clone: get a list of files to clone from store

mercurial/hg.py

@@ -1,303 +1,286 b''
 # hg.py - repository classes for mercurial
 #
 # Copyright 2005-2007 Matt Mackall <[e-mail]>
 # Copyright 2006 Vadim Gelfer <[e-mail]>
 #
 # This software may be used and distributed according to the terms
 # of the GNU General Public License, incorporated herein by reference.
 
 from i18n import _
 import localrepo, bundlerepo, httprepo, sshrepo, statichttprepo
 import errno, lock, os, shutil, util, extensions
 import merge as _merge
 import verify as _verify
 
 def _local(path):
     return (os.path.isfile(util.drop_scheme('file', path)) and
             bundlerepo or localrepo)
 
 def parseurl(url, revs=[]):
     '''parse url#branch, returning url, branch + revs'''
 
     if '#' not in url:
         return url, (revs or None), None
 
     url, rev = url.split('#', 1)
     return url, revs + [rev], rev
 
 schemes = {
     'bundle': bundlerepo,
     'file': _local,
     'http': httprepo,
     'https': httprepo,
     'ssh': sshrepo,
     'static-http': statichttprepo,
 }
 
 def _lookup(path):
     scheme = 'file'
     if path:
         c = path.find(':')
         if c > 0:
             scheme = path[:c]
     thing = schemes.get(scheme) or schemes['file']
     try:
         return thing(path)
     except TypeError:
         return thing
 
 def islocal(repo):
     '''return true if repo or path is local'''
     if isinstance(repo, str):
         try:
             return _lookup(repo).islocal(repo)
         except AttributeError:
             return False
     return repo.local()
 
 def repository(ui, path='', create=False):
     """return a repository object for the specified path"""
     repo = _lookup(path).instance(ui, path, create)
     ui = getattr(repo, "ui", ui)
     for name, module in extensions.extensions():
         hook = getattr(module, 'reposetup', None)
         if hook:
             hook(ui, repo)
     return repo
 
 def defaultdest(source):
     '''return default destination of clone if none is given'''
     return os.path.basename(os.path.normpath(source))
 
 def localpath(path):
     if path.startswith('file://localhost/'):
         return path[16:]
     if path.startswith('file://'):
         return path[7:]
     if path.startswith('file:'):
         return path[5:]
     return path
 
 def clone(ui, source, dest=None, pull=False, rev=None, update=True,
           stream=False):
     """Make a copy of an existing repository.
 
     Create a copy of an existing repository in a new directory.  The
     source and destination are URLs, as passed to the repository
     function.  Returns a pair of repository objects, the source and
     newly created destination.
 
     The location of the source is added to the new repository's
     .hg/hgrc file, as the default to be used for future pulls and
     pushes.
 
     If an exception is raised, the partly cloned/updated destination
     repository will be deleted.
 
     Arguments:
 
     source: repository object or URL
 
     dest: URL of destination repository to create (defaults to base
     name of source repository)
 
     pull: always pull from source repository, even in local case
 
     stream: stream raw data uncompressed from repository (fast over
     LAN, slow over WAN)
 
     rev: revision to clone up to (implies pull=True)
 
     update: update working directory after clone completes, if
     destination is local repository (True means update to default rev,
     anything else is treated as a revision)
     """
 
     if isinstance(source, str):
         origsource = ui.expandpath(source)
         source, rev, checkout = parseurl(origsource, rev)
         src_repo = repository(ui, source)
     else:
         src_repo = source
         origsource = source = src_repo.url()
         checkout = None
 
     if dest is None:
         dest = defaultdest(source)
         ui.status(_("destination directory: %s\n") % dest)
 
     dest = localpath(dest)
     source = localpath(source)
 
     if os.path.exists(dest):
         raise util.Abort(_("destination '%s' already exists") % dest)
 
     class DirCleanup(object):
         def __init__(self, dir_):
             self.rmtree = shutil.rmtree
             self.dir_ = dir_
         def close(self):
             self.dir_ = None
         def __del__(self):
             if self.dir_:
                 self.rmtree(self.dir_, True)
 
