subrepo: don't crash when git repo is missing

mercurial/subrepo.py

@@ -1,1001 +1,1015
 # subrepo.py - sub-repository handling for Mercurial
 #
 # Copyright 2009-2010 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.
 
 import errno, os, re, xml.dom.minidom, shutil, urlparse, posixpath
 import stat, subprocess, tarfile
 from i18n import _
 import config, util, node, error, cmdutil
 hg = None
 
 nullstate = ('', '', 'empty')
 
 def state(ctx, ui):
     """return a state dict, mapping subrepo paths configured in .hgsub
     to tuple: (source from .hgsub, revision from .hgsubstate, kind
     (key in types dict))
     """
     p = config.config()
     def read(f, sections=None, remap=None):
         if f in ctx:
             try:
                 data = ctx[f].data()
             except IOError, err:
                 if err.errno != errno.ENOENT:
                     raise
                 # handle missing subrepo spec files as removed
                 ui.warn(_("warning: subrepo spec file %s not found\n") % f)
                 return
             p.parse(f, data, sections, remap, read)
         else:
             raise util.Abort(_("subrepo spec file %s not found") % f)
 
     if '.hgsub' in ctx:
         read('.hgsub')
 
     for path, src in ui.configitems('subpaths'):
         p.set('subpaths', path, src, ui.configsource('subpaths', path))
 
     rev = {}
     if '.hgsubstate' in ctx:
         try:
             for l in ctx['.hgsubstate'].data().splitlines():
                 revision, path = l.split(" ", 1)
                 rev[path] = revision
         except IOError, err:
             if err.errno != errno.ENOENT:
                 raise
 
     state = {}
     for path, src in p[''].items():
         kind = 'hg'
         if src.startswith('['):
             if ']' not in src:
                 raise util.Abort(_('missing ] in subrepo source'))
             kind, src = src.split(']', 1)
             kind = kind[1:]
 
         for pattern, repl in p.items('subpaths'):
             # Turn r'C:\foo\bar' into r'C:\\foo\\bar' since re.sub
             # does a string decode.
             repl = repl.encode('string-escape')
             # However, we still want to allow back references to go
             # through unharmed, so we turn r'\\1' into r'\1'. Again,
             # extra escapes are needed because re.sub string decodes.
             repl = re.sub(r'\\\\([0-9]+)', r'\\\1', repl)
             try:
                 src = re.sub(pattern, repl, src, 1)
             except re.error, e:
                 raise util.Abort(_("bad subrepository pattern in %s: %s")
                                  % (p.source('subpaths', pattern), e))
 
         state[path] = (src.strip(), rev.get(path, ''), kind)
 
     return state
 
 def writestate(repo, state):
     """rewrite .hgsubstate in (outer) repo with these subrepo states"""
     repo.wwrite('.hgsubstate',
                 ''.join(['%s %s\n' % (state[s][1], s)
                          for s in sorted(state)]), '')
 
 def submerge(repo, wctx, mctx, actx, overwrite):
     """delegated from merge.applyupdates: merging of .hgsubstate file
     in working context, merging context and ancestor context"""
     if mctx == actx: # backwards?
         actx = wctx.p1()
     s1 = wctx.substate
     s2 = mctx.substate
     sa = actx.substate
     sm = {}
 
     repo.ui.debug("subrepo merge %s %s %s\n" % (wctx, mctx, actx))
 
     def debug(s, msg, r=""):
         if r:
             r = "%s:%s:%s" % r
         repo.ui.debug("  subrepo %s: %s %s\n" % (s, msg, r))
 
     for s, l in s1.items():
         a = sa.get(s, nullstate)
         ld = l # local state with possible dirty flag for compares
         if wctx.sub(s).dirty():
             ld = (l[0], l[1] + "+")
         if wctx == actx: # overwrite
             a = ld
 
         if s in s2:
             r = s2[s]
             if ld == r or r == a: # no change or local is newer
                 sm[s] = l
                 continue
             elif ld == a: # other side changed
                 debug(s, "other changed, get", r)
                 wctx.sub(s).get(r, overwrite)
                 sm[s] = r
             elif ld[0] != r[0]: # sources differ
                 if repo.ui.promptchoice(
                     _(' subrepository sources for %s differ\n'
                       'use (l)ocal source (%s) or (r)emote source (%s)?')
                       % (s, l[0], r[0]),
                       (_('&Local'), _('&Remote')), 0):
                     debug(s, "prompt changed, get", r)
                     wctx.sub(s).get(r, overwrite)
                     sm[s] = r
             elif ld[1] == a[1]: # local side is unchanged
                 debug(s, "other side changed, get", r)
                 wctx.sub(s).get(r, overwrite)
                 sm[s] = r
             else:
                 debug(s, "both sides changed, merge with", r)
                 wctx.sub(s).merge(r)
                 sm[s] = l
         elif ld == a: # remote removed, local unchanged
             debug(s, "remote removed, remove")
             wctx.sub(s).remove()
         else:
             if repo.ui.promptchoice(
                 _(' local changed subrepository %s which remote removed\n'
                   'use (c)hanged version or (d)elete?') % s,
                 (_('&Changed'), _('&Delete')), 0):
                 debug(s, "prompt remove")
                 wctx.sub(s).remove()
 
     for s, r in s2.items():
         if s in s1:
             continue
         elif s not in sa:
             debug(s, "remote added, get", r)
             mctx.sub(s).get(r)
             sm[s] = r
         elif r != sa[s]:
             if repo.ui.promptchoice(
                 _(' remote changed subrepository %s which local removed\n'
                   'use (c)hanged version or (d)elete?') % s,
                 (_('&Changed'), _('&Delete')), 0) == 0:
                 debug(s, "prompt recreate", r)
                 wctx.sub(s).get(r)
                 sm[s] = r
 
