windows: recompute flags when committing a merge (issue1802)

Before this patch, Windows always did the wrong thing with exec bits
when committing a merge: consult the flags in first parent.

Now we manually recompute the result of merging flags at commit time,
which almost always does the right thing (except when there are
conflicts between symlink and exec flags).

To do this, we:

- pull flag synthesis out into its own function
- delay building this function unless it's needed
- add a merge case that compares flags in local and other against the ancestor

This has been tested in multiple ways on Linux:

- running the whole test suite with both old and new code in place,
  checking for differences in each flags() result
- running the whole test suite while comparing real on-disk flags
  against synthetic ones for merges
- test-issue1802 (from Martin Geisler) which disables exec bit
  checking on Unix

tests/test-issue1802.t

@@ -0,0 +1,69 b''
+Create extension that can disable exec checks:
+
+  $ cat > noexec.py <<EOF
+  > from mercurial import extensions, util
+  > def setflags(orig, f, l, x):
+  >     pass
+  > def checkexec(orig, path):
+  >     return False
+  > def extsetup(ui):
+  >     extensions.wrapfunction(util, 'setflags', setflags)
+  >     extensions.wrapfunction(util, 'checkexec', checkexec)
+  > EOF
+
+  $ hg init unix-repo
+  $ cd unix-repo
+  $ touch a
+  $ hg add a
+  $ hg commit -m 'unix: add a'
+  $ hg clone . ../win-repo
+  updating to branch default
+  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
+  $ chmod +x a
+  $ hg commit -m 'unix: chmod a'
+  $ hg manifest -v
+  755 * a
+
+  $ cd ../win-repo
+
+  $ touch b
+  $ hg add b
+  $ hg commit -m 'win: add b'
+
+  $ hg manifest -v
+  644   a
+  644   b
+
+  $ hg pull
+  pulling from $TESTTMP/unix-repo
+  searching for changes
+  adding changesets
+  adding manifests
+  adding file changes
+  added 1 changesets with 0 changes to 0 files (+1 heads)
+  (run 'hg heads' to see heads, 'hg merge' to merge)
+
+  $ hg manifest -v -r tip
+  755 * a
+
+Simulate a Windows merge:
+
+  $ hg --config extensions.n=$TESTTMP/noexec.py merge --debug
+    searching for copies back to rev 1
+    unmatched files in local:
+     b
+  resolving manifests
+   overwrite None partial False
+   ancestor a03b0deabf2b local d6fa54f68ae1+ remote 2d8bcf2dda39
+   a: update permissions -> e
+  updating: a 1/1 files (100.00%)
+  0 files updated, 0 files merged, 0 files removed, 0 files unresolved
+  (branch merge, don't forget to commit)
+
+Simulate a Windows commit:
+
+  $ hg --config extensions.n=$TESTTMP/noexec.py commit -m 'win: merge'
+
+  $ hg manifest -v
+  755 * a
+  644   b

mercurial/context.py

@@ -1,1116 +1,1137 b''
 # 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 version 2 or any later version.
 
 from node import nullid, nullrev, short, hex
 from i18n import _
 import ancestor, bdiff, error, util, scmutil, subrepo, patch, encoding
 import match as matchmod
 import os, errno, stat
 
 propertycache = util.propertycache
 
 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
         if isinstance(changeid, (long, int)):
             self._rev = changeid
             self._node = self._repo.changelog.node(changeid)
         else:
             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
 
     @propertycache
     def _changeset(self):
         return self._repo.changelog.read(self.node())
 
     @propertycache
     def _manifest(self):
         return self._repo.manifest.read(self._changeset[0])
 
     @propertycache
     def _manifestdelta(self):
         return self._repo.manifest.readdelta(self._changeset[0])
 
     @propertycache
     def _parents(self):
         p = self._repo.changelog.parentrevs(self._rev)
         if p[1] == nullrev:
             p = p[:-1]
         return [changectx(self._repo, x) for x in p]
 
     @propertycache
     def substate(self):
         return subrepo.state(self, self._repo.ui)
 
     def __contains__(self, key):
         return key in self._manifest
 
     def __getitem__(self, key):
         return self.filectx(key)
 
     def __iter__(self):
         for f in sorted(self._manifest):
             yield f
 
     def changeset(self):
         return self._changeset
     def manifest(self):
         return self._manifest
     def manifestnode(self):
         return self._changeset[0]
 
     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 encoding.tolocal(self._changeset[5].get("branch"))
     def extra(self):
         return self._changeset[5]
     def tags(self):
         return self._repo.nodetags(self._node)
     def bookmarks(self):
         return self._repo.nodebookmarks(self._node)
     def hidden(self):
         return self._rev in self._repo.changelog.hiddenrevs
 
     def parents(self):
         """return contexts for each parent changeset"""
         return self._parents
 
     def p1(self):
         return self._parents[0]
 
     def p2(self):
         if len(self._parents) == 2:
             return self._parents[1]
         return changectx(self._repo, -1)
 
