fncache: avoid loading the filename cache when not actually modifying it

With time, fncache can become very large. The mozilla-central repo for example,
has a 31M and growing fncache file. Loading this file takes time (280ms for the
mozilla-central repository).

In many scenarios, we don't need to load fncache at all. For example, when
committing changes to existing files, or pushing such commits to another clone.
This patch detects when a name is added via store.vfs(), and only loads the
cache if a) the data metadata file doesn't already exist, or b) when opening
for appending, the data or metadata file exists but has size  (a transaction
rollback leaves behind such files).

Benchmarks (run on Macos 10.13 on a 2017-model Macbook Pro with Core i7 2.9GHz
and flash drive), each test without and with patch run 5 times:

* committing to an existing file, against the mozilla-central repository.
  Baseline real time average 2.3736, with patch 1.9884.

* unbundling a large changeset consisting *only* of existing-file modifications
  (159 revisions, 1050 modifications, mozilla-central
  4a250a0e4f29:beea9ac7d823), into a clone limited to the ancestor revision of
  that revset). Baseline real time average 1.5048, with patch 1.3108.

mercurial/store.py

@@ -1,582 +1,592 b''
 # store.py - repository store handling for Mercurial
 #
 # Copyright 2008 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 __future__ import absolute_import
 
 import errno
 import hashlib
 import os
 import stat
 
 from .i18n import _
 from . import (
     error,
     node,
     policy,
     pycompat,
     util,
     vfs as vfsmod,
 )
 
 parsers = policy.importmod(r'parsers')
 
 # This avoids a collision between a file named foo and a dir named
 # foo.i or foo.d
 def _encodedir(path):
     '''
     >>> _encodedir(b'data/foo.i')
     'data/foo.i'
     >>> _encodedir(b'data/foo.i/bla.i')
     'data/foo.i.hg/bla.i'
     >>> _encodedir(b'data/foo.i.hg/bla.i')
     'data/foo.i.hg.hg/bla.i'
     >>> _encodedir(b'data/foo.i\\ndata/foo.i/bla.i\\ndata/foo.i.hg/bla.i\\n')
     'data/foo.i\\ndata/foo.i.hg/bla.i\\ndata/foo.i.hg.hg/bla.i\\n'
     '''
     return (path
             .replace(".hg/", ".hg.hg/")
             .replace(".i/", ".i.hg/")
             .replace(".d/", ".d.hg/"))
 
 encodedir = getattr(parsers, 'encodedir', _encodedir)
 
 def decodedir(path):
     '''
     >>> decodedir(b'data/foo.i')
     'data/foo.i'
     >>> decodedir(b'data/foo.i.hg/bla.i')
     'data/foo.i/bla.i'
     >>> decodedir(b'data/foo.i.hg.hg/bla.i')
     'data/foo.i.hg/bla.i'
     '''
     if ".hg/" not in path:
         return path
     return (path
             .replace(".d.hg/", ".d/")
             .replace(".i.hg/", ".i/")
             .replace(".hg.hg/", ".hg/"))
 
 def _reserved():
     ''' characters that are problematic for filesystems
 
     * ascii escapes (0..31)
     * ascii hi (126..255)
     * windows specials
 
     these characters will be escaped by encodefunctions
     '''
     winreserved = [ord(x) for x in u'\\:*?"<>|']
     for x in range(32):
         yield x
     for x in range(126, 256):
         yield x
     for x in winreserved:
         yield x
 
 def _buildencodefun():
     '''
     >>> enc, dec = _buildencodefun()
 
     >>> enc(b'nothing/special.txt')
     'nothing/special.txt'
     >>> dec(b'nothing/special.txt')
     'nothing/special.txt'
 
     >>> enc(b'HELLO')
     '_h_e_l_l_o'
     >>> dec(b'_h_e_l_l_o')
     'HELLO'
 
     >>> enc(b'hello:world?')
     'hello~3aworld~3f'
     >>> dec(b'hello~3aworld~3f')
     'hello:world?'
 
     >>> enc(b'the\\x07quick\\xADshot')
     'the~07quick~adshot'
     >>> dec(b'the~07quick~adshot')
     'the\\x07quick\\xadshot'
     '''
     e = '_'
     xchr = pycompat.bytechr
     asciistr = list(map(xchr, range(127)))
     capitals = list(range(ord("A"), ord("Z") + 1))
 
     cmap = dict((x, x) for x in asciistr)
     for x in _reserved():
         cmap[xchr(x)] = "~%02x" % x
     for x in capitals + [ord(e)]:
         cmap[xchr(x)] = e + xchr(x).lower()
 
     dmap = {}
     for k, v in cmap.iteritems():
         dmap[v] = k
     def decode(s):
         i = 0
         while i < len(s):
             for l in xrange(1, 4):
                 try:
                     yield dmap[s[i:i + l]]
                     i += l
                     break
                 except KeyError:
                     pass
             else:
                 raise KeyError
     return (lambda s: ''.join([cmap[s[c:c + 1]] for c in xrange(len(s))]),
             lambda s: ''.join(list(decode(s))))
 
