largefiles: factor out procedures to update standins for pre-committing

This patch factors out procedures to update standins for
pre-committing. This is one of preparations to avoid execution of such
procedures according to invocation context.

For example, resuming automated committing (e.g. "hg rebase
--continue") should update standins at the 1st commit, because
largefiles in the working directory may be modified manually. But on
the other hand, it should avoid updating standins at subsequent
committings for efficiency reason.

For simplicity, this patch just moves procedures mechanically only
with replacing below.

  - "self"            => "repo"
  - "lfutil."         => (none)
  - "orig" invocation => returning "match"

Using "fstandin" instead "standin" as the name of local variable for
the loop below is the only special care, because the latter shadows
the same name function in "lfutil.py".

  [before]
                for standin in standins:
                    lfile = lfutil.splitstandin(standin)
                    if lfdirstate[lfile] != 'r':
                        lfutil.updatestandin(self, standin)

  [after]
    for fstandin in standins:
        lfile = splitstandin(fstandin)
        if lfdirstate[lfile] != 'r':
            updatestandin(repo, fstandin)

hgext/largefiles/lfutil.py

@@ -1,429 +1,539 b''
 # Copyright 2009-2010 Gregory P. Ward
 # Copyright 2009-2010 Intelerad Medical Systems Incorporated
 # Copyright 2010-2011 Fog Creek Software
 # Copyright 2010-2011 Unity Technologies
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 '''largefiles utility code: must not import other modules in this package.'''
 
 import os
 import platform
 import shutil
 import stat
+import copy
 
 from mercurial import dirstate, httpconnection, match as match_, util, scmutil
 from mercurial.i18n import _
 from mercurial import node
 
 shortname = '.hglf'
 shortnameslash = shortname + '/'
 longname = 'largefiles'
 
 
 # -- Private worker functions ------------------------------------------
 
 def getminsize(ui, assumelfiles, opt, default=10):
     lfsize = opt
     if not lfsize and assumelfiles:
         lfsize = ui.config(longname, 'minsize', default=default)
     if lfsize:
         try:
             lfsize = float(lfsize)
         except ValueError:
             raise util.Abort(_('largefiles: size must be number (not %s)\n')
                              % lfsize)
     if lfsize is None:
         raise util.Abort(_('minimum size for largefiles must be specified'))
     return lfsize
 
 def link(src, dest):
     util.makedirs(os.path.dirname(dest))
     try:
         util.oslink(src, dest)
     except OSError:
         # if hardlinks fail, fallback on atomic copy
         dst = util.atomictempfile(dest)
         for chunk in util.filechunkiter(open(src, 'rb')):
             dst.write(chunk)
         dst.close()
         os.chmod(dest, os.stat(src).st_mode)
 
 def usercachepath(ui, hash):
     path = ui.configpath(longname, 'usercache', None)
     if path:
         path = os.path.join(path, hash)
     else:
         if os.name == 'nt':
             appdata = os.getenv('LOCALAPPDATA', os.getenv('APPDATA'))
             if appdata:
                 path = os.path.join(appdata, longname, hash)
         elif platform.system() == 'Darwin':
             home = os.getenv('HOME')
             if home:
                 path = os.path.join(home, 'Library', 'Caches',
                                     longname, hash)
         elif os.name == 'posix':
             path = os.getenv('XDG_CACHE_HOME')
             if path:
                 path = os.path.join(path, longname, hash)
             else:
                 home = os.getenv('HOME')
                 if home:
                     path = os.path.join(home, '.cache', longname, hash)
         else:
             raise util.Abort(_('unknown operating system: %s\n') % os.name)
     return path
 
 def inusercache(ui, hash):
     path = usercachepath(ui, hash)
     return path and os.path.exists(path)
 
 def findfile(repo, hash):
     if instore(repo, hash):
         repo.ui.note(_('found %s in store\n') % hash)
         return storepath(repo, hash)
     elif inusercache(repo.ui, hash):
         repo.ui.note(_('found %s in system cache\n') % hash)
         path = storepath(repo, hash)
         link(usercachepath(repo.ui, hash), path)
         return path
     return None
 
 class largefilesdirstate(dirstate.dirstate):
     def __getitem__(self, key):
         return super(largefilesdirstate, self).__getitem__(unixpath(key))
     def normal(self, f):
         return super(largefilesdirstate, self).normal(unixpath(f))
     def remove(self, f):
         return super(largefilesdirstate, self).remove(unixpath(f))
     def add(self, f):
         return super(largefilesdirstate, self).add(unixpath(f))
     def drop(self, f):
         return super(largefilesdirstate, self).drop(unixpath(f))
     def forget(self, f):
         return super(largefilesdirstate, self).forget(unixpath(f))
     def normallookup(self, f):
         return super(largefilesdirstate, self).normallookup(unixpath(f))
     def _ignore(self, f):
         return False
 