     # record merged .hgsubstate
     writestate(repo, sm)
 
 def _updateprompt(ui, sub, dirty, local, remote):
     if dirty:
         msg = (_(' subrepository sources for %s differ\n'
                  'use (l)ocal source (%s) or (r)emote source (%s)?\n')
                % (subrelpath(sub), local, remote))
     else:
         msg = (_(' subrepository sources for %s differ (in checked out version)\n'
                  'use (l)ocal source (%s) or (r)emote source (%s)?\n')
                % (subrelpath(sub), local, remote))
     return ui.promptchoice(msg, (_('&Local'), _('&Remote')), 0)
 
 def reporelpath(repo):
     """return path to this (sub)repo as seen from outermost repo"""
     parent = repo
     while hasattr(parent, '_subparent'):
         parent = parent._subparent
     return repo.root[len(parent.root)+1:]
 
 def subrelpath(sub):
     """return path to this subrepo as seen from outermost repo"""
     if hasattr(sub, '_relpath'):
         return sub._relpath
     if not hasattr(sub, '_repo'):
         return sub._path
     return reporelpath(sub._repo)
 
 def _abssource(repo, push=False, abort=True):
     """return pull/push path of repo - either based on parent repo .hgsub info
     or on the top repo config. Abort or return None if no source found."""
     if hasattr(repo, '_subparent'):
         source = repo._subsource
         if source.startswith('/') or '://' in source:
             return source
         parent = _abssource(repo._subparent, push, abort=False)
         if parent:
             if '://' in parent:
                 if parent[-1] == '/':
                     parent = parent[:-1]
                 r = urlparse.urlparse(parent + '/' + source)
                 r = urlparse.urlunparse((r[0], r[1],
                                          posixpath.normpath(r[2]),
                                          r[3], r[4], r[5]))
                 return r
             else: # plain file system path
                 return posixpath.normpath(os.path.join(parent, repo._subsource))
     else: # recursion reached top repo
         if hasattr(repo, '_subtoppath'):
             return repo._subtoppath
         if push and repo.ui.config('paths', 'default-push'):
             return repo.ui.config('paths', 'default-push')
         if repo.ui.config('paths', 'default'):
             return repo.ui.config('paths', 'default')
     if abort:
         raise util.Abort(_("default path for subrepository %s not found") %
             reporelpath(repo))
 
 def itersubrepos(ctx1, ctx2):
     """find subrepos in ctx1 or ctx2"""
     # Create a (subpath, ctx) mapping where we prefer subpaths from
     # ctx1. The subpaths from ctx2 are important when the .hgsub file
     # has been modified (in ctx2) but not yet committed (in ctx1).
     subpaths = dict.fromkeys(ctx2.substate, ctx2)
     subpaths.update(dict.fromkeys(ctx1.substate, ctx1))
     for subpath, ctx in sorted(subpaths.iteritems()):
         yield subpath, ctx.sub(subpath)
 
 def subrepo(ctx, path):
     """return instance of the right subrepo class for subrepo in path"""
     # subrepo inherently violates our import layering rules
     # because it wants to make repo objects from deep inside the stack
     # so we manually delay the circular imports to not break
     # scripts that don't use our demand-loading
     global hg
     import hg as h
     hg = h
 
     util.path_auditor(ctx._repo.root)(path)
     state = ctx.substate.get(path, nullstate)
     if state[2] not in types:
         raise util.Abort(_('unknown subrepo type %s') % state[2])
     return types[state[2]](ctx, path, state[:2])
 
 # subrepo classes need to implement the following abstract class:
 
 class abstractsubrepo(object):
 
     def dirty(self, ignoreupdate=False):
         """returns true if the dirstate of the subrepo is dirty or does not
         match current stored state. If ignoreupdate is true, only check
         whether the subrepo has uncommitted changes in its dirstate.
         """
         raise NotImplementedError
 
     def checknested(self, path):
         """check if path is a subrepository within this repository"""
         return False
 
     def commit(self, text, user, date):
         """commit the current changes to the subrepo with the given
         log message. Use given user and date if possible. Return the
         new state of the subrepo.
         """
         raise NotImplementedError
 
     def remove(self):
         """remove the subrepo
 
         (should verify the dirstate is not dirty first)
         """
         raise NotImplementedError
 
     def get(self, state, overwrite=False):
         """run whatever commands are needed to put the subrepo into
         this state
         """
         raise NotImplementedError
 
     def merge(self, state):
         """merge currently-saved state with the new state."""
         raise NotImplementedError
 
     def push(self, force):
         """perform whatever action is analogous to 'hg push'
 
         This may be a no-op on some systems.
         """
         raise NotImplementedError
 
     def add(self, ui, match, dryrun, prefix):
         return []
 
     def status(self, rev2, **opts):
         return [], [], [], [], [], [], []
 
     def diff(self, diffopts, node2, match, prefix, **opts):
         pass
 
     def outgoing(self, ui, dest, opts):
         return 1
 
     def incoming(self, ui, source, opts):
         return 1
 
     def files(self):
         """return filename iterator"""
         raise NotImplementedError
 
     def filedata(self, name):
         """return file data"""
         raise NotImplementedError
 
     def fileflags(self, name):
         """return file flags"""
         return ''
 
     def archive(self, ui, archiver, prefix):
         files = self.files()
         total = len(files)
         relpath = subrelpath(self)
         ui.progress(_('archiving (%s)') % relpath, 0,
                     unit=_('files'), total=total)
         for i, name in enumerate(files):
             flags = self.fileflags(name)
             mode = 'x' in flags and 0755 or 0644
             symlink = 'l' in flags
             archiver.addfile(os.path.join(prefix, self._path, name),
                              mode, symlink, self.filedata(name))
             ui.progress(_('archiving (%s)') % relpath, i + 1,
                         unit=_('files'), total=total)
         ui.progress(_('archiving (%s)') % relpath, None)
 
