store: only add new entries to the fncache file

Newly added fncache entries were not added to the in-memory cache,
making it possible for 'hg convert' to cause duplicates in
.hg/store/fncache.

Duplicates in the fncache file are harmless, but excessive numbers
of duplicates from large converted repositories may slow down
execution speed considerably.

mercurial/store.py

@@ -1,333 +1,333
 # 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 i18n import _
 import osutil, util
 import os, stat
 
 _sha = util.sha1
 
 # This avoids a collision between a file named foo and a dir named
 # foo.i or foo.d
 def encodedir(path):
     if not path.startswith('data/'):
         return path
     return (path
             .replace(".hg/", ".hg.hg/")
             .replace(".i/", ".i.hg/")
             .replace(".d/", ".d.hg/"))
 
 def decodedir(path):
     if not path.startswith('data/'):
         return path
     return (path
             .replace(".d.hg/", ".d/")
             .replace(".i.hg/", ".i/")
             .replace(".hg.hg/", ".hg/"))
 
 def _buildencodefun():
     e = '_'
     win_reserved = [ord(x) for x in '\\:*?"<>|']
     cmap = dict([(chr(x), chr(x)) for x in xrange(127)])
     for x in (range(32) + range(126, 256) + win_reserved):
         cmap[chr(x)] = "~%02x" % x
     for x in range(ord("A"), ord("Z")+1) + [ord(e)]:
         cmap[chr(x)] = e + chr(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[c] for c in encodedir(s)]),
             lambda s: decodedir("".join(list(decode(s)))))
 
 encodefilename, decodefilename = _buildencodefun()
 
 def _build_lower_encodefun():
     win_reserved = [ord(x) for x in '\\:*?"<>|']
     cmap = dict([(chr(x), chr(x)) for x in xrange(127)])
     for x in (range(32) + range(126, 256) + win_reserved):
         cmap[chr(x)] = "~%02x" % x
     for x in range(ord("A"), ord("Z")+1):
         cmap[chr(x)] = chr(x).lower()
     return lambda s: "".join([cmap[c] for c in s])
 
 lowerencode = _build_lower_encodefun()
 
 _windows_reserved_filenames = '''con prn aux nul
     com1 com2 com3 com4 com5 com6 com7 com8 com9
     lpt1 lpt2 lpt3 lpt4 lpt5 lpt6 lpt7 lpt8 lpt9'''.split()
 def auxencode(path):
     res = []
     for n in path.split('/'):
         if n:
             base = n.split('.')[0]
             if base and (base in _windows_reserved_filenames):
                 # encode third letter ('aux' -> 'au~78')
                 ec = "~%02x" % ord(n[2])
                 n = n[0:2] + ec + n[3:]
             if n[-1] in '. ':
                 # encode last period or space ('foo...' -> 'foo..~2e')
                 n = n[:-1] + "~%02x" % ord(n[-1])
         res.append(n)
     return '/'.join(res)
 
 MAX_PATH_LEN_IN_HGSTORE = 120
 DIR_PREFIX_LEN = 8
 _MAX_SHORTENED_DIRS_LEN = 8 * (DIR_PREFIX_LEN + 1) - 4
 def hybridencode(path):
     '''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 MAX_PATH_LEN_IN_HGSTORE, a
     non-reversible hybrid hashing of the path is done instead.
     This encoding uses up to DIR_PREFIX_LEN characters of all directory
     levels of the lowerencoded path, but not more levels than can fit into
     _MAX_SHORTENED_DIRS_LEN.
     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 MAX_PATH_LEN_IN_HGSTORE 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.
     '''
     if not path.startswith('data/'):
         return path
     # escape directories ending with .i and .d
     path = encodedir(path)
     ndpath = path[len('data/'):]
     res = 'data/' + auxencode(encodefilename(ndpath))
     if len(res) > MAX_PATH_LEN_IN_HGSTORE:
         digest = _sha(path).hexdigest()
         aep = auxencode(lowerencode(ndpath))
         _root, ext = os.path.splitext(aep)
         parts = aep.split('/')
         basename = parts[-1]
         sdirs = []
         for p in parts[:-1]:
             d = p[:DIR_PREFIX_LEN]
             if d[-1] in '. ':
                 # Windows can't access dirs ending in period or space
                 d = d[:-1] + '_'
             t = '/'.join(sdirs) + '/' + d
             if len(t) > _MAX_SHORTENED_DIRS_LEN:
                 break
             sdirs.append(d)
         dirs = '/'.join(sdirs)
         if len(dirs) > 0:
             dirs += '/'
         res = 'dh/' + dirs + digest + ext
         space_left = MAX_PATH_LEN_IN_HGSTORE - len(res)
         if space_left > 0:
             filler = basename[:space_left]
             res = 'dh/' + dirs + filler + digest + ext
     return res
 
 def _calcmode(path):
     try:
         # files in .hg/ will be created using this mode
         mode = os.stat(path).st_mode
             # avoid some useless chmods
         if (0777 & ~util.umask) == (0777 & mode):
             mode = None
     except OSError:
         mode = None
     return mode
 