     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 ancestors(self):
         for a in self._repo.changelog.ancestors(self._rev):
             yield changectx(self._repo, a)
 
     def descendants(self):
         for d in self._repo.changelog.descendants(self._rev):
             yield changectx(self._repo, d)
 
     def _fileinfo(self, path):
         if '_manifest' in self.__dict__:
             try:
                 return self._manifest[path], self._manifest.flags(path)
             except KeyError:
                 raise error.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 error.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 error.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 ancestor(self, c2):
         """
         return the ancestor context of self and c2
         """
         # deal with workingctxs
         n2 = c2._node
         if n2 is None:
             n2 = c2._parents[0]._node
         n = self._repo.changelog.ancestor(self._node, n2)
         return changectx(self._repo, n)
 
     def walk(self, match):
         fset = set(match.files())
         # for dirstate.walk, files=['.'] means "walk the whole tree".
         # follow that here, too
         fset.discard('.')
         for fn in self:
             for ffn in fset:
                 # match if the file is the exact name or a directory
                 if ffn == fn or fn.startswith("%s/" % ffn):
                     fset.remove(ffn)
                     break
             if match(fn):
                 yield fn
         for fn in sorted(fset):
             if match.bad(fn, _('no such file in rev %s') % self) and match(fn):
                 yield fn
 
     def sub(self, path):
         return subrepo.subrepo(self, path)
 
     def match(self, pats=[], include=None, exclude=None, default='glob'):
         r = self._repo
         return matchmod.match(r.root, r.getcwd(), pats,
                               include, exclude, default,
                               auditor=r.auditor, ctx=self)
 
     def diff(self, ctx2=None, match=None, **opts):
         """Returns a diff generator for the given contexts and matcher"""
         if ctx2 is None:
             ctx2 = self.p1()
         if ctx2 is not None and not isinstance(ctx2, changectx):
             ctx2 = self._repo[ctx2]
         diffopts = patch.diffopts(self._repo.ui, opts)
         return patch.diff(self._repo, ctx2.node(), self.node(),
                           match=match, opts=diffopts)
 
 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), \
                 ("bad args: changeid=%r, fileid=%r, changectx=%r"
                  % (changeid, fileid, changectx))
 
         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
 
     @propertycache
     def _changectx(self):
         return changectx(self._repo, self._changeid)
 
     @propertycache
     def _filelog(self):
         return self._repo.file(self._path)
 
     @propertycache
     def _changeid(self):
         if '_changectx' in self.__dict__:
             return self._changectx.rev()
         else:
             return self._filelog.linkrev(self._filerev)
 
     @propertycache
     def _filenode(self):
         if '_fileid' in self.__dict__:
             return self._filelog.lookup(self._fileid)
         else:
             return self._changectx.filenode(self._path)
 
     @propertycache
     def _filerev(self):
         return self._filelog.rev(self._filenode)
 
     @propertycache
     def _repopath(self):
         return self._path
 
     def __nonzero__(self):
         try:
             self._filenode
             return True
         except error.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._filenode))
         except AttributeError:
             return id(self)
 
     def __eq__(self, other):
         try:
             return (self._path == other._path
                     and self._filenode == other._filenode)
         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._filerev)
 
     def linkrev(self):
         return self._filelog.linkrev(self._filerev)
     def node(self):
         return self._changectx.node()
     def hex(self):
         return hex(self.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 extra(self):
         return self._changectx.extra()
     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, fctx):
         """compare with other file context
 
         returns True if different than fctx.
         """
         if (fctx._filerev is None and self._repo._encodefilterpats
             or self.size() == fctx.size()):
             return self._filelog.cmp(self._filenode, fctx.data())
 
         return True
 
     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 error.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 p1(self):
         return self.parents()[0]
 
     def p2(self):
         p = self.parents()
         if len(p) == 2:
             return p[1]
         return filectx(self._repo, self._path, fileid=-1, filelog=self._filelog)
 
     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.lrucachefunc(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.lrucachefunc(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
 
         # This algorithm would prefer to be recursive, but Python is a
         # bit recursion-hostile. Instead we do an iterative
         # depth-first search.
 
         visit = [base]
         hist = {}
         pcache = {}
         needed = {base: 1}
         while visit:
             f = visit[-1]
             if f not in pcache:
                 pcache[f] = parents(f)
 
             ready = True
             pl = pcache[f]
             for p in pl:
                 if p not in hist:
                     ready = False
                     visit.append(p)
                     needed[p] = needed.get(p, 0) + 1
             if ready:
                 visit.pop()
                 curr = decorate(f.data(), f)
                 for p in pl:
                     curr = pair(hist[p], curr)
                     if needed[p] == 1:
                         del hist[p]
                     else:
                         needed[p] -= 1
 
                 hist[f] = curr
                 pcache[f] = []
 
         return zip(hist[base][0], hist[base][1].splitlines(True))
 
     def ancestor(self, fc2, actx=None):
         """
         find the common ancestor file context, if any, of self, and fc2
 