 
 class hgsubrepo(abstractsubrepo):
     def __init__(self, ctx, path, state):
         self._path = path
         self._state = state
         r = ctx._repo
         root = r.wjoin(path)
         create = False
         if not os.path.exists(os.path.join(root, '.hg')):
             create = True
             util.makedirs(root)
         self._repo = hg.repository(r.ui, root, create=create)
         self._repo._subparent = r
         self._repo._subsource = state[0]
 
         if create:
             fp = self._repo.opener("hgrc", "w", text=True)
             fp.write('[paths]\n')
 
             def addpathconfig(key, value):
                 if value:
                     fp.write('%s = %s\n' % (key, value))
                     self._repo.ui.setconfig('paths', key, value)
 
             defpath = _abssource(self._repo, abort=False)
             defpushpath = _abssource(self._repo, True, abort=False)
             addpathconfig('default', defpath)
             if defpath != defpushpath:
                 addpathconfig('default-push', defpushpath)
             fp.close()
 
     def add(self, ui, match, dryrun, prefix):
         return cmdutil.add(ui, self._repo, match, dryrun, True,
                            os.path.join(prefix, self._path))
 
     def status(self, rev2, **opts):
         try:
             rev1 = self._state[1]
             ctx1 = self._repo[rev1]
             ctx2 = self._repo[rev2]
             return self._repo.status(ctx1, ctx2, **opts)
         except error.RepoLookupError, inst:
             self._repo.ui.warn(_('warning: error "%s" in subrepository "%s"\n')
                                % (inst, subrelpath(self)))
             return [], [], [], [], [], [], []
 
     def diff(self, diffopts, node2, match, prefix, **opts):
         try:
             node1 = node.bin(self._state[1])
             # We currently expect node2 to come from substate and be
             # in hex format
             if node2 is not None:
                 node2 = node.bin(node2)
             cmdutil.diffordiffstat(self._repo.ui, self._repo, diffopts,
                                    node1, node2, match,
                                    prefix=os.path.join(prefix, self._path),
                                    listsubrepos=True, **opts)
         except error.RepoLookupError, inst:
             self._repo.ui.warn(_('warning: error "%s" in subrepository "%s"\n')
                                % (inst, subrelpath(self)))
 
     def archive(self, ui, archiver, prefix):
         abstractsubrepo.archive(self, ui, archiver, prefix)
 
         rev = self._state[1]
         ctx = self._repo[rev]
         for subpath in ctx.substate:
             s = subrepo(ctx, subpath)
             s.archive(ui, archiver, os.path.join(prefix, self._path))
 
     def dirty(self, ignoreupdate=False):
         r = self._state[1]
         if r == '' and not ignoreupdate: # no state recorded
             return True
         w = self._repo[None]
         if w.p1() != self._repo[r] and not ignoreupdate:
             # different version checked out
             return True
         return w.dirty() # working directory changed
 
     def checknested(self, path):
         return self._repo._checknested(self._repo.wjoin(path))
 
     def commit(self, text, user, date):
         self._repo.ui.debug("committing subrepo %s\n" % subrelpath(self))
         n = self._repo.commit(text, user, date)
         if not n:
             return self._repo['.'].hex() # different version checked out
         return node.hex(n)
 
     def remove(self):
         # we can't fully delete the repository as it may contain
         # local-only history
         self._repo.ui.note(_('removing subrepo %s\n') % subrelpath(self))
         hg.clean(self._repo, node.nullid, False)
 
     def _get(self, state):
         source, revision, kind = state
         try:
             self._repo.lookup(revision)
         except error.RepoError:
             self._repo._subsource = source
             srcurl = _abssource(self._repo)
             self._repo.ui.status(_('pulling subrepo %s from %s\n')
                                  % (subrelpath(self), srcurl))
             other = hg.repository(self._repo.ui, srcurl)
             self._repo.pull(other)
 
     def get(self, state, overwrite=False):
         self._get(state)
         source, revision, kind = state
         self._repo.ui.debug("getting subrepo %s\n" % self._path)
         hg.clean(self._repo, revision, False)
 
     def merge(self, state):
         self._get(state)
         cur = self._repo['.']
         dst = self._repo[state[1]]
         anc = dst.ancestor(cur)
 
         def mergefunc():
             if anc == cur:
                 self._repo.ui.debug("updating subrepo %s\n" % subrelpath(self))
                 hg.update(self._repo, state[1])
             elif anc == dst:
                 self._repo.ui.debug("skipping subrepo %s\n" % subrelpath(self))
             else:
                 self._repo.ui.debug("merging subrepo %s\n" % subrelpath(self))
                 hg.merge(self._repo, state[1], remind=False)
 
         wctx = self._repo[None]
         if self.dirty():
             if anc != dst:
                 if _updateprompt(self._repo.ui, self, wctx.dirty(), cur, dst):
                     mergefunc()
             else:
                 mergefunc()
         else:
             mergefunc()
 
     def push(self, force):
         # push subrepos depth-first for coherent ordering
         c = self._repo['']
         subs = c.substate # only repos that are committed
         for s in sorted(subs):
             if not c.sub(s).push(force):
                 return False
 
         dsturl = _abssource(self._repo, True)
         self._repo.ui.status(_('pushing subrepo %s to %s\n') %
             (subrelpath(self), dsturl))
         other = hg.repository(self._repo.ui, dsturl)
         return self._repo.push(other, force)
 
     def outgoing(self, ui, dest, opts):
         return hg.outgoing(ui, self._repo, _abssource(self._repo, True), opts)
 
     def incoming(self, ui, source, opts):
         return hg.incoming(ui, self._repo, _abssource(self._repo, False), opts)
 
     def files(self):
         rev = self._state[1]
         ctx = self._repo[rev]
         return ctx.manifest()
 
     def filedata(self, name):
         rev = self._state[1]
         return self._repo[rev][name].data()
 
     def fileflags(self, name):
         rev = self._state[1]
         ctx = self._repo[rev]
         return ctx.flags(name)
 
 
 class svnsubrepo(abstractsubrepo):
     def __init__(self, ctx, path, state):
         self._path = path
         self._state = state
         self._ctx = ctx
         self._ui = ctx._repo.ui
 