 _encodefname, _decodefname = _buildencodefun()
 
 def encodefilename(s):
     '''
     >>> encodefilename(b'foo.i/bar.d/bla.hg/hi:world?/HELLO')
     'foo.i.hg/bar.d.hg/bla.hg.hg/hi~3aworld~3f/_h_e_l_l_o'
     '''
     return _encodefname(encodedir(s))
 
 def decodefilename(s):
     '''
     >>> decodefilename(b'foo.i.hg/bar.d.hg/bla.hg.hg/hi~3aworld~3f/_h_e_l_l_o')
     'foo.i/bar.d/bla.hg/hi:world?/HELLO'
     '''
     return decodedir(_decodefname(s))
 
 def _buildlowerencodefun():
     '''
     >>> f = _buildlowerencodefun()
     >>> f(b'nothing/special.txt')
     'nothing/special.txt'
     >>> f(b'HELLO')
     'hello'
     >>> f(b'hello:world?')
     'hello~3aworld~3f'
     >>> f(b'the\\x07quick\\xADshot')
     'the~07quick~adshot'
     '''
     xchr = pycompat.bytechr
     cmap = dict([(xchr(x), xchr(x)) for x in xrange(127)])
     for x in _reserved():
         cmap[xchr(x)] = "~%02x" % x
     for x in range(ord("A"), ord("Z") + 1):
         cmap[xchr(x)] = xchr(x).lower()
     def lowerencode(s):
         return "".join([cmap[c] for c in pycompat.iterbytestr(s)])
     return lowerencode
 
 lowerencode = getattr(parsers, 'lowerencode', None) or _buildlowerencodefun()
 
 # Windows reserved names: con, prn, aux, nul, com1..com9, lpt1..lpt9
 _winres3 = ('aux', 'con', 'prn', 'nul') # length 3
 _winres4 = ('com', 'lpt')               # length 4 (with trailing 1..9)
 def _auxencode(path, dotencode):
     '''
     Encodes filenames containing names reserved by Windows or which end in
     period or space. Does not touch other single reserved characters c.
     Specifically, c in '\\:*?"<>|' or ord(c) <= 31 are *not* encoded here.
     Additionally encodes space or period at the beginning, if dotencode is
     True. Parameter path is assumed to be all lowercase.
     A segment only needs encoding if a reserved name appears as a
     basename (e.g. "aux", "aux.foo"). A directory or file named "foo.aux"
     doesn't need encoding.
 
     >>> s = b'.foo/aux.txt/txt.aux/con/prn/nul/foo.'
     >>> _auxencode(s.split(b'/'), True)
     ['~2efoo', 'au~78.txt', 'txt.aux', 'co~6e', 'pr~6e', 'nu~6c', 'foo~2e']
     >>> s = b'.com1com2/lpt9.lpt4.lpt1/conprn/com0/lpt0/foo.'
     >>> _auxencode(s.split(b'/'), False)
     ['.com1com2', 'lp~749.lpt4.lpt1', 'conprn', 'com0', 'lpt0', 'foo~2e']
     >>> _auxencode([b'foo. '], True)
     ['foo.~20']
     >>> _auxencode([b' .foo'], True)
     ['~20.foo']
     '''
     for i, n in enumerate(path):
         if not n:
             continue
         if dotencode and n[0] in '. ':
             n = "~%02x" % ord(n[0:1]) + n[1:]
             path[i] = n
         else:
             l = n.find('.')
             if l == -1:
                 l = len(n)
             if ((l == 3 and n[:3] in _winres3) or
                 (l == 4 and n[3:4] <= '9' and n[3:4] >= '1'
                         and n[:3] in _winres4)):
                 # encode third letter ('aux' -> 'au~78')
                 ec = "~%02x" % ord(n[2:3])
                 n = n[0:2] + ec + n[3:]
                 path[i] = n
         if n[-1] in '. ':
             # encode last period or space ('foo...' -> 'foo..~2e')
             path[i] = n[:-1] + "~%02x" % ord(n[-1:])
     return path
 
 _maxstorepathlen = 120
 _dirprefixlen = 8
 _maxshortdirslen = 8 * (_dirprefixlen + 1) - 4
 
 def _hashencode(path, dotencode):
     digest = node.hex(hashlib.sha1(path).digest())
     le = lowerencode(path[5:]).split('/') # skips prefix 'data/' or 'meta/'
     parts = _auxencode(le, dotencode)
     basename = parts[-1]
     _root, ext = os.path.splitext(basename)
     sdirs = []
     sdirslen = 0
     for p in parts[:-1]:
         d = p[:_dirprefixlen]
         if d[-1] in '. ':
             # Windows can't access dirs ending in period or space
             d = d[:-1] + '_'
         if sdirslen == 0:
             t = len(d)
         else:
             t = sdirslen + 1 + len(d)
             if t > _maxshortdirslen:
                 break
         sdirs.append(d)
         sdirslen = t
     dirs = '/'.join(sdirs)
     if len(dirs) > 0:
         dirs += '/'
     res = 'dh/' + dirs + digest + ext
     spaceleft = _maxstorepathlen - len(res)
     if spaceleft > 0:
         filler = basename[:spaceleft]
         res = 'dh/' + dirs + filler + digest + ext
     return res
 
 def _hybridencode(path, dotencode):
     '''encodes path with a length limit
 
     Encodes all paths that begin with 'data/', according to the following.
 