 def openlfdirstate(ui, repo, create=True):
     '''
     Return a dirstate object that tracks largefiles: i.e. its root is
     the repo root, but it is saved in .hg/largefiles/dirstate.
     '''
     lfstoredir = repo.join(longname)
     opener = scmutil.opener(lfstoredir)
     lfdirstate = largefilesdirstate(opener, ui, repo.root,
                                      repo.dirstate._validate)
 
     # If the largefiles dirstate does not exist, populate and create
     # it. This ensures that we create it on the first meaningful
     # largefiles operation in a new clone.
     if create and not os.path.exists(os.path.join(lfstoredir, 'dirstate')):
         matcher = getstandinmatcher(repo)
         standins = repo.dirstate.walk(matcher, [], False, False)
 
         if len(standins) > 0:
             util.makedirs(lfstoredir)
 
         for standin in standins:
             lfile = splitstandin(standin)
             lfdirstate.normallookup(lfile)
     return lfdirstate
 
 def lfdirstatestatus(lfdirstate, repo):
     wctx = repo['.']
     match = match_.always(repo.root, repo.getcwd())
     unsure, s = lfdirstate.status(match, [], False, False, False)
     modified, clean = s.modified, s.clean
     for lfile in unsure:
         try:
             fctx = wctx[standin(lfile)]
         except LookupError:
             fctx = None
         if not fctx or fctx.data().strip() != hashfile(repo.wjoin(lfile)):
             modified.append(lfile)
         else:
             clean.append(lfile)
             lfdirstate.normal(lfile)
     return s
 
 def listlfiles(repo, rev=None, matcher=None):
     '''return a list of largefiles in the working copy or the
     specified changeset'''
 
     if matcher is None:
         matcher = getstandinmatcher(repo)
 
     # ignore unknown files in working directory
     return [splitstandin(f)
             for f in repo[rev].walk(matcher)
             if rev is not None or repo.dirstate[f] != '?']
 
 def instore(repo, hash):
     return os.path.exists(storepath(repo, hash))
 
 def storepath(repo, hash):
     return repo.join(os.path.join(longname, hash))
 
 def copyfromcache(repo, hash, filename):
     '''Copy the specified largefile from the repo or system cache to
     filename in the repository. Return true on success or false if the
     file was not found in either cache (which should not happened:
     this is meant to be called only after ensuring that the needed
     largefile exists in the cache).'''
     path = findfile(repo, hash)
     if path is None:
         return False
     util.makedirs(os.path.dirname(repo.wjoin(filename)))
     # The write may fail before the file is fully written, but we
     # don't use atomic writes in the working copy.
     shutil.copy(path, repo.wjoin(filename))
     return True
 
 def copytostore(repo, rev, file, uploaded=False):
     hash = readstandin(repo, file, rev)
     if instore(repo, hash):
         return
     copytostoreabsolute(repo, repo.wjoin(file), hash)
 
 def copyalltostore(repo, node):
     '''Copy all largefiles in a given revision to the store'''
 
     ctx = repo[node]
     for filename in ctx.files():
         if isstandin(filename) and filename in ctx.manifest():
             realfile = splitstandin(filename)
             copytostore(repo, ctx.node(), realfile)
 
 
 def copytostoreabsolute(repo, file, hash):
     if inusercache(repo.ui, hash):
         link(usercachepath(repo.ui, hash), storepath(repo, hash))
     elif not getattr(repo, "_isconverting", False):
         util.makedirs(os.path.dirname(storepath(repo, hash)))
         dst = util.atomictempfile(storepath(repo, hash),
                                   createmode=repo.store.createmode)
         for chunk in util.filechunkiter(open(file, 'rb')):
             dst.write(chunk)
         dst.close()
         linktousercache(repo, hash)
 
 def linktousercache(repo, hash):
     path = usercachepath(repo.ui, hash)
     if path:
         link(storepath(repo, hash), path)
 
 def getstandinmatcher(repo, pats=[], opts={}):
     '''Return a match object that applies pats to the standin directory'''
     standindir = repo.wjoin(shortname)
     if pats:
         pats = [os.path.join(standindir, pat) for pat in pats]
     else:
         # no patterns: relative to repo root
         pats = [standindir]
     # no warnings about missing files or directories
     match = scmutil.match(repo[None], pats, opts)
     match.bad = lambda f, msg: None
     return match
 