     def _svncommand(self, commands, filename=''):
         path = os.path.join(self._ctx._repo.origroot, self._path, filename)
         cmd = ['svn'] + commands + [path]
         env = dict(os.environ)
         # Avoid localized output, preserve current locale for everything else.
         env['LC_MESSAGES'] = 'C'
         p = subprocess.Popen(cmd, bufsize=-1, close_fds=util.closefds,
                              stdout=subprocess.PIPE, stderr=subprocess.PIPE,
                              universal_newlines=True, env=env)
         stdout, stderr = p.communicate()
         stderr = stderr.strip()
         if stderr:
             raise util.Abort(stderr)
         return stdout
 
     def _wcrevs(self):
         # Get the working directory revision as well as the last
         # commit revision so we can compare the subrepo state with
         # both. We used to store the working directory one.
         output = self._svncommand(['info', '--xml'])
         doc = xml.dom.minidom.parseString(output)
         entries = doc.getElementsByTagName('entry')
         lastrev, rev = '0', '0'
         if entries:
             rev = str(entries[0].getAttribute('revision')) or '0'
             commits = entries[0].getElementsByTagName('commit')
             if commits:
                 lastrev = str(commits[0].getAttribute('revision')) or '0'
         return (lastrev, rev)
 
     def _wcrev(self):
         return self._wcrevs()[0]
 
     def _wcchanged(self):
         """Return (changes, extchanges) where changes is True
         if the working directory was changed, and extchanges is
         True if any of these changes concern an external entry.
         """
         output = self._svncommand(['status', '--xml'])
         externals, changes = [], []
         doc = xml.dom.minidom.parseString(output)
         for e in doc.getElementsByTagName('entry'):
             s = e.getElementsByTagName('wc-status')
             if not s:
                 continue
             item = s[0].getAttribute('item')
             props = s[0].getAttribute('props')
             path = e.getAttribute('path')
             if item == 'external':
                 externals.append(path)
             if (item not in ('', 'normal', 'unversioned', 'external')
                 or props not in ('', 'none')):
                 changes.append(path)
         for path in changes:
             for ext in externals:
                 if path == ext or path.startswith(ext + os.sep):
                     return True, True
         return bool(changes), False
 
     def dirty(self, ignoreupdate=False):
         if not self._wcchanged()[0]:
             if self._state[1] in self._wcrevs() or ignoreupdate:
                 return False
         return True
 
     def commit(self, text, user, date):
         # user and date are out of our hands since svn is centralized
         changed, extchanged = self._wcchanged()
         if not changed:
             return self._wcrev()
         if extchanged:
             # Do not try to commit externals
             raise util.Abort(_('cannot commit svn externals'))
         commitinfo = self._svncommand(['commit', '-m', text])
         self._ui.status(commitinfo)
         newrev = re.search('Committed revision ([0-9]+).', commitinfo)
         if not newrev:
             raise util.Abort(commitinfo.splitlines()[-1])
         newrev = newrev.groups()[0]
         self._ui.status(self._svncommand(['update', '-r', newrev]))
         return newrev
 
     def remove(self):
         if self.dirty():
             self._ui.warn(_('not removing repo %s because '
                             'it has changes.\n' % self._path))
             return
         self._ui.note(_('removing subrepo %s\n') % self._path)
 
         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)
 
         path = self._ctx._repo.wjoin(self._path)
         shutil.rmtree(path, onerror=onerror)
         try:
             os.removedirs(os.path.dirname(path))
         except OSError:
             pass
 
     def get(self, state, overwrite=False):
         if overwrite:
             self._svncommand(['revert', '--recursive'])
         status = self._svncommand(['checkout', '--force', state[0],
                                    '--revision', state[1]])
         if not re.search('Checked out revision [0-9]+.', status):
             raise util.Abort(status.splitlines()[-1])
         self._ui.status(status)
 
     def merge(self, state):
         old = self._state[1]
         new = state[1]
         if new != self._wcrev():
             dirty = old == self._wcrev() or self._wcchanged()[0]
             if _updateprompt(self._ui, self, dirty, self._wcrev(), new):
                 self.get(state, False)
 
     def push(self, force):
         # push is a no-op for SVN
         return True
 
     def files(self):
         output = self._svncommand(['list'])
         # This works because svn forbids \n in filenames.
         return output.splitlines()
 
     def filedata(self, name):
         return self._svncommand(['cat'], name)
 
 
 class gitsubrepo(abstractsubrepo):
     def __init__(self, ctx, path, state):
         # TODO add git version check.
         self._state = state
         self._ctx = ctx
         self._path = path
         self._relpath = os.path.join(reporelpath(ctx._repo), path)
         self._abspath = ctx._repo.wjoin(path)
         self._subparent = ctx._repo
         self._ui = ctx._repo.ui
 
     def _gitcommand(self, commands, env=None, stream=False):
         return self._gitdir(commands, env=env, stream=stream)[0]
 
     def _gitdir(self, commands, env=None, stream=False):
         return self._gitnodir(commands, env=env, stream=stream,
                               cwd=self._abspath)
 
     def _gitnodir(self, commands, env=None, stream=False, cwd=None):
         """Calls the git command
 
