()
 
         # parse the changegroup data, otherwise we will block
         # in case of sshrepo because we don't know the end of the stream
 
         # an empty chunkgroup is the end of the changegroup
         # a changegroup has at least 2 chunkgroups (changelog and manifest).
         # after that, an empty chunkgroup is the end of the changegroup
         for chunk in cg.getchunks():
             fh.write(z.compress(chunk))
         fh.write(z.flush())
         cleanup = None
         return filename
     finally:
         if fh is not None:
             fh.close()
         if cleanup is not None:
             if filename and vfs:
                 vfs.unlink(cleanup)
             else:
                 os.unlink(cleanup)
 
 def decompressor(fh, alg):
     if alg == 'UN':
         return fh
     elif alg == 'GZ':
         def generator(f):
             zd = zlib.decompressobj()
             for chunk in util.filechunkiter(f):
                 yield zd.decompress(chunk)
     elif alg == 'BZ':
         def generator(f):
             zd = bz2.BZ2Decompressor()
             zd.decompress("BZ")
             for chunk in util.filechunkiter(f, 4096):
                 yield zd.decompress(chunk)
     else:
         raise util.Abort("unknown bundle compression '%s'" % alg)
     return util.chunkbuffer(generator(fh))
 
 class cg1unpacker(object):
     deltaheader = _CHANGEGROUPV1_DELTA_HEADER
     deltaheadersize = struct.calcsize(deltaheader)
     def __init__(self, fh, alg):
         self._stream = decompressor(fh, alg)
         self._type = alg
         self.callback = None
     def compressed(self):
         return self._type != 'UN'
     def read(self, l):
         return self._stream.read(l)
     def seek(self, pos):
         return self._stream.seek(pos)
     def tell(self):
         return self._stream.tell()
     def close(self):
         return self._stream.close()
 
     def chunklength(self):
         d = readexactly(self._stream, 4)
         l = struct.unpack(">l", d)[0]
         if l <= 4:
             if l:
                 raise util.Abort(_("invalid chunk length %d") % l)
             return 0
         if self.callback:
             self.callback()
         return l - 4
 
     def changelogheader(self):
         """v10 does not have a changelog header chunk"""
         return {}
 
     def manifestheader(self):
         """v10 does not have a manifest header chunk"""
         return {}
 
     def filelogheader(self):
         """return the header of the filelogs chunk, v10 only has the filename"""
         l = self.chunklength()
         if not l:
             return {}
         fname = readexactly(self._stream, l)
         return {'filename': fname}
 
     def _deltaheader(self, headertuple, prevnode):
         node, p1, p2, cs = headertuple
         if prevnode is None:
             deltabase = p1
         else:
             deltabase = prevnode
         return node, p1, p2, deltabase, cs
 
     def deltachunk(self, prevnode):
         l = self.chunklength()
         if not l:
             return {}
         headerdata = readexactly(self._stream, self.deltaheadersize)
         header = struct.unpack(self.deltaheader, headerdata)
         delta = readexactly(self._stream, l - self.deltaheadersize)
         node, p1, p2, deltabase, cs = self._deltaheader(header, prevnode)
         return {'node': node, 'p1': p1, 'p2': p2, 'cs': cs,
                 'deltabase': deltabase, 'delta': delta}
 
     def getchunks(self):
         """returns all the chunks contains in the bundle
 
         Used when you need to forward the binary stream to a file or another
         network API. To do so, it parse the changegroup data, otherwise it will
         block in case of sshrepo because it don't know the end of the stream.
         """
         # an empty chunkgroup is the end of the changegroup
         # a changegroup has at least 2 chunkgroups (changelog and manifest).
         # after that, an empty chunkgroup is the end of the changegroup
         empty = False
         count = 0
         while not empty or count <= 2:
             empty = True
             count += 1
             while True:
                 chunk = getchunk(self)
                 if not chunk:
                     break
                 empty = False
                 yield chunkheader(len(chunk))
                 pos = 0
                 while pos < len(chunk):
                     next = pos + 2**20
                     yield chunk[pos:next]
                     pos = next
             yield closechunk()
 
 class cg2unpacker(cg1unpacker):
     deltaheader = _CHANGEGROUPV2_DELTA_HEADER
     deltaheadersize = struct.calcsize(deltaheader)
 
     def _deltaheader(self, headertuple, prevnode):
         node, p1, p2, deltabase, cs = headertuple
         return node, p1, p2, deltabase, cs
 
 class headerlessfixup(object):
     def __init__(self, fh, h):
         self._h = h
         self._fh = fh
     def read(self, n):
         if self._h:
             d, self._h = self._h[:n], self._h[n:]
             if len(d) < n:
                 d += readexactly(self._fh, n - len(d))
             return d
         return readexactly(self._fh, n)
 
 class cg1packer(object):
     deltaheader = _CHANGEGROUPV1_DELTA_HEADER
     def __init__(self, repo, bundlecaps=None):
         """Given a source repo, construct a bundler.
 
         bundlecaps is optional and can be used to specify the set of
         capabilities which can be used to build the bundle.
         """
         # Set of capabilities we can use to build the bundle.
         if bundlecaps is None:
             bundlecaps = set()
         self._bundlecaps = bundlecaps
         self._changelog = repo.changelog
         self._manifest = repo.manifest
         reorder = repo.ui.config('bundle', 'reorder', 'auto')
         if reorder == 'auto':
             reorder = None
         else:
             reorder = util.parsebool(reorder)
         self._repo = repo
         self._reorder = reorder
         self._progress = repo.ui.progress
     def close(self):
         return closechunk()
 
     def fileheader(self, fname):
         return chunkheader(len(fname)) + fname
 
     def group(self, nodelist, revlog, lookup, units=None, reorder=None):
         """Calculate a delta group, yielding a sequence of changegroup chunks
         (strings).
 
         Given a list of changeset revs, return a set of deltas and
         metadata corresponding to nodes. The first delta is
         first parent(nodelist[0]) -> nodelist[0], the receiver is
         guaranteed to have this parent as it has all history before
         these changesets. In the case firstparent is nullrev the
         changegroup starts with a full revision.
 
         If units is not None, progress detail will be generated, units specifies
         the type of revlog that is touched (changelog, manifest, etc.).
         """
         # if we don't have any revisions touched by these changesets, bail
         if len(nodelist) == 0:
             yield self.close()
             return
 
         # for generaldelta revlogs, we linearize the revs; this will both be
         # much quicker and generate a much smaller bundle
         if (revlog._generaldelta and reorder is not False) or reorder:
             dag = dagutil.revlogdag(revlog)
             revs = set(revlog.rev(n) for n in nodelist)
             revs = dag.linearize(revs)
         else:
             revs = sorted([revlog.rev(n) for n in nodelist])
 
         # add the parent of the first rev
         p = revlog.parentrevs(revs[0])[0]
         revs.insert(0, p)
 
         # build deltas
         total = len(revs) - 1
         msgbundling = _('bundling')
         for r in xrange(len(revs) - 1):
             if units is not None:
                 self._progress(msgbundling, r + 1, unit=units, total=total)
             prev, curr = revs[r], revs[r + 1]
             linknode = lookup(revlog.node(curr))
             for c in self.revchunk(revlog, curr, prev, linknode):
                 yield c
 
         yield self.close()
 
     # filter any nodes that claim to be part of the known set
     def prune(self, revlog, missing, commonrevs, source):
         rr, rl = revlog.rev, revlog.linkrev
         return [n for n in missing if rl(rr(n)) not in commonrevs]
 
     def generate(self, commonrevs, clnodes, fastpathlinkrev, source):
         '''yield a sequence of changegroup chunks (strings)'''
         repo = self._repo
         cl = self._changelog
         mf = self._manifest
         reorder = self._reorder
         progress = self._progress
 
         # for progress output
         msgbundling = _('bundling')
 
         mfs = {} # needed manifests
         fnodes = {} # needed file nodes
         changedfiles = set()
 
         # Callback for the changelog, used to collect changed files and manifest
         # nodes.
         # Returns the linkrev node (identity in the changelog case).
         def lookupcl(x):
             c = cl.read(x)
             changedfiles.update(c[3])
             # record the first changeset introducing this manifest version
             mfs.setdefault(c[0], x)
             return x
 
         # Callback for the manifest, used to collect linkrevs for filelog
         # revisions.
         # Returns the linkrev node (collected in lookupcl).
         def lookupmf(x):
             clnode = mfs[x]
             if not fastpathlinkrev:
                 mdata = mf.readfast(x)
                 for f, n in mdata.iteritems():
                     if f in changedfiles:
                         # record the first changeset introducing this filelog
                         # version
                         fnodes[f].setdefault(n, clnode)
             return clnode
 
         for chunk in self.group(clnodes, cl, lookupcl, units=_('changesets'),
                                 reorder=reorder):
             yield chunk
         progress(msgbundling, None)
 
         for f in changedfiles:
             fnodes[f] = {}
         mfnodes = self.prune(mf, mfs, commonrevs, source)
         for chunk in self.group(mfnodes, mf, lookupmf, units=_('manifests'),
                                 reorder=reorder):
             yield chunk
         progress(msgbundling, None)
 
         mfs.clear()
         needed = set(cl.rev(x) for x in clnodes)
 
         def linknodes(filerevlog, fname):
             if fastpathlinkrev:
                 llr = filerevlog.linkrev
                 def genfilenodes():
                     for r in filerevlog:
                         linkrev = llr(r)
                         if linkrev in needed:
                             yield filerevlog.node(r), cl.node(linkrev)
                 fnodes[fname] = dict(genfilenodes())
             return fnodes.get(fname, {})
 
         for chunk in self.generatefiles(changedfiles, linknodes, commonrevs,
                                         source):
             yield chunk
 
         yield self.close()
         progress(msgbundling, None)
 
         if clnodes:
             repo.hook('outgoing', node=hex(clnodes[0]), source=source)
 
     def generatefiles(self, changedfiles, linknodes, commonrevs, source):
         repo = self._repo
         progress = self._progress
         reorder = self._reorder
         msgbundling = _('bundling')
 
         total = len(changedfiles)
         # for progress output
         msgfiles = _('files')
         for i, fname in enumerate(sorted(changedfiles)):
             filerevlog = repo.file(fname)
             if not filerevlog:
                 raise util.Abort(_("empty or missing revlog for %s") % fname)
 
             linkrevnodes = linknodes(filerevlog, fname)
             # Lookup for filenodes, we collected the linkrev nodes above in the
             # fastpath case and with lookupmf in the slowpath case.
             def lookupfilelog(x):
                 return linkrevnodes[x]
 
             filenodes = self.prune(filerevlog, linkrevnodes, commonrevs, source)
             if filenodes:
                 progress(msgbundling, i + 1, item=fname, unit=msgfiles,
                          total=total)
                 yield self.fileheader(fname)
                 for chunk in self.group(filenodes, filerevlog, lookupfilelog,
                                         reorder=reorder):
                     yield chunk
 
     def deltaparent(self, revlog, rev, p1, p2, prev):
         return prev
 
     def revchunk(self, revlog, rev, prev, linknode):
         node = revlog.node(rev)
         p1, p2 = revlog.parentrevs(rev)
         base = self.deltaparent(revlog, rev, p1, p2, prev)
 
         prefix = ''
         if base == nullrev:
             delta = revlog.revision(node)
             prefix = mdiff.trivialdiffheader(len(delta))
         else:
             delta = revlog.revdiff(base, rev)
         p1n, p2n = revlog.parents(node)
         basenode = revlog.node(base)
         meta = self.builddeltaheader(node, p1n, p2n, basenode, linknode)
         meta += prefix
         l = len(meta) + len(delta)
         yield chunkheader(l)
         yield meta
         yield delta
     def builddeltaheader(self, node, p1n, p2n, basenode, linknode):
         # do nothing with basenode, it is implicitly the previous one in HG10
         return struct.pack(self.deltaheader, node, p1n, p2n, linknode)
 
 class cg2packer(cg1packer):
 
