vfs: add a `register_file` method on the vfs class

This is used by the fncache vfs to register new file. Until now, `fncache` have
been doing this "automatically" by monitoring write pattern. However this is
fragile and when we copy files in place by other means, we need something more
robuts. So we add an explicit method to do so.

Differential Revision: https://phab.mercurial-scm.org/D10851

mercurial/store.py

@@ -1,824 +1,829 b''
 # store.py - repository store handling for Mercurial
 #
 # Copyright 2008 Olivia 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 functools
 import os
 import re
 import stat
 
 from .i18n import _
 from .pycompat import getattr
 from .node import hex
 from . import (
     changelog,
     error,
     manifest,
     policy,
     pycompat,
     util,
     vfs as vfsmod,
 )
 from .utils import hashutil
 
 parsers = policy.importmod('parsers')
 # how much bytes should be read from fncache in one read
 # It is done to prevent loading large fncache files into memory
 fncache_chunksize = 10 ** 6
 
 
 def _matchtrackedpath(path, matcher):
     """parses a fncache entry and returns whether the entry is tracking a path
     matched by matcher or not.
 
     If matcher is None, returns True"""
 
     if matcher is None:
         return True
     path = decodedir(path)
     if path.startswith(b'data/'):
         return matcher(path[len(b'data/') : -len(b'.i')])
     elif path.startswith(b'meta/'):
         return matcher.visitdir(path[len(b'meta/') : -len(b'/00manifest.i')])
 
     raise error.ProgrammingError(b"cannot decode path %s" % path)
 
 
 # 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(b".hg/", b".hg.hg/")
         .replace(b".i/", b".i.hg/")
         .replace(b".d/", b".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 b".hg/" not in path:
         return path
     return (
         path.replace(b".d.hg/", b".d/")
         .replace(b".i.hg/", b".i/")
         .replace(b".hg.hg/", b".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 = b'_'
     xchr = pycompat.bytechr
     asciistr = list(map(xchr, range(127)))
     capitals = list(range(ord(b"A"), ord(b"Z") + 1))
 
     cmap = {x: x for x in asciistr}
     for x in _reserved():
         cmap[xchr(x)] = b"~%02x" % x
     for x in capitals + [ord(e)]:
         cmap[xchr(x)] = e + xchr(x).lower()
 
     dmap = {}
     for k, v in pycompat.iteritems(cmap):
         dmap[v] = k
 
     def decode(s):
         i = 0
         while i < len(s):
             for l in pycompat.xrange(1, 4):
                 try:
                     yield dmap[s[i : i + l]]
                     i += l
                     break
                 except KeyError:
                     pass
             else:
                 raise KeyError
 
     return (
         lambda s: b''.join(
             [cmap[s[c : c + 1]] for c in pycompat.xrange(len(s))]
         ),
         lambda s: b''.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 = {xchr(x): xchr(x) for x in pycompat.xrange(127)}
     for x in _reserved():
         cmap[xchr(x)] = b"~%02x" % x
     for x in range(ord(b"A"), ord(b"Z") + 1):
         cmap[xchr(x)] = xchr(x).lower()
 
     def lowerencode(s):
         return b"".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 = (b'aux', b'con', b'prn', b'nul')  # length 3
 _winres4 = (b'com', b'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 b'. ':
             n = b"~%02x" % ord(n[0:1]) + n[1:]
             path[i] = n
         else:
             l = n.find(b'.')
             if l == -1:
                 l = len(n)
             if (l == 3 and n[:3] in _winres3) or (
                 l == 4
                 and n[3:4] <= b'9'
                 and n[3:4] >= b'1'
                 and n[:3] in _winres4
             ):
                 # encode third letter ('aux' -> 'au~78')
                 ec = b"~%02x" % ord(n[2:3])
                 n = n[0:2] + ec + n[3:]
                 path[i] = n
         if n[-1] in b'. ':
             # encode last period or space ('foo...' -> 'foo..~2e')
             path[i] = n[:-1] + b"~%02x" % ord(n[-1:])
     return path
 
 
 _maxstorepathlen = 120
 _dirprefixlen = 8
 _maxshortdirslen = 8 * (_dirprefixlen + 1) - 4
 
 
 def _hashencode(path, dotencode):
     digest = hex(hashutil.sha1(path).digest())
     le = lowerencode(path[5:]).split(b'/')  # 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 b'. ':
             # Windows can't access dirs ending in period or space
             d = d[:-1] + b'_'
         if sdirslen == 0:
             t = len(d)
         else:
             t = sdirslen + 1 + len(d)
             if t > _maxshortdirslen:
                 break
         sdirs.append(d)
         sdirslen = t
     dirs = b'/'.join(sdirs)
     if len(dirs) > 0:
         dirs += b'/'
     res = b'dh/' + dirs + digest + ext
     spaceleft = _maxstorepathlen - len(res)
     if spaceleft > 0:
         filler = basename[:spaceleft]
         res = b'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(b'/')
     res = b'/'.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(b'/')
     res = b'/'.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 = [
     b'bookmarks',
     b'narrowspec',
     b'data',
     b'meta',
     b'00manifest.d',
     b'00manifest.i',
     b'00changelog.d',
     b'00changelog.i',
     b'phaseroots',
     b'obsstore',
     b'requires',
 ]
 