         The methods tries to call the git command. versions previor to 1.6.0
         are not supported and very probably fail.
         """
         self._ui.debug('%s: git %s\n' % (self._relpath, ' '.join(commands)))
         # unless ui.quiet is set, print git's stderr,
         # which is mostly progress and useful info
         errpipe = None
         if self._ui.quiet:
             errpipe = open(os.devnull, 'w')
         p = subprocess.Popen(['git'] + commands, bufsize=-1, cwd=cwd, env=env,
                              close_fds=util.closefds,
                              stdout=subprocess.PIPE, stderr=errpipe)
         if stream:
             return p.stdout, None
 
         retdata = p.stdout.read().strip()
         # wait for the child to exit to avoid race condition.
         p.wait()
 
         if p.returncode != 0 and p.returncode != 1:
             # there are certain error codes that are ok
             command = commands[0]
             if command in ('cat-file', 'symbolic-ref'):
                 return retdata, p.returncode
             # for all others, abort
             raise util.Abort('git %s error %d in %s' %
                              (command, p.returncode, self._relpath))
 
         return retdata, p.returncode
 
+    def _gitmissing(self):
+        return not os.path.exists(os.path.join(self._abspath, '.git'))
+
     def _gitstate(self):
         return self._gitcommand(['rev-parse', 'HEAD'])
 
     def _gitcurrentbranch(self):
         current, err = self._gitdir(['symbolic-ref', 'HEAD', '--quiet'])
         if err:
             current = None
         return current
 
     def _githavelocally(self, revision):
         out, code = self._gitdir(['cat-file', '-e', revision])
         return code == 0
 
     def _gitisancestor(self, r1, r2):
         base = self._gitcommand(['merge-base', r1, r2])
         return base == r1
 
     def _gitbranchmap(self):
         '''returns 2 things:
         a map from git branch to revision
         a map from revision to branches'''
         branch2rev = {}
         rev2branch = {}
 
         out = self._gitcommand(['for-each-ref', '--format',
                                 '%(objectname) %(refname)'])
         for line in out.split('\n'):
             revision, ref = line.split(' ')
             if (not ref.startswith('refs/heads/') and
                 not ref.startswith('refs/remotes/')):
                 continue
             if ref.startswith('refs/remotes/') and ref.endswith('/HEAD'):
                 continue # ignore remote/HEAD redirects
             branch2rev[ref] = revision
             rev2branch.setdefault(revision, []).append(ref)
         return branch2rev, rev2branch
 
     def _gittracking(self, branches):
         'return map of remote branch to local tracking branch'
         # assumes no more than one local tracking branch for each remote
         tracking = {}
         for b in branches:
             if b.startswith('refs/remotes/'):
                 continue
             remote = self._gitcommand(['config', 'branch.%s.remote' % b])
             if remote:
                 ref = self._gitcommand(['config', 'branch.%s.merge' % b])
                 tracking['refs/remotes/%s/%s' %
                          (remote, ref.split('/', 2)[2])] = b
         return tracking
 
     def _abssource(self, source):
         self._subsource = source
         return _abssource(self)
 
     def _fetch(self, source, revision):
-        if not os.path.exists(os.path.join(self._abspath, '.git')):
+        if self._gitmissing():
             self._ui.status(_('cloning subrepo %s\n') % self._relpath)
             self._gitnodir(['clone', self._abssource(source), self._abspath])
         if self._githavelocally(revision):
             return
         self._ui.status(_('pulling subrepo %s\n') % self._relpath)
         # try only origin: the originally cloned repo
         self._gitcommand(['fetch'])
         if not self._githavelocally(revision):
             raise util.Abort(_("revision %s does not exist in subrepo %s\n") %
                                (revision, self._relpath))
 
     def dirty(self, ignoreupdate=False):
+        if self._gitmissing():
+            return True
         if not ignoreupdate and self._state[1] != self._gitstate():
             # different version checked out
             return True
         # check for staged changes or modified files; ignore untracked files
         out, code = self._gitdir(['diff-index', '--quiet', 'HEAD'])
         return code == 1
 
     def get(self, state, overwrite=False):
         source, revision, kind = state
         self._fetch(source, revision)
         # if the repo was set to be bare, unbare it
         if self._gitcommand(['config', '--bool', 'core.bare']) == 'true':
             self._gitcommand(['config', 'core.bare', 'false'])
             if self._gitstate() == revision:
                 self._gitcommand(['reset', '--hard', 'HEAD'])
                 return
         elif self._gitstate() == revision:
             if overwrite:
                 # first reset the index to unmark new files for commit, because 
                 # reset --hard will otherwise throw away files added for commit,
                 # not just unmark them.
                 self._gitcommand(['reset', 'HEAD'])
                 self._gitcommand(['reset', '--hard', 'HEAD'])
             return
         branch2rev, rev2branch = self._gitbranchmap()
 
         def checkout(args):
             cmd = ['checkout']
             if overwrite:
                 # first reset the index to unmark new files for commit, because
                 # the -f option will otherwise throw away files added for
                 # commit, not just unmark them.
                 self._gitcommand(['reset', 'HEAD'])
                 cmd.append('-f')
             self._gitcommand(cmd + args)
 
         def rawcheckout():
             # no branch to checkout, check it out with no branch
             self._ui.warn(_('checking out detached HEAD in subrepo %s\n') %
                           self._relpath)
             self._ui.warn(_('check out a git branch if you intend '
                             'to make changes\n'))
             checkout(['-q', revision])
 
         if revision not in rev2branch:
             rawcheckout()
             return
         branches = rev2branch[revision]
         firstlocalbranch = None
         for b in branches:
             if b == 'refs/heads/master':
                 # master trumps all other branches
                 checkout(['refs/heads/master'])
                 return
             if not firstlocalbranch and not b.startswith('refs/remotes/'):
                 firstlocalbranch = b
         if firstlocalbranch:
             checkout([firstlocalbranch])
             return
 
         tracking = self._gittracking(branch2rev.keys())
         # choose a remote branch already tracked if possible
         remote = branches[0]
         if remote not in tracking:
             for b in branches:
                 if b in tracking:
                     remote = b
                     break
 
         if remote not in tracking:
             # create a new local tracking branch
             local = remote.split('/', 2)[2]
             checkout(['-b', local, remote])
         elif self._gitisancestor(branch2rev[tracking[remote]], remote):
             # When updating to a tracked remote branch,
             # if the local tracking branch is downstream of it,
             # a normal `git pull` would have performed a "fast-forward merge"
             # which is equivalent to updating the local branch to the remote.
             # Since we are only looking at branching at update, we need to
             # detect this situation and perform this action lazily.
             if tracking[remote] != self._gitcurrentbranch():
                 checkout([tracking[remote]])
             self._gitcommand(['merge', '--ff', remote])
         else:
             # a real merge would be required, just checkout the revision
             rawcheckout()
 
