dirstate.walk: eliminate src from yield

- restructure 'step two' to stat unseen files
- remove extra stat logic in status
- remove redundant src

mercurial/context.py

@@ -1,774 +1,774
 # context.py - changeset and file context objects for mercurial
 #
 # Copyright 2006, 2007 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms
 # of the GNU General Public License, incorporated herein by reference.
 
 from node import nullid, nullrev, short, hex
 from i18n import _
 import ancestor, bdiff, revlog, util, os, errno
 
 class changectx(object):
     """A changecontext object makes access to data related to a particular
     changeset convenient."""
     def __init__(self, repo, changeid=''):
         """changeid is a revision number, node, or tag"""
         if changeid == '':
             changeid = '.'
         self._repo = repo
         self._node = self._repo.lookup(changeid)
         self._rev = self._repo.changelog.rev(self._node)
 
     def __str__(self):
         return short(self.node())
 
     def __int__(self):
         return self.rev()
 
     def __repr__(self):
         return "<changectx %s>" % str(self)
 
     def __hash__(self):
         try:
             return hash(self._rev)
         except AttributeError:
             return id(self)
 
     def __eq__(self, other):
         try:
             return self._rev == other._rev
         except AttributeError:
             return False
 
     def __ne__(self, other):
         return not (self == other)
 
     def __nonzero__(self):
         return self._rev != nullrev
 
     def __getattr__(self, name):
         if name == '_changeset':
             self._changeset = self._repo.changelog.read(self.node())
             return self._changeset
         elif name == '_manifest':
             self._manifest = self._repo.manifest.read(self._changeset[0])
             return self._manifest
         elif name == '_manifestdelta':
             md = self._repo.manifest.readdelta(self._changeset[0])
             self._manifestdelta = md
             return self._manifestdelta
         elif name == '_parents':
             p = self._repo.changelog.parents(self._node)
             if p[1] == nullid:
                 p = p[:-1]
             self._parents = [changectx(self._repo, x) for x in p]
             return self._parents
         else:
             raise AttributeError, name
 
     def __contains__(self, key):
         return key in self._manifest
 
     def __getitem__(self, key):
         return self.filectx(key)
 
     def __iter__(self):
         for f in util.sort(self._manifest):
             yield f
 
     def changeset(self): return self._changeset
     def manifest(self): return self._manifest
 
     def rev(self): return self._rev
     def node(self): return self._node
     def hex(self): return hex(self._node)
     def user(self): return self._changeset[1]
     def date(self): return self._changeset[2]
     def files(self): return self._changeset[3]
     def description(self): return self._changeset[4]
     def branch(self): return self._changeset[5].get("branch")
     def extra(self): return self._changeset[5]
     def tags(self): return self._repo.nodetags(self._node)
 
     def parents(self):
         """return contexts for each parent changeset"""
         return self._parents
 
     def children(self):
         """return contexts for each child changeset"""
         c = self._repo.changelog.children(self._node)
         return [changectx(self._repo, x) for x in c]
 
     def _fileinfo(self, path):
         if '_manifest' in self.__dict__:
             try:
                 return self._manifest[path], self._manifest.flags(path)
             except KeyError:
                 raise revlog.LookupError(self._node, path,
                                          _('not found in manifest'))
         if '_manifestdelta' in self.__dict__ or path in self.files():
             if path in self._manifestdelta:
                 return self._manifestdelta[path], self._manifestdelta.flags(path)
         node, flag = self._repo.manifest.find(self._changeset[0], path)
         if not node:
             raise revlog.LookupError(self._node, path,
                                      _('not found in manifest'))
 
         return node, flag
 
     def filenode(self, path):
         return self._fileinfo(path)[0]
 
     def flags(self, path):
         try:
             return self._fileinfo(path)[1]
         except revlog.LookupError:
             return ''
 
     def filectx(self, path, fileid=None, filelog=None):
         """get a file context from this changeset"""
         if fileid is None:
             fileid = self.filenode(path)
         return filectx(self._repo, path, fileid=fileid,
                        changectx=self, filelog=filelog)
 
     def filectxs(self):
         """generate a file context for each file in this changeset's
            manifest"""
         for f in util.sort(mf):
             yield self.filectx(f, fileid=mf[f])
 
     def ancestor(self, c2):
         """
         return the ancestor context of self and c2
         """
         n = self._repo.changelog.ancestor(self._node, c2._node)
         return changectx(self._repo, n)
 
     def walk(self, match):
         fdict = dict.fromkeys(match.files())
         # for dirstate.walk, files=['.'] means "walk the whole tree".
         # follow that here, too
         fdict.pop('.', None)
         for fn in self:
             for ffn in fdict:
                 # match if the file is the exact name or a directory
                 if ffn == fn or fn.startswith("%s/" % ffn):
                     del fdict[ffn]
                     break
             if match(fn):
                 yield fn
         for fn in util.sort(fdict):
             if match.bad(fn, 'No such file in rev ' + str(self)) and match(fn):
                 yield fn
 
 class filectx(object):
     """A filecontext object makes access to data related to a particular
        filerevision convenient."""
     def __init__(self, repo, path, changeid=None, fileid=None,
                  filelog=None, changectx=None):
         """changeid can be a changeset revision, node, or tag.
            fileid can be a file revision or node."""
         self._repo = repo
         self._path = path
 
         assert (changeid is not None
                 or fileid is not None
                 or changectx is not None)
 