 def composestandinmatcher(repo, rmatcher):
     '''Return a matcher that accepts standins corresponding to the
     files accepted by rmatcher. Pass the list of files in the matcher
     as the paths specified by the user.'''
     smatcher = getstandinmatcher(repo, rmatcher.files())
     isstandin = smatcher.matchfn
     def composedmatchfn(f):
         return isstandin(f) and rmatcher.matchfn(splitstandin(f))
     smatcher.matchfn = composedmatchfn
 
     return smatcher
 
 def standin(filename):
     '''Return the repo-relative path to the standin for the specified big
     file.'''
     # Notes:
     # 1) Some callers want an absolute path, but for instance addlargefiles
     #    needs it repo-relative so it can be passed to repo[None].add().  So
     #    leave it up to the caller to use repo.wjoin() to get an absolute path.
     # 2) Join with '/' because that's what dirstate always uses, even on
     #    Windows. Change existing separator to '/' first in case we are
     #    passed filenames from an external source (like the command line).
     return shortnameslash + util.pconvert(filename)
 
 def isstandin(filename):
     '''Return true if filename is a big file standin. filename must be
     in Mercurial's internal form (slash-separated).'''
     return filename.startswith(shortnameslash)
 
 def splitstandin(filename):
     # Split on / because that's what dirstate always uses, even on Windows.
     # Change local separator to / first just in case we are passed filenames
     # from an external source (like the command line).
     bits = util.pconvert(filename).split('/', 1)
     if len(bits) == 2 and bits[0] == shortname:
         return bits[1]
     else:
         return None
 
 def updatestandin(repo, standin):
     file = repo.wjoin(splitstandin(standin))
     if os.path.exists(file):
         hash = hashfile(file)
         executable = getexecutable(file)
         writestandin(repo, standin, hash, executable)
 
 def readstandin(repo, filename, node=None):
     '''read hex hash from standin for filename at given node, or working
     directory if no node is given'''
     return repo[node][standin(filename)].data().strip()
 
 def writestandin(repo, standin, hash, executable):
     '''write hash to <repo.root>/<standin>'''
     repo.wwrite(standin, hash + '\n', executable and 'x' or '')
 
 def copyandhash(instream, outfile):
     '''Read bytes from instream (iterable) and write them to outfile,
     computing the SHA-1 hash of the data along the way. Return the hash.'''
     hasher = util.sha1('')
     for data in instream:
         hasher.update(data)
         outfile.write(data)
     return hasher.hexdigest()
 
 def hashrepofile(repo, file):
     return hashfile(repo.wjoin(file))
 
 def hashfile(file):
     if not os.path.exists(file):
         return ''
     hasher = util.sha1('')
     fd = open(file, 'rb')
     for data in util.filechunkiter(fd, 128 * 1024):
         hasher.update(data)
     fd.close()
     return hasher.hexdigest()
 
 def getexecutable(filename):
     mode = os.stat(filename).st_mode
     return ((mode & stat.S_IXUSR) and
             (mode & stat.S_IXGRP) and
             (mode & stat.S_IXOTH))
 
 def urljoin(first, second, *arg):
     def join(left, right):
         if not left.endswith('/'):
             left += '/'
         if right.startswith('/'):
             right = right[1:]
         return left + right
 
     url = join(first, second)
     for a in arg:
         url = join(url, a)
     return url
 
 def hexsha1(data):
     """hexsha1 returns the hex-encoded sha1 sum of the data in the file-like
     object data"""
     h = util.sha1()
     for chunk in util.filechunkiter(data):
         h.update(chunk)
     return h.hexdigest()
 
 def httpsendfile(ui, filename):
     return httpconnection.httpsendfile(ui, filename, 'rb')
 
 def unixpath(path):
     '''Return a version of path normalized for use with the lfdirstate.'''
     return util.pconvert(os.path.normpath(path))
 
 def islfilesrepo(repo):
     if ('largefiles' in repo.requirements and
             util.any(shortnameslash in f[0] for f in repo.store.datafiles())):
         return True
 
     return util.any(openlfdirstate(repo.ui, repo, False))
 
 class storeprotonotcapable(Exception):
     def __init__(self, storetypes):
         self.storetypes = storetypes
 