     Default encoding (reversible):
 
     Encodes all uppercase letters 'X' as '_x'. All reserved or illegal
     characters are encoded as '~xx', where xx is the two digit hex code
     of the character (see encodefilename).
     Relevant path components consisting of Windows reserved filenames are
     masked by encoding the third character ('aux' -> 'au~78', see _auxencode).
 
     Hashed encoding (not reversible):
 
     If the default-encoded path is longer than _maxstorepathlen, a
     non-reversible hybrid hashing of the path is done instead.
     This encoding uses up to _dirprefixlen characters of all directory
     levels of the lowerencoded path, but not more levels than can fit into
     _maxshortdirslen.
     Then follows the filler followed by the sha digest of the full path.
     The filler is the beginning of the basename of the lowerencoded path
     (the basename is everything after the last path separator). The filler
     is as long as possible, filling in characters from the basename until
     the encoded path has _maxstorepathlen characters (or all chars of the
     basename have been taken).
     The extension (e.g. '.i' or '.d') is preserved.
 
     The string 'data/' at the beginning is replaced with 'dh/', if the hashed
     encoding was used.
     '''
     path = encodedir(path)
     ef = _encodefname(path).split('/')
     res = '/'.join(_auxencode(ef, dotencode))
     if len(res) > _maxstorepathlen:
         res = _hashencode(path, dotencode)
     return res
 
 def _pathencode(path):
     de = encodedir(path)
     if len(path) > _maxstorepathlen:
         return _hashencode(de, True)
     ef = _encodefname(de).split('/')
     res = '/'.join(_auxencode(ef, True))
     if len(res) > _maxstorepathlen:
         return _hashencode(de, True)
     return res
 
 _pathencode = getattr(parsers, 'pathencode', _pathencode)
 
 def _plainhybridencode(f):
     return _hybridencode(f, False)
 
 def _calcmode(vfs):
     try:
         # files in .hg/ will be created using this mode
         mode = vfs.stat().st_mode
             # avoid some useless chmods
         if (0o777 & ~util.umask) == (0o777 & mode):
             mode = None
     except OSError:
         mode = None
     return mode
 
 _data = ('data meta 00manifest.d 00manifest.i 00changelog.d 00changelog.i'
          ' phaseroots obsstore')
 
 def isrevlog(f, kind, st):
     return kind == stat.S_IFREG and f[-2:] in ('.i', '.d')
 
 class basicstore(object):
     '''base class for local repository stores'''
     def __init__(self, path, vfstype):
         vfs = vfstype(path)
         self.path = vfs.base
         self.createmode = _calcmode(vfs)
         vfs.createmode = self.createmode
         self.rawvfs = vfs
         self.vfs = vfsmod.filtervfs(vfs, encodedir)
         self.opener = self.vfs
 
     def join(self, f):
         return self.path + '/' + encodedir(f)
 
     def _walk(self, relpath, recurse, filefilter=isrevlog):
         '''yields (unencoded, encoded, size)'''
         path = self.path
         if relpath:
             path += '/' + relpath
         striplen = len(self.path) + 1
         l = []
         if self.rawvfs.isdir(path):
             visit = [path]
             readdir = self.rawvfs.readdir
             while visit:
                 p = visit.pop()
                 for f, kind, st in readdir(p, stat=True):
                     fp = p + '/' + f
                     if filefilter(f, kind, st):
                         n = util.pconvert(fp[striplen:])
                         l.append((decodedir(n), n, st.st_size))
                     elif kind == stat.S_IFDIR and recurse:
                         visit.append(fp)
         l.sort()
         return l
 
     def datafiles(self):
         return self._walk('data', True) + self._walk('meta', True)
 
     def topfiles(self):
         # yield manifest before changelog
         return reversed(self._walk('', False))
 
     def walk(self):
         '''yields (unencoded, encoded, size)'''
         # yield data files first
         for x in self.datafiles():
             yield x
         for x in self.topfiles():
             yield x
 
     def copylist(self):
         return ['requires'] + _data.split()
 
     def write(self, tr):
         pass
 
     def invalidatecaches(self):
         pass
 
     def markremoved(self, fn):
         pass
 
     def __contains__(self, path):
         '''Checks if the store contains path'''
         path = "/".join(("data", path))
         # file?
         if self.vfs.exists(path + ".i"):
             return True
         # dir?
         if not path.endswith("/"):
             path = path + "/"
         return self.vfs.exists(path)
 
