merge with stable

tests/test-casecollision-merge.t

@@ -0,0 +1,109
+run only on case-insensitive filesystems
+
+  $ "$TESTDIR/hghave" icasefs || exit 80
+
+################################
+test for branch merging
+################################
+
+  $ hg init repo1
+  $ cd repo1
+
+create base revision
+
+  $ echo base > base.txt
+  $ hg add base.txt
+  $ hg commit -m 'base'
+
+add same file in different case on both heads
+
+  $ echo a > a.txt
+  $ hg add a.txt
+  $ hg commit -m 'add a.txt'
+
+  $ hg update 0
+  0 files updated, 0 files merged, 1 files removed, 0 files unresolved
+
+  $ echo A > A.TXT
+  $ hg add A.TXT
+  $ hg commit -m 'add A.TXT'
+  created new head
+
+merge another, and fail with case-folding collision
+
+  $ hg merge
+  abort: case-folding collision between a.txt and A.TXT
+  [255]
+
+check clean-ness of working directory
+
+  $ hg status
+  $ hg parents --template '{rev}\n'
+  2
+  $ cd ..
+
+################################
+test for linear updates
+################################
+
+  $ hg init repo2
+  $ cd repo2
+
+create base revision (rev:0)
+
+  $ hg import --bypass --exact - <<EOF
+  > # HG changeset patch
+  > # User null
+  > # Date 1 0
+  > # Node ID e1bdf414b0ea9c831fd3a14e94a0a18e1410f98b
+  > # Parent  0000000000000000000000000000000000000000
+  > add a
+  > 
+  > diff --git a/a b/a
+  > new file mode 100644
+  > --- /dev/null
+  > +++ b/a
+  > @@ -0,0 +1,3 @@
+  > +this is line 1
+  > +this is line 2
+  > +this is line 3
+  > EOF
+  applying patch from stdin
+
+create rename revision (rev:1)
+
+  $ hg import --bypass --exact - <<EOF
+  > # HG changeset patch
+  > # User null
+  > # Date 1 0
+  > # Node ID 9dca9f19bb91851bc693544b598b0740629edfad
+  > # Parent  e1bdf414b0ea9c831fd3a14e94a0a18e1410f98b
+  > rename a to A
+  > 
+  > diff --git a/a b/A
+  > rename from a
+  > rename to A
+  > EOF
+  applying patch from stdin
+
+update to base revision, and modify 'a'
+
+  $ hg update 0
+  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
+  $ echo 'this is added line' >> a
+
+update to current tip linearly
+
+  $ hg update 1
+  merging a and A to A
+  0 files updated, 1 files merged, 0 files removed, 0 files unresolved
+
+check status and contents of file
+
+  $ hg status -A
+  M A
+  $ cat A
+  this is line 1
+  this is line 2
+  this is line 3
+  this is added line

tests/test-issue3084.t

@@ -0,0 +1,108
+
+  $ echo "[extensions]" >> $HGRCPATH
+  $ echo "largefiles =" >> $HGRCPATH
+
+Create the repository outside $HOME since largefiles write to
+$HOME/.cache/largefiles.
+
+  $ hg init test
+  $ cd test
+  $ echo "root" > root
+  $ hg add root
+  $ hg commit -m "Root commit"
+
+  $ echo "large" > foo
+  $ hg add --large foo
+  $ hg commit -m "Add foo as a largefile"
+
+  $ hg update -r 0
+  0 files updated, 0 files merged, 1 files removed, 0 files unresolved
+  getting changed largefiles
+  0 largefiles updated, 1 removed
+
+  $ echo "normal" > foo
+  $ hg add foo
+  $ hg commit -m "Add foo as normal file"
+  created new head
+
+Normal file in the working copy, keeping the normal version:
+
+  $ echo "n" | hg merge --config ui.interactive=Yes
+  foo has been turned into a largefile
+  use (l)argefile or keep as (n)ormal file? 0 files updated, 0 files merged, 1 files removed, 0 files unresolved
+  (branch merge, don't forget to commit)
+
+  $ hg status
+  $ cat foo
+  normal
+
+Normal file in the working copy, keeping the largefile version:
+
+  $ hg update -q -C
+  $ echo "l" | hg merge --config ui.interactive=Yes
+  foo has been turned into a largefile
+  use (l)argefile or keep as (n)ormal file? 1 files updated, 0 files merged, 1 files removed, 0 files unresolved
+  (branch merge, don't forget to commit)
+  getting changed largefiles
+  1 largefiles updated, 0 removed
+
+  $ hg status
+  M foo
+
+  $ hg diff --nodates
+  diff -r fa129ab6b5a7 .hglf/foo
+  --- /dev/null
+  +++ b/.hglf/foo
+  @@ -0,0 +1,1 @@
+  +7f7097b041ccf68cc5561e9600da4655d21c6d18
+  diff -r fa129ab6b5a7 foo
+  --- a/foo
+  +++ /dev/null
+  @@ -1,1 +0,0 @@
+  -normal
+
+  $ cat foo
+  large
+
+Largefile in the working copy, keeping the normal version:
+
+  $ hg update -q -C -r 1
+  $ echo "n" | hg merge --config ui.interactive=Yes
+  foo has been turned into a normal file
+  keep as (l)argefile or use (n)ormal file? 1 files updated, 0 files merged, 1 files removed, 0 files unresolved
+  (branch merge, don't forget to commit)
+  getting changed largefiles
+  0 largefiles updated, 0 removed
+
+  $ hg status
+  M foo
+
+  $ hg diff --nodates
+  diff -r ff521236428a .hglf/foo
+  --- a/.hglf/foo
+  +++ /dev/null
+  @@ -1,1 +0,0 @@
+  -7f7097b041ccf68cc5561e9600da4655d21c6d18
+  diff -r ff521236428a foo
+  --- /dev/null
+  +++ b/foo
+  @@ -0,0 +1,1 @@
+  +normal
+
+  $ cat foo
+  normal
+
+Largefile in the working copy, keeping the largefile version:
+
+  $ hg update -q -C -r 1
+  $ echo "l" | hg merge --config ui.interactive=Yes
+  foo has been turned into a normal file
+  keep as (l)argefile or use (n)ormal file? 0 files updated, 0 files merged, 1 files removed, 0 files unresolved
+  (branch merge, don't forget to commit)
+  getting changed largefiles
+  1 largefiles updated, 0 removed
+
+  $ hg status
+
+  $ cat foo
+  large

hgext/largefiles/lfcommands.py

@@ -1,475 +1,479
 # 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.
 
 '''High-level command function for lfconvert, plus the cmdtable.'''
 
 import os
 import shutil
 
 from mercurial import util, match as match_, hg, node, context, error
 from mercurial.i18n import _
 
 import lfutil
 import basestore
 
 # -- Commands ----------------------------------------------------------
 
 def lfconvert(ui, src, dest, *pats, **opts):
     '''convert a normal repository to a largefiles repository
 
     Convert repository SOURCE to a new repository DEST, identical to
     SOURCE except that certain files will be converted as largefiles:
     specifically, any file that matches any PATTERN *or* whose size is
     above the minimum size threshold is converted as a largefile. The
     size used to determine whether or not to track a file as a
     largefile is the size of the first version of the file. The
     minimum size can be specified either with --size or in
     configuration as ``largefiles.size``.
 
     After running this command you will need to make sure that
     largefiles is enabled anywhere you intend to push the new
     repository.
 
     Use --to-normal to convert largefiles back to normal files; after
     this, the DEST repository can be used without largefiles at all.'''
 
     if opts['to_normal']:
         tolfile = False
     else:
         tolfile = True
         size = lfutil.getminsize(ui, True, opts.get('size'), default=None)
 
     if not hg.islocal(src):
         raise util.Abort(_('%s is not a local Mercurial repo') % src)
     if not hg.islocal(dest):
         raise util.Abort(_('%s is not a local Mercurial repo') % dest)
 
     rsrc = hg.repository(ui, src)
     ui.status(_('initializing destination %s\n') % dest)
     rdst = hg.repository(ui, dest, create=True)
 
     success = False
     try:
         # Lock destination to prevent modification while it is converted to.
         # Don't need to lock src because we are just reading from its history
         # which can't change.
         dst_lock = rdst.lock()
 
         # Get a list of all changesets in the source.  The easy way to do this
         # is to simply walk the changelog, using changelog.nodesbewteen().
         # Take a look at mercurial/revlog.py:639 for more details.
         # Use a generator instead of a list to decrease memory usage
         ctxs = (rsrc[ctx] for ctx in rsrc.changelog.nodesbetween(None,
             rsrc.heads())[0])
         revmap = {node.nullid: node.nullid}
         if tolfile:
             lfiles = set()
             normalfiles = set()
             if not pats:
                 pats = ui.configlist(lfutil.longname, 'patterns', default=[])
             if pats:
                 matcher = match_.match(rsrc.root, '', list(pats))
             else:
                 matcher = None
 
             lfiletohash = {}
             for ctx in ctxs:
                 ui.progress(_('converting revisions'), ctx.rev(),
                     unit=_('revision'), total=rsrc['tip'].rev())
                 _lfconvert_addchangeset(rsrc, rdst, ctx, revmap,
                     lfiles, normalfiles, matcher, size, lfiletohash)
             ui.progress(_('converting revisions'), None)
 
             if os.path.exists(rdst.wjoin(lfutil.shortname)):
                 shutil.rmtree(rdst.wjoin(lfutil.shortname))
 
             for f in lfiletohash.keys():
                 if os.path.isfile(rdst.wjoin(f)):
                     os.unlink(rdst.wjoin(f))
                 try:
                     os.removedirs(os.path.dirname(rdst.wjoin(f)))
                 except OSError:
                     pass
 
             # If there were any files converted to largefiles, add largefiles
             # to the destination repository's requirements.
             if lfiles:
                 rdst.requirements.add('largefiles')
                 rdst._writerequirements()
         else:
             for ctx in ctxs:
                 ui.progress(_('converting revisions'), ctx.rev(),
                     unit=_('revision'), total=rsrc['tip'].rev())
                 _addchangeset(ui, rsrc, rdst, ctx, revmap)
 
             ui.progress(_('converting revisions'), None)
         success = True
     finally:
         if not success:
             # we failed, remove the new directory
             shutil.rmtree(rdst.root)
         dst_lock.release()
 
 def _addchangeset(ui, rsrc, rdst, ctx, revmap):
  # Convert src parents to dst parents
     parents = []
     for p in ctx.parents():
         parents.append(revmap[p.node()])
     while len(parents) < 2:
         parents.append(node.nullid)
 
     # Generate list of changed files
     files = set(ctx.files())
     if node.nullid not in parents:
         mc = ctx.manifest()
         mp1 = ctx.parents()[0].manifest()
         mp2 = ctx.parents()[1].manifest()
         files |= (set(mp1) | set(mp2)) - set(mc)
         for f in mc:
             if mc[f] != mp1.get(f, None) or mc[f] != mp2.get(f, None):
                 files.add(f)
 
     def getfilectx(repo, memctx, f):
         if lfutil.standin(f) in files:
             # if the file isn't in the manifest then it was removed
             # or renamed, raise IOError to indicate this
             try:
                 fctx = ctx.filectx(lfutil.standin(f))
             except error.LookupError:
                 raise IOError()
             renamed = fctx.renamed()
             if renamed:
                 renamed = lfutil.splitstandin(renamed[0])
 
             hash = fctx.data().strip()
             path = lfutil.findfile(rsrc, hash)
             ### TODO: What if the file is not cached?
             data = ''
             fd = None
             try:
                 fd = open(path, 'rb')
                 data = fd.read()
             finally:
                 if fd:
                     fd.close()
             return context.memfilectx(f, data, 'l' in fctx.flags(),
                                       'x' in fctx.flags(), renamed)
         else:
             try:
                 fctx = ctx.filectx(f)
             except error.LookupError:
                 raise IOError()
             renamed = fctx.renamed()
             if renamed:
                 renamed = renamed[0]
             data = fctx.data()
             if f == '.hgtags':
                 newdata = []
                 for line in data.splitlines():
                     id, name = line.split(' ', 1)
                     newdata.append('%s %s\n' % (node.hex(revmap[node.bin(id)]),
                         name))
                 data = ''.join(newdata)
             return context.memfilectx(f, data, 'l' in fctx.flags(),
                                       'x' in fctx.flags(), renamed)
 
     dstfiles = []
     for file in files:
         if lfutil.isstandin(file):
             dstfiles.append(lfutil.splitstandin(file))
         else:
             dstfiles.append(file)
     # Commit
     mctx = context.memctx(rdst, parents, ctx.description(), dstfiles,
                           getfilectx, ctx.user(), ctx.date(), ctx.extra())
     ret = rdst.commitctx(mctx)
     rdst.dirstate.setparents(ret)
     revmap[ctx.node()] = rdst.changelog.tip()
 
 def _lfconvert_addchangeset(rsrc, rdst, ctx, revmap, lfiles, normalfiles,
         matcher, size, lfiletohash):
     # Convert src parents to dst parents
     parents = []
     for p in ctx.parents():
         parents.append(revmap[p.node()])
     while len(parents) < 2:
         parents.append(node.nullid)
 
     # Generate list of changed files
     files = set(ctx.files())
     if node.nullid not in parents:
         mc = ctx.manifest()
         mp1 = ctx.parents()[0].manifest()
         mp2 = ctx.parents()[1].manifest()
         files |= (set(mp1) | set(mp2)) - set(mc)
         for f in mc:
             if mc[f] != mp1.get(f, None) or mc[f] != mp2.get(f, None):
                 files.add(f)
 
     dstfiles = []
     for f in files:
         if f not in lfiles and f not in normalfiles:
             islfile = _islfile(f, ctx, matcher, size)
             # If this file was renamed or copied then copy
             # the lfileness of its predecessor
             if f in ctx.manifest():
                 fctx = ctx.filectx(f)
                 renamed = fctx.renamed()
                 renamedlfile = renamed and renamed[0] in lfiles
                 islfile |= renamedlfile
                 if 'l' in fctx.flags():
                     if renamedlfile:
                         raise util.Abort(
                             _('renamed/copied largefile %s becomes symlink')
                             % f)
                     islfile = False
             if islfile:
                 lfiles.add(f)
             else:
                 normalfiles.add(f)
 
         if f in lfiles:
             dstfiles.append(lfutil.standin(f))
             # largefile in manifest if it has not been removed/renamed
             if f in ctx.manifest():
                 if 'l' in ctx.filectx(f).flags():
                     if renamed and renamed[0] in lfiles:
                         raise util.Abort(_('largefile %s becomes symlink') % f)
 
                 # largefile was modified, update standins
                 fullpath = rdst.wjoin(f)
                 util.makedirs(os.path.dirname(fullpath))
                 m = util.sha1('')
                 m.update(ctx[f].data())
                 hash = m.hexdigest()
                 if f not in lfiletohash or lfiletohash[f] != hash:
                     try:
                         fd = open(fullpath, 'wb')
                         fd.write(ctx[f].data())
                     finally:
                         if fd:
                             fd.close()
                     executable = 'x' in ctx[f].flags()
                     os.chmod(fullpath, lfutil.getmode(executable))
                     lfutil.writestandin(rdst, lfutil.standin(f), hash,
                         executable)
                     lfiletohash[f] = hash
         else:
             # normal file
             dstfiles.append(f)
 
     def getfilectx(repo, memctx, f):
         if lfutil.isstandin(f):
             # if the file isn't in the manifest then it was removed
             # or renamed, raise IOError to indicate this
             srcfname = lfutil.splitstandin(f)
             try:
                 fctx = ctx.filectx(srcfname)
             except error.LookupError:
                 raise IOError()
             renamed = fctx.renamed()
             if renamed:
                 # standin is always a largefile because largefile-ness
                 # doesn't change after rename or copy
                 renamed = lfutil.standin(renamed[0])
 
             return context.memfilectx(f, lfiletohash[srcfname] + '\n', 'l' in
                 fctx.flags(), 'x' in fctx.flags(), renamed)
         else:
             try:
                 fctx = ctx.filectx(f)
             except error.LookupError:
                 raise IOError()
             renamed = fctx.renamed()
             if renamed:
                 renamed = renamed[0]
 
             data = fctx.data()
             if f == '.hgtags':
                 newdata = []
                 for line in data.splitlines():
                     id, name = line.split(' ', 1)
                     newdata.append('%s %s\n' % (node.hex(revmap[node.bin(id)]),
                         name))
                 data = ''.join(newdata)
             return context.memfilectx(f, data, 'l' in fctx.flags(),
                                       'x' in fctx.flags(), renamed)
 
     # Commit
     mctx = context.memctx(rdst, parents, ctx.description(), dstfiles,
                           getfilectx, ctx.user(), ctx.date(), ctx.extra())
     ret = rdst.commitctx(mctx)
     rdst.dirstate.setparents(ret)
     revmap[ctx.node()] = rdst.changelog.tip()
 
 def _islfile(file, ctx, matcher, size):
     '''Return true if file should be considered a largefile, i.e.
     matcher matches it or it is larger than size.'''
     # never store special .hg* files as largefiles
     if file == '.hgtags' or file == '.hgignore' or file == '.hgsigs':
         return False
     if matcher and matcher(file):
         return True
     try:
         return ctx.filectx(file).size() >= size * 1024 * 1024
     except error.LookupError:
         return False
 
 def uploadlfiles(ui, rsrc, rdst, files):
     '''upload largefiles to the central store'''
 
     if not files:
         return
 
     store = basestore._openstore(rsrc, rdst, put=True)
 
     at = 0
     files = filter(lambda h: not store.exists(h), files)
     for hash in files:
         ui.progress(_('uploading largefiles'), at, unit='largefile',
                     total=len(files))
         source = lfutil.findfile(rsrc, hash)
         if not source:
             raise util.Abort(_('largefile %s missing from store'
                                ' (needs to be uploaded)') % hash)
         # XXX check for errors here
         store.put(source, hash)
         at += 1
     ui.progress(_('uploading largefiles'), None)
 
 def verifylfiles(ui, repo, all=False, contents=False):
     '''Verify that every big file revision in the current changeset
     exists in the central store.  With --contents, also verify that
     the contents of each big file revision are correct (SHA-1 hash
     matches the revision ID).  With --all, check every changeset in
     this repository.'''
     if all:
         # Pass a list to the function rather than an iterator because we know a
         # list will work.
         revs = range(len(repo))
     else:
         revs = ['.']
 
     store = basestore._openstore(repo)
     return store.verify(revs, contents=contents)
 
 def cachelfiles(ui, repo, node):
     '''cachelfiles ensures that all largefiles needed by the specified revision
     are present in the repository's largefile cache.
 
     returns a tuple (cached, missing).  cached is the list of files downloaded
     by this operation; missing is the list of files that were needed but could
     not be found.'''
     lfiles = lfutil.listlfiles(repo, node)
     toget = []
 
     for lfile in lfiles:
         expectedhash = repo[node][lfutil.standin(lfile)].data().strip()
         # if it exists and its hash matches, it might have been locally
         # modified before updating and the user chose 'local'.  in this case,
         # it will not be in any store, so don't look for it.
         if ((not os.path.exists(repo.wjoin(lfile)) or
              expectedhash != lfutil.hashfile(repo.wjoin(lfile))) and
             not lfutil.findfile(repo, expectedhash)):
             toget.append((lfile, expectedhash))
 
     if toget:
         store = basestore._openstore(repo)
         ret = store.get(toget)
         return ret
 
     return ([], [])
 
 def updatelfiles(ui, repo, filelist=None, printmessage=True):
     wlock = repo.wlock()
     try:
         lfdirstate = lfutil.openlfdirstate(ui, repo)
         lfiles = set(lfutil.listlfiles(repo)) | set(lfdirstate)
 
         if filelist is not None:
             lfiles = [f for f in lfiles if f in filelist]
 
         printed = False
         if printmessage and lfiles:
             ui.status(_('getting changed largefiles\n'))
             printed = True
             cachelfiles(ui, repo, '.')
 
         updated, removed = 0, 0
         for i in map(lambda f: _updatelfile(repo, lfdirstate, f), lfiles):
             # increment the appropriate counter according to _updatelfile's
             # return value
             updated += i > 0 and i or 0
             removed -= i < 0 and i or 0
             if printmessage and (removed or updated) and not printed:
                 ui.status(_('getting changed largefiles\n'))
                 printed = True
 
         lfdirstate.write()
         if printed and printmessage:
             ui.status(_('%d largefiles updated, %d removed\n') % (updated,
                 removed))
     finally:
         wlock.release()
 
 def _updatelfile(repo, lfdirstate, lfile):
     '''updates a single largefile and copies the state of its standin from
     the repository's dirstate to its state in the lfdirstate.
 
     returns 1 if the file was modified, -1 if the file was removed, 0 if the
     file was unchanged, and None if the needed largefile was missing from the
     cache.'''
     ret = 0
     abslfile = repo.wjoin(lfile)
     absstandin = repo.wjoin(lfutil.standin(lfile))
     if os.path.exists(absstandin):
         if os.path.exists(absstandin+'.orig'):
             shutil.copyfile(abslfile, abslfile+'.orig')
         expecthash = lfutil.readstandin(repo, lfile)
         if (expecthash != '' and
             (not os.path.exists(abslfile) or
              expecthash != lfutil.hashfile(abslfile))):
             if not lfutil.copyfromcache(repo, expecthash, lfile):
                 # use normallookup() to allocate entry in largefiles dirstate,
                 # because lack of it misleads lfiles_repo.status() into
                 # recognition that such cache missing files are REMOVED.
                 lfdirstate.normallookup(lfile)
                 return None # don't try to set the mode
             ret = 1
         mode = os.stat(absstandin).st_mode
         if mode != os.stat(abslfile).st_mode:
             os.chmod(abslfile, mode)
             ret = 1
     else:
-        if os.path.exists(abslfile):
+        # Remove lfiles for which the standin is deleted, unless the
+        # lfile is added to the repository again. This happens when a
+        # largefile is converted back to a normal file: the standin
+        # disappears, but a new (normal) file appears as the lfile.
+        if os.path.exists(abslfile) and lfile not in repo[None]:
             os.unlink(abslfile)
             ret = -1
     state = repo.dirstate[lfutil.standin(lfile)]
     if state == 'n':
         lfdirstate.normal(lfile)
     elif state == 'r':
         lfdirstate.remove(lfile)
     elif state == 'a':
         lfdirstate.add(lfile)
     elif state == '?':
         lfdirstate.drop(lfile)
     return ret
 
 # -- hg commands declarations ------------------------------------------------
 
 cmdtable = {
     'lfconvert': (lfconvert,
                   [('s', 'size', '',
                     _('minimum size (MB) for files to be converted '
                       'as largefiles'),
                     'SIZE'),
                   ('', 'to-normal', False,
                    _('convert from a largefiles repo to a normal repo')),
                   ],
                   _('hg lfconvert SOURCE DEST [FILE ...]')),
     }

hgext/largefiles/overrides.py

@@ -1,825 +1,909
 # 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.
 
 '''Overridden Mercurial commands and functions for the largefiles extension'''
 
 import os
 import copy
 
 from mercurial import hg, commands, util, cmdutil, scmutil, match as match_, \
     node, archival, error, merge
 from mercurial.i18n import _
 from mercurial.node import hex
 from hgext import rebase
 
 import lfutil
 import lfcommands
 
 def installnormalfilesmatchfn(manifest):
     '''overrides scmutil.match so that the matcher it returns will ignore all
     largefiles'''
     oldmatch = None # for the closure
     def override_match(ctx, pats=[], opts={}, globbed=False,
             default='relpath'):
         match = oldmatch(ctx, pats, opts, globbed, default)
         m = copy.copy(match)
         notlfile = lambda f: not (lfutil.isstandin(f) or lfutil.standin(f) in
                 manifest)
         m._files = filter(notlfile, m._files)
         m._fmap = set(m._files)
         orig_matchfn = m.matchfn
         m.matchfn = lambda f: notlfile(f) and orig_matchfn(f) or None
         return m
     oldmatch = installmatchfn(override_match)
 
 def installmatchfn(f):
     oldmatch = scmutil.match
     setattr(f, 'oldmatch', oldmatch)
     scmutil.match = f
     return oldmatch
 
 def restorematchfn():
     '''restores scmutil.match to what it was before installnormalfilesmatchfn
     was called.  no-op if scmutil.match is its original function.
 