 def getstandinsstate(repo):
     standins = []
     matcher = getstandinmatcher(repo)
     for standin in repo.dirstate.walk(matcher, [], False, False):
         lfile = splitstandin(standin)
         try:
             hash = readstandin(repo, lfile)
         except IOError:
             hash = None
         standins.append((lfile, hash))
     return standins
 
 def synclfdirstate(repo, lfdirstate, lfile, normallookup):
     lfstandin = standin(lfile)
     if lfstandin in repo.dirstate:
         stat = repo.dirstate._map[lfstandin]
         state, mtime = stat[0], stat[3]
     else:
         state, mtime = '?', -1
     if state == 'n':
         if normallookup or mtime < 0:
             # state 'n' doesn't ensure 'clean' in this case
             lfdirstate.normallookup(lfile)
         else:
             lfdirstate.normal(lfile)
     elif state == 'm':
         lfdirstate.normallookup(lfile)
     elif state == 'r':
         lfdirstate.remove(lfile)
     elif state == 'a':
         lfdirstate.add(lfile)
     elif state == '?':
         lfdirstate.drop(lfile)
 
 def markcommitted(orig, ctx, node):
     repo = ctx._repo
 
     orig(node)
 
     lfdirstate = openlfdirstate(repo.ui, repo)
     for f in ctx.files():
         if isstandin(f):
             lfile = splitstandin(f)
             synclfdirstate(repo, lfdirstate, lfile, False)
     lfdirstate.write()
 
 def getlfilestoupdate(oldstandins, newstandins):
     changedstandins = set(oldstandins).symmetric_difference(set(newstandins))
     filelist = []
     for f in changedstandins:
         if f[0] not in filelist:
             filelist.append(f[0])
     return filelist
 
 def getlfilestoupload(repo, missing, addfunc):
     for n in missing:
         parents = [p for p in repo.changelog.parents(n) if p != node.nullid]
         ctx = repo[n]
         files = set(ctx.files())
         if len(parents) == 2:
             mc = ctx.manifest()
             mp1 = ctx.parents()[0].manifest()
             mp2 = ctx.parents()[1].manifest()
             for f in mp1:
                 if f not in mc:
                     files.add(f)
             for f in mp2:
                 if f not in mc:
                     files.add(f)
             for f in mc:
                 if mc[f] != mp1.get(f, None) or mc[f] != mp2.get(f, None):
                     files.add(f)
         for fn in files:
             if isstandin(fn) and fn in ctx:
                 addfunc(fn, ctx[fn].data().strip())
+
+def updatestandinsbymatch(repo, match):
+    '''Update standins in the working directory according to specified match
+
+    This returns (possibly modified) ``match`` object to be used for
+    subsequent commit process.
+    '''
+
+    ui = repo.ui
+
+    # Case 0: Automated committing
+    #
+    # While automated committing (like rebase, transplant
+    # and so on), this code path is used to avoid:
+    # (1) updating standins, because standins should
+    #     be already updated at this point
+    # (2) aborting when standins are matched by "match",
+    #     because automated committing may specify them directly
+    #
+    if getattr(repo, "_isrebasing", False) or \
+            getattr(repo, "_istransplanting", False):
+        return match
+
+    # Case 1: user calls commit with no specific files or
+    # include/exclude patterns: refresh and commit all files that
+    # are "dirty".
+    if match is None or match.always():
+        # Spend a bit of time here to get a list of files we know
+        # are modified so we can compare only against those.
+        # It can cost a lot of time (several seconds)
+        # otherwise to update all standins if the largefiles are
+        # large.
+        lfdirstate = openlfdirstate(ui, repo)
+        dirtymatch = match_.always(repo.root, repo.getcwd())
+        unsure, s = lfdirstate.status(dirtymatch, [], False, False,
+                                      False)
+        modifiedfiles = unsure + s.modified + s.added + s.removed
+        lfiles = listlfiles(repo)
+        # this only loops through largefiles that exist (not
+        # removed/renamed)
+        for lfile in lfiles:
+            if lfile in modifiedfiles:
+                if os.path.exists(
+                        repo.wjoin(standin(lfile))):
+                    # this handles the case where a rebase is being
+                    # performed and the working copy is not updated
+                    # yet.
+                    if os.path.exists(repo.wjoin(lfile)):
+                        updatestandin(repo,
+                            standin(lfile))
+
+        return match
+
+    lfiles = listlfiles(repo)
+    match._files = repo._subdirlfs(match.files(), lfiles)
+
+    # Case 2: user calls commit with specified patterns: refresh
+    # any matching big files.
+    smatcher = composestandinmatcher(repo, match)
+    standins = repo.dirstate.walk(smatcher, [], False, False)
+
+    # No matching big files: get out of the way and pass control to
+    # the usual commit() method.
+    if not standins:
+        return match
+
+    # Refresh all matching big files.  It's possible that the
+    # commit will end up failing, in which case the big files will
+    # stay refreshed.  No harm done: the user modified them and
+    # asked to commit them, so sooner or later we're going to
+    # refresh the standins.  Might as well leave them refreshed.
+    lfdirstate = openlfdirstate(ui, repo)
+    for fstandin in standins:
+        lfile = splitstandin(fstandin)
+        if lfdirstate[lfile] != 'r':
+            updatestandin(repo, fstandin)
+
+    # Cook up a new matcher that only matches regular files or
+    # standins corresponding to the big files requested by the
+    # user.  Have to modify _files to prevent commit() from
+    # complaining "not tracked" for big files.
+    match = copy.copy(match)
+    origmatchfn = match.matchfn
+
+    # Check both the list of largefiles and the list of
+    # standins because if a largefile was removed, it
+    # won't be in the list of largefiles at this point
+    match._files += sorted(standins)
+
+    actualfiles = []
+    for f in match._files:
+        fstandin = standin(f)
+
+        # ignore known largefiles and standins
+        if f in lfiles or fstandin in standins:
+            continue
+
+        actualfiles.append(f)
+    match._files = actualfiles
+
+    def matchfn(f):
+        if origmatchfn(f):
+            return f not in lfiles
+        else:
+            return f in standins
+
+    match.matchfn = matchfn
+
+    return match