     def commit(self, text, user, date):
+        if self._gitmissing():
+            raise util.Abort(_("subrepo %s is missing") % self._relpath)
         cmd = ['commit', '-a', '-m', text]
         env = os.environ.copy()
         if user:
             cmd += ['--author', user]
         if date:
             # git's date parser silently ignores when seconds < 1e9
             # convert to ISO8601
             env['GIT_AUTHOR_DATE'] = util.datestr(date,
                                                   '%Y-%m-%dT%H:%M:%S %1%2')
         self._gitcommand(cmd, env=env)
         # make sure commit works otherwise HEAD might not exist under certain
         # circumstances
         return self._gitstate()
 
     def merge(self, state):
         source, revision, kind = state
         self._fetch(source, revision)
         base = self._gitcommand(['merge-base', revision, self._state[1]])
         out, code = self._gitdir(['diff-index', '--quiet', 'HEAD'])
 
         def mergefunc():
             if base == revision:
                 self.get(state) # fast forward merge
             elif base != self._state[1]:
                 self._gitcommand(['merge', '--no-commit', revision])
 
         if self.dirty():
             if self._gitstate() != revision:
                 dirty = self._gitstate() == self._state[1] or code != 0
                 if _updateprompt(self._ui, self, dirty,
                                  self._state[1][:7], revision[:7]):
                     mergefunc()
         else:
             mergefunc()
 
     def push(self, force):
+        if self._gitmissing():
+            raise util.Abort(_("subrepo %s is missing") % self._relpath)
         # if a branch in origin contains the revision, nothing to do
         branch2rev, rev2branch = self._gitbranchmap()
         if self._state[1] in rev2branch:
             for b in rev2branch[self._state[1]]:
                 if b.startswith('refs/remotes/origin/'):
                     return True
         for b, revision in branch2rev.iteritems():
             if b.startswith('refs/remotes/origin/'):
                 if self._gitisancestor(self._state[1], revision):
                     return True
         # otherwise, try to push the currently checked out branch
         cmd = ['push']
         if force:
             cmd.append('--force')
 
         current = self._gitcurrentbranch()
         if current:
             # determine if the current branch is even useful
             if not self._gitisancestor(self._state[1], current):
                 self._ui.warn(_('unrelated git branch checked out '
                                 'in subrepo %s\n') % self._relpath)
                 return False
             self._ui.status(_('pushing branch %s of subrepo %s\n') %
                             (current.split('/', 2)[2], self._relpath))
             self._gitcommand(cmd + ['origin', current])
             return True
         else:
             self._ui.warn(_('no branch checked out in subrepo %s\n'
                             'cannot push revision %s') %
                           (self._relpath, self._state[1]))
             return False
 
     def remove(self):
+        if self._gitmissing():
+            return
         if self.dirty():
             self._ui.warn(_('not removing repo %s because '
                             'it has changes.\n') % self._relpath)
             return
         # we can't fully delete the repository as it may contain
         # local-only history
         self._ui.note(_('removing subrepo %s\n') % self._relpath)
         self._gitcommand(['config', 'core.bare', 'true'])
         for f in os.listdir(self._abspath):
             if f == '.git':
                 continue
             path = os.path.join(self._abspath, f)
             if os.path.isdir(path) and not os.path.islink(path):
                 shutil.rmtree(path)
             else:
                 os.remove(path)
 
     def archive(self, ui, archiver, prefix):
         source, revision = self._state
         self._fetch(source, revision)
 
         # Parse git's native archive command.
         # This should be much faster than manually traversing the trees
         # and objects with many subprocess calls.
         tarstream = self._gitcommand(['archive', revision], stream=True)
         tar = tarfile.open(fileobj=tarstream, mode='r|')
         relpath = subrelpath(self)
         ui.progress(_('archiving (%s)') % relpath, 0, unit=_('files'))
         for i, info in enumerate(tar):
             if info.isdir():
                 continue
             if info.issym():
                 data = info.linkname
             else:
                 data = tar.extractfile(info).read()
             archiver.addfile(os.path.join(prefix, self._path, info.name),
                              info.mode, info.issym(), data)
             ui.progress(_('archiving (%s)') % relpath, i + 1,
                         unit=_('files'))
         ui.progress(_('archiving (%s)') % relpath, None)
 
 
     def status(self, rev2, **opts):
+        if self._gitmissing():
+            # if the repo is missing, return no results
+            return [], [], [], [], [], [], []
         rev1 = self._state[1]
         modified, added, removed = [], [], []
         if rev2:
             command = ['diff-tree', rev1, rev2]
         else:
             command = ['diff-index', rev1]
         out = self._gitcommand(command)
         for line in out.split('\n'):
             tab = line.find('\t')
             if tab == -1:
                 continue
             status, f = line[tab - 1], line[tab + 1:]
             if status == 'M':
                 modified.append(f)
             elif status == 'A':
                 added.append(f)
             elif status == 'D':
                 removed.append(f)
 
         deleted = unknown = ignored = clean = []
         return modified, added, removed, deleted, unknown, ignored, clean
 
 types = {
     'hg': hgsubrepo,
     'svn': svnsubrepo,
     'git': gitsubrepo,
     }

tests/test-subrepo-git.t

@@ -1,454 +1,474
   $ "$TESTDIR/hghave" git || exit 80
 
 make git commits repeatable
 
   $ GIT_AUTHOR_NAME='test'; export GIT_AUTHOR_NAME
   $ GIT_AUTHOR_EMAIL='[e-mail]'; export GIT_AUTHOR_EMAIL
   $ GIT_AUTHOR_DATE='1234567891 +0000'; export GIT_AUTHOR_DATE
   $ GIT_COMMITTER_NAME="$GIT_AUTHOR_NAME"; export GIT_COMMITTER_NAME
   $ GIT_COMMITTER_EMAIL="$GIT_AUTHOR_EMAIL"; export GIT_COMMITTER_EMAIL
   $ GIT_COMMITTER_DATE="$GIT_AUTHOR_DATE"; export GIT_COMMITTER_DATE
 
 root hg repo
 
   $ hg init t
   $ cd t
   $ echo a > a
   $ hg add a
   $ hg commit -m a
   $ cd ..
 