         if filelog:
             self._filelog = filelog
 
         if changeid is not None:
             self._changeid = changeid
         if changectx is not None:
             self._changectx = changectx
         if fileid is not None:
             self._fileid = fileid
 
     def __getattr__(self, name):
         if name == '_changectx':
             self._changectx = changectx(self._repo, self._changeid)
             return self._changectx
         elif name == '_filelog':
             self._filelog = self._repo.file(self._path)
             return self._filelog
         elif name == '_changeid':
             if '_changectx' in self.__dict__:
                 self._changeid = self._changectx.rev()
             else:
                 self._changeid = self._filelog.linkrev(self._filenode)
             return self._changeid
         elif name == '_filenode':
             if '_fileid' in self.__dict__:
                 self._filenode = self._filelog.lookup(self._fileid)
             else:
                 self._filenode = self._changectx.filenode(self._path)
             return self._filenode
         elif name == '_filerev':
             self._filerev = self._filelog.rev(self._filenode)
             return self._filerev
         elif name == '_repopath':
             self._repopath = self._path
             return self._repopath
         else:
             raise AttributeError, name
 
     def __nonzero__(self):
         try:
             n = self._filenode
             return True
         except revlog.LookupError:
             # file is missing
             return False
 
     def __str__(self):
         return "%s@%s" % (self.path(), short(self.node()))
 
     def __repr__(self):
         return "<filectx %s>" % str(self)
 
     def __hash__(self):
         try:
             return hash((self._path, self._fileid))
         except AttributeError:
             return id(self)
 
     def __eq__(self, other):
         try:
             return (self._path == other._path
                     and self._fileid == other._fileid)
         except AttributeError:
             return False
 
     def __ne__(self, other):
         return not (self == other)
 
     def filectx(self, fileid):
         '''opens an arbitrary revision of the file without
         opening a new filelog'''
         return filectx(self._repo, self._path, fileid=fileid,
                        filelog=self._filelog)
 
     def filerev(self): return self._filerev
     def filenode(self): return self._filenode
     def flags(self): return self._changectx.flags(self._path)
     def filelog(self): return self._filelog
 
     def rev(self):
         if '_changectx' in self.__dict__:
             return self._changectx.rev()
         if '_changeid' in self.__dict__:
             return self._changectx.rev()
         return self._filelog.linkrev(self._filenode)
 
     def linkrev(self): return self._filelog.linkrev(self._filenode)
     def node(self): return self._changectx.node()
     def user(self): return self._changectx.user()
     def date(self): return self._changectx.date()
     def files(self): return self._changectx.files()
     def description(self): return self._changectx.description()
     def branch(self): return self._changectx.branch()
     def manifest(self): return self._changectx.manifest()
     def changectx(self): return self._changectx
 
     def data(self): return self._filelog.read(self._filenode)
     def path(self): return self._path
     def size(self): return self._filelog.size(self._filerev)
 
     def cmp(self, text): return self._filelog.cmp(self._filenode, text)
 
     def renamed(self):
         """check if file was actually renamed in this changeset revision
 
         If rename logged in file revision, we report copy for changeset only
         if file revisions linkrev points back to the changeset in question
         or both changeset parents contain different file revisions.
         """
 
         renamed = self._filelog.renamed(self._filenode)
         if not renamed:
             return renamed
 
         if self.rev() == self.linkrev():
             return renamed
 
         name = self.path()
         fnode = self._filenode
         for p in self._changectx.parents():
             try:
                 if fnode == p.filenode(name):
                     return None
             except revlog.LookupError:
                 pass
         return renamed
 
     def parents(self):
         p = self._path
         fl = self._filelog
         pl = [(p, n, fl) for n in self._filelog.parents(self._filenode)]
 
         r = self._filelog.renamed(self._filenode)
         if r:
             pl[0] = (r[0], r[1], None)
 
         return [filectx(self._repo, p, fileid=n, filelog=l)
                 for p,n,l in pl if n != nullid]
 
     def children(self):
         # hard for renames
         c = self._filelog.children(self._filenode)
         return [filectx(self._repo, self._path, fileid=x,
                         filelog=self._filelog) for x in c]
 
     def annotate(self, follow=False, linenumber=None):
         '''returns a list of tuples of (ctx, line) for each line
         in the file, where ctx is the filectx of the node where
         that line was last changed.
         This returns tuples of ((ctx, linenumber), line) for each line,
         if "linenumber" parameter is NOT "None".
         In such tuples, linenumber means one at the first appearance
         in the managed file.
         To reduce annotation cost,
         this returns fixed value(False is used) as linenumber,
         if "linenumber" parameter is "False".'''
 
         def decorate_compat(text, rev):
             return ([rev] * len(text.splitlines()), text)
 
         def without_linenumber(text, rev):
             return ([(rev, False)] * len(text.splitlines()), text)
 
         def with_linenumber(text, rev):
             size = len(text.splitlines())
             return ([(rev, i) for i in xrange(1, size + 1)], text)
 
         decorate = (((linenumber is None) and decorate_compat) or
                     (linenumber and with_linenumber) or
                     without_linenumber)
 
         def pair(parent, child):
             for a1, a2, b1, b2 in bdiff.blocks(parent[1], child[1]):
                 child[0][b1:b2] = parent[0][a1:a2]
             return child
 
         getlog = util.cachefunc(lambda x: self._repo.file(x))
         def getctx(path, fileid):
             log = path == self._path and self._filelog or getlog(path)
             return filectx(self._repo, path, fileid=fileid, filelog=log)
         getctx = util.cachefunc(getctx)
 
         def parents(f):
             # we want to reuse filectx objects as much as possible
             p = f._path
             if f._filerev is None: # working dir
                 pl = [(n.path(), n.filerev()) for n in f.parents()]
             else:
                 pl = [(p, n) for n in f._filelog.parentrevs(f._filerev)]
 
             if follow:
                 r = f.renamed()
                 if r:
                     pl[0] = (r[0], getlog(r[0]).rev(r[1]))
 
             return [getctx(p, n) for p, n in pl if n != nullrev]
 