 _data = 'data 00manifest.d 00manifest.i 00changelog.d  00changelog.i'
 
 class basicstore(object):
     '''base class for local repository stores'''
     def __init__(self, path, opener, pathjoiner):
         self.pathjoiner = pathjoiner
         self.path = path
         self.createmode = _calcmode(path)
         op = opener(self.path)
         op.createmode = self.createmode
         self.opener = lambda f, *args, **kw: op(encodedir(f), *args, **kw)
 
     def join(self, f):
         return self.pathjoiner(self.path, encodedir(f))
 
     def _walk(self, relpath, recurse):
         '''yields (unencoded, encoded, size)'''
         path = self.pathjoiner(self.path, relpath)
         striplen = len(self.path) + len(os.sep)
         l = []
         if os.path.isdir(path):
             visit = [path]
             while visit:
                 p = visit.pop()
                 for f, kind, st in osutil.listdir(p, stat=True):
                     fp = self.pathjoiner(p, f)
                     if kind == stat.S_IFREG and f[-2:] in ('.d', '.i'):
                         n = util.pconvert(fp[striplen:])
                         l.append((decodedir(n), n, st.st_size))
                     elif kind == stat.S_IFDIR and recurse:
                         visit.append(fp)
         return sorted(l)
 
     def datafiles(self):
         return self._walk('data', True)
 
     def walk(self):
         '''yields (unencoded, encoded, size)'''
         # yield data files first
         for x in self.datafiles():
             yield x
         # yield manifest before changelog
         for x in reversed(self._walk('', False)):
             yield x
 
     def copylist(self):
         return ['requires'] + _data.split()
 
 class encodedstore(basicstore):
     def __init__(self, path, opener, pathjoiner):
         self.pathjoiner = pathjoiner
         self.path = self.pathjoiner(path, 'store')
         self.createmode = _calcmode(self.path)
         op = opener(self.path)
         op.createmode = self.createmode
         self.opener = lambda f, *args, **kw: op(encodefilename(f), *args, **kw)
 
     def datafiles(self):
         for a, b, size in self._walk('data', True):
             try:
                 a = decodefilename(a)
             except KeyError:
                 a = None
             yield a, b, size
 
     def join(self, f):
         return self.pathjoiner(self.path, encodefilename(f))
 
     def copylist(self):
         return (['requires', '00changelog.i'] +
                 [self.pathjoiner('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, opener):
         self.opener = opener
         self.entries = None
 
     def _load(self):
         '''fill the entries from the fncache file'''
         self.entries = set()
         try:
             fp = self.opener('fncache', mode='rb')
         except IOError:
             # skip nonexistent file
             return
         for n, line in enumerate(fp):
             if (len(line) < 2) or (line[-1] != '\n'):
                 t = _('invalid entry in fncache, line %s') % (n + 1)
                 raise util.Abort(t)
             self.entries.add(decodedir(line[:-1]))
         fp.close()
 
     def rewrite(self, files):
         fp = self.opener('fncache', mode='wb')
         for p in files:
             fp.write(encodedir(p) + '\n')
         fp.close()
         self.entries = set(files)
 
     def add(self, fn):
         if self.entries is None:
             self._load()
+        if fn not in self.entries:
         self.opener('fncache', 'ab').write(encodedir(fn) + '\n')
+            self.entries.add(fn)
 