 REVLOG_FILES_MAIN_EXT = (b'.i', b'i.tmpcensored')
 REVLOG_FILES_OTHER_EXT = (
     b'.idx',
     b'.d',
     b'.dat',
     b'.n',
     b'.nd',
     b'.sda',
     b'd.tmpcensored',
 )
 # files that are "volatile" and might change between listing and streaming
 #
 # note: the ".nd" file are nodemap data and won't "change" but they might be
 # deleted.
 REVLOG_FILES_VOLATILE_EXT = (b'.n', b'.nd')
 
 # some exception to the above matching
 EXCLUDED = re.compile(b'.*undo\.[^/]+\.(nd?|i)$')
 
 
 def is_revlog(f, kind, st):
     if kind != stat.S_IFREG:
         return None
     return revlog_type(f)
 
 
 def revlog_type(f):
     if f.endswith(REVLOG_FILES_MAIN_EXT) and EXCLUDED.match(f) is None:
         return FILEFLAGS_REVLOG_MAIN
     elif f.endswith(REVLOG_FILES_OTHER_EXT) and EXCLUDED.match(f) is None:
         t = FILETYPE_FILELOG_OTHER
         if f.endswith(REVLOG_FILES_VOLATILE_EXT):
             t |= FILEFLAGS_VOLATILE
         return t
     return None
 
 
 # the file is part of changelog data
 FILEFLAGS_CHANGELOG = 1 << 13
 # the file is part of manifest data
 FILEFLAGS_MANIFESTLOG = 1 << 12
 # the file is part of filelog data
 FILEFLAGS_FILELOG = 1 << 11
 # file that are not directly part of a revlog
 FILEFLAGS_OTHER = 1 << 10
 
 # the main entry point for a revlog
 FILEFLAGS_REVLOG_MAIN = 1 << 1
 # a secondary file for a revlog
 FILEFLAGS_REVLOG_OTHER = 1 << 0
 
 # files that are "volatile" and might change between listing and streaming
 FILEFLAGS_VOLATILE = 1 << 20
 
 FILETYPE_CHANGELOG_MAIN = FILEFLAGS_CHANGELOG | FILEFLAGS_REVLOG_MAIN
 FILETYPE_CHANGELOG_OTHER = FILEFLAGS_CHANGELOG | FILEFLAGS_REVLOG_OTHER
 FILETYPE_MANIFESTLOG_MAIN = FILEFLAGS_MANIFESTLOG | FILEFLAGS_REVLOG_MAIN
 FILETYPE_MANIFESTLOG_OTHER = FILEFLAGS_MANIFESTLOG | FILEFLAGS_REVLOG_OTHER
 FILETYPE_FILELOG_MAIN = FILEFLAGS_FILELOG | FILEFLAGS_REVLOG_MAIN
 FILETYPE_FILELOG_OTHER = FILEFLAGS_FILELOG | FILEFLAGS_REVLOG_OTHER
 FILETYPE_OTHER = FILEFLAGS_OTHER
 
 
 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 + b'/' + encodedir(f)
 
     def _walk(self, relpath, recurse):
         '''yields (unencoded, encoded, size)'''
         path = self.path
         if relpath:
             path += b'/' + 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 + b'/' + f
                     rl_type = is_revlog(f, kind, st)
                     if rl_type is not None:
                         n = util.pconvert(fp[striplen:])
                         l.append((rl_type, decodedir(n), n, st.st_size))
                     elif kind == stat.S_IFDIR and recurse:
                         visit.append(fp)
         l.sort()
         return l
 
     def changelog(self, trypending, concurrencychecker=None):
         return changelog.changelog(
             self.vfs,
             trypending=trypending,
             concurrencychecker=concurrencychecker,
         )
 
     def manifestlog(self, repo, storenarrowmatch):
         rootstore = manifest.manifestrevlog(repo.nodeconstants, self.vfs)
         return manifest.manifestlog(self.vfs, repo, rootstore, storenarrowmatch)
 
     def datafiles(self, matcher=None):
         files = self._walk(b'data', True) + self._walk(b'meta', True)
         for (t, u, e, s) in files:
             yield (FILEFLAGS_FILELOG | t, u, e, s)
 