         # use linkrev to find the first changeset where self appeared
         if self.rev() != self.linkrev():
             base = self.filectx(self.filerev())
         else:
             base = self
 
         # find all ancestors
         needed = {base: 1}
         visit = [base]
         files = [base._path]
         while visit:
             f = visit.pop(0)
             for p in parents(f):
                 if p not in needed:
                     needed[p] = 1
                     visit.append(p)
                     if p._path not in files:
                         files.append(p._path)
                 else:
                     # count how many times we'll use this
                     needed[p] += 1
 
         # sort by revision (per file) which is a topological order
         visit = []
         for f in files:
             fn = [(n.rev(), n) for n in needed if n._path == f]
             visit.extend(fn)
 
         hist = {}
         for r, f in util.sort(visit):
             curr = decorate(f.data(), f)
             for p in parents(f):
                 if p != nullid:
                     curr = pair(hist[p], curr)
                     # trim the history of unneeded revs
                     needed[p] -= 1
                     if not needed[p]:
                         del hist[p]
             hist[f] = curr
 
         return zip(hist[f][0], hist[f][1].splitlines(1))
 
     def ancestor(self, fc2):
         """
         find the common ancestor file context, if any, of self, and fc2
         """
 
         acache = {}
 
         # prime the ancestor cache for the working directory
         for c in (self, fc2):
             if c._filerev == None:
                 pl = [(n.path(), n.filenode()) for n in c.parents()]
                 acache[(c._path, None)] = pl
 
         flcache = {self._repopath:self._filelog, fc2._repopath:fc2._filelog}
         def parents(vertex):
             if vertex in acache:
                 return acache[vertex]
             f, n = vertex
             if f not in flcache:
                 flcache[f] = self._repo.file(f)
             fl = flcache[f]
             pl = [(f, p) for p in fl.parents(n) if p != nullid]
             re = fl.renamed(n)
             if re:
                 pl.append(re)
             acache[vertex] = pl
             return pl
 
         a, b = (self._path, self._filenode), (fc2._path, fc2._filenode)
         v = ancestor.ancestor(a, b, parents)
         if v:
             f, n = v
             return filectx(self._repo, f, fileid=n, filelog=flcache[f])
 
         return None
 
 class workingctx(changectx):
     """A workingctx object makes access to data related to
     the current working directory convenient.
     parents - a pair of parent nodeids, or None to use the dirstate.
     date - any valid date string or (unixtime, offset), or None.
     user - username string, or None.
     extra - a dictionary of extra values, or None.
     changes - a list of file lists as returned by localrepo.status()
                or None to use the repository status.
     """
     def __init__(self, repo, parents=None, text="", user=None, date=None,
                  extra=None, changes=None):
         self._repo = repo
         self._rev = None
         self._node = None
         self._text = text
         if date:
             self._date = util.parsedate(date)
         if user:
             self._user = user
         if parents:
             self._parents = [changectx(self._repo, p) for p in parents]
         if changes:
             self._status = list(changes)
 
         self._extra = {}
         if extra:
             self._extra = extra.copy()
         if 'branch' not in self._extra:
             branch = self._repo.dirstate.branch()
             try:
                 branch = branch.decode('UTF-8').encode('UTF-8')
             except UnicodeDecodeError:
                 raise util.Abort(_('branch name not in UTF-8!'))
             self._extra['branch'] = branch
         if self._extra['branch'] == '':
             self._extra['branch'] = 'default'
 
     def __str__(self):
         return str(self._parents[0]) + "+"
 
     def __nonzero__(self):
         return True
 
     def __contains__(self, key):
         return self._dirstate[f] not in "?r"
 
     def __getattr__(self, name):
         if name == '_status':
             self._status = self._repo.status(unknown=True)
             return self._status
         elif name == '_user':
             self._user = self._repo.ui.username()
             return self._user
         elif name == '_date':
             self._date = util.makedate()
             return self._date
         if name == '_manifest':
             self._buildmanifest()
             return self._manifest
         elif name == '_parents':
             p = self._repo.dirstate.parents()
             if p[1] == nullid:
                 p = p[:-1]
             self._parents = [changectx(self._repo, x) for x in p]
             return self._parents
         else:
             raise AttributeError, name
 
     def _buildmanifest(self):
         """generate a manifest corresponding to the working directory"""
 
         man = self._parents[0].manifest().copy()
         copied = self._repo.dirstate.copies()
         cf = lambda x: man.flags(copied.get(x, x))
         ff = self._repo.dirstate.flagfunc(cf)
         modified, added, removed, deleted, unknown = self._status[:5]
         for i, l in (("a", added), ("m", modified), ("u", unknown)):
             for f in l:
                 man[f] = man.get(copied.get(f, f), nullid) + i
                 try:
                     man.set(f, ff(f))
                 except OSError:
                     pass
 
         for f in deleted + removed:
             if f in man:
                 del man[f]
 