         If actx is given, it must be the changectx of the common ancestor
         of self's and fc2's respective changesets.
         """
 
         if actx is None:
             actx = self.changectx().ancestor(fc2.changectx())
 
         # the trivial case: changesets are unrelated, files must be too
         if not actx:
             return None
 
         # the easy case: no (relevant) renames
         if fc2.path() == self.path() and self.path() in actx:
             return actx[self.path()]
         acache = {}
 
         # prime the ancestor cache for the working directory
         for c in (self, fc2):
             if c._filerev is 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
 
     def ancestors(self):
         visit = {}
         c = self
         while True:
             for parent in c.parents():
                 visit[(parent.rev(), parent.node())] = parent
             if not visit:
                 break
             c = visit.pop(max(visit))
             yield c
 
 class workingctx(changectx):
     """A workingctx object makes access to data related to
     the current working directory convenient.
     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, 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 changes:
             self._status = list(changes[:4])
             self._unknown = changes[4]
             self._ignored = changes[5]
             self._clean = changes[6]
         else:
             self._unknown = None
             self._ignored = None
             self._clean = None
 
         self._extra = {}
         if extra:
             self._extra = extra.copy()
         if 'branch' not in self._extra:
             try:
                 branch = encoding.fromlocal(self._repo.dirstate.branch())
             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 __repr__(self):
         return "<workingctx %s>" % str(self)
 
     def __nonzero__(self):
         return True
 
     def __contains__(self, key):
         return self._repo.dirstate[key] not in "?r"
 
+    def _buildflagfunc(self):
+        # Create a fallback function for getting file flags when the
+        # filesystem doesn't support them
+
+        copiesget = self._repo.dirstate.copies().get
+
+        if len(self._parents) < 2:
+            # when we have one parent, it's easy: copy from parent
+            man = self._parents[0].manifest()
+            def func(f):
+                f = copiesget(f, f)
+                return man.flags(f)
+        else:
+            # merges are tricky: we try to reconstruct the unstored
+            # result from the merge (issue1802)
+            p1, p2 = self._parents
+            pa = p1.ancestor(p2)
+            m1, m2, ma = p1.manifest(), p2.manifest(), pa.manifest()
+
+            def func(f):
+                f = copiesget(f, f) # may be wrong for merges with copies
+                fl1, fl2, fla = m1.flags(f), m2.flags(f), ma.flags(f)
+                if fl1 == fl2:
+                    return fl1
+                if fl1 == fla:
+                    return fl2
+                if fl2 == fla:
+                    return fl1
+                return '' # punt for conflicts
+
+        return func
+
+    @propertycache
+    def _flagfunc(self):
+        return self._repo.dirstate.flagfunc(self._buildflagfunc)
+
     @propertycache
     def _manifest(self):
         """generate a manifest corresponding to the working directory"""
 
         if self._unknown is None:
             self.status(unknown=True)
 
         man = self._parents[0].manifest().copy()
-        copied = self._repo.dirstate.copies()
         if len(self._parents) > 1:
             man2 = self.p2().manifest()
             def getman(f):
                 if f in man:
                     return man
                 return man2
         else:
             getman = lambda f: man
-        def cf(f):
-            f = copied.get(f, f)
-            return getman(f).flags(f)
-        ff = self._repo.dirstate.flagfunc(cf)
+
+        copied = self._repo.dirstate.copies()
+        ff = self._flagfunc
         modified, added, removed, deleted = self._status
         unknown = self._unknown
         for i, l in (("a", added), ("m", modified), ("u", unknown)):
             for f in l:
                 orig = copied.get(f, f)
                 man[f] = getman(orig).get(orig, nullid) + i
                 try:
                     man.set(f, ff(f))
                 except OSError:
                     pass
 
         for f in deleted + removed:
             if f in man:
                 del man[f]
 
         return man
 
     def __iter__(self):
         d = self._repo.dirstate
         for f in d:
             if d[f] != 'r':
                 yield f
 
     @propertycache
     def _status(self):
         return self._repo.status()[:4]
 
     @propertycache
     def _user(self):
         return self._repo.ui.username()
 
     @propertycache
     def _date(self):
         return util.makedate()
 
     @propertycache
     def _parents(self):
         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
 
     def status(self, ignored=False, clean=False, unknown=False):
         """Explicit status query
         Unless this method is used to query the working copy status, the
         _status property will implicitly read the status using its default
         arguments."""
         stat = self._repo.status(ignored=ignored, clean=clean, unknown=unknown)
         self._unknown = self._ignored = self._clean = None
         if unknown:
             self._unknown = stat[4]
         if ignored:
             self._ignored = stat[5]
         if clean:
             self._clean = stat[6]
         self._status = stat[:4]
         return stat
 