     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 fncachestore(basicstore):
     def __init__(self, path, opener, pathjoiner):
         self.pathjoiner = pathjoiner
         self.path = self.pathjoiner(path, 'store')
         self.createmode = _calcmode(self.path)
         op = opener(self.path)
         op.createmode = self.createmode
         fnc = fncache(op)
         self.fncache = fnc
 
         def fncacheopener(path, mode='r', *args, **kw):
-            if (mode not in ('r', 'rb')
-                and path.startswith('data/')
-                and path not in fnc):
+            if mode not in ('r', 'rb') and path.startswith('data/'):
                 fnc.add(path)
             return op(hybridencode(path), mode, *args, **kw)
         self.opener = fncacheopener
 
     def join(self, f):
         return self.pathjoiner(self.path, hybridencode(f))
 
     def datafiles(self):
         rewrite = False
         existing = []
         pjoin = self.pathjoiner
         spath = self.path
         for f in self.fncache:
             ef = hybridencode(f)
             try:
                 st = os.stat(pjoin(spath, ef))
                 yield f, ef, st.st_size
                 existing.append(f)
             except OSError:
                 # nonexistent entry
                 rewrite = True
         if rewrite:
             # rewrite fncache to remove nonexistent entries
             # (may be caused by rollback / strip)
             self.fncache.rewrite(existing)
 
     def copylist(self):
         d = _data + ' dh fncache'
         return (['requires', '00changelog.i'] +
                 [self.pathjoiner('store', f) for f in d.split()])
 
 def store(requirements, path, opener, pathjoiner=None):
     pathjoiner = pathjoiner or os.path.join
     if 'store' in requirements:
         if 'fncache' in requirements:
             return fncachestore(path, opener, pathjoiner)
         return encodedstore(path, opener, pathjoiner)
     return basicstore(path, opener, pathjoiner)

tests/test-convert

@@ -1,59 +1,65
 #!/bin/sh
 
 cat >> $HGRCPATH <<EOF
 [extensions]
 convert=
 [convert]
 hg.saverev=False
 EOF
 
 hg help convert
 
 hg init a
 cd a
 echo a > a
 hg ci -d'0 0' -Ama
 hg cp a b
 hg ci -d'1 0' -mb
 hg rm a
 hg ci -d'2 0' -mc
 hg mv b a
 hg ci -d'3 0' -md
 echo a >> a
 hg ci -d'4 0' -me
 
 cd ..
 hg convert a 2>&1 | grep -v 'subversion python bindings could not be loaded'
 hg --cwd a-hg pull ../a
 
 touch bogusfile
 echo % should fail
 hg convert a bogusfile
 
 mkdir bogusdir
 chmod 000 bogusdir
 
 echo % should fail
 hg convert a bogusdir
 
 echo % should succeed
 chmod 700 bogusdir
 hg convert a bogusdir
 
 echo % test pre and post conversion actions
 echo 'include b' > filemap
 hg convert --debug --filemap filemap a partialb | \
     grep 'run hg'
 
 echo % converting empty dir should fail "nicely"
 mkdir emptydir
 # override $PATH to ensure p4 not visible; use $PYTHON in case we're
 # running from a devel copy, not a temp installation
 PATH=$BINDIR $PYTHON $BINDIR/hg convert emptydir 2>&1 | sed 's,file://.*/emptydir,.../emptydir,g'
 
 echo % convert with imaginary source type
 hg convert --source-type foo a a-foo
 echo % convert with imaginary sink type
 hg convert --dest-type foo a a-foo
 
+echo
+echo % "testing: convert must not produce duplicate entries in fncache"
+hg convert a b
+echo % "contents of fncache file:"
+cat b/.hg/store/fncache
+
 true

tests/test-convert.out

@@ -1,275 +1,289
 hg convert [OPTION]... SOURCE [DEST [REVMAP]]
 
 convert a foreign SCM repository to a Mercurial one.
 
     Accepted source formats [identifiers]:
 
     - Mercurial [hg]
     - CVS [cvs]
     - Darcs [darcs]
     - git [git]
     - Subversion [svn]
     - Monotone [mtn]
     - GNU Arch [gnuarch]
     - Bazaar [bzr]
     - Perforce [p4]
 
     Accepted destination formats [identifiers]:
 
     - Mercurial [hg]
     - Subversion [svn] (history on branches is not preserved)
 
     If no revision is given, all revisions will be converted. Otherwise,
     convert will only import up to the named revision (given in a format
     understood by the source).
 
     If no destination directory name is specified, it defaults to the basename
     of the source with '-hg' appended. If the destination repository doesn't
     exist, it will be created.
 
     By default, all sources except Mercurial will use --branchsort. Mercurial
     uses --sourcesort to preserve original revision numbers order. Sort modes
     have the following effects:
 
     --branchsort  convert from parent to child revision when possible, which
                   means branches are usually converted one after the other. It
                   generates more compact repositories.
     --datesort    sort revisions by date. Converted repositories have good-
                   looking changelogs but are often an order of magnitude
                   larger than the same ones generated by --branchsort.
     --sourcesort  try to preserve source revisions order, only supported by
                   Mercurial sources.
 