     def topfiles(self):
         # yield manifest before changelog
         files = reversed(self._walk(b'', False))
         for (t, u, e, s) in files:
             if u.startswith(b'00changelog'):
                 yield (FILEFLAGS_CHANGELOG | t, u, e, s)
             elif u.startswith(b'00manifest'):
                 yield (FILEFLAGS_MANIFESTLOG | t, u, e, s)
             else:
                 yield (FILETYPE_OTHER | t, u, e, s)
 
     def walk(self, matcher=None):
         """return file related to data storage (ie: revlogs)
 
         yields (file_type, unencoded, encoded, size)
 
         if a matcher is passed, storage files of only those tracked paths
         are passed with matches the matcher
         """
         # yield data files first
         for x in self.datafiles(matcher):
             yield x
         for x in self.topfiles():
             yield x
 
     def copylist(self):
         return _data
 
     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 = b"/".join((b"data", path))
         # file?
         if self.vfs.exists(path + b".i"):
             return True
         # dir?
         if not path.endswith(b"/"):
             path = path + b"/"
         return self.vfs.exists(path)
 
 
 class encodedstore(basicstore):
     def __init__(self, path, vfstype):
         vfs = vfstype(path + b'/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, matcher=None):
         for t, a, b, size in super(encodedstore, self).datafiles():
             try:
                 a = decodefilename(a)
             except KeyError:
                 a = None
             if a is not None and not _matchtrackedpath(a, matcher):
                 continue
             yield t, a, b, size
 
     def join(self, f):
         return self.path + b'/' + encodefilename(f)
 
     def copylist(self):
         return [b'requires', b'00changelog.i'] + [b'store/' + f for f in _data]
 
 
 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
         # set of new additions to fncache
         self.addls = set()
 
     def ensureloaded(self, warn=None):
         """read the fncache file if not already read.
 
         If the file on disk is corrupted, raise. If warn is provided,
         warn and keep going instead."""
         if self.entries is None:
             self._load(warn)
 
     def _load(self, warn=None):
         '''fill the entries from the fncache file'''
         self._dirty = False
         try:
             fp = self.vfs(b'fncache', mode=b'rb')
         except IOError:
             # skip nonexistent file
             self.entries = set()
             return
 
         self.entries = set()
         chunk = b''
         for c in iter(functools.partial(fp.read, fncache_chunksize), b''):
             chunk += c
             try:
                 p = chunk.rindex(b'\n')
                 self.entries.update(decodedir(chunk[: p + 1]).splitlines())
                 chunk = chunk[p + 1 :]
             except ValueError:
                 # substring '\n' not found, maybe the entry is bigger than the
                 # chunksize, so let's keep iterating
                 pass
 
         if chunk:
             msg = _(b"fncache does not ends with a newline")
             if warn:
                 warn(msg + b'\n')
             else:
                 raise error.Abort(
                     msg,
                     hint=_(
                         b"use 'hg debugrebuildfncache' to "
                         b"rebuild the fncache"
                     ),
                 )
         self._checkentries(fp, warn)
         fp.close()
 
     def _checkentries(self, fp, warn):
         """make sure there is no empty string in entries"""
         if b'' in self.entries:
             fp.seek(0)
             for n, line in enumerate(util.iterfile(fp)):
                 if not line.rstrip(b'\n'):
                     t = _(b'invalid entry in fncache, line %d') % (n + 1)
                     if warn:
                         warn(t + b'\n')
                     else:
                         raise error.Abort(t)
 
     def write(self, tr):
         if self._dirty:
             assert self.entries is not None
             self.entries = self.entries | self.addls
             self.addls = set()
             tr.addbackup(b'fncache')
             fp = self.vfs(b'fncache', mode=b'wb', atomictemp=True)
             if self.entries:
                 fp.write(encodedir(b'\n'.join(self.entries) + b'\n'))
             fp.close()
             self._dirty = False
         if self.addls:
             # if we have just new entries, let's append them to the fncache
             tr.addbackup(b'fncache')
             fp = self.vfs(b'fncache', mode=b'ab', atomictemp=True)
             if self.addls:
                 fp.write(encodedir(b'\n'.join(self.addls) + b'\n'))
             fp.close()
             self.entries = None
             self.addls = set()
 
     def add(self, fn):
         if self.entries is None:
             self._load()
         if fn not in self.entries:
             self.addls.add(fn)
 
     def remove(self, fn):
         if self.entries is None:
             self._load()
         if fn in self.addls:
             self.addls.remove(fn)
             return
         try:
             self.entries.remove(fn)
             self._dirty = True
         except KeyError:
             pass
 
     def __contains__(self, fn):
         if fn in self.addls:
             return True
         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 | self.addls)
 
 
 class _fncachevfs(vfsmod.proxyvfs):
     def __init__(self, vfs, fnc, encode):
         vfsmod.proxyvfs.__init__(self, vfs)
         self.fncache = fnc
         self.encode = encode
 
     def __call__(self, path, mode=b'r', *args, **kw):
         encoded = self.encode(path)
         if mode not in (b'r', b'rb') and (
             path.startswith(b'data/') or path.startswith(b'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 b'r+' 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(encoded, mode, *args, **kw)
 
     def join(self, path):
         if path:
             return self.vfs.join(self.encode(path))
         else:
             return self.vfs.join(path)
 