         self._manifest = man
 
     def manifest(self): return self._manifest
 
     def user(self): return self._user or self._repo.ui.username()
     def date(self): return self._date
     def description(self): return self._text
     def files(self):
         return util.sort(self._status[0] + self._status[1] + self._status[2])
 
     def modified(self): return self._status[0]
     def added(self): return self._status[1]
     def removed(self): return self._status[2]
     def deleted(self): return self._status[3]
     def unknown(self): return self._status[4]
     def clean(self): return self._status[5]
     def branch(self): return self._extra['branch']
     def extra(self): return self._extra
 
     def tags(self):
         t = []
         [t.extend(p.tags()) for p in self.parents()]
         return t
 
     def children(self):
         return []
 
     def flags(self, path):
         if '_manifest' in self.__dict__:
             try:
                 return self._manifest.flags(path)
             except KeyError:
                 return ''
 
         pnode = self._parents[0].changeset()[0]
         orig = self._repo.dirstate.copies().get(path, path)
         node, flag = self._repo.manifest.find(pnode, orig)
         try:
             ff = self._repo.dirstate.flagfunc(lambda x: flag or '')
             return ff(path)
         except OSError:
             pass
 
         if not node or path in self.deleted() or path in self.removed():
             return ''
         return flag
 
     def filectx(self, path, filelog=None):
         """get a file context from the working directory"""
         return workingfilectx(self._repo, path, workingctx=self,
                               filelog=filelog)
 
     def ancestor(self, c2):
         """return the ancestor context of self and c2"""
         return self._parents[0].ancestor(c2) # punt on two parents for now
 
     def walk(self, match):
-        for src, fn, st in self._repo.dirstate.walk(match, True, False):
+        for fn, st in self._repo.dirstate.walk(match, True, False):
             yield fn
 
 class workingfilectx(filectx):
     """A workingfilectx object makes access to data related to a particular
        file in the working directory convenient."""
     def __init__(self, repo, path, filelog=None, workingctx=None):
         """changeid can be a changeset revision, node, or tag.
            fileid can be a file revision or node."""
         self._repo = repo
         self._path = path
         self._changeid = None
         self._filerev = self._filenode = None
 
         if filelog:
             self._filelog = filelog
         if workingctx:
             self._changectx = workingctx
 
     def __getattr__(self, name):
         if name == '_changectx':
             self._changectx = workingctx(self._repo)
             return self._changectx
         elif name == '_repopath':
             self._repopath = (self._repo.dirstate.copied(self._path)
                               or self._path)
             return self._repopath
         elif name == '_filelog':
             self._filelog = self._repo.file(self._repopath)
             return self._filelog
         else:
             raise AttributeError, name
 
     def __nonzero__(self):
         return True
 
     def __str__(self):
         return "%s@%s" % (self.path(), self._changectx)
 
     def filectx(self, fileid):
         '''opens an arbitrary revision of the file without
         opening a new filelog'''
         return filectx(self._repo, self._repopath, fileid=fileid,
                        filelog=self._filelog)
 
     def rev(self):
         if '_changectx' in self.__dict__:
             return self._changectx.rev()
         return self._filelog.linkrev(self._filenode)
 
     def data(self): return self._repo.wread(self._path)
     def renamed(self):
         rp = self._repopath
         if rp == self._path:
             return None
         return rp, self._changectx._parents[0]._manifest.get(rp, nullid)
 
     def parents(self):
         '''return parent filectxs, following copies if necessary'''
         p = self._path
         rp = self._repopath
         pcl = self._changectx._parents
         fl = self._filelog
         pl = [(rp, pcl[0]._manifest.get(rp, nullid), fl)]
         if len(pcl) > 1:
             if rp != p:
                 fl = None
             pl.append((p, pcl[1]._manifest.get(p, nullid), fl))
 
         return [filectx(self._repo, p, fileid=n, filelog=l)
                 for p,n,l in pl if n != nullid]
 
     def children(self):
         return []
 
     def size(self): return os.stat(self._repo.wjoin(self._path)).st_size
     def date(self):
         t, tz = self._changectx.date()
         try:
             return (int(os.lstat(self._repo.wjoin(self._path)).st_mtime), tz)
         except OSError, err:
             if err.errno != errno.ENOENT: raise
             return (t, tz)
 
     def cmp(self, text): return self._repo.wread(self._path) == text
 
 class memctx(object):
     """A memctx is a subset of changectx supposed to be built on memory
     and passed to commit functions.
 
     NOTE: this interface and the related memfilectx are experimental and
     may change without notice.
 
     parents - a pair of parent nodeids.
     filectxfn - a callable taking (repo, memctx, path) arguments and
     returning a memctx object.
     date - any valid date string or (unixtime, offset), or None.
     user - username string, or None.
     extra - a dictionary of extra values, or None.
     """
     def __init__(self, repo, parents, text, files, filectxfn, user=None,
                  date=None, extra=None):
         self._repo = repo
         self._rev = None
         self._node = None
         self._text = text
         self._date = date and util.parsedate(date) or util.makedate()
         self._user = user
         parents = [(p or nullid) for p in parents]
         p1, p2 = parents
         self._parents = [changectx(self._repo, p) for p in (p1, p2)]
         files = util.sort(list(files))
         self._status = [files, [], [], [], []]
         self._filectxfn = filectxfn
 
         self._extra = extra and extra.copy() or {}
         if 'branch' not in self._extra:
             self._extra['branch'] = 'default'
         elif self._extra.get('branch') == '':
             self._extra['branch'] = 'default'
 