     src_lock = dest_lock = dir_cleanup = None
     try:
         if islocal(dest):
             dir_cleanup = DirCleanup(dest)
 
         abspath = origsource
         copy = False
         if src_repo.cancopy() and islocal(dest):
             abspath = os.path.abspath(util.drop_scheme('file', origsource))
             copy = not pull and not rev
 
         if copy:
             try:
                 # we use a lock here because if we race with commit, we
                 # can end up with extra data in the cloned revlogs that's
                 # not pointed to by changesets, thus causing verify to
                 # fail
                 src_lock = src_repo.lock()
             except lock.LockException:
                 copy = False
 
         if copy:
-            def force_copy(src, dst):
-                if not os.path.exists(src):
-                    # Tolerate empty source repository and optional files
-                    return
-                util.copyfiles(src, dst)
-
-            src_store = os.path.realpath(src_repo.spath)
             if not os.path.exists(dest):
                 os.mkdir(dest)
             try:
                 dest_path = os.path.realpath(os.path.join(dest, ".hg"))
                 os.mkdir(dest_path)
             except OSError, inst:
                 if inst.errno == errno.EEXIST:
                     dir_cleanup.close()
                     raise util.Abort(_("destination '%s' already exists")
                                      % dest)
                 raise
-            if src_repo.spath != src_repo.path:
-                # XXX racy
-                dummy_changelog = os.path.join(dest_path, "00changelog.i")
-                # copy the dummy changelog
-                force_copy(src_repo.join("00changelog.i"), dummy_changelog)
-                dest_store = os.path.join(dest_path, "store")
-                os.mkdir(dest_store)
-            else:
-                dest_store = dest_path
-            # copy the requires file
-            force_copy(src_repo.join("requires"),
-                       os.path.join(dest_path, "requires"))
-            # we lock here to avoid premature writing to the target
-            dest_lock = lock.lock(os.path.join(dest_store, "lock"))
 
-            files = ("data",
-                     "00manifest.d", "00manifest.i",
-                     "00changelog.d", "00changelog.i")
-            for f in files:
-                src = os.path.join(src_store, f)
-                dst = os.path.join(dest_store, f)
-                force_copy(src, dst)
+            for f in src_repo.store.copylist():
+                src = os.path.join(src_repo.path, f)
+                if os.path.exists(src):
+                    dst = os.path.join(dest_path, f)
+                    dstbase = os.path.dirname(dst)
+                    if not os.path.exists(dstbase):
+                        os.mkdir(dstbase)
+                    if dst.endswith('data'):
+                        # lock to avoid premature writing to the target
+                        dest_lock = lock.lock(os.path.join(dstbase, "lock"))
+                    util.copyfiles(src, dst)
 
             # we need to re-init the repo after manually copying the data
             # into it
             dest_repo = repository(ui, dest)
 
         else:
             try:
                 dest_repo = repository(ui, dest, create=True)
             except OSError, inst:
                 if inst.errno == errno.EEXIST:
                     dir_cleanup.close()
                     raise util.Abort(_("destination '%s' already exists")
                                      % dest)
                 raise
 
             revs = None
             if rev:
                 if 'lookup' not in src_repo.capabilities:
                     raise util.Abort(_("src repository does not support revision "
                                        "lookup and so doesn't support clone by "
                                        "revision"))
                 revs = [src_repo.lookup(r) for r in rev]
 
             if dest_repo.local():
                 dest_repo.clone(src_repo, heads=revs, stream=stream)
             elif src_repo.local():
                 src_repo.push(dest_repo, revs=revs)
             else:
                 raise util.Abort(_("clone from remote to remote not supported"))
 
         if dir_cleanup:
             dir_cleanup.close()
 
         if dest_repo.local():
             fp = dest_repo.opener("hgrc", "w", text=True)
             fp.write("[paths]\n")
             fp.write("default = %s\n" % abspath)
             fp.close()
 