+    def register_file(self, path):
+        """generic hook point to lets fncache steer its stew"""
+        if path.startswith(b'data/') or path.startswith(b'meta/'):
+            self.fncache.add(path)
+
 
 class fncachestore(basicstore):
     def __init__(self, path, vfstype, dotencode):
         if dotencode:
             encode = _pathencode
         else:
             encode = _plainhybridencode
         self.encode = encode
         vfs = vfstype(path + b'/store')
         self.path = vfs.base
         self.pathsep = self.path + b'/'
         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, matcher=None):
         for f in sorted(self.fncache):
             if not _matchtrackedpath(f, matcher):
                 continue
             ef = self.encode(f)
             try:
                 t = revlog_type(f)
                 assert t is not None, f
                 t |= FILEFLAGS_FILELOG
                 yield t, f, ef, self.getsize(ef)
             except OSError as err:
                 if err.errno != errno.ENOENT:
                     raise
 
     def copylist(self):
         d = (
             b'bookmarks',
             b'narrowspec',
             b'data',
             b'meta',
             b'dh',
             b'fncache',
             b'phaseroots',
             b'obsstore',
             b'00manifest.d',
             b'00manifest.i',
             b'00changelog.d',
             b'00changelog.i',
             b'requires',
         )
         return [b'requires', b'00changelog.i'] + [b'store/' + f for f in d]
 
     def write(self, tr):
         self.fncache.write(tr)
 
     def invalidatecaches(self):
         self.fncache.entries = None
         self.fncache.addls = set()
 
     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 = b"/".join((b"data", path))
         # check for files (exact match)
         e = path + b'.i'
         if e in self.fncache and self._exists(e):
             return True
         # now check for directories (prefix match)
         if not path.endswith(b'/'):
             path += b'/'
         for e in self.fncache:
             if e.startswith(path) and self._exists(e):
                 return True
         return False

mercurial/vfs.py

@@ -1,751 +1,754 b''
 # vfs.py - Mercurial 'vfs' classes
 #
 #  Copyright Olivia 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 contextlib
 import errno
 import os
 import shutil
 import stat
 import threading
 
 from .i18n import _
 from .pycompat import (
     delattr,
     getattr,
     setattr,
 )
 from . import (
     encoding,
     error,
     pathutil,
     pycompat,
     util,
 )
 
 
 def _avoidambig(path, oldstat):
     """Avoid file stat ambiguity forcibly
 
     This function causes copying ``path`` file, if it is owned by
     another (see issue5418 and issue5584 for detail).
     """
 
     def checkandavoid():
         newstat = util.filestat.frompath(path)
         # return whether file stat ambiguity is (already) avoided
         return not newstat.isambig(oldstat) or newstat.avoidambig(path, oldstat)
 
     if not checkandavoid():
         # simply copy to change owner of path to get privilege to
         # advance mtime (see issue5418)
         util.rename(util.mktempcopy(path), path)
         checkandavoid()
 
 
 class abstractvfs(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 __call__(self, path, mode=b'rb', **kwargs):
         raise NotImplementedError
 
     def _auditpath(self, path, mode):
         raise NotImplementedError
 
     def join(self, path, *insidef):
         raise NotImplementedError
 
     def tryread(self, path):
         '''gracefully return an empty string for missing files'''
         try:
             return self.read(path)
         except IOError as inst:
             if inst.errno != errno.ENOENT:
                 raise
         return b""
 
     def tryreadlines(self, path, mode=b'rb'):
         '''gracefully return an empty array for missing files'''
         try:
             return self.readlines(path, mode=mode)
         except IOError as inst:
             if inst.errno != errno.ENOENT:
                 raise
         return []
 
     @util.propertycache
     def open(self):
         """Open ``path`` file, which is relative to vfs root.
 