     def __str__(self):
         return str(self._parents[0]) + "+"
 
     def __int__(self):
         return self._rev
 
     def __nonzero__(self):
         return True
 
     def user(self): return self._user or self._repo.ui.username()
     def date(self): return self._date
     def description(self): return self._text
     def files(self): return self.modified()
     def modified(self): return self._status[0]
     def added(self): return self._status[1]
     def removed(self): return self._status[2]
     def deleted(self): return self._status[3]
     def unknown(self): return self._status[4]
     def clean(self): return self._status[5]
     def branch(self): return self._extra['branch']
     def extra(self): return self._extra
     def flags(self, f): return self[f].flags()
 
     def parents(self):
         """return contexts for each parent changeset"""
         return self._parents
 
     def filectx(self, path, filelog=None):
         """get a file context from the working directory"""
         return self._filectxfn(self._repo, self, path)
 
 class memfilectx(object):
     """A memfilectx is a subset of filectx supposed to be built by client
     code and passed to commit functions.
     """
     def __init__(self, path, data, islink, isexec, copied):
         """copied is the source file path, or None."""
         self._path = path
         self._data = data
         self._flags = (islink and 'l' or '') + (isexec and 'x' or '')
         self._copied = None
         if copied:
             self._copied = (copied, nullid)
 
     def __nonzero__(self): return True
     def __str__(self): return "%s@%s" % (self.path(), self._changectx)
     def path(self): return self._path
     def data(self): return self._data
     def flags(self): return self._flags
     def isexec(self): return 'x' in self._flags
     def islink(self): return 'l' in self._flags
     def renamed(self): return self._copied
 

mercurial/dirstate.py

@@ -1,670 +1,658
 """
 dirstate.py - working directory tracking 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, incorporated herein by reference.
 """
 
 from node import nullid
 from i18n import _
 import struct, os, bisect, stat, util, errno, ignore
 import cStringIO, osutil, sys
 
 _unknown = ('?', 0, 0, 0)
 _format = ">cllll"
 
 def _finddirs(path):
     pos = len(path)
     while 1:
         pos = path.rfind('/', 0, pos)
         if pos == -1:
             break
         yield path[:pos]
 
 class dirstate(object):
 
     def __init__(self, opener, ui, root):
         self._opener = opener
         self._root = root
         self._dirty = False
         self._dirtypl = False
         self._ui = ui
 
     def __getattr__(self, name):
         if name == '_map':
             self._read()
             return self._map
         elif name == '_copymap':
             self._read()
             return self._copymap
         elif name == '_foldmap':
             _foldmap = {}
             for name in self._map:
                 norm = os.path.normcase(os.path.normpath(name))
                 _foldmap[norm] = name
             self._foldmap = _foldmap
             return self._foldmap
         elif name == '_branch':
             try:
                 self._branch = (self._opener("branch").read().strip()
                                 or "default")
             except IOError:
                 self._branch = "default"
             return self._branch
         elif name == '_pl':
             self._pl = [nullid, nullid]
             try:
                 st = self._opener("dirstate").read(40)
                 if len(st) == 40:
                     self._pl = st[:20], st[20:40]
             except IOError, err:
                 if err.errno != errno.ENOENT: raise
             return self._pl
         elif name == '_dirs':
             dirs = {}
             for f,s in self._map.items():
                 if s[0] != 'r':
                     for base in _finddirs(f):
                         dirs[base] = dirs.get(base, 0) + 1
             self._dirs = dirs
             return self._dirs
         elif name == '_ignore':
             files = [self._join('.hgignore')]
             for name, path in self._ui.configitems("ui"):
                 if name == 'ignore' or name.startswith('ignore.'):
                     files.append(os.path.expanduser(path))
             self._ignore = ignore.ignore(self._root, files, self._ui.warn)
             return self._ignore
         elif name == '_slash':
             self._slash = self._ui.configbool('ui', 'slash') and os.sep != '/'
             return self._slash
         elif name == '_checklink':
             self._checklink = util.checklink(self._root)
             return self._checklink
         elif name == '_checkexec':
             self._checkexec = util.checkexec(self._root)
             return self._checkexec
         elif name == '_checkcase':
             self._checkcase = not util.checkcase(self._join('.hg'))
             return self._checkcase
         elif name == 'normalize':
             if self._checkcase:
                 self.normalize = self._normalize
             else:
                 self.normalize = lambda x: x
             return self.normalize
         else:
             raise AttributeError, name
 
     def _join(self, f):
         return os.path.join(self._root, f)
 
     def flagfunc(self, fallback):
         if self._checklink:
             if self._checkexec:
                 def f(x):
                     p = os.path.join(self._root, x)
                     if os.path.islink(p):
                         return 'l'
                     if util.is_exec(p):
                         return 'x'
                     return ''
                 return f
             def f(x):
                 if os.path.islink(os.path.join(self._root, x)):
                     return 'l'
                 if 'x' in fallback(x):
                     return 'x'
                 return ''
             return f
         if self._checkexec:
             def f(x):
                 if 'l' in fallback(x):
                     return 'l'
                 if util.is_exec(os.path.join(self._root, x)):
                     return 'x'
                 return ''
             return f
         return fallback
 
     def getcwd(self):
         cwd = os.getcwd()
         if cwd == self._root: return ''
         # self._root ends with a path separator if self._root is '/' or 'C:\'
         rootsep = self._root
         if not util.endswithsep(rootsep):
             rootsep += os.sep
         if cwd.startswith(rootsep):
             return cwd[len(rootsep):]
         else:
             # we're outside the repo. return an absolute path.
             return cwd
 