     Note that n calls to installnormalfilesmatchfn will require n calls to
     restore matchfn to reverse'''
     scmutil.match = getattr(scmutil.match, 'oldmatch', scmutil.match)
 
 # -- Wrappers: modify existing commands --------------------------------
 
 # Add works by going through the files that the user wanted to add and
 # checking if they should be added as largefiles. Then it makes a new
 # matcher which matches only the normal files and runs the original
 # version of add.
 def override_add(orig, ui, repo, *pats, **opts):
     large = opts.pop('large', None)
     lfsize = lfutil.getminsize(
         ui, lfutil.islfilesrepo(repo), opts.pop('lfsize', None))
 
     lfmatcher = None
     if os.path.exists(repo.wjoin(lfutil.shortname)):
         lfpats = ui.configlist(lfutil.longname, 'patterns', default=[])
         if lfpats:
             lfmatcher = match_.match(repo.root, '', list(lfpats))
 
     lfnames = []
     m = scmutil.match(repo[None], pats, opts)
     m.bad = lambda x, y: None
     wctx = repo[None]
     for f in repo.walk(m):
         exact = m.exact(f)
         lfile = lfutil.standin(f) in wctx
         nfile = f in wctx
         exists = lfile or nfile
 
         # Don't warn the user when they attempt to add a normal tracked file.
         # The normal add code will do that for us.
         if exact and exists:
             if lfile:
                 ui.warn(_('%s already a largefile\n') % f)
             continue
 
         if exact or not exists:
             abovemin = (lfsize and
                         os.lstat(repo.wjoin(f)).st_size >= lfsize * 1024 * 1024)
             if large or abovemin or (lfmatcher and lfmatcher(f)):
                 lfnames.append(f)
                 if ui.verbose or not exact:
                     ui.status(_('adding %s as a largefile\n') % m.rel(f))
 
     bad = []
     standins = []
 
     # Need to lock, otherwise there could be a race condition between
     # when standins are created and added to the repo.
     wlock = repo.wlock()
     try:
         if not opts.get('dry_run'):
             lfdirstate = lfutil.openlfdirstate(ui, repo)
             for f in lfnames:
                 standinname = lfutil.standin(f)
                 lfutil.writestandin(repo, standinname, hash='',
                     executable=lfutil.getexecutable(repo.wjoin(f)))
                 standins.append(standinname)
                 if lfdirstate[f] == 'r':
                     lfdirstate.normallookup(f)
                 else:
                     lfdirstate.add(f)
             lfdirstate.write()
             bad += [lfutil.splitstandin(f)
                     for f in lfutil.repo_add(repo, standins)
                     if f in m.files()]
     finally:
         wlock.release()
 
     installnormalfilesmatchfn(repo[None].manifest())
     result = orig(ui, repo, *pats, **opts)
     restorematchfn()
 
     return (result == 1 or bad) and 1 or 0
 
 def override_remove(orig, ui, repo, *pats, **opts):
     manifest = repo[None].manifest()
     installnormalfilesmatchfn(manifest)
     orig(ui, repo, *pats, **opts)
     restorematchfn()
 
     after, force = opts.get('after'), opts.get('force')
     if not pats and not after:
         raise util.Abort(_('no files specified'))
     m = scmutil.match(repo[None], pats, opts)
     try:
         repo.lfstatus = True
         s = repo.status(match=m, clean=True)
     finally:
         repo.lfstatus = False
     modified, added, deleted, clean = [[f for f in list
                                         if lfutil.standin(f) in manifest]
                                        for list in [s[0], s[1], s[3], s[6]]]
 
     def warn(files, reason):
         for f in files:
             ui.warn(_('not removing %s: %s (use -f to force removal)\n')
                     % (m.rel(f), reason))
 
     if force:
         remove, forget = modified + deleted + clean, added
     elif after:
         remove, forget = deleted, []
         warn(modified + added + clean, _('file still exists'))
     else:
         remove, forget = deleted + clean, []
         warn(modified, _('file is modified'))
         warn(added, _('file has been marked for add'))
 
     for f in sorted(remove + forget):
         if ui.verbose or not m.exact(f):
             ui.status(_('removing %s\n') % m.rel(f))
 
     # Need to lock because standin files are deleted then removed from the
     # repository and we could race inbetween.
     wlock = repo.wlock()
     try:
         lfdirstate = lfutil.openlfdirstate(ui, repo)
         for f in remove:
             if not after:
                 os.unlink(repo.wjoin(f))
                 currentdir = os.path.split(f)[0]
                 while currentdir and not os.listdir(repo.wjoin(currentdir)):
                     os.rmdir(repo.wjoin(currentdir))
                     currentdir = os.path.split(currentdir)[0]
             lfdirstate.remove(f)
         lfdirstate.write()
 
         forget = [lfutil.standin(f) for f in forget]
         remove = [lfutil.standin(f) for f in remove]
         lfutil.repo_forget(repo, forget)
         lfutil.repo_remove(repo, remove, unlink=True)
     finally:
         wlock.release()
 
 def override_status(orig, ui, repo, *pats, **opts):
     try:
         repo.lfstatus = True
         return orig(ui, repo, *pats, **opts)
     finally:
         repo.lfstatus = False
 
 def override_log(orig, ui, repo, *pats, **opts):
     try:
         repo.lfstatus = True
         orig(ui, repo, *pats, **opts)
     finally:
         repo.lfstatus = False
 
 def override_verify(orig, ui, repo, *pats, **opts):
     large = opts.pop('large', False)
     all = opts.pop('lfa', False)
     contents = opts.pop('lfc', False)
 
     result = orig(ui, repo, *pats, **opts)
     if large:
         result = result or lfcommands.verifylfiles(ui, repo, all, contents)
     return result
 
 # Override needs to refresh standins so that update's normal merge
 # will go through properly. Then the other update hook (overriding repo.update)
 # will get the new files. Filemerge is also overriden so that the merge
 # will merge standins correctly.
 def override_update(orig, ui, repo, *pats, **opts):
     lfdirstate = lfutil.openlfdirstate(ui, repo)
     s = lfdirstate.status(match_.always(repo.root, repo.getcwd()), [], False,
         False, False)
     (unsure, modified, added, removed, missing, unknown, ignored, clean) = s
 
     # Need to lock between the standins getting updated and their
     # largefiles getting updated
     wlock = repo.wlock()
     try:
         if opts['check']:
             mod = len(modified) > 0
             for lfile in unsure:
                 standin = lfutil.standin(lfile)
                 if repo['.'][standin].data().strip() != \
                         lfutil.hashfile(repo.wjoin(lfile)):
                     mod = True
                 else:
                     lfdirstate.normal(lfile)
             lfdirstate.write()
             if mod:
                 raise util.Abort(_('uncommitted local changes'))
         # XXX handle removed differently
         if not opts['clean']:
             for lfile in unsure + modified + added:
                 lfutil.updatestandin(repo, lfutil.standin(lfile))
     finally:
         wlock.release()
     return orig(ui, repo, *pats, **opts)
 
+# Before starting the manifest merge, merge.updates will call
+# _checkunknown to check if there are any files in the merged-in
+# changeset that collide with unknown files in the working copy.
+#
+# The largefiles are seen as unknown, so this prevents us from merging
+# in a file 'foo' if we already have a largefile with the same name.
+#
+# The overridden function filters the unknown files by removing any
+# largefiles. This makes the merge proceed and we can then handle this
+# case further in the overridden manifestmerge function below.
+def override_checkunknown(origfn, wctx, mctx, folding):
+    origunknown = wctx.unknown()
+    wctx._unknown = filter(lambda f: lfutil.standin(f) not in wctx, origunknown)
+    try:
+        return origfn(wctx, mctx, folding)
+    finally:
+        wctx._unknown = origunknown
+
+# The manifest merge handles conflicts on the manifest level. We want
+# to handle changes in largefile-ness of files at this level too.
+#
+# The strategy is to run the original manifestmerge and then process
+# the action list it outputs. There are two cases we need to deal with:
+#
+# 1. Normal file in p1, largefile in p2. Here the largefile is
+#    detected via its standin file, which will enter the working copy
+#    with a "get" action. It is not "merge" since the standin is all
+#    Mercurial is concerned with at this level -- the link to the
+#    existing normal file is not relevant here.
+#
+# 2. Largefile in p1, normal file in p2. Here we get a "merge" action
+#    since the largefile will be present in the working copy and
+#    different from the normal file in p2. Mercurial therefore
+#    triggers a merge action.
+#
+# In both cases, we prompt the user and emit new actions to either
+# remove the standin (if the normal file was kept) or to remove the
+# normal file and get the standin (if the largefile was kept). The
+# default prompt answer is to use the largefile version since it was
+# presumably changed on purpose.
+#
+# Finally, the merge.applyupdates function will then take care of
+# writing the files into the working copy and lfcommands.updatelfiles
+# will update the largefiles.
+def override_manifestmerge(origfn, repo, p1, p2, pa, overwrite, partial):
+    actions = origfn(repo, p1, p2, pa, overwrite, partial)
+    processed = []
+
+    for action in actions:
+        if overwrite:
+            processed.append(action)
+            continue
+        f, m = action[:2]
+
+        choices = (_('&Largefile'), _('&Normal file'))
+        if m == "g" and lfutil.splitstandin(f) in p1 and f in p2:
+            # Case 1: normal file in the working copy, largefile in
+            # the second parent
+            lfile = lfutil.splitstandin(f)
+            standin = f
+            msg = _('%s has been turned into a largefile\n'
+                    'use (l)argefile or keep as (n)ormal file?') % lfile
+            if repo.ui.promptchoice(msg, choices, 0) == 0:
+                processed.append((lfile, "r"))
+                processed.append((standin, "g", p2.flags(standin)))
+            else:
+                processed.append((standin, "r"))
+        elif m == "m" and lfutil.standin(f) in p1 and f in p2:
+            # Case 2: largefile in the working copy, normal file in
+            # the second parent
+            standin = lfutil.standin(f)
+            lfile = f
+            msg = _('%s has been turned into a normal file\n'
+                    'keep as (l)argefile or use (n)ormal file?') % lfile
+            if repo.ui.promptchoice(msg, choices, 0) == 0:
+                processed.append((lfile, "r"))
+            else:
+                processed.append((standin, "r"))
+                processed.append((lfile, "g", p2.flags(lfile)))
+        else:
+            processed.append(action)
+
+    return processed
+
 # Override filemerge to prompt the user about how they wish to merge
 # largefiles. This will handle identical edits, and copy/rename +
 # edit without prompting the user.
 def override_filemerge(origfn, repo, mynode, orig, fcd, fco, fca):
     # Use better variable names here. Because this is a wrapper we cannot
     # change the variable names in the function declaration.
     fcdest, fcother, fcancestor = fcd, fco, fca
     if not lfutil.isstandin(orig):
         return origfn(repo, mynode, orig, fcdest, fcother, fcancestor)
     else:
         if not fcother.cmp(fcdest): # files identical?
             return None
 
         # backwards, use working dir parent as ancestor
         if fcancestor == fcother:
             fcancestor = fcdest.parents()[0]
 
         if orig != fcother.path():
             repo.ui.status(_('merging %s and %s to %s\n')
                            % (lfutil.splitstandin(orig),
                               lfutil.splitstandin(fcother.path()),
                               lfutil.splitstandin(fcdest.path())))
         else:
             repo.ui.status(_('merging %s\n')
                            % lfutil.splitstandin(fcdest.path()))
 
         if fcancestor.path() != fcother.path() and fcother.data() == \
                 fcancestor.data():
             return 0
         if fcancestor.path() != fcdest.path() and fcdest.data() == \
                 fcancestor.data():
             repo.wwrite(fcdest.path(), fcother.data(), fcother.flags())
             return 0
 
         if repo.ui.promptchoice(_('largefile %s has a merge conflict\n'
                              'keep (l)ocal or take (o)ther?') %
                              lfutil.splitstandin(orig),
                              (_('&Local'), _('&Other')), 0) == 0:
             return 0
         else:
             repo.wwrite(fcdest.path(), fcother.data(), fcother.flags())
             return 0
 
 # Copy first changes the matchers to match standins instead of
 # largefiles.  Then it overrides util.copyfile in that function it
 # checks if the destination largefile already exists. It also keeps a
 # list of copied files so that the largefiles can be copied and the
 # dirstate updated.
 def override_copy(orig, ui, repo, pats, opts, rename=False):
     # doesn't remove largefile on rename
     if len(pats) < 2:
         # this isn't legal, let the original function deal with it
         return orig(ui, repo, pats, opts, rename)
 
     def makestandin(relpath):
         path = scmutil.canonpath(repo.root, repo.getcwd(), relpath)
         return os.path.join(repo.wjoin(lfutil.standin(path)))
 
     fullpats = scmutil.expandpats(pats)
     dest = fullpats[-1]
 
     if os.path.isdir(dest):
         if not os.path.isdir(makestandin(dest)):
             os.makedirs(makestandin(dest))
     # This could copy both lfiles and normal files in one command,
     # but we don't want to do that. First replace their matcher to
     # only match normal files and run it, then replace it to just
     # match largefiles and run it again.
     nonormalfiles = False
     nolfiles = False
     try:
         try:
             installnormalfilesmatchfn(repo[None].manifest())
             result = orig(ui, repo, pats, opts, rename)
         except util.Abort, e:
             if str(e) != 'no files to copy':
                 raise e
             else:
                 nonormalfiles = True
             result = 0
     finally:
         restorematchfn()
 
     # The first rename can cause our current working directory to be removed.
     # In that case there is nothing left to copy/rename so just quit.
     try:
         repo.getcwd()
     except OSError:
         return result
 
     try:
         try:
             # When we call orig below it creates the standins but we don't add them
             # to the dir state until later so lock during that time.
             wlock = repo.wlock()
 
             manifest = repo[None].manifest()
             oldmatch = None # for the closure
             def override_match(ctx, pats=[], opts={}, globbed=False,
                     default='relpath'):
                 newpats = []
                 # The patterns were previously mangled to add the standin
                 # directory; we need to remove that now
                 for pat in pats:
                     if match_.patkind(pat) is None and lfutil.shortname in pat:
                         newpats.append(pat.replace(lfutil.shortname, ''))
                     else:
                         newpats.append(pat)
                 match = oldmatch(ctx, newpats, opts, globbed, default)
                 m = copy.copy(match)
                 lfile = lambda f: lfutil.standin(f) in manifest
                 m._files = [lfutil.standin(f) for f in m._files if lfile(f)]
                 m._fmap = set(m._files)
                 orig_matchfn = m.matchfn
                 m.matchfn = lambda f: (lfutil.isstandin(f) and
                                     lfile(lfutil.splitstandin(f)) and
                                     orig_matchfn(lfutil.splitstandin(f)) or
                                     None)
                 return m
             oldmatch = installmatchfn(override_match)
             listpats = []
             for pat in pats:
                 if match_.patkind(pat) is not None:
                     listpats.append(pat)
                 else:
                     listpats.append(makestandin(pat))
 
             try:
                 origcopyfile = util.copyfile
                 copiedfiles = []
                 def override_copyfile(src, dest):
                     if (lfutil.shortname in src and
                         dest.startswith(repo.wjoin(lfutil.shortname))):
                         destlfile = dest.replace(lfutil.shortname, '')
                         if not opts['force'] and os.path.exists(destlfile):
                             raise IOError('',
                                 _('destination largefile already exists'))
                     copiedfiles.append((src, dest))
                     origcopyfile(src, dest)
 
                 util.copyfile = override_copyfile
                 result += orig(ui, repo, listpats, opts, rename)
             finally:
                 util.copyfile = origcopyfile
 
             lfdirstate = lfutil.openlfdirstate(ui, repo)
             for (src, dest) in copiedfiles:
                 if (lfutil.shortname in src and
                     dest.startswith(repo.wjoin(lfutil.shortname))):
                     srclfile = src.replace(repo.wjoin(lfutil.standin('')), '')
                     destlfile = dest.replace(repo.wjoin(lfutil.standin('')), '')
                     destlfiledir = os.path.dirname(destlfile) or '.'
                     if not os.path.isdir(destlfiledir):
                         os.makedirs(destlfiledir)
                     if rename:
                         os.rename(repo.wjoin(srclfile), repo.wjoin(destlfile))
                         lfdirstate.remove(srclfile)
                     else:
                         util.copyfile(srclfile, destlfile)
                     lfdirstate.add(destlfile)
             lfdirstate.write()
         except util.Abort, e:
             if str(e) != 'no files to copy':
                 raise e
             else:
                 nolfiles = True
     finally:
         restorematchfn()
         wlock.release()
 
     if nolfiles and nonormalfiles:
         raise util.Abort(_('no files to copy'))
 
     return result
 
 # When the user calls revert, we have to be careful to not revert any
 # changes to other largefiles accidentally. This means we have to keep
 # track of the largefiles that are being reverted so we only pull down
 # the necessary largefiles.
 #
 # Standins are only updated (to match the hash of largefiles) before
 # commits. Update the standins then run the original revert, changing
 # the matcher to hit standins instead of largefiles. Based on the
 # resulting standins update the largefiles. Then return the standins
 # to their proper state
 def override_revert(orig, ui, repo, *pats, **opts):
     # Because we put the standins in a bad state (by updating them)
     # and then return them to a correct state we need to lock to
     # prevent others from changing them in their incorrect state.
     wlock = repo.wlock()
     try:
         lfdirstate = lfutil.openlfdirstate(ui, repo)
         (modified, added, removed, missing, unknown, ignored, clean) = \
             lfutil.lfdirstate_status(lfdirstate, repo, repo['.'].rev())
         for lfile in modified:
             lfutil.updatestandin(repo, lfutil.standin(lfile))
 
         try:
             ctx = repo[opts.get('rev')]
             oldmatch = None # for the closure
             def override_match(ctx, pats=[], opts={}, globbed=False,
                     default='relpath'):
                 match = oldmatch(ctx, pats, opts, globbed, default)
                 m = copy.copy(match)
                 def tostandin(f):
                     if lfutil.standin(f) in ctx or lfutil.standin(f) in ctx:
                         return lfutil.standin(f)
                     elif lfutil.standin(f) in repo[None]:
                         return None
                     return f
                 m._files = [tostandin(f) for f in m._files]
                 m._files = [f for f in m._files if f is not None]
                 m._fmap = set(m._files)
                 orig_matchfn = m.matchfn
                 def matchfn(f):
                     if lfutil.isstandin(f):
                         # We need to keep track of what largefiles are being
                         # matched so we know which ones to update later --
                         # otherwise we accidentally revert changes to other
                         # largefiles. This is repo-specific, so duckpunch the
                         # repo object to keep the list of largefiles for us
                         # later.
                         if orig_matchfn(lfutil.splitstandin(f)) and \
                                 (f in repo[None] or f in ctx):
                             lfileslist = getattr(repo, '_lfilestoupdate', [])
                             lfileslist.append(lfutil.splitstandin(f))
                             repo._lfilestoupdate = lfileslist
                             return True
                         else:
                             return False
                     return orig_matchfn(f)
                 m.matchfn = matchfn
                 return m
             oldmatch = installmatchfn(override_match)
             scmutil.match
             matches = override_match(repo[None], pats, opts)
             orig(ui, repo, *pats, **opts)
         finally:
             restorematchfn()
         lfileslist = getattr(repo, '_lfilestoupdate', [])
         lfcommands.updatelfiles(ui, repo, filelist=lfileslist,
                                 printmessage=False)
 
         # empty out the largefiles list so we start fresh next time
         repo._lfilestoupdate = []
         for lfile in modified:
             if lfile in lfileslist:
                 if os.path.exists(repo.wjoin(lfutil.standin(lfile))) and lfile\
                         in repo['.']:
                     lfutil.writestandin(repo, lfutil.standin(lfile),
                         repo['.'][lfile].data().strip(),
                         'x' in repo['.'][lfile].flags())
         lfdirstate = lfutil.openlfdirstate(ui, repo)
         for lfile in added:
             standin = lfutil.standin(lfile)
             if standin not in ctx and (standin in matches or opts.get('all')):
                 if lfile in lfdirstate:
                     lfdirstate.drop(lfile)
                 util.unlinkpath(repo.wjoin(standin))
         lfdirstate.write()
     finally:
         wlock.release()
 
 def hg_update(orig, repo, node):
     result = orig(repo, node)
     # XXX check if it worked first
     lfcommands.updatelfiles(repo.ui, repo)
     return result
 
 def hg_clean(orig, repo, node, show_stats=True):
     result = orig(repo, node, show_stats)
     lfcommands.updatelfiles(repo.ui, repo)
     return result
 
 def hg_merge(orig, repo, node, force=None, remind=True):
     result = orig(repo, node, force, remind)
     lfcommands.updatelfiles(repo.ui, repo)
     return result
 
 # When we rebase a repository with remotely changed largefiles, we need to
 # take some extra care so that the largefiles are correctly updated in the
 # working copy
 def override_pull(orig, ui, repo, source=None, **opts):
     if opts.get('rebase', False):
         repo._isrebasing = True
         try:
             if opts.get('update'):
                  del opts['update']
                  ui.debug('--update and --rebase are not compatible, ignoring '
                           'the update flag\n')
             del opts['rebase']
             cmdutil.bailifchanged(repo)
             revsprepull = len(repo)
             origpostincoming = commands.postincoming
             def _dummy(*args, **kwargs):
                 pass
             commands.postincoming = _dummy
             repo.lfpullsource = source
             if not source:
                 source = 'default'
             try:
                 result = commands.pull(ui, repo, source, **opts)
             finally:
                 commands.postincoming = origpostincoming
             revspostpull = len(repo)
             if revspostpull > revsprepull:
                 result = result or rebase.rebase(ui, repo)
         finally:
             repo._isrebasing = False
     else:
         repo.lfpullsource = source
         if not source:
             source = 'default'
         result = orig(ui, repo, source, **opts)
     return result
 
 def override_rebase(orig, ui, repo, **opts):
     repo._isrebasing = True
     try:
         orig(ui, repo, **opts)
     finally:
         repo._isrebasing = False
 
 def override_archive(orig, repo, dest, node, kind, decode=True, matchfn=None,
             prefix=None, mtime=None, subrepos=None):
     # No need to lock because we are only reading history and
     # largefile caches, neither of which are modified.
     lfcommands.cachelfiles(repo.ui, repo, node)
 
     if kind not in archival.archivers:
         raise util.Abort(_("unknown archive type '%s'") % kind)
 
     ctx = repo[node]
 
     if kind == 'files':
         if prefix:
             raise util.Abort(
                 _('cannot give prefix when archiving to files'))
     else:
         prefix = archival.tidyprefix(dest, kind, prefix)
 
     def write(name, mode, islink, getdata):
         if matchfn and not matchfn(name):
             return
         data = getdata()
         if decode:
             data = repo.wwritedata(name, data)
         archiver.addfile(prefix + name, mode, islink, data)
 
     archiver = archival.archivers[kind](dest, mtime or ctx.date()[0])
 
     if repo.ui.configbool("ui", "archivemeta", True):
         def metadata():
             base = 'repo: %s\nnode: %s\nbranch: %s\n' % (
                 hex(repo.changelog.node(0)), hex(node), ctx.branch())
 
             tags = ''.join('tag: %s\n' % t for t in ctx.tags()
                            if repo.tagtype(t) == 'global')
             if not tags:
                 repo.ui.pushbuffer()
                 opts = {'template': '{latesttag}\n{latesttagdistance}',
                         'style': '', 'patch': None, 'git': None}
                 cmdutil.show_changeset(repo.ui, repo, opts).show(ctx)
                 ltags, dist = repo.ui.popbuffer().split('\n')
                 tags = ''.join('latesttag: %s\n' % t for t in ltags.split(':'))
                 tags += 'latesttagdistance: %s\n' % dist
 
             return base + tags
 
         write('.hg_archival.txt', 0644, False, metadata)
 
     for f in ctx:
         ff = ctx.flags(f)
         getdata = ctx[f].data
         if lfutil.isstandin(f):
             path = lfutil.findfile(repo, getdata().strip())
             f = lfutil.splitstandin(f)
 
             def getdatafn():
                 fd = None
                 try:
                     fd = open(path, 'rb')
                     return fd.read()
                 finally:
                     if fd:
                         fd.close()
 
             getdata = getdatafn
         write(f, 'x' in ff and 0755 or 0644, 'l' in ff, getdata)
 
     if subrepos:
         for subpath in ctx.substate:
             sub = ctx.sub(subpath)
             sub.archive(repo.ui, archiver, prefix)
 
     archiver.done()
 
 # If a largefile is modified, the change is not reflected in its
 # standin until a commit. cmdutil.bailifchanged() raises an exception
 # if the repo has uncommitted changes. Wrap it to also check if
 # largefiles were changed. This is used by bisect and backout.
 def override_bailifchanged(orig, repo):
     orig(repo)
     repo.lfstatus = True
     modified, added, removed, deleted = repo.status()[:4]
     repo.lfstatus = False
     if modified or added or removed or deleted:
         raise util.Abort(_('outstanding uncommitted changes'))
 
 # Fetch doesn't use cmdutil.bail_if_changed so override it to add the check
 def override_fetch(orig, ui, repo, *pats, **opts):
     repo.lfstatus = True
     modified, added, removed, deleted = repo.status()[:4]
     repo.lfstatus = False
     if modified or added or removed or deleted:
         raise util.Abort(_('outstanding uncommitted changes'))
     return orig(ui, repo, *pats, **opts)
 
 def override_forget(orig, ui, repo, *pats, **opts):
     installnormalfilesmatchfn(repo[None].manifest())
     orig(ui, repo, *pats, **opts)
     restorematchfn()
     m = scmutil.match(repo[None], pats, opts)
 
     try:
         repo.lfstatus = True
         s = repo.status(match=m, clean=True)
     finally:
         repo.lfstatus = False
     forget = sorted(s[0] + s[1] + s[3] + s[6])
     forget = [f for f in forget if lfutil.standin(f) in repo[None].manifest()]
 
     for f in forget:
         if lfutil.standin(f) not in repo.dirstate and not \
                 os.path.isdir(m.rel(lfutil.standin(f))):
             ui.warn(_('not removing %s: file is already untracked\n')
                     % m.rel(f))
 
     for f in forget:
         if ui.verbose or not m.exact(f):
             ui.status(_('removing %s\n') % m.rel(f))
 
     # Need to lock because standin files are deleted then removed from the
     # repository and we could race inbetween.
     wlock = repo.wlock()
     try:
         lfdirstate = lfutil.openlfdirstate(ui, repo)
         for f in forget:
             if lfdirstate[f] == 'a':
                 lfdirstate.drop(f)
             else:
                 lfdirstate.remove(f)
         lfdirstate.write()
         lfutil.repo_remove(repo, [lfutil.standin(f) for f in forget],
             unlink=True)
     finally:
         wlock.release()
 
 def getoutgoinglfiles(ui, repo, dest=None, **opts):
     dest = ui.expandpath(dest or 'default-push', dest or 'default')
     dest, branches = hg.parseurl(dest, opts.get('branch'))
     revs, checkout = hg.addbranchrevs(repo, repo, branches, opts.get('rev'))
     if revs:
         revs = [repo.lookup(rev) for rev in revs]
 
     remoteui = hg.remoteui
 
     try:
         remote = hg.repository(remoteui(repo, opts), dest)
     except error.RepoError:
         return None
     o = lfutil.findoutgoing(repo, remote, False)
     if not o:
         return None
     o = repo.changelog.nodesbetween(o, revs)[0]
     if opts.get('newest_first'):
         o.reverse()
 
     toupload = set()
     for n in o:
         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)
         toupload = toupload.union(
             set([f for f in files if lfutil.isstandin(f) and f in ctx]))
     return toupload
 
 def override_outgoing(orig, ui, repo, dest=None, **opts):
     orig(ui, repo, dest, **opts)
 
     if opts.pop('large', None):
         toupload = getoutgoinglfiles(ui, repo, dest, **opts)
         if toupload is None:
             ui.status(_('largefiles: No remote repo\n'))
         else:
             ui.status(_('largefiles to upload:\n'))
             for file in toupload:
                 ui.status(lfutil.splitstandin(file) + '\n')
             ui.status('\n')
 
 def override_summary(orig, ui, repo, *pats, **opts):
     orig(ui, repo, *pats, **opts)
 
     if opts.pop('large', None):
         toupload = getoutgoinglfiles(ui, repo, None, **opts)
         if toupload is None:
             ui.status(_('largefiles: No remote repo\n'))
         else:
             ui.status(_('largefiles: %d to upload\n') % len(toupload))
 
 def override_addremove(orig, ui, repo, *pats, **opts):
     # Check if the parent or child has largefiles; if so, disallow
     # addremove. If there is a symlink in the manifest then getting
     # the manifest throws an exception: catch it and let addremove
     # deal with it.
     try:
         manifesttip = set(repo['tip'].manifest())
     except util.Abort:
         manifesttip = set()
     try:
         manifestworking = set(repo[None].manifest())
     except util.Abort:
         manifestworking = set()
 
     # Manifests are only iterable so turn them into sets then union
     for file in manifesttip.union(manifestworking):
         if file.startswith(lfutil.shortname):
             raise util.Abort(
                 _('addremove cannot be run on a repo with largefiles'))
 
     return orig(ui, repo, *pats, **opts)
 