         Newly created directories are marked as "not to be indexed by
         the content indexing service", if ``notindexed`` is specified
         for "write" mode access.
         """
         return self.__call__
 
     def read(self, path):
         with self(path, b'rb') as fp:
             return fp.read()
 
     def readlines(self, path, mode=b'rb'):
         with self(path, mode=mode) as fp:
             return fp.readlines()
 
     def write(self, path, data, backgroundclose=False, **kwargs):
         with self(path, b'wb', backgroundclose=backgroundclose, **kwargs) as fp:
             return fp.write(data)
 
     def writelines(self, path, data, mode=b'wb', notindexed=False):
         with self(path, mode=mode, notindexed=notindexed) as fp:
             return fp.writelines(data)
 
     def append(self, path, data):
         with self(path, b'ab') as fp:
             return fp.write(data)
 
     def basename(self, path):
         """return base element of a path (as os.path.basename would do)
 
         This exists to allow handling of strange encoding if needed."""
         return os.path.basename(path)
 
     def chmod(self, path, mode):
         return os.chmod(self.join(path), mode)
 
     def dirname(self, path):
         """return dirname element of a path (as os.path.dirname would do)
 
         This exists to allow handling of strange encoding if needed."""
         return os.path.dirname(path)
 
     def exists(self, path=None):
         return os.path.exists(self.join(path))
 
     def fstat(self, fp):
         return util.fstat(fp)
 
     def isdir(self, path=None):
         return os.path.isdir(self.join(path))
 
     def isfile(self, path=None):
         return os.path.isfile(self.join(path))
 
     def islink(self, path=None):
         return os.path.islink(self.join(path))
 
     def isfileorlink(self, path=None):
         """return whether path is a regular file or a symlink
 
         Unlike isfile, this doesn't follow symlinks."""
         try:
             st = self.lstat(path)
         except OSError:
             return False
         mode = st.st_mode
         return stat.S_ISREG(mode) or stat.S_ISLNK(mode)
 
     def reljoin(self, *paths):
         """join various elements of a path together (as os.path.join would do)
 
         The vfs base is not injected so that path stay relative. This exists
         to allow handling of strange encoding if needed."""
         return os.path.join(*paths)
 
     def split(self, path):
         """split top-most element of a path (as os.path.split would do)
 
         This exists to allow handling of strange encoding if needed."""
         return os.path.split(path)
 
     def lexists(self, path=None):
         return os.path.lexists(self.join(path))
 
     def lstat(self, path=None):
         return os.lstat(self.join(path))
 
     def listdir(self, path=None):
         return os.listdir(self.join(path))
 
     def makedir(self, path=None, notindexed=True):
         return util.makedir(self.join(path), notindexed)
 
     def makedirs(self, path=None, mode=None):
         return util.makedirs(self.join(path), mode)
 
     def makelock(self, info, path):
         return util.makelock(info, self.join(path))
 
     def mkdir(self, path=None):
         return os.mkdir(self.join(path))
 
     def mkstemp(self, suffix=b'', prefix=b'tmp', dir=None):
         fd, name = pycompat.mkstemp(
             suffix=suffix, prefix=prefix, dir=self.join(dir)
         )
         dname, fname = util.split(name)
         if dir:
             return fd, os.path.join(dir, fname)
         else:
             return fd, fname
 
     def readdir(self, path=None, stat=None, skip=None):
         return util.listdir(self.join(path), stat, skip)
 
     def readlock(self, path):
         return util.readlock(self.join(path))
 
     def rename(self, src, dst, checkambig=False):
         """Rename from src to dst
 
         checkambig argument is used with util.filestat, and is useful
         only if destination file is guarded by any lock
         (e.g. repo.lock or repo.wlock).
 
         To avoid file stat ambiguity forcibly, checkambig=True involves
         copying ``src`` file, if it is owned by another. Therefore, use
         checkambig=True only in limited cases (see also issue5418 and
         issue5584 for detail).
         """
         self._auditpath(dst, b'w')
         srcpath = self.join(src)
         dstpath = self.join(dst)
         oldstat = checkambig and util.filestat.frompath(dstpath)
         if oldstat and oldstat.stat:
             ret = util.rename(srcpath, dstpath)
             _avoidambig(dstpath, oldstat)
             return ret
         return util.rename(srcpath, dstpath)
 
     def readlink(self, path):
         return util.readlink(self.join(path))
 
     def removedirs(self, path=None):
         """Remove a leaf directory and all empty intermediate ones"""
         return util.removedirs(self.join(path))
 
     def rmdir(self, path=None):
         """Remove an empty directory."""
         return os.rmdir(self.join(path))
 
     def rmtree(self, path=None, ignore_errors=False, forcibly=False):
         """Remove a directory tree recursively
 
