synced with latest vcs

rhodecode/lib/vcs/__init__.py

@@ -1,43 +1,43 b''
 # -*- coding: utf-8 -*-
 """
     vcs
     ~~~
 
     Various version Control System (vcs) management abstraction layer for
     Python.
 
     :created_on: Apr 8, 2010
     :copyright: (c) 2010-2011 by Marcin Kuzminski, Lukasz Balcerzak.
 """
 
-VERSION = (0, 4, 0, 'dev')
+VERSION = (0, 5, 0, 'dev')
 
 __version__ = '.'.join((str(each) for each in VERSION[:4]))
 
 __all__ = [
     'get_version', 'get_repo', 'get_backend',
     'VCSError', 'RepositoryError', 'ChangesetError'
 ]
 
 import sys
 from rhodecode.lib.vcs.backends import get_repo, get_backend
 from rhodecode.lib.vcs.exceptions import VCSError, RepositoryError, ChangesetError
 
 
 def get_version():
     """
     Returns shorter version (digit parts only) as string.
     """
     return '.'.join((str(each) for each in VERSION[:3]))
 
 
 def main(argv=None):
     if argv is None:
         argv = sys.argv
     from rhodecode.lib.vcs.cli import ExecutionManager
     manager = ExecutionManager(argv)
     manager.execute()
     return 0
 
 if __name__ == '__main__':
     sys.exit(main(sys.argv))

rhodecode/lib/vcs/backends/git/changeset.py

@@ -1,552 +1,553 b''
 import re
 from itertools import chain
 from dulwich import objects
 from subprocess import Popen, PIPE
 
 from rhodecode.lib.vcs.conf import settings
 from rhodecode.lib.vcs.backends.base import BaseChangeset, EmptyChangeset
 from rhodecode.lib.vcs.exceptions import (
     RepositoryError, ChangesetError, NodeDoesNotExistError, VCSError,
     ChangesetDoesNotExistError, ImproperArchiveTypeError
 )
 from rhodecode.lib.vcs.nodes import (
     FileNode, DirNode, NodeKind, RootNode, RemovedFileNode, SubModuleNode,
     ChangedFileNodesGenerator, AddedFileNodesGenerator, RemovedFileNodesGenerator
 )
 from rhodecode.lib.vcs.utils import (
     safe_unicode, safe_str, safe_int, date_fromtimestamp
 )
 from rhodecode.lib.vcs.utils.lazy import LazyProperty
 
 
 class GitChangeset(BaseChangeset):
     """
     Represents state of the repository at single revision.
     """
 
     def __init__(self, repository, revision):
         self._stat_modes = {}
         self.repository = repository
 
         try:
-            commit = self.repository._repo.get_object(revision)
+            commit = self.repository._repo[revision]
             if isinstance(commit, objects.Tag):
                 revision = commit.object[1]
                 commit = self.repository._repo.get_object(commit.object[1])
         except KeyError:
             raise RepositoryError("Cannot get object with id %s" % revision)
         self.raw_id = revision
         self.id = self.raw_id
         self.short_id = self.raw_id[:12]
         self._commit = commit
-
         self._tree_id = commit.tree
         self._committer_property = 'committer'
         self._author_property = 'author'
         self._date_property = 'commit_time'
         self._date_tz_property = 'commit_timezone'
         self.revision = repository.revisions.index(revision)
 
-        self.message = safe_unicode(commit.message)
-
         self.nodes = {}
         self._paths = {}
 
     @LazyProperty
+    def message(self):
+        return safe_unicode(self._commit.message)
+
+    @LazyProperty
     def committer(self):
         return safe_unicode(getattr(self._commit, self._committer_property))
 
     @LazyProperty
     def author(self):
         return safe_unicode(getattr(self._commit, self._author_property))
 
     @LazyProperty
     def date(self):
         return date_fromtimestamp(getattr(self._commit, self._date_property),
                                   getattr(self._commit, self._date_tz_property))
 
     @LazyProperty
     def _timestamp(self):
         return getattr(self._commit, self._date_property)
 
     @LazyProperty
     def status(self):
         """
         Returns modified, added, removed, deleted files for current changeset
         """
         return self.changed, self.added, self.removed
 
     @LazyProperty
     def tags(self):
         _tags = []
         for tname, tsha in self.repository.tags.iteritems():
             if tsha == self.raw_id:
                 _tags.append(tname)
         return _tags
 
     @LazyProperty
     def branch(self):
 
         heads = self.repository._heads(reverse=False)
 
         ref = heads.get(self.raw_id)
         if ref:
             return safe_unicode(ref)
 
     def _fix_path(self, path):
         """
         Paths are stored without trailing slash so we need to get rid off it if
         needed.
         """
         if path.endswith('/'):
             path = path.rstrip('/')
         return path
 
     def _get_id_for_path(self, path):
 
         # FIXME: Please, spare a couple of minutes and make those codes cleaner;
         if not path in self._paths:
             path = path.strip('/')
             # set root tree
             tree = self.repository._repo[self._tree_id]
             if path == '':
                 self._paths[''] = tree.id
                 return tree.id
             splitted = path.split('/')
             dirs, name = splitted[:-1], splitted[-1]
             curdir = ''
 
             # initially extract things from root dir
             for item, stat, id in tree.iteritems():
                 if curdir:
                     name = '/'.join((curdir, item))
                 else:
                     name = item
                 self._paths[name] = id
                 self._stat_modes[name] = stat
 
             for dir in dirs:
                 if curdir:
                     curdir = '/'.join((curdir, dir))
                 else:
                     curdir = dir
                 dir_id = None
                 for item, stat, id in tree.iteritems():
                     if dir == item:
                         dir_id = id
                 if dir_id:
                     # Update tree
                     tree = self.repository._repo[dir_id]
                     if not isinstance(tree, objects.Tree):
                         raise ChangesetError('%s is not a directory' % curdir)
                 else:
                     raise ChangesetError('%s have not been found' % curdir)
 