hgext/largefiles/reposetup.py

@@ -1,456 +1,357 b''
 # Copyright 2009-2010 Gregory P. Ward
 # Copyright 2009-2010 Intelerad Medical Systems Incorporated
 # Copyright 2010-2011 Fog Creek Software
 # Copyright 2010-2011 Unity Technologies
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 '''setup for largefiles repositories: reposetup'''
 import copy
 import os
 
 from mercurial import error, manifest, match as match_, util
 from mercurial.i18n import _
 from mercurial import localrepo, scmutil
 
 import lfcommands
 import lfutil
 
 def reposetup(ui, repo):
     # wire repositories should be given new wireproto functions
     # by "proto.wirereposetup()" via "hg.wirepeersetupfuncs"
     if not repo.local():
         return
 
     class lfilesrepo(repo.__class__):
         lfstatus = False
         def status_nolfiles(self, *args, **kwargs):
             return super(lfilesrepo, self).status(*args, **kwargs)
 
         # When lfstatus is set, return a context that gives the names
         # of largefiles instead of their corresponding standins and
         # identifies the largefiles as always binary, regardless of
         # their actual contents.
         def __getitem__(self, changeid):
             ctx = super(lfilesrepo, self).__getitem__(changeid)
             if self.lfstatus:
                 class lfilesmanifestdict(manifest.manifestdict):
                     def __contains__(self, filename):
                         orig = super(lfilesmanifestdict, self).__contains__
                         return orig(filename) or orig(lfutil.standin(filename))
                 class lfilesctx(ctx.__class__):
                     def files(self):
                         filenames = super(lfilesctx, self).files()
                         return [lfutil.splitstandin(f) or f for f in filenames]
                     def manifest(self):
                         man1 = super(lfilesctx, self).manifest()
                         man1.__class__ = lfilesmanifestdict
                         return man1
                     def filectx(self, path, fileid=None, filelog=None):
                         orig = super(lfilesctx, self).filectx
                         try:
                             if filelog is not None:
                                 result = orig(path, fileid, filelog)
                             else:
                                 result = orig(path, fileid)
                         except error.LookupError:
                             # Adding a null character will cause Mercurial to
                             # identify this as a binary file.
                             if filelog is not None:
                                 result = orig(lfutil.standin(path), fileid,
                                               filelog)
                             else:
                                 result = orig(lfutil.standin(path), fileid)
                             olddata = result.data
                             result.data = lambda: olddata() + '\0'
                         return result
                 ctx.__class__ = lfilesctx
             return ctx
 
         # Figure out the status of big files and insert them into the
         # appropriate list in the result. Also removes standin files
         # from the listing. Revert to the original status if
         # self.lfstatus is False.
         # XXX large file status is buggy when used on repo proxy.
         # XXX this needs to be investigated.
         @localrepo.unfilteredmethod
         def status(self, node1='.', node2=None, match=None, ignored=False,
                 clean=False, unknown=False, listsubrepos=False):
             listignored, listclean, listunknown = ignored, clean, unknown
             orig = super(lfilesrepo, self).status
             if not self.lfstatus:
                 return orig(node1, node2, match, listignored, listclean,
                             listunknown, listsubrepos)
 