 # Calling purge with --all will cause the largefiles to be deleted.
 # Override repo.status to prevent this from happening.
 def override_purge(orig, ui, repo, *dirs, **opts):
     oldstatus = repo.status
     def override_status(node1='.', node2=None, match=None, ignored=False,
                         clean=False, unknown=False, listsubrepos=False):
         r = oldstatus(node1, node2, match, ignored, clean, unknown,
                       listsubrepos)
         lfdirstate = lfutil.openlfdirstate(ui, repo)
         modified, added, removed, deleted, unknown, ignored, clean = r
         unknown = [f for f in unknown if lfdirstate[f] == '?']
         ignored = [f for f in ignored if lfdirstate[f] == '?']
         return modified, added, removed, deleted, unknown, ignored, clean
     repo.status = override_status
     orig(ui, repo, *dirs, **opts)
     repo.status = oldstatus
 
 def override_rollback(orig, ui, repo, **opts):
     result = orig(ui, repo, **opts)
     merge.update(repo, node=None, branchmerge=False, force=True,
         partial=lfutil.isstandin)
     lfdirstate = lfutil.openlfdirstate(ui, repo)
     lfiles = lfutil.listlfiles(repo)
     oldlfiles = lfutil.listlfiles(repo, repo[None].parents()[0].rev())
     for file in lfiles:
         if file in oldlfiles:
             lfdirstate.normallookup(file)
         else:
             lfdirstate.add(file)
     lfdirstate.write()
     return result
 
 def override_transplant(orig, ui, repo, *revs, **opts):
     result = orig(ui, repo, *revs, **opts)
     lfcommands.updatelfiles(repo.ui, repo)
     return result

hgext/largefiles/reposetup.py

@@ -1,437 +1,447
 # 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 types
 import os
 
 from mercurial import context, error, manifest, match as match_, node, util
 from mercurial.i18n import _
 
 import lfcommands
 import proto
 import lfutil
 
 def reposetup(ui, repo):
     # wire repositories should be given new wireproto functions but not the
     # other largefiles modifications
     if not repo.local():
         return proto.wirereposetup(ui, repo)
 
     for name in ('status', 'commitctx', 'commit', 'push'):
         method = getattr(repo, name)
         #if not (isinstance(method, types.MethodType) and
         #        method.im_func is repo.__class__.commitctx.im_func):
         if (isinstance(method, types.FunctionType) and
             method.func_name == 'wrap'):
             ui.warn(_('largefiles: repo method %r appears to have already been'
                     ' wrapped by another extension: '
                     'largefiles may behave incorrectly\n')
                     % name)
 
     class lfiles_repo(repo.__class__):
         lfstatus = False
         def status_nolfiles(self, *args, **kwargs):
             return super(lfiles_repo, 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(lfiles_repo, self).__getitem__(changeid)
             if self.lfstatus:
                 class lfiles_manifestdict(manifest.manifestdict):
                     def __contains__(self, filename):
                         if super(lfiles_manifestdict,
                                 self).__contains__(filename):
                             return True
                         return super(lfiles_manifestdict,
                             self).__contains__(lfutil.standin(filename))
                 class lfiles_ctx(ctx.__class__):
                     def files(self):
                         filenames = super(lfiles_ctx, self).files()
                         return [lfutil.splitstandin(f) or f for f in filenames]
                     def manifest(self):
                         man1 = super(lfiles_ctx, self).manifest()
                         man1.__class__ = lfiles_manifestdict
                         return man1
                     def filectx(self, path, fileid=None, filelog=None):
                         try:
                             result = super(lfiles_ctx, self).filectx(path,
                                 fileid, filelog)
                         except error.LookupError:
                             # Adding a null character will cause Mercurial to
                             # identify this as a binary file.
                             result = super(lfiles_ctx, self).filectx(
                                 lfutil.standin(path), fileid, filelog)
                             olddata = result.data
                             result.data = lambda: olddata() + '\0'
                         return result
                 ctx.__class__ = lfiles_ctx
             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.
         def status(self, node1='.', node2=None, match=None, ignored=False,
                 clean=False, unknown=False, listsubrepos=False):
             listignored, listclean, listunknown = ignored, clean, unknown
             if not self.lfstatus:
                 return super(lfiles_repo, self).status(node1, node2, match,
                     listignored, listclean, listunknown, listsubrepos)
             else:
                 # some calls in this function rely on the old version of status
                 self.lfstatus = False
                 if isinstance(node1, context.changectx):
                     ctx1 = node1
                 else:
                     ctx1 = repo[node1]
                 if isinstance(node2, context.changectx):
                     ctx2 = node2
                 else:
                     ctx2 = repo[node2]
                 working = ctx2.rev() is None
                 parentworking = working and ctx1 == self['.']
 
                 def inctx(file, ctx):
                     try:
                         if ctx.rev() is None:
                             return file in ctx.manifest()
                         ctx[file]
                         return True
                     except KeyError:
                         return False
 
                 if match is None:
                     match = match_.always(self.root, self.getcwd())
 
                 # First check if there were files specified on the
                 # command line.  If there were, and none of them were
                 # largefiles, we should just bail here and let super
                 # handle it -- thus gaining a big performance boost.
                 lfdirstate = lfutil.openlfdirstate(ui, self)
                 if match.files() and not match.anypats():
                     matchedfiles = [f for f in match.files() if f in lfdirstate]
                     if not matchedfiles:
                         return super(lfiles_repo, self).status(node1, node2,
                                 match, listignored, listclean,
                                 listunknown, listsubrepos)
 
                 # Create a copy of match that matches standins instead
                 # of largefiles.
                 def tostandin(file):
                     if inctx(lfutil.standin(file), ctx2):
                         return lfutil.standin(file)
                     return file
 
                 # Create a function that we can use to override what is
                 # normally the ignore matcher.  We've already checked
                 # for ignored files on the first dirstate walk, and
                 # unecessarily re-checking here causes a huge performance
                 # hit because lfdirstate only knows about largefiles
                 def _ignoreoverride(self):
                     return False
 
                 m = copy.copy(match)
                 m._files = [tostandin(f) for f in m._files]
 
                 # Get ignored files here even if we weren't asked for them; we
                 # must use the result here for filtering later
                 result = super(lfiles_repo, self).status(node1, node2, m,
                     True, clean, unknown, listsubrepos)
                 if working:
                     # hold the wlock while we read largefiles and
                     # update the lfdirstate
                     wlock = repo.wlock()
                     try:
                         # Any non-largefiles that were explicitly listed must be
                         # taken out or lfdirstate.status will report an error.
                         # The status of these files was already computed using
                         # super's status.
                         # Override lfdirstate's ignore matcher to not do
                         # anything
                         orig_ignore = lfdirstate._ignore
                         lfdirstate._ignore = _ignoreoverride
 
                         match._files = [f for f in match._files if f in
                             lfdirstate]
                         # Don't waste time getting the ignored and unknown
                         # files again; we already have them
                         s = lfdirstate.status(match, [], False,
                                 listclean, False)
                         (unsure, modified, added, removed, missing, unknown,
                                 ignored, clean) = s
                         # Replace the list of ignored and unknown files with
                         # the previously caclulated lists, and strip out the
                         # largefiles
                         lfiles = set(lfdirstate._map)
                         ignored = set(result[5]).difference(lfiles)
                         unknown = set(result[4]).difference(lfiles)
                         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:
                                     clean.append(lfile)
                                     lfdirstate.normal(lfile)
                             lfdirstate.write()
                         else:
                             tocheck = unsure + modified + added + clean
                             modified, added, clean = [], [], []
 
                             for lfile in tocheck:
                                 standin = lfutil.standin(lfile)
                                 if inctx(standin, ctx1):
                                     if ctx1[standin].data().strip() != \
                                             lfutil.hashfile(self.wjoin(lfile)):
                                         modified.append(lfile)
                                     else:
                                         clean.append(lfile)
                                 else:
                                     added.append(lfile)
                         # Replace the original ignore function
                         lfdirstate._ignore = orig_ignore
                     finally:
                         wlock.release()
 
                     for standin in ctx1.manifest():
                         if not lfutil.isstandin(standin):
                             continue
                         lfile = lfutil.splitstandin(standin)
                         if not match(lfile):
                             continue
                         if lfile not in lfdirstate:
                             removed.append(lfile)
-                    # Handle unknown and ignored differently
-                    lfiles = (modified, added, removed, missing, [], [], clean)
+
+                    # 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 it's still in
+                    # lfdirstate. This happens in merges where files
+                    # change type.
+                    removed = [f for f in removed if f not in repo.dirstate]
+                    result[2] = [f for f in result[2] if f not in lfdirstate]
+
                     # Unknown files
                     unknown = set(unknown).difference(ignored)
                     result[4] = [f for f in unknown
                                  if (repo.dirstate[f] == '?' and
                                      not lfutil.isstandin(f))]
                     # Ignored files were calculated earlier by the dirstate,
                     # and we already stripped out the largefiles from the list
                     result[5] = ignored
                     # combine normal files and largefiles
                     normals = [[fn for fn in filelist
                                 if not lfutil.isstandin(fn)]
                                for filelist in result]
+                    lfiles = (modified, added, removed, missing, [], [], clean)
                     result = [sorted(list1 + list2)
                               for (list1, list2) in zip(normals, lfiles)]
                 else:
                     def toname(f):
                         if lfutil.isstandin(f):
                             return lfutil.splitstandin(f)
                         return f
                     result = [[toname(f) for f in items] for items in result]
 
                 if not listunknown:
                     result[4] = []
                 if not listignored:
                     result[5] = []
                 if not listclean:
                     result[6] = []
                 self.lfstatus = True
                 return result
 
         # As part of committing, copy all of the largefiles into the
         # cache.
         def commitctx(self, *args, **kwargs):
             node = super(lfiles_repo, self).commitctx(*args, **kwargs)
             ctx = self[node]
             for filename in ctx.files():
                 if lfutil.isstandin(filename) and filename in ctx.manifest():
                     realfile = lfutil.splitstandin(filename)
                     lfutil.copytostore(self, ctx.node(), realfile)
 
             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(lfiles_repo, self).commit
 
             wlock = repo.wlock()
             try:
                 if getattr(repo, "_isrebasing", False):
                     # We have to take the time to pull down the new
                     # largefiles now. Otherwise if we are rebasing,
                     # any largefiles that were modified in the
                     # destination changesets get overwritten, either
                     # by the rebase or in the first commit after the
                     # rebase.
                     lfcommands.updatelfiles(repo.ui, repo)
                 # 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
                     (not match.anypats() and not match.files())):
                     # 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(repo.root, repo.getcwd())
                     s = lfdirstate.status(dirtymatch, [], False, False, False)
                     modifiedfiles = []
                     for i in s:
                         modifiedfiles.extend(i)
                     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))
                                     lfdirstate.normal(lfile)
                     for lfile in lfdirstate:
                         if lfile in modifiedfiles:
                             if not os.path.exists(
                                     repo.wjoin(lfutil.standin(lfile))):
                                 lfdirstate.drop(lfile)
                     lfdirstate.write()
 
                     return orig(text=text, user=user, date=date, match=match,
                                     force=force, editor=editor, extra=extra)
 
                 for f in match.files():
                     if lfutil.isstandin(f):
                         raise util.Abort(
                             _('file "%s" is a largefile standin') % f,
                             hint=('commit the largefile itself instead'))
 
                 # Case 2: user calls commit with specified patterns: refresh
                 # any matching big files.
                 smatcher = lfutil.composestandinmatcher(self, match)
                 standins = lfutil.dirstate_walk(self.dirstate, smatcher)
 
                 # 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)
                         lfdirstate.normal(lfile)
                     else:
                         lfdirstate.drop(lfile)
                 lfdirstate.write()
 
                 # 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.
                 lfiles = lfutil.listlfiles(repo)
                 match = copy.copy(match)
                 orig_matchfn = 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
 
                     # append directory separator to avoid collisions
                     if not fstandin.endswith(os.sep):
                         fstandin += os.sep
 
                     # prevalidate matching standin directories
                     if util.any(st for st in match._files
                                    if st.startswith(fstandin)):
                         continue
                     actualfiles.append(f)
                 match._files = actualfiles
 
                 def matchfn(f):
                     if orig_matchfn(f):
                         return f not in lfiles
                     else:
                         return f in standins
 
                 match.matchfn = matchfn
                 return orig(text=text, user=user, date=date, match=match,
                                 force=force, editor=editor, extra=extra)
             finally:
                 wlock.release()
 
         def push(self, remote, force=False, revs=None, newbranch=False):
             o = lfutil.findoutgoing(repo, remote, force)
             if o:
                 toupload = set()
                 o = repo.changelog.nodesbetween(o, revs)[0]
                 for n in o:
                     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)
 
                     toupload = toupload.union(
                         set([ctx[f].data().strip()
                              for f in files
                              if lfutil.isstandin(f) and f in ctx]))
                 lfcommands.uploadlfiles(ui, self, remote, toupload)
             return super(lfiles_repo, self).push(remote, force, revs,
                 newbranch)
 
     repo.__class__ = lfiles_repo
 
     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)
     ui.setconfig('hooks', 'commit.lfiles', checkrequireslfiles)

hgext/largefiles/uisetup.py

@@ -1,138 +1,142
 # 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 extension: uisetup'''
 
 from mercurial import archival, cmdutil, commands, extensions, filemerge, hg, \
-    httprepo, localrepo, sshrepo, sshserver, wireproto
+    httprepo, localrepo, merge, sshrepo, sshserver, wireproto
 from mercurial.i18n import _
 from mercurial.hgweb import hgweb_mod, protocol
 
 import overrides
 import proto
 
 def uisetup(ui):
     # Disable auto-status for some commands which assume that all
     # files in the result are under Mercurial's control
 
     entry = extensions.wrapcommand(commands.table, 'add',
                                    overrides.override_add)
     addopt = [('', 'large', None, _('add as largefile')),
               ('', 'lfsize', '', _('add all files above this size '
                                    '(in megabytes) as largefiles '
                                    '(default: 10)'))]
     entry[1].extend(addopt)
 
     entry = extensions.wrapcommand(commands.table, 'addremove',
             overrides.override_addremove)
     entry = extensions.wrapcommand(commands.table, 'remove',
                                    overrides.override_remove)
     entry = extensions.wrapcommand(commands.table, 'forget',
                                    overrides.override_forget)
     entry = extensions.wrapcommand(commands.table, 'status',
                                    overrides.override_status)
     entry = extensions.wrapcommand(commands.table, 'log',
                                    overrides.override_log)
     entry = extensions.wrapcommand(commands.table, 'rollback',
                                    overrides.override_rollback)
     entry = extensions.wrapcommand(commands.table, 'verify',
                                    overrides.override_verify)
 
     verifyopt = [('', 'large', None, _('verify largefiles')),
                  ('', 'lfa', None,
                      _('verify all revisions of largefiles not just current')),
                  ('', 'lfc', None,
                      _('verify largefile contents not just existence'))]
     entry[1].extend(verifyopt)
 
     entry = extensions.wrapcommand(commands.table, 'outgoing',
         overrides.override_outgoing)
     outgoingopt = [('', 'large', None, _('display outgoing largefiles'))]
     entry[1].extend(outgoingopt)
     entry = extensions.wrapcommand(commands.table, 'summary',
                                    overrides.override_summary)
     summaryopt = [('', 'large', None, _('display outgoing largefiles'))]
     entry[1].extend(summaryopt)
 
     entry = extensions.wrapcommand(commands.table, 'update',
                                    overrides.override_update)
     entry = extensions.wrapcommand(commands.table, 'pull',
                                    overrides.override_pull)
+    entry = extensions.wrapfunction(merge, '_checkunknown',
+                                    overrides.override_checkunknown)
+    entry = extensions.wrapfunction(merge, 'manifestmerge',
+                                    overrides.override_manifestmerge)
     entry = extensions.wrapfunction(filemerge, 'filemerge',
                                     overrides.override_filemerge)
     entry = extensions.wrapfunction(cmdutil, 'copy',
                                     overrides.override_copy)
 
     # Backout calls revert so we need to override both the command and the
     # function
     entry = extensions.wrapcommand(commands.table, 'revert',
                                    overrides.override_revert)
     entry = extensions.wrapfunction(commands, 'revert',
                                     overrides.override_revert)
 
     # clone uses hg._update instead of hg.update even though they are the
     # same function... so wrap both of them)
     extensions.wrapfunction(hg, 'update', overrides.hg_update)
     extensions.wrapfunction(hg, '_update', overrides.hg_update)
     extensions.wrapfunction(hg, 'clean', overrides.hg_clean)
     extensions.wrapfunction(hg, 'merge', overrides.hg_merge)
 
     extensions.wrapfunction(archival, 'archive', overrides.override_archive)
     extensions.wrapfunction(cmdutil, 'bailifchanged',
                             overrides.override_bailifchanged)
 
     # create the new wireproto commands ...
     wireproto.commands['putlfile'] = (proto.putlfile, 'sha')
     wireproto.commands['getlfile'] = (proto.getlfile, 'sha')
     wireproto.commands['statlfile'] = (proto.statlfile, 'sha')
 
     # ... and wrap some existing ones
     wireproto.commands['capabilities'] = (proto.capabilities, '')
     wireproto.commands['heads'] = (proto.heads, '')
     wireproto.commands['lheads'] = (wireproto.heads, '')
 
     # make putlfile behave the same as push and {get,stat}lfile behave
     # the same as pull w.r.t. permissions checks
     hgweb_mod.perms['putlfile'] = 'push'
     hgweb_mod.perms['getlfile'] = 'pull'
     hgweb_mod.perms['statlfile'] = 'pull'
 
     # the hello wireproto command uses wireproto.capabilities, so it won't see
     # our largefiles capability unless we replace the actual function as well.
     proto.capabilities_orig = wireproto.capabilities
     wireproto.capabilities = proto.capabilities
 
     # these let us reject non-largefiles clients and make them display
     # our error messages
     protocol.webproto.refuseclient = proto.webproto_refuseclient
     sshserver.sshserver.refuseclient = proto.sshproto_refuseclient
 
     # can't do this in reposetup because it needs to have happened before
     # wirerepo.__init__ is called
     proto.ssh_oldcallstream = sshrepo.sshrepository._callstream
     proto.http_oldcallstream = httprepo.httprepository._callstream
     sshrepo.sshrepository._callstream = proto.sshrepo_callstream
     httprepo.httprepository._callstream = proto.httprepo_callstream
 
     # don't die on seeing a repo with the largefiles requirement
     localrepo.localrepository.supported |= set(['largefiles'])
 
     # override some extensions' stuff as well
     for name, module in extensions.extensions():
         if name == 'fetch':
             extensions.wrapcommand(getattr(module, 'cmdtable'), 'fetch',
                 overrides.override_fetch)
         if name == 'purge':
             extensions.wrapcommand(getattr(module, 'cmdtable'), 'purge',
                 overrides.override_purge)
         if name == 'rebase':
             extensions.wrapcommand(getattr(module, 'cmdtable'), 'rebase',
                 overrides.override_rebase)
         if name == 'transplant':
             extensions.wrapcommand(getattr(module, 'cmdtable'), 'transplant',
                 overrides.override_transplant)

hgext/progress.py

@@ -1,302 +1,306
 # progress.py show progress bars for some actions
 #
 # Copyright (C) 2010 Augie Fackler <[e-mail]>
 #
 # This program is free software; you can redistribute it and/or modify it
 # under the terms of the GNU General Public License as published by the
 # Free Software Foundation; either version 2 of the License, or (at your
 # option) any later version.
 #
 # This program is distributed in the hope that it will be useful, but
 # WITHOUT ANY WARRANTY; without even the implied warranty of
 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General
 # Public License for more details.
 #
 # You should have received a copy of the GNU General Public License along
 # with this program; if not, write to the Free Software Foundation, Inc.,
 # 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
 
 """show progress bars for some actions
 
 This extension uses the progress information logged by hg commands
 to draw progress bars that are as informative as possible. Some progress
 bars only offer indeterminate information, while others have a definite
 end point.
 
 The following settings are available::
 
   [progress]
   delay = 3 # number of seconds (float) before showing the progress bar
   changedelay = 1 # changedelay: minimum delay before showing a new topic.
                   # If set to less than 3 * refresh, that value will
                   # be used instead.
   refresh = 0.1 # time in seconds between refreshes of the progress bar
   format = topic bar number estimate # format of the progress bar
   width = <none> # if set, the maximum width of the progress information
                  # (that is, min(width, term width) will be used)
   clear-complete = True # clear the progress bar after it's done
   disable = False # if true, don't show a progress bar
   assume-tty = False # if true, ALWAYS show a progress bar, unless
                      # disable is given
 
 Valid entries for the format field are topic, bar, number, unit,
 estimate, speed, and item. item defaults to the last 20 characters of
 the item, but this can be changed by adding either ``-<num>`` which
 would take the last num characters, or ``+<num>`` for the first num
 characters.
 """
 
 import sys
 import time
 
 from mercurial import util
 from mercurial.i18n import _
 
 def spacejoin(*args):
     return ' '.join(s for s in args if s)
 
 def shouldprint(ui):
     return util.isatty(sys.stderr) or ui.configbool('progress', 'assume-tty')
 
 def fmtremaining(seconds):
     if seconds < 60:
         # i18n: format XX seconds as "XXs"
         return _("%02ds") % (seconds)
     minutes = seconds // 60
     if minutes < 60:
         seconds -= minutes * 60
         # i18n: format X minutes and YY seconds as "XmYYs"
         return _("%dm%02ds") % (minutes, seconds)
     # we're going to ignore seconds in this case
     minutes += 1
     hours = minutes // 60
     minutes -= hours * 60
     if hours < 30:
         # i18n: format X hours and YY minutes as "XhYYm"
         return _("%dh%02dm") % (hours, minutes)
     # we're going to ignore minutes in this case
     hours += 1
     days = hours // 24
     hours -= days * 24
     if days < 15:
         # i18n: format X days and YY hours as "XdYYh"
         return _("%dd%02dh") % (days, hours)
     # we're going to ignore hours in this case
     days += 1
     weeks = days // 7
     days -= weeks * 7
     if weeks < 55:
         # i18n: format X weeks and YY days as "XwYYd"
         return _("%dw%02dd") % (weeks, days)
     # we're going to ignore days and treat a year as 52 weeks
     weeks += 1
     years = weeks // 52
     weeks -= years * 52
     # i18n: format X years and YY weeks as "XyYYw"
     return _("%dy%02dw") % (years, weeks)
 
 class progbar(object):
     def __init__(self, ui):
         self.ui = ui
         self.resetstate()
 
     def resetstate(self):
         self.topics = []
         self.topicstates = {}
         self.starttimes = {}
         self.startvals = {}
         self.printed = False
         self.lastprint = time.time() + float(self.ui.config(
             'progress', 'delay', default=3))
         self.lasttopic = None
         self.indetcount = 0
         self.refresh = float(self.ui.config(
             'progress', 'refresh', default=0.1))
         self.changedelay = max(3 * self.refresh,
                                float(self.ui.config(
                                    'progress', 'changedelay', default=1)))
         self.order = self.ui.configlist(
             'progress', 'format',
             default=['topic', 'bar', 'number', 'estimate'])
 
     def show(self, now, topic, pos, item, unit, total):
         if not shouldprint(self.ui):
             return
         termwidth = self.width()
         self.printed = True
         head = ''
         needprogress = False
         tail = ''
         for indicator in self.order:
             add = ''
             if indicator == 'topic':
                 add = topic
             elif indicator == 'number':
                 if total:
                     add = ('% ' + str(len(str(total))) +
                            's/%s') % (pos, total)
                 else:
                     add = str(pos)
             elif indicator.startswith('item') and item:
                 slice = 'end'
                 if '-' in indicator:
                     wid = int(indicator.split('-')[1])
                 elif '+' in indicator:
                     slice = 'beginning'
                     wid = int(indicator.split('+')[1])
                 else:
                     wid = 20
                 if slice == 'end':
                     add = item[-wid:]
                 else:
                     add = item[:wid]
                 add += (wid - len(add)) * ' '
             elif indicator == 'bar':
                 add = ''
                 needprogress = True
             elif indicator == 'unit' and unit:
                 add = unit
             elif indicator == 'estimate':
                 add = self.estimate(topic, pos, total, now)
             elif indicator == 'speed':
                 add = self.speed(topic, pos, unit, now)
             if not needprogress:
                 head = spacejoin(head, add)
             else:
                 tail = spacejoin(tail, add)
         if needprogress:
             used = 0
             if head:
                 used += len(head) + 1
             if tail:
                 used += len(tail) + 1
             progwidth = termwidth - used - 3
             if total and pos <= total:
                 amt = pos * progwidth // total
                 bar = '=' * (amt - 1)
                 if amt > 0:
                     bar += '>'
                 bar += ' ' * (progwidth - amt)
             else:
                 progwidth -= 3
                 self.indetcount += 1
                 # mod the count by twice the width so we can make the
                 # cursor bounce between the right and left sides
                 amt = self.indetcount % (2 * progwidth)
                 amt -= progwidth
                 bar = (' ' * int(progwidth - abs(amt)) + '<=>' +
                        ' ' * int(abs(amt)))
             prog = ''.join(('[', bar , ']'))
             out = spacejoin(head, prog, tail)
         else:
             out = spacejoin(head, tail)
         sys.stderr.write('\r' + out[:termwidth])
         self.lasttopic = topic
         sys.stderr.flush()
 
     def clear(self):
         if not shouldprint(self.ui):
             return
         sys.stderr.write('\r%s\r' % (' ' * self.width()))
 
     def complete(self):
         if not shouldprint(self.ui):
             return
         if self.ui.configbool('progress', 'clear-complete', default=True):
             self.clear()
         else:
             sys.stderr.write('\n')
         sys.stderr.flush()
 
     def width(self):
         tw = self.ui.termwidth()
         return min(int(self.ui.config('progress', 'width', default=tw)), tw)
 
     def estimate(self, topic, pos, total, now):
         if total is None:
             return ''
         initialpos = self.startvals[topic]
         target = total - initialpos
         delta = pos - initialpos
         if delta > 0:
             elapsed = now - self.starttimes[topic]
             if elapsed > float(
                 self.ui.config('progress', 'estimate', default=2)):
                 seconds = (elapsed * (target - delta)) // delta + 1
                 return fmtremaining(seconds)
         return ''
 
     def speed(self, topic, pos, unit, now):
         initialpos = self.startvals[topic]
         delta = pos - initialpos
         elapsed = now - self.starttimes[topic]
         if elapsed > float(
             self.ui.config('progress', 'estimate', default=2)):
             return _('%d %s/sec') % (delta / elapsed, unit)
         return ''
 
     def progress(self, topic, pos, item='', unit='', total=None):
         now = time.time()
         if pos is None:
             self.starttimes.pop(topic, None)
             self.startvals.pop(topic, None)
             self.topicstates.pop(topic, None)
             # reset the progress bar if this is the outermost topic
             if self.topics and self.topics[0] == topic and self.printed:
                 self.complete()
                 self.resetstate()
             # truncate the list of topics assuming all topics within
             # this one are also closed
             if topic in self.topics:
               self.topics = self.topics[:self.topics.index(topic)]
         else:
             if topic not in self.topics:
                 self.starttimes[topic] = now
                 self.startvals[topic] = pos
                 self.topics.append(topic)
             self.topicstates[topic] = pos, item, unit, total
             if now - self.lastprint >= self.refresh and self.topics:
                 if (self.lasttopic is None # first time we printed
                     # not a topic change
                     or topic == self.lasttopic
                     # it's been long enough we should print anyway
                     or now - self.lastprint >= self.changedelay):
                     self.lastprint = now
                     self.show(now, topic, *self.topicstates[topic])
 
 _singleton = None
 
 def uisetup(ui):
     global _singleton
     class progressui(ui.__class__):
         _progbar = None
 
+        def _quiet(self):
+            return self.debugflag or self.quiet
+
         def progress(self, *args, **opts):
-            self._progbar.progress(*args, **opts)
+            if not self._quiet():
+                self._progbar.progress(*args, **opts)
             return super(progressui, self).progress(*args, **opts)
 
         def write(self, *args, **opts):
-            if self._progbar.printed:
+            if not self._quiet() and self._progbar.printed:
                 self._progbar.clear()
             return super(progressui, self).write(*args, **opts)
 
         def write_err(self, *args, **opts):
-            if self._progbar.printed:
+            if not self._quiet() and self._progbar.printed:
                 self._progbar.clear()
             return super(progressui, self).write_err(*args, **opts)
 
     # Apps that derive a class from ui.ui() can use
     # setconfig('progress', 'disable', 'True') to disable this extension
     if ui.configbool('progress', 'disable'):
         return
     if shouldprint(ui) and not ui.debugflag and not ui.quiet:
         ui.__class__ = progressui
         # we instantiate one globally shared progress bar to avoid
         # competing progress bars when multiple UI objects get created
         if not progressui._progbar:
             if _singleton is None:
                 _singleton = progbar(ui)
             progressui._progbar = _singleton
 
 def reposetup(ui, repo):
     uisetup(repo.ui)

hgext/win32mbcs.py

@@ -1,160 +1,160
 # win32mbcs.py -- MBCS filename support for Mercurial
 #
 # Copyright (c) 2008 Shun-ichi Goto <[e-mail]>
 #
 # Version: 0.3
 # Author:  Shun-ichi Goto <[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.
 #
 
 '''allow the use of MBCS paths with problematic encodings
 
 Some MBCS encodings are not good for some path operations (i.e.
 splitting path, case conversion, etc.) with its encoded bytes. We call
 such a encoding (i.e. shift_jis and big5) as "problematic encoding".
 This extension can be used to fix the issue with those encodings by
 wrapping some functions to convert to Unicode string before path
 operation.
 
 This extension is useful for:
 
 - Japanese Windows users using shift_jis encoding.
 - Chinese Windows users using big5 encoding.
 - All users who use a repository with one of problematic encodings on
   case-insensitive file system.
 
 This extension is not needed for:
 
 - Any user who use only ASCII chars in path.
 - Any user who do not use any of problematic encodings.
 