                 # cache all items from the given traversed tree
                 for item, stat, id in tree.iteritems():
                     if curdir:
                         name = '/'.join((curdir, item))
                     else:
                         name = item
                     self._paths[name] = id
                     self._stat_modes[name] = stat
             if not path in self._paths:
                 raise NodeDoesNotExistError("There is no file nor directory "
                     "at the given path '%s' at revision %s"
                     % (path, self.short_id))
         return self._paths[path]
 
     def _get_kind(self, path):
         obj = self.repository._repo[self._get_id_for_path(path)]
         if isinstance(obj, objects.Blob):
             return NodeKind.FILE
         elif isinstance(obj, objects.Tree):
             return NodeKind.DIR
 
     def _get_filectx(self, path):
         path = self._fix_path(path)
         if self._get_kind(path) != NodeKind.FILE:
             raise ChangesetError("File does not exist for revision %s at "
                 " '%s'" % (self.raw_id, path))
         return path
 
     def _get_file_nodes(self):
         return chain(*(t[2] for t in self.walk()))
 
     @LazyProperty
     def parents(self):
         """
         Returns list of parents changesets.
         """
         return [self.repository.get_changeset(parent)
                 for parent in self._commit.parents]
 
     @LazyProperty
     def children(self):
         """
         Returns list of children changesets.
         """
         rev_filter = _git_path = settings.GIT_REV_FILTER
         so, se = self.repository.run_git_command(
             "rev-list %s --children | grep '^%s'" % (rev_filter, self.raw_id)
         )
 
         children = []
         for l in so.splitlines():
             childs = l.split(' ')[1:]
             children.extend(childs)
         return [self.repository.get_changeset(cs) for cs in children]
 
     def next(self, branch=None):
 
         if branch and self.branch != branch:
             raise VCSError('Branch option used on changeset not belonging '
                            'to that branch')
 
         def _next(changeset, branch):
             try:
                 next_ = changeset.revision + 1
                 next_rev = changeset.repository.revisions[next_]
             except IndexError:
                 raise ChangesetDoesNotExistError
             cs = changeset.repository.get_changeset(next_rev)
 
             if branch and branch != cs.branch:
                 return _next(cs, branch)
 
             return cs
 
         return _next(self, branch)
 
     def prev(self, branch=None):
         if branch and self.branch != branch:
             raise VCSError('Branch option used on changeset not belonging '
                            'to that branch')
 
         def _prev(changeset, branch):
             try:
                 prev_ = changeset.revision - 1
                 if prev_ < 0:
                     raise IndexError
                 prev_rev = changeset.repository.revisions[prev_]
             except IndexError:
                 raise ChangesetDoesNotExistError
 
             cs = changeset.repository.get_changeset(prev_rev)
 
             if branch and branch != cs.branch:
                 return _prev(cs, branch)
 
             return cs
 
         return _prev(self, branch)
 
     def diff(self, ignore_whitespace=True, context=3):
         rev1 = self.parents[0] if self.parents else self.repository.EMPTY_CHANGESET
         rev2 = self
         return ''.join(self.repository.get_diff(rev1, rev2,
                                     ignore_whitespace=ignore_whitespace,
                                     context=context))
 
     def get_file_mode(self, path):
         """
         Returns stat mode of the file at the given ``path``.
         """
         # ensure path is traversed
         self._get_id_for_path(path)
         return self._stat_modes[path]
 
     def get_file_content(self, path):
         """
         Returns content of the file at given ``path``.
         """
         id = self._get_id_for_path(path)
         blob = self.repository._repo[id]
         return blob.as_pretty_string()
 
     def get_file_size(self, path):
         """
         Returns size of the file at given ``path``.
         """
         id = self._get_id_for_path(path)
         blob = self.repository._repo[id]
         return blob.raw_length()
 
     def get_file_changeset(self, path):
         """
         Returns last commit of the file at the given ``path``.
         """
         return self.get_file_history(path, limit=1)[0]
 
     def get_file_history(self, path, limit=None):
         """
         Returns history of file as reversed list of ``Changeset`` objects for
         which file at given ``path`` has been modified.
 
         TODO: This function now uses os underlying 'git' and 'grep' commands
         which is generally not good. Should be replaced with algorithm
         iterating commits.
         """
         self._get_filectx(path)
         cs_id = safe_str(self.id)
         f_path = safe_str(path)
 
         if limit:
             cmd = 'log -n %s --pretty="format: %%H" -s -p %s -- "%s"' % (
                       safe_int(limit, 0), cs_id, f_path
                    )
 
         else:
             cmd = 'log --pretty="format: %%H" -s -p %s -- "%s"' % (
                       cs_id, f_path
                    )
         so, se = self.repository.run_git_command(cmd)
         ids = re.findall(r'[0-9a-fA-F]{40}', so)
         return [self.repository.get_changeset(id) for id in ids]
 
     def get_file_history_2(self, path):
         """
         Returns history of file as reversed list of ``Changeset`` objects for
         which file at given ``path`` has been modified.
 
         """
         self._get_filectx(path)
         from dulwich.walk import Walker
         include = [self.id]
         walker = Walker(self.repository._repo.object_store, include,
                         paths=[path], max_entries=1)
         return [self.repository.get_changeset(sha)
                 for sha in (x.commit.id for x in walker)]
 
     def get_file_annotate(self, path):
         """
         Returns a generator of four element tuples with
             lineno, sha, changeset lazy loader and line
 
         TODO: This function now uses os underlying 'git' command which is
         generally not good. Should be replaced with algorithm iterating
         commits.
         """
         cmd = 'blame -l --root -r %s -- "%s"' % (self.id, path)
         # -l     ==> outputs long shas (and we need all 40 characters)
         # --root ==> doesn't put '^' character for bounderies
         # -r sha ==> blames for the given revision
         so, se = self.repository.run_git_command(cmd)
 
         for i, blame_line in enumerate(so.split('\n')[:-1]):
             ln_no = i + 1
             sha, line = re.split(r' ', blame_line, 1)
             yield (ln_no, sha, lambda: self.repository.get_changeset(sha), line)
 
     def fill_archive(self, stream=None, kind='tgz', prefix=None,
                      subrepos=False):
         """
         Fills up given stream.
 