     deltaheader = _CHANGEGROUPV2_DELTA_HEADER
 
     def group(self, nodelist, revlog, lookup, units=None, reorder=None):
         if (revlog._generaldelta and reorder is not True):
             reorder = False
         return cg1packer.group(self, nodelist, revlog, lookup,
                                units=units, reorder=reorder)
 
     def deltaparent(self, revlog, rev, p1, p2, prev):
         dp = revlog.deltaparent(rev)
         # avoid storing full revisions; pick prev in those cases
         # also pick prev when we can't be sure remote has dp
         if dp == nullrev or (dp != p1 and dp != p2 and dp != prev):
             return prev
         return dp
 
     def builddeltaheader(self, node, p1n, p2n, basenode, linknode):
         return struct.pack(self.deltaheader, node, p1n, p2n, basenode, linknode)
 
 packermap = {'01': (cg1packer, cg1unpacker),
              '02': (cg2packer, cg2unpacker)}
 
 def _changegroupinfo(repo, nodes, source):
     if repo.ui.verbose or source == 'bundle':
         repo.ui.status(_("%d changesets found\n") % len(nodes))
     if repo.ui.debugflag:
         repo.ui.debug("list of changesets:\n")
         for node in nodes:
             repo.ui.debug("%s\n" % hex(node))
 
 def getsubsetraw(repo, outgoing, bundler, source, fastpath=False):
     repo = repo.unfiltered()
     commonrevs = outgoing.common
     csets = outgoing.missing
     heads = outgoing.missingheads
     # We go through the fast path if we get told to, or if all (unfiltered
     # heads have been requested (since we then know there all linkrevs will
     # be pulled by the client).
     heads.sort()
     fastpathlinkrev = fastpath or (
             repo.filtername is None and heads == sorted(repo.heads()))
 
     repo.hook('preoutgoing', throw=True, source=source)
     _changegroupinfo(repo, csets, source)
     return bundler.generate(commonrevs, csets, fastpathlinkrev, source)
 
 def getsubset(repo, outgoing, bundler, source, fastpath=False):
     gengroup = getsubsetraw(repo, outgoing, bundler, source, fastpath)
     return cg1unpacker(util.chunkbuffer(gengroup), 'UN')
 
 def changegroupsubset(repo, roots, heads, source):
     """Compute a changegroup consisting of all the nodes that are
     descendants of any of the roots and ancestors of any of the heads.
     Return a chunkbuffer object whose read() method will return
     successive changegroup chunks.
 
     It is fairly complex as determining which filenodes and which
     manifest nodes need to be included for the changeset to be complete
     is non-trivial.
 
     Another wrinkle is doing the reverse, figuring out which changeset in
     the changegroup a particular filenode or manifestnode belongs to.
     """
     cl = repo.changelog
     if not roots:
         roots = [nullid]
     # TODO: remove call to nodesbetween.
     csets, roots, heads = cl.nodesbetween(roots, heads)
     discbases = []
     for n in roots:
         discbases.extend([p for p in cl.parents(n) if p != nullid])
     outgoing = discovery.outgoing(cl, discbases, heads)
     bundler = cg1packer(repo)
     return getsubset(repo, outgoing, bundler, source)
 
 def getlocalchangegroupraw(repo, source, outgoing, bundlecaps=None,
                            version='01'):
     """Like getbundle, but taking a discovery.outgoing as an argument.
 
     This is only implemented for local repos and reuses potentially
     precomputed sets in outgoing. Returns a raw changegroup generator."""
     if not outgoing.missing:
         return None
     bundler = packermap[version][0](repo, bundlecaps)
     return getsubsetraw(repo, outgoing, bundler, source)
 
 def getlocalchangegroup(repo, source, outgoing, bundlecaps=None):
     """Like getbundle, but taking a discovery.outgoing as an argument.
 
     This is only implemented for local repos and reuses potentially
     precomputed sets in outgoing."""
     if not outgoing.missing:
         return None
     bundler = cg1packer(repo, bundlecaps)
     return getsubset(repo, outgoing, bundler, source)
 
 def _computeoutgoing(repo, heads, common):
     """Computes which revs are outgoing given a set of common
     and a set of heads.
 
     This is a separate function so extensions can have access to
     the logic.
 
     Returns a discovery.outgoing object.
     """
     cl = repo.changelog
     if common:
         hasnode = cl.hasnode
         common = [n for n in common if hasnode(n)]
     else:
         common = [nullid]
     if not heads:
         heads = cl.heads()
     return discovery.outgoing(cl, common, heads)
 
 def getchangegroupraw(repo, source, heads=None, common=None, bundlecaps=None,
                       version='01'):
     """Like changegroupsubset, but returns the set difference between the
     ancestors of heads and the ancestors common.
 
     If heads is None, use the local heads. If common is None, use [nullid].
 
     If version is None, use a version '1' changegroup.
 
     The nodes in common might not all be known locally due to the way the
     current discovery protocol works. Returns a raw changegroup generator.
     """
     outgoing = _computeoutgoing(repo, heads, common)
     return getlocalchangegroupraw(repo, source, outgoing, bundlecaps=bundlecaps,
                                   version=version)
 
 def getchangegroup(repo, source, heads=None, common=None, bundlecaps=None):
     """Like changegroupsubset, but returns the set difference between the
     ancestors of heads and the ancestors common.
 
     If heads is None, use the local heads. If common is None, use [nullid].
 
     The nodes in common might not all be known locally due to the way the
     current discovery protocol works.
     """
     outgoing = _computeoutgoing(repo, heads, common)
     return getlocalchangegroup(repo, source, outgoing, bundlecaps=bundlecaps)
 
 def changegroup(repo, basenodes, source):
     # to avoid a race we use changegroupsubset() (issue1320)
     return changegroupsubset(repo, basenodes, repo.heads(), source)
 
 def addchangegroupfiles(repo, source, revmap, trp, pr, needfiles):
     revisions = 0
     files = 0
     while True:
         chunkdata = source.filelogheader()
         if not chunkdata:
             break
         f = chunkdata["filename"]
         repo.ui.debug("adding %s revisions\n" % f)
         pr()
         fl = repo.file(f)
         o = len(fl)
         if not fl.addgroup(source, revmap, trp):
             raise util.Abort(_("received file revlog group is empty"))
         revisions += len(fl) - o
         files += 1
         if f in needfiles:
             needs = needfiles[f]
             for new in xrange(o, len(fl)):
                 n = fl.node(new)
                 if n in needs:
                     needs.remove(n)
                 else:
                     raise util.Abort(
                         _("received spurious file revlog entry"))
             if not needs:
                 del needfiles[f]
     repo.ui.progress(_('files'), None)
 
     for f, needs in needfiles.iteritems():
         fl = repo.file(f)
         for n in needs:
             try:
                 fl.rev(n)
             except error.LookupError:
                 raise util.Abort(
                     _('missing file data for %s:%s - run hg verify') %
                     (f, hex(n)))
 
     return revisions, files
 
 def addchangegroup(repo, source, srctype, url, emptyok=False,
                    targetphase=phases.draft):
     """Add the changegroup returned by source.read() to this repo.
     srctype is a string like 'push', 'pull', or 'unbundle'.  url is
     the URL of the repo where this changegroup is coming from.
 
     Return an integer summarizing the change to this repo:
     - nothing changed or no source: 0
     - more heads than before: 1+added heads (2..n)
     - fewer heads than before: -1-removed heads (-2..-n)
     - number of heads stays the same: 1
     """
     repo = repo.unfiltered()
     def csmap(x):
         repo.ui.debug("add changeset %s\n" % short(x))
         return len(cl)
 
     def revmap(x):
         return cl.rev(x)
 
     if not source:
         return 0
 
     changesets = files = revisions = 0
     efiles = set()
 
     tr = repo.transaction("\n".join([srctype, util.hidepassword(url)]))
     # The transaction could have been created before and already carries source
     # information. In this case we use the top level data. We overwrite the
     # argument because we need to use the top level value (if they exist) in
     # this function.
     srctype = tr.hookargs.setdefault('source', srctype)
     url = tr.hookargs.setdefault('url', url)
 
     # write changelog data to temp files so concurrent readers will not see
     # inconsistent view
     cl = repo.changelog
     cl.delayupdate(tr)
     oldheads = cl.heads()
     try:
         repo.hook('prechangegroup', throw=True, **tr.hookargs)
 
         trp = weakref.proxy(tr)
         # pull off the changeset group
         repo.ui.status(_("adding changesets\n"))
         clstart = len(cl)
         class prog(object):
             step = _('changesets')
             count = 1
             ui = repo.ui
             total = None
             def __call__(repo):
                 repo.ui.progress(repo.step, repo.count, unit=_('chunks'),
                                  total=repo.total)
                 repo.count += 1
         pr = prog()
         source.callback = pr
 
         source.changelogheader()
         srccontent = cl.addgroup(source, csmap, trp)
         if not (srccontent or emptyok):
             raise util.Abort(_("received changelog group is empty"))
         clend = len(cl)
         changesets = clend - clstart
         for c in xrange(clstart, clend):
             efiles.update(repo[c].files())
         efiles = len(efiles)
         repo.ui.progress(_('changesets'), None)
 
         # pull off the manifest group
         repo.ui.status(_("adding manifests\n"))
         pr.step = _('manifests')
         pr.count = 1
         pr.total = changesets # manifests <= changesets
         # no need to check for empty manifest group here:
         # if the result of the merge of 1 and 2 is the same in 3 and 4,
         # no new manifest will be created and the manifest group will
         # be empty during the pull
         source.manifestheader()
         repo.manifest.addgroup(source, revmap, trp)
         repo.ui.progress(_('manifests'), None)
 
         needfiles = {}
         if repo.ui.configbool('server', 'validate', default=False):
             # validate incoming csets have their manifests
             for cset in xrange(clstart, clend):
                 mfest = repo.changelog.read(repo.changelog.node(cset))[0]
                 mfest = repo.manifest.readdelta(mfest)
                 # store file nodes we must see
                 for f, n in mfest.iteritems():
                     needfiles.setdefault(f, set()).add(n)
 
         # process the files
         repo.ui.status(_("adding file changes\n"))
         pr.step = _('files')
         pr.count = 1
         pr.total = efiles
         source.callback = None
 
         newrevs, newfiles = addchangegroupfiles(repo, source, revmap, trp, pr,
                                                 needfiles)
         revisions += newrevs
         files += newfiles
 
         dh = 0
         if oldheads:
             heads = cl.heads()
             dh = len(heads) - len(oldheads)
             for h in heads:
                 if h not in oldheads and repo[h].closesbranch():
                     dh -= 1
         htext = ""
         if dh:
             htext = _(" (%+d heads)") % dh
 
         repo.ui.status(_("added %d changesets"
                          " with %d changes to %d files%s\n")
                          % (changesets, revisions, files, htext))
         repo.invalidatevolatilesets()
 
         if changesets > 0:
             p = lambda: tr.writepending() and repo.root or ""
             if 'node' not in tr.hookargs:
                 tr.hookargs['node'] = hex(cl.node(clstart))
                 hookargs = dict(tr.hookargs)
             else:
                 hookargs = dict(tr.hookargs)
                 hookargs['node'] = hex(cl.node(clstart))
             repo.hook('pretxnchangegroup', throw=True, pending=p, **hookargs)
 
         added = [cl.node(r) for r in xrange(clstart, clend)]
         publishing = repo.ui.configbool('phases', 'publish', True)
         if srctype in ('push', 'serve'):
             # Old servers can not push the boundary themselves.
             # New servers won't push the boundary if changeset already
             # exists locally as secret
             #
             # We should not use added here but the list of all change in
             # the bundle
             if publishing:
                 phases.advanceboundary(repo, tr, phases.public, srccontent)
             else:
                 # Those changesets have been pushed from the outside, their
                 # phases are going to be pushed alongside. Therefor
                 # `targetphase` is ignored.
                 phases.advanceboundary(repo, tr, phases.draft, srccontent)
                 phases.retractboundary(repo, tr, phases.draft, added)
         elif srctype != 'strip':
             # publishing only alter behavior during push
             #
             # strip should not touch boundary at all
             phases.retractboundary(repo, tr, targetphase, added)
 