 Note that there are some limitations on using this extension:
 
 - You should use single encoding in one repository.
 - If the repository path ends with 0x5c, .hg/hgrc cannot be read.
 - win32mbcs is not compatible with fixutf8 extension.
 
 By default, win32mbcs uses encoding.encoding decided by Mercurial.
 You can specify the encoding by config option::
 
  [win32mbcs]
  encoding = sjis
 
 It is useful for the users who want to commit with UTF-8 log message.
 '''
 
 import os, sys
 from mercurial.i18n import _
 from mercurial import util, encoding
 
 _encoding = None                                # see extsetup
 
 def decode(arg):
     if isinstance(arg, str):
         uarg = arg.decode(_encoding)
         if arg == uarg.encode(_encoding):
             return uarg
         raise UnicodeError("Not local encoding")
     elif isinstance(arg, tuple):
         return tuple(map(decode, arg))
     elif isinstance(arg, list):
         return map(decode, arg)
     elif isinstance(arg, dict):
         for k, v in arg.items():
             arg[k] = decode(v)
     return arg
 
 def encode(arg):
     if isinstance(arg, unicode):
         return arg.encode(_encoding)
     elif isinstance(arg, tuple):
         return tuple(map(encode, arg))
     elif isinstance(arg, list):
         return map(encode, arg)
     elif isinstance(arg, dict):
         for k, v in arg.items():
             arg[k] = encode(v)
     return arg
 
 def appendsep(s):
     # ensure the path ends with os.sep, appending it if necessary.
     try:
         us = decode(s)
     except UnicodeError:
         us = s
     if us and us[-1] not in ':/\\':
         s += os.sep
     return s
 
 def wrapper(func, args, kwds):
     # check argument is unicode, then call original
     for arg in args:
         if isinstance(arg, unicode):
             return func(*args, **kwds)
 
     try:
         # convert arguments to unicode, call func, then convert back
         return encode(func(*decode(args), **decode(kwds)))
     except UnicodeError:
         raise util.Abort(_("[win32mbcs] filename conversion failed with"
                          " %s encoding\n") % (_encoding))
 
 def wrapperforlistdir(func, args, kwds):
     # Ensure 'path' argument ends with os.sep to avoids
     # misinterpreting last 0x5c of MBCS 2nd byte as path separator.
     if args:
         args = list(args)
         args[0] = appendsep(args[0])
     if 'path' in kwds:
         kwds['path'] = appendsep(kwds['path'])
     return func(*args, **kwds)
 
 def wrapname(name, wrapper):
     module, name = name.rsplit('.', 1)
     module = sys.modules[module]
     func = getattr(module, name)
     def f(*args, **kwds):
         return wrapper(func, args, kwds)
     try:
         f.__name__ = func.__name__                # fail with python23
     except Exception:
         pass
     setattr(module, name, f)
 
 # List of functions to be wrapped.
 # NOTE: os.path.dirname() and os.path.basename() are safe because
 #       they use result of os.path.split()
 funcs = '''os.path.join os.path.split os.path.splitext
- os.path.splitunc os.path.normpath os.path.normcase os.makedirs
+ os.path.splitunc os.path.normpath os.makedirs
  mercurial.util.endswithsep mercurial.util.splitpath mercurial.util.checkcase
  mercurial.util.fspath mercurial.util.pconvert mercurial.util.normpath
  mercurial.util.checkwinfilename mercurial.util.checkosfilename'''
 
 # codec and alias names of sjis and big5 to be faked.
 problematic_encodings = '''big5 big5-tw csbig5 big5hkscs big5-hkscs
  hkscs cp932 932 ms932 mskanji ms-kanji shift_jis csshiftjis shiftjis
  sjis s_jis shift_jis_2004 shiftjis2004 sjis_2004 sjis2004
  shift_jisx0213 shiftjisx0213 sjisx0213 s_jisx0213 950 cp950 ms950 '''
 
 def extsetup(ui):
     # TODO: decide use of config section for this extension
     if not os.path.supports_unicode_filenames:
         ui.warn(_("[win32mbcs] cannot activate on this platform.\n"))
         return
     # determine encoding for filename
     global _encoding
     _encoding = ui.config('win32mbcs', 'encoding', encoding.encoding)
     # fake is only for relevant environment.
     if _encoding.lower() in problematic_encodings.split():
         for f in funcs.split():
             wrapname(f, wrapper)
         wrapname("mercurial.osutil.listdir", wrapperforlistdir)
         # Check sys.args manually instead of using ui.debug() because
         # command line options is not yet applied when
         # extensions.loadall() is called.
         if '--debug' in sys.argv:
             ui.write("[win32mbcs] activated with encoding: %s\n"
                      % _encoding)
 

mercurial/changelog.py

@@ -1,244 +1,255
 # changelog.py - changelog class for mercurial
 #
 # Copyright 2005-2007 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from node import bin, hex, nullid
 from i18n import _
 import util, error, revlog, encoding
 
 def _string_escape(text):
     """
     >>> d = {'nl': chr(10), 'bs': chr(92), 'cr': chr(13), 'nul': chr(0)}
     >>> s = "ab%(nl)scd%(bs)s%(bs)sn%(nul)sab%(cr)scd%(bs)s%(nl)s" % d
     >>> s
     'ab\\ncd\\\\\\\\n\\x00ab\\rcd\\\\\\n'
     >>> res = _string_escape(s)
     >>> s == res.decode('string_escape')
     True
     """
     # subset of the string_escape codec
     text = text.replace('\\', '\\\\').replace('\n', '\\n').replace('\r', '\\r')
     return text.replace('\0', '\\0')
 
 def decodeextra(text):
+    """
+    >>> decodeextra(encodeextra({'foo': 'bar', 'baz': chr(0) + '2'}))
+    {'foo': 'bar', 'baz': '\\x002'}
+    >>> decodeextra(encodeextra({'foo': 'bar', 'baz': chr(92) + chr(0) + '2'}))
+    {'foo': 'bar', 'baz': '\\\\\\x002'}
+    """
     extra = {}
     for l in text.split('\0'):
         if l:
+            if '\\0' in l:
+                # fix up \0 without getting into trouble with \\0
+                l = l.replace('\\\\', '\\\\\n')
+                l = l.replace('\\0', '\0')
+                l = l.replace('\n', '')
             k, v = l.decode('string_escape').split(':', 1)
             extra[k] = v
     return extra
 
 def encodeextra(d):
     # keys must be sorted to produce a deterministic changelog entry
     items = [_string_escape('%s:%s' % (k, d[k])) for k in sorted(d)]
     return "\0".join(items)
 
 class appender(object):
     '''the changelog index must be updated last on disk, so we use this class
     to delay writes to it'''
     def __init__(self, fp, buf):
         self.data = buf
         self.fp = fp
         self.offset = fp.tell()
         self.size = util.fstat(fp).st_size
 
     def end(self):
         return self.size + len("".join(self.data))
     def tell(self):
         return self.offset
     def flush(self):
         pass
     def close(self):
         self.fp.close()
 
     def seek(self, offset, whence=0):
         '''virtual file offset spans real file and data'''
         if whence == 0:
             self.offset = offset
         elif whence == 1:
             self.offset += offset
         elif whence == 2:
             self.offset = self.end() + offset
         if self.offset < self.size:
             self.fp.seek(self.offset)
 
     def read(self, count=-1):
         '''only trick here is reads that span real file and data'''
         ret = ""
         if self.offset < self.size:
             s = self.fp.read(count)
             ret = s
             self.offset += len(s)
             if count > 0:
                 count -= len(s)
         if count != 0:
             doff = self.offset - self.size
             self.data.insert(0, "".join(self.data))
             del self.data[1:]
             s = self.data[0][doff:doff + count]
             self.offset += len(s)
             ret += s
         return ret
 
     def write(self, s):
         self.data.append(str(s))
         self.offset += len(s)
 
 def delayopener(opener, target, divert, buf):
     def o(name, mode='r'):
         if name != target:
             return opener(name, mode)
         if divert:
             return opener(name + ".a", mode.replace('a', 'w'))
         # otherwise, divert to memory
         return appender(opener(name, mode), buf)
     return o
 
 class changelog(revlog.revlog):
     def __init__(self, opener):
         revlog.revlog.__init__(self, opener, "00changelog.i")
         if self._initempty:
             # changelogs don't benefit from generaldelta
             self.version &= ~revlog.REVLOGGENERALDELTA
             self._generaldelta = False
         self._realopener = opener
         self._delayed = False
         self._divert = False
         # hiddenrevs: revs that should be hidden by command and tools
         self.hiddenrevs = set()
 
     def delayupdate(self):
         "delay visibility of index updates to other readers"
         self._delayed = True
         self._divert = (len(self) == 0)
         self._delaybuf = []
         self.opener = delayopener(self._realopener, self.indexfile,
                                   self._divert, self._delaybuf)
 
     def finalize(self, tr):
         "finalize index updates"
         self._delayed = False
         self.opener = self._realopener
         # move redirected index data back into place
         if self._divert:
             nfile = self.opener(self.indexfile + ".a")
             n = nfile.name
             nfile.close()
             util.rename(n, n[:-2])
         elif self._delaybuf:
             fp = self.opener(self.indexfile, 'a')
             fp.write("".join(self._delaybuf))
             fp.close()
             self._delaybuf = []
         # split when we're done
         self.checkinlinesize(tr)
 
     def readpending(self, file):
         r = revlog.revlog(self.opener, file)
         self.index = r.index
         self.nodemap = r.nodemap
         self._chunkcache = r._chunkcache
 
     def writepending(self):
         "create a file containing the unfinalized state for pretxnchangegroup"
         if self._delaybuf:
             # make a temporary copy of the index
             fp1 = self._realopener(self.indexfile)
             fp2 = self._realopener(self.indexfile + ".a", "w")
             fp2.write(fp1.read())
             # add pending data
             fp2.write("".join(self._delaybuf))
             fp2.close()
             # switch modes so finalize can simply rename
             self._delaybuf = []
             self._divert = True
 
         if self._divert:
             return True
 
         return False
 
     def checkinlinesize(self, tr, fp=None):
         if not self._delayed:
             revlog.revlog.checkinlinesize(self, tr, fp)
 
     def read(self, node):
         """
         format used:
         nodeid\n        : manifest node in ascii
         user\n          : user, no \n or \r allowed
         time tz extra\n : date (time is int or float, timezone is int)
                         : extra is metadatas, encoded and separated by '\0'
                         : older versions ignore it
         files\n\n       : files modified by the cset, no \n or \r allowed
         (.*)            : comment (free text, ideally utf-8)
 
         changelog v0 doesn't use extra
         """
         text = self.revision(node)
         if not text:
             return (nullid, "", (0, 0), [], "", {'branch': 'default'})
         last = text.index("\n\n")
         desc = encoding.tolocal(text[last + 2:])
         l = text[:last].split('\n')
         manifest = bin(l[0])
         user = encoding.tolocal(l[1])
 
         extra_data = l[2].split(' ', 2)
         if len(extra_data) != 3:
             time = float(extra_data.pop(0))
             try:
                 # various tools did silly things with the time zone field.
                 timezone = int(extra_data[0])
             except ValueError:
                 timezone = 0
             extra = {}
         else:
             time, timezone, extra = extra_data
             time, timezone = float(time), int(timezone)
             extra = decodeextra(extra)
         if not extra.get('branch'):
             extra['branch'] = 'default'
         files = l[3:]
         return (manifest, user, (time, timezone), files, desc, extra)
 
     def add(self, manifest, files, desc, transaction, p1, p2,
                   user, date=None, extra=None):
         # Convert to UTF-8 encoded bytestrings as the very first
         # thing: calling any method on a localstr object will turn it
         # into a str object and the cached UTF-8 string is thus lost.
         user, desc = encoding.fromlocal(user), encoding.fromlocal(desc)
 
         user = user.strip()
         # An empty username or a username with a "\n" will make the
         # revision text contain two "\n\n" sequences -> corrupt
         # repository since read cannot unpack the revision.
         if not user:
             raise error.RevlogError(_("empty username"))
         if "\n" in user:
             raise error.RevlogError(_("username %s contains a newline")
                                     % repr(user))
 
         # strip trailing whitespace and leading and trailing empty lines
         desc = '\n'.join([l.rstrip() for l in desc.splitlines()]).strip('\n')
 
         if date:
             parseddate = "%d %d" % util.parsedate(date)
         else:
             parseddate = "%d %d" % util.makedate()
         if extra:
             branch = extra.get("branch")
             if branch in ("default", ""):
                 del extra["branch"]
             elif branch in (".", "null", "tip"):
                 raise error.RevlogError(_('the name \'%s\' is reserved')
                                         % branch)
         if extra:
             extra = encodeextra(extra)
             parseddate = "%s %s" % (parseddate, extra)
         l = [hex(manifest), user, parseddate] + sorted(files) + ["", desc]
         text = "\n".join(l)
         return self.addrevision(text, transaction, len(self), p1, p2)

mercurial/dirstate.py

@@ -1,725 +1,730
 # 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 _normroot(self):
+        return util.normcase(self._root)
+
+    @propertycache
     def _foldmap(self):
         f = {}
         for name in self._map:
             f[util.normcase(name)] = name
+        f['.'] = '.' # prevents useless util.fspath() invocation
         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, 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:
             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
         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'''
         if f in self._map:
             self._dirty = True
             self._droppath(f)
             del self._map[f]
 
     def _normalize(self, path, isknown):
         normed = util.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))
+                                util.fspath(normed, self._normroot))
         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)

mercurial/encoding.py

@@ -1,173 +1,192
 # encoding.py - character transcoding support for Mercurial
 #
 #  Copyright 2005-2009 Matt Mackall <[e-mail]> and others
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 import error
 import unicodedata, locale, os
 
 def _getpreferredencoding():
     '''
     On darwin, getpreferredencoding ignores the locale environment and
     always returns mac-roman. http://bugs.python.org/issue6202 fixes this
     for Python 2.7 and up. This is the same corrected code for earlier
     Python versions.
 
     However, we can't use a version check for this method, as some distributions
     patch Python to fix this. Instead, we use it as a 'fixer' for the mac-roman
     encoding, as it is unlikely that this encoding is the actually expected.
     '''
     try:
         locale.CODESET
     except AttributeError:
         # Fall back to parsing environment variables :-(
         return locale.getdefaultlocale()[1]
 
     oldloc = locale.setlocale(locale.LC_CTYPE)
     locale.setlocale(locale.LC_CTYPE, "")
     result = locale.nl_langinfo(locale.CODESET)
     locale.setlocale(locale.LC_CTYPE, oldloc)
 
     return result
 
 _encodingfixers = {
     '646': lambda: 'ascii',
     'ANSI_X3.4-1968': lambda: 'ascii',
     'mac-roman': _getpreferredencoding
 }
 
 try:
     encoding = os.environ.get("HGENCODING")
     if not encoding:
         encoding = locale.getpreferredencoding() or 'ascii'
         encoding = _encodingfixers.get(encoding, lambda: encoding)()
 except locale.Error:
     encoding = 'ascii'
 encodingmode = os.environ.get("HGENCODINGMODE", "strict")
 fallbackencoding = 'ISO-8859-1'
 
 class localstr(str):
     '''This class allows strings that are unmodified to be
     round-tripped to the local encoding and back'''
     def __new__(cls, u, l):
         s = str.__new__(cls, l)
         s._utf8 = u
         return s
     def __hash__(self):
         return hash(self._utf8) # avoid collisions in local string space
 
 def tolocal(s):
     """
     Convert a string from internal UTF-8 to local encoding
 
     All internal strings should be UTF-8 but some repos before the
     implementation of locale support may contain latin1 or possibly
     other character sets. We attempt to decode everything strictly
     using UTF-8, then Latin-1, and failing that, we use UTF-8 and
     replace unknown characters.
 
     The localstr class is used to cache the known UTF-8 encoding of
     strings next to their local representation to allow lossless
     round-trip conversion back to UTF-8.
 
     >>> u = 'foo: \\xc3\\xa4' # utf-8
     >>> l = tolocal(u)
     >>> l
     'foo: ?'
     >>> fromlocal(l)
     'foo: \\xc3\\xa4'
     >>> u2 = 'foo: \\xc3\\xa1'
     >>> d = { l: 1, tolocal(u2): 2 }
     >>> d # no collision
     {'foo: ?': 1, 'foo: ?': 2}
     >>> 'foo: ?' in d
     False
     >>> l1 = 'foo: \\xe4' # historical latin1 fallback
     >>> l = tolocal(l1)
     >>> l
     'foo: ?'
     >>> fromlocal(l) # magically in utf-8
     'foo: \\xc3\\xa4'
     """
 
     for e in ('UTF-8', fallbackencoding):
         try:
             u = s.decode(e) # attempt strict decoding
             r = u.encode(encoding, "replace")
             if u == r.decode(encoding):
                 # r is a safe, non-lossy encoding of s
                 return r
             elif e == 'UTF-8':
                 return localstr(s, r)
             else:
                 return localstr(u.encode('UTF-8'), r)
 
         except LookupError, k:
             raise error.Abort("%s, please check your locale settings" % k)
         except UnicodeDecodeError:
             pass
     u = s.decode("utf-8", "replace") # last ditch
     return u.encode(encoding, "replace") # can't round-trip
 
 def fromlocal(s):
     """
     Convert a string from the local character encoding to UTF-8
 
     We attempt to decode strings using the encoding mode set by
     HGENCODINGMODE, which defaults to 'strict'. In this mode, unknown
     characters will cause an error message. Other modes include
     'replace', which replaces unknown characters with a special
     Unicode character, and 'ignore', which drops the character.
     """
 
     # can we do a lossless round-trip?
     if isinstance(s, localstr):
         return s._utf8
 
     try:
         return s.decode(encoding, encodingmode).encode("utf-8")
     except UnicodeDecodeError, inst:
         sub = s[max(0, inst.start - 10):inst.start + 10]
         raise error.Abort("decoding near '%s': %s!" % (sub, inst))
     except LookupError, k:
         raise error.Abort("%s, please check your locale settings" % k)
 
 # How to treat ambiguous-width characters. Set to 'wide' to treat as wide.
 wide = (os.environ.get("HGENCODINGAMBIGUOUS", "narrow") == "wide"
         and "WFA" or "WF")
 
 def colwidth(s):
     "Find the column width of a string for display in the local encoding"
     return ucolwidth(s.decode(encoding, 'replace'))
 
 def ucolwidth(d):
     "Find the column width of a Unicode string for display"
     eaw = getattr(unicodedata, 'east_asian_width', None)
     if eaw is not None:
         return sum([eaw(c) in wide and 2 or 1 for c in d])
     return len(d)
 
 def getcols(s, start, c):
     '''Use colwidth to find a c-column substring of s starting at byte
     index start'''
     for x in xrange(start + c, len(s)):
         t = s[start:x]
         if colwidth(t) == c:
             return t
 
 def lower(s):
     "best-effort encoding-aware case-folding of local string s"
     try:
         if isinstance(s, localstr):
             u = s._utf8.decode("utf-8")
         else:
             u = s.decode(encoding, encodingmode)
 
         lu = u.lower()
         if u == lu:
             return s # preserve localstring
         return lu.encode(encoding)
     except UnicodeError:
         return s.lower() # we don't know how to fold this except in ASCII
+    except LookupError, k:
+        raise error.Abort(k, hint="please check your locale settings")
+
+def upper(s):
+    "best-effort encoding-aware case-folding of local string s"
+    try:
+        if isinstance(s, localstr):
+            u = s._utf8.decode("utf-8")
+        else:
+            u = s.decode(encoding, encodingmode)
+
+        uu = u.upper()
+        if u == uu:
+            return s # preserve localstring
+        return uu.encode(encoding)
+    except UnicodeError:
+        return s.upper() # we don't know how to fold this except in ASCII
+    except LookupError, k:
+        raise error.Abort(k, hint="please check your locale settings")

mercurial/merge.py

@@ -1,576 +1,584
 # merge.py - directory-level update/merge handling 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, hex, bin
 from i18n import _
 import scmutil, util, filemerge, copies, subrepo
 import errno, os, shutil
 
 class mergestate(object):
     '''track 3-way merge state of individual files'''
     def __init__(self, repo):
         self._repo = repo
         self._dirty = False
         self._read()
     def reset(self, node=None):
         self._state = {}
         if node:
             self._local = node
         shutil.rmtree(self._repo.join("merge"), True)
         self._dirty = False
     def _read(self):
         self._state = {}
         try:
             f = self._repo.opener("merge/state")
             for i, l in enumerate(f):
                 if i == 0:
                     self._local = bin(l[:-1])
                 else:
                     bits = l[:-1].split("\0")
                     self._state[bits[0]] = bits[1:]
             f.close()
         except IOError, err:
             if err.errno != errno.ENOENT:
                 raise
         self._dirty = False
     def commit(self):
         if self._dirty:
             f = self._repo.opener("merge/state", "w")
             f.write(hex(self._local) + "\n")
             for d, v in self._state.iteritems():
                 f.write("\0".join([d] + v) + "\n")
             f.close()
             self._dirty = False
     def add(self, fcl, fco, fca, fd, flags):
         hash = util.sha1(fcl.path()).hexdigest()
         self._repo.opener.write("merge/" + hash, fcl.data())
         self._state[fd] = ['u', hash, fcl.path(), fca.path(),
                            hex(fca.filenode()), fco.path(), flags]
         self._dirty = True
     def __contains__(self, dfile):
         return dfile in self._state
     def __getitem__(self, dfile):
         return self._state[dfile][0]
     def __iter__(self):
         l = self._state.keys()
         l.sort()
         for f in l:
             yield f
     def mark(self, dfile, state):
         self._state[dfile][0] = state
         self._dirty = True
     def resolve(self, dfile, wctx, octx):
         if self[dfile] == 'r':
             return 0
         state, hash, lfile, afile, anode, ofile, flags = self._state[dfile]
         f = self._repo.opener("merge/" + hash)
         self._repo.wwrite(dfile, f.read(), flags)
         f.close()
         fcd = wctx[dfile]
         fco = octx[ofile]
         fca = self._repo.filectx(afile, fileid=anode)
         r = filemerge.filemerge(self._repo, self._local, lfile, fcd, fco, fca)
         if r is None:
             # no real conflict
             del self._state[dfile]
         elif not r:
             self.mark(dfile, 'r')
         return r
 
 def _checkunknown(wctx, mctx, folding):
     "check for collisions between unknown files and files in mctx"
     if folding:
         foldf = util.normcase
     else:
         foldf = lambda fn: fn
     folded = {}
     for fn in mctx:
         folded[foldf(fn)] = fn
     for fn in wctx.unknown():
         f = foldf(fn)
         if f in folded and mctx[folded[f]].cmp(wctx[f]):
             raise util.Abort(_("untracked file in working directory differs"
                                " from file in requested revision: '%s'") % fn)
 
-def _checkcollision(mctx):
+def _checkcollision(mctx, wctx):
     "check for case folding collisions in the destination context"
     folded = {}
     for fn in mctx:
         fold = util.normcase(fn)
         if fold in folded:
             raise util.Abort(_("case-folding collision between %s and %s")
                              % (fn, folded[fold]))
         folded[fold] = fn
 
+    if wctx:
+        for fn in wctx:
+            fold = util.normcase(fn)
+            mfn = folded.get(fold, None)
+            if mfn and (mfn != fn):
+                raise util.Abort(_("case-folding collision between %s and %s")
+                                 % (mfn, fn))
+
 def _forgetremoved(wctx, mctx, branchmerge):
     """
     Forget removed files
 
     If we're jumping between revisions (as opposed to merging), and if
     neither the working directory nor the target rev has the file,
     then we need to remove it from the dirstate, to prevent the
     dirstate from listing the file when it is no longer in the
     manifest.
 
     If we're merging, and the other revision has removed a file
     that is not present in the working directory, we need to mark it
     as removed.
     """
 
     action = []
     state = branchmerge and 'r' or 'f'
     for f in wctx.deleted():
         if f not in mctx:
             action.append((f, state))
 
     if not branchmerge:
         for f in wctx.removed():
             if f not in mctx:
                 action.append((f, "f"))
 
     return action
 
 def manifestmerge(repo, p1, p2, pa, overwrite, partial):
     """
     Merge p1 and p2 with ancestor pa and generate merge action list
 
     overwrite = whether we clobber working files
     partial = function to filter file lists
     """
 
     def fmerge(f, f2, fa):
         """merge flags"""
         a, m, n = ma.flags(fa), m1.flags(f), m2.flags(f2)
         if m == n: # flags agree
             return m # unchanged
         if m and n and not a: # flags set, don't agree, differ from parent
             r = repo.ui.promptchoice(
                 _(" conflicting flags for %s\n"
                   "(n)one, e(x)ec or sym(l)ink?") % f,
                 (_("&None"), _("E&xec"), _("Sym&link")), 0)
             if r == 1:
                 return "x" # Exec
             if r == 2:
                 return "l" # Symlink
             return ""
         if m and m != a: # changed from a to m
             return m
         if n and n != a: # changed from a to n
             return n
         return '' # flag was cleared
 
     def act(msg, m, f, *args):
         repo.ui.debug(" %s: %s -> %s\n" % (f, msg, m))
         action.append((f, m) + args)
 
     action, copy = [], {}
 
     if overwrite:
         pa = p1
     elif pa == p2: # backwards
         pa = p1.p1()
     elif pa and repo.ui.configbool("merge", "followcopies", True):
         dirs = repo.ui.configbool("merge", "followdirs", True)
         copy, diverge = copies.copies(repo, p1, p2, pa, dirs)
         for of, fl in diverge.iteritems():
             act("divergent renames", "dr", of, fl)
 
     repo.ui.note(_("resolving manifests\n"))
     repo.ui.debug(" overwrite: %s, partial: %s\n"
                   % (bool(overwrite), bool(partial)))
     repo.ui.debug(" ancestor: %s, local: %s, remote: %s\n" % (pa, p1, p2))
 
     m1, m2, ma = p1.manifest(), p2.manifest(), pa.manifest()
     copied = set(copy.values())
 
     if '.hgsubstate' in m1:
         # check whether sub state is modified
         for s in p1.substate:
             if p1.sub(s).dirty():
                 m1['.hgsubstate'] += "+"
                 break
 
     # Compare manifests
     for f, n in m1.iteritems():
         if partial and not partial(f):
             continue
         if f in m2:
             rflags = fmerge(f, f, f)
             a = ma.get(f, nullid)
             if n == m2[f] or m2[f] == a: # same or local newer
                 # is file locally modified or flags need changing?
                 # dirstate flags may need to be made current
                 if m1.flags(f) != rflags or n[20:]:
                     act("update permissions", "e", f, rflags)
             elif n == a: # remote newer
                 act("remote is newer", "g", f, rflags)
             else: # both changed
                 act("versions differ", "m", f, f, f, rflags, False)
         elif f in copied: # files we'll deal with on m2 side
             pass
         elif f in copy:
             f2 = copy[f]
             if f2 not in m2: # directory rename
                 act("remote renamed directory to " + f2, "d",
                     f, None, f2, m1.flags(f))
             else: # case 2 A,B/B/B or case 4,21 A/B/B
                 act("local copied/moved to " + f2, "m",
                     f, f2, f, fmerge(f, f2, f2), False)
         elif f in ma: # clean, a different, no remote
             if n != ma[f]:
                 if repo.ui.promptchoice(
                     _(" local changed %s which remote deleted\n"
                       "use (c)hanged version or (d)elete?") % f,
                     (_("&Changed"), _("&Delete")), 0):
                     act("prompt delete", "r", f)
                 else:
                     act("prompt keep", "a", f)
             elif n[20:] == "a": # added, no remote
                 act("remote deleted", "f", f)
             elif n[20:] != "u":
                 act("other deleted", "r", f)
 
     for f, n in m2.iteritems():
         if partial and not partial(f):
             continue
         if f in m1 or f in copied: # files already visited
             continue
         if f in copy:
             f2 = copy[f]
             if f2 not in m1: # directory rename
                 act("local renamed directory to " + f2, "d",
                     None, f, f2, m2.flags(f))
             elif f2 in m2: # rename case 1, A/A,B/A
                 act("remote copied to " + f, "m",
                     f2, f, f, fmerge(f2, f, f2), False)
             else: # case 3,20 A/B/A
                 act("remote moved to " + f, "m",
                     f2, f, f, fmerge(f2, f, f2), True)
         elif f not in ma:
             act("remote created", "g", f, m2.flags(f))
         elif n != ma[f]:
             if repo.ui.promptchoice(
                 _("remote changed %s which local deleted\n"
                   "use (c)hanged version or leave (d)eleted?") % f,
                 (_("&Changed"), _("&Deleted")), 0) == 0:
                 act("prompt recreating", "g", f, m2.flags(f))
 
     return action
 
 def actionkey(a):
     return a[1] == 'r' and -1 or 0, a
 
 def applyupdates(repo, action, wctx, mctx, actx, overwrite):
     """apply the merge action list to the working directory
 
     wctx is the working copy context
     mctx is the context to be merged into the working copy
     actx is the context of the common ancestor
 