         :param stream: file like object.
         :param kind: one of following: ``zip``, ``tgz`` or ``tbz2``.
             Default: ``tgz``.
         :param prefix: name of root directory in archive.
             Default is repository name and changeset's raw_id joined with dash
             (``repo-tip.<KIND>``).
         :param subrepos: include subrepos in this archive.
 
         :raise ImproperArchiveTypeError: If given kind is wrong.
         :raise VcsError: If given stream is None
 
         """
         allowed_kinds = settings.ARCHIVE_SPECS.keys()
         if kind not in allowed_kinds:
             raise ImproperArchiveTypeError('Archive kind not supported use one'
                 'of %s', allowed_kinds)
 
         if prefix is None:
             prefix = '%s-%s' % (self.repository.name, self.short_id)
         elif prefix.startswith('/'):
             raise VCSError("Prefix cannot start with leading slash")
         elif prefix.strip() == '':
             raise VCSError("Prefix cannot be empty")
 
         if kind == 'zip':
             frmt = 'zip'
         else:
             frmt = 'tar'
         _git_path = settings.GIT_EXECUTABLE_PATH
         cmd = '%s archive --format=%s --prefix=%s/ %s' % (_git_path,
                                                 frmt, prefix, self.raw_id)
         if kind == 'tgz':
             cmd += ' | gzip -9'
         elif kind == 'tbz2':
             cmd += ' | bzip2 -9'
 
         if stream is None:
             raise VCSError('You need to pass in a valid stream for filling'
                            ' with archival data')
         popen = Popen(cmd, stdout=PIPE, stderr=PIPE, shell=True,
             cwd=self.repository.path)
 
         buffer_size = 1024 * 8
         chunk = popen.stdout.read(buffer_size)
         while chunk:
             stream.write(chunk)
             chunk = popen.stdout.read(buffer_size)
         # Make sure all descriptors would be read
         popen.communicate()
 
     def get_nodes(self, path):
         if self._get_kind(path) != NodeKind.DIR:
             raise ChangesetError("Directory does not exist for revision %s at "
                 " '%s'" % (self.revision, path))
         path = self._fix_path(path)
         id = self._get_id_for_path(path)
         tree = self.repository._repo[id]
         dirnodes = []
         filenodes = []
         als = self.repository.alias
         for name, stat, id in tree.iteritems():
             if objects.S_ISGITLINK(stat):
                 dirnodes.append(SubModuleNode(name, url=None, changeset=id,
                                               alias=als))
                 continue
 
             obj = self.repository._repo.get_object(id)
             if path != '':
                 obj_path = '/'.join((path, name))
             else:
                 obj_path = name
             if obj_path not in self._stat_modes:
                 self._stat_modes[obj_path] = stat
             if isinstance(obj, objects.Tree):
                 dirnodes.append(DirNode(obj_path, changeset=self))
             elif isinstance(obj, objects.Blob):
                 filenodes.append(FileNode(obj_path, changeset=self, mode=stat))
             else:
                 raise ChangesetError("Requested object should be Tree "
                                      "or Blob, is %r" % type(obj))
         nodes = dirnodes + filenodes
         for node in nodes:
             if not node.path in self.nodes:
                 self.nodes[node.path] = node
         nodes.sort()
         return nodes
 
     def get_node(self, path):
         if isinstance(path, unicode):
             path = path.encode('utf-8')
         path = self._fix_path(path)
         if not path in self.nodes:
             try:
                 id_ = self._get_id_for_path(path)
             except ChangesetError:
                 raise NodeDoesNotExistError("Cannot find one of parents' "
                     "directories for a given path: %s" % path)
 
             _GL = lambda m: m and objects.S_ISGITLINK(m)
             if _GL(self._stat_modes.get(path)):
                 node = SubModuleNode(path, url=None, changeset=id_,
                                      alias=self.repository.alias)
             else:
                 obj = self.repository._repo.get_object(id_)
 
                 if isinstance(obj, objects.Tree):
                     if path == '':
                         node = RootNode(changeset=self)
                     else:
                         node = DirNode(path, changeset=self)
                     node._tree = obj
                 elif isinstance(obj, objects.Blob):
                     node = FileNode(path, changeset=self)
                     node._blob = obj
                 else:
                     raise NodeDoesNotExistError("There is no file nor directory "
                         "at the given path '%s' at revision %s"
                         % (path, self.short_id))
             # cache node
             self.nodes[path] = node
         return self.nodes[path]
 
     @LazyProperty
     def affected_files(self):
         """
         Get's a fast accessible file changes for given changeset
         """
         added, modified, deleted = self._changes_cache
         return list(added.union(modified).union(deleted))
 
     @LazyProperty
     def _diff_name_status(self):
         output = []
         for parent in self.parents:
             cmd = 'diff --name-status %s %s --encoding=utf8' % (parent.raw_id,
                                                                 self.raw_id)
             so, se = self.repository.run_git_command(cmd)
             output.append(so.strip())
         return '\n'.join(output)
 
     @LazyProperty
     def _changes_cache(self):
         added = set()
         modified = set()
         deleted = set()
         _r = self.repository._repo
 
         parents = self.parents
         if not self.parents:
             parents = [EmptyChangeset()]
         for parent in parents:
             if isinstance(parent, EmptyChangeset):
                 oid = None
             else:
                 oid = _r[parent.raw_id].tree
             changes = _r.object_store.tree_changes(oid, _r[self.raw_id].tree)
             for (oldpath, newpath), (_, _), (_, _) in changes:
                 if newpath and oldpath:
                     modified.add(newpath)
                 elif newpath and not oldpath:
                     added.add(newpath)
                 elif not newpath and oldpath:
                     deleted.add(oldpath)
         return added, modified, deleted
 
     def _get_paths_for_status(self, status):
         """
         Returns sorted list of paths for given ``status``.
 