     def pathto(self, f, cwd=None):
         if cwd is None:
             cwd = self.getcwd()
         path = util.pathto(self._root, cwd, f)
         if self._slash:
             return util.normpath(path)
         return path
 
     def __getitem__(self, key):
         ''' current states:
         n  normal
         m  needs merging
         r  marked for removal
         a  marked for addition
         ?  not tracked'''
         return self._map.get(key, ("?",))[0]
 
     def __contains__(self, key):
         return key in self._map
 
     def __iter__(self):
         for x in util.sort(self._map):
             yield x
 
     def parents(self):
         return self._pl
 
     def branch(self):
         return self._branch
 
     def setparents(self, p1, p2=nullid):
         self._dirty = self._dirtypl = True
         self._pl = p1, p2
 
     def setbranch(self, branch):
         self._branch = branch
         self._opener("branch", "w").write(branch + '\n')
 
     def _read(self):
         self._map = {}
         self._copymap = {}
         if not self._dirtypl:
             self._pl = [nullid, nullid]
         try:
             st = self._opener("dirstate").read()
         except IOError, err:
             if err.errno != errno.ENOENT: raise
             return
         if not st:
             return
 
         if not self._dirtypl:
             self._pl = [st[:20], st[20: 40]]
 
         # deref fields so they will be local in loop
         dmap = self._map
         copymap = self._copymap
         unpack = struct.unpack
         e_size = struct.calcsize(_format)
         pos1 = 40
         l = len(st)
 
         # the inner loop
         while pos1 < l:
             pos2 = pos1 + e_size
             e = unpack(">cllll", st[pos1:pos2]) # a literal here is faster
             pos1 = pos2 + e[4]
             f = st[pos2:pos1]
             if '\0' in f:
                 f, c = f.split('\0')
                 copymap[f] = c
             dmap[f] = e # we hold onto e[4] because making a subtuple is slow
 
     def invalidate(self):
         for a in "_map _copymap _foldmap _branch _pl _dirs _ignore".split():
             if a in self.__dict__:
                 delattr(self, a)
         self._dirty = False
 
     def copy(self, source, dest):
         if source == dest:
             return
         self._dirty = True
         self._copymap[dest] = source
 
     def copied(self, file):
         return self._copymap.get(file, None)
 
     def copies(self):
         return self._copymap
 
     def _droppath(self, f):
         if self[f] not in "?r" and "_dirs" in self.__dict__:
             dirs = self._dirs
             for base in _finddirs(f):
                 if dirs[base] == 1:
                     del dirs[base]
                 else:
                     dirs[base] -= 1
 
     def _addpath(self, f, check=False):
         oldstate = self[f]
         if check or oldstate == "r":
             if '\r' in f or '\n' in f:
                 raise util.Abort(
                     _("'\\n' and '\\r' disallowed in filenames: %r") % f)
             if f in self._dirs:
                 raise util.Abort(_('directory %r already in dirstate') % f)
             # shadows
             for d in _finddirs(f):
                 if d in self._dirs:
                     break
                 if d in self._map and self[d] != 'r':
                     raise util.Abort(
                         _('file %r in dirstate clashes with %r') % (d, f))
         if oldstate in "?r" and "_dirs" in self.__dict__:
             dirs = self._dirs
             for base in _finddirs(f):
                 dirs[base] = dirs.get(base, 0) + 1
 
     def normal(self, f):
         'mark a file normal and clean'
         self._dirty = True
         self._addpath(f)
         s = os.lstat(self._join(f))
         self._map[f] = ('n', s.st_mode, s.st_size, s.st_mtime, 0)
         if f in self._copymap:
             del self._copymap[f]
 
     def normallookup(self, f):
         'mark a file normal, but possibly dirty'
         if self._pl[1] != nullid and f in self._map:
             # if there is a merge going on and the file was either
             # in state 'm' or dirty before being removed, restore that state.
             entry = self._map[f]
             if entry[0] == 'r' and entry[2] in (-1, -2):
                 source = self._copymap.get(f)
                 if entry[2] == -1:
                     self.merge(f)
                 elif entry[2] == -2:
                     self.normaldirty(f)
                 if source:
                     self.copy(source, f)
                 return
             if entry[0] == 'm' or entry[0] == 'n' and entry[2] == -2:
                 return
         self._dirty = True
         self._addpath(f)
         self._map[f] = ('n', 0, -1, -1, 0)
         if f in self._copymap:
             del self._copymap[f]
 
     def normaldirty(self, f):
         'mark a file normal, but dirty'
         self._dirty = True
         self._addpath(f)
         self._map[f] = ('n', 0, -2, -1, 0)
         if f in self._copymap:
             del self._copymap[f]
 
     def add(self, f):
         'mark a file added'
         self._dirty = True
         self._addpath(f, True)
         self._map[f] = ('a', 0, -1, -1, 0)
         if f in self._copymap:
             del self._copymap[f]
 
     def remove(self, f):
         'mark a file removed'
         self._dirty = True
         self._droppath(f)
         size = 0
         if self._pl[1] != nullid and f in self._map:
             entry = self._map[f]
             if entry[0] == 'm':
                 size = -1
             elif entry[0] == 'n' and entry[2] == -2:
                 size = -2
         self._map[f] = ('r', 0, size, 0, 0)
         if size == 0 and f in self._copymap:
             del self._copymap[f]
 