 class encodedstore(basicstore):
     def __init__(self, path, vfstype):
         vfs = vfstype(path + '/store')
         self.path = vfs.base
         self.createmode = _calcmode(vfs)
         vfs.createmode = self.createmode
         self.rawvfs = vfs
         self.vfs = vfsmod.filtervfs(vfs, encodefilename)
         self.opener = self.vfs
 
     def datafiles(self):
         for a, b, size in super(encodedstore, self).datafiles():
             try:
                 a = decodefilename(a)
             except KeyError:
                 a = None
             yield a, b, size
 
     def join(self, f):
         return self.path + '/' + encodefilename(f)
 
     def copylist(self):
         return (['requires', '00changelog.i'] +
                 ['store/' + f for f in _data.split()])
 
 class fncache(object):
     # the filename used to be partially encoded
     # hence the encodedir/decodedir dance
     def __init__(self, vfs):
         self.vfs = vfs
         self.entries = None
         self._dirty = False
 
     def _load(self):
         '''fill the entries from the fncache file'''
         self._dirty = False
         try:
             fp = self.vfs('fncache', mode='rb')
         except IOError:
             # skip nonexistent file
             self.entries = set()
             return
         self.entries = set(decodedir(fp.read()).splitlines())
         if '' in self.entries:
             fp.seek(0)
             for n, line in enumerate(util.iterfile(fp)):
                 if not line.rstrip('\n'):
                     t = _('invalid entry in fncache, line %d') % (n + 1)
                     raise error.Abort(t)
         fp.close()
 
     def write(self, tr):
         if self._dirty:
             tr.addbackup('fncache')
             fp = self.vfs('fncache', mode='wb', atomictemp=True)
             if self.entries:
                 fp.write(encodedir('\n'.join(self.entries) + '\n'))
             fp.close()
             self._dirty = False
 
     def add(self, fn):
         if self.entries is None:
             self._load()
         if fn not in self.entries:
             self._dirty = True
             self.entries.add(fn)
 
     def remove(self, fn):
         if self.entries is None:
             self._load()
         try:
             self.entries.remove(fn)
             self._dirty = True
         except KeyError:
             pass
 
     def __contains__(self, fn):
         if self.entries is None:
             self._load()
         return fn in self.entries
 
     def __iter__(self):
         if self.entries is None:
             self._load()
         return iter(self.entries)
 
 class _fncachevfs(vfsmod.abstractvfs, vfsmod.proxyvfs):
     def __init__(self, vfs, fnc, encode):
         vfsmod.proxyvfs.__init__(self, vfs)
         self.fncache = fnc
         self.encode = encode
 
     def __call__(self, path, mode='r', *args, **kw):
+        encoded = self.encode(path)
         if mode not in ('r', 'rb') and (path.startswith('data/') or
                                         path.startswith('meta/')):
+            # do not trigger a fncache load when adding a file that already is
+            # known to exist.
+            notload = self.fncache.entries is None and self.vfs.exists(encoded)
+            if notload and 'a' in mode and not self.vfs.stat(encoded).st_size:
+                # when appending to an existing file, if the file has size zero,
+                # it should be considered as missing. Such zero-size files are
+                # the result of truncation when a transaction is aborted.
+                notload = False
+            if not notload:
             self.fncache.add(path)
-        return self.vfs(self.encode(path), mode, *args, **kw)
+        return self.vfs(encoded, mode, *args, **kw)
 
     def join(self, path):
         if path:
             return self.vfs.join(self.encode(path))
         else:
             return self.vfs.join(path)
 
 class fncachestore(basicstore):
     def __init__(self, path, vfstype, dotencode):
         if dotencode:
             encode = _pathencode
         else:
             encode = _plainhybridencode
         self.encode = encode
         vfs = vfstype(path + '/store')
         self.path = vfs.base
         self.pathsep = self.path + '/'
         self.createmode = _calcmode(vfs)
         vfs.createmode = self.createmode
         self.rawvfs = vfs
         fnc = fncache(vfs)
         self.fncache = fnc
         self.vfs = _fncachevfs(vfs, fnc, encode)
         self.opener = self.vfs
 
     def join(self, f):
         return self.pathsep + self.encode(f)
 
     def getsize(self, path):
         return self.rawvfs.stat(path).st_size
 
     def datafiles(self):
         for f in sorted(self.fncache):
             ef = self.encode(f)
             try:
                 yield f, ef, self.getsize(ef)
             except OSError as err:
                 if err.errno != errno.ENOENT:
                     raise
 
     def copylist(self):
         d = ('data meta dh fncache phaseroots obsstore'
              ' 00manifest.d 00manifest.i 00changelog.d 00changelog.i')
         return (['requires', '00changelog.i'] +
                 ['store/' + f for f in d.split()])
 
     def write(self, tr):
         self.fncache.write(tr)
 
     def invalidatecaches(self):
         self.fncache.entries = None
 
     def markremoved(self, fn):
         self.fncache.remove(fn)
 