-        # make changelog see real files again
-        cl.finalize(trp)
 
         tr.close()
 
         if changesets > 0:
             if srctype != 'strip':
                 # During strip, branchcache is invalid but coming call to
                 # `destroyed` will repair it.
                 # In other case we can safely update cache on disk.
                 branchmap.updatecache(repo.filtered('served'))
 
             def runhooks():
                 # These hooks run when the lock releases, not when the
                 # transaction closes. So it's possible for the changelog
                 # to have changed since we last saw it.
                 if clstart >= len(repo):
                     return
 
                 # forcefully update the on-disk branch cache
                 repo.ui.debug("updating the branch cache\n")
                 repo.hook("changegroup", **hookargs)
 
                 for n in added:
                     args = hookargs.copy()
                     args['node'] = hex(n)
                     repo.hook("incoming", **args)
 
                 newheads = [h for h in repo.heads() if h not in oldheads]
                 repo.ui.log("incoming",
                             "%s incoming changes - new heads: %s\n",
                             len(added),
                             ', '.join([hex(c[:6]) for c in newheads]))
             repo._afterlock(runhooks)
 
     finally:
         tr.release()
     # never return 0 here:
     if dh < 0:
         return dh - 1
     else:
         return dh + 1

mercurial/changelog.py

@@ -1,375 +1,378
 # changelog.py - changelog class for mercurial
 #
 # Copyright 2005-2007 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
+import weakref
 from node import bin, hex, nullid
 from i18n import _
 import util, error, revlog, encoding
 
 _defaultextra = {'branch': 'default'}
 
 def _string_escape(text):
     """
     >>> d = {'nl': chr(10), 'bs': chr(92), 'cr': chr(13), 'nul': chr(0)}
     >>> s = "ab%(nl)scd%(bs)s%(bs)sn%(nul)sab%(cr)scd%(bs)s%(nl)s" % d
     >>> s
     'ab\\ncd\\\\\\\\n\\x00ab\\rcd\\\\\\n'
     >>> res = _string_escape(s)
     >>> s == res.decode('string_escape')
     True
     """
     # subset of the string_escape codec
     text = text.replace('\\', '\\\\').replace('\n', '\\n').replace('\r', '\\r')
     return text.replace('\0', '\\0')
 
 def decodeextra(text):
     """
     >>> sorted(decodeextra(encodeextra({'foo': 'bar', 'baz': chr(0) + '2'})
     ...                    ).iteritems())
     [('baz', '\\x002'), ('branch', 'default'), ('foo', 'bar')]
     >>> sorted(decodeextra(encodeextra({'foo': 'bar',
     ...                                 'baz': chr(92) + chr(0) + '2'})
     ...                    ).iteritems())
     [('baz', '\\\\\\x002'), ('branch', 'default'), ('foo', 'bar')]
     """
     extra = _defaultextra.copy()
     for l in text.split('\0'):
         if l:
             if '\\0' in l:
                 # fix up \0 without getting into trouble with \\0
                 l = l.replace('\\\\', '\\\\\n')
                 l = l.replace('\\0', '\0')
                 l = l.replace('\n', '')
             k, v = l.decode('string_escape').split(':', 1)
             extra[k] = v
     return extra
 
 def encodeextra(d):
     # keys must be sorted to produce a deterministic changelog entry
     items = [_string_escape('%s:%s' % (k, d[k])) for k in sorted(d)]
     return "\0".join(items)
 
 def stripdesc(desc):
     """strip trailing whitespace and leading and trailing empty lines"""
     return '\n'.join([l.rstrip() for l in desc.splitlines()]).strip('\n')
 
 class appender(object):
     '''the changelog index must be updated last on disk, so we use this class
     to delay writes to it'''
     def __init__(self, vfs, name, mode, buf):
         self.data = buf
         fp = vfs(name, mode)
         self.fp = fp
         self.offset = fp.tell()
         self.size = vfs.fstat(fp).st_size
 
     def end(self):
         return self.size + len("".join(self.data))
     def tell(self):
         return self.offset
     def flush(self):
         pass
     def close(self):
         self.fp.close()
 
     def seek(self, offset, whence=0):
         '''virtual file offset spans real file and data'''
         if whence == 0:
             self.offset = offset
         elif whence == 1:
             self.offset += offset
         elif whence == 2:
             self.offset = self.end() + offset
         if self.offset < self.size:
             self.fp.seek(self.offset)
 
     def read(self, count=-1):
         '''only trick here is reads that span real file and data'''
         ret = ""
         if self.offset < self.size:
             s = self.fp.read(count)
             ret = s
             self.offset += len(s)
             if count > 0:
                 count -= len(s)
         if count != 0:
             doff = self.offset - self.size
             self.data.insert(0, "".join(self.data))
             del self.data[1:]
             s = self.data[0][doff:doff + count]
             self.offset += len(s)
             ret += s
         return ret
 
     def write(self, s):
         self.data.append(str(s))
         self.offset += len(s)
 
 def _divertopener(opener, target):
     """build an opener that writes in 'target.a' instead of 'target'"""
     def _divert(name, mode='r'):
         if name != target:
             return opener(name, mode)
         return opener(name + ".a", mode)
     return _divert
 
 def _delayopener(opener, target, buf):
     """build an opener that stores chunks in 'buf' instead of 'target'"""
     def _delay(name, mode='r'):
         if name != target:
             return opener(name, mode)
         return appender(opener, name, mode, buf)
     return _delay
 
 class changelog(revlog.revlog):
     def __init__(self, opener):
         revlog.revlog.__init__(self, opener, "00changelog.i")
         if self._initempty:
             # changelogs don't benefit from generaldelta
             self.version &= ~revlog.REVLOGGENERALDELTA
             self._generaldelta = False
         self._realopener = opener
         self._delayed = False
         self._delaybuf = None
         self._divert = False
         self.filteredrevs = frozenset()
 
     def tip(self):
         """filtered version of revlog.tip"""
         for i in xrange(len(self) -1, -2, -1):
             if i not in self.filteredrevs:
                 return self.node(i)
 
     def __iter__(self):
         """filtered version of revlog.__iter__"""
         if len(self.filteredrevs) == 0:
             return revlog.revlog.__iter__(self)
 
         def filterediter():
             for i in xrange(len(self)):
                 if i not in self.filteredrevs:
                     yield i
 
         return filterediter()
 
     def revs(self, start=0, stop=None):
         """filtered version of revlog.revs"""
         for i in super(changelog, self).revs(start, stop):
             if i not in self.filteredrevs:
                 yield i
 
     @util.propertycache
     def nodemap(self):
         # XXX need filtering too
         self.rev(self.node(0))
         return self._nodecache
 
     def hasnode(self, node):
         """filtered version of revlog.hasnode"""
         try:
             i = self.rev(node)
             return i not in self.filteredrevs
         except KeyError:
             return False
 
     def headrevs(self):
         if self.filteredrevs:
             try:
                 return self.index.headrevsfiltered(self.filteredrevs)
             # AttributeError covers non-c-extension environments and
             # old c extensions without filter handling.
             except AttributeError:
                 return self._headrevs()
 
         return super(changelog, self).headrevs()
 
     def strip(self, *args, **kwargs):
         # XXX make something better than assert
         # We can't expect proper strip behavior if we are filtered.
         assert not self.filteredrevs
         super(changelog, self).strip(*args, **kwargs)
 
     def rev(self, node):
         """filtered version of revlog.rev"""
         r = super(changelog, self).rev(node)
         if r in self.filteredrevs:
             raise error.FilteredLookupError(hex(node), self.indexfile,
                                             _('filtered node'))
         return r
 
     def node(self, rev):
         """filtered version of revlog.node"""
         if rev in self.filteredrevs:
             raise error.FilteredIndexError(rev)
         return super(changelog, self).node(rev)
 
     def linkrev(self, rev):
         """filtered version of revlog.linkrev"""
         if rev in self.filteredrevs:
             raise error.FilteredIndexError(rev)
         return super(changelog, self).linkrev(rev)
 
     def parentrevs(self, rev):
         """filtered version of revlog.parentrevs"""
         if rev in self.filteredrevs:
             raise error.FilteredIndexError(rev)
         return super(changelog, self).parentrevs(rev)
 
     def flags(self, rev):
         """filtered version of revlog.flags"""
         if rev in self.filteredrevs:
             raise error.FilteredIndexError(rev)
         return super(changelog, self).flags(rev)
 
     def delayupdate(self, tr):
         "delay visibility of index updates to other readers"
 
         if not self._delayed:
             if len(self) == 0:
                 self._divert = True
                 if self._realopener.exists(self.indexfile + '.a'):
                     self._realopener.unlink(self.indexfile + '.a')
                 self.opener = _divertopener(self._realopener, self.indexfile)
             else:
                 self._delaybuf = []
                 self.opener = _delayopener(self._realopener, self.indexfile,
                                            self._delaybuf)
         self._delayed = True
         tr.addpending('cl-%i' % id(self), self._writepending)
+        trp = weakref.proxy(tr)
+        tr.addfinalize('cl-%i' % id(self), lambda: self._finalize(trp))
 
-    def finalize(self, tr):
+    def _finalize(self, tr):
         "finalize index updates"
         self._delayed = False
         self.opener = self._realopener
         # move redirected index data back into place
         if self._divert:
             assert not self._delaybuf
             tmpname = self.indexfile + ".a"
             nfile = self.opener.open(tmpname)
             nfile.close()
             self.opener.rename(tmpname, self.indexfile)
         elif self._delaybuf:
             fp = self.opener(self.indexfile, 'a')
             fp.write("".join(self._delaybuf))
             fp.close()
             self._delaybuf = None
         self._divert = False
         # split when we're done
         self.checkinlinesize(tr)
 
     def readpending(self, file):
         r = revlog.revlog(self.opener, file)
         self.index = r.index
         self.nodemap = r.nodemap
         self._nodecache = r._nodecache
         self._chunkcache = r._chunkcache
 
     def _writepending(self):
         "create a file containing the unfinalized state for pretxnchangegroup"
         if self._delaybuf:
             # make a temporary copy of the index
             fp1 = self._realopener(self.indexfile)
             fp2 = self._realopener(self.indexfile + ".a", "w")
             fp2.write(fp1.read())
             # add pending data
             fp2.write("".join(self._delaybuf))
             fp2.close()
             # switch modes so finalize can simply rename
             self._delaybuf = None
             self._divert = True
             self.opener = _divertopener(self._realopener, self.indexfile)
 
         if self._divert:
             return True
 
         return False
 
     def checkinlinesize(self, tr, fp=None):
         if not self._delayed:
             revlog.revlog.checkinlinesize(self, tr, fp)
 
     def read(self, node):
         """
         format used:
         nodeid\n        : manifest node in ascii
         user\n          : user, no \n or \r allowed
         time tz extra\n : date (time is int or float, timezone is int)
                         : extra is metadata, encoded and separated by '\0'
                         : older versions ignore it
         files\n\n       : files modified by the cset, no \n or \r allowed
         (.*)            : comment (free text, ideally utf-8)
 
         changelog v0 doesn't use extra
         """
         text = self.revision(node)
         if not text:
             return (nullid, "", (0, 0), [], "", _defaultextra)
         last = text.index("\n\n")
         desc = encoding.tolocal(text[last + 2:])
         l = text[:last].split('\n')
         manifest = bin(l[0])
         user = encoding.tolocal(l[1])
 
         tdata = l[2].split(' ', 2)
         if len(tdata) != 3:
             time = float(tdata[0])
             try:
                 # various tools did silly things with the time zone field.
                 timezone = int(tdata[1])
             except ValueError:
                 timezone = 0
             extra = _defaultextra
         else:
             time, timezone = float(tdata[0]), int(tdata[1])
             extra = decodeextra(tdata[2])
 
         files = l[3:]
         return (manifest, user, (time, timezone), files, desc, extra)
 
     def add(self, manifest, files, desc, transaction, p1, p2,
                   user, date=None, extra=None):
         # Convert to UTF-8 encoded bytestrings as the very first
         # thing: calling any method on a localstr object will turn it
         # into a str object and the cached UTF-8 string is thus lost.
         user, desc = encoding.fromlocal(user), encoding.fromlocal(desc)
 
         user = user.strip()
         # An empty username or a username with a "\n" will make the
         # revision text contain two "\n\n" sequences -> corrupt
         # repository since read cannot unpack the revision.
         if not user:
             raise error.RevlogError(_("empty username"))
         if "\n" in user:
             raise error.RevlogError(_("username %s contains a newline")
                                     % repr(user))
 
         desc = stripdesc(desc)
 
         if date:
             parseddate = "%d %d" % util.parsedate(date)
         else:
             parseddate = "%d %d" % util.makedate()
         if extra:
             branch = extra.get("branch")
             if branch in ("default", ""):
                 del extra["branch"]
             elif branch in (".", "null", "tip"):
                 raise error.RevlogError(_('the name \'%s\' is reserved')
                                         % branch)
         if extra:
             extra = encodeextra(extra)
             parseddate = "%s %s" % (parseddate, extra)
         l = [hex(manifest), user, parseddate] + sorted(files) + ["", desc]
         text = "\n".join(l)
         return self.addrevision(text, transaction, len(self), p1, p2)
 
     def branchinfo(self, rev):
         """return the branch name and open/close state of a revision
 