     def merge(self, f):
         'mark a file merged'
         self._dirty = True
         s = os.lstat(self._join(f))
         self._addpath(f)
         self._map[f] = ('m', s.st_mode, s.st_size, s.st_mtime, 0)
         if f in self._copymap:
             del self._copymap[f]
 
     def forget(self, f):
         'forget a file'
         self._dirty = True
         try:
             self._droppath(f)
             del self._map[f]
         except KeyError:
             self._ui.warn(_("not in dirstate: %s\n") % f)
 
     def _normalize(self, path):
         normpath = os.path.normcase(os.path.normpath(path))
         if normpath in self._foldmap:
             return self._foldmap[normpath]
         elif os.path.exists(path):
             self._foldmap[normpath] = util.fspath(path, self._root)
             return self._foldmap[normpath]
         else:
             return path
 
     def clear(self):
         self._map = {}
         if "_dirs" in self.__dict__:
             delattr(self, "_dirs");
         self._copymap = {}
         self._pl = [nullid, nullid]
         self._dirty = True
 
     def rebuild(self, parent, files):
         self.clear()
         for f in files:
             if 'x' in files.flags(f):
                 self._map[f] = ('n', 0777, -1, 0, 0)
             else:
                 self._map[f] = ('n', 0666, -1, 0, 0)
         self._pl = (parent, nullid)
         self._dirty = True
 
     def write(self):
         if not self._dirty:
             return
         st = self._opener("dirstate", "w", atomictemp=True)
 
         try:
             gran = int(self._ui.config('dirstate', 'granularity', 1))
         except ValueError:
             gran = 1
         limit = sys.maxint
         if gran > 0:
             limit = util.fstat(st).st_mtime - gran
 
         cs = cStringIO.StringIO()
         copymap = self._copymap
         pack = struct.pack
         write = cs.write
         write("".join(self._pl))
         for f, e in self._map.iteritems():
             if f in copymap:
                 f = "%s\0%s" % (f, copymap[f])
             if e[3] > limit and e[0] == 'n':
                 e = (e[0], 0, -1, -1, 0)
             e = pack(_format, e[0], e[1], e[2], e[3], len(f))
             write(e)
             write(f)
         st.write(cs.getvalue())
         st.rename()
         self._dirty = self._dirtypl = False
 
     def _filter(self, files):
         ret = {}
         unknown = []
 
         for x in files:
             if x == '.':
                 return self._map.copy()
             if x not in self._map:
                 unknown.append(x)
             else:
                 ret[x] = self._map[x]
 
         if not unknown:
             return ret
 
         b = util.sort(self._map)
         blen = len(b)
 
         for x in unknown:
             bs = bisect.bisect(b, "%s%s" % (x, '/'))
             while bs < blen:
                 s = b[bs]
                 if len(s) > len(x) and s.startswith(x):
                     ret[s] = self._map[s]
                 else:
                     break
                 bs += 1
         return ret
 
     def _supported(self, f, mode, verbose=False):
         if stat.S_ISREG(mode) or stat.S_ISLNK(mode):
             return True
         if verbose:
             kind = 'unknown'
             if stat.S_ISCHR(mode): kind = _('character device')
             elif stat.S_ISBLK(mode): kind = _('block device')
             elif stat.S_ISFIFO(mode): kind = _('fifo')
             elif stat.S_ISSOCK(mode): kind = _('socket')
             elif stat.S_ISDIR(mode): kind = _('directory')
             self._ui.warn(_('%s: unsupported file type (type is %s)\n')
                           % (self.pathto(f), kind))
         return False
 
     def _dirignore(self, f):
         if f == '.':
             return False
         if self._ignore(f):
             return True
         for p in _finddirs(f):
             if self._ignore(p):
                 return True
         return False
 
     def walk(self, match, unknown, ignored):
         '''
         walk recursively through the directory tree, finding all files
         matched by the match function
 
-        results are yielded in a tuple (src, filename, st), where src
-        is one of:
-        'f' the file was found in the directory tree
-        'm' the file was only in the dirstate and not in the tree
-
+        results are yielded in a tuple (filename, stat), where stat
         and st is the stat result if the file was found in the directory.
         '''
 
         def fwarn(f, msg):
             self._ui.warn('%s: %s\n' % (self.pathto(ff), msg))
             return False
         badfn = fwarn
         if hasattr(match, 'bad'):
             badfn = match.bad
 
         # walk all files by default
         files = match.files()
         if not files:
             files = ['.']
             dc = self._map.copy()
         else:
             files = util.unique(files)
             dc = self._filter(files)
 
         def imatch(file_):
             if file_ not in dc and self._ignore(file_):
                 return False
             return match(file_)
 
         # TODO: don't walk unknown directories if unknown and ignored are False
         ignore = self._ignore
         dirignore = self._dirignore
         if ignored:
             imatch = match
             ignore = util.never
             dirignore = util.never
 
         # self._root may end with a path separator when self._root == '/'
         common_prefix_len = len(self._root)
         if not util.endswithsep(self._root):
             common_prefix_len += 1
 
         normpath = util.normpath
         listdir = osutil.listdir
         lstat = os.lstat
         bisect_left = bisect.bisect_left
         isdir = os.path.isdir
         pconvert = util.pconvert
         join = os.path.join
         s_isdir = stat.S_ISDIR
         supported = self._supported
         _join = self._join
         known = {'.hg': 1}
 