     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 sorted(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):
         assert self._unknown is not None  # must call status first
         return self._unknown
     def ignored(self):
         assert self._ignored is not None  # must call status first
         return self._ignored
     def clean(self):
         assert self._clean is not None  # must call status first
         return self._clean
     def branch(self):
         return encoding.tolocal(self._extra['branch'])
     def extra(self):
         return self._extra
 
     def tags(self):
         t = []
         for p in self.parents():
             t.extend(p.tags())
         return t
 
     def bookmarks(self):
         b = []
         for p in self.parents():
             b.extend(p.bookmarks())
         return b
 
     def children(self):
         return []
 
     def flags(self, path):
         if '_manifest' in self.__dict__:
             try:
                 return self._manifest.flags(path)
             except KeyError:
                 return ''
 
-        orig = self._repo.dirstate.copies().get(path, path)
-
-        def findflag(ctx):
-            mnode = ctx.changeset()[0]
-            node, flag = self._repo.manifest.find(mnode, orig)
-            ff = self._repo.dirstate.flagfunc(lambda x: flag or '')
-            try:
-                return ff(path)
-            except OSError:
-                pass
-
-        flag = findflag(self._parents[0])
-        if flag is None and len(self.parents()) > 1:
-            flag = findflag(self._parents[1])
-        if flag is None or self._repo.dirstate[path] == 'r':
+        try:
+            return self._flagfunc(path)
+        except OSError:
             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):
         return sorted(self._repo.dirstate.walk(match, self.substate.keys(),
                                                True, False))
 
     def dirty(self, missing=False):
         "check whether a working directory is modified"
         # check subrepos first
         for s in self.substate:
             if self.sub(s).dirty():
                 return True
         # check current working dir
         return (self.p2() or self.branch() != self.p1().branch() or
                 self.modified() or self.added() or self.removed() or
                 (missing and self.deleted()))
 
     def add(self, list, prefix=""):
         join = lambda f: os.path.join(prefix, f)
         wlock = self._repo.wlock()
         ui, ds = self._repo.ui, self._repo.dirstate
         try:
             rejected = []
             for f in list:
                 scmutil.checkportable(ui, join(f))
                 p = self._repo.wjoin(f)
                 try:
                     st = os.lstat(p)
                 except OSError:
                     ui.warn(_("%s does not exist!\n") % join(f))
                     rejected.append(f)
                     continue
                 if st.st_size > 10000000:
                     ui.warn(_("%s: up to %d MB of RAM may be required "
                               "to manage this file\n"
                               "(use 'hg revert %s' to cancel the "
                               "pending addition)\n")
                               % (f, 3 * st.st_size // 1000000, join(f)))
                 if not (stat.S_ISREG(st.st_mode) or stat.S_ISLNK(st.st_mode)):
                     ui.warn(_("%s not added: only files and symlinks "
                               "supported currently\n") % join(f))
                     rejected.append(p)
                 elif ds[f] in 'amn':
                     ui.warn(_("%s already tracked!\n") % join(f))
                 elif ds[f] == 'r':
                     ds.normallookup(f)
                 else:
                     ds.add(f)
             return rejected
         finally:
             wlock.release()
 
     def forget(self, files):
         wlock = self._repo.wlock()
         try:
             for f in files:
                 if self._repo.dirstate[f] != 'a':
                     self._repo.dirstate.remove(f)
                 elif f not in self._repo.dirstate:
                     self._repo.ui.warn(_("%s not tracked!\n") % f)
                 else:
                     self._repo.dirstate.drop(f)
         finally:
             wlock.release()
 
     def ancestors(self):
         for a in self._repo.changelog.ancestors(
             *[p.rev() for p in self._parents]):
             yield changectx(self._repo, a)
 
     def undelete(self, list):
         pctxs = self.parents()
         wlock = self._repo.wlock()
         try:
             for f in list:
                 if self._repo.dirstate[f] != 'r':
                     self._repo.ui.warn(_("%s not removed!\n") % f)
                 else:
                     fctx = f in pctxs[0] and pctxs[0][f] or pctxs[1][f]
                     t = fctx.data()
                     self._repo.wwrite(f, t, fctx.flags())
                     self._repo.dirstate.normal(f)
         finally:
             wlock.release()
 
     def copy(self, source, dest):
         p = self._repo.wjoin(dest)
         if not os.path.lexists(p):
             self._repo.ui.warn(_("%s does not exist!\n") % dest)
         elif not (os.path.isfile(p) or os.path.islink(p)):
             self._repo.ui.warn(_("copy failed: %s is not a file or a "
                                  "symbolic link\n") % dest)
         else:
             wlock = self._repo.wlock()
             try:
                 if self._repo.dirstate[dest] in '?r':
                     self._repo.dirstate.add(dest)
                 self._repo.dirstate.copy(source, dest)
             finally:
                 wlock.release()
 