             if update:
                 dest_repo.ui.status(_("updating working directory\n"))
                 if update is not True:
                     checkout = update
                 elif not checkout:
                     try:
                         checkout = dest_repo.lookup("default")
                     except:
                         checkout = dest_repo.changelog.tip()
                 _update(dest_repo, checkout)
 
         return src_repo, dest_repo
     finally:
         del src_lock, dest_lock, dir_cleanup
 
 def _showstats(repo, stats):
     stats = ((stats[0], _("updated")),
              (stats[1], _("merged")),
              (stats[2], _("removed")),
              (stats[3], _("unresolved")))
     note = ", ".join([_("%d files %s") % s for s in stats])
     repo.ui.status("%s\n" % note)
 
 def _update(repo, node): return update(repo, node)
 
 def update(repo, node):
     """update the working directory to node, merging linear changes"""
     pl = repo.parents()
     stats = _merge.update(repo, node, False, False, None)
     _showstats(repo, stats)
     if stats[3]:
         repo.ui.status(_("use 'hg resolve' to retry unresolved file merges\n"))
     return stats[3] > 0
 
 def clean(repo, node, show_stats=True):
     """forcibly switch the working directory to node, clobbering changes"""
     stats = _merge.update(repo, node, False, True, None)
     if show_stats: _showstats(repo, stats)
     return stats[3] > 0
 
 def merge(repo, node, force=None, remind=True):
     """branch merge with node, resolving changes"""
     stats = _merge.update(repo, node, True, force, False)
     _showstats(repo, stats)
     if stats[3]:
         pl = repo.parents()
         repo.ui.status(_("use 'hg resolve' to retry unresolved file merges\n"))
     elif remind:
         repo.ui.status(_("(branch merge, don't forget to commit)\n"))
     return stats[3] > 0
 
 def revert(repo, node, choose):
     """revert changes to revision in node without updating dirstate"""
     return _merge.update(repo, node, False, True, choose)[3] > 0
 
 def verify(repo):
     """verify the consistency of a repository"""
     return _verify.verify(repo)

mercurial/store.py

@@ -1,115 +1,124 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, incorporated herein by reference.
 
 import os, stat, osutil, util
 
 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 s]),
             lambda s: "".join(list(decode(s))))
 
 encodefilename, decodefilename = _buildencodefun()
 
 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:
     '''base class for local repository stores'''
     def __init__(self, path, opener):
         self.path = path
         self.createmode = _calcmode(path)
         self.opener = opener(self.path)
         self.opener.createmode = self.createmode
 
     def join(self, f):
         return os.path.join(self.path, f)
 
     def _walk(self, relpath, recurse):
         '''yields (unencoded, encoded, size)'''
         path = os.path.join(self.path, relpath)
         striplen = len(self.path) + len(os.sep)
         prefix = path[striplen:]
         l = []
         if os.path.isdir(path):
             visit = [path]
             while visit:
                 p = visit.pop()
                 for f, kind, st in osutil.listdir(p, stat=True):
                     fp = os.path.join(p, f)
                     if kind == stat.S_IFREG and f[-2:] in ('.d', '.i'):
                         n = util.pconvert(fp[striplen:])
                         l.append((n, n, st.st_size))
                     elif kind == stat.S_IFDIR and recurse:
                         visit.append(fp)
         return util.sort(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
         meta = self._walk('', False)
         meta.reverse()
         for x in meta:
             yield x
 
+    def copylist(self):
+        return ['requires'] + _data.split()
+
 class encodedstore(basicstore):
     def __init__(self, path, opener):
         self.path = os.path.join(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 os.path.join(self.path, encodefilename(f))
 
+    def copylist(self):
+        return (['requires', '00changelog.i'] +
+                ['store/' + f for f in _data.split()])
+
 def store(requirements, path, opener):
     if 'store' in requirements:
         return encodedstore(path, opener)
     return basicstore(path, opener)