         :param status: one of: *added*, *modified* or *deleted*
         """
         added, modified, deleted = self._changes_cache
         return sorted({
             'added': list(added),
             'modified': list(modified),
             'deleted': list(deleted)}[status]
         )
 
     @LazyProperty
     def added(self):
         """
         Returns list of added ``FileNode`` objects.
         """
         if not self.parents:
             return list(self._get_file_nodes())
         return AddedFileNodesGenerator([n for n in
                                 self._get_paths_for_status('added')], self)
 
     @LazyProperty
     def changed(self):
         """
         Returns list of modified ``FileNode`` objects.
         """
         if not self.parents:
             return []
         return ChangedFileNodesGenerator([n for n in
                                 self._get_paths_for_status('modified')], self)
 
     @LazyProperty
     def removed(self):
         """
         Returns list of removed ``FileNode`` objects.
         """
         if not self.parents:
             return []
         return RemovedFileNodesGenerator([n for n in
                                 self._get_paths_for_status('deleted')], self)

rhodecode/lib/vcs/backends/git/repository.py

@@ -1,688 +1,694 b''
 # -*- coding: utf-8 -*-
 """
     vcs.backends.git.repository
     ~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
     Git repository implementation.
 
     :created_on: Apr 8, 2010
     :copyright: (c) 2010-2011 by Marcin Kuzminski, Lukasz Balcerzak.
 """
 
 import os
 import re
 import time
 import urllib
 import urllib2
 import logging
 import posixpath
 import string
 
 from dulwich.objects import Tag
 from dulwich.repo import Repo, NotGitRepository
 
 from rhodecode.lib.vcs import subprocessio
 from rhodecode.lib.vcs.backends.base import BaseRepository, CollectionGenerator
 from rhodecode.lib.vcs.conf import settings
 
 from rhodecode.lib.vcs.exceptions import (
     BranchDoesNotExistError, ChangesetDoesNotExistError, EmptyRepositoryError,
     RepositoryError, TagAlreadyExistError, TagDoesNotExistError
 )
 from rhodecode.lib.vcs.utils import safe_unicode, makedate, date_fromtimestamp
 from rhodecode.lib.vcs.utils.lazy import LazyProperty
 from rhodecode.lib.vcs.utils.ordered_dict import OrderedDict
 from rhodecode.lib.vcs.utils.paths import abspath, get_user_home
 
 from rhodecode.lib.vcs.utils.hgcompat import (
     hg_url, httpbasicauthhandler, httpdigestauthhandler
 )
 
 from .changeset import GitChangeset
 from .config import ConfigFile
 from .inmemory import GitInMemoryChangeset
 from .workdir import GitWorkdir
 
+SHA_PATTERN = re.compile(r'^[[0-9a-fA-F]{12}|[0-9a-fA-F]{40}]$')
+
 log = logging.getLogger(__name__)
 
 
 class GitRepository(BaseRepository):
     """
     Git repository backend.
     """
     DEFAULT_BRANCH_NAME = 'master'
     scm = 'git'
 
     def __init__(self, repo_path, create=False, src_url=None,
                  update_after_clone=False, bare=False):
 
         self.path = abspath(repo_path)
         repo = self._get_repo(create, src_url, update_after_clone, bare)
         self.bare = repo.bare
 
-        self._config_files = [
-            bare and abspath(self.path, 'config')
-                     or abspath(self.path, '.git', 'config'),
-            abspath(get_user_home(), '.gitconfig'),
-        ]
+    @property
+    def _config_files(self):
+        return [
+            self.bare and abspath(self.path, 'config')
+                      or abspath(self.path, '.git', 'config'),
+             abspath(get_user_home(), '.gitconfig'),
+         ]
 
     @property
     def _repo(self):
         return Repo(self.path)
 
     @property
     def head(self):
         try:
             return self._repo.head()
         except KeyError:
             return None
 
     @LazyProperty
     def revisions(self):
         """
         Returns list of revisions' ids, in ascending order.  Being lazy
         attribute allows external tools to inject shas from cache.
         """
         return self._get_all_revisions()
 
     @classmethod
     def _run_git_command(cls, cmd, **opts):
         """
         Runs given ``cmd`` as git command and returns tuple
         (stdout, stderr).
 
         :param cmd: git command to be executed
         :param opts: env options to pass into Subprocess command
         """
 
         if '_bare' in opts:
             _copts = []
             del opts['_bare']
         else:
             _copts = ['-c', 'core.quotepath=false', ]
         safe_call = False
         if '_safe' in opts:
             #no exc on failure
             del opts['_safe']
             safe_call = True
 
         _str_cmd = False
         if isinstance(cmd, basestring):
             cmd = [cmd]
             _str_cmd = True
 
         gitenv = os.environ
         # need to clean fix GIT_DIR !
         if 'GIT_DIR' in gitenv:
             del gitenv['GIT_DIR']
         gitenv['GIT_CONFIG_NOGLOBAL'] = '1'
 
         _git_path = settings.GIT_EXECUTABLE_PATH
         cmd = [_git_path] + _copts + cmd
         if _str_cmd:
             cmd = ' '.join(cmd)
         try:
             _opts = dict(
                 env=gitenv,
                 shell=False,
             )
             _opts.update(opts)
             p = subprocessio.SubprocessIOChunker(cmd, **_opts)
         except (EnvironmentError, OSError), err:
             tb_err = ("Couldn't run git command (%s).\n"
                       "Original error was:%s\n" % (cmd, err))
             log.error(tb_err)
             if safe_call:
                 return '', err
             else:
                 raise RepositoryError(tb_err)
 
         return ''.join(p.output), ''.join(p.error)
 
     def run_git_command(self, cmd):
         opts = {}
         if os.path.isdir(self.path):
             opts['cwd'] = self.path
         return self._run_git_command(cmd, **opts)
 
     @classmethod
     def _check_url(cls, url):
         """
         Functon will check given url and try to verify if it's a valid
         link. Sometimes it may happened that mercurial will issue basic
         auth request that can cause whole API to hang when used from python
         or other external calls.
 