     Return a tuple of counts (updated, merged, removed, unresolved) that
     describes how many files were affected by the update.
     """
 
     updated, merged, removed, unresolved = 0, 0, 0, 0
     ms = mergestate(repo)
     ms.reset(wctx.p1().node())
     moves = []
     action.sort(key=actionkey)
 
     # prescan for merges
     for a in action:
         f, m = a[:2]
         if m == 'm': # merge
             f2, fd, flags, move = a[2:]
             if f == '.hgsubstate': # merged internally
                 continue
             repo.ui.debug("preserving %s for resolve of %s\n" % (f, fd))
             fcl = wctx[f]
             fco = mctx[f2]
             if mctx == actx: # backwards, use working dir parent as ancestor
                 if fcl.parents():
                     fca = fcl.p1()
                 else:
                     fca = repo.filectx(f, fileid=nullrev)
             else:
                 fca = fcl.ancestor(fco, actx)
             if not fca:
                 fca = repo.filectx(f, fileid=nullrev)
             ms.add(fcl, fco, fca, fd, flags)
             if f != fd and move:
                 moves.append(f)
 
     audit = scmutil.pathauditor(repo.root)
 
     # remove renamed files after safely stored
     for f in moves:
         if os.path.lexists(repo.wjoin(f)):
             repo.ui.debug("removing %s\n" % f)
             audit(f)
             os.unlink(repo.wjoin(f))
 
     numupdates = len(action)
     for i, a in enumerate(action):
         f, m = a[:2]
         repo.ui.progress(_('updating'), i + 1, item=f, total=numupdates,
                          unit=_('files'))
         if f and f[0] == "/":
             continue
         if m == "r": # remove
             repo.ui.note(_("removing %s\n") % f)
             audit(f)
             if f == '.hgsubstate': # subrepo states need updating
                 subrepo.submerge(repo, wctx, mctx, wctx, overwrite)
             try:
                 util.unlinkpath(repo.wjoin(f))
             except OSError, inst:
                 if inst.errno != errno.ENOENT:
                     repo.ui.warn(_("update failed to remove %s: %s!\n") %
                                  (f, inst.strerror))
             removed += 1
         elif m == "m": # merge
             if f == '.hgsubstate': # subrepo states need updating
                 subrepo.submerge(repo, wctx, mctx, wctx.ancestor(mctx), overwrite)
                 continue
             f2, fd, flags, move = a[2:]
             repo.wopener.audit(fd)
             r = ms.resolve(fd, wctx, mctx)
             if r is not None and r > 0:
                 unresolved += 1
             else:
                 if r is None:
                     updated += 1
                 else:
                     merged += 1
             util.setflags(repo.wjoin(fd), 'l' in flags, 'x' in flags)
             if (move and repo.dirstate.normalize(fd) != f
                 and os.path.lexists(repo.wjoin(f))):
                 repo.ui.debug("removing %s\n" % f)
                 audit(f)
                 os.unlink(repo.wjoin(f))
         elif m == "g": # get
             flags = a[2]
             repo.ui.note(_("getting %s\n") % f)
             t = mctx.filectx(f).data()
             repo.wwrite(f, t, flags)
             t = None
             updated += 1
             if f == '.hgsubstate': # subrepo states need updating
                 subrepo.submerge(repo, wctx, mctx, wctx, overwrite)
         elif m == "d": # directory rename
             f2, fd, flags = a[2:]
             if f:
                 repo.ui.note(_("moving %s to %s\n") % (f, fd))
                 audit(f)
                 t = wctx.filectx(f).data()
                 repo.wwrite(fd, t, flags)
                 util.unlinkpath(repo.wjoin(f))
             if f2:
                 repo.ui.note(_("getting %s to %s\n") % (f2, fd))
                 t = mctx.filectx(f2).data()
                 repo.wwrite(fd, t, flags)
             updated += 1
         elif m == "dr": # divergent renames
             fl = a[2]
             repo.ui.warn(_("note: possible conflict - %s was renamed "
                            "multiple times to:\n") % f)
             for nf in fl:
                 repo.ui.warn(" %s\n" % nf)
         elif m == "e": # exec
             flags = a[2]
             repo.wopener.audit(f)
             util.setflags(repo.wjoin(f), 'l' in flags, 'x' in flags)
     ms.commit()
     repo.ui.progress(_('updating'), None, total=numupdates, unit=_('files'))
 
     return updated, merged, removed, unresolved
 
 def recordupdates(repo, action, branchmerge):
     "record merge actions to the dirstate"
 
     for a in action:
         f, m = a[:2]
         if m == "r": # remove
             if branchmerge:
                 repo.dirstate.remove(f)
             else:
                 repo.dirstate.drop(f)
         elif m == "a": # re-add
             if not branchmerge:
                 repo.dirstate.add(f)
         elif m == "f": # forget
             repo.dirstate.drop(f)
         elif m == "e": # exec change
             repo.dirstate.normallookup(f)
         elif m == "g": # get
             if branchmerge:
                 repo.dirstate.otherparent(f)
             else:
                 repo.dirstate.normal(f)
         elif m == "m": # merge
             f2, fd, flag, move = a[2:]
             if branchmerge:
                 # We've done a branch merge, mark this file as merged
                 # so that we properly record the merger later
                 repo.dirstate.merge(fd)
                 if f != f2: # copy/rename
                     if move:
                         repo.dirstate.remove(f)
                     if f != fd:
                         repo.dirstate.copy(f, fd)
                     else:
                         repo.dirstate.copy(f2, fd)
             else:
                 # We've update-merged a locally modified file, so
                 # we set the dirstate to emulate a normal checkout
                 # of that file some time in the past. Thus our
                 # merge will appear as a normal local file
                 # modification.
                 if f2 == fd: # file not locally copied/moved
                     repo.dirstate.normallookup(fd)
                 if move:
                     repo.dirstate.drop(f)
         elif m == "d": # directory rename
             f2, fd, flag = a[2:]
             if not f2 and f not in repo.dirstate:
                 # untracked file moved
                 continue
             if branchmerge:
                 repo.dirstate.add(fd)
                 if f:
                     repo.dirstate.remove(f)
                     repo.dirstate.copy(f, fd)
                 if f2:
                     repo.dirstate.copy(f2, fd)
             else:
                 repo.dirstate.normal(fd)
                 if f:
                     repo.dirstate.drop(f)
 
 def update(repo, node, branchmerge, force, partial, ancestor=None):
     """
     Perform a merge between the working directory and the given node
 
     node = the node to update to, or None if unspecified
     branchmerge = whether to merge between branches
     force = whether to force branch merging or file overwriting
     partial = a function to filter file lists (dirstate not updated)
 
     The table below shows all the behaviors of the update command
     given the -c and -C or no options, whether the working directory
     is dirty, whether a revision is specified, and the relationship of
     the parent rev to the target rev (linear, on the same named
     branch, or on another named branch).
 
     This logic is tested by test-update-branches.t.
 
     -c  -C  dirty  rev  |  linear   same  cross
      n   n    n     n   |    ok     (1)     x
      n   n    n     y   |    ok     ok     ok
      n   n    y     *   |   merge   (2)    (2)
      n   y    *     *   |    ---  discard  ---
      y   n    y     *   |    ---    (3)    ---
      y   n    n     *   |    ---    ok     ---
      y   y    *     *   |    ---    (4)    ---
 
     x = can't happen
     * = don't-care
     1 = abort: crosses branches (use 'hg merge' or 'hg update -c')
     2 = abort: crosses branches (use 'hg merge' to merge or
                  use 'hg update -C' to discard changes)
     3 = abort: uncommitted local changes
     4 = incompatible options (checked in commands.py)
 
     Return the same tuple as applyupdates().
     """
 
     onode = node
     wlock = repo.wlock()
     try:
         wc = repo[None]
         if node is None:
             # tip of current branch
             try:
                 node = repo.branchtags()[wc.branch()]
             except KeyError:
                 if wc.branch() == "default": # no default branch!
                     node = repo.lookup("tip") # update to tip
                 else:
                     raise util.Abort(_("branch %s not found") % wc.branch())
         overwrite = force and not branchmerge
         pl = wc.parents()
         p1, p2 = pl[0], repo[node]
         if ancestor:
             pa = repo[ancestor]
         else:
             pa = p1.ancestor(p2)
 
         fp1, fp2, xp1, xp2 = p1.node(), p2.node(), str(p1), str(p2)
 
         ### check phase
         if not overwrite and len(pl) > 1:
             raise util.Abort(_("outstanding uncommitted merges"))
         if branchmerge:
             if pa == p2:
                 raise util.Abort(_("merging with a working directory ancestor"
                                    " has no effect"))
             elif pa == p1:
                 if p1.branch() == p2.branch():
                     raise util.Abort(_("nothing to merge"),
                                      hint=_("use 'hg update' "
                                             "or check 'hg heads'"))
             if not force and (wc.files() or wc.deleted()):
                 raise util.Abort(_("outstanding uncommitted changes"),
                                  hint=_("use 'hg status' to list changes"))
             for s in wc.substate:
                 if wc.sub(s).dirty():
                     raise util.Abort(_("outstanding uncommitted changes in "
                                        "subrepository '%s'") % s)
 
         elif not overwrite:
             if pa == p1 or pa == p2: # linear
                 pass # all good
             elif wc.dirty(missing=True):
                 raise util.Abort(_("crosses branches (merge branches or use"
                                    " --clean to discard changes)"))
             elif onode is None:
                 raise util.Abort(_("crosses branches (merge branches or update"
                                    " --check to force update)"))
             else:
                 # Allow jumping branches if clean and specific rev given
                 overwrite = True
 
         ### calculate phase
         action = []
         wc.status(unknown=True) # prime cache
         folding = not util.checkcase(repo.path)
         if not force:
             _checkunknown(wc, p2, folding)
         if folding:
-            _checkcollision(p2)
+            _checkcollision(p2, branchmerge and p1)
         action += _forgetremoved(wc, p2, branchmerge)
         action += manifestmerge(repo, wc, p2, pa, overwrite, partial)
 
         ### apply phase
         if not branchmerge: # just jump to the new rev
             fp1, fp2, xp1, xp2 = fp2, nullid, xp2, ''
         if not partial:
             repo.hook('preupdate', throw=True, parent1=xp1, parent2=xp2)
 
         stats = applyupdates(repo, action, wc, p2, pa, overwrite)
 
         if not partial:
             repo.dirstate.setparents(fp1, fp2)
             recordupdates(repo, action, branchmerge)
             if not branchmerge:
                 repo.dirstate.setbranch(p2.branch())
     finally:
         wlock.release()
 
     if not partial:
         repo.hook('update', parent1=xp1, parent2=xp2, error=stats[3])
     return stats

mercurial/posix.py

@@ -1,416 +1,419
 # posix.py - Posix utility function implementations for Mercurial
 #
 #  Copyright 2005-2009 Matt Mackall <[e-mail]> and others
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from i18n import _
 import os, sys, errno, stat, getpass, pwd, grp, tempfile, unicodedata
 
 posixfile = open
 nulldev = '/dev/null'
 normpath = os.path.normpath
 samestat = os.path.samestat
 oslink = os.link
 unlink = os.unlink
 rename = os.rename
 expandglobs = False
 
 umask = os.umask(0)
 os.umask(umask)
 
 def openhardlinks():
     '''return true if it is safe to hold open file handles to hardlinks'''
     return True
 
 def nlinks(name):
     '''return number of hardlinks for the given file'''
     return os.lstat(name).st_nlink
 
 def parsepatchoutput(output_line):
     """parses the output produced by patch and returns the filename"""
     pf = output_line[14:]
     if os.sys.platform == 'OpenVMS':
         if pf[0] == '`':
             pf = pf[1:-1] # Remove the quotes
     else:
         if pf.startswith("'") and pf.endswith("'") and " " in pf:
             pf = pf[1:-1] # Remove the quotes
     return pf
 
 def sshargs(sshcmd, host, user, port):
     '''Build argument list for ssh'''
     args = user and ("%s@%s" % (user, host)) or host
     return port and ("%s -p %s" % (args, port)) or args
 
 def isexec(f):
     """check whether a file is executable"""
     return (os.lstat(f).st_mode & 0100 != 0)
 
 def setflags(f, l, x):
     s = os.lstat(f).st_mode
     if l:
         if not stat.S_ISLNK(s):
             # switch file to link
             fp = open(f)
             data = fp.read()
             fp.close()
             os.unlink(f)
             try:
                 os.symlink(data, f)
             except OSError:
                 # failed to make a link, rewrite file
                 fp = open(f, "w")
                 fp.write(data)
                 fp.close()
         # no chmod needed at this point
         return
     if stat.S_ISLNK(s):
         # switch link to file
         data = os.readlink(f)
         os.unlink(f)
         fp = open(f, "w")
         fp.write(data)
         fp.close()
         s = 0666 & ~umask # avoid restatting for chmod
 
     sx = s & 0100
     if x and not sx:
         # Turn on +x for every +r bit when making a file executable
         # and obey umask.
         os.chmod(f, s | (s & 0444) >> 2 & ~umask)
     elif not x and sx:
         # Turn off all +x bits
         os.chmod(f, s & 0666)
 
 def copymode(src, dst, mode=None):
     '''Copy the file mode from the file at path src to dst.
     If src doesn't exist, we're using mode instead. If mode is None, we're
     using umask.'''
     try:
         st_mode = os.lstat(src).st_mode & 0777
     except OSError, inst:
         if inst.errno != errno.ENOENT:
             raise
         st_mode = mode
         if st_mode is None:
             st_mode = ~umask
         st_mode &= 0666
     os.chmod(dst, st_mode)
 
 def checkexec(path):
     """
     Check whether the given path is on a filesystem with UNIX-like exec flags
 
     Requires a directory (like /foo/.hg)
     """
 
     # VFAT on some Linux versions can flip mode but it doesn't persist
     # a FS remount. Frequently we can detect it if files are created
     # with exec bit on.
 
     try:
         EXECFLAGS = stat.S_IXUSR | stat.S_IXGRP | stat.S_IXOTH
         fh, fn = tempfile.mkstemp(dir=path, prefix='hg-checkexec-')
         try:
             os.close(fh)
             m = os.stat(fn).st_mode & 0777
             new_file_has_exec = m & EXECFLAGS
             os.chmod(fn, m ^ EXECFLAGS)
             exec_flags_cannot_flip = ((os.stat(fn).st_mode & 0777) == m)
         finally:
             os.unlink(fn)
     except (IOError, OSError):
         # we don't care, the user probably won't be able to commit anyway
         return False
     return not (new_file_has_exec or exec_flags_cannot_flip)
 
 def checklink(path):
     """check whether the given path is on a symlink-capable filesystem"""
     # mktemp is not racy because symlink creation will fail if the
     # file already exists
     name = tempfile.mktemp(dir=path, prefix='hg-checklink-')
     try:
         os.symlink(".", name)
         os.unlink(name)
         return True
     except (OSError, AttributeError):
         return False
 
 def checkosfilename(path):
     '''Check that the base-relative path is a valid filename on this platform.
     Returns None if the path is ok, or a UI string describing the problem.'''
     pass # on posix platforms, every path is ok
 
 def setbinary(fd):
     pass
 
 def pconvert(path):
     return path
 
 def localpath(path):
     return path
 
 def samefile(fpath1, fpath2):
     """Returns whether path1 and path2 refer to the same file. This is only
     guaranteed to work for files, not directories."""
     return os.path.samefile(fpath1, fpath2)
 
 def samedevice(fpath1, fpath2):
     """Returns whether fpath1 and fpath2 are on the same device. This is only
     guaranteed to work for files, not directories."""
     st1 = os.lstat(fpath1)
     st2 = os.lstat(fpath2)
     return st1.st_dev == st2.st_dev
 
+encodinglower = None
+encodingupper = None
+
 # os.path.normcase is a no-op, which doesn't help us on non-native filesystems
 def normcase(path):
     return path.lower()
 
 if sys.platform == 'darwin':
     import fcntl # only needed on darwin, missing on jython
 
     def normcase(path):
         try:
             u = path.decode('utf-8')
         except UnicodeDecodeError:
             # percent-encode any characters that don't round-trip
             p2 = path.decode('utf-8', 'ignore').encode('utf-8')
             s = ""
             pos = 0
             for c in path:
                 if p2[pos:pos + 1] == c:
                     s += c
                     pos += 1
                 else:
                     s += "%%%02X" % ord(c)
             u = s.decode('utf-8')
 
         # Decompose then lowercase (HFS+ technote specifies lower)
         return unicodedata.normalize('NFD', u).lower().encode('utf-8')
 
     def realpath(path):
         '''
         Returns the true, canonical file system path equivalent to the given
         path.
 
         Equivalent means, in this case, resulting in the same, unique
         file system link to the path. Every file system entry, whether a file,
         directory, hard link or symbolic link or special, will have a single
         path preferred by the system, but may allow multiple, differing path
         lookups to point to it.
 
         Most regular UNIX file systems only allow a file system entry to be
         looked up by its distinct path. Obviously, this does not apply to case
         insensitive file systems, whether case preserving or not. The most
         complex issue to deal with is file systems transparently reencoding the
         path, such as the non-standard Unicode normalisation required for HFS+
         and HFSX.
         '''
         # Constants copied from /usr/include/sys/fcntl.h
         F_GETPATH = 50
         O_SYMLINK = 0x200000
 
         try:
             fd = os.open(path, O_SYMLINK)
         except OSError, err:
             if err.errno == errno.ENOENT:
                 return path
             raise
 
         try:
             return fcntl.fcntl(fd, F_GETPATH, '\0' * 1024).rstrip('\0')
         finally:
             os.close(fd)
 elif sys.version_info < (2, 4, 2, 'final'):
     # Workaround for http://bugs.python.org/issue1213894 (os.path.realpath
     # didn't resolve symlinks that were the first component of the path.)
     def realpath(path):
         if os.path.isabs(path):
             return os.path.realpath(path)
         else:
             return os.path.realpath('./' + path)
 else:
     # Fallback to the likely inadequate Python builtin function.
     realpath = os.path.realpath
 
 def shellquote(s):
     if os.sys.platform == 'OpenVMS':
         return '"%s"' % s
     else:
         return "'%s'" % s.replace("'", "'\\''")
 
 def quotecommand(cmd):
     return cmd
 
 def popen(command, mode='r'):
     return os.popen(command, mode)
 
 def testpid(pid):
     '''return False if pid dead, True if running or not sure'''
     if os.sys.platform == 'OpenVMS':
         return True
     try:
         os.kill(pid, 0)
         return True
     except OSError, inst:
         return inst.errno != errno.ESRCH
 
 def explainexit(code):
     """return a 2-tuple (desc, code) describing a subprocess status
     (codes from kill are negative - not os.system/wait encoding)"""
     if code >= 0:
         return _("exited with status %d") % code, code
     return _("killed by signal %d") % -code, -code
 
 def isowner(st):
     """Return True if the stat object st is from the current user."""
     return st.st_uid == os.getuid()
 
 def findexe(command):
     '''Find executable for command searching like which does.
     If command is a basename then PATH is searched for command.
     PATH isn't searched if command is an absolute or relative path.
     If command isn't found None is returned.'''
     if sys.platform == 'OpenVMS':
         return command
 
     def findexisting(executable):
         'Will return executable if existing file'
         if os.path.isfile(executable) and os.access(executable, os.X_OK):
             return executable
         return None
 
     if os.sep in command:
         return findexisting(command)
 
     for path in os.environ.get('PATH', '').split(os.pathsep):
         executable = findexisting(os.path.join(path, command))
         if executable is not None:
             return executable
     return None
 
 def setsignalhandler():
     pass
 
 def statfiles(files):
     'Stat each file in files and yield stat or None if file does not exist.'
     lstat = os.lstat
     for nf in files:
         try:
             st = lstat(nf)
         except OSError, err:
             if err.errno not in (errno.ENOENT, errno.ENOTDIR):
                 raise
             st = None
         yield st
 
 def getuser():
     '''return name of current user'''
     return getpass.getuser()
 
 def username(uid=None):
     """Return the name of the user with the given uid.
 
     If uid is None, return the name of the current user."""
 
     if uid is None:
         uid = os.getuid()
     try:
         return pwd.getpwuid(uid)[0]
     except KeyError:
         return str(uid)
 
 def groupname(gid=None):
     """Return the name of the group with the given gid.
 
     If gid is None, return the name of the current group."""
 
     if gid is None:
         gid = os.getgid()
     try:
         return grp.getgrgid(gid)[0]
     except KeyError:
         return str(gid)
 
 def groupmembers(name):
     """Return the list of members of the group with the given
     name, KeyError if the group does not exist.
     """
     return list(grp.getgrnam(name).gr_mem)
 
 def spawndetached(args):
     return os.spawnvp(os.P_NOWAIT | getattr(os, 'P_DETACH', 0),
                       args[0], args)
 
 def gethgcmd():
     return sys.argv[:1]
 
 def termwidth():
     try:
         import termios, array, fcntl
         for dev in (sys.stderr, sys.stdout, sys.stdin):
             try:
                 try:
                     fd = dev.fileno()
                 except AttributeError:
                     continue
                 if not os.isatty(fd):
                     continue
                 arri = fcntl.ioctl(fd, termios.TIOCGWINSZ, '\0' * 8)
                 width = array.array('h', arri)[1]
                 if width > 0:
                     return width
             except ValueError:
                 pass
             except IOError, e:
                 if e[0] == errno.EINVAL:
                     pass
                 else:
                     raise
     except ImportError:
         pass
     return 80
 
 def makedir(path, notindexed):
     os.mkdir(path)
 
 def unlinkpath(f):
     """unlink and remove the directory if it is empty"""
     os.unlink(f)
     # try removing directories that might now be empty
     try:
         os.removedirs(os.path.dirname(f))
     except OSError:
         pass
 
 def lookupreg(key, name=None, scope=None):
     return None
 
 def hidewindow():
     """Hide current shell window.
 
     Used to hide the window opened when starting asynchronous
     child process under Windows, unneeded on other systems.
     """
     pass
 
 class cachestat(object):
     def __init__(self, path):
         self.stat = os.stat(path)
 
     def cacheable(self):
         return bool(self.stat.st_ino)
 
     def __eq__(self, other):
         try:
             return self.stat == other.stat
         except AttributeError:
             return False
 
     def __ne__(self, other):
         return not self == other
 
 def executablepath():
     return None # available on Windows only

mercurial/scmutil.py

@@ -1,803 +1,813
 # scmutil.py - Mercurial core utility functions
 #
 #  Copyright 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 i18n import _
 import util, error, osutil, revset, similar, encoding
 import match as matchmod
 import os, errno, re, stat, sys, glob
 
 def checkfilename(f):
     '''Check that the filename f is an acceptable filename for a tracked file'''
     if '\r' in f or '\n' in f:
         raise util.Abort(_("'\\n' and '\\r' disallowed in filenames: %r") % f)
 
 def checkportable(ui, f):
     '''Check if filename f is portable and warn or abort depending on config'''
     checkfilename(f)
     abort, warn = checkportabilityalert(ui)
     if abort or warn:
         msg = util.checkwinfilename(f)
         if msg:
             msg = "%s: %r" % (msg, f)
             if abort:
                 raise util.Abort(msg)
             ui.warn(_("warning: %s\n") % msg)
 
 def checkportabilityalert(ui):
     '''check if the user's config requests nothing, a warning, or abort for
     non-portable filenames'''
     val = ui.config('ui', 'portablefilenames', 'warn')
     lval = val.lower()
     bval = util.parsebool(val)
     abort = os.name == 'nt' or lval == 'abort'
     warn = bval or lval == 'warn'
     if bval is None and not (warn or abort or lval == 'ignore'):
         raise error.ConfigError(
             _("ui.portablefilenames value is invalid ('%s')") % val)
     return abort, warn
 
 class casecollisionauditor(object):
     def __init__(self, ui, abort, existingiter):
         self._ui = ui
         self._abort = abort
         self._map = {}
         for f in existingiter:
             self._map[encoding.lower(f)] = f
 
     def __call__(self, f):
         fl = encoding.lower(f)
         map = self._map
         if fl in map and map[fl] != f:
             msg = _('possible case-folding collision for %s') % f
             if self._abort:
                 raise util.Abort(msg)
             self._ui.warn(_("warning: %s\n") % msg)
         map[fl] = f
 
 class pathauditor(object):
     '''ensure that a filesystem path contains no banned components.
     the following properties of a path are checked:
 
     - ends with a directory separator
     - under top-level .hg
     - starts at the root of a windows drive
     - contains ".."
     - traverses a symlink (e.g. a/symlink_here/b)
     - inside a nested repository (a callback can be used to approve
       some nested repositories, e.g., subrepositories)
     '''
 
     def __init__(self, root, callback=None):
         self.audited = set()
         self.auditeddir = set()
         self.root = root
         self.callback = callback
+        if os.path.lexists(root) and not util.checkcase(root):
+            self.normcase = util.normcase
+        else:
+            self.normcase = lambda x: x
 
     def __call__(self, path):
         '''Check the relative path.
         path may contain a pattern (e.g. foodir/**.txt)'''
 
-        if path in self.audited:
+        normpath = self.normcase(path)
+        if normpath in self.audited:
             return
         # AIX ignores "/" at end of path, others raise EISDIR.
         if util.endswithsep(path):
             raise util.Abort(_("path ends in directory separator: %s") % path)
-        normpath = os.path.normcase(path)
-        parts = util.splitpath(normpath)
+        parts = util.splitpath(path)
         if (os.path.splitdrive(path)[0]
             or parts[0].lower() in ('.hg', '.hg.', '')
             or os.pardir in parts):
             raise util.Abort(_("path contains illegal component: %s") % path)
         if '.hg' in path.lower():
             lparts = [p.lower() for p in parts]
             for p in '.hg', '.hg.':
                 if p in lparts[1:]:
                     pos = lparts.index(p)
                     base = os.path.join(*parts[:pos])
                     raise util.Abort(_("path '%s' is inside nested repo %r")
                                      % (path, base))
 
+        normparts = util.splitpath(normpath)
+        assert len(parts) == len(normparts)
+
         parts.pop()
+        normparts.pop()
         prefixes = []
         while parts:
             prefix = os.sep.join(parts)
-            if prefix in self.auditeddir:
+            normprefix = os.sep.join(normparts)
+            if normprefix in self.auditeddir:
                 break
             curpath = os.path.join(self.root, prefix)
             try:
                 st = os.lstat(curpath)
             except OSError, err:
                 # EINVAL can be raised as invalid path syntax under win32.
                 # They must be ignored for patterns can be checked too.
                 if err.errno not in (errno.ENOENT, errno.ENOTDIR, errno.EINVAL):
                     raise
             else:
                 if stat.S_ISLNK(st.st_mode):
                     raise util.Abort(
                         _('path %r traverses symbolic link %r')
                         % (path, prefix))
                 elif (stat.S_ISDIR(st.st_mode) and
                       os.path.isdir(os.path.join(curpath, '.hg'))):
                     if not self.callback or not self.callback(curpath):
                         raise util.Abort(_("path '%s' is inside nested repo %r") %
                                          (path, prefix))
-            prefixes.append(prefix)
+            prefixes.append(normprefix)
             parts.pop()
+            normparts.pop()
 