 new external git repo
 
   $ mkdir gitroot
   $ cd gitroot
   $ git init -q
   $ echo g > g
   $ git add g
   $ git commit -q -m g
 
 add subrepo clone
 
   $ cd ../t
   $ echo 's = [git]../gitroot' > .hgsub
   $ git clone -q ../gitroot s
   $ hg add .hgsub
   $ hg commit -m 'new git subrepo'
   committing subrepository s
   $ hg debugsub
   path s
    source   ../gitroot
    revision da5f5b1d8ffcf62fb8327bcd3c89a4367a6018e7
 
 record a new commit from upstream from a different branch
 
   $ cd ../gitroot
   $ git checkout -q -b testing
   $ echo gg >> g
   $ git commit -q -a -m gg
 
   $ cd ../t/s
   $ git pull -q >/dev/null 2>/dev/null
   $ git checkout -q -b testing origin/testing >/dev/null
 
   $ cd ..
   $ hg status --subrepos
   M s/g
   $ hg commit -m 'update git subrepo'
   committing subrepository s
   $ hg debugsub
   path s
    source   ../gitroot
    revision 126f2a14290cd5ce061fdedc430170e8d39e1c5a
 
 make $GITROOT pushable, by replacing it with a clone with nothing checked out
 
   $ cd ..
   $ git clone gitroot gitrootbare --bare -q
   $ rm -rf gitroot
   $ mv gitrootbare gitroot
 
 clone root
 
   $ cd t
   $ hg clone . ../tc
   updating to branch default
   cloning subrepo s
   3 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ cd ../tc
   $ hg debugsub
   path s
    source   ../gitroot
    revision 126f2a14290cd5ce061fdedc430170e8d39e1c5a
 
 update to previous substate
 
   $ hg update 1 -q
   $ cat s/g
   g
   $ hg debugsub
   path s
    source   ../gitroot
    revision da5f5b1d8ffcf62fb8327bcd3c89a4367a6018e7
 
 clone root, make local change
 
   $ cd ../t
   $ hg clone . ../ta
   updating to branch default
   cloning subrepo s
   3 files updated, 0 files merged, 0 files removed, 0 files unresolved
 
   $ cd ../ta
   $ echo ggg >> s/g
   $ hg status --subrepos
   M s/g
   $ hg commit -m ggg
   committing subrepository s
   $ hg debugsub
   path s
    source   ../gitroot
    revision 79695940086840c99328513acbe35f90fcd55e57
 
 clone root separately, make different local change
 
   $ cd ../t
   $ hg clone . ../tb
   updating to branch default
   cloning subrepo s
   3 files updated, 0 files merged, 0 files removed, 0 files unresolved
 
   $ cd ../tb/s
   $ echo f > f
   $ git add f
   $ cd ..
 
   $ hg status --subrepos
   A s/f
   $ hg commit -m f
   committing subrepository s
   $ hg debugsub
   path s
    source   ../gitroot
    revision aa84837ccfbdfedcdcdeeedc309d73e6eb069edc
 
 user b push changes
 
   $ hg push 2>/dev/null
   pushing to $TESTTMP/t
   pushing branch testing of subrepo s
   searching for changes
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 1 changes to 1 files
 
 user a pulls, merges, commits
 
   $ cd ../ta
   $ hg pull
   pulling from $TESTTMP/t
   searching for changes
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 1 changes to 1 files (+1 heads)
   (run 'hg heads' to see heads, 'hg merge' to merge)
   $ hg merge 2>/dev/null
   pulling subrepo s
   0 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ cat s/f
   f
   $ cat s/g
   g
   gg
   ggg
   $ hg commit -m 'merge'
   committing subrepository s
   $ hg status --subrepos --rev 1:5
   M .hgsubstate
   M s/g
   A s/f
   $ hg debugsub
   path s
    source   ../gitroot
    revision f47b465e1bce645dbf37232a00574aa1546ca8d3
   $ hg push 2>/dev/null
   pushing to $TESTTMP/t
   pushing branch testing of subrepo s
   searching for changes
   adding changesets
   adding manifests
   adding file changes
   added 2 changesets with 2 changes to 1 files
 
 make upstream git changes
 
   $ cd ..
   $ git clone -q gitroot gitclone
   $ cd gitclone
   $ echo ff >> f
   $ git commit -q -a -m ff
   $ echo fff >> f
   $ git commit -q -a -m fff
   $ git push origin testing 2>/dev/null
 
 make and push changes to hg without updating the subrepo
 
   $ cd ../t
   $ hg clone . ../td
   updating to branch default
   cloning subrepo s
   checking out detached HEAD in subrepo s
   check out a git branch if you intend to make changes
   3 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ cd ../td
   $ echo aa >> a
   $ hg commit -m aa
   $ hg push
   pushing to $TESTTMP/t
   searching for changes
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 1 changes to 1 files
 
 sync to upstream git, distribute changes
 
   $ cd ../ta
   $ hg pull -u -q
   $ cd s
   $ git pull -q >/dev/null 2>/dev/null
   $ cd ..
   $ hg commit -m 'git upstream sync'
   committing subrepository s
   $ hg debugsub
   path s
    source   ../gitroot
    revision 32a343883b74769118bb1d3b4b1fbf9156f4dddc
   $ hg push -q
 