         On failures it'll raise urllib2.HTTPError
         """
 
         # check first if it's not an local url
         if os.path.isdir(url) or url.startswith('file:'):
             return True
 
         if('+' in url[:url.find('://')]):
             url = url[url.find('+') + 1:]
 
         handlers = []
         test_uri, authinfo = hg_url(url).authinfo()
         if not test_uri.endswith('info/refs'):
             test_uri = test_uri.rstrip('/') + '/info/refs'
         if authinfo:
             #create a password manager
             passmgr = urllib2.HTTPPasswordMgrWithDefaultRealm()
             passmgr.add_password(*authinfo)
 
             handlers.extend((httpbasicauthhandler(passmgr),
                              httpdigestauthhandler(passmgr)))
 
         o = urllib2.build_opener(*handlers)
         o.addheaders = [('User-Agent', 'git/1.7.8.0')]  # fake some git
 
         q = {"service": 'git-upload-pack'}
         qs = '?%s' % urllib.urlencode(q)
         cu = "%s%s" % (test_uri, qs)
         req = urllib2.Request(cu, None, {})
 
         try:
             resp = o.open(req)
             return resp.code == 200
         except Exception, e:
             # means it cannot be cloned
             raise urllib2.URLError("[%s] %s" % (url, e))
 
     def _get_repo(self, create, src_url=None, update_after_clone=False,
                   bare=False):
         if create and os.path.exists(self.path):
             raise RepositoryError("Location already exist")
         if src_url and not create:
             raise RepositoryError("Create should be set to True if src_url is "
                                   "given (clone operation creates repository)")
         try:
             if create and src_url:
                 GitRepository._check_url(src_url)
                 self.clone(src_url, update_after_clone, bare)
                 return Repo(self.path)
             elif create:
                 os.mkdir(self.path)
                 if bare:
                     return Repo.init_bare(self.path)
                 else:
                     return Repo.init(self.path)
             else:
                 return self._repo
         except (NotGitRepository, OSError), err:
             raise RepositoryError(err)
 
     def _get_all_revisions(self):
         # we must check if this repo is not empty, since later command
         # fails if it is. And it's cheaper to ask than throw the subprocess
         # errors
         try:
             self._repo.head()
         except KeyError:
             return []
 
         rev_filter = _git_path = settings.GIT_REV_FILTER
         cmd = 'rev-list %s --reverse --date-order' % (rev_filter)
         try:
             so, se = self.run_git_command(cmd)
         except RepositoryError:
             # Can be raised for empty repositories
             return []
         return so.splitlines()
 
     def _get_all_revisions2(self):
         #alternate implementation using dulwich
         includes = [x[1][0] for x in self._parsed_refs.iteritems()
                     if x[1][1] != 'T']
         return [c.commit.id for c in self._repo.get_walker(include=includes)]
 
     def _get_revision(self, revision):
         """
         For git backend we always return integer here. This way we ensure
         that changset's revision attribute would become integer.
         """
-        pattern = re.compile(r'^[[0-9a-fA-F]{12}|[0-9a-fA-F]{40}]$')
-        is_bstr = lambda o: isinstance(o, (str, unicode))
+
         is_null = lambda o: len(o) == revision.count('0')
 
-        if len(self.revisions) == 0:
+        try:
+            self.revisions[0]
+        except (KeyError, IndexError):
             raise EmptyRepositoryError("There are no changesets yet")
 
         if revision in (None, '', 'tip', 'HEAD', 'head', -1):
-            revision = self.revisions[-1]
+            return self.revisions[-1]
 
-        if ((is_bstr(revision) and revision.isdigit() and len(revision) < 12)
+        is_bstr = isinstance(revision, (str, unicode))
+        if ((is_bstr and revision.isdigit() and len(revision) < 12)
             or isinstance(revision, int) or is_null(revision)):
             try:
                 revision = self.revisions[int(revision)]
             except Exception:
                 raise ChangesetDoesNotExistError("Revision %s does not exist "
                     "for this repository" % (revision))
 
-        elif is_bstr(revision):
+        elif is_bstr:
             # get by branch/tag name
             _ref_revision = self._parsed_refs.get(revision)
-            _tags_shas = self.tags.values()
             if _ref_revision:  # and _ref_revision[1] in ['H', 'RH', 'T']:
                 return _ref_revision[0]
 
+            _tags_shas = self.tags.values()
             # maybe it's a tag ? we don't have them in self.revisions
-            elif revision in _tags_shas:
+            if revision in _tags_shas:
                 return _tags_shas[_tags_shas.index(revision)]
 
-            elif not pattern.match(revision) or revision not in self.revisions:
+            elif not SHA_PATTERN.match(revision) or revision not in self.revisions:
                 raise ChangesetDoesNotExistError("Revision %s does not exist "
                     "for this repository" % (revision))
 
         # Ensure we return full id
-        if not pattern.match(str(revision)):
+        if not SHA_PATTERN.match(str(revision)):
             raise ChangesetDoesNotExistError("Given revision %s not recognized"
                 % revision)
         return revision
 
     def _get_archives(self, archive_name='tip'):
 
         for i in [('zip', '.zip'), ('gz', '.tar.gz'), ('bz2', '.tar.bz2')]:
                 yield {"type": i[0], "extension": i[1], "node": archive_name}
 
     def _get_url(self, url):
         """
         Returns normalized url. If schema is not given, would fall to
         filesystem (``file:///``) schema.
         """
         url = str(url)
         if url != 'default' and not '://' in url:
             url = ':///'.join(('file', url))
         return url
 
     def get_hook_location(self):
         """
         returns absolute path to location where hooks are stored
         """
         loc = os.path.join(self.path, 'hooks')
         if not self.bare:
             loc = os.path.join(self.path, '.git', 'hooks')
         return loc
 
     @LazyProperty
     def name(self):
         return os.path.basename(self.path)
 
     @LazyProperty
     def last_change(self):
         """
         Returns last change made on this repository as datetime object
         """
         return date_fromtimestamp(self._get_mtime(), makedate()[1])
 
     def _get_mtime(self):
         try:
             return time.mktime(self.get_changeset().date.timetuple())
         except RepositoryError:
             idx_loc = '' if self.bare else '.git'
             # fallback to filesystem
             in_path = os.path.join(self.path, idx_loc, "index")
             he_path = os.path.join(self.path, idx_loc, "HEAD")
             if os.path.exists(in_path):
                 return os.stat(in_path).st_mtime
             else:
                 return os.stat(he_path).st_mtime
 
     @LazyProperty
     def description(self):
         idx_loc = '' if self.bare else '.git'
         undefined_description = u'unknown'
         description_path = os.path.join(self.path, idx_loc, 'description')
         if os.path.isfile(description_path):
             return safe_unicode(open(description_path).read())
         else:
             return undefined_description
 