         This function exists because creating a changectx object
         just to access this is costly."""
         extra = self.read(rev)[5]
         return encoding.tolocal(extra.get("branch")), 'close' in extra

mercurial/localrepo.py

@@ -1,1804 +1,1803
 # localrepo.py - read/write repository class for mercurial
 #
 # Copyright 2005-2007 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 from node import hex, nullid, short
 from i18n import _
 import urllib
 import peer, changegroup, subrepo, pushkey, obsolete, repoview
 import changelog, dirstate, filelog, manifest, context, bookmarks, phases
 import lock as lockmod
 import transaction, store, encoding, exchange, bundle2
 import scmutil, util, extensions, hook, error, revset
 import match as matchmod
 import merge as mergemod
 import tags as tagsmod
 from lock import release
 import weakref, errno, os, time, inspect
 import branchmap, pathutil
 propertycache = util.propertycache
 filecache = scmutil.filecache
 
 class repofilecache(filecache):
     """All filecache usage on repo are done for logic that should be unfiltered
     """
 
     def __get__(self, repo, type=None):
         return super(repofilecache, self).__get__(repo.unfiltered(), type)
     def __set__(self, repo, value):
         return super(repofilecache, self).__set__(repo.unfiltered(), value)
     def __delete__(self, repo):
         return super(repofilecache, self).__delete__(repo.unfiltered())
 
 class storecache(repofilecache):
     """filecache for files in the store"""
     def join(self, obj, fname):
         return obj.sjoin(fname)
 
 class unfilteredpropertycache(propertycache):
     """propertycache that apply to unfiltered repo only"""
 
     def __get__(self, repo, type=None):
         unfi = repo.unfiltered()
         if unfi is repo:
             return super(unfilteredpropertycache, self).__get__(unfi)
         return getattr(unfi, self.name)
 
 class filteredpropertycache(propertycache):
     """propertycache that must take filtering in account"""
 
     def cachevalue(self, obj, value):
         object.__setattr__(obj, self.name, value)
 
 
 def hasunfilteredcache(repo, name):
     """check if a repo has an unfilteredpropertycache value for <name>"""
     return name in vars(repo.unfiltered())
 
 def unfilteredmethod(orig):
     """decorate method that always need to be run on unfiltered version"""
     def wrapper(repo, *args, **kwargs):
         return orig(repo.unfiltered(), *args, **kwargs)
     return wrapper
 
 moderncaps = set(('lookup', 'branchmap', 'pushkey', 'known', 'getbundle',
                   'unbundle'))
 legacycaps = moderncaps.union(set(['changegroupsubset']))
 
 class localpeer(peer.peerrepository):
     '''peer for a local repo; reflects only the most recent API'''
 
     def __init__(self, repo, caps=moderncaps):
         peer.peerrepository.__init__(self)
         self._repo = repo.filtered('served')
         self.ui = repo.ui
         self._caps = repo._restrictcapabilities(caps)
         self.requirements = repo.requirements
         self.supportedformats = repo.supportedformats
 
     def close(self):
         self._repo.close()
 
     def _capabilities(self):
         return self._caps
 
     def local(self):
         return self._repo
 
     def canpush(self):
         return True
 
     def url(self):
         return self._repo.url()
 
     def lookup(self, key):
         return self._repo.lookup(key)
 
     def branchmap(self):
         return self._repo.branchmap()
 
     def heads(self):
         return self._repo.heads()
 
     def known(self, nodes):
         return self._repo.known(nodes)
 
     def getbundle(self, source, heads=None, common=None, bundlecaps=None,
                   format='HG10', **kwargs):
         cg = exchange.getbundle(self._repo, source, heads=heads,
                                 common=common, bundlecaps=bundlecaps, **kwargs)
         if bundlecaps is not None and 'HG2Y' in bundlecaps:
             # When requesting a bundle2, getbundle returns a stream to make the
             # wire level function happier. We need to build a proper object
             # from it in local peer.
             cg = bundle2.unbundle20(self.ui, cg)
         return cg
 
     # TODO We might want to move the next two calls into legacypeer and add
     # unbundle instead.
 
     def unbundle(self, cg, heads, url):
         """apply a bundle on a repo
 
         This function handles the repo locking itself."""
         try:
             cg = exchange.readbundle(self.ui, cg, None)
             ret = exchange.unbundle(self._repo, cg, heads, 'push', url)
             if util.safehasattr(ret, 'getchunks'):
                 # This is a bundle20 object, turn it into an unbundler.
                 # This little dance should be dropped eventually when the API
                 # is finally improved.
                 stream = util.chunkbuffer(ret.getchunks())
                 ret = bundle2.unbundle20(self.ui, stream)
             return ret
         except error.PushRaced, exc:
             raise error.ResponseError(_('push failed:'), str(exc))
 
     def lock(self):
         return self._repo.lock()
 
     def addchangegroup(self, cg, source, url):
         return changegroup.addchangegroup(self._repo, cg, source, url)
 
     def pushkey(self, namespace, key, old, new):
         return self._repo.pushkey(namespace, key, old, new)
 
     def listkeys(self, namespace):
         return self._repo.listkeys(namespace)
 
     def debugwireargs(self, one, two, three=None, four=None, five=None):
         '''used to test argument passing over the wire'''
         return "%s %s %s %s %s" % (one, two, three, four, five)
 
 class locallegacypeer(localpeer):
     '''peer extension which implements legacy methods too; used for tests with
     restricted capabilities'''
 
     def __init__(self, repo):
         localpeer.__init__(self, repo, caps=legacycaps)
 
     def branches(self, nodes):
         return self._repo.branches(nodes)
 
     def between(self, pairs):
         return self._repo.between(pairs)
 
     def changegroup(self, basenodes, source):
         return changegroup.changegroup(self._repo, basenodes, source)
 
     def changegroupsubset(self, bases, heads, source):
         return changegroup.changegroupsubset(self._repo, bases, heads, source)
 
 class localrepository(object):
 
     supportedformats = set(('revlogv1', 'generaldelta'))
     _basesupported = supportedformats | set(('store', 'fncache', 'shared',
                                              'dotencode'))
     openerreqs = set(('revlogv1', 'generaldelta'))
     requirements = ['revlogv1']
     filtername = None
 
     # a list of (ui, featureset) functions.
     # only functions defined in module of enabled extensions are invoked
     featuresetupfuncs = set()
 
     def _baserequirements(self, create):
         return self.requirements[:]
 
     def __init__(self, baseui, path=None, create=False):
         self.wvfs = scmutil.vfs(path, expandpath=True, realpath=True)
         self.wopener = self.wvfs
         self.root = self.wvfs.base
         self.path = self.wvfs.join(".hg")
         self.origroot = path
         self.auditor = pathutil.pathauditor(self.root, self._checknested)
         self.vfs = scmutil.vfs(self.path)
         self.opener = self.vfs
         self.baseui = baseui
         self.ui = baseui.copy()
         self.ui.copy = baseui.copy # prevent copying repo configuration
         # A list of callback to shape the phase if no data were found.
         # Callback are in the form: func(repo, roots) --> processed root.
         # This list it to be filled by extension during repo setup
         self._phasedefaults = []
         try:
             self.ui.readconfig(self.join("hgrc"), self.root)
             extensions.loadall(self.ui)
         except IOError:
             pass
 
         if self.featuresetupfuncs:
             self.supported = set(self._basesupported) # use private copy
             extmods = set(m.__name__ for n, m
                           in extensions.extensions(self.ui))
             for setupfunc in self.featuresetupfuncs:
                 if setupfunc.__module__ in extmods:
                     setupfunc(self.ui, self.supported)
         else:
             self.supported = self._basesupported
 
         if not self.vfs.isdir():
             if create:
                 if not self.wvfs.exists():
                     self.wvfs.makedirs()
                 self.vfs.makedir(notindexed=True)
                 requirements = self._baserequirements(create)
                 if self.ui.configbool('format', 'usestore', True):
                     self.vfs.mkdir("store")
                     requirements.append("store")
                     if self.ui.configbool('format', 'usefncache', True):
                         requirements.append("fncache")
                         if self.ui.configbool('format', 'dotencode', True):
                             requirements.append('dotencode')
                     # create an invalid changelog
                     self.vfs.append(
                         "00changelog.i",
                         '\0\0\0\2' # represents revlogv2
                         ' dummy changelog to prevent using the old repo layout'
                     )
                 if self.ui.configbool('format', 'generaldelta', False):
                     requirements.append("generaldelta")
                 requirements = set(requirements)
             else:
                 raise error.RepoError(_("repository %s not found") % path)
         elif create:
             raise error.RepoError(_("repository %s already exists") % path)
         else:
             try:
                 requirements = scmutil.readrequires(self.vfs, self.supported)
             except IOError, inst:
                 if inst.errno != errno.ENOENT:
                     raise
                 requirements = set()
 
         self.sharedpath = self.path
         try:
             vfs = scmutil.vfs(self.vfs.read("sharedpath").rstrip('\n'),
                               realpath=True)
             s = vfs.base
             if not vfs.exists():
                 raise error.RepoError(
                     _('.hg/sharedpath points to nonexistent directory %s') % s)
             self.sharedpath = s
         except IOError, inst:
             if inst.errno != errno.ENOENT:
                 raise
 
         self.store = store.store(requirements, self.sharedpath, scmutil.vfs)
         self.spath = self.store.path
         self.svfs = self.store.vfs
         self.sopener = self.svfs
         self.sjoin = self.store.join
         self.vfs.createmode = self.store.createmode
         self._applyrequirements(requirements)
         if create:
             self._writerequirements()
 
 
         self._branchcaches = {}
         self.filterpats = {}
         self._datafilters = {}
         self._transref = self._lockref = self._wlockref = None
 
         # A cache for various files under .hg/ that tracks file changes,
         # (used by the filecache decorator)
         #
         # Maps a property name to its util.filecacheentry
         self._filecache = {}
 
         # hold sets of revision to be filtered
         # should be cleared when something might have changed the filter value:
         # - new changesets,
         # - phase change,
         # - new obsolescence marker,
         # - working directory parent change,
         # - bookmark changes
         self.filteredrevcache = {}
 
     def close(self):
         pass
 
     def _restrictcapabilities(self, caps):
         # bundle2 is not ready for prime time, drop it unless explicitly
         # required by the tests (or some brave tester)
         if self.ui.configbool('experimental', 'bundle2-exp', False):
             caps = set(caps)
             capsblob = bundle2.encodecaps(bundle2.getrepocaps(self))
             caps.add('bundle2-exp=' + urllib.quote(capsblob))
         return caps
 
     def _applyrequirements(self, requirements):
         self.requirements = requirements
         self.sopener.options = dict((r, 1) for r in requirements
                                            if r in self.openerreqs)
         chunkcachesize = self.ui.configint('format', 'chunkcachesize')
         if chunkcachesize is not None:
             self.sopener.options['chunkcachesize'] = chunkcachesize
 
     def _writerequirements(self):
         reqfile = self.opener("requires", "w")
         for r in sorted(self.requirements):
             reqfile.write("%s\n" % r)
         reqfile.close()
 
     def _checknested(self, path):
         """Determine if path is a legal nested repository."""
         if not path.startswith(self.root):
             return False
         subpath = path[len(self.root) + 1:]
         normsubpath = util.pconvert(subpath)
 