     def _exists(self, f):
         ef = self.encode(f)
         try:
             self.getsize(ef)
             return True
         except OSError as err:
             if err.errno != errno.ENOENT:
                 raise
             # nonexistent entry
             return False
 
     def __contains__(self, path):
         '''Checks if the store contains path'''
         path = "/".join(("data", path))
         # check for files (exact match)
         e = path + '.i'
         if e in self.fncache and self._exists(e):
             return True
         # now check for directories (prefix match)
         if not path.endswith('/'):
             path += '/'
         for e in self.fncache:
             if e.startswith(path) and self._exists(e):
                 return True
         return False
 
 def store(requirements, path, vfstype):
     if 'store' in requirements:
         if 'fncache' in requirements:
             return fncachestore(path, vfstype, 'dotencode' in requirements)
         return encodedstore(path, vfstype)
     return basicstore(path, vfstype)

tests/test-fncache.t

@@ -1,438 +1,508 b''
 #require repofncache
 
 Init repo1:
 
   $ hg init repo1
   $ cd repo1
   $ echo "some text" > a
   $ hg add
   adding a
   $ hg ci -m first
   $ cat .hg/store/fncache | sort
   data/a.i
 
 Testing a.i/b:
 
   $ mkdir a.i
   $ echo "some other text" > a.i/b
   $ hg add
   adding a.i/b
   $ hg ci -m second
   $ cat .hg/store/fncache | sort
   data/a.i
   data/a.i.hg/b.i
 
 Testing a.i.hg/c:
 
   $ mkdir a.i.hg
   $ echo "yet another text" > a.i.hg/c
   $ hg add
   adding a.i.hg/c
   $ hg ci -m third
   $ cat .hg/store/fncache | sort
   data/a.i
   data/a.i.hg.hg/c.i
   data/a.i.hg/b.i
 
 Testing verify:
 
   $ hg verify
   checking changesets
   checking manifests
   crosschecking files in changesets and manifests
   checking files
   3 files, 3 changesets, 3 total revisions
 
   $ rm .hg/store/fncache
 
   $ hg verify
   checking changesets
   checking manifests
   crosschecking files in changesets and manifests
   checking files
    warning: revlog 'data/a.i' not in fncache!
    warning: revlog 'data/a.i.hg/c.i' not in fncache!
    warning: revlog 'data/a.i/b.i' not in fncache!
   3 files, 3 changesets, 3 total revisions
   3 warnings encountered!
   hint: run "hg debugrebuildfncache" to recover from corrupt fncache
 
 Follow the hint to make sure it works
 
   $ hg debugrebuildfncache
   adding data/a.i
   adding data/a.i.hg/c.i
   adding data/a.i/b.i
   3 items added, 0 removed from fncache
 
   $ hg verify
   checking changesets
   checking manifests
   crosschecking files in changesets and manifests
   checking files
   3 files, 3 changesets, 3 total revisions
 
   $ cd ..
 
 Non store repo:
 
   $ hg --config format.usestore=False init foo
   $ cd foo
   $ mkdir tst.d
   $ echo foo > tst.d/foo
   $ hg ci -Amfoo
   adding tst.d/foo
   $ find .hg | sort
   .hg
   .hg/00changelog.i
   .hg/00manifest.i
   .hg/cache
   .hg/cache/branch2-served
   .hg/cache/rbc-names-v1
   .hg/cache/rbc-revs-v1
   .hg/data
   .hg/data/tst.d.hg
   .hg/data/tst.d.hg/foo.i
   .hg/dirstate
   .hg/fsmonitor.state (fsmonitor !)
   .hg/last-message.txt
   .hg/phaseroots
   .hg/requires
   .hg/undo
   .hg/undo.backup.dirstate
   .hg/undo.backupfiles
   .hg/undo.bookmarks
   .hg/undo.branch
   .hg/undo.desc
   .hg/undo.dirstate
   .hg/undo.phaseroots
   $ cd ..
 
 Non fncache repo:
 
   $ hg --config format.usefncache=False init bar
   $ cd bar
   $ mkdir tst.d
   $ echo foo > tst.d/Foo
   $ hg ci -Amfoo
   adding tst.d/Foo
   $ find .hg | sort
   .hg
   .hg/00changelog.i
   .hg/cache
   .hg/cache/branch2-served
   .hg/cache/rbc-names-v1
   .hg/cache/rbc-revs-v1
   .hg/dirstate
   .hg/fsmonitor.state (fsmonitor !)
   .hg/last-message.txt
   .hg/requires
   .hg/store
   .hg/store/00changelog.i
   .hg/store/00manifest.i
   .hg/store/data
   .hg/store/data/tst.d.hg
   .hg/store/data/tst.d.hg/_foo.i
   .hg/store/phaseroots
   .hg/store/undo
   .hg/store/undo.backupfiles
   .hg/store/undo.phaseroots
   .hg/undo.backup.dirstate
   .hg/undo.bookmarks
   .hg/undo.branch
   .hg/undo.desc
   .hg/undo.dirstate
   $ cd ..
 