     @LazyProperty
     def contact(self):
         undefined_contact = u'Unknown'
         return undefined_contact
 
     @property
     def branches(self):
         if not self.revisions:
             return {}
         sortkey = lambda ctx: ctx[0]
         _branches = [(x[0], x[1][0])
                      for x in self._parsed_refs.iteritems() if x[1][1] == 'H']
         return OrderedDict(sorted(_branches, key=sortkey, reverse=False))
 
     @LazyProperty
     def tags(self):
         return self._get_tags()
 
     def _get_tags(self):
         if not self.revisions:
             return {}
 
         sortkey = lambda ctx: ctx[0]
         _tags = [(x[0], x[1][0])
                  for x in self._parsed_refs.iteritems() if x[1][1] == 'T']
         return OrderedDict(sorted(_tags, key=sortkey, reverse=True))
 
     def tag(self, name, user, revision=None, message=None, date=None,
             **kwargs):
         """
         Creates and returns a tag for the given ``revision``.
 
         :param name: name for new tag
         :param user: full username, i.e.: "Joe Doe <[e-mail]>"
         :param revision: changeset id for which new tag would be created
         :param message: message of the tag's commit
         :param date: date of tag's commit
 
         :raises TagAlreadyExistError: if tag with same name already exists
         """
         if name in self.tags:
             raise TagAlreadyExistError("Tag %s already exists" % name)
         changeset = self.get_changeset(revision)
         message = message or "Added tag %s for commit %s" % (name,
             changeset.raw_id)
         self._repo.refs["refs/tags/%s" % name] = changeset._commit.id
 
         self._parsed_refs = self._get_parsed_refs()
         self.tags = self._get_tags()
         return changeset
 
     def remove_tag(self, name, user, message=None, date=None):
         """
         Removes tag with the given ``name``.
 
         :param name: name of the tag to be removed
         :param user: full username, i.e.: "Joe Doe <[e-mail]>"
         :param message: message of the tag's removal commit
         :param date: date of tag's removal commit
 
         :raises TagDoesNotExistError: if tag with given name does not exists
         """
         if name not in self.tags:
             raise TagDoesNotExistError("Tag %s does not exist" % name)
         tagpath = posixpath.join(self._repo.refs.path, 'refs', 'tags', name)
         try:
             os.remove(tagpath)
             self._parsed_refs = self._get_parsed_refs()
             self.tags = self._get_tags()
         except OSError, e:
             raise RepositoryError(e.strerror)
 
     @LazyProperty
     def _parsed_refs(self):
         return self._get_parsed_refs()
 
     def _get_parsed_refs(self):
         # cache the property
         _repo = self._repo
         refs = _repo.get_refs()
         keys = [('refs/heads/', 'H'),
                 ('refs/remotes/origin/', 'RH'),
                 ('refs/tags/', 'T')]
         _refs = {}
         for ref, sha in refs.iteritems():
             for k, type_ in keys:
                 if ref.startswith(k):
                     _key = ref[len(k):]
                     if type_ == 'T':
                         obj = _repo.get_object(sha)
                         if isinstance(obj, Tag):
                             sha = _repo.get_object(sha).object[1]
                     _refs[_key] = [sha, type_]
                     break
         return _refs
 
     def _heads(self, reverse=False):
         refs = self._repo.get_refs()
         heads = {}
 
         for key, val in refs.items():
             for ref_key in ['refs/heads/', 'refs/remotes/origin/']:
                 if key.startswith(ref_key):
                     n = key[len(ref_key):]
                     if n not in ['HEAD']:
                         heads[n] = val
 
         return heads if reverse else dict((y, x) for x, y in heads.iteritems())
 
     def get_changeset(self, revision=None):
         """
         Returns ``GitChangeset`` object representing commit from git repository
         at the given revision or head (most recent commit) if None given.
         """
         if isinstance(revision, GitChangeset):
             return revision
         revision = self._get_revision(revision)
         changeset = GitChangeset(repository=self, revision=revision)
         return changeset
 
     def get_changesets(self, start=None, end=None, start_date=None,
            end_date=None, branch_name=None, reverse=False):
         """
         Returns iterator of ``GitChangeset`` objects from start to end (both
         are inclusive), in ascending date order (unless ``reverse`` is set).
 
         :param start: changeset ID, as str; first returned changeset
         :param end: changeset ID, as str; last returned changeset
         :param start_date: if specified, changesets with commit date less than
           ``start_date`` would be filtered out from returned set
         :param end_date: if specified, changesets with commit date greater than
           ``end_date`` would be filtered out from returned set
         :param branch_name: if specified, changesets not reachable from given
           branch would be filtered out from returned set
         :param reverse: if ``True``, returned generator would be reversed
           (meaning that returned changesets would have descending date order)
 
         :raise BranchDoesNotExistError: If given ``branch_name`` does not
             exist.
         :raise ChangesetDoesNotExistError: If changeset for given ``start`` or
           ``end`` could not be found.
 
         """
         if branch_name and branch_name not in self.branches:
             raise BranchDoesNotExistError("Branch '%s' not found" \
                                           % branch_name)
         # %H at format means (full) commit hash, initial hashes are retrieved
         # in ascending date order
         cmd_template = 'log --date-order --reverse --pretty=format:"%H"'
         cmd_params = {}
         if start_date:
             cmd_template += ' --since "$since"'
             cmd_params['since'] = start_date.strftime('%m/%d/%y %H:%M:%S')
         if end_date:
             cmd_template += ' --until "$until"'
             cmd_params['until'] = end_date.strftime('%m/%d/%y %H:%M:%S')
         if branch_name:
             cmd_template += ' $branch_name'
             cmd_params['branch_name'] = branch_name
         else:
             rev_filter = _git_path = settings.GIT_REV_FILTER
             cmd_template += ' %s' % (rev_filter)
 
         cmd = string.Template(cmd_template).safe_substitute(**cmd_params)
         revs = self.run_git_command(cmd)[0].splitlines()
         start_pos = 0
         end_pos = len(revs)
         if start:
             _start = self._get_revision(start)
             try:
                 start_pos = revs.index(_start)
             except ValueError:
                 pass
 
         if end is not None:
             _end = self._get_revision(end)
             try:
                 end_pos = revs.index(_end)
             except ValueError:
                 pass
 
         if None not in [start, end] and start_pos > end_pos:
             raise RepositoryError('start cannot be after end')
 
         if end_pos is not None:
             end_pos += 1
 
         revs = revs[start_pos:end_pos]
         if reverse:
             revs = reversed(revs)
         return CollectionGenerator(self, revs)
 
     def get_diff(self, rev1, rev2, path=None, ignore_whitespace=False,
                  context=3):
         """
         Returns (git like) *diff*, as plain text. Shows changes introduced by
         ``rev2`` since ``rev1``.
 