         # XXX: Checking against the current working copy is wrong in
         # the sense that it can reject things like
         #
         #   $ hg cat -r 10 sub/x.txt
         #
         # if sub/ is no longer a subrepository in the working copy
         # parent revision.
         #
         # However, it can of course also allow things that would have
         # been rejected before, such as the above cat command if sub/
         # is a subrepository now, but was a normal directory before.
         # The old path auditor would have rejected by mistake since it
         # panics when it sees sub/.hg/.
         #
         # All in all, checking against the working copy seems sensible
         # since we want to prevent access to nested repositories on
         # the filesystem *now*.
         ctx = self[None]
         parts = util.splitpath(subpath)
         while parts:
             prefix = '/'.join(parts)
             if prefix in ctx.substate:
                 if prefix == normsubpath:
                     return True
                 else:
                     sub = ctx.sub(prefix)
                     return sub.checknested(subpath[len(prefix) + 1:])
             else:
                 parts.pop()
         return False
 
     def peer(self):
         return localpeer(self) # not cached to avoid reference cycle
 
     def unfiltered(self):
         """Return unfiltered version of the repository
 
         Intended to be overwritten by filtered repo."""
         return self
 
     def filtered(self, name):
         """Return a filtered version of a repository"""
         # build a new class with the mixin and the current class
         # (possibly subclass of the repo)
         class proxycls(repoview.repoview, self.unfiltered().__class__):
             pass
         return proxycls(self, name)
 
     @repofilecache('bookmarks')
     def _bookmarks(self):
         return bookmarks.bmstore(self)
 
     @repofilecache('bookmarks.current')
     def _bookmarkcurrent(self):
         return bookmarks.readcurrent(self)
 
     def bookmarkheads(self, bookmark):
         name = bookmark.split('@', 1)[0]
         heads = []
         for mark, n in self._bookmarks.iteritems():
             if mark.split('@', 1)[0] == name:
                 heads.append(n)
         return heads
 
     @storecache('phaseroots')
     def _phasecache(self):
         return phases.phasecache(self, self._phasedefaults)
 
     @storecache('obsstore')
     def obsstore(self):
         # read default format for new obsstore.
         defaultformat = self.ui.configint('format', 'obsstore-version', None)
         # rely on obsstore class default when possible.
         kwargs = {}
         if defaultformat is not None:
             kwargs['defaultformat'] = defaultformat
         readonly = not obsolete.isenabled(self, obsolete.createmarkersopt)
         store = obsolete.obsstore(self.sopener, readonly=readonly,
                                   **kwargs)
         if store and readonly:
             # message is rare enough to not be translated
             msg = 'obsolete feature not enabled but %i markers found!\n'
             self.ui.warn(msg % len(list(store)))
         return store
 
     @storecache('00changelog.i')
     def changelog(self):
         c = changelog.changelog(self.sopener)
         if 'HG_PENDING' in os.environ:
             p = os.environ['HG_PENDING']
             if p.startswith(self.root):
                 c.readpending('00changelog.i.a')
         return c
 
     @storecache('00manifest.i')
     def manifest(self):
         return manifest.manifest(self.sopener)
 
     @repofilecache('dirstate')
     def dirstate(self):
         warned = [0]
         def validate(node):
             try:
                 self.changelog.rev(node)
                 return node
             except error.LookupError:
                 if not warned[0]:
                     warned[0] = True
                     self.ui.warn(_("warning: ignoring unknown"
                                    " working parent %s!\n") % short(node))
                 return nullid
 
         return dirstate.dirstate(self.opener, self.ui, self.root, validate)
 
     def __getitem__(self, changeid):
         if changeid is None:
             return context.workingctx(self)
         return context.changectx(self, changeid)
 
     def __contains__(self, changeid):
         try:
             return bool(self.lookup(changeid))
         except error.RepoLookupError:
             return False
 
     def __nonzero__(self):
         return True
 
     def __len__(self):
         return len(self.changelog)
 
     def __iter__(self):
         return iter(self.changelog)
 
     def revs(self, expr, *args):
         '''Return a list of revisions matching the given revset'''
         expr = revset.formatspec(expr, *args)
         m = revset.match(None, expr)
         return m(self, revset.spanset(self))
 
     def set(self, expr, *args):
         '''
         Yield a context for each matching revision, after doing arg
         replacement via revset.formatspec
         '''
         for r in self.revs(expr, *args):
             yield self[r]
 
     def url(self):
         return 'file:' + self.root
 
     def hook(self, name, throw=False, **args):
         """Call a hook, passing this repo instance.
 
         This a convenience method to aid invoking hooks. Extensions likely
         won't call this unless they have registered a custom hook or are
         replacing code that is expected to call a hook.
         """
         return hook.hook(self.ui, self, name, throw, **args)
 
     @unfilteredmethod
     def _tag(self, names, node, message, local, user, date, extra={},
              editor=False):
         if isinstance(names, str):
             names = (names,)
 
         branches = self.branchmap()
         for name in names:
             self.hook('pretag', throw=True, node=hex(node), tag=name,
                       local=local)
             if name in branches:
                 self.ui.warn(_("warning: tag %s conflicts with existing"
                 " branch name\n") % name)
 
         def writetags(fp, names, munge, prevtags):
             fp.seek(0, 2)
             if prevtags and prevtags[-1] != '\n':
                 fp.write('\n')
             for name in names:
                 m = munge and munge(name) or name
                 if (self._tagscache.tagtypes and
                     name in self._tagscache.tagtypes):
                     old = self.tags().get(name, nullid)
                     fp.write('%s %s\n' % (hex(old), m))
                 fp.write('%s %s\n' % (hex(node), m))
             fp.close()
 
         prevtags = ''
         if local:
             try:
                 fp = self.opener('localtags', 'r+')
             except IOError:
                 fp = self.opener('localtags', 'a')
             else:
                 prevtags = fp.read()
 
             # local tags are stored in the current charset
             writetags(fp, names, None, prevtags)
             for name in names:
                 self.hook('tag', node=hex(node), tag=name, local=local)
             return
 
         try:
             fp = self.wfile('.hgtags', 'rb+')
         except IOError, e:
             if e.errno != errno.ENOENT:
                 raise
             fp = self.wfile('.hgtags', 'ab')
         else:
             prevtags = fp.read()
 
         # committed tags are stored in UTF-8
         writetags(fp, names, encoding.fromlocal, prevtags)
 
         fp.close()
 
         self.invalidatecaches()
 
         if '.hgtags' not in self.dirstate:
             self[None].add(['.hgtags'])
 
         m = matchmod.exact(self.root, '', ['.hgtags'])
         tagnode = self.commit(message, user, date, extra=extra, match=m,
                               editor=editor)
 
         for name in names:
             self.hook('tag', node=hex(node), tag=name, local=local)
 
         return tagnode
 
     def tag(self, names, node, message, local, user, date, editor=False):
         '''tag a revision with one or more symbolic names.
 
         names is a list of strings or, when adding a single tag, names may be a
         string.
 
         if local is True, the tags are stored in a per-repository file.
         otherwise, they are stored in the .hgtags file, and a new
         changeset is committed with the change.
 
         keyword arguments:
 
         local: whether to store tags in non-version-controlled file
         (default False)
 
         message: commit message to use if committing
 
         user: name of user to use if committing
 
         date: date tuple to use if committing'''
 
         if not local:
             m = matchmod.exact(self.root, '', ['.hgtags'])
             if util.any(self.status(match=m, unknown=True, ignored=True)):
                 raise util.Abort(_('working copy of .hgtags is changed'),
                                  hint=_('please commit .hgtags manually'))
 
         self.tags() # instantiate the cache
         self._tag(names, node, message, local, user, date, editor=editor)
 
     @filteredpropertycache
     def _tagscache(self):
         '''Returns a tagscache object that contains various tags related
         caches.'''
 
         # This simplifies its cache management by having one decorated
         # function (this one) and the rest simply fetch things from it.
         class tagscache(object):
             def __init__(self):
                 # These two define the set of tags for this repository. tags
                 # maps tag name to node; tagtypes maps tag name to 'global' or
                 # 'local'. (Global tags are defined by .hgtags across all
                 # heads, and local tags are defined in .hg/localtags.)
                 # They constitute the in-memory cache of tags.
                 self.tags = self.tagtypes = None
 
                 self.nodetagscache = self.tagslist = None
 
         cache = tagscache()
         cache.tags, cache.tagtypes = self._findtags()
 
         return cache
 
     def tags(self):
         '''return a mapping of tag to node'''
         t = {}
         if self.changelog.filteredrevs:
             tags, tt = self._findtags()
         else:
             tags = self._tagscache.tags
         for k, v in tags.iteritems():
             try:
                 # ignore tags to unknown nodes
                 self.changelog.rev(v)
                 t[k] = v
             except (error.LookupError, ValueError):
                 pass
         return t
 
     def _findtags(self):
         '''Do the hard work of finding tags.  Return a pair of dicts
         (tags, tagtypes) where tags maps tag name to node, and tagtypes
         maps tag name to a string like \'global\' or \'local\'.
         Subclasses or extensions are free to add their own tags, but
         should be aware that the returned dicts will be retained for the
         duration of the localrepo object.'''
 
         # XXX what tagtype should subclasses/extensions use?  Currently
         # mq and bookmarks add tags, but do not set the tagtype at all.
         # Should each extension invent its own tag type?  Should there
         # be one tagtype for all such "virtual" tags?  Or is the status
         # quo fine?
 
         alltags = {}                    # map tag name to (node, hist)
         tagtypes = {}
 
         tagsmod.findglobaltags(self.ui, self, alltags, tagtypes)
         tagsmod.readlocaltags(self.ui, self, alltags, tagtypes)
 
         # Build the return dicts.  Have to re-encode tag names because
         # the tags module always uses UTF-8 (in order not to lose info
         # writing to the cache), but the rest of Mercurial wants them in
         # local encoding.
         tags = {}
         for (name, (node, hist)) in alltags.iteritems():
             if node != nullid:
                 tags[encoding.tolocal(name)] = node
         tags['tip'] = self.changelog.tip()
         tagtypes = dict([(encoding.tolocal(name), value)
                          for (name, value) in tagtypes.iteritems()])
         return (tags, tagtypes)
 
     def tagtype(self, tagname):
         '''
         return the type of the given tag. result can be:
 