-        self.audited.add(path)
+        self.audited.add(normpath)
         # only add prefixes to the cache after checking everything: we don't
         # want to add "foo/bar/baz" before checking if there's a "foo/.hg"
         self.auditeddir.update(prefixes)
 
 class abstractopener(object):
     """Abstract base class; cannot be instantiated"""
 
     def __init__(self, *args, **kwargs):
         '''Prevent instantiation; don't call this from subclasses.'''
         raise NotImplementedError('attempted instantiating ' + str(type(self)))
 
     def read(self, path):
         fp = self(path, 'rb')
         try:
             return fp.read()
         finally:
             fp.close()
 
     def write(self, path, data):
         fp = self(path, 'wb')
         try:
             return fp.write(data)
         finally:
             fp.close()
 
     def append(self, path, data):
         fp = self(path, 'ab')
         try:
             return fp.write(data)
         finally:
             fp.close()
 
 class opener(abstractopener):
     '''Open files relative to a base directory
 
     This class is used to hide the details of COW semantics and
     remote file access from higher level code.
     '''
     def __init__(self, base, audit=True):
         self.base = base
         self._audit = audit
         if audit:
             self.auditor = pathauditor(base)
         else:
             self.auditor = util.always
         self.createmode = None
         self._trustnlink = None
 
     @util.propertycache
     def _cansymlink(self):
         return util.checklink(self.base)
 
     def _fixfilemode(self, name):
         if self.createmode is None:
             return
         os.chmod(name, self.createmode & 0666)
 
     def __call__(self, path, mode="r", text=False, atomictemp=False):
         if self._audit:
             r = util.checkosfilename(path)
             if r:
                 raise util.Abort("%s: %r" % (r, path))
         self.auditor(path)
         f = os.path.join(self.base, path)
 
         if not text and "b" not in mode:
             mode += "b" # for that other OS
 
         nlink = -1
         dirname, basename = os.path.split(f)
         # If basename is empty, then the path is malformed because it points
         # to a directory. Let the posixfile() call below raise IOError.
         if basename and mode not in ('r', 'rb'):
             if atomictemp:
                 if not os.path.isdir(dirname):
                     util.makedirs(dirname, self.createmode)
                 return util.atomictempfile(f, mode, self.createmode)
             try:
                 if 'w' in mode:
                     util.unlink(f)
                     nlink = 0
                 else:
                     # nlinks() may behave differently for files on Windows
                     # shares if the file is open.
                     fd = util.posixfile(f)
                     nlink = util.nlinks(f)
                     if nlink < 1:
                         nlink = 2 # force mktempcopy (issue1922)
                     fd.close()
             except (OSError, IOError), e:
                 if e.errno != errno.ENOENT:
                     raise
                 nlink = 0
                 if not os.path.isdir(dirname):
                     util.makedirs(dirname, self.createmode)
             if nlink > 0:
                 if self._trustnlink is None:
                     self._trustnlink = nlink > 1 or util.checknlink(f)
                 if nlink > 1 or not self._trustnlink:
                     util.rename(util.mktempcopy(f), f)
         fp = util.posixfile(f, mode)
         if nlink == 0:
             self._fixfilemode(f)
         return fp
 
     def symlink(self, src, dst):
         self.auditor(dst)
         linkname = os.path.join(self.base, dst)
         try:
             os.unlink(linkname)
         except OSError:
             pass
 
         dirname = os.path.dirname(linkname)
         if not os.path.exists(dirname):
             util.makedirs(dirname, self.createmode)
 
         if self._cansymlink:
             try:
                 os.symlink(src, linkname)
             except OSError, err:
                 raise OSError(err.errno, _('could not symlink to %r: %s') %
                               (src, err.strerror), linkname)
         else:
             f = self(dst, "w")
             f.write(src)
             f.close()
             self._fixfilemode(dst)
 
     def audit(self, path):
         self.auditor(path)
 
 class filteropener(abstractopener):
     '''Wrapper opener for filtering filenames with a function.'''
 
     def __init__(self, opener, filter):
         self._filter = filter
         self._orig = opener
 
     def __call__(self, path, *args, **kwargs):
         return self._orig(self._filter(path), *args, **kwargs)
 
 def canonpath(root, cwd, myname, auditor=None):
     '''return the canonical path of myname, given cwd and root'''
     if util.endswithsep(root):
         rootsep = root
     else:
         rootsep = root + os.sep
     name = myname
     if not os.path.isabs(name):
         name = os.path.join(root, cwd, name)
     name = os.path.normpath(name)
     if auditor is None:
         auditor = pathauditor(root)
     if name != rootsep and name.startswith(rootsep):
         name = name[len(rootsep):]
         auditor(name)
         return util.pconvert(name)
     elif name == root:
         return ''
     else:
         # Determine whether `name' is in the hierarchy at or beneath `root',
         # by iterating name=dirname(name) until that causes no change (can't
         # check name == '/', because that doesn't work on windows).  For each
         # `name', compare dev/inode numbers.  If they match, the list `rel'
         # holds the reversed list of components making up the relative file
         # name we want.
         root_st = os.stat(root)
         rel = []
         while True:
             try:
                 name_st = os.stat(name)
             except OSError:
                 break
             if util.samestat(name_st, root_st):
                 if not rel:
                     # name was actually the same as root (maybe a symlink)
                     return ''
                 rel.reverse()
                 name = os.path.join(*rel)
                 auditor(name)
                 return util.pconvert(name)
             dirname, basename = os.path.split(name)
             rel.append(basename)
             if dirname == name:
                 break
             name = dirname
 
         raise util.Abort('%s not under root' % myname)
 
 def walkrepos(path, followsym=False, seen_dirs=None, recurse=False):
     '''yield every hg repository under path, recursively.'''
     def errhandler(err):
         if err.filename == path:
             raise err
     samestat = getattr(os.path, 'samestat', None)
     if followsym and samestat is not None:
         def adddir(dirlst, dirname):
             match = False
             dirstat = os.stat(dirname)
             for lstdirstat in dirlst:
                 if samestat(dirstat, lstdirstat):
                     match = True
                     break
             if not match:
                 dirlst.append(dirstat)
             return not match
     else:
         followsym = False
 
     if (seen_dirs is None) and followsym:
         seen_dirs = []
         adddir(seen_dirs, path)
     for root, dirs, files in os.walk(path, topdown=True, onerror=errhandler):
         dirs.sort()
         if '.hg' in dirs:
             yield root # found a repository
             qroot = os.path.join(root, '.hg', 'patches')
             if os.path.isdir(os.path.join(qroot, '.hg')):
                 yield qroot # we have a patch queue repo here
             if recurse:
                 # avoid recursing inside the .hg directory
                 dirs.remove('.hg')
             else:
                 dirs[:] = [] # don't descend further
         elif followsym:
             newdirs = []
             for d in dirs:
                 fname = os.path.join(root, d)
                 if adddir(seen_dirs, fname):
                     if os.path.islink(fname):
                         for hgname in walkrepos(fname, True, seen_dirs):
                             yield hgname
                     else:
                         newdirs.append(d)
             dirs[:] = newdirs
 
 def osrcpath():
     '''return default os-specific hgrc search path'''
     path = systemrcpath()
     path.extend(userrcpath())
     path = [os.path.normpath(f) for f in path]
     return path
 
 _rcpath = None
 
 def rcpath():
     '''return hgrc search path. if env var HGRCPATH is set, use it.
     for each item in path, if directory, use files ending in .rc,
     else use item.
     make HGRCPATH empty to only look in .hg/hgrc of current repo.
     if no HGRCPATH, use default os-specific path.'''
     global _rcpath
     if _rcpath is None:
         if 'HGRCPATH' in os.environ:
             _rcpath = []
             for p in os.environ['HGRCPATH'].split(os.pathsep):
                 if not p:
                     continue
                 p = util.expandpath(p)
                 if os.path.isdir(p):
                     for f, kind in osutil.listdir(p):
                         if f.endswith('.rc'):
                             _rcpath.append(os.path.join(p, f))
                 else:
                     _rcpath.append(p)
         else:
             _rcpath = osrcpath()
     return _rcpath
 
 if os.name != 'nt':
 
     def rcfiles(path):
         rcs = [os.path.join(path, 'hgrc')]
         rcdir = os.path.join(path, 'hgrc.d')
         try:
             rcs.extend([os.path.join(rcdir, f)
                         for f, kind in osutil.listdir(rcdir)
                         if f.endswith(".rc")])
         except OSError:
             pass
         return rcs
 
     def systemrcpath():
         path = []
         # old mod_python does not set sys.argv
         if len(getattr(sys, 'argv', [])) > 0:
             p = os.path.dirname(os.path.dirname(sys.argv[0]))
             path.extend(rcfiles(os.path.join(p, 'etc/mercurial')))
         path.extend(rcfiles('/etc/mercurial'))
         return path
 
     def userrcpath():
         return [os.path.expanduser('~/.hgrc')]
 
 else:
 
     _HKEY_LOCAL_MACHINE = 0x80000002L
 
     def systemrcpath():
         '''return default os-specific hgrc search path'''
         rcpath = []
         filename = util.executablepath()
         # Use mercurial.ini found in directory with hg.exe
         progrc = os.path.join(os.path.dirname(filename), 'mercurial.ini')
         if os.path.isfile(progrc):
             rcpath.append(progrc)
             return rcpath
         # Use hgrc.d found in directory with hg.exe
         progrcd = os.path.join(os.path.dirname(filename), 'hgrc.d')
         if os.path.isdir(progrcd):
             for f, kind in osutil.listdir(progrcd):
                 if f.endswith('.rc'):
                     rcpath.append(os.path.join(progrcd, f))
             return rcpath
         # else look for a system rcpath in the registry
         value = util.lookupreg('SOFTWARE\\Mercurial', None,
                                _HKEY_LOCAL_MACHINE)
         if not isinstance(value, str) or not value:
             return rcpath
         value = value.replace('/', os.sep)
         for p in value.split(os.pathsep):
             if p.lower().endswith('mercurial.ini'):
                 rcpath.append(p)
             elif os.path.isdir(p):
                 for f, kind in osutil.listdir(p):
                     if f.endswith('.rc'):
                         rcpath.append(os.path.join(p, f))
         return rcpath
 
     def userrcpath():
         '''return os-specific hgrc search path to the user dir'''
         home = os.path.expanduser('~')
         path = [os.path.join(home, 'mercurial.ini'),
                 os.path.join(home, '.hgrc')]
         userprofile = os.environ.get('USERPROFILE')
         if userprofile:
             path.append(os.path.join(userprofile, 'mercurial.ini'))
             path.append(os.path.join(userprofile, '.hgrc'))
         return path
 
 def revsingle(repo, revspec, default='.'):
     if not revspec:
         return repo[default]
 
     l = revrange(repo, [revspec])
     if len(l) < 1:
         raise util.Abort(_('empty revision set'))
     return repo[l[-1]]
 
 def revpair(repo, revs):
     if not revs:
         return repo.dirstate.p1(), None
 
     l = revrange(repo, revs)
 
     if len(l) == 0:
         return repo.dirstate.p1(), None
 
     if len(l) == 1:
         return repo.lookup(l[0]), None
 
     return repo.lookup(l[0]), repo.lookup(l[-1])
 
 _revrangesep = ':'
 
 def revrange(repo, revs):
     """Yield revision as strings from a list of revision specifications."""
 
     def revfix(repo, val, defval):
         if not val and val != 0 and defval is not None:
             return defval
         return repo.changelog.rev(repo.lookup(val))
 
     seen, l = set(), []
     for spec in revs:
         # attempt to parse old-style ranges first to deal with
         # things like old-tag which contain query metacharacters
         try:
             if isinstance(spec, int):
                 seen.add(spec)
                 l.append(spec)
                 continue
 
             if _revrangesep in spec:
                 start, end = spec.split(_revrangesep, 1)
                 start = revfix(repo, start, 0)
                 end = revfix(repo, end, len(repo) - 1)
                 step = start > end and -1 or 1
                 for rev in xrange(start, end + step, step):
                     if rev in seen:
                         continue
                     seen.add(rev)
                     l.append(rev)
                 continue
             elif spec and spec in repo: # single unquoted rev
                 rev = revfix(repo, spec, None)
                 if rev in seen:
                     continue
                 seen.add(rev)
                 l.append(rev)
                 continue
         except error.RepoLookupError:
             pass
 
         # fall through to new-style queries if old-style fails
         m = revset.match(repo.ui, spec)
         for r in m(repo, range(len(repo))):
             if r not in seen:
                 l.append(r)
         seen.update(l)
 
     return l
 
 def expandpats(pats):
     if not util.expandglobs:
         return list(pats)
     ret = []
     for p in pats:
         kind, name = matchmod._patsplit(p, None)
         if kind is None:
             try:
                 globbed = glob.glob(name)
             except re.error:
                 globbed = [name]
             if globbed:
                 ret.extend(globbed)
                 continue
         ret.append(p)
     return ret
 
 def match(ctx, pats=[], opts={}, globbed=False, default='relpath'):
     if pats == ("",):
         pats = []
     if not globbed and default == 'relpath':
         pats = expandpats(pats or [])
 
     m = ctx.match(pats, opts.get('include'), opts.get('exclude'),
                          default)
     def badfn(f, msg):
         ctx._repo.ui.warn("%s: %s\n" % (m.rel(f), msg))
     m.bad = badfn
     return m
 
 def matchall(repo):
     return matchmod.always(repo.root, repo.getcwd())
 
 def matchfiles(repo, files):
     return matchmod.exact(repo.root, repo.getcwd(), files)
 
 def addremove(repo, pats=[], opts={}, dry_run=None, similarity=None):
     if dry_run is None:
         dry_run = opts.get('dry_run')
     if similarity is None:
         similarity = float(opts.get('similarity') or 0)
     # we'd use status here, except handling of symlinks and ignore is tricky
     added, unknown, deleted, removed = [], [], [], []
     audit_path = pathauditor(repo.root)
     m = match(repo[None], pats, opts)
     for abs in repo.walk(m):
         target = repo.wjoin(abs)
         good = True
         try:
             audit_path(abs)
         except (OSError, util.Abort):
             good = False
         rel = m.rel(abs)
         exact = m.exact(abs)
         if good and abs not in repo.dirstate:
             unknown.append(abs)
             if repo.ui.verbose or not exact:
                 repo.ui.status(_('adding %s\n') % ((pats and rel) or abs))
         elif repo.dirstate[abs] != 'r' and (not good or not os.path.lexists(target)
             or (os.path.isdir(target) and not os.path.islink(target))):
             deleted.append(abs)
             if repo.ui.verbose or not exact:
                 repo.ui.status(_('removing %s\n') % ((pats and rel) or abs))
         # for finding renames
         elif repo.dirstate[abs] == 'r':
             removed.append(abs)
         elif repo.dirstate[abs] == 'a':
             added.append(abs)
     copies = {}
     if similarity > 0:
         for old, new, score in similar.findrenames(repo,
                 added + unknown, removed + deleted, similarity):
             if repo.ui.verbose or not m.exact(old) or not m.exact(new):
                 repo.ui.status(_('recording removal of %s as rename to %s '
                                  '(%d%% similar)\n') %
                                (m.rel(old), m.rel(new), score * 100))
             copies[new] = old
 
     if not dry_run:
         wctx = repo[None]
         wlock = repo.wlock()
         try:
             wctx.forget(deleted)
             wctx.add(unknown)
             for new, old in copies.iteritems():
                 wctx.copy(old, new)
         finally:
             wlock.release()
 
 def updatedir(ui, repo, patches, similarity=0):
     '''Update dirstate after patch application according to metadata'''
     if not patches:
         return []
     copies = []
     removes = set()
     cfiles = patches.keys()
     cwd = repo.getcwd()
     if cwd:
         cfiles = [util.pathto(repo.root, cwd, f) for f in patches.keys()]
     for f in patches:
         gp = patches[f]
         if not gp:
             continue
         if gp.op == 'RENAME':
             copies.append((gp.oldpath, gp.path))
             removes.add(gp.oldpath)
         elif gp.op == 'COPY':
             copies.append((gp.oldpath, gp.path))
         elif gp.op == 'DELETE':
             removes.add(gp.path)
 
     wctx = repo[None]
     for src, dst in copies:
         dirstatecopy(ui, repo, wctx, src, dst, cwd=cwd)
     if (not similarity) and removes:
         wctx.remove(sorted(removes), True)
 
     for f in patches:
         gp = patches[f]
         if gp and gp.mode:
             islink, isexec = gp.mode
             dst = repo.wjoin(gp.path)
             # patch won't create empty files
             if gp.op == 'ADD' and not os.path.lexists(dst):
                 flags = (isexec and 'x' or '') + (islink and 'l' or '')
                 repo.wwrite(gp.path, '', flags)
             util.setflags(dst, islink, isexec)
     addremove(repo, cfiles, similarity=similarity)
     files = patches.keys()
     files.extend([r for r in removes if r not in files])
     return sorted(files)
 
 def dirstatecopy(ui, repo, wctx, src, dst, dryrun=False, cwd=None):
     """Update the dirstate to reflect the intent of copying src to dst. For
     different reasons it might not end with dst being marked as copied from src.
     """
     origsrc = repo.dirstate.copied(src) or src
     if dst == origsrc: # copying back a copy?
         if repo.dirstate[dst] not in 'mn' and not dryrun:
             repo.dirstate.normallookup(dst)
     else:
         if repo.dirstate[origsrc] == 'a' and origsrc == src:
             if not ui.quiet:
                 ui.warn(_("%s has not been committed yet, so no copy "
                           "data will be stored for %s.\n")
                         % (repo.pathto(origsrc, cwd), repo.pathto(dst, cwd)))
             if repo.dirstate[dst] in '?r' and not dryrun:
                 wctx.add([dst])
         elif not dryrun:
             wctx.copy(origsrc, dst)
 
 def readrequires(opener, supported):
     '''Reads and parses .hg/requires and checks if all entries found
     are in the list of supported features.'''
     requirements = set(opener.read("requires").splitlines())
     missings = []
     for r in requirements:
         if r not in supported:
             if not r or not r[0].isalnum():
                 raise error.RequirementError(_(".hg/requires file is corrupt"))
             missings.append(r)
     missings.sort()
     if missings:
         raise error.RequirementError(_("unknown repository format: "
             "requires features '%s' (upgrade Mercurial)") % "', '".join(missings))
     return requirements
 
 class filecacheentry(object):
     def __init__(self, path):
         self.path = path
         self.cachestat = filecacheentry.stat(self.path)
 
         if self.cachestat:
             self._cacheable = self.cachestat.cacheable()
         else:
             # None means we don't know yet
             self._cacheable = None
 
     def refresh(self):
         if self.cacheable():
             self.cachestat = filecacheentry.stat(self.path)
 
     def cacheable(self):
         if self._cacheable is not None:
             return self._cacheable
 
         # we don't know yet, assume it is for now
         return True
 
     def changed(self):
         # no point in going further if we can't cache it
         if not self.cacheable():
             return True
 
         newstat = filecacheentry.stat(self.path)
 
         # we may not know if it's cacheable yet, check again now
         if newstat and self._cacheable is None:
             self._cacheable = newstat.cacheable()
 
             # check again
             if not self._cacheable:
                 return True
 
         if self.cachestat != newstat:
             self.cachestat = newstat
             return True
         else:
             return False
 
     @staticmethod
     def stat(path):
         try:
             return util.cachestat(path)
         except OSError, e:
             if e.errno != errno.ENOENT:
                 raise
 
 class filecache(object):
     '''A property like decorator that tracks a file under .hg/ for updates.
 
     Records stat info when called in _filecache.
 
     On subsequent calls, compares old stat info with new info, and recreates
     the object when needed, updating the new stat info in _filecache.
 
     Mercurial either atomic renames or appends for files under .hg,
     so to ensure the cache is reliable we need the filesystem to be able
     to tell us if a file has been replaced. If it can't, we fallback to
     recreating the object on every call (essentially the same behaviour as
     propertycache).'''
     def __init__(self, path, instore=False):
         self.path = path
         self.instore = instore
 
     def __call__(self, func):
         self.func = func
         self.name = func.__name__
         return self
 
     def __get__(self, obj, type=None):
         entry = obj._filecache.get(self.name)
 
         if entry:
             if entry.changed():
                 entry.obj = self.func(obj)
         else:
             path = self.instore and obj.sjoin(self.path) or obj.join(self.path)
 
             # We stat -before- creating the object so our cache doesn't lie if
             # a writer modified between the time we read and stat
             entry = filecacheentry(path)
             entry.obj = self.func(obj)
 
             obj._filecache[self.name] = entry
 
         setattr(obj, self.name, entry.obj)
         return entry.obj

mercurial/util.py

@@ -1,1744 +1,1751
 # util.py - Mercurial utility functions and platform specfic implementations
 #
 #  Copyright 2005 K. Thananchayan <[e-mail]>
 #  Copyright 2005-2007 Matt Mackall <[e-mail]>
 #  Copyright 2006 Vadim Gelfer <[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.
 
 """Mercurial utility functions and platform specfic implementations.
 
 This contains helper routines that are independent of the SCM core and
 hide platform-specific details from the core.
 """
 
 from i18n import _
 import error, osutil, encoding
 import errno, re, shutil, sys, tempfile, traceback
 import os, time, datetime, calendar, textwrap, signal
 import imp, socket, urllib
 
 if os.name == 'nt':
     import windows as platform
 else:
     import posix as platform
 
+platform.encodinglower = encoding.lower
+platform.encodingupper = encoding.upper
+
 cachestat = platform.cachestat
 checkexec = platform.checkexec
 checklink = platform.checklink
 copymode = platform.copymode
 executablepath = platform.executablepath
 expandglobs = platform.expandglobs
 explainexit = platform.explainexit
 findexe = platform.findexe
 gethgcmd = platform.gethgcmd
 getuser = platform.getuser
 groupmembers = platform.groupmembers
 groupname = platform.groupname
 hidewindow = platform.hidewindow
 isexec = platform.isexec
 isowner = platform.isowner
 localpath = platform.localpath
 lookupreg = platform.lookupreg
 makedir = platform.makedir
 nlinks = platform.nlinks
 normpath = platform.normpath
 normcase = platform.normcase
 nulldev = platform.nulldev
 openhardlinks = platform.openhardlinks
 oslink = platform.oslink
 parsepatchoutput = platform.parsepatchoutput
 pconvert = platform.pconvert
 popen = platform.popen
 posixfile = platform.posixfile
 quotecommand = platform.quotecommand
 realpath = platform.realpath
 rename = platform.rename
 samedevice = platform.samedevice
 samefile = platform.samefile
 samestat = platform.samestat
 setbinary = platform.setbinary
 setflags = platform.setflags
 setsignalhandler = platform.setsignalhandler
 shellquote = platform.shellquote
 spawndetached = platform.spawndetached
 sshargs = platform.sshargs
 statfiles = platform.statfiles
 termwidth = platform.termwidth
 testpid = platform.testpid
 umask = platform.umask
 unlink = platform.unlink
 unlinkpath = platform.unlinkpath
 username = platform.username
 
 # Python compatibility
 
 _notset = object()
 
 def safehasattr(thing, attr):
     return getattr(thing, attr, _notset) is not _notset
 
 def sha1(s=''):
     '''
     Low-overhead wrapper around Python's SHA support
 
     >>> f = _fastsha1
     >>> a = sha1()
     >>> a = f()
     >>> a.hexdigest()
     'da39a3ee5e6b4b0d3255bfef95601890afd80709'
     '''
 
     return _fastsha1(s)
 
 def _fastsha1(s=''):
     # This function will import sha1 from hashlib or sha (whichever is
     # available) and overwrite itself with it on the first call.
     # Subsequent calls will go directly to the imported function.
     if sys.version_info >= (2, 5):
         from hashlib import sha1 as _sha1
     else:
         from sha import sha as _sha1
     global _fastsha1, sha1
     _fastsha1 = sha1 = _sha1
     return _sha1(s)
 
 try:
     buffer = buffer
 except NameError:
     if sys.version_info[0] < 3:
         def buffer(sliceable, offset=0):
             return sliceable[offset:]
     else:
         def buffer(sliceable, offset=0):
             return memoryview(sliceable)[offset:]
 
 import subprocess
 closefds = os.name == 'posix'
 
 def popen2(cmd, env=None, newlines=False):
     # Setting bufsize to -1 lets the system decide the buffer size.
     # The default for bufsize is 0, meaning unbuffered. This leads to
     # poor performance on Mac OS X: http://bugs.python.org/issue4194
     p = subprocess.Popen(cmd, shell=True, bufsize=-1,
                          close_fds=closefds,
                          stdin=subprocess.PIPE, stdout=subprocess.PIPE,
                          universal_newlines=newlines,
                          env=env)
     return p.stdin, p.stdout
 
 def popen3(cmd, env=None, newlines=False):
     p = subprocess.Popen(cmd, shell=True, bufsize=-1,
                          close_fds=closefds,
                          stdin=subprocess.PIPE, stdout=subprocess.PIPE,
                          stderr=subprocess.PIPE,
                          universal_newlines=newlines,
                          env=env)
     return p.stdin, p.stdout, p.stderr
 
 def version():
     """Return version information if available."""
     try:
         import __version__
         return __version__.version
     except ImportError:
         return 'unknown'
 
 # used by parsedate
 defaultdateformats = (
     '%Y-%m-%d %H:%M:%S',
     '%Y-%m-%d %I:%M:%S%p',
     '%Y-%m-%d %H:%M',
     '%Y-%m-%d %I:%M%p',
     '%Y-%m-%d',
     '%m-%d',
     '%m/%d',
     '%m/%d/%y',
     '%m/%d/%Y',
     '%a %b %d %H:%M:%S %Y',
     '%a %b %d %I:%M:%S%p %Y',
     '%a, %d %b %Y %H:%M:%S',        #  GNU coreutils "/bin/date --rfc-2822"
     '%b %d %H:%M:%S %Y',
     '%b %d %I:%M:%S%p %Y',
     '%b %d %H:%M:%S',
     '%b %d %I:%M:%S%p',
     '%b %d %H:%M',
     '%b %d %I:%M%p',
     '%b %d %Y',
     '%b %d',
     '%H:%M:%S',
     '%I:%M:%S%p',
     '%H:%M',
     '%I:%M%p',
 )
 
 extendeddateformats = defaultdateformats + (
     "%Y",
     "%Y-%m",
     "%b",
     "%b %Y",
     )
 
 def cachefunc(func):
     '''cache the result of function calls'''
     # XXX doesn't handle keywords args
     cache = {}
     if func.func_code.co_argcount == 1:
         # we gain a small amount of time because
         # we don't need to pack/unpack the list
         def f(arg):
             if arg not in cache:
                 cache[arg] = func(arg)
             return cache[arg]
     else:
         def f(*args):
             if args not in cache:
                 cache[args] = func(*args)
             return cache[args]
 
     return f
 
 def lrucachefunc(func):
     '''cache most recent results of function calls'''
     cache = {}
     order = []
     if func.func_code.co_argcount == 1:
         def f(arg):
             if arg not in cache:
                 if len(cache) > 20:
                     del cache[order.pop(0)]
                 cache[arg] = func(arg)
             else:
                 order.remove(arg)
             order.append(arg)
             return cache[arg]
     else:
         def f(*args):
             if args not in cache:
                 if len(cache) > 20:
                     del cache[order.pop(0)]
                 cache[args] = func(*args)
             else:
                 order.remove(args)
             order.append(args)
             return cache[args]
 
     return f
 
 class propertycache(object):
     def __init__(self, func):
         self.func = func
         self.name = func.__name__
     def __get__(self, obj, type=None):
         result = self.func(obj)
         setattr(obj, self.name, result)
         return result
 
 def pipefilter(s, cmd):
     '''filter string S through command CMD, returning its output'''
     p = subprocess.Popen(cmd, shell=True, close_fds=closefds,
                          stdin=subprocess.PIPE, stdout=subprocess.PIPE)
     pout, perr = p.communicate(s)
     return pout
 
 def tempfilter(s, cmd):
     '''filter string S through a pair of temporary files with CMD.
     CMD is used as a template to create the real command to be run,
     with the strings INFILE and OUTFILE replaced by the real names of
     the temporary files generated.'''
     inname, outname = None, None
     try:
         infd, inname = tempfile.mkstemp(prefix='hg-filter-in-')
         fp = os.fdopen(infd, 'wb')
         fp.write(s)
         fp.close()
         outfd, outname = tempfile.mkstemp(prefix='hg-filter-out-')
         os.close(outfd)
         cmd = cmd.replace('INFILE', inname)
         cmd = cmd.replace('OUTFILE', outname)
         code = os.system(cmd)
         if sys.platform == 'OpenVMS' and code & 1:
             code = 0
         if code:
             raise Abort(_("command '%s' failed: %s") %
                         (cmd, explainexit(code)))
         fp = open(outname, 'rb')
         r = fp.read()
         fp.close()
         return r
     finally:
         try:
             if inname:
                 os.unlink(inname)
         except OSError:
             pass
         try:
             if outname:
                 os.unlink(outname)
         except OSError:
             pass
 
 filtertable = {
     'tempfile:': tempfilter,
     'pipe:': pipefilter,
     }
 
 def filter(s, cmd):
     "filter a string through a command that transforms its input to its output"
     for name, fn in filtertable.iteritems():
         if cmd.startswith(name):
             return fn(s, cmd[len(name):].lstrip())
     return pipefilter(s, cmd)
 
 def binary(s):
     """return true if a string is binary data"""
     return bool(s and '\0' in s)
 
 def increasingchunks(source, min=1024, max=65536):
     '''return no less than min bytes per chunk while data remains,
     doubling min after each chunk until it reaches max'''
     def log2(x):
         if not x:
             return 0
         i = 0
         while x:
             x >>= 1
             i += 1
         return i - 1
 
     buf = []
     blen = 0
     for chunk in source:
         buf.append(chunk)
         blen += len(chunk)
         if blen >= min:
             if min < max:
                 min = min << 1
                 nmin = 1 << log2(blen)
                 if nmin > min:
                     min = nmin
                 if min > max:
                     min = max
             yield ''.join(buf)
             blen = 0
             buf = []
     if buf:
         yield ''.join(buf)
 
 Abort = error.Abort
 
 def always(fn):
     return True
 
 def never(fn):
     return False
 
 def pathto(root, n1, n2):
     '''return the relative path from one place to another.
     root should use os.sep to separate directories
     n1 should use os.sep to separate directories
     n2 should use "/" to separate directories
     returns an os.sep-separated path.
 