 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
 
     @propertycache
     def _changectx(self):
         return workingctx(self._repo)
 
     def __nonzero__(self):
         return True
 
     def __str__(self):
         return "%s@%s" % (self.path(), self._changectx)
 
     def __repr__(self):
         return "<workingfilectx %s>" % str(self)
 
     def data(self):
         return self._repo.wread(self._path)
     def renamed(self):
         rp = self._repo.dirstate.copied(self._path)
         if not rp:
             return None
         return rp, self._changectx._parents[0]._manifest.get(rp, nullid)
 
     def parents(self):
         '''return parent filectxs, following copies if necessary'''
         def filenode(ctx, path):
             return ctx._manifest.get(path, nullid)
 
         path = self._path
         fl = self._filelog
         pcl = self._changectx._parents
         renamed = self.renamed()
 
         if renamed:
             pl = [renamed + (None,)]
         else:
             pl = [(path, filenode(pcl[0], path), fl)]
 
         for pc in pcl[1:]:
             pl.append((path, filenode(pc, path), 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.lstat(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, fctx):
         """compare with other file context
 
         returns True if different than fctx.
         """
         # fctx should be a filectx (not a wfctx)
         # invert comparison to reuse the same code path
         return fctx.cmp(self)
 
 class memctx(object):
     """Use memctx to perform in-memory commits via localrepo.commitctx().
 
     Revision information is supplied at initialization time while
     related files data and is made available through a callback
     mechanism.  'repo' is the current localrepo, 'parents' is a
     sequence of two parent revisions identifiers (pass None for every
     missing parent), 'text' is the commit message and 'files' lists
     names of files touched by the revision (normalized and relative to
     repository root).
 
     filectxfn(repo, memctx, path) is a callable receiving the
     repository, the current memctx object and the normalized path of
     requested file, relative to repository root. It is fired by the
     commit function for every file in 'files', but calls order is
     undefined. If the file is available in the revision being
     committed (updated or added), filectxfn returns a memfilectx
     object. If the file was removed, filectxfn raises an
     IOError. Moved files are represented by marking the source file
     removed and the new file added with copy information (see
     memfilectx).
 
     user receives the committer name and defaults to current
     repository username, date is the commit date in any format
     supported by util.parsedate() and defaults to current date, extra
     is a dictionary of metadata or is left empty.
     """
     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 = sorted(set(files))
         self._status = [files, [], [], [], []]
         self._filectxfn = filectxfn
 
         self._extra = extra and extra.copy() or {}
         if 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 __getitem__(self, key):
         return self.filectx(key)
 
     def p1(self):
         return self._parents[0]
     def p2(self):
         return self._parents[1]
 
     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 ignored(self):
         return self._status[5]
     def clean(self):
         return self._status[6]
     def branch(self):
         return encoding.tolocal(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)
 
     def commit(self):
         """commit context to the repo"""
         return self._repo.commitctx(self)
 
 class memfilectx(object):
     """memfilectx represents an in-memory file to commit.
 
     See memctx for more details.
     """
     def __init__(self, path, data, islink=False, isexec=False, copied=None):
         """
         path is the normalized file path relative to repository root.
         data is the file content as a string.
         islink is True if the file is a symbolic link.
         isexec is True if the file is executable.
         copied is the source file path if current file was copied in the
         revision being committed, 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

@@ -1,721 +1,724 b''
 # 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 version 2 or any later version.
 
 from node import nullid
 from i18n import _
 import scmutil, util, ignore, osutil, parsers, encoding
 import struct, os, stat, errno
 import cStringIO
 
 _format = ">cllll"
 propertycache = util.propertycache
 
 def _finddirs(path):
     pos = path.rfind('/')
     while pos != -1:
         yield path[:pos]
         pos = path.rfind('/', 0, pos)
 
 def _incdirs(dirs, path):
     for base in _finddirs(path):
         if base in dirs:
             dirs[base] += 1
             return
         dirs[base] = 1
 
 def _decdirs(dirs, path):
     for base in _finddirs(path):
         if dirs[base] > 1:
             dirs[base] -= 1
             return
         del dirs[base]
 
 class dirstate(object):
 
     def __init__(self, opener, ui, root, validate):
         '''Create a new dirstate object.
 
         opener is an open()-like callable that can be used to open the
         dirstate file; root is the root of the directory tracked by
         the dirstate.
         '''
         self._opener = opener
         self._validate = validate
         self._root = root
         self._rootdir = os.path.join(root, '')
         self._dirty = False
         self._dirtypl = False
         self._lastnormaltime = None
         self._ui = ui
 
     @propertycache
     def _map(self):
         '''Return the dirstate contents as a map from filename to
         (state, mode, size, time).'''
         self._read()
         return self._map
 
     @propertycache
     def _copymap(self):
         self._read()
         return self._copymap
 
     @propertycache
     def _foldmap(self):
         f = {}
         for name in self._map:
             f[os.path.normcase(name)] = name
         return f
 
     @propertycache
     def _branch(self):
         try:
             return self._opener.read("branch").strip() or "default"
         except IOError:
             return "default"
 