         'local'  : a local tag
         'global' : a global tag
         None     : tag does not exist
         '''
 
         return self._tagscache.tagtypes.get(tagname)
 
     def tagslist(self):
         '''return a list of tags ordered by revision'''
         if not self._tagscache.tagslist:
             l = []
             for t, n in self.tags().iteritems():
                 l.append((self.changelog.rev(n), t, n))
             self._tagscache.tagslist = [(t, n) for r, t, n in sorted(l)]
 
         return self._tagscache.tagslist
 
     def nodetags(self, node):
         '''return the tags associated with a node'''
         if not self._tagscache.nodetagscache:
             nodetagscache = {}
             for t, n in self._tagscache.tags.iteritems():
                 nodetagscache.setdefault(n, []).append(t)
             for tags in nodetagscache.itervalues():
                 tags.sort()
             self._tagscache.nodetagscache = nodetagscache
         return self._tagscache.nodetagscache.get(node, [])
 
     def nodebookmarks(self, node):
         marks = []
         for bookmark, n in self._bookmarks.iteritems():
             if n == node:
                 marks.append(bookmark)
         return sorted(marks)
 
     def branchmap(self):
         '''returns a dictionary {branch: [branchheads]} with branchheads
         ordered by increasing revision number'''
         branchmap.updatecache(self)
         return self._branchcaches[self.filtername]
 
     def branchtip(self, branch):
         '''return the tip node for a given branch'''
         try:
             return self.branchmap().branchtip(branch)
         except KeyError:
             raise error.RepoLookupError(_("unknown branch '%s'") % branch)
 
     def lookup(self, key):
         return self[key].node()
 
     def lookupbranch(self, key, remote=None):
         repo = remote or self
         if key in repo.branchmap():
             return key
 
         repo = (remote and remote.local()) and remote or self
         return repo[key].branch()
 
     def known(self, nodes):
         nm = self.changelog.nodemap
         pc = self._phasecache
         result = []
         for n in nodes:
             r = nm.get(n)
             resp = not (r is None or pc.phase(self, r) >= phases.secret)
             result.append(resp)
         return result
 
     def local(self):
         return self
 
     def cancopy(self):
         # so statichttprepo's override of local() works
         if not self.local():
             return False
         if not self.ui.configbool('phases', 'publish', True):
             return True
         # if publishing we can't copy if there is filtered content
         return not self.filtered('visible').changelog.filteredrevs
 
     def join(self, f, *insidef):
         return os.path.join(self.path, f, *insidef)
 
     def wjoin(self, f, *insidef):
         return os.path.join(self.root, f, *insidef)
 
     def file(self, f):
         if f[0] == '/':
             f = f[1:]
         return filelog.filelog(self.sopener, f)
 
     def changectx(self, changeid):
         return self[changeid]
 
     def parents(self, changeid=None):
         '''get list of changectxs for parents of changeid'''
         return self[changeid].parents()
 
     def setparents(self, p1, p2=nullid):
         self.dirstate.beginparentchange()
         copies = self.dirstate.setparents(p1, p2)
         pctx = self[p1]
         if copies:
             # Adjust copy records, the dirstate cannot do it, it
             # requires access to parents manifests. Preserve them
             # only for entries added to first parent.
             for f in copies:
                 if f not in pctx and copies[f] in pctx:
                     self.dirstate.copy(copies[f], f)
         if p2 == nullid:
             for f, s in sorted(self.dirstate.copies().items()):
                 if f not in pctx and s not in pctx:
                     self.dirstate.copy(None, f)
         self.dirstate.endparentchange()
 
     def filectx(self, path, changeid=None, fileid=None):
         """changeid can be a changeset revision, node, or tag.
            fileid can be a file revision or node."""
         return context.filectx(self, path, changeid, fileid)
 
     def getcwd(self):
         return self.dirstate.getcwd()
 
     def pathto(self, f, cwd=None):
         return self.dirstate.pathto(f, cwd)
 
     def wfile(self, f, mode='r'):
         return self.wopener(f, mode)
 
     def _link(self, f):
         return self.wvfs.islink(f)
 
     def _loadfilter(self, filter):
         if filter not in self.filterpats:
             l = []
             for pat, cmd in self.ui.configitems(filter):
                 if cmd == '!':
                     continue
                 mf = matchmod.match(self.root, '', [pat])
                 fn = None
                 params = cmd
                 for name, filterfn in self._datafilters.iteritems():
                     if cmd.startswith(name):
                         fn = filterfn
                         params = cmd[len(name):].lstrip()
                         break
                 if not fn:
                     fn = lambda s, c, **kwargs: util.filter(s, c)
                 # Wrap old filters not supporting keyword arguments
                 if not inspect.getargspec(fn)[2]:
                     oldfn = fn
                     fn = lambda s, c, **kwargs: oldfn(s, c)
                 l.append((mf, fn, params))
             self.filterpats[filter] = l
         return self.filterpats[filter]
 
     def _filter(self, filterpats, filename, data):
         for mf, fn, cmd in filterpats:
             if mf(filename):
                 self.ui.debug("filtering %s through %s\n" % (filename, cmd))
                 data = fn(data, cmd, ui=self.ui, repo=self, filename=filename)
                 break
 
         return data
 
     @unfilteredpropertycache
     def _encodefilterpats(self):
         return self._loadfilter('encode')
 
     @unfilteredpropertycache
     def _decodefilterpats(self):
         return self._loadfilter('decode')
 
     def adddatafilter(self, name, filter):
         self._datafilters[name] = filter
 
     def wread(self, filename):
         if self._link(filename):
             data = self.wvfs.readlink(filename)
         else:
             data = self.wopener.read(filename)
         return self._filter(self._encodefilterpats, filename, data)
 
     def wwrite(self, filename, data, flags):
         data = self._filter(self._decodefilterpats, filename, data)
         if 'l' in flags:
             self.wopener.symlink(data, filename)
         else:
             self.wopener.write(filename, data)
             if 'x' in flags:
                 self.wvfs.setflags(filename, False, True)
 
     def wwritedata(self, filename, data):
         return self._filter(self._decodefilterpats, filename, data)
 
     def transaction(self, desc, report=None):
         tr = self._transref and self._transref() or None
         if tr and tr.running():
             return tr.nest()
 
         # abort here if the journal already exists
         if self.svfs.exists("journal"):
             raise error.RepoError(
                 _("abandoned transaction found"),
                 hint=_("run 'hg recover' to clean up transaction"))
 
         def onclose():
             self.store.write(self._transref())
 
         self._writejournal(desc)
         renames = [(vfs, x, undoname(x)) for vfs, x in self._journalfiles()]
         rp = report and report or self.ui.warn
         tr = transaction.transaction(rp, self.sopener,
                                      "journal",
                                      aftertrans(renames),
                                      self.store.createmode,
                                      onclose)
         self._transref = weakref.ref(tr)
         return tr
 
     def _journalfiles(self):
         return ((self.svfs, 'journal'),
                 (self.vfs, 'journal.dirstate'),
                 (self.vfs, 'journal.branch'),
                 (self.vfs, 'journal.desc'),
                 (self.vfs, 'journal.bookmarks'),
                 (self.svfs, 'journal.phaseroots'))
 
     def undofiles(self):
         return [(vfs, undoname(x)) for vfs, x in self._journalfiles()]
 
     def _writejournal(self, desc):
         self.opener.write("journal.dirstate",
                           self.opener.tryread("dirstate"))
         self.opener.write("journal.branch",
                           encoding.fromlocal(self.dirstate.branch()))
         self.opener.write("journal.desc",
                           "%d\n%s\n" % (len(self), desc))
         self.opener.write("journal.bookmarks",
                           self.opener.tryread("bookmarks"))
         self.sopener.write("journal.phaseroots",
                            self.sopener.tryread("phaseroots"))
 
     def recover(self):
         lock = self.lock()
         try:
             if self.svfs.exists("journal"):
                 self.ui.status(_("rolling back interrupted transaction\n"))
                 transaction.rollback(self.sopener, "journal",
                                      self.ui.warn)
                 self.invalidate()
                 return True
             else:
                 self.ui.warn(_("no interrupted transaction available\n"))
                 return False
         finally:
             lock.release()
 
     def rollback(self, dryrun=False, force=False):
         wlock = lock = None
         try:
             wlock = self.wlock()
             lock = self.lock()
             if self.svfs.exists("undo"):
                 return self._rollback(dryrun, force)
             else:
                 self.ui.warn(_("no rollback information available\n"))
                 return 1
         finally:
             release(lock, wlock)
 
     @unfilteredmethod # Until we get smarter cache management
     def _rollback(self, dryrun, force):
         ui = self.ui
         try:
             args = self.opener.read('undo.desc').splitlines()
             (oldlen, desc, detail) = (int(args[0]), args[1], None)
             if len(args) >= 3:
                 detail = args[2]
             oldtip = oldlen - 1
 
             if detail and ui.verbose:
                 msg = (_('repository tip rolled back to revision %s'
                          ' (undo %s: %s)\n')
                        % (oldtip, desc, detail))
             else:
                 msg = (_('repository tip rolled back to revision %s'
                          ' (undo %s)\n')
                        % (oldtip, desc))
         except IOError:
             msg = _('rolling back unknown transaction\n')
             desc = None
 
         if not force and self['.'] != self['tip'] and desc == 'commit':
             raise util.Abort(
                 _('rollback of last commit while not checked out '
                   'may lose data'), hint=_('use -f to force'))
 
         ui.status(msg)
         if dryrun:
             return 0
 
         parents = self.dirstate.parents()
         self.destroying()
         transaction.rollback(self.sopener, 'undo', ui.warn)
         if self.vfs.exists('undo.bookmarks'):
             self.vfs.rename('undo.bookmarks', 'bookmarks')
         if self.svfs.exists('undo.phaseroots'):
             self.svfs.rename('undo.phaseroots', 'phaseroots')
         self.invalidate()
 
         parentgone = (parents[0] not in self.changelog.nodemap or
                       parents[1] not in self.changelog.nodemap)
         if parentgone:
             self.vfs.rename('undo.dirstate', 'dirstate')
             try:
                 branch = self.opener.read('undo.branch')
                 self.dirstate.setbranch(encoding.tolocal(branch))
             except IOError:
                 ui.warn(_('named branch could not be reset: '
                           'current branch is still \'%s\'\n')
                         % self.dirstate.branch())
 
             self.dirstate.invalidate()
             parents = tuple([p.rev() for p in self.parents()])
             if len(parents) > 1:
                 ui.status(_('working directory now based on '
                             'revisions %d and %d\n') % parents)
             else:
                 ui.status(_('working directory now based on '
                             'revision %d\n') % parents)
         # TODO: if we know which new heads may result from this rollback, pass
         # them to destroy(), which will prevent the branchhead cache from being
         # invalidated.
         self.destroyed()
         return 0
 
     def invalidatecaches(self):
 
         if '_tagscache' in vars(self):
             # can't use delattr on proxy
             del self.__dict__['_tagscache']
 
         self.unfiltered()._branchcaches.clear()
         self.invalidatevolatilesets()
 
     def invalidatevolatilesets(self):
         self.filteredrevcache.clear()
         obsolete.clearobscaches(self)
 
     def invalidatedirstate(self):
         '''Invalidates the dirstate, causing the next call to dirstate
         to check if it was modified since the last time it was read,
         rereading it if it has.
 
         This is different to dirstate.invalidate() that it doesn't always
         rereads the dirstate. Use dirstate.invalidate() if you want to
         explicitly read the dirstate again (i.e. restoring it to a previous
         known good state).'''
         if hasunfilteredcache(self, 'dirstate'):
             for k in self.dirstate._filecache:
                 try:
                     delattr(self.dirstate, k)
                 except AttributeError:
                     pass
             delattr(self.unfiltered(), 'dirstate')
 
     def invalidate(self):
         unfiltered = self.unfiltered() # all file caches are stored unfiltered
         for k in self._filecache:
             # dirstate is invalidated separately in invalidatedirstate()
             if k == 'dirstate':
                 continue
 
             try:
                 delattr(unfiltered, k)
             except AttributeError:
                 pass
         self.invalidatecaches()
         self.store.invalidatecaches()
 
     def invalidateall(self):
         '''Fully invalidates both store and non-store parts, causing the
         subsequent operation to reread any outside changes.'''
         # extension should hook this to invalidate its caches
         self.invalidate()
         self.invalidatedirstate()
 
     def _lock(self, vfs, lockname, wait, releasefn, acquirefn, desc):
         try:
             l = lockmod.lock(vfs, lockname, 0, releasefn, desc=desc)
         except error.LockHeld, inst:
             if not wait:
                 raise
             self.ui.warn(_("waiting for lock on %s held by %r\n") %
                          (desc, inst.locker))
             # default to 600 seconds timeout
             l = lockmod.lock(vfs, lockname,
                              int(self.ui.config("ui", "timeout", "600")),
                              releasefn, desc=desc)
             self.ui.warn(_("got lock after %s seconds\n") % l.delay)
         if acquirefn:
             acquirefn()
         return l
 
     def _afterlock(self, callback):
         """add a callback to the current repository lock.
 