             # some calls in this function rely on the old version of status
             self.lfstatus = False
             ctx1 = self[node1]
             ctx2 = self[node2]
             working = ctx2.rev() is None
             parentworking = working and ctx1 == self['.']
 
             if match is None:
                 match = match_.always(self.root, self.getcwd())
 
             wlock = None
             try:
                 try:
                     # updating the dirstate is optional
                     # so we don't wait on the lock
                     wlock = self.wlock(False)
                 except error.LockError:
                     pass
 
                 # First check if paths or patterns were specified on the
                 # command line.  If there were, and they don't match any
                 # largefiles, we should just bail here and let super
                 # handle it -- thus gaining a big performance boost.
                 lfdirstate = lfutil.openlfdirstate(ui, self)
                 if not match.always():
                     for f in lfdirstate:
                         if match(f):
                             break
                     else:
                         return orig(node1, node2, match, listignored, listclean,
                                     listunknown, listsubrepos)
 
                 # Create a copy of match that matches standins instead
                 # of largefiles.
                 def tostandins(files):
                     if not working:
                         return files
                     newfiles = []
                     dirstate = self.dirstate
                     for f in files:
                         sf = lfutil.standin(f)
                         if sf in dirstate:
                             newfiles.append(sf)
                         elif sf in dirstate.dirs():
                             # Directory entries could be regular or
                             # standin, check both
                             newfiles.extend((f, sf))
                         else:
                             newfiles.append(f)
                     return newfiles
 
                 m = copy.copy(match)
                 m._files = tostandins(m._files)
 
                 result = orig(node1, node2, m, ignored, clean, unknown,
                               listsubrepos)
                 if working:
 
                     def sfindirstate(f):
                         sf = lfutil.standin(f)
                         dirstate = self.dirstate
                         return sf in dirstate or sf in dirstate.dirs()
 
                     match._files = [f for f in match._files
                                     if sfindirstate(f)]
                     # Don't waste time getting the ignored and unknown
                     # files from lfdirstate
                     unsure, s = lfdirstate.status(match, [], False, listclean,
                                                   False)
                     (modified, added, removed, clean) = (s.modified, s.added,
                                                          s.removed, s.clean)
                     if parentworking:
                         for lfile in unsure:
                             standin = lfutil.standin(lfile)
                             if standin not in ctx1:
                                 # from second parent
                                 modified.append(lfile)
                             elif ctx1[standin].data().strip() \
                                     != lfutil.hashfile(self.wjoin(lfile)):
                                 modified.append(lfile)
                             else:
                                 if listclean:
                                     clean.append(lfile)
                                 lfdirstate.normal(lfile)
                     else:
                         tocheck = unsure + modified + added + clean
                         modified, added, clean = [], [], []
 
                         for lfile in tocheck:
                             standin = lfutil.standin(lfile)
                             if standin in ctx1:
                                 abslfile = self.wjoin(lfile)
                                 if ((ctx1[standin].data().strip() !=
                                      lfutil.hashfile(abslfile)) or
                                     (('x' in ctx1.flags(standin)) !=
                                      bool(lfutil.getexecutable(abslfile)))):
                                     modified.append(lfile)
                                 elif listclean:
                                     clean.append(lfile)
                             else:
                                 added.append(lfile)
 
                         # at this point, 'removed' contains largefiles
                         # marked as 'R' in the working context.
                         # then, largefiles not managed also in the target
                         # context should be excluded from 'removed'.
                         removed = [lfile for lfile in removed
                                    if lfutil.standin(lfile) in ctx1]
 
                     # Standins no longer found in lfdirstate has been
                     # removed
                     for standin in ctx1.walk(lfutil.getstandinmatcher(self)):
                         lfile = lfutil.splitstandin(standin)
                         if not match(lfile):
                             continue
                         if lfile not in lfdirstate:
                             removed.append(lfile)
 