     @propertycache
     def _pl(self):
         try:
             fp = self._opener("dirstate")
             st = fp.read(40)
             fp.close()
             l = len(st)
             if l == 40:
                 return st[:20], st[20:40]
             elif l > 0 and l < 40:
                 raise util.Abort(_('working directory state appears damaged!'))
         except IOError, err:
             if err.errno != errno.ENOENT:
                 raise
         return [nullid, nullid]
 
     @propertycache
     def _dirs(self):
         dirs = {}
         for f, s in self._map.iteritems():
             if s[0] != 'r':
                 _incdirs(dirs, f)
         return dirs
 
     @propertycache
     def _ignore(self):
         files = [self._join('.hgignore')]
         for name, path in self._ui.configitems("ui"):
             if name == 'ignore' or name.startswith('ignore.'):
                 files.append(util.expandpath(path))
         return ignore.ignore(self._root, files, self._ui.warn)
 
     @propertycache
     def _slash(self):
         return self._ui.configbool('ui', 'slash') and os.sep != '/'
 
     @propertycache
     def _checklink(self):
         return util.checklink(self._root)
 
     @propertycache
     def _checkexec(self):
         return util.checkexec(self._root)
 
     @propertycache
     def _checkcase(self):
         return not util.checkcase(self._join('.hg'))
 
     def _join(self, f):
         # much faster than os.path.join()
         # it's safe because f is always a relative path
         return self._rootdir + f
 
-    def flagfunc(self, fallback):
+    def flagfunc(self, buildfallback):
+        if self._checklink and self._checkexec:
+            def f(x):
+                p = self._join(x)
+                if os.path.islink(p):
+                    return 'l'
+                if util.isexec(p):
+                    return 'x'
+                return ''
+            return f
+
+        fallback = buildfallback()
         if self._checklink:
-            if self._checkexec:
-                def f(x):
-                    p = self._join(x)
-                    if os.path.islink(p):
-                        return 'l'
-                    if util.isexec(p):
-                        return 'x'
-                    return ''
-                return f
             def f(x):
                 if os.path.islink(self._join(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.isexec(self._join(x)):
                     return 'x'
                 return ''
             return f
-        return fallback
+        else:
+            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):
         '''Return the current state of key (a filename) in the dirstate.
 
         States are:
           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 sorted(self._map):
             yield x
 
     def parents(self):
         return [self._validate(p) for p in self._pl]
 
     def p1(self):
         return self._validate(self._pl[0])
 
     def p2(self):
         return self._validate(self._pl[1])
 
     def branch(self):
         return encoding.tolocal(self._branch)
 
     def setparents(self, p1, p2=nullid):
         self._dirty = self._dirtypl = True
         self._pl = p1, p2
 
     def setbranch(self, branch):
         if branch in ['tip', '.', 'null']:
             raise util.Abort(_('the name \'%s\' is reserved') % branch)
         self._branch = encoding.fromlocal(branch)
         self._opener.write("branch", self._branch + '\n')
 
     def _read(self):
         self._map = {}
         self._copymap = {}
         try:
             st = self._opener.read("dirstate")
         except IOError, err:
             if err.errno != errno.ENOENT:
                 raise
             return
         if not st:
             return
 
         p = parsers.parse_dirstate(self._map, self._copymap, st)
         if not self._dirtypl:
             self._pl = p
 
     def invalidate(self):
         for a in ("_map", "_copymap", "_foldmap", "_branch", "_pl", "_dirs",
                 "_ignore"):
             if a in self.__dict__:
                 delattr(self, a)
         self._lastnormaltime = None
         self._dirty = False
 
     def copy(self, source, dest):
         """Mark dest as a copy of source. Unmark dest if source is None."""
         if source == dest:
             return
         self._dirty = True
         if source is not None:
             self._copymap[dest] = source
         elif dest in self._copymap:
             del self._copymap[dest]
 
     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__:
             _decdirs(self._dirs, f)
 
     def _addpath(self, f, check=False):
         oldstate = self[f]
         if check or oldstate == "r":
             scmutil.checkfilename(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__:
             _incdirs(self._dirs, f)
 
     def normal(self, f):
         '''Mark a file normal and clean.'''
         self._dirty = True
         self._addpath(f)
         s = os.lstat(self._join(f))
         mtime = int(s.st_mtime)
         self._map[f] = ('n', s.st_mode, s.st_size, mtime)
         if f in self._copymap:
             del self._copymap[f]
         if mtime > self._lastnormaltime:
             # Remember the most recent modification timeslot for status(),
             # to make sure we won't miss future size-preserving file content
             # modifications that happen within the same timeslot.
             self._lastnormaltime = mtime
 