         The callback will be executed on lock release."""
         l = self._lockref and self._lockref()
         if l:
             l.postrelease.append(callback)
         else:
             callback()
 
     def lock(self, wait=True):
         '''Lock the repository store (.hg/store) and return a weak reference
         to the lock. Use this before modifying the store (e.g. committing or
         stripping). If you are opening a transaction, get a lock as well.)'''
         l = self._lockref and self._lockref()
         if l is not None and l.held:
             l.lock()
             return l
 
         def unlock():
             for k, ce in self._filecache.items():
                 if k == 'dirstate' or k not in self.__dict__:
                     continue
                 ce.refresh()
 
         l = self._lock(self.svfs, "lock", wait, unlock,
                        self.invalidate, _('repository %s') % self.origroot)
         self._lockref = weakref.ref(l)
         return l
 
     def wlock(self, wait=True):
         '''Lock the non-store parts of the repository (everything under
         .hg except .hg/store) and return a weak reference to the lock.
         Use this before modifying files in .hg.'''
         l = self._wlockref and self._wlockref()
         if l is not None and l.held:
             l.lock()
             return l
 
         def unlock():
             if self.dirstate.pendingparentchange():
                 self.dirstate.invalidate()
             else:
                 self.dirstate.write()
 
             self._filecache['dirstate'].refresh()
 
         l = self._lock(self.vfs, "wlock", wait, unlock,
                        self.invalidatedirstate, _('working directory of %s') %
                        self.origroot)
         self._wlockref = weakref.ref(l)
         return l
 
     def _filecommit(self, fctx, manifest1, manifest2, linkrev, tr, changelist):
         """
         commit an individual file as part of a larger transaction
         """
 
         fname = fctx.path()
         text = fctx.data()
         flog = self.file(fname)
         fparent1 = manifest1.get(fname, nullid)
         fparent2 = manifest2.get(fname, nullid)
 
         meta = {}
         copy = fctx.renamed()
         if copy and copy[0] != fname:
             # Mark the new revision of this file as a copy of another
             # file.  This copy data will effectively act as a parent
             # of this new revision.  If this is a merge, the first
             # parent will be the nullid (meaning "look up the copy data")
             # and the second one will be the other parent.  For example:
             #
             # 0 --- 1 --- 3   rev1 changes file foo
             #   \       /     rev2 renames foo to bar and changes it
             #    \- 2 -/      rev3 should have bar with all changes and
             #                      should record that bar descends from
             #                      bar in rev2 and foo in rev1
             #
             # this allows this merge to succeed:
             #
             # 0 --- 1 --- 3   rev4 reverts the content change from rev2
             #   \       /     merging rev3 and rev4 should use bar@rev2
             #    \- 2 --- 4        as the merge base
             #
 
             cfname = copy[0]
             crev = manifest1.get(cfname)
             newfparent = fparent2
 
             if manifest2: # branch merge
                 if fparent2 == nullid or crev is None: # copied on remote side
                     if cfname in manifest2:
                         crev = manifest2[cfname]
                         newfparent = fparent1
 
             # find source in nearest ancestor if we've lost track
             if not crev:
                 self.ui.debug(" %s: searching for copy revision for %s\n" %
                               (fname, cfname))
                 for ancestor in self[None].ancestors():
                     if cfname in ancestor:
                         crev = ancestor[cfname].filenode()
                         break
 
             if crev:
                 self.ui.debug(" %s: copy %s:%s\n" % (fname, cfname, hex(crev)))
                 meta["copy"] = cfname
                 meta["copyrev"] = hex(crev)
                 fparent1, fparent2 = nullid, newfparent
             else:
                 self.ui.warn(_("warning: can't find ancestor for '%s' "
                                "copied from '%s'!\n") % (fname, cfname))
 
         elif fparent1 == nullid:
             fparent1, fparent2 = fparent2, nullid
         elif fparent2 != nullid:
             # is one parent an ancestor of the other?
             fparentancestors = flog.commonancestorsheads(fparent1, fparent2)
             if fparent1 in fparentancestors:
                 fparent1, fparent2 = fparent2, nullid
             elif fparent2 in fparentancestors:
                 fparent2 = nullid
 
         # is the file changed?
         if fparent2 != nullid or flog.cmp(fparent1, text) or meta:
             changelist.append(fname)
             return flog.add(text, meta, tr, linkrev, fparent1, fparent2)
         # are just the flags changed during merge?
         elif fname in manifest1 and manifest1.flags(fname) != fctx.flags():
             changelist.append(fname)
 
         return fparent1
 
     @unfilteredmethod
     def commit(self, text="", user=None, date=None, match=None, force=False,
                editor=False, extra={}):
         """Add a new revision to current repository.
 
         Revision information is gathered from the working directory,
         match can be used to filter the committed files. If editor is
         supplied, it is called to get a commit message.
         """
 
         def fail(f, msg):
             raise util.Abort('%s: %s' % (f, msg))
 
         if not match:
             match = matchmod.always(self.root, '')
 
         if not force:
             vdirs = []
             match.explicitdir = vdirs.append
             match.bad = fail
 
         wlock = self.wlock()
         try:
             wctx = self[None]
             merge = len(wctx.parents()) > 1
 
             if (not force and merge and match and
                 (match.files() or match.anypats())):
                 raise util.Abort(_('cannot partially commit a merge '
                                    '(do not specify files or patterns)'))
 
             status = self.status(match=match, clean=force)
             if force:
                 status.modified.extend(status.clean) # mq may commit clean files
 
             # check subrepos
             subs = []
             commitsubs = set()
             newstate = wctx.substate.copy()
             # only manage subrepos and .hgsubstate if .hgsub is present
             if '.hgsub' in wctx:
                 # we'll decide whether to track this ourselves, thanks
                 for c in status.modified, status.added, status.removed:
                     if '.hgsubstate' in c:
                         c.remove('.hgsubstate')
 
                 # compare current state to last committed state
                 # build new substate based on last committed state
                 oldstate = wctx.p1().substate
                 for s in sorted(newstate.keys()):
                     if not match(s):
                         # ignore working copy, use old state if present
                         if s in oldstate:
                             newstate[s] = oldstate[s]
                             continue
                         if not force:
                             raise util.Abort(
                                 _("commit with new subrepo %s excluded") % s)
                     if wctx.sub(s).dirty(True):
                         if not self.ui.configbool('ui', 'commitsubrepos'):
                             raise util.Abort(
                                 _("uncommitted changes in subrepo %s") % s,
                                 hint=_("use --subrepos for recursive commit"))
                         subs.append(s)
                         commitsubs.add(s)
                     else:
                         bs = wctx.sub(s).basestate()
                         newstate[s] = (newstate[s][0], bs, newstate[s][2])
                         if oldstate.get(s, (None, None, None))[1] != bs:
                             subs.append(s)
 
                 # check for removed subrepos
                 for p in wctx.parents():
                     r = [s for s in p.substate if s not in newstate]
                     subs += [s for s in r if match(s)]
                 if subs:
                     if (not match('.hgsub') and
                         '.hgsub' in (wctx.modified() + wctx.added())):
                         raise util.Abort(
                             _("can't commit subrepos without .hgsub"))
                     status.modified.insert(0, '.hgsubstate')
 
             elif '.hgsub' in status.removed:
                 # clean up .hgsubstate when .hgsub is removed
                 if ('.hgsubstate' in wctx and
                     '.hgsubstate' not in (status.modified + status.added +
                                           status.removed)):
                     status.removed.insert(0, '.hgsubstate')
 
             # make sure all explicit patterns are matched
             if not force and match.files():
                 matched = set(status.modified + status.added + status.removed)
 
                 for f in match.files():
                     f = self.dirstate.normalize(f)
                     if f == '.' or f in matched or f in wctx.substate:
                         continue
                     if f in status.deleted:
                         fail(f, _('file not found!'))
                     if f in vdirs: # visited directory
                         d = f + '/'
                         for mf in matched:
                             if mf.startswith(d):
                                 break
                         else:
                             fail(f, _("no match under directory!"))
                     elif f not in self.dirstate:
                         fail(f, _("file not tracked!"))
 
             cctx = context.workingctx(self, text, user, date, extra, status)
 
             if (not force and not extra.get("close") and not merge
                 and not cctx.files()
                 and wctx.branch() == wctx.p1().branch()):
                 return None
 
             if merge and cctx.deleted():
                 raise util.Abort(_("cannot commit merge with missing files"))
 
             ms = mergemod.mergestate(self)
             for f in status.modified:
                 if f in ms and ms[f] == 'u':
                     raise util.Abort(_("unresolved merge conflicts "
                                        "(see hg help resolve)"))
 
             if editor:
                 cctx._text = editor(self, cctx, subs)
             edited = (text != cctx._text)
 
             # Save commit message in case this transaction gets rolled back
             # (e.g. by a pretxncommit hook).  Leave the content alone on
             # the assumption that the user will use the same editor again.
             msgfn = self.savecommitmessage(cctx._text)
 
             # commit subs and write new state
             if subs:
                 for s in sorted(commitsubs):
                     sub = wctx.sub(s)
                     self.ui.status(_('committing subrepository %s\n') %
                         subrepo.subrelpath(sub))
                     sr = sub.commit(cctx._text, user, date)
                     newstate[s] = (newstate[s][0], sr)
                 subrepo.writestate(self, newstate)
 
             p1, p2 = self.dirstate.parents()
             hookp1, hookp2 = hex(p1), (p2 != nullid and hex(p2) or '')
             try:
                 self.hook("precommit", throw=True, parent1=hookp1,
                           parent2=hookp2)
                 ret = self.commitctx(cctx, True)
             except: # re-raises
                 if edited:
                     self.ui.write(
                         _('note: commit message saved in %s\n') % msgfn)
                 raise
 
             # update bookmarks, dirstate and mergestate
             bookmarks.update(self, [p1, p2], ret)
             cctx.markcommitted(ret)
             ms.reset()
         finally:
             wlock.release()
 
         def commithook(node=hex(ret), parent1=hookp1, parent2=hookp2):
             # hack for command that use a temporary commit (eg: histedit)
             # temporary commit got stripped before hook release
             if node in self:
                 self.hook("commit", node=node, parent1=parent1,
                           parent2=parent2)
         self._afterlock(commithook)
         return ret
 
     @unfilteredmethod
     def commitctx(self, ctx, error=False):
         """Add a new revision to current repository.
         Revision information is passed via the context argument.
         """
 
         tr = None
         p1, p2 = ctx.p1(), ctx.p2()
         user = ctx.user()
 
         lock = self.lock()
         try:
             tr = self.transaction("commit")
             trp = weakref.proxy(tr)
 
             if ctx.files():
                 m1 = p1.manifest()
                 m2 = p2.manifest()
                 m = m1.copy()
 
                 # check in files
                 added = []
                 changed = []
                 removed = list(ctx.removed())
                 linkrev = len(self)
                 for f in sorted(ctx.modified() + ctx.added()):
                     self.ui.note(f + "\n")
                     try:
                         fctx = ctx[f]
                         if fctx is None:
                             removed.append(f)
                         else:
                             added.append(f)
                             m[f] = self._filecommit(fctx, m1, m2, linkrev,
                                                     trp, changed)
                             m.setflag(f, fctx.flags())
                     except OSError, inst:
                         self.ui.warn(_("trouble committing %s!\n") % f)
                         raise
                     except IOError, inst:
                         errcode = getattr(inst, 'errno', errno.ENOENT)
                         if error or errcode and errcode != errno.ENOENT:
                             self.ui.warn(_("trouble committing %s!\n") % f)
                         raise
 
                 # update manifest
                 removed = [f for f in sorted(removed) if f in m1 or f in m2]
                 drop = [f for f in removed if f in m]
                 for f in drop:
                     del m[f]
                 mn = self.manifest.add(m, trp, linkrev,
                                        p1.manifestnode(), p2.manifestnode(),
                                        added, drop)
                 files = changed + removed
             else:
                 mn = p1.manifestnode()
                 files = []
 
             # update changelog
             self.changelog.delayupdate(tr)
             n = self.changelog.add(mn, files, ctx.description(),
                                    trp, p1.node(), p2.node(),
                                    user, ctx.date(), ctx.extra().copy())
             p = lambda: tr.writepending() and self.root or ""
             xp1, xp2 = p1.hex(), p2 and p2.hex() or ''
             self.hook('pretxncommit', throw=True, node=hex(n), parent1=xp1,
                       parent2=xp2, pending=p)
-            self.changelog.finalize(trp)
             # set the new commit is proper phase
             targetphase = subrepo.newcommitphase(self.ui, ctx)
             if targetphase:
                 # retract boundary do not alter parent changeset.
                 # if a parent have higher the resulting phase will
                 # be compliant anyway
                 #
                 # if minimal phase was 0 we don't need to retract anything
                 phases.retractboundary(self, tr, targetphase, [n])
             tr.close()
             branchmap.updatecache(self.filtered('served'))
             return n
         finally:
             if tr:
                 tr.release()
             lock.release()
 
     @unfilteredmethod
     def destroying(self):
         '''Inform the repository that nodes are about to be destroyed.
         Intended for use by strip and rollback, so there's a common
         place for anything that has to be done before destroying history.
 