     If <REVMAP> isn't given, it will be put in a default location
     (<dest>/.hg/shamap by default). The <REVMAP> is a simple text file that
     maps each source commit ID to the destination ID for that revision, like
     so:
 
       <source ID> <destination ID>
 
     If the file doesn't exist, it's automatically created. It's updated on
     each commit copied, so convert-repo can be interrupted and can be run
     repeatedly to copy new commits.
 
     The [username mapping] file is a simple text file that maps each source
     commit author to a destination commit author. It is handy for source SCMs
     that use unix logins to identify authors (eg: CVS). One line per author
     mapping and the line format is: srcauthor=whatever string you want
 
     The filemap is a file that allows filtering and remapping of files and
     directories. Comment lines start with '#'. Each line can contain one of
     the following directives:
 
       include path/to/file
 
       exclude path/to/file
 
       rename from/file to/file
 
     The 'include' directive causes a file, or all files under a directory, to
     be included in the destination repository, and the exclusion of all other
     files and directories not explicitly included. The 'exclude' directive
     causes files or directories to be omitted. The 'rename' directive renames
     a file or directory. To rename from a subdirectory into the root of the
     repository, use '.' as the path to rename to.
 
     The splicemap is a file that allows insertion of synthetic history,
     letting you specify the parents of a revision. This is useful if you want
     to e.g. give a Subversion merge two parents, or graft two disconnected
     series of history together. Each entry contains a key, followed by a
     space, followed by one or two comma-separated values. The key is the
     revision ID in the source revision control system whose parents should be
     modified (same format as a key in .hg/shamap). The values are the revision
     IDs (in either the source or destination revision control system) that
     should be used as the new parents for that node. For example, if you have
     merged "release-1.0" into "trunk", then you should specify the revision on
     "trunk" as the first parent and the one on the "release-1.0" branch as the
     second.
 
     The branchmap is a file that allows you to rename a branch when it is
     being brought in from whatever external repository. When used in
     conjunction with a splicemap, it allows for a powerful combination to help
     fix even the most badly mismanaged repositories and turn them into nicely
     structured Mercurial repositories. The branchmap contains lines of the
     form "original_branch_name new_branch_name". "original_branch_name" is the
     name of the branch in the source repository, and "new_branch_name" is the
     name of the branch is the destination repository. This can be used to (for
     instance) move code in one repository from "default" to a named branch.
 
     Mercurial Source
     ----------------
 
     --config convert.hg.ignoreerrors=False    (boolean)
         ignore integrity errors when reading. Use it to fix Mercurial
         repositories with missing revlogs, by converting from and to
         Mercurial.
     --config convert.hg.saverev=False         (boolean)
         store original revision ID in changeset (forces target IDs to change)
     --config convert.hg.startrev=0            (hg revision identifier)
         convert start revision and its descendants
 
     CVS Source
     ----------
 
     CVS source will use a sandbox (i.e. a checked-out copy) from CVS to
     indicate the starting point of what will be converted. Direct access to
     the repository files is not needed, unless of course the repository is
     :local:. The conversion uses the top level directory in the sandbox to
     find the CVS repository, and then uses CVS rlog commands to find files to
     convert. This means that unless a filemap is given, all files under the
     starting directory will be converted, and that any directory
     reorganization in the CVS sandbox is ignored.
 
     The options shown are the defaults.
 
     --config convert.cvsps.cache=True         (boolean)
         Set to False to disable remote log caching, for testing and debugging
         purposes.
     --config convert.cvsps.fuzz=60            (integer)
         Specify the maximum time (in seconds) that is allowed between commits
         with identical user and log message in a single changeset. When very
         large files were checked in as part of a changeset then the default
         may not be long enough.
     --config convert.cvsps.mergeto='{{mergetobranch ([-\w]+)}}'
         Specify a regular expression to which commit log messages are matched.
         If a match occurs, then the conversion process will insert a dummy
         revision merging the branch on which this log message occurs to the
         branch indicated in the regex.
     --config convert.cvsps.mergefrom='{{mergefrombranch ([-\w]+)}}'
         Specify a regular expression to which commit log messages are matched.
         If a match occurs, then the conversion process will add the most
         recent revision on the branch indicated in the regex as the second
         parent of the changeset.
     --config hook.cvslog
         Specify a Python function to be called at the end of gathering the CVS
         log. The function is passed a list with the log entries, and can
         modify the entries in-place, or add or delete them.
     --config hook.cvschangesets
         Specify a Python function to be called after the changesets are
         calculated from the the CVS log. The function is passed a list with
         the changeset entries, and can modify the changesets in-place, or add
         or delete them.
 