         :param rev1: Entry point from which diff is shown. Can be
           ``self.EMPTY_CHANGESET`` - in this case, patch showing all
           the changes since empty state of the repository until ``rev2``
         :param rev2: Until which revision changes should be shown.
         :param ignore_whitespace: If set to ``True``, would not show whitespace
           changes. Defaults to ``False``.
         :param context: How many lines before/after changed lines should be
           shown. Defaults to ``3``.
         """
         flags = ['-U%s' % context, '--full-index', '--binary', '-p', '-M', '--abbrev=40']
         if ignore_whitespace:
             flags.append('-w')
 
         if hasattr(rev1, 'raw_id'):
             rev1 = getattr(rev1, 'raw_id')
 
         if hasattr(rev2, 'raw_id'):
             rev2 = getattr(rev2, 'raw_id')
 
         if rev1 == self.EMPTY_CHANGESET:
             rev2 = self.get_changeset(rev2).raw_id
             cmd = ' '.join(['show'] + flags + [rev2])
         else:
             rev1 = self.get_changeset(rev1).raw_id
             rev2 = self.get_changeset(rev2).raw_id
             cmd = ' '.join(['diff'] + flags + [rev1, rev2])
 
         if path:
             cmd += ' -- "%s"' % path
 
         stdout, stderr = self.run_git_command(cmd)
         # If we used 'show' command, strip first few lines (until actual diff
         # starts)
         if rev1 == self.EMPTY_CHANGESET:
             lines = stdout.splitlines()
             x = 0
             for line in lines:
                 if line.startswith('diff'):
                     break
                 x += 1
             # Append new line just like 'diff' command do
             stdout = '\n'.join(lines[x:]) + '\n'
         return stdout
 
     @LazyProperty
     def in_memory_changeset(self):
         """
         Returns ``GitInMemoryChangeset`` object for this repository.
         """
         return GitInMemoryChangeset(self)
 
     def clone(self, url, update_after_clone=True, bare=False):
         """
         Tries to clone changes from external location.
 
         :param update_after_clone: If set to ``False``, git won't checkout
           working directory
         :param bare: If set to ``True``, repository would be cloned into
           *bare* git repository (no working directory at all).
         """
         url = self._get_url(url)
         cmd = ['clone']
         if bare:
             cmd.append('--bare')
         elif not update_after_clone:
             cmd.append('--no-checkout')
         cmd += ['--', '"%s"' % url, '"%s"' % self.path]
         cmd = ' '.join(cmd)
         # If error occurs run_git_command raises RepositoryError already
         self.run_git_command(cmd)
 
     def pull(self, url):
         """
         Tries to pull changes from external location.
         """
         url = self._get_url(url)
         cmd = ['pull']
         cmd.append("--ff-only")
         cmd.append(url)
         cmd = ' '.join(cmd)
         # If error occurs run_git_command raises RepositoryError already
         self.run_git_command(cmd)
 
     def fetch(self, url):
         """
         Tries to pull changes from external location.
         """
         url = self._get_url(url)
         so, se = self.run_git_command('ls-remote -h %s' % url)
         refs = []
         for line in (x for x in so.splitlines()):
             sha, ref = line.split('\t')
             refs.append(ref)
         refs = ' '.join(('+%s:%s' % (r, r) for r in refs))
         cmd = '''fetch %s -- %s''' % (url, refs)
         self.run_git_command(cmd)
 
     @LazyProperty
     def workdir(self):
         """
         Returns ``Workdir`` instance for this repository.
         """
         return GitWorkdir(self)
 
     def get_config_value(self, section, name, config_file=None):
         """
         Returns configuration value for a given [``section``] and ``name``.
 
         :param section: Section we want to retrieve value from
         :param name: Name of configuration we want to retrieve
         :param config_file: A path to file which should be used to retrieve
           configuration from (might also be a list of file paths)
         """
         if config_file is None:
             config_file = []
         elif isinstance(config_file, basestring):
             config_file = [config_file]
 
         def gen_configs():
             for path in config_file + self._config_files:
                 try:
                     yield ConfigFile.from_path(path)
                 except (IOError, OSError, ValueError):
                     continue
 
         for config in gen_configs():
             try:
                 return config.get(section, name)
             except KeyError:
                 continue
         return None
 
     def get_user_name(self, config_file=None):
         """
         Returns user's name from global configuration file.
 
         :param config_file: A path to file which should be used to retrieve
           configuration from (might also be a list of file paths)
         """
         return self.get_config_value('user', 'name', config_file)
 
     def get_user_email(self, config_file=None):
         """
         Returns user's email from global configuration file.
 
         :param config_file: A path to file which should be used to retrieve
           configuration from (might also be a list of file paths)
         """
         return self.get_config_value('user', 'email', config_file)

rhodecode/lib/vcs/backends/hg/repository.py

@@ -1,569 +1,571 b''
 # -*- coding: utf-8 -*-
 """
     vcs.backends.hg.repository
     ~~~~~~~~~~~~~~~~~~~~~~~~~~
 
     Mercurial repository implementation.
 