     If n1 is a relative path, it's assumed it's
     relative to root.
     n2 should always be relative to root.
     '''
     if not n1:
         return localpath(n2)
     if os.path.isabs(n1):
         if os.path.splitdrive(root)[0] != os.path.splitdrive(n1)[0]:
             return os.path.join(root, localpath(n2))
         n2 = '/'.join((pconvert(root), n2))
     a, b = splitpath(n1), n2.split('/')
     a.reverse()
     b.reverse()
     while a and b and a[-1] == b[-1]:
         a.pop()
         b.pop()
     b.reverse()
     return os.sep.join((['..'] * len(a)) + b) or '.'
 
 _hgexecutable = None
 
 def mainfrozen():
     """return True if we are a frozen executable.
 
     The code supports py2exe (most common, Windows only) and tools/freeze
     (portable, not much used).
     """
     return (safehasattr(sys, "frozen") or # new py2exe
             safehasattr(sys, "importers") or # old py2exe
             imp.is_frozen("__main__")) # tools/freeze
 
 def hgexecutable():
     """return location of the 'hg' executable.
 
     Defaults to $HG or 'hg' in the search path.
     """
     if _hgexecutable is None:
         hg = os.environ.get('HG')
         mainmod = sys.modules['__main__']
         if hg:
             _sethgexecutable(hg)
         elif mainfrozen():
             _sethgexecutable(sys.executable)
         elif os.path.basename(getattr(mainmod, '__file__', '')) == 'hg':
             _sethgexecutable(mainmod.__file__)
         else:
             exe = findexe('hg') or os.path.basename(sys.argv[0])
             _sethgexecutable(exe)
     return _hgexecutable
 
 def _sethgexecutable(path):
     """set location of the 'hg' executable"""
     global _hgexecutable
     _hgexecutable = path
 
 def system(cmd, environ={}, cwd=None, onerr=None, errprefix=None, out=None):
     '''enhanced shell command execution.
     run with environment maybe modified, maybe in different dir.
 
     if command fails and onerr is None, return status.  if ui object,
     print error message and return status, else raise onerr object as
     exception.
 
     if out is specified, it is assumed to be a file-like object that has a
     write() method. stdout and stderr will be redirected to out.'''
     try:
         sys.stdout.flush()
     except Exception:
         pass
     def py2shell(val):
         'convert python object into string that is useful to shell'
         if val is None or val is False:
             return '0'
         if val is True:
             return '1'
         return str(val)
     origcmd = cmd
     cmd = quotecommand(cmd)
     env = dict(os.environ)
     env.update((k, py2shell(v)) for k, v in environ.iteritems())
     env['HG'] = hgexecutable()
     if out is None or out == sys.__stdout__:
         rc = subprocess.call(cmd, shell=True, close_fds=closefds,
                              env=env, cwd=cwd)
     else:
         proc = subprocess.Popen(cmd, shell=True, close_fds=closefds,
                                 env=env, cwd=cwd, stdout=subprocess.PIPE,
                                 stderr=subprocess.STDOUT)
         for line in proc.stdout:
             out.write(line)
         proc.wait()
         rc = proc.returncode
     if sys.platform == 'OpenVMS' and rc & 1:
         rc = 0
     if rc and onerr:
         errmsg = '%s %s' % (os.path.basename(origcmd.split(None, 1)[0]),
                             explainexit(rc)[0])
         if errprefix:
             errmsg = '%s: %s' % (errprefix, errmsg)
         try:
             onerr.warn(errmsg + '\n')
         except AttributeError:
             raise onerr(errmsg)
     return rc
 
 def checksignature(func):
     '''wrap a function with code to check for calling errors'''
     def check(*args, **kwargs):
         try:
             return func(*args, **kwargs)
         except TypeError:
             if len(traceback.extract_tb(sys.exc_info()[2])) == 1:
                 raise error.SignatureError
             raise
 
     return check
 
 def copyfile(src, dest):
     "copy a file, preserving mode and atime/mtime"
     if os.path.islink(src):
         try:
             os.unlink(dest)
         except OSError:
             pass
         os.symlink(os.readlink(src), dest)
     else:
         try:
             shutil.copyfile(src, dest)
             shutil.copymode(src, dest)
         except shutil.Error, inst:
             raise Abort(str(inst))
 
 def copyfiles(src, dst, hardlink=None):
     """Copy a directory tree using hardlinks if possible"""
 
     if hardlink is None:
         hardlink = (os.stat(src).st_dev ==
                     os.stat(os.path.dirname(dst)).st_dev)
 
     num = 0
     if os.path.isdir(src):
         os.mkdir(dst)
         for name, kind in osutil.listdir(src):
             srcname = os.path.join(src, name)
             dstname = os.path.join(dst, name)
             hardlink, n = copyfiles(srcname, dstname, hardlink)
             num += n
     else:
         if hardlink:
             try:
                 oslink(src, dst)
             except (IOError, OSError):
                 hardlink = False
                 shutil.copy(src, dst)
         else:
             shutil.copy(src, dst)
         num += 1
 
     return hardlink, num
 
 _winreservednames = '''con prn aux nul
     com1 com2 com3 com4 com5 com6 com7 com8 com9
     lpt1 lpt2 lpt3 lpt4 lpt5 lpt6 lpt7 lpt8 lpt9'''.split()
 _winreservedchars = ':*?"<>|'
 def checkwinfilename(path):
     '''Check that the base-relative path is a valid filename on Windows.
     Returns None if the path is ok, or a UI string describing the problem.
 
     >>> checkwinfilename("just/a/normal/path")
     >>> checkwinfilename("foo/bar/con.xml")
     "filename contains 'con', which is reserved on Windows"
     >>> checkwinfilename("foo/con.xml/bar")
     "filename contains 'con', which is reserved on Windows"
     >>> checkwinfilename("foo/bar/xml.con")
     >>> checkwinfilename("foo/bar/AUX/bla.txt")
     "filename contains 'AUX', which is reserved on Windows"
     >>> checkwinfilename("foo/bar/bla:.txt")
     "filename contains ':', which is reserved on Windows"
     >>> checkwinfilename("foo/bar/b\07la.txt")
     "filename contains '\\\\x07', which is invalid on Windows"
     >>> checkwinfilename("foo/bar/bla ")
     "filename ends with ' ', which is not allowed on Windows"
     >>> checkwinfilename("../bar")
     '''
     for n in path.replace('\\', '/').split('/'):
         if not n:
             continue
         for c in n:
             if c in _winreservedchars:
                 return _("filename contains '%s', which is reserved "
                          "on Windows") % c
             if ord(c) <= 31:
                 return _("filename contains %r, which is invalid "
                          "on Windows") % c
         base = n.split('.')[0]
         if base and base.lower() in _winreservednames:
             return _("filename contains '%s', which is reserved "
                      "on Windows") % base
         t = n[-1]
         if t in '. ' and n not in '..':
             return _("filename ends with '%s', which is not allowed "
                      "on Windows") % t
 
 if os.name == 'nt':
     checkosfilename = checkwinfilename
 else:
     checkosfilename = platform.checkosfilename
 
 def makelock(info, pathname):
     try:
         return os.symlink(info, pathname)
     except OSError, why:
         if why.errno == errno.EEXIST:
             raise
     except AttributeError: # no symlink in os
         pass
 
     ld = os.open(pathname, os.O_CREAT | os.O_WRONLY | os.O_EXCL)
     os.write(ld, info)
     os.close(ld)
 
 def readlock(pathname):
     try:
         return os.readlink(pathname)
     except OSError, why:
         if why.errno not in (errno.EINVAL, errno.ENOSYS):
             raise
     except AttributeError: # no symlink in os
         pass
     fp = posixfile(pathname)
     r = fp.read()
     fp.close()
     return r
 
 def fstat(fp):
     '''stat file object that may not have fileno method.'''
     try:
         return os.fstat(fp.fileno())
     except AttributeError:
         return os.stat(fp.name)
 
 # File system features
 
 def checkcase(path):
     """
     Check whether the given path is on a case-sensitive filesystem
 
     Requires a path (like /foo/.hg) ending with a foldable final
     directory component.
     """
     s1 = os.stat(path)
     d, b = os.path.split(path)
-    p2 = os.path.join(d, b.upper())
-    if path == p2:
-        p2 = os.path.join(d, b.lower())
+    b2 = b.upper()
+    if b == b2:
+        b2 = b.lower()
+        if b == b2:
+            return True # no evidence against case sensitivity
+    p2 = os.path.join(d, b2)
     try:
         s2 = os.stat(p2)
         if s2 == s1:
             return False
         return True
     except OSError:
         return True
 
 _fspathcache = {}
 def fspath(name, root):
     '''Get name in the case stored in the filesystem
 
     The name is either relative to root, or it is an absolute path starting
     with root. Note that this function is unnecessary, and should not be
     called, for case-sensitive filesystems (simply because it's expensive).
+
+    Both name and root should be normcase-ed.
     '''
     # If name is absolute, make it relative
-    if name.lower().startswith(root.lower()):
+    if name.startswith(root):
         l = len(root)
         if name[l] == os.sep or name[l] == os.altsep:
             l = l + 1
         name = name[l:]
 
     if not os.path.lexists(os.path.join(root, name)):
         return None
 
     seps = os.sep
     if os.altsep:
         seps = seps + os.altsep
     # Protect backslashes. This gets silly very quickly.
     seps.replace('\\','\\\\')
     pattern = re.compile(r'([^%s]+)|([%s]+)' % (seps, seps))
-    dir = os.path.normcase(os.path.normpath(root))
+    dir = os.path.normpath(root)
     result = []
     for part, sep in pattern.findall(name):
         if sep:
             result.append(sep)
             continue
 
         if dir not in _fspathcache:
             _fspathcache[dir] = os.listdir(dir)
         contents = _fspathcache[dir]
 
-        lpart = part.lower()
         lenp = len(part)
         for n in contents:
-            if lenp == len(n) and n.lower() == lpart:
+            if lenp == len(n) and normcase(n) == part:
                 result.append(n)
                 break
         else:
             # Cannot happen, as the file exists!
             result.append(part)
-        dir = os.path.join(dir, lpart)
+        dir = os.path.join(dir, part)
 
     return ''.join(result)
 
 def checknlink(testfile):
     '''check whether hardlink count reporting works properly'''
 
     # testfile may be open, so we need a separate file for checking to
     # work around issue2543 (or testfile may get lost on Samba shares)
     f1 = testfile + ".hgtmp1"
     if os.path.lexists(f1):
         return False
     try:
         posixfile(f1, 'w').close()
     except IOError:
         return False
 
     f2 = testfile + ".hgtmp2"
     fd = None
     try:
         try:
             oslink(f1, f2)
         except OSError:
             return False
 
         # nlinks() may behave differently for files on Windows shares if
         # the file is open.
         fd = posixfile(f2)
         return nlinks(f2) > 1
     finally:
         if fd is not None:
             fd.close()
         for f in (f1, f2):
             try:
                 os.unlink(f)
             except OSError:
                 pass
 
     return False
 
 def endswithsep(path):
     '''Check path ends with os.sep or os.altsep.'''
     return path.endswith(os.sep) or os.altsep and path.endswith(os.altsep)
 
 def splitpath(path):
     '''Split path by os.sep.
     Note that this function does not use os.altsep because this is
     an alternative of simple "xxx.split(os.sep)".
     It is recommended to use os.path.normpath() before using this
     function if need.'''
     return path.split(os.sep)
 
 def gui():
     '''Are we running in a GUI?'''
     if sys.platform == 'darwin':
         if 'SSH_CONNECTION' in os.environ:
             # handle SSH access to a box where the user is logged in
             return False
         elif getattr(osutil, 'isgui', None):
             # check if a CoreGraphics session is available
             return osutil.isgui()
         else:
             # pure build; use a safe default
             return True
     else:
         return os.name == "nt" or os.environ.get("DISPLAY")
 
 def mktempcopy(name, emptyok=False, createmode=None):
     """Create a temporary file with the same contents from name
 
     The permission bits are copied from the original file.
 
     If the temporary file is going to be truncated immediately, you
     can use emptyok=True as an optimization.
 
     Returns the name of the temporary file.
     """
     d, fn = os.path.split(name)
     fd, temp = tempfile.mkstemp(prefix='.%s-' % fn, dir=d)
     os.close(fd)
     # Temporary files are created with mode 0600, which is usually not
     # what we want.  If the original file already exists, just copy
     # its mode.  Otherwise, manually obey umask.
     copymode(name, temp, createmode)
     if emptyok:
         return temp
     try:
         try:
             ifp = posixfile(name, "rb")
         except IOError, inst:
             if inst.errno == errno.ENOENT:
                 return temp
             if not getattr(inst, 'filename', None):
                 inst.filename = name
             raise
         ofp = posixfile(temp, "wb")
         for chunk in filechunkiter(ifp):
             ofp.write(chunk)
         ifp.close()
         ofp.close()
     except:
         try: os.unlink(temp)
         except: pass
         raise
     return temp
 
 class atomictempfile(object):
     '''writeable file object that atomically updates a file
 
     All writes will go to a temporary copy of the original file. Call
     close() when you are done writing, and atomictempfile will rename
     the temporary copy to the original name, making the changes
     visible. If the object is destroyed without being closed, all your
     writes are discarded.
     '''
     def __init__(self, name, mode='w+b', createmode=None):
         self.__name = name      # permanent name
         self._tempname = mktempcopy(name, emptyok=('w' in mode),
                                     createmode=createmode)
         self._fp = posixfile(self._tempname, mode)
 
         # delegated methods
         self.write = self._fp.write
         self.fileno = self._fp.fileno
 
     def close(self):
         if not self._fp.closed:
             self._fp.close()
             rename(self._tempname, localpath(self.__name))
 
     def discard(self):
         if not self._fp.closed:
             try:
                 os.unlink(self._tempname)
             except OSError:
                 pass
             self._fp.close()
 
     def __del__(self):
         if safehasattr(self, '_fp'): # constructor actually did something
             self.discard()
 
 def makedirs(name, mode=None):
     """recursive directory creation with parent mode inheritance"""
     try:
         os.mkdir(name)
     except OSError, err:
         if err.errno == errno.EEXIST:
             return
         if err.errno != errno.ENOENT or not name:
             raise
         parent = os.path.dirname(os.path.abspath(name))
         if parent == name:
             raise
         makedirs(parent, mode)
         os.mkdir(name)
     if mode is not None:
         os.chmod(name, mode)
 
 def readfile(path):
     fp = open(path, 'rb')
     try:
         return fp.read()
     finally:
         fp.close()
 
 def writefile(path, text):
     fp = open(path, 'wb')
     try:
         fp.write(text)
     finally:
         fp.close()
 
 def appendfile(path, text):
     fp = open(path, 'ab')
     try:
         fp.write(text)
     finally:
         fp.close()
 
 class chunkbuffer(object):
     """Allow arbitrary sized chunks of data to be efficiently read from an
     iterator over chunks of arbitrary size."""
 
     def __init__(self, in_iter):
         """in_iter is the iterator that's iterating over the input chunks.
         targetsize is how big a buffer to try to maintain."""
         def splitbig(chunks):
             for chunk in chunks:
                 if len(chunk) > 2**20:
                     pos = 0
                     while pos < len(chunk):
                         end = pos + 2 ** 18
                         yield chunk[pos:end]
                         pos = end
                 else:
                     yield chunk
         self.iter = splitbig(in_iter)
         self._queue = []
 
     def read(self, l):
         """Read L bytes of data from the iterator of chunks of data.
         Returns less than L bytes if the iterator runs dry."""
         left = l
         buf = ''
         queue = self._queue
         while left > 0:
             # refill the queue
             if not queue:
                 target = 2**18
                 for chunk in self.iter:
                     queue.append(chunk)
                     target -= len(chunk)
                     if target <= 0:
                         break
                 if not queue:
                     break
 
             chunk = queue.pop(0)
             left -= len(chunk)
             if left < 0:
                 queue.insert(0, chunk[left:])
                 buf += chunk[:left]
             else:
                 buf += chunk
 
         return buf
 
 def filechunkiter(f, size=65536, limit=None):
     """Create a generator that produces the data in the file size
     (default 65536) bytes at a time, up to optional limit (default is
     to read all data).  Chunks may be less than size bytes if the
     chunk is the last chunk in the file, or the file is a socket or
     some other type of file that sometimes reads less data than is
     requested."""
     assert size >= 0
     assert limit is None or limit >= 0
     while True:
         if limit is None:
             nbytes = size
         else:
             nbytes = min(limit, size)
         s = nbytes and f.read(nbytes)
         if not s:
             break
         if limit:
             limit -= len(s)
         yield s
 
 def makedate():
     ct = time.time()
     if ct < 0:
         hint = _("check your clock")
         raise Abort(_("negative timestamp: %d") % ct, hint=hint)
     delta = (datetime.datetime.utcfromtimestamp(ct) -
              datetime.datetime.fromtimestamp(ct))
     tz = delta.days * 86400 + delta.seconds
     return ct, tz
 
 def datestr(date=None, format='%a %b %d %H:%M:%S %Y %1%2'):
     """represent a (unixtime, offset) tuple as a localized time.
     unixtime is seconds since the epoch, and offset is the time zone's
     number of seconds away from UTC. if timezone is false, do not
     append time zone to string."""
     t, tz = date or makedate()
     if t < 0:
         t = 0   # time.gmtime(lt) fails on Windows for lt < -43200
         tz = 0
     if "%1" in format or "%2" in format:
         sign = (tz > 0) and "-" or "+"
         minutes = abs(tz) // 60
         format = format.replace("%1", "%c%02d" % (sign, minutes // 60))
         format = format.replace("%2", "%02d" % (minutes % 60))
     try:
         t = time.gmtime(float(t) - tz)
     except ValueError:
         # time was out of range
         t = time.gmtime(sys.maxint)
     s = time.strftime(format, t)
     return s
 
 def shortdate(date=None):
     """turn (timestamp, tzoff) tuple into iso 8631 date."""
     return datestr(date, format='%Y-%m-%d')
 
 def strdate(string, format, defaults=[]):
     """parse a localized time string and return a (unixtime, offset) tuple.
     if the string cannot be parsed, ValueError is raised."""
     def timezone(string):
         tz = string.split()[-1]
         if tz[0] in "+-" and len(tz) == 5 and tz[1:].isdigit():
             sign = (tz[0] == "+") and 1 or -1
             hours = int(tz[1:3])
             minutes = int(tz[3:5])
             return -sign * (hours * 60 + minutes) * 60
         if tz == "GMT" or tz == "UTC":
             return 0
         return None
 
     # NOTE: unixtime = localunixtime + offset
     offset, date = timezone(string), string
     if offset is not None:
         date = " ".join(string.split()[:-1])
 
     # add missing elements from defaults
     usenow = False # default to using biased defaults
     for part in ("S", "M", "HI", "d", "mb", "yY"): # decreasing specificity
         found = [True for p in part if ("%"+p) in format]
         if not found:
             date += "@" + defaults[part][usenow]
             format += "@%" + part[0]
         else:
             # We've found a specific time element, less specific time
             # elements are relative to today
             usenow = True
 
     timetuple = time.strptime(date, format)
     localunixtime = int(calendar.timegm(timetuple))
     if offset is None:
         # local timezone
         unixtime = int(time.mktime(timetuple))
         offset = unixtime - localunixtime
     else:
         unixtime = localunixtime + offset
     return unixtime, offset
 
 def parsedate(date, formats=None, bias={}):
     """parse a localized date/time and return a (unixtime, offset) tuple.
 
     The date may be a "unixtime offset" string or in one of the specified
     formats. If the date already is a (unixtime, offset) tuple, it is returned.
     """
     if not date:
         return 0, 0
     if isinstance(date, tuple) and len(date) == 2:
         return date
     if not formats:
         formats = defaultdateformats
     date = date.strip()
     try:
         when, offset = map(int, date.split(' '))
     except ValueError:
         # fill out defaults
         now = makedate()
         defaults = {}
         for part in ("d", "mb", "yY", "HI", "M", "S"):
             # this piece is for rounding the specific end of unknowns
             b = bias.get(part)
             if b is None:
                 if part[0] in "HMS":
                     b = "00"
                 else:
                     b = "0"
 
             # this piece is for matching the generic end to today's date
             n = datestr(now, "%" + part[0])
 
             defaults[part] = (b, n)
 
         for format in formats:
             try:
                 when, offset = strdate(date, format, defaults)
             except (ValueError, OverflowError):
                 pass
             else:
                 break
         else:
             raise Abort(_('invalid date: %r') % date)
     # validate explicit (probably user-specified) date and
     # time zone offset. values must fit in signed 32 bits for
     # current 32-bit linux runtimes. timezones go from UTC-12
     # to UTC+14
     if abs(when) > 0x7fffffff:
         raise Abort(_('date exceeds 32 bits: %d') % when)
     if when < 0:
         raise Abort(_('negative date value: %d') % when)
     if offset < -50400 or offset > 43200:
         raise Abort(_('impossible time zone offset: %d') % offset)
     return when, offset
 
 def matchdate(date):
     """Return a function that matches a given date match specifier
 
     Formats include:
 
     '{date}' match a given date to the accuracy provided
 
     '<{date}' on or before a given date
 
     '>{date}' on or after a given date
 
     >>> p1 = parsedate("10:29:59")
     >>> p2 = parsedate("10:30:00")
     >>> p3 = parsedate("10:30:59")
     >>> p4 = parsedate("10:31:00")
     >>> p5 = parsedate("Sep 15 10:30:00 1999")
     >>> f = matchdate("10:30")
     >>> f(p1[0])
     False
     >>> f(p2[0])
     True
     >>> f(p3[0])
     True
     >>> f(p4[0])
     False
     >>> f(p5[0])
     False
     """
 
     def lower(date):
         d = dict(mb="1", d="1")
         return parsedate(date, extendeddateformats, d)[0]
 
     def upper(date):
         d = dict(mb="12", HI="23", M="59", S="59")
         for days in ("31", "30", "29"):
             try:
                 d["d"] = days
                 return parsedate(date, extendeddateformats, d)[0]
             except:
                 pass
         d["d"] = "28"
         return parsedate(date, extendeddateformats, d)[0]
 
     date = date.strip()
 
     if not date:
         raise Abort(_("dates cannot consist entirely of whitespace"))
     elif date[0] == "<":
         if not date[1:]:
             raise Abort(_("invalid day spec, use '<DATE'"))
         when = upper(date[1:])
         return lambda x: x <= when
     elif date[0] == ">":
         if not date[1:]:
             raise Abort(_("invalid day spec, use '>DATE'"))
         when = lower(date[1:])
         return lambda x: x >= when
     elif date[0] == "-":
         try:
             days = int(date[1:])
         except ValueError:
             raise Abort(_("invalid day spec: %s") % date[1:])
         if days < 0:
             raise Abort(_("%s must be nonnegative (see 'hg help dates')")
                 % date[1:])
         when = makedate()[0] - days * 3600 * 24
         return lambda x: x >= when
     elif " to " in date:
         a, b = date.split(" to ")
         start, stop = lower(a), upper(b)
         return lambda x: x >= start and x <= stop
     else:
         start, stop = lower(date), upper(date)
         return lambda x: x >= start and x <= stop
 
 def shortuser(user):
     """Return a short representation of a user name or email address."""
     f = user.find('@')
     if f >= 0:
         user = user[:f]
     f = user.find('<')
     if f >= 0:
         user = user[f + 1:]
     f = user.find(' ')
     if f >= 0:
         user = user[:f]
     f = user.find('.')
     if f >= 0:
         user = user[:f]
     return user
 
 def email(author):
     '''get email of author.'''
     r = author.find('>')
     if r == -1:
         r = None
     return author[author.find('<') + 1:r]
 
 def _ellipsis(text, maxlength):
     if len(text) <= maxlength:
         return text, False
     else:
         return "%s..." % (text[:maxlength - 3]), True
 
 def ellipsis(text, maxlength=400):
     """Trim string to at most maxlength (default: 400) characters."""
     try:
         # use unicode not to split at intermediate multi-byte sequence
         utext, truncated = _ellipsis(text.decode(encoding.encoding),
                                      maxlength)
         if not truncated:
             return text
         return utext.encode(encoding.encoding)
     except (UnicodeDecodeError, UnicodeEncodeError):
         return _ellipsis(text, maxlength)[0]
 
 def bytecount(nbytes):
     '''return byte count formatted as readable string, with units'''
 
     units = (
         (100, 1 << 30, _('%.0f GB')),
         (10, 1 << 30, _('%.1f GB')),
         (1, 1 << 30, _('%.2f GB')),
         (100, 1 << 20, _('%.0f MB')),
         (10, 1 << 20, _('%.1f MB')),
         (1, 1 << 20, _('%.2f MB')),
         (100, 1 << 10, _('%.0f KB')),
         (10, 1 << 10, _('%.1f KB')),
         (1, 1 << 10, _('%.2f KB')),
         (1, 1, _('%.0f bytes')),
         )
 
     for multiplier, divisor, format in units:
         if nbytes >= divisor * multiplier:
             return format % (nbytes / float(divisor))
     return units[-1][2] % nbytes
 
 def uirepr(s):
     # Avoid double backslash in Windows path repr()
     return repr(s).replace('\\\\', '\\')
 
 # delay import of textwrap
 def MBTextWrapper(**kwargs):
     class tw(textwrap.TextWrapper):
         """
         Extend TextWrapper for width-awareness.
 