   $ cd ../tb
   $ hg pull -q
   $ hg update 2>/dev/null
   pulling subrepo s
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg debugsub
   path s
    source   ../gitroot
    revision 32a343883b74769118bb1d3b4b1fbf9156f4dddc
 
 update to a revision without the subrepo, keeping the local git repository
 
   $ cd ../t
   $ hg up 0
   0 files updated, 0 files merged, 2 files removed, 0 files unresolved
   $ ls -a s
   .
   ..
   .git
 
   $ hg up 2
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ ls -a s
   .
   ..
   .git
   g
 
 archive subrepos
 
   $ cd ../tc
   $ hg pull -q
   $ hg archive --subrepos -r 5 ../archive 2>/dev/null
   pulling subrepo s
   $ cd ../archive
   $ cat s/f
   f
   $ cat s/g
   g
   gg
   ggg
 
 create nested repo
 
   $ cd ..
   $ hg init outer
   $ cd outer
   $ echo b>b
   $ hg add b
   $ hg commit -m b
 
   $ hg clone ../t inner
   updating to branch default
   cloning subrepo s
   3 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ echo inner = inner > .hgsub
   $ hg add .hgsub
   $ hg commit -m 'nested sub'
   committing subrepository inner
 
 nested commit
 
   $ echo ffff >> inner/s/f
   $ hg status --subrepos
   M inner/s/f
   $ hg commit -m nested
   committing subrepository inner
   committing subrepository inner/s
 
 nested archive
 
   $ hg archive --subrepos ../narchive
   $ ls ../narchive/inner/s | grep -v pax_global_header
   f
   g
 
 relative source expansion
 
   $ cd ..
   $ mkdir d
   $ hg clone t d/t
   updating to branch default
   cloning subrepo s
   3 files updated, 0 files merged, 0 files removed, 0 files unresolved
 
+Don't crash if the subrepo is missing
+
+  $ hg clone t missing -q
+  $ cd missing
+  $ rm -rf s
+  $ hg status -S
+  $ hg sum | grep commit
+  commit: 1 subrepos
+  $ hg push -q
+  abort: subrepo s is missing
+  [255]
+  $ hg commit -qm missing
+  abort: subrepo s is missing
+  [255]
+  $ hg update -C
+  cloning subrepo s
+  1 files updated, 0 files merged, 0 files removed, 0 files unresolved
+  $ hg sum | grep commit
+  commit: (clean)
+
 Check hg update --clean
   $ cd $TESTTMP/ta
   $ echo  > s/g
   $ cd s
   $ echo c1 > f1
   $ echo c1 > f2
   $ git add f1
   $ cd ..
   $ hg status -S
   M s/g
   A s/f1
   $ ls s
   f
   f1
   f2
   g
   $ hg update --clean
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg status -S
   $ ls s
   f
   f1
   f2
   g
 
 Sticky subrepositories, no changes
   $ cd $TESTTMP/ta
   $ hg id -n
   7
   $ cd s
   $ git rev-parse HEAD
   32a343883b74769118bb1d3b4b1fbf9156f4dddc
   $ cd ..
   $ hg update 1 > /dev/null 2>&1
   $ hg id -n
   1
   $ cd s
   $ git rev-parse HEAD
   da5f5b1d8ffcf62fb8327bcd3c89a4367a6018e7
   $ cd ..
 
 Sticky subrepositorys, file changes
   $ touch s/f1
   $ cd s
   $ git add f1
   $ cd ..
   $ hg id -n
   1
   $ cd s
   $ git rev-parse HEAD
   da5f5b1d8ffcf62fb8327bcd3c89a4367a6018e7
   $ cd ..
   $ hg update 4
    subrepository sources for s differ
   use (l)ocal source (da5f5b1) or (r)emote source (aa84837)?
    l
   0 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg id -n
   4+
   $ cd s
   $ git rev-parse HEAD
   da5f5b1d8ffcf62fb8327bcd3c89a4367a6018e7
   $ cd ..
   $ hg update --clean tip > /dev/null 2>&1 
 
 Sticky subrepository, revision updates
   $ hg id -n
   7
   $ cd s
   $ git rev-parse HEAD
   32a343883b74769118bb1d3b4b1fbf9156f4dddc
   $ cd ..
   $ cd s
   $ git checkout aa84837ccfbdfedcdcdeeedc309d73e6eb069edc
   Previous HEAD position was 32a3438... fff
   HEAD is now at aa84837... f
   $ cd ..
   $ hg update 1
    subrepository sources for s differ (in checked out version)
   use (l)ocal source (32a3438) or (r)emote source (da5f5b1)?
    l
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg id -n
   1+
   $ cd s
   $ git rev-parse HEAD
   aa84837ccfbdfedcdcdeeedc309d73e6eb069edc
   $ cd ..
 
 Sticky subrepository, file changes and revision updates
   $ touch s/f1
   $ cd s
   $ git add f1
   $ git rev-parse HEAD
   aa84837ccfbdfedcdcdeeedc309d73e6eb069edc
   $ cd ..
   $ hg id -n
   1+
   $ hg update 7
    subrepository sources for s differ
   use (l)ocal source (32a3438) or (r)emote source (32a3438)?
    l
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg id -n
   7
   $ cd s
   $ git rev-parse HEAD
   aa84837ccfbdfedcdcdeeedc309d73e6eb069edc
   $ cd ..
 
 Sticky repository, update --clean
   $ hg update --clean tip
   Previous HEAD position was aa84837... f
   HEAD is now at 32a3438... fff
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg id -n
   7
   $ cd s
   $ git rev-parse HEAD
   32a343883b74769118bb1d3b4b1fbf9156f4dddc
   $ cd ..
 
 Test subrepo already at intended revision:
   $ cd s
   $ git checkout 32a343883b74769118bb1d3b4b1fbf9156f4dddc
   HEAD is now at 32a3438... fff
   $ cd ..
   $ hg update 1
   Previous HEAD position was 32a3438... fff
   HEAD is now at da5f5b1... g
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg id -n
   1
   $ cd s
   $ git rev-parse HEAD
   da5f5b1d8ffcf62fb8327bcd3c89a4367a6018e7
   $ cd ..