         # recursion free walker, faster than os.walk.
         def findfiles(s):
             work = [s]
             wadd = work.append
             found = []
             add = found.append
             if hasattr(match, 'dir'):
                 match.dir(normpath(s[common_prefix_len:]))
             while work:
                 top = work.pop()
                 entries = listdir(top, stat=True)
                 # nd is the top of the repository dir tree
                 nd = normpath(top[common_prefix_len:])
                 if nd == '.':
                     nd = ''
                 else:
                     # do not recurse into a repo contained in this
                     # one. use bisect to find .hg directory so speed
                     # is good on big directory.
                     names = [e[0] for e in entries]
                     hg = bisect_left(names, '.hg')
                     if hg < len(names) and names[hg] == '.hg':
                         if isdir(join(top, '.hg')):
                             continue
                 for f, kind, st in entries:
                     np = pconvert(join(nd, f))
                     nn = self.normalize(np)
                     if np in known:
                         continue
                     known[nn] = 1
                     p = join(top, f)
                     # don't trip over symlinks
                     if kind == stat.S_IFDIR:
                         if not ignore(np):
                             wadd(p)
                             if hasattr(match, 'dir'):
                                 match.dir(np)
                         if np in dc and match(np):
-                            add((nn, 'm', st))
+                            add((nn, None))
                     elif imatch(np):
                         if supported(np, st.st_mode):
-                            add((nn, 'f', st))
+                            add((nn, st))
                         elif np in dc:
-                            add((nn, 'm', st))
+                            add((nn, None))
             return util.sort(found)
 
         # step one, find all files that match our criteria
         for ff in util.sort(files):
             nf = normpath(ff)
             nn = self.normalize(nf)
             f = _join(ff)
             try:
                 st = lstat(f)
             except OSError, inst:
                 found = False
                 for fn in dc:
                     if nf == fn or (fn.startswith(nf) and fn[len(nf)] == '/'):
                         found = True
                         break
                 if not found:
                     if inst.errno != errno.ENOENT:
                         fwarn(ff, inst.strerror)
                     elif badfn(ff, inst.strerror) and imatch(nf):
-                        yield 'f', ff, None
+                        yield ff, None
                 continue
             if s_isdir(st.st_mode):
                 if not dirignore(nf):
-                    for f, src, st in findfiles(f):
-                        yield src, f, st
+                    for e in findfiles(f):
+                        yield e
             else:
                 if nn in known:
                     continue
                 known[nn] = 1
                 if match(nf):
                     if supported(ff, st.st_mode, verbose=True):
-                        yield 'f', nn, st
+                        yield nn, st
                     elif ff in dc:
-                        yield 'm', nf, st
+                        yield nf, None
 
         # step two run through anything left in the dc hash and yield
         # if we haven't already seen it
-        for k in util.sort(dc):
-            if k in known:
+        for f in util.sort(dc):
+            if f in known:
                 continue
-            known[k] = 1
-            if imatch(k):
-                yield 'm', k, None
+            known[f] = 1
+            if imatch(f):
+                try:
+                    st = lstat(_join(f))
+                    if supported(f, st.st_mode):
+                        yield f, st
+                        continue
+                except OSError, inst:
+                    if inst.errno not in (errno.ENOENT, errno.ENOTDIR):
+                        raise
+                yield f, None
 
     def status(self, match, ignored, clean, unknown):
         listignored, listclean, listunknown = ignored, clean, unknown
         lookup, modified, added, unknown, ignored = [], [], [], [], []
         removed, deleted, clean = [], [], []
 
         _join = self._join
         lstat = os.lstat
         cmap = self._copymap
         dmap = self._map
         ladd = lookup.append
         madd = modified.append
         aadd = added.append
         uadd = unknown.append
         iadd = ignored.append
         radd = removed.append
         dadd = deleted.append
         cadd = clean.append
 
-        for src, fn, st in self.walk(match, listunknown, listignored):
+        for fn, st in self.walk(match, listunknown, listignored):
             if fn not in dmap:
                 if (listignored or match.exact(fn)) and self._dirignore(fn):
                     if listignored:
                         iadd(fn)
                 elif listunknown:
                     uadd(fn)
                 continue
 
             state, mode, size, time, foo = dmap[fn]
 
-            if src == 'm':
-                nonexistent = True
-                if not st:
-                    try:
-                        st = lstat(_join(fn))
-                    except OSError, inst:
-                        if inst.errno not in (errno.ENOENT, errno.ENOTDIR):
-                            raise
-                        st = None
-                    # We need to re-check that it is a valid file
-                    if st and self._supported(fn, st.st_mode):
-                        nonexistent = False
-                if nonexistent and state in "nma":
-                    dadd(fn)
-                    continue
-            # check the common case first
-            if state == 'n':
-                if not st:
-                    st = lstat(_join(fn))
+            if not st and state in "nma":
+                dadd(fn)
+            elif state == 'n':
                 if (size >= 0 and
                     (size != st.st_size
                      or ((mode ^ st.st_mode) & 0100 and self._checkexec))
                     or size == -2
                     or fn in self._copymap):
                     madd(fn)
                 elif time != int(st.st_mtime):
                     ladd(fn)
                 elif listclean:
                     cadd(fn)
             elif state == 'm':
                 madd(fn)
             elif state == 'a':
                 aadd(fn)
             elif state == 'r':
                 radd(fn)
 
         return (lookup, modified, added, removed, deleted, unknown, ignored,
                 clean)