     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' (-1) or coming from other parent (-2) 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.otherparent(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)
         if f in self._copymap:
             del self._copymap[f]
 
     def otherparent(self, f):
         '''Mark as coming from the other parent, always dirty.'''
         if self._pl[1] == nullid:
             raise util.Abort(_("setting %r to other parent "
                                "only allowed in merges") % f)
         self._dirty = True
         self._addpath(f)
         self._map[f] = ('n', 0, -2, -1)
         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)
         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:
             # backup the previous state
             entry = self._map[f]
             if entry[0] == 'm': # merge
                 size = -1
             elif entry[0] == 'n' and entry[2] == -2: # other parent
                 size = -2
         self._map[f] = ('r', 0, size, 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, int(s.st_mtime))
         if f in self._copymap:
             del self._copymap[f]
 
     def drop(self, f):
         '''Drop a file from the dirstate'''
         self._dirty = True
         self._droppath(f)
         del self._map[f]
 
     def _normalize(self, path, isknown):
         normed = os.path.normcase(path)
         folded = self._foldmap.get(normed, None)
         if folded is None:
             if isknown or not os.path.lexists(os.path.join(self._root, path)):
                 folded = path
             else:
                 folded = self._foldmap.setdefault(normed,
                                 util.fspath(path, self._root))
         return folded
 
     def normalize(self, path, isknown=False):
         '''
         normalize the case of a pathname when on a casefolding filesystem
 
         isknown specifies whether the filename came from walking the
         disk, to avoid extra filesystem access
 
         The normalized case is determined based on the following precedence:
 
         - version of name already stored in the dirstate
         - version of name stored on disk
         - version provided via command arguments
         '''
 
         if self._checkcase:
             return self._normalize(path, isknown)
         return path
 
     def clear(self):
         self._map = {}
         if "_dirs" in self.__dict__:
             delattr(self, "_dirs")
         self._copymap = {}
         self._pl = [nullid, nullid]
         self._lastnormaltime = None
         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)
             else:
                 self._map[f] = ('n', 0666, -1, 0)
         self._pl = (parent, nullid)
         self._dirty = True
 
     def write(self):
         if not self._dirty:
             return
         st = self._opener("dirstate", "w", atomictemp=True)
 
         # use the modification time of the newly created temporary file as the
         # filesystem's notion of 'now'
         now = int(util.fstat(st).st_mtime)
 
         cs = cStringIO.StringIO()
         copymap = self._copymap
         pack = struct.pack
         write = cs.write
         write("".join(self._pl))
         for f, e in self._map.iteritems():
             if e[0] == 'n' and e[3] == now:
                 # The file was last modified "simultaneously" with the current
                 # write to dirstate (i.e. within the same second for file-
                 # systems with a granularity of 1 sec). This commonly happens
                 # for at least a couple of files on 'update'.
                 # The user could change the file without changing its size
                 # within the same second. Invalidate the file's stat data in
                 # dirstate, forcing future 'status' calls to compare the
                 # contents of the file. This prevents mistakenly treating such
                 # files as clean.
                 e = (e[0], 0, -1, -1)   # mark entry as 'unset'
                 self._map[f] = e
 
             if f in copymap:
                 f = "%s\0%s" % (f, copymap[f])
             e = pack(_format, e[0], e[1], e[2], e[3], len(f))
             write(e)
             write(f)
         st.write(cs.getvalue())
         st.close()
         self._lastnormaltime = None
         self._dirty = self._dirtypl = 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, subrepos, unknown, ignored):
         '''
         Walk recursively through the directory tree, finding all files
         matched by match.
 
         Return a dict mapping filename to stat-like object (either
         mercurial.osutil.stat instance or return value of os.stat()).
         '''
 
         def fwarn(f, msg):
             self._ui.warn('%s: %s\n' % (self.pathto(f), msg))
             return False
 
         def badtype(mode):
             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')
             return _('unsupported file type (type is %s)') % kind
 
         ignore = self._ignore
         dirignore = self._dirignore
         if ignored:
             ignore = util.never
             dirignore = util.never
         elif not unknown:
             # if unknown and ignored are False, skip step 2
             ignore = util.always
             dirignore = util.always
 
         matchfn = match.matchfn
         badfn = match.bad
         dmap = self._map
         normpath = util.normpath
         listdir = osutil.listdir
         lstat = os.lstat
         getkind = stat.S_IFMT
         dirkind = stat.S_IFDIR
         regkind = stat.S_IFREG
         lnkkind = stat.S_IFLNK
         join = self._join
         work = []
         wadd = work.append
 
         exact = skipstep3 = False
         if matchfn == match.exact: # match.exact
             exact = True
             dirignore = util.always # skip step 2
         elif match.files() and not match.anypats(): # match.match, no patterns
             skipstep3 = True
 
         if self._checkcase:
             normalize = self._normalize
             skipstep3 = False
         else:
             normalize = lambda x, y: x
 
         files = sorted(match.files())
         subrepos.sort()
         i, j = 0, 0
         while i < len(files) and j < len(subrepos):
             subpath = subrepos[j] + "/"
             if files[i] < subpath:
                 i += 1
                 continue
             while i < len(files) and files[i].startswith(subpath):
                 del files[i]
             j += 1
 
         if not files or '.' in files:
             files = ['']
         results = dict.fromkeys(subrepos)
         results['.hg'] = None
 