         Neither number of 'bytes' in any encoding nor 'characters' is
         appropriate to calculate terminal columns for specified string.
 
         Original TextWrapper implementation uses built-in 'len()' directly,
         so overriding is needed to use width information of each characters.
 
         In addition, characters classified into 'ambiguous' width are
         treated as wide in east asian area, but as narrow in other.
 
         This requires use decision to determine width of such characters.
         """
         def __init__(self, **kwargs):
             textwrap.TextWrapper.__init__(self, **kwargs)
 
             # for compatibility between 2.4 and 2.6
             if getattr(self, 'drop_whitespace', None) is None:
                 self.drop_whitespace = kwargs.get('drop_whitespace', True)
 
         def _cutdown(self, ucstr, space_left):
             l = 0
             colwidth = encoding.ucolwidth
             for i in xrange(len(ucstr)):
                 l += colwidth(ucstr[i])
                 if space_left < l:
                     return (ucstr[:i], ucstr[i:])
             return ucstr, ''
 
         # overriding of base class
         def _handle_long_word(self, reversed_chunks, cur_line, cur_len, width):
             space_left = max(width - cur_len, 1)
 
             if self.break_long_words:
                 cut, res = self._cutdown(reversed_chunks[-1], space_left)
                 cur_line.append(cut)
                 reversed_chunks[-1] = res
             elif not cur_line:
                 cur_line.append(reversed_chunks.pop())
 
         # this overriding code is imported from TextWrapper of python 2.6
         # to calculate columns of string by 'encoding.ucolwidth()'
         def _wrap_chunks(self, chunks):
             colwidth = encoding.ucolwidth
 
             lines = []
             if self.width <= 0:
                 raise ValueError("invalid width %r (must be > 0)" % self.width)
 
             # Arrange in reverse order so items can be efficiently popped
             # from a stack of chucks.
             chunks.reverse()
 
             while chunks:
 
                 # Start the list of chunks that will make up the current line.
                 # cur_len is just the length of all the chunks in cur_line.
                 cur_line = []
                 cur_len = 0
 
                 # Figure out which static string will prefix this line.
                 if lines:
                     indent = self.subsequent_indent
                 else:
                     indent = self.initial_indent
 
                 # Maximum width for this line.
                 width = self.width - len(indent)
 
                 # First chunk on line is whitespace -- drop it, unless this
                 # is the very beginning of the text (ie. no lines started yet).
                 if self.drop_whitespace and chunks[-1].strip() == '' and lines:
                     del chunks[-1]
 
                 while chunks:
                     l = colwidth(chunks[-1])
 
                     # Can at least squeeze this chunk onto the current line.
                     if cur_len + l <= width:
                         cur_line.append(chunks.pop())
                         cur_len += l
 
                     # Nope, this line is full.
                     else:
                         break
 
                 # The current line is full, and the next chunk is too big to
                 # fit on *any* line (not just this one).
                 if chunks and colwidth(chunks[-1]) > width:
                     self._handle_long_word(chunks, cur_line, cur_len, width)
 
                 # If the last chunk on this line is all whitespace, drop it.
                 if (self.drop_whitespace and
                     cur_line and cur_line[-1].strip() == ''):
                     del cur_line[-1]
 
                 # Convert current line back to a string and store it in list
                 # of all lines (return value).
                 if cur_line:
                     lines.append(indent + ''.join(cur_line))
 
             return lines
 
     global MBTextWrapper
     MBTextWrapper = tw
     return tw(**kwargs)
 
 def wrap(line, width, initindent='', hangindent=''):
     maxindent = max(len(hangindent), len(initindent))
     if width <= maxindent:
         # adjust for weird terminal size
         width = max(78, maxindent + 1)
     line = line.decode(encoding.encoding, encoding.encodingmode)
     initindent = initindent.decode(encoding.encoding, encoding.encodingmode)
     hangindent = hangindent.decode(encoding.encoding, encoding.encodingmode)
     wrapper = MBTextWrapper(width=width,
                             initial_indent=initindent,
                             subsequent_indent=hangindent)
     return wrapper.fill(line).encode(encoding.encoding)
 
 def iterlines(iterator):
     for chunk in iterator:
         for line in chunk.splitlines():
             yield line
 
 def expandpath(path):
     return os.path.expanduser(os.path.expandvars(path))
 
 def hgcmd():
     """Return the command used to execute current hg
 
     This is different from hgexecutable() because on Windows we want
     to avoid things opening new shell windows like batch files, so we
     get either the python call or current executable.
     """
     if mainfrozen():
         return [sys.executable]
     return gethgcmd()
 
 def rundetached(args, condfn):
     """Execute the argument list in a detached process.
 
     condfn is a callable which is called repeatedly and should return
     True once the child process is known to have started successfully.
     At this point, the child process PID is returned. If the child
     process fails to start or finishes before condfn() evaluates to
     True, return -1.
     """
     # Windows case is easier because the child process is either
     # successfully starting and validating the condition or exiting
     # on failure. We just poll on its PID. On Unix, if the child
     # process fails to start, it will be left in a zombie state until
     # the parent wait on it, which we cannot do since we expect a long
     # running process on success. Instead we listen for SIGCHLD telling
     # us our child process terminated.
     terminated = set()
     def handler(signum, frame):
         terminated.add(os.wait())
     prevhandler = None
     SIGCHLD = getattr(signal, 'SIGCHLD', None)
     if SIGCHLD is not None:
         prevhandler = signal.signal(SIGCHLD, handler)
     try:
         pid = spawndetached(args)
         while not condfn():
             if ((pid in terminated or not testpid(pid))
                 and not condfn()):
                 return -1
             time.sleep(0.1)
         return pid
     finally:
         if prevhandler is not None:
             signal.signal(signal.SIGCHLD, prevhandler)
 
 try:
     any, all = any, all
 except NameError:
     def any(iterable):
         for i in iterable:
             if i:
                 return True
         return False
 
     def all(iterable):
         for i in iterable:
             if not i:
                 return False
         return True
 
 def interpolate(prefix, mapping, s, fn=None, escape_prefix=False):
     """Return the result of interpolating items in the mapping into string s.
 
     prefix is a single character string, or a two character string with
     a backslash as the first character if the prefix needs to be escaped in
     a regular expression.
 
     fn is an optional function that will be applied to the replacement text
     just before replacement.
 
     escape_prefix is an optional flag that allows using doubled prefix for
     its escaping.
     """
     fn = fn or (lambda s: s)
     patterns = '|'.join(mapping.keys())
     if escape_prefix:
         patterns += '|' + prefix
         if len(prefix) > 1:
             prefix_char = prefix[1:]
         else:
             prefix_char = prefix
         mapping[prefix_char] = prefix_char
     r = re.compile(r'%s(%s)' % (prefix, patterns))
     return r.sub(lambda x: fn(mapping[x.group()[1:]]), s)
 
 def getport(port):
     """Return the port for a given network service.
 
     If port is an integer, it's returned as is. If it's a string, it's
     looked up using socket.getservbyname(). If there's no matching
     service, util.Abort is raised.
     """
     try:
         return int(port)
     except ValueError:
         pass
 
     try:
         return socket.getservbyname(port)
     except socket.error:
         raise Abort(_("no port number associated with service '%s'") % port)
 
 _booleans = {'1': True, 'yes': True, 'true': True, 'on': True, 'always': True,
              '0': False, 'no': False, 'false': False, 'off': False,
              'never': False}
 
 def parsebool(s):
     """Parse s into a boolean.
 
     If s is not a valid boolean, returns None.
     """
     return _booleans.get(s.lower(), None)
 
 _hexdig = '0123456789ABCDEFabcdef'
 _hextochr = dict((a + b, chr(int(a + b, 16)))
                  for a in _hexdig for b in _hexdig)
 
 def _urlunquote(s):
     """unquote('abc%20def') -> 'abc def'."""
     res = s.split('%')
     # fastpath
     if len(res) == 1:
         return s
     s = res[0]
     for item in res[1:]:
         try:
             s += _hextochr[item[:2]] + item[2:]
         except KeyError:
             s += '%' + item
         except UnicodeDecodeError:
             s += unichr(int(item[:2], 16)) + item[2:]
     return s
 
 class url(object):
     r"""Reliable URL parser.
 
     This parses URLs and provides attributes for the following
     components:
 
     <scheme>://<user>:<passwd>@<host>:<port>/<path>?<query>#<fragment>
 
     Missing components are set to None. The only exception is
     fragment, which is set to '' if present but empty.
 
     If parsefragment is False, fragment is included in query. If
     parsequery is False, query is included in path. If both are
     False, both fragment and query are included in path.
 
     See http://www.ietf.org/rfc/rfc2396.txt for more information.
 
     Note that for backward compatibility reasons, bundle URLs do not
     take host names. That means 'bundle://../' has a path of '../'.
 
     Examples:
 
     >>> url('http://www.ietf.org/rfc/rfc2396.txt')
     <url scheme: 'http', host: 'www.ietf.org', path: 'rfc/rfc2396.txt'>
     >>> url('ssh://[::1]:2200//home/joe/repo')
     <url scheme: 'ssh', host: '[::1]', port: '2200', path: '/home/joe/repo'>
     >>> url('file:///home/joe/repo')
     <url scheme: 'file', path: '/home/joe/repo'>
     >>> url('file:///c:/temp/foo/')
     <url scheme: 'file', path: 'c:/temp/foo/'>
     >>> url('bundle:foo')
     <url scheme: 'bundle', path: 'foo'>
     >>> url('bundle://../foo')
     <url scheme: 'bundle', path: '../foo'>
     >>> url(r'c:\foo\bar')
     <url path: 'c:\\foo\\bar'>
     >>> url(r'\\blah\blah\blah')
     <url path: '\\\\blah\\blah\\blah'>
     >>> url(r'\\blah\blah\blah#baz')
     <url path: '\\\\blah\\blah\\blah', fragment: 'baz'>
 
     Authentication credentials:
 
     >>> url('ssh://joe:xyz@x/repo')
     <url scheme: 'ssh', user: 'joe', passwd: 'xyz', host: 'x', path: 'repo'>
     >>> url('ssh://joe@x/repo')
     <url scheme: 'ssh', user: 'joe', host: 'x', path: 'repo'>
 
     Query strings and fragments:
 
     >>> url('http://host/a?b#c')
     <url scheme: 'http', host: 'host', path: 'a', query: 'b', fragment: 'c'>
     >>> url('http://host/a?b#c', parsequery=False, parsefragment=False)
     <url scheme: 'http', host: 'host', path: 'a?b#c'>
     """
 
     _safechars = "!~*'()+"
     _safepchars = "/!~*'()+:"
     _matchscheme = re.compile(r'^[a-zA-Z0-9+.\-]+:').match
 
     def __init__(self, path, parsequery=True, parsefragment=True):
         # We slowly chomp away at path until we have only the path left
         self.scheme = self.user = self.passwd = self.host = None
         self.port = self.path = self.query = self.fragment = None
         self._localpath = True
         self._hostport = ''
         self._origpath = path
 
         if parsefragment and '#' in path:
             path, self.fragment = path.split('#', 1)
             if not path:
                 path = None
 
         # special case for Windows drive letters and UNC paths
         if hasdriveletter(path) or path.startswith(r'\\'):
             self.path = path
             return
 
         # For compatibility reasons, we can't handle bundle paths as
         # normal URLS
         if path.startswith('bundle:'):
             self.scheme = 'bundle'
             path = path[7:]
             if path.startswith('//'):
                 path = path[2:]
             self.path = path
             return
 
         if self._matchscheme(path):
             parts = path.split(':', 1)
             if parts[0]:
                 self.scheme, path = parts
                 self._localpath = False
 
         if not path:
             path = None
             if self._localpath:
                 self.path = ''
                 return
         else:
             if self._localpath:
                 self.path = path
                 return
 
             if parsequery and '?' in path:
                 path, self.query = path.split('?', 1)
                 if not path:
                     path = None
                 if not self.query:
                     self.query = None
 
             # // is required to specify a host/authority
             if path and path.startswith('//'):
                 parts = path[2:].split('/', 1)
                 if len(parts) > 1:
                     self.host, path = parts
                     path = path
                 else:
                     self.host = parts[0]
                     path = None
                 if not self.host:
                     self.host = None
                     # path of file:///d is /d
                     # path of file:///d:/ is d:/, not /d:/
                     if path and not hasdriveletter(path):
                         path = '/' + path
 
             if self.host and '@' in self.host:
                 self.user, self.host = self.host.rsplit('@', 1)
                 if ':' in self.user:
                     self.user, self.passwd = self.user.split(':', 1)
                 if not self.host:
                     self.host = None
 
             # Don't split on colons in IPv6 addresses without ports
             if (self.host and ':' in self.host and
                 not (self.host.startswith('[') and self.host.endswith(']'))):
                 self._hostport = self.host
                 self.host, self.port = self.host.rsplit(':', 1)
                 if not self.host:
                     self.host = None
 
             if (self.host and self.scheme == 'file' and
                 self.host not in ('localhost', '127.0.0.1', '[::1]')):
                 raise Abort(_('file:// URLs can only refer to localhost'))
 
         self.path = path
 
         # leave the query string escaped
         for a in ('user', 'passwd', 'host', 'port',
                   'path', 'fragment'):
             v = getattr(self, a)
             if v is not None:
                 setattr(self, a, _urlunquote(v))
 
     def __repr__(self):
         attrs = []
         for a in ('scheme', 'user', 'passwd', 'host', 'port', 'path',
                   'query', 'fragment'):
             v = getattr(self, a)
             if v is not None:
                 attrs.append('%s: %r' % (a, v))
         return '<url %s>' % ', '.join(attrs)
 
     def __str__(self):
         r"""Join the URL's components back into a URL string.
 
         Examples:
 
         >>> str(url('http://user:pw@host:80/c:/bob?fo:oo#ba:ar'))
         'http://user:pw@host:80/c:/bob?fo:oo#ba:ar'
         >>> str(url('http://user:pw@host:80/?foo=bar&baz=42'))
         'http://user:pw@host:80/?foo=bar&baz=42'
         >>> str(url('http://user:pw@host:80/?foo=bar%3dbaz'))
         'http://user:pw@host:80/?foo=bar%3dbaz'
         >>> str(url('ssh://user:pw@[::1]:2200//home/joe#'))
         'ssh://user:pw@[::1]:2200//home/joe#'
         >>> str(url('http://localhost:80//'))
         'http://localhost:80//'
         >>> str(url('http://localhost:80/'))
         'http://localhost:80/'
         >>> str(url('http://localhost:80'))
         'http://localhost:80/'
         >>> str(url('bundle:foo'))
         'bundle:foo'
         >>> str(url('bundle://../foo'))
         'bundle:../foo'
         >>> str(url('path'))
         'path'
         >>> str(url('file:///tmp/foo/bar'))
         'file:///tmp/foo/bar'
         >>> str(url('file:///c:/tmp/foo/bar'))
         'file:///c:/tmp/foo/bar'
         >>> print url(r'bundle:foo\bar')
         bundle:foo\bar
         """
         if self._localpath:
             s = self.path
             if self.scheme == 'bundle':
                 s = 'bundle:' + s
             if self.fragment:
                 s += '#' + self.fragment
             return s
 
         s = self.scheme + ':'
         if self.user or self.passwd or self.host:
             s += '//'
         elif self.scheme and (not self.path or self.path.startswith('/')
                               or hasdriveletter(self.path)):
             s += '//'
             if hasdriveletter(self.path):
                 s += '/'
         if self.user:
             s += urllib.quote(self.user, safe=self._safechars)
         if self.passwd:
             s += ':' + urllib.quote(self.passwd, safe=self._safechars)
         if self.user or self.passwd:
             s += '@'
         if self.host:
             if not (self.host.startswith('[') and self.host.endswith(']')):
                 s += urllib.quote(self.host)
             else:
                 s += self.host
         if self.port:
             s += ':' + urllib.quote(self.port)
         if self.host:
             s += '/'
         if self.path:
             # TODO: similar to the query string, we should not unescape the
             # path when we store it, the path might contain '%2f' = '/',
             # which we should *not* escape.
             s += urllib.quote(self.path, safe=self._safepchars)
         if self.query:
             # we store the query in escaped form.
             s += '?' + self.query
         if self.fragment is not None:
             s += '#' + urllib.quote(self.fragment, safe=self._safepchars)
         return s
 
     def authinfo(self):
         user, passwd = self.user, self.passwd
         try:
             self.user, self.passwd = None, None
             s = str(self)
         finally:
             self.user, self.passwd = user, passwd
         if not self.user:
             return (s, None)
         # authinfo[1] is passed to urllib2 password manager, and its
         # URIs must not contain credentials. The host is passed in the
         # URIs list because Python < 2.4.3 uses only that to search for
         # a password.
         return (s, (None, (s, self.host),
                     self.user, self.passwd or ''))
 
     def isabs(self):
         if self.scheme and self.scheme != 'file':
             return True # remote URL
         if hasdriveletter(self.path):
             return True # absolute for our purposes - can't be joined()
         if self.path.startswith(r'\\'):
             return True # Windows UNC path
         if self.path.startswith('/'):
             return True # POSIX-style
         return False
 
     def localpath(self):
         if self.scheme == 'file' or self.scheme == 'bundle':
             path = self.path or '/'
             # For Windows, we need to promote hosts containing drive
             # letters to paths with drive letters.
             if hasdriveletter(self._hostport):
                 path = self._hostport + '/' + self.path
             elif (self.host is not None and self.path
                   and not hasdriveletter(path)):
                 path = '/' + path
             return path
         return self._origpath
 
 def hasscheme(path):
     return bool(url(path).scheme)
 
 def hasdriveletter(path):
     return path and path[1:2] == ':' and path[0:1].isalpha()
 
 def urllocalpath(path):
     return url(path, parsequery=False, parsefragment=False).localpath()
 
 def hidepassword(u):
     '''hide user credential in a url string'''
     u = url(u)
     if u.passwd:
         u.passwd = '***'
     return str(u)
 
 def removeauth(u):
     '''remove all authentication information from a url string'''
     u = url(u)
     u.user = u.passwd = None
     return str(u)
 
 def isatty(fd):
     try:
         return fd.isatty()
     except AttributeError:
         return False

mercurial/windows.py

@@ -1,315 +1,319
 # windows.py - Windows utility function implementations for Mercurial
 #
 #  Copyright 2005-2009 Matt Mackall <[e-mail]> and others
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from i18n import _
 import osutil
 import errno, msvcrt, os, re, sys
 
 import win32
 executablepath = win32.executablepath
 getuser = win32.getuser
 hidewindow = win32.hidewindow
 lookupreg = win32.lookupreg
 makedir = win32.makedir
 nlinks = win32.nlinks
 oslink = win32.oslink
 samedevice = win32.samedevice
 samefile = win32.samefile
 setsignalhandler = win32.setsignalhandler
 spawndetached = win32.spawndetached
 termwidth = win32.termwidth
 testpid = win32.testpid
 unlink = win32.unlink
 
 nulldev = 'NUL:'
 umask = 0022
 
 # wrap osutil.posixfile to provide friendlier exceptions
 def posixfile(name, mode='r', buffering=-1):
     try:
         return osutil.posixfile(name, mode, buffering)
     except WindowsError, err:
         raise IOError(err.errno, '%s: %s' % (name, err.strerror))
 posixfile.__doc__ = osutil.posixfile.__doc__
 
 class winstdout(object):
     '''stdout on windows misbehaves if sent through a pipe'''
 
     def __init__(self, fp):
         self.fp = fp
 
     def __getattr__(self, key):
         return getattr(self.fp, key)
 
     def close(self):
         try:
             self.fp.close()
         except IOError:
             pass
 
     def write(self, s):
         try:
             # This is workaround for "Not enough space" error on
             # writing large size of data to console.
             limit = 16000
             l = len(s)
             start = 0
             self.softspace = 0
             while start < l:
                 end = start + limit
                 self.fp.write(s[start:end])
                 start = end
         except IOError, inst:
             if inst.errno != 0:
                 raise
             self.close()
             raise IOError(errno.EPIPE, 'Broken pipe')
 
     def flush(self):
         try:
             return self.fp.flush()
         except IOError, inst:
             if inst.errno != errno.EINVAL:
                 raise
             self.close()
             raise IOError(errno.EPIPE, 'Broken pipe')
 
 sys.__stdout__ = sys.stdout = winstdout(sys.stdout)
 
 def _is_win_9x():
     '''return true if run on windows 95, 98 or me.'''
     try:
         return sys.getwindowsversion()[3] == 1
     except AttributeError:
         return 'command' in os.environ.get('comspec', '')
 
 def openhardlinks():
     return not _is_win_9x()
 
 def parsepatchoutput(output_line):
     """parses the output produced by patch and returns the filename"""
     pf = output_line[14:]
     if pf[0] == '`':
         pf = pf[1:-1] # Remove the quotes
     return pf
 
 def sshargs(sshcmd, host, user, port):
     '''Build argument list for ssh or Plink'''
     pflag = 'plink' in sshcmd.lower() and '-P' or '-p'
     args = user and ("%s@%s" % (user, host)) or host
     return port and ("%s %s %s" % (args, pflag, port)) or args
 
 def setflags(f, l, x):
     pass
 
 def copymode(src, dst, mode=None):
     pass
 
 def checkexec(path):
     return False
 
 def checklink(path):
     return False
 
 def setbinary(fd):
     # When run without console, pipes may expose invalid
     # fileno(), usually set to -1.
     fno = getattr(fd, 'fileno', None)
     if fno is not None and fno() >= 0:
         msvcrt.setmode(fno(), os.O_BINARY)
 
 def pconvert(path):
     return '/'.join(path.split(os.sep))
 
 def localpath(path):
     return path.replace('/', '\\')
 
 def normpath(path):
     return pconvert(os.path.normpath(path))
 
-normcase = os.path.normcase
+encodinglower = None
+encodingupper = None
+
+def normcase(path):
+    return encodingupper(path)
 
 def realpath(path):
     '''
     Returns the true, canonical file system path equivalent to the given
     path.
     '''
     # TODO: There may be a more clever way to do this that also handles other,
     # less common file systems.
     return os.path.normpath(normcase(os.path.realpath(path)))
 
 def samestat(s1, s2):
     return False
 
 # A sequence of backslashes is special iff it precedes a double quote:
 # - if there's an even number of backslashes, the double quote is not
 #   quoted (i.e. it ends the quoted region)
 # - if there's an odd number of backslashes, the double quote is quoted
 # - in both cases, every pair of backslashes is unquoted into a single
 #   backslash
 # (See http://msdn2.microsoft.com/en-us/library/a1y7w461.aspx )
 # So, to quote a string, we must surround it in double quotes, double
 # the number of backslashes that preceed double quotes and add another
 # backslash before every double quote (being careful with the double
 # quote we've appended to the end)
 _quotere = None
 def shellquote(s):
     global _quotere
     if _quotere is None:
         _quotere = re.compile(r'(\\*)("|\\$)')
     return '"%s"' % _quotere.sub(r'\1\1\\\2', s)
 
 def quotecommand(cmd):
     """Build a command string suitable for os.popen* calls."""
     if sys.version_info < (2, 7, 1):
         # Python versions since 2.7.1 do this extra quoting themselves
         return '"' + cmd + '"'
     return cmd
 
 def popen(command, mode='r'):
     # Work around "popen spawned process may not write to stdout
     # under windows"
     # http://bugs.python.org/issue1366
     command += " 2> %s" % nulldev
     return os.popen(quotecommand(command), mode)
 
 def explainexit(code):
     return _("exited with status %d") % code, code
 
 # if you change this stub into a real check, please try to implement the
 # username and groupname functions above, too.
 def isowner(st):
     return True
 
 def findexe(command):
     '''Find executable for command searching like cmd.exe does.
     If command is a basename then PATH is searched for command.
     PATH isn't searched if command is an absolute or relative path.
     An extension from PATHEXT is found and added if not present.
     If command isn't found None is returned.'''
     pathext = os.environ.get('PATHEXT', '.COM;.EXE;.BAT;.CMD')
     pathexts = [ext for ext in pathext.lower().split(os.pathsep)]
     if os.path.splitext(command)[1].lower() in pathexts:
         pathexts = ['']
 
     def findexisting(pathcommand):
         'Will append extension (if needed) and return existing file'
         for ext in pathexts:
             executable = pathcommand + ext
             if os.path.exists(executable):
                 return executable
         return None
 
     if os.sep in command:
         return findexisting(command)
 
     for path in os.environ.get('PATH', '').split(os.pathsep):
         executable = findexisting(os.path.join(path, command))
         if executable is not None:
             return executable
     return findexisting(os.path.expanduser(os.path.expandvars(command)))
 
 def statfiles(files):
     '''Stat each file in files and yield stat or None if file does not exist.
     Cluster and cache stat per directory to minimize number of OS stat calls.'''
     dircache = {} # dirname -> filename -> status | None if file does not exist
     for nf in files:
         nf  = normcase(nf)
         dir, base = os.path.split(nf)
         if not dir:
             dir = '.'
         cache = dircache.get(dir, None)
         if cache is None:
             try:
                 dmap = dict([(normcase(n), s)
                     for n, k, s in osutil.listdir(dir, True)])
             except OSError, err:
                 # handle directory not found in Python version prior to 2.5
                 # Python <= 2.4 returns native Windows code 3 in errno
                 # Python >= 2.5 returns ENOENT and adds winerror field
                 # EINVAL is raised if dir is not a directory.
                 if err.errno not in (3, errno.ENOENT, errno.EINVAL,
                                      errno.ENOTDIR):
                     raise
                 dmap = {}
             cache = dircache.setdefault(dir, dmap)
         yield cache.get(base, None)
 
 def username(uid=None):
     """Return the name of the user with the given uid.
 
     If uid is None, return the name of the current user."""
     return None
 
 def groupname(gid=None):
     """Return the name of the group with the given gid.
 
     If gid is None, return the name of the current group."""
     return None
 
 def _removedirs(name):
     """special version of os.removedirs that does not remove symlinked
     directories or junction points if they actually contain files"""
     if osutil.listdir(name):
         return
     os.rmdir(name)
     head, tail = os.path.split(name)
     if not tail:
         head, tail = os.path.split(head)
     while head and tail:
         try:
             if osutil.listdir(head):
                 return
             os.rmdir(head)
         except (ValueError, OSError):
             break
         head, tail = os.path.split(head)
 
 def unlinkpath(f):
     """unlink and remove the directory if it is empty"""
     unlink(f)
     # try removing directories that might now be empty
     try:
         _removedirs(os.path.dirname(f))
     except OSError:
         pass
 
 def rename(src, dst):
     '''atomically rename file src to dst, replacing dst if it exists'''
     try:
         os.rename(src, dst)
     except OSError, e:
         if e.errno != errno.EEXIST:
             raise
         unlink(dst)
         os.rename(src, dst)
 
 def gethgcmd():
     return [sys.executable] + sys.argv[:1]
 
 def termwidth():
     # cmd.exe does not handle CR like a unix console, the CR is
     # counted in the line length. On 80 columns consoles, if 80
     # characters are written, the following CR won't apply on the
     # current line but on the new one. Keep room for it.
     return 79
 
 def groupmembers(name):
     # Don't support groups on Windows for now
     raise KeyError()
 
 def isexec(f):
     return False
 
 class cachestat(object):
     def __init__(self, path):
         pass
 
     def cacheable(self):
         return False
 
 expandglobs = True