 Encoding of reserved / long paths in the store
 
   $ hg init r2
   $ cd r2
   $ cat <<EOF > .hg/hgrc
   > [ui]
   > portablefilenames = ignore
   > EOF
 
   $ hg import -q --bypass - <<EOF
   > # HG changeset patch
   > # User test
   > # Date 0 0
   > # Node ID 1c7a2f7cb77be1a0def34e4c7cabc562ad98fbd7
   > # Parent  0000000000000000000000000000000000000000
   > 1
   > 
   > diff --git a/12345678/12345678/12345678/12345678/12345678/12345678/12345678/12345/xxxxxxxxx-xxxxxxxxx-xxxxxxxxx-123456789-12.3456789-12345-ABCDEFGHIJKLMNOPRSTUVWXYZ-abcdefghjiklmnopqrstuvwxyz b/12345678/12345678/12345678/12345678/12345678/12345678/12345678/12345/xxxxxxxxx-xxxxxxxxx-xxxxxxxxx-123456789-12.3456789-12345-ABCDEFGHIJKLMNOPRSTUVWXYZ-abcdefghjiklmnopqrstuvwxyz
   > new file mode 100644
   > --- /dev/null
   > +++ b/12345678/12345678/12345678/12345678/12345678/12345678/12345678/12345/xxxxxxxxx-xxxxxxxxx-xxxxxxxxx-123456789-12.3456789-12345-ABCDEFGHIJKLMNOPRSTUVWXYZ-abcdefghjiklmnopqrstuvwxyz
   > @@ -0,0 +1,1 @@
   > +foo
   > diff --git a/AUX/SECOND/X.PRN/FOURTH/FI:FTH/SIXTH/SEVENTH/EIGHTH/NINETH/TENTH/ELEVENTH/LOREMIPSUM.TXT b/AUX/SECOND/X.PRN/FOURTH/FI:FTH/SIXTH/SEVENTH/EIGHTH/NINETH/TENTH/ELEVENTH/LOREMIPSUM.TXT
   > new file mode 100644
   > --- /dev/null
   > +++ b/AUX/SECOND/X.PRN/FOURTH/FI:FTH/SIXTH/SEVENTH/EIGHTH/NINETH/TENTH/ELEVENTH/LOREMIPSUM.TXT
   > @@ -0,0 +1,1 @@
   > +foo
   > diff --git a/Project Planning/Resources/AnotherLongDirectoryName/Followedbyanother/AndAnother/AndThenAnExtremelyLongFileName.txt b/Project Planning/Resources/AnotherLongDirectoryName/Followedbyanother/AndAnother/AndThenAnExtremelyLongFileName.txt
   > new file mode 100644
   > --- /dev/null
   > +++ b/Project Planning/Resources/AnotherLongDirectoryName/Followedbyanother/AndAnother/AndThenAnExtremelyLongFileName.txt	
   > @@ -0,0 +1,1 @@
   > +foo
   > diff --git a/bla.aux/prn/PRN/lpt/com3/nul/coma/foo.NUL/normal.c b/bla.aux/prn/PRN/lpt/com3/nul/coma/foo.NUL/normal.c
   > new file mode 100644
   > --- /dev/null
   > +++ b/bla.aux/prn/PRN/lpt/com3/nul/coma/foo.NUL/normal.c
   > @@ -0,0 +1,1 @@
   > +foo
   > diff --git a/enterprise/openesbaddons/contrib-imola/corba-bc/netbeansplugin/wsdlExtension/src/main/java/META-INF/services/org.netbeans.modules.xml.wsdl.bindingsupport.spi.ExtensibilityElementTemplateProvider b/enterprise/openesbaddons/contrib-imola/corba-bc/netbeansplugin/wsdlExtension/src/main/java/META-INF/services/org.netbeans.modules.xml.wsdl.bindingsupport.spi.ExtensibilityElementTemplateProvider
   > new file mode 100644
   > --- /dev/null
   > +++ b/enterprise/openesbaddons/contrib-imola/corba-bc/netbeansplugin/wsdlExtension/src/main/java/META-INF/services/org.netbeans.modules.xml.wsdl.bindingsupport.spi.ExtensibilityElementTemplateProvider
   > @@ -0,0 +1,1 @@
   > +foo
   > EOF
 
   $ find .hg/store -name *.i  | sort
   .hg/store/00changelog.i
   .hg/store/00manifest.i
   .hg/store/data/bla.aux/pr~6e/_p_r_n/lpt/co~6d3/nu~6c/coma/foo._n_u_l/normal.c.i
   .hg/store/dh/12345678/12345678/12345678/12345678/12345678/12345678/12345678/12345/xxxxxx168e07b38e65eff86ab579afaaa8e30bfbe0f35f.i
   .hg/store/dh/au~78/second/x.prn/fourth/fi~3afth/sixth/seventh/eighth/nineth/tenth/loremia20419e358ddff1bf8751e38288aff1d7c32ec05.i
   .hg/store/dh/enterpri/openesba/contrib-/corba-bc/netbeans/wsdlexte/src/main/java/org.net7018f27961fdf338a598a40c4683429e7ffb9743.i
   .hg/store/dh/project_/resource/anotherl/followed/andanoth/andthenanextremelylongfilename0d8e1f4187c650e2f1fdca9fd90f786bc0976b6b.i
 
   $ cd ..
 