     An additional "debugcvsps" Mercurial command allows the builtin changeset
     merging code to be run without doing a conversion. Its parameters and
     output are similar to that of cvsps 2.1. Please see the command help for
     more details.
 
     Subversion Source
     -----------------
 
     Subversion source detects classical trunk/branches/tags layouts. By
     default, the supplied "svn://repo/path/" source URL is converted as a
     single branch. If "svn://repo/path/trunk" exists it replaces the default
     branch. If "svn://repo/path/branches" exists, its subdirectories are
     listed as possible branches. If "svn://repo/path/tags" exists, it is
     looked for tags referencing converted branches. Default "trunk",
     "branches" and "tags" values can be overridden with following options. Set
     them to paths relative to the source URL, or leave them blank to disable
     auto detection.
 
     --config convert.svn.branches=branches    (directory name)
         specify the directory containing branches
     --config convert.svn.tags=tags            (directory name)
         specify the directory containing tags
     --config convert.svn.trunk=trunk          (directory name)
         specify the name of the trunk branch
 
     Source history can be retrieved starting at a specific revision, instead
     of being integrally converted. Only single branch conversions are
     supported.
 
     --config convert.svn.startrev=0           (svn revision number)
         specify start Subversion revision.
 
     Perforce Source
     ---------------
 
     The Perforce (P4) importer can be given a p4 depot path or a client
     specification as source. It will convert all files in the source to a flat
     Mercurial repository, ignoring labels, branches and integrations. Note
     that when a depot path is given you then usually should specify a target
     directory, because otherwise the target may be named ...-hg.
 
     It is possible to limit the amount of source history to be converted by
     specifying an initial Perforce revision.
 
     --config convert.p4.startrev=0            (perforce changelist number)
         specify initial Perforce revision.
 
     Mercurial Destination
     ---------------------
 
     --config convert.hg.clonebranches=False   (boolean)
         dispatch source branches in separate clones.
     --config convert.hg.tagsbranch=default    (branch name)
         tag revisions branch name
     --config convert.hg.usebranchnames=True   (boolean)
         preserve branch names
 
 options:
 
  -A --authors      username mapping filename
  -d --dest-type    destination repository type
     --filemap      remap file names using contents of file
  -r --rev          import up to target revision REV
  -s --source-type  source repository type
     --splicemap    splice synthesized history into place
     --branchmap    change branch names while converting
     --branchsort   try to sort changesets by branches
     --datesort     try to sort changesets by date
     --sourcesort   preserve source changesets order
 
 use "hg -v help convert" to show global options
 adding a
 assuming destination a-hg
 initializing destination a-hg repository
 scanning source...
 sorting...
 converting...
 4 a
 3 b
 2 c
 1 d
 0 e
 pulling from ../a
 searching for changes
 no changes found
 % should fail
 initializing destination bogusfile repository
 abort: cannot create new bundle repository
 % should fail
 abort: Permission denied: bogusdir
 % should succeed
 initializing destination bogusdir repository
 scanning source...
 sorting...
 converting...
 4 a
 3 b
 2 c
 1 d
 0 e
 % test pre and post conversion actions
 run hg source pre-conversion action
 run hg sink pre-conversion action
 run hg sink post-conversion action
 run hg source post-conversion action
 % converting empty dir should fail nicely
 assuming destination emptydir-hg
 initializing destination emptydir-hg repository
 emptydir does not look like a CVS checkout
 emptydir does not look like a Git repo
 emptydir does not look like a Subversion repo
 emptydir is not a local Mercurial repo
 emptydir does not look like a darcs repo
 emptydir does not look like a monotone repo
 emptydir does not look like a GNU Arch repo
 emptydir does not look like a Bazaar repo
 cannot find required "p4" tool
 abort: emptydir: missing or unsupported repository
 % convert with imaginary source type
 initializing destination a-foo repository
 abort: foo: invalid source repository type
 % convert with imaginary sink type
 abort: foo: invalid destination repository type
+
+% testing: convert must not produce duplicate entries in fncache
+initializing destination b repository
+scanning source...
+sorting...
+converting...
+4 a
+3 b
+2 c
+1 d
+0 e
+% contents of fncache file:
+data/a.i
+data/b.i