         This is mostly useful for saving state that is in memory and waiting
         to be flushed when the current lock is released. Because a call to
         destroyed is imminent, the repo will be invalidated causing those
         changes to stay in memory (waiting for the next unlock), or vanish
         completely.
         '''
         # When using the same lock to commit and strip, the phasecache is left
         # dirty after committing. Then when we strip, the repo is invalidated,
         # causing those changes to disappear.
         if '_phasecache' in vars(self):
             self._phasecache.write()
 
     @unfilteredmethod
     def destroyed(self):
         '''Inform the repository that nodes have been destroyed.
         Intended for use by strip and rollback, so there's a common
         place for anything that has to be done after destroying history.
         '''
         # When one tries to:
         # 1) destroy nodes thus calling this method (e.g. strip)
         # 2) use phasecache somewhere (e.g. commit)
         #
         # then 2) will fail because the phasecache contains nodes that were
         # removed. We can either remove phasecache from the filecache,
         # causing it to reload next time it is accessed, or simply filter
         # the removed nodes now and write the updated cache.
         self._phasecache.filterunknown(self)
         self._phasecache.write()
 
         # update the 'served' branch cache to help read only server process
         # Thanks to branchcache collaboration this is done from the nearest
         # filtered subset and it is expected to be fast.
         branchmap.updatecache(self.filtered('served'))
 
         # Ensure the persistent tag cache is updated.  Doing it now
         # means that the tag cache only has to worry about destroyed
         # heads immediately after a strip/rollback.  That in turn
         # guarantees that "cachetip == currenttip" (comparing both rev
         # and node) always means no nodes have been added or destroyed.
 
         # XXX this is suboptimal when qrefresh'ing: we strip the current
         # head, refresh the tag cache, then immediately add a new head.
         # But I think doing it this way is necessary for the "instant
         # tag cache retrieval" case to work.
         self.invalidate()
 
     def walk(self, match, node=None):
         '''
         walk recursively through the directory tree or a given
         changeset, finding all files matched by the match
         function
         '''
         return self[node].walk(match)
 
     def status(self, node1='.', node2=None, match=None,
                ignored=False, clean=False, unknown=False,
                listsubrepos=False):
         '''a convenience method that calls node1.status(node2)'''
         return self[node1].status(node2, match, ignored, clean, unknown,
                                   listsubrepos)
 
     def heads(self, start=None):
         heads = self.changelog.heads(start)
         # sort the output in rev descending order
         return sorted(heads, key=self.changelog.rev, reverse=True)
 
     def branchheads(self, branch=None, start=None, closed=False):
         '''return a (possibly filtered) list of heads for the given branch
 
         Heads are returned in topological order, from newest to oldest.
         If branch is None, use the dirstate branch.
         If start is not None, return only heads reachable from start.
         If closed is True, return heads that are marked as closed as well.
         '''
         if branch is None:
             branch = self[None].branch()
         branches = self.branchmap()
         if branch not in branches:
             return []
         # the cache returns heads ordered lowest to highest
         bheads = list(reversed(branches.branchheads(branch, closed=closed)))
         if start is not None:
             # filter out the heads that cannot be reached from startrev
             fbheads = set(self.changelog.nodesbetween([start], bheads)[2])
             bheads = [h for h in bheads if h in fbheads]
         return bheads
 
     def branches(self, nodes):
         if not nodes:
             nodes = [self.changelog.tip()]
         b = []
         for n in nodes:
             t = n
             while True:
                 p = self.changelog.parents(n)
                 if p[1] != nullid or p[0] == nullid:
                     b.append((t, n, p[0], p[1]))
                     break
                 n = p[0]
         return b
 
     def between(self, pairs):
         r = []
 
         for top, bottom in pairs:
             n, l, i = top, [], 0
             f = 1
 
             while n != bottom and n != nullid:
                 p = self.changelog.parents(n)[0]
                 if i == f:
                     l.append(n)
                     f = f * 2
                 n = p
                 i += 1
 
             r.append(l)
 
         return r
 
     def checkpush(self, pushop):
         """Extensions can override this function if additional checks have
         to be performed before pushing, or call it if they override push
         command.
         """
         pass
 
     @unfilteredpropertycache
     def prepushoutgoinghooks(self):
         """Return util.hooks consists of "(repo, remote, outgoing)"
         functions, which are called before pushing changesets.
         """
         return util.hooks()
 
     def stream_in(self, remote, requirements):
         lock = self.lock()
         try:
             # Save remote branchmap. We will use it later
             # to speed up branchcache creation
             rbranchmap = None
             if remote.capable("branchmap"):
                 rbranchmap = remote.branchmap()
 
             fp = remote.stream_out()
             l = fp.readline()
             try:
                 resp = int(l)
             except ValueError:
                 raise error.ResponseError(
                     _('unexpected response from remote server:'), l)
             if resp == 1:
                 raise util.Abort(_('operation forbidden by server'))
             elif resp == 2:
                 raise util.Abort(_('locking the remote repository failed'))
             elif resp != 0:
                 raise util.Abort(_('the server sent an unknown error code'))
             self.ui.status(_('streaming all changes\n'))
             l = fp.readline()
             try:
                 total_files, total_bytes = map(int, l.split(' ', 1))
             except (ValueError, TypeError):
                 raise error.ResponseError(
                     _('unexpected response from remote server:'), l)
             self.ui.status(_('%d files to transfer, %s of data\n') %
                            (total_files, util.bytecount(total_bytes)))
             handled_bytes = 0
             self.ui.progress(_('clone'), 0, total=total_bytes)
             start = time.time()
 
             tr = self.transaction(_('clone'))
             try:
                 for i in xrange(total_files):
                     # XXX doesn't support '\n' or '\r' in filenames
                     l = fp.readline()
                     try:
                         name, size = l.split('\0', 1)
                         size = int(size)
                     except (ValueError, TypeError):
                         raise error.ResponseError(
                             _('unexpected response from remote server:'), l)
                     if self.ui.debugflag:
                         self.ui.debug('adding %s (%s)\n' %
                                       (name, util.bytecount(size)))
                     # for backwards compat, name was partially encoded
                     ofp = self.sopener(store.decodedir(name), 'w')
                     for chunk in util.filechunkiter(fp, limit=size):
                         handled_bytes += len(chunk)
                         self.ui.progress(_('clone'), handled_bytes,
                                          total=total_bytes)
                         ofp.write(chunk)
                     ofp.close()
                 tr.close()
             finally:
                 tr.release()
 
             # Writing straight to files circumvented the inmemory caches
             self.invalidate()
 
             elapsed = time.time() - start
             if elapsed <= 0:
                 elapsed = 0.001
             self.ui.progress(_('clone'), None)
             self.ui.status(_('transferred %s in %.1f seconds (%s/sec)\n') %
                            (util.bytecount(total_bytes), elapsed,
                             util.bytecount(total_bytes / elapsed)))
 
             # new requirements = old non-format requirements +
             #                    new format-related
             # requirements from the streamed-in repository
             requirements.update(set(self.requirements) - self.supportedformats)
             self._applyrequirements(requirements)
             self._writerequirements()
 
             if rbranchmap:
                 rbheads = []
                 closed = []
                 for bheads in rbranchmap.itervalues():
                     rbheads.extend(bheads)
                     for h in bheads:
                         r = self.changelog.rev(h)
                         b, c = self.changelog.branchinfo(r)
                         if c:
                             closed.append(h)
 
                 if rbheads:
                     rtiprev = max((int(self.changelog.rev(node))
                             for node in rbheads))
                     cache = branchmap.branchcache(rbranchmap,
                                                   self[rtiprev].node(),
                                                   rtiprev,
                                                   closednodes=closed)
                     # Try to stick it as low as possible
                     # filter above served are unlikely to be fetch from a clone
                     for candidate in ('base', 'immutable', 'served'):
                         rview = self.filtered(candidate)
                         if cache.validfor(rview):
                             self._branchcaches[candidate] = cache
                             cache.write(rview)
                             break
             self.invalidate()
             return len(self.heads()) + 1
         finally:
             lock.release()
 
     def clone(self, remote, heads=[], stream=False):
         '''clone remote repository.
 
         keyword arguments:
         heads: list of revs to clone (forces use of pull)
         stream: use streaming clone if possible'''
 
         # now, all clients that can request uncompressed clones can
         # read repo formats supported by all servers that can serve
         # them.
 
         # if revlog format changes, client will have to check version
         # and format flags on "stream" capability, and use
         # uncompressed only if compatible.
 
         if not stream:
             # if the server explicitly prefers to stream (for fast LANs)
             stream = remote.capable('stream-preferred')
 
         if stream and not heads:
             # 'stream' means remote revlog format is revlogv1 only
             if remote.capable('stream'):
                 self.stream_in(remote, set(('revlogv1',)))
             else:
                 # otherwise, 'streamreqs' contains the remote revlog format
                 streamreqs = remote.capable('streamreqs')
                 if streamreqs:
                     streamreqs = set(streamreqs.split(','))
                     # if we support it, stream in and adjust our requirements
                     if not streamreqs - self.supportedformats:
                         self.stream_in(remote, streamreqs)
 
         quiet = self.ui.backupconfig('ui', 'quietbookmarkmove')
         try:
             self.ui.setconfig('ui', 'quietbookmarkmove', True, 'clone')
             ret = exchange.pull(self, remote, heads).cgresult
         finally:
             self.ui.restoreconfig(quiet)
         return ret
 
     def pushkey(self, namespace, key, old, new):
         self.hook('prepushkey', throw=True, namespace=namespace, key=key,
                   old=old, new=new)
         self.ui.debug('pushing key for "%s:%s"\n' % (namespace, key))
         ret = pushkey.push(self, namespace, key, old, new)
         self.hook('pushkey', namespace=namespace, key=key, old=old, new=new,
                   ret=ret)
         return ret
 
     def listkeys(self, namespace):
         self.hook('prelistkeys', throw=True, namespace=namespace)
         self.ui.debug('listing keys for "%s"\n' % namespace)
         values = pushkey.list(self, namespace)
         self.hook('listkeys', namespace=namespace, values=values)
         return values
 
     def debugwireargs(self, one, two, three=None, four=None, five=None):
         '''used to test argument passing over the wire'''
         return "%s %s %s %s %s" % (one, two, three, four, five)
 
     def savecommitmessage(self, text):
         fp = self.opener('last-message.txt', 'wb')
         try:
             fp.write(text)
         finally:
             fp.close()
         return self.pathto(fp.name[len(self.root) + 1:])
 
 # used to avoid circular references so destructors work
 def aftertrans(files):
     renamefiles = [tuple(t) for t in files]
     def a():
         for vfs, src, dest in renamefiles:
             try:
                 vfs.rename(src, dest)
             except OSError: # journal file does not yet exist
                 pass
     return a
 
 def undoname(fn):
     base, name = os.path.split(fn)
     assert name.startswith('journal')
     return os.path.join(base, name.replace('journal', 'undo', 1))
 
 def instance(ui, path, create):
     return localrepository(ui, util.urllocalpath(path), create)
 
 def islocal(path):
     return True