                     # Filter result lists
                     result = list(result)
 
                     # Largefiles are not really removed when they're
                     # still in the normal dirstate. Likewise, normal
                     # files are not really removed if they are still in
                     # lfdirstate. This happens in merges where files
                     # change type.
                     removed = [f for f in removed
                                if f not in self.dirstate]
                     result[2] = [f for f in result[2]
                                  if f not in lfdirstate]
 
                     lfiles = set(lfdirstate._map)
                     # Unknown files
                     result[4] = set(result[4]).difference(lfiles)
                     # Ignored files
                     result[5] = set(result[5]).difference(lfiles)
                     # combine normal files and largefiles
                     normals = [[fn for fn in filelist
                                 if not lfutil.isstandin(fn)]
                                for filelist in result]
                     lfstatus = (modified, added, removed, s.deleted, [], [],
                                 clean)
                     result = [sorted(list1 + list2)
                               for (list1, list2) in zip(normals, lfstatus)]
                 else: # not against working directory
                     result = [[lfutil.splitstandin(f) or f for f in items]
                               for items in result]
 
                 if wlock:
                     lfdirstate.write()
 
             finally:
                 if wlock:
                     wlock.release()
 
             self.lfstatus = True
             return scmutil.status(*result)
 
         # As part of committing, copy all of the largefiles into the
         # cache.
         def commitctx(self, ctx, *args, **kwargs):
             node = super(lfilesrepo, self).commitctx(ctx, *args, **kwargs)
             lfutil.copyalltostore(self, node)
             class lfilesctx(ctx.__class__):
                 def markcommitted(self, node):
                     orig = super(lfilesctx, self).markcommitted
                     return lfutil.markcommitted(orig, self, node)
             ctx.__class__ = lfilesctx
             return node
 
         # Before commit, largefile standins have not had their
         # contents updated to reflect the hash of their largefile.
         # Do that here.
         def commit(self, text="", user=None, date=None, match=None,
                 force=False, editor=False, extra={}):
             orig = super(lfilesrepo, self).commit
 
             wlock = self.wlock()
             try:
-                # Case 0: Automated committing
-                #
-                # While automated committing (like rebase, transplant
-                # and so on), this code path is used to avoid:
-                # (1) updating standins, because standins should
-                #     be already updated at this point
-                # (2) aborting when standins are matched by "match",
-                #     because automated committing may specify them directly
-                #
-                if getattr(self, "_isrebasing", False) or \
-                        getattr(self, "_istransplanting", False):
-                    result = orig(text=text, user=user, date=date, match=match,
-                                    force=force, editor=editor, extra=extra)
-                    return result
-                # Case 1: user calls commit with no specific files or
-                # include/exclude patterns: refresh and commit all files that
-                # are "dirty".
-                if match is None or match.always():
-                    # Spend a bit of time here to get a list of files we know
-                    # are modified so we can compare only against those.
-                    # It can cost a lot of time (several seconds)
-                    # otherwise to update all standins if the largefiles are
-                    # large.
-                    lfdirstate = lfutil.openlfdirstate(ui, self)
-                    dirtymatch = match_.always(self.root, self.getcwd())
-                    unsure, s = lfdirstate.status(dirtymatch, [], False, False,
-                                                  False)
-                    modifiedfiles = unsure + s.modified + s.added + s.removed
-                    lfiles = lfutil.listlfiles(self)
-                    # this only loops through largefiles that exist (not
-                    # removed/renamed)
-                    for lfile in lfiles:
-                        if lfile in modifiedfiles:
-                            if os.path.exists(
-                                    self.wjoin(lfutil.standin(lfile))):
-                                # this handles the case where a rebase is being
-                                # performed and the working copy is not updated
-                                # yet.
-                                if os.path.exists(self.wjoin(lfile)):
-                                    lfutil.updatestandin(self,
-                                        lfutil.standin(lfile))
-
-                    result = orig(text=text, user=user, date=date, match=match,
-                                    force=force, editor=editor, extra=extra)
-
-                    return result
-
-                lfiles = lfutil.listlfiles(self)
-                match._files = self._subdirlfs(match.files(), lfiles)
-
-                # Case 2: user calls commit with specified patterns: refresh
-                # any matching big files.
-                smatcher = lfutil.composestandinmatcher(self, match)
-                standins = self.dirstate.walk(smatcher, [], False, False)
-
-                # No matching big files: get out of the way and pass control to
-                # the usual commit() method.
-                if not standins:
-                    return orig(text=text, user=user, date=date, match=match,
-                                    force=force, editor=editor, extra=extra)
-
-                # Refresh all matching big files.  It's possible that the
-                # commit will end up failing, in which case the big files will
-                # stay refreshed.  No harm done: the user modified them and
-                # asked to commit them, so sooner or later we're going to
-                # refresh the standins.  Might as well leave them refreshed.
-                lfdirstate = lfutil.openlfdirstate(ui, self)
-                for standin in standins:
-                    lfile = lfutil.splitstandin(standin)
-                    if lfdirstate[lfile] != 'r':
-                        lfutil.updatestandin(self, standin)
-
-                # Cook up a new matcher that only matches regular files or
-                # standins corresponding to the big files requested by the
-                # user.  Have to modify _files to prevent commit() from
-                # complaining "not tracked" for big files.
-                match = copy.copy(match)
-                origmatchfn = match.matchfn
-
-                # Check both the list of largefiles and the list of
-                # standins because if a largefile was removed, it
-                # won't be in the list of largefiles at this point
-                match._files += sorted(standins)
-
-                actualfiles = []
-                for f in match._files:
-                    fstandin = lfutil.standin(f)
-
-                    # ignore known largefiles and standins
-                    if f in lfiles or fstandin in standins:
-                        continue
-
-                    actualfiles.append(f)
-                match._files = actualfiles
-
-                def matchfn(f):
-                    if origmatchfn(f):
-                        return f not in lfiles
-                    else:
-                        return f in standins
-
-                match.matchfn = matchfn
+                match = lfutil.updatestandinsbymatch(self, match)
                 result = orig(text=text, user=user, date=date, match=match,
                                 force=force, editor=editor, extra=extra)
                 return result
             finally:
                 wlock.release()
 