         If ``forcibly``, this tries to remove READ-ONLY files, too.
         """
         if forcibly:
 
             def onerror(function, path, excinfo):
                 if function is not os.remove:
                     raise
                 # read-only files cannot be unlinked under Windows
                 s = os.stat(path)
                 if (s.st_mode & stat.S_IWRITE) != 0:
                     raise
                 os.chmod(path, stat.S_IMODE(s.st_mode) | stat.S_IWRITE)
                 os.remove(path)
 
         else:
             onerror = None
         return shutil.rmtree(
             self.join(path), ignore_errors=ignore_errors, onerror=onerror
         )
 
     def setflags(self, path, l, x):
         return util.setflags(self.join(path), l, x)
 
     def stat(self, path=None):
         return os.stat(self.join(path))
 
     def unlink(self, path=None):
         return util.unlink(self.join(path))
 
     def tryunlink(self, path=None):
         """Attempt to remove a file, ignoring missing file errors."""
         util.tryunlink(self.join(path))
 
     def unlinkpath(self, path=None, ignoremissing=False, rmdir=True):
         return util.unlinkpath(
             self.join(path), ignoremissing=ignoremissing, rmdir=rmdir
         )
 
     def utime(self, path=None, t=None):
         return os.utime(self.join(path), t)
 
     def walk(self, path=None, onerror=None):
         """Yield (dirpath, dirs, files) tuple for each directories under path
 
         ``dirpath`` is relative one from the root of this vfs. This
         uses ``os.sep`` as path separator, even you specify POSIX
         style ``path``.
 
         "The root of this vfs" is represented as empty ``dirpath``.
         """
         root = os.path.normpath(self.join(None))
         # when dirpath == root, dirpath[prefixlen:] becomes empty
         # because len(dirpath) < prefixlen.
         prefixlen = len(pathutil.normasprefix(root))
         for dirpath, dirs, files in os.walk(self.join(path), onerror=onerror):
             yield (dirpath[prefixlen:], dirs, files)
 
     @contextlib.contextmanager
     def backgroundclosing(self, ui, expectedcount=-1):
         """Allow files to be closed asynchronously.
 
         When this context manager is active, ``backgroundclose`` can be passed
         to ``__call__``/``open`` to result in the file possibly being closed
         asynchronously, on a background thread.
         """
         # Sharing backgroundfilecloser between threads is complex and using
         # multiple instances puts us at risk of running out of file descriptors
         # only allow to use backgroundfilecloser when in main thread.
         if not isinstance(
             threading.current_thread(),
             threading._MainThread,  # pytype: disable=module-attr
         ):
             yield
             return
         vfs = getattr(self, 'vfs', self)
         if getattr(vfs, '_backgroundfilecloser', None):
             raise error.Abort(
                 _(b'can only have 1 active background file closer')
             )
 
         with backgroundfilecloser(ui, expectedcount=expectedcount) as bfc:
             try:
                 vfs._backgroundfilecloser = (
                     bfc  # pytype: disable=attribute-error
                 )
                 yield bfc
             finally:
                 vfs._backgroundfilecloser = (
                     None  # pytype: disable=attribute-error
                 )
 
+    def register_file(self, path):
+        """generic hook point to lets fncache steer its stew"""
+
 
 class vfs(abstractvfs):
     """Operate 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.
 
     'cacheaudited' should be enabled only if (a) vfs object is short-lived, or
     (b) the base directory is managed by hg and considered sort-of append-only.
     See pathutil.pathauditor() for details.
     """
 
     def __init__(
         self,
         base,
         audit=True,
         cacheaudited=False,
         expandpath=False,
         realpath=False,
     ):
         if expandpath:
             base = util.expandpath(base)
         if realpath:
             base = os.path.realpath(base)
         self.base = base
         self._audit = audit
         if audit:
             self.audit = pathutil.pathauditor(self.base, cached=cacheaudited)
         else:
             self.audit = lambda path, mode=None: True
         self.createmode = None
         self._trustnlink = None
         self.options = {}
 
     @util.propertycache
     def _cansymlink(self):
         return util.checklink(self.base)
 
     @util.propertycache
     def _chmod(self):
         return util.checkexec(self.base)
 
     def _fixfilemode(self, name):
         if self.createmode is None or not self._chmod:
             return
         os.chmod(name, self.createmode & 0o666)
 
     def _auditpath(self, path, mode):
         if self._audit:
             if os.path.isabs(path) and path.startswith(self.base):
                 path = os.path.relpath(path, self.base)
             r = util.checkosfilename(path)
             if r:
                 raise error.Abort(b"%s: %r" % (r, path))
             self.audit(path, mode=mode)
 
     def __call__(
         self,
         path,
         mode=b"r",
         atomictemp=False,
         notindexed=False,
         backgroundclose=False,
         checkambig=False,
         auditpath=True,
         makeparentdirs=True,
     ):
         """Open ``path`` file, which is relative to vfs root.
 
         By default, parent directories are created as needed. Newly created
         directories are marked as "not to be indexed by the content indexing
         service", if ``notindexed`` is specified for "write" mode access.
         Set ``makeparentdirs=False`` to not create directories implicitly.
 
         If ``backgroundclose`` is passed, the file may be closed asynchronously.
         It can only be used if the ``self.backgroundclosing()`` context manager
         is active. This should only be specified if the following criteria hold:
 
         1. There is a potential for writing thousands of files. Unless you
            are writing thousands of files, the performance benefits of
            asynchronously closing files is not realized.
         2. Files are opened exactly once for the ``backgroundclosing``
            active duration and are therefore free of race conditions between
            closing a file on a background thread and reopening it. (If the
            file were opened multiple times, there could be unflushed data
            because the original file handle hasn't been flushed/closed yet.)
 