 Aborting lock does not prevent fncache writes
 
   $ cat > exceptionext.py <<EOF
   > from __future__ import absolute_import
   > import os
   > from mercurial import commands, error, extensions
   > 
   > def lockexception(orig, vfs, lockname, wait, releasefn, *args, **kwargs):
   >     def releasewrap():
   >         l.held = False # ensure __del__ is a noop
   >         raise error.Abort("forced lock failure")
   >     l = orig(vfs, lockname, wait, releasewrap, *args, **kwargs)
   >     return l
   > 
   > def reposetup(ui, repo):
   >     extensions.wrapfunction(repo, '_lock', lockexception)
   > 
   > cmdtable = {}
   > 
   > # wrap "commit" command to prevent wlock from being '__del__()'-ed
   > # at the end of dispatching (for intentional "forced lcok failure")
   > def commitwrap(orig, ui, repo, *pats, **opts):
   >     repo = repo.unfiltered() # to use replaced repo._lock certainly
   >     wlock = repo.wlock()
   >     try:
   >         return orig(ui, repo, *pats, **opts)
   >     finally:
   >         # multiple 'relase()' is needed for complete releasing wlock,
   >         # because "forced" abort at last releasing store lock
   >         # prevents wlock from being released at same 'lockmod.release()'
   >         for i in range(wlock.held):
   >             wlock.release()
   > 
   > def extsetup(ui):
   >     extensions.wrapcommand(commands.table, b"commit", commitwrap)
   > EOF
   $ extpath=`pwd`/exceptionext.py
   $ hg init fncachetxn
   $ cd fncachetxn
   $ printf "[extensions]\nexceptionext=$extpath\n" >> .hg/hgrc
   $ touch y
   $ hg ci -qAm y
   abort: forced lock failure
   [255]
   $ cat .hg/store/fncache
   data/y.i
 
 Aborting transaction prevents fncache change
 
   $ cat > ../exceptionext.py <<EOF
   > from __future__ import absolute_import
   > import os
   > from mercurial import commands, error, extensions, localrepo
   > 
   > def wrapper(orig, self, *args, **kwargs):
   >     tr = orig(self, *args, **kwargs)
   >     def fail(tr):
   >         raise error.Abort(b"forced transaction failure")
   >     # zzz prefix to ensure it sorted after store.write
   >     tr.addfinalize(b'zzz-forcefails', fail)
   >     return tr
   > 
   > def uisetup(ui):
   >     extensions.wrapfunction(
   >         localrepo.localrepository, b'transaction', wrapper)
   > 
   > cmdtable = {}
   > 
   > EOF
 
 Clean cached version
   $ rm -f "${extpath}c"
   $ rm -Rf "`dirname $extpath`/__pycache__"
 
   $ touch z
   $ hg ci -qAm z
   transaction abort!
   rollback completed
   abort: forced transaction failure
   [255]
   $ cat .hg/store/fncache
   data/y.i
 
 Aborted transactions can be recovered later
 
   $ cat > ../exceptionext.py <<EOF
   > from __future__ import absolute_import
   > import os
   > from mercurial import (
   >   commands,
   >   error,
   >   extensions,
   >   localrepo,
   >   transaction,
   > )
   > 
   > def trwrapper(orig, self, *args, **kwargs):
   >     tr = orig(self, *args, **kwargs)
   >     def fail(tr):
   >         raise error.Abort("forced transaction failure")
   >     # zzz prefix to ensure it sorted after store.write
   >     tr.addfinalize('zzz-forcefails', fail)
   >     return tr
   > 
   > def abortwrapper(orig, self, *args, **kwargs):
   >     raise error.Abort("forced transaction failure")
   > 
   > def uisetup(ui):
   >     extensions.wrapfunction(localrepo.localrepository, 'transaction',
   >                             trwrapper)
   >     extensions.wrapfunction(transaction.transaction, '_abort',
   >                             abortwrapper)
   > 
   > cmdtable = {}
   > 
   > EOF
 
 Clean cached versions
   $ rm -f "${extpath}c"
   $ rm -Rf "`dirname $extpath`/__pycache__"
 
   $ hg up -q 1
   $ touch z
   $ hg ci -qAm z 2>/dev/null
   [255]
   $ cat .hg/store/fncache | sort
   data/y.i
   data/z.i
   $ hg recover
   rolling back interrupted transaction
   checking changesets
   checking manifests
   crosschecking files in changesets and manifests
   checking files
   1 files, 1 changesets, 1 total revisions
   $ cat .hg/store/fncache
   data/y.i
 