         def push(self, remote, force=False, revs=None, newbranch=False):
             if remote.local():
                 missing = set(self.requirements) - remote.local().supported
                 if missing:
                     msg = _("required features are not"
                             " supported in the destination:"
                             " %s") % (', '.join(sorted(missing)))
                     raise util.Abort(msg)
             return super(lfilesrepo, self).push(remote, force=force, revs=revs,
                 newbranch=newbranch)
 
+        # TODO: _subdirlfs should be moved into "lfutil.py", because
+        # it is referred only from "lfutil.updatestandinsbymatch"
         def _subdirlfs(self, files, lfiles):
             '''
             Adjust matched file list
             If we pass a directory to commit whose only commitable files
             are largefiles, the core commit code aborts before finding
             the largefiles.
             So we do the following:
             For directories that only have largefiles as matches,
             we explicitly add the largefiles to the match list and remove
             the directory.
             In other cases, we leave the match list unmodified.
             '''
             actualfiles = []
             dirs = []
             regulars = []
 
             for f in files:
                 if lfutil.isstandin(f + '/'):
                     raise util.Abort(
                         _('file "%s" is a largefile standin') % f,
                         hint=('commit the largefile itself instead'))
                 # Scan directories
                 if os.path.isdir(self.wjoin(f)):
                     dirs.append(f)
                 else:
                     regulars.append(f)
 
             for f in dirs:
                 matcheddir = False
                 d = self.dirstate.normalize(f) + '/'
                 # Check for matched normal files
                 for mf in regulars:
                     if self.dirstate.normalize(mf).startswith(d):
                         actualfiles.append(f)
                         matcheddir = True
                         break
                 if not matcheddir:
                     # If no normal match, manually append
                     # any matching largefiles
                     for lf in lfiles:
                         if self.dirstate.normalize(lf).startswith(d):
                             actualfiles.append(lf)
                             if not matcheddir:
                                 actualfiles.append(lfutil.standin(f))
                                 matcheddir = True
                 # Nothing in dir, so readd it
                 # and let commit reject it
                 if not matcheddir:
                     actualfiles.append(f)
 
             # Always add normal files
             actualfiles += regulars
             return actualfiles
 
     repo.__class__ = lfilesrepo
 
     def prepushoutgoinghook(local, remote, outgoing):
         if outgoing.missing:
             toupload = set()
             addfunc = lambda fn, lfhash: toupload.add(lfhash)
             lfutil.getlfilestoupload(local, outgoing.missing, addfunc)
             lfcommands.uploadlfiles(ui, local, remote, toupload)
     repo.prepushoutgoinghooks.add("largefiles", prepushoutgoinghook)
 
     def checkrequireslfiles(ui, repo, **kwargs):
         if 'largefiles' not in repo.requirements and util.any(
                 lfutil.shortname+'/' in f[0] for f in repo.store.datafiles()):
             repo.requirements.add('largefiles')
             repo._writerequirements()
 
     ui.setconfig('hooks', 'changegroup.lfiles', checkrequireslfiles,
                  'largefiles')
     ui.setconfig('hooks', 'commit.lfiles', checkrequireslfiles, 'largefiles')