         ``checkambig`` argument is passed to atomictempfile (valid
         only for writing), and is useful only if target file is
         guarded by any lock (e.g. repo.lock or repo.wlock).
 
         To avoid file stat ambiguity forcibly, checkambig=True involves
         copying ``path`` file opened in "append" mode (e.g. for
         truncation), if it is owned by another. Therefore, use
         combination of append mode and checkambig=True only in limited
         cases (see also issue5418 and issue5584 for detail).
         """
         if auditpath:
             self._auditpath(path, mode)
         f = self.join(path)
 
         if b"b" not in mode:
             mode += b"b"  # for that other OS
 
         nlink = -1
         if mode not in (b'r', b'rb'):
             dirname, basename = util.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:
                 if atomictemp:
                     if makeparentdirs:
                         util.makedirs(dirname, self.createmode, notindexed)
                     return util.atomictempfile(
                         f, mode, self.createmode, checkambig=checkambig
                     )
                 try:
                     if b'w' in mode:
                         util.unlink(f)
                         nlink = 0
                     else:
                         # nlinks() may behave differently for files on Windows
                         # shares if the file is open.
                         with util.posixfile(f):
                             nlink = util.nlinks(f)
                             if nlink < 1:
                                 nlink = 2  # force mktempcopy (issue1922)
                 except (OSError, IOError) as e:
                     if e.errno != errno.ENOENT:
                         raise
                     nlink = 0
                     if makeparentdirs:
                         util.makedirs(dirname, self.createmode, notindexed)
                 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)
 
         if checkambig:
             if mode in (b'r', b'rb'):
                 raise error.Abort(
                     _(
                         b'implementation error: mode %s is not'
                         b' valid for checkambig=True'
                     )
                     % mode
                 )
             fp = checkambigatclosing(fp)
 
         if backgroundclose and isinstance(
             threading.current_thread(),
             threading._MainThread,  # pytype: disable=module-attr
         ):
             if (
                 not self._backgroundfilecloser  # pytype: disable=attribute-error
             ):
                 raise error.Abort(
                     _(
                         b'backgroundclose can only be used when a '
                         b'backgroundclosing context manager is active'
                     )
                 )
 
             fp = delayclosedfile(
                 fp,
                 self._backgroundfilecloser,  # pytype: disable=attribute-error
             )
 
         return fp
 
     def symlink(self, src, dst):
         self.audit(dst)
         linkname = self.join(dst)
         util.tryunlink(linkname)
 
         util.makedirs(os.path.dirname(linkname), self.createmode)
 
         if self._cansymlink:
             try:
                 os.symlink(src, linkname)
             except OSError as err:
                 raise OSError(
                     err.errno,
                     _(b'could not symlink to %r: %s')
                     % (src, encoding.strtolocal(err.strerror)),
                     linkname,
                 )
         else:
             self.write(dst, src)
 
     def join(self, path, *insidef):
         if path:
             return os.path.join(self.base, path, *insidef)
         else:
             return self.base
 
 
 opener = vfs
 
 
 class proxyvfs(abstractvfs):
     def __init__(self, vfs):
         self.vfs = vfs
 
     def _auditpath(self, path, mode):
         return self.vfs._auditpath(path, mode)
 
     @property
     def options(self):
         return self.vfs.options
 
     @options.setter
     def options(self, value):
         self.vfs.options = value
 
 
 class filtervfs(proxyvfs, abstractvfs):
     '''Wrapper vfs for filtering filenames with a function.'''
 
     def __init__(self, vfs, filter):
         proxyvfs.__init__(self, vfs)
         self._filter = filter
 
     def __call__(self, path, *args, **kwargs):
         return self.vfs(self._filter(path), *args, **kwargs)
 
     def join(self, path, *insidef):
         if path:
             return self.vfs.join(self._filter(self.vfs.reljoin(path, *insidef)))
         else:
             return self.vfs.join(path)
 
 
 filteropener = filtervfs
 
 
 class readonlyvfs(proxyvfs):
     '''Wrapper vfs preventing any writing.'''
 