   $ cd ..
 
 debugrebuildfncache does nothing unless repo has fncache requirement
 
   $ hg --config format.usefncache=false init nofncache
   $ cd nofncache
   $ hg debugrebuildfncache
   (not rebuilding fncache because repository does not support fncache)
 
   $ cd ..
 
 debugrebuildfncache works on empty repository
 
   $ hg init empty
   $ cd empty
   $ hg debugrebuildfncache
   fncache already up to date
   $ cd ..
 
 debugrebuildfncache on an up to date repository no-ops
 
   $ hg init repo
   $ cd repo
   $ echo initial > foo
   $ echo initial > .bar
   $ hg commit -A -m initial
   adding .bar
   adding foo
 
   $ cat .hg/store/fncache | sort
   data/.bar.i
   data/foo.i
 
   $ hg debugrebuildfncache
   fncache already up to date
 
 debugrebuildfncache restores deleted fncache file
 
   $ rm -f .hg/store/fncache
   $ hg debugrebuildfncache
   adding data/.bar.i
   adding data/foo.i
   2 items added, 0 removed from fncache
 
   $ cat .hg/store/fncache | sort
   data/.bar.i
   data/foo.i
 
 Rebuild after rebuild should no-op
 
   $ hg debugrebuildfncache
   fncache already up to date
 
 A single missing file should get restored, an extra file should be removed
 
   $ cat > .hg/store/fncache << EOF
   > data/foo.i
   > data/bad-entry.i
   > EOF
 
   $ hg debugrebuildfncache
   removing data/bad-entry.i
   adding data/.bar.i
   1 items added, 1 removed from fncache
 
   $ cat .hg/store/fncache | sort
   data/.bar.i
   data/foo.i
 
   $ cd ..
 
 Try a simple variation without dotencode to ensure fncache is ignorant of encoding
 
   $ hg --config format.dotencode=false init nodotencode
   $ cd nodotencode
   $ echo initial > foo
   $ echo initial > .bar
   $ hg commit -A -m initial
   adding .bar
   adding foo
 
   $ cat .hg/store/fncache | sort
   data/.bar.i
   data/foo.i
 
   $ rm .hg/store/fncache
   $ hg debugrebuildfncache
   adding data/.bar.i
   adding data/foo.i
   2 items added, 0 removed from fncache
 
   $ cat .hg/store/fncache | sort
   data/.bar.i
   data/foo.i
+
+  $ cd ..
+
+In repositories that have accumulated a large number of files over time, the
+fncache file is going to be large. If we possibly can avoid loading it, so much the better.
+The cache should not loaded when committing changes to existing files, or when unbundling
+changesets that only contain changes to existing files:
+
+  $ cat > fncacheloadwarn.py << EOF
+  > from __future__ import absolute_import
+  > from mercurial import extensions, store
+  > 
+  > def extsetup(ui):
+  >     def wrapstore(orig, requirements, *args):
+  >         store = orig(requirements, *args)
+  >         if 'store' in requirements and 'fncache' in requirements:
+  >             instrumentfncachestore(store, ui)
+  >         return store
+  >     extensions.wrapfunction(store, 'store', wrapstore)
+  > 
+  > def instrumentfncachestore(fncachestore, ui):
+  >     class instrumentedfncache(type(fncachestore.fncache)):
+  >         def _load(self):
+  >             ui.warn('fncache load triggered!\n')
+  >             super(instrumentedfncache, self)._load()
+  >     fncachestore.fncache.__class__ = instrumentedfncache
+  > EOF
+
+  $ fncachextpath=`pwd`/fncacheloadwarn.py
+  $ hg init nofncacheload
+  $ cd nofncacheload
+  $ printf "[extensions]\nfncacheloadwarn=$fncachextpath\n" >> .hg/hgrc
+
+A new file should trigger a load, as we'd want to update the fncache set in that case:
+
+  $ touch foo
+  $ hg ci -qAm foo
+  fncache load triggered!
+
+But modifying that file should not:
+
+  $ echo bar >> foo
+  $ hg ci -qm foo
+
+If a transaction has been aborted, the zero-size truncated index file will
+not prevent the fncache from being loaded; rather than actually abort
+a transaction, we simulate the situation by creating a zero-size index file:
+
+  $ touch .hg/store/data/bar.i
+  $ touch bar
+  $ hg ci -qAm bar
+  fncache load triggered!
+
+Unbundling should follow the same rules; existing files should not cause a load:
+
+  $ hg clone -q . tobundle
+  $ echo 'new line' > tobundle/bar
+  $ hg -R tobundle ci -qm bar
+  $ hg -R tobundle bundle -q barupdated.hg
+  $ hg unbundle -q barupdated.hg
+
+but adding new files should:
+
+  $ touch tobundle/newfile
+  $ hg -R tobundle ci -qAm newfile
+  $ hg -R tobundle bundle -q newfile.hg
+  $ hg unbundle -q newfile.hg
+  fncache load triggered!
+
+  $ cd ..