         # step 1: find all explicit files
         for ff in files:
             nf = normalize(normpath(ff), False)
             if nf in results:
                 continue
 
             try:
                 st = lstat(join(nf))
                 kind = getkind(st.st_mode)
                 if kind == dirkind:
                     skipstep3 = False
                     if nf in dmap:
                         #file deleted on disk but still in dirstate
                         results[nf] = None
                     match.dir(nf)
                     if not dirignore(nf):
                         wadd(nf)
                 elif kind == regkind or kind == lnkkind:
                     results[nf] = st
                 else:
                     badfn(ff, badtype(kind))
                     if nf in dmap:
                         results[nf] = None
             except OSError, inst:
                 if nf in dmap: # does it exactly match a file?
                     results[nf] = None
                 else: # does it match a directory?
                     prefix = nf + "/"
                     for fn in dmap:
                         if fn.startswith(prefix):
                             match.dir(nf)
                             skipstep3 = False
                             break
                     else:
                         badfn(ff, inst.strerror)
 
         # step 2: visit subdirectories
         while work:
             nd = work.pop()
             skip = None
             if nd == '.':
                 nd = ''
             else:
                 skip = '.hg'
             try:
                 entries = listdir(join(nd), stat=True, skip=skip)
             except OSError, inst:
                 if inst.errno == errno.EACCES:
                     fwarn(nd, inst.strerror)
                     continue
                 raise
             for f, kind, st in entries:
                 nf = normalize(nd and (nd + "/" + f) or f, True)
                 if nf not in results:
                     if kind == dirkind:
                         if not ignore(nf):
                             match.dir(nf)
                             wadd(nf)
                         if nf in dmap and matchfn(nf):
                             results[nf] = None
                     elif kind == regkind or kind == lnkkind:
                         if nf in dmap:
                             if matchfn(nf):
                                 results[nf] = st
                         elif matchfn(nf) and not ignore(nf):
                             results[nf] = st
                     elif nf in dmap and matchfn(nf):
                         results[nf] = None
 
         # step 3: report unseen items in the dmap hash
         if not skipstep3 and not exact:
             visit = sorted([f for f in dmap if f not in results and matchfn(f)])
             for nf, st in zip(visit, util.statfiles([join(i) for i in visit])):
                 if not st is None and not getkind(st.st_mode) in (regkind, lnkkind):
                     st = None
                 results[nf] = st
         for s in subrepos:
             del results[s]
         del results['.hg']
         return results
 
     def status(self, match, subrepos, ignored, clean, unknown):
         '''Determine the status of the working copy relative to the
         dirstate and return a tuple of lists (unsure, modified, added,
         removed, deleted, unknown, ignored, clean), where:
 
           unsure:
             files that might have been modified since the dirstate was
             written, but need to be read to be sure (size is the same
             but mtime differs)
           modified:
             files that have definitely been modified since the dirstate
             was written (different size or mode)
           added:
             files that have been explicitly added with hg add
           removed:
             files that have been explicitly removed with hg remove
           deleted:
             files that have been deleted through other means ("missing")
           unknown:
             files not in the dirstate that are not ignored
           ignored:
             files not in the dirstate that are ignored
             (by _dirignore())
           clean:
             files that have definitely not been modified since the
             dirstate was written
         '''
         listignored, listclean, listunknown = ignored, clean, unknown
         lookup, modified, added, unknown, ignored = [], [], [], [], []
         removed, deleted, clean = [], [], []
 
         dmap = self._map
         ladd = lookup.append            # aka "unsure"
         madd = modified.append
         aadd = added.append
         uadd = unknown.append
         iadd = ignored.append
         radd = removed.append
         dadd = deleted.append
         cadd = clean.append
 
         lnkkind = stat.S_IFLNK
 
         for fn, st in self.walk(match, subrepos, listunknown,
                                 listignored).iteritems():
             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 = dmap[fn]
 
             if not st and state in "nma":
                 dadd(fn)
             elif state == 'n':
                 # The "mode & lnkkind != lnkkind or self._checklink"
                 # lines are an expansion of "islink => checklink"
                 # where islink means "is this a link?" and checklink
                 # means "can we check links?".
                 mtime = int(st.st_mtime)
                 if (size >= 0 and
                     (size != st.st_size
                      or ((mode ^ st.st_mode) & 0100 and self._checkexec))
                     and (mode & lnkkind != lnkkind or self._checklink)
                     or size == -2 # other parent
                     or fn in self._copymap):
                     madd(fn)
                 elif (mtime != time
                       and (mode & lnkkind != lnkkind or self._checklink)):
                     ladd(fn)
                 elif mtime == self._lastnormaltime:
                     # fn may have been changed in the same timeslot without
                     # changing its size. This can happen if we quickly do
                     # multiple commits in a single transaction.
                     # Force lookup, so we don't miss such a racy file change.
                     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)