     def __init__(self, vfs):
         proxyvfs.__init__(self, vfs)
 
     def __call__(self, path, mode=b'r', *args, **kw):
         if mode not in (b'r', b'rb'):
             raise error.Abort(_(b'this vfs is read only'))
         return self.vfs(path, mode, *args, **kw)
 
     def join(self, path, *insidef):
         return self.vfs.join(path, *insidef)
 
 
 class closewrapbase(object):
     """Base class of wrapper, which hooks closing
 
     Do not instantiate outside of the vfs layer.
     """
 
     def __init__(self, fh):
         object.__setattr__(self, '_origfh', fh)
 
     def __getattr__(self, attr):
         return getattr(self._origfh, attr)
 
     def __setattr__(self, attr, value):
         return setattr(self._origfh, attr, value)
 
     def __delattr__(self, attr):
         return delattr(self._origfh, attr)
 
     def __enter__(self):
         self._origfh.__enter__()
         return self
 
     def __exit__(self, exc_type, exc_value, exc_tb):
         raise NotImplementedError('attempted instantiating ' + str(type(self)))
 
     def close(self):
         raise NotImplementedError('attempted instantiating ' + str(type(self)))
 
 
 class delayclosedfile(closewrapbase):
     """Proxy for a file object whose close is delayed.
 
     Do not instantiate outside of the vfs layer.
     """
 
     def __init__(self, fh, closer):
         super(delayclosedfile, self).__init__(fh)
         object.__setattr__(self, '_closer', closer)
 
     def __exit__(self, exc_type, exc_value, exc_tb):
         self._closer.close(self._origfh)
 
     def close(self):
         self._closer.close(self._origfh)
 
 
 class backgroundfilecloser(object):
     """Coordinates background closing of file handles on multiple threads."""
 
     def __init__(self, ui, expectedcount=-1):
         self._running = False
         self._entered = False
         self._threads = []
         self._threadexception = None
 
         # Only Windows/NTFS has slow file closing. So only enable by default
         # on that platform. But allow to be enabled elsewhere for testing.
         defaultenabled = pycompat.iswindows
         enabled = ui.configbool(b'worker', b'backgroundclose', defaultenabled)
 
         if not enabled:
             return
 
         # There is overhead to starting and stopping the background threads.
         # Don't do background processing unless the file count is large enough
         # to justify it.
         minfilecount = ui.configint(b'worker', b'backgroundcloseminfilecount')
         # FUTURE dynamically start background threads after minfilecount closes.
         # (We don't currently have any callers that don't know their file count)
         if expectedcount > 0 and expectedcount < minfilecount:
             return
 
         maxqueue = ui.configint(b'worker', b'backgroundclosemaxqueue')
         threadcount = ui.configint(b'worker', b'backgroundclosethreadcount')
 
         ui.debug(
             b'starting %d threads for background file closing\n' % threadcount
         )
 
         self._queue = pycompat.queue.Queue(maxsize=maxqueue)
         self._running = True
 
         for i in range(threadcount):
             t = threading.Thread(target=self._worker, name='backgroundcloser')
             self._threads.append(t)
             t.start()
 
     def __enter__(self):
         self._entered = True
         return self
 
     def __exit__(self, exc_type, exc_value, exc_tb):
         self._running = False
 
         # Wait for threads to finish closing so open files don't linger for
         # longer than lifetime of context manager.
         for t in self._threads:
             t.join()
 
     def _worker(self):
         """Main routine for worker thread."""
         while True:
             try:
                 fh = self._queue.get(block=True, timeout=0.100)
                 # Need to catch or the thread will terminate and
                 # we could orphan file descriptors.
                 try:
                     fh.close()
                 except Exception as e:
                     # Stash so can re-raise from main thread later.
                     self._threadexception = e
             except pycompat.queue.Empty:
                 if not self._running:
                     break
 
     def close(self, fh):
         """Schedule a file for closing."""
         if not self._entered:
             raise error.Abort(
                 _(b'can only call close() when context manager active')
             )
 
         # If a background thread encountered an exception, raise now so we fail
         # fast. Otherwise we may potentially go on for minutes until the error
         # is acted on.
         if self._threadexception:
             e = self._threadexception
             self._threadexception = None
             raise e
 
         # If we're not actively running, close synchronously.
         if not self._running:
             fh.close()
             return
 
         self._queue.put(fh, block=True, timeout=None)
 
 
 class checkambigatclosing(closewrapbase):
     """Proxy for a file object, to avoid ambiguity of file stat
 
     See also util.filestat for detail about "ambiguity of file stat".
 
     This proxy is useful only if the target file is guarded by any
     lock (e.g. repo.lock or repo.wlock)
 
     Do not instantiate outside of the vfs layer.
     """
 
     def __init__(self, fh):
         super(checkambigatclosing, self).__init__(fh)
         object.__setattr__(self, '_oldstat', util.filestat.frompath(fh.name))
 
     def _checkambig(self):
         oldstat = self._oldstat
         if oldstat.stat:
             _avoidambig(self._origfh.name, oldstat)
 
     def __exit__(self, exc_type, exc_value, exc_tb):
         self._origfh.__exit__(exc_type, exc_value, exc_tb)
         self._checkambig()
 
     def close(self):
         self._origfh.close()
         self._checkambig()