     :created_on: Apr 8, 2010
     :copyright: (c) 2010-2011 by Marcin Kuzminski, Lukasz Balcerzak.
 """
 
 import os
 import time
 import urllib
 import urllib2
 import logging
 import datetime
 
 
 from rhodecode.lib.vcs.backends.base import BaseRepository, CollectionGenerator
 from rhodecode.lib.vcs.conf import settings
 
 from rhodecode.lib.vcs.exceptions import (
     BranchDoesNotExistError, ChangesetDoesNotExistError, EmptyRepositoryError,
     RepositoryError, VCSError, TagAlreadyExistError, TagDoesNotExistError
 )
 from rhodecode.lib.vcs.utils import (
     author_email, author_name, date_fromtimestamp, makedate, safe_unicode
 )
 from rhodecode.lib.vcs.utils.lazy import LazyProperty
 from rhodecode.lib.vcs.utils.ordered_dict import OrderedDict
 from rhodecode.lib.vcs.utils.paths import abspath
 from rhodecode.lib.vcs.utils.hgcompat import (
     ui, nullid, match, patch, diffopts, clone, get_contact, pull,
     localrepository, RepoLookupError, Abort, RepoError, hex, scmutil, hg_url,
     httpbasicauthhandler, httpdigestauthhandler
 )
 
 from .changeset import MercurialChangeset
 from .inmemory import MercurialInMemoryChangeset
 from .workdir import MercurialWorkdir
 
 log = logging.getLogger(__name__)
 
 
 class MercurialRepository(BaseRepository):
     """
     Mercurial repository backend
     """
     DEFAULT_BRANCH_NAME = 'default'
     scm = 'hg'
 
     def __init__(self, repo_path, create=False, baseui=None, src_url=None,
                  update_after_clone=False):
         """
         Raises RepositoryError if repository could not be find at the given
         ``repo_path``.
 
         :param repo_path: local path of the repository
         :param create=False: if set to True, would try to create repository if
            it does not exist rather than raising exception
         :param baseui=None: user data
         :param src_url=None: would try to clone repository from given location
         :param update_after_clone=False: sets update of working copy after
           making a clone
         """
 
         if not isinstance(repo_path, str):
             raise VCSError('Mercurial backend requires repository path to '
                            'be instance of <str> got %s instead' %
                            type(repo_path))
 
         self.path = abspath(repo_path)
         self.baseui = baseui or ui.ui()
         # We've set path and ui, now we can set _repo itself
         self._repo = self._get_repo(create, src_url, update_after_clone)
 
     @property
     def _empty(self):
         """
         Checks if repository is empty without any changesets
         """
         # TODO: Following raises errors when using InMemoryChangeset...
         # return len(self._repo.changelog) == 0
         return len(self.revisions) == 0
 
     @LazyProperty
     def revisions(self):
         """
         Returns list of revisions' ids, in ascending order.  Being lazy
         attribute allows external tools to inject shas from cache.
         """
         return self._get_all_revisions()
 
     @LazyProperty
     def name(self):
         return os.path.basename(self.path)
 
     @LazyProperty
     def branches(self):
         return self._get_branches()
 
     @LazyProperty
     def allbranches(self):
         """
         List all branches, including closed branches.
         """
         return self._get_branches(closed=True)
 
     def _get_branches(self, closed=False):
         """
         Get's branches for this repository
         Returns only not closed branches by default
 
         :param closed: return also closed branches for mercurial
         """
 
         if self._empty:
             return {}
 
         def _branchtags(localrepo):
             """
             Patched version of mercurial branchtags to not return the closed
             branches
 
             :param localrepo: locarepository instance
             """
 
             bt = {}
             bt_closed = {}
             for bn, heads in localrepo.branchmap().iteritems():
                 tip = heads[-1]
                 if 'close' in localrepo.changelog.read(tip)[5]:
                     bt_closed[bn] = tip
                 else:
                     bt[bn] = tip
 
             if closed:
                 bt.update(bt_closed)
             return bt
 
         sortkey = lambda ctx: ctx[0]  # sort by name
         _branches = [(safe_unicode(n), hex(h),) for n, h in
                      _branchtags(self._repo).items()]
 
         return OrderedDict(sorted(_branches, key=sortkey, reverse=False))
 
     @LazyProperty
     def tags(self):
         """
         Get's tags for this repository
         """
         return self._get_tags()
 
     def _get_tags(self):
         if self._empty:
             return {}
 
         sortkey = lambda ctx: ctx[0]  # sort by name
         _tags = [(safe_unicode(n), hex(h),) for n, h in
                  self._repo.tags().items()]
 
         return OrderedDict(sorted(_tags, key=sortkey, reverse=True))
 
     def tag(self, name, user, revision=None, message=None, date=None,
             **kwargs):
         """
         Creates and returns a tag for the given ``revision``.
 
         :param name: name for new tag
         :param user: full username, i.e.: "Joe Doe <[e-mail]>"
         :param revision: changeset id for which new tag would be created
         :param message: message of the tag's commit
         :param date: date of tag's commit
 
         :raises TagAlreadyExistError: if tag with same name already exists
         """
         if name in self.tags:
             raise TagAlreadyExistError("Tag %s already exists" % name)
         changeset = self.get_changeset(revision)
         local = kwargs.setdefault('local', False)
 
         if message is None:
             message = "Added tag %s for changeset %s" % (name,
                 changeset.short_id)
 
         if date is None:
             date = datetime.datetime.now().ctime()
 
         try:
             self._repo.tag(name, changeset._ctx.node(), message, local, user,
                 date)
         except Abort, e:
             raise RepositoryError(e.message)
 
         # Reinitialize tags
         self.tags = self._get_tags()
         tag_id = self.tags[name]
 
         return self.get_changeset(revision=tag_id)
 
     def remove_tag(self, name, user, message=None, date=None):
         """
         Removes tag with the given ``name``.
 
         :param name: name of the tag to be removed
         :param user: full username, i.e.: "Joe Doe <[e-mail]>"
         :param message: message of the tag's removal commit
         :param date: date of tag's removal commit
 
         :raises TagDoesNotExistError: if tag with given name does not exists
         """
         if name not in self.tags:
             raise TagDoesNotExistError("Tag %s does not exist" % name)
         if message is None:
             message = "Removed tag %s" % name
         if date is None:
             date = datetime.datetime.now().ctime()
         local = False
 
         try:
             self._repo.tag(name, nullid, message, local, user, date)
             self.tags = self._get_tags()
         except Abort, e:
             raise RepositoryError(e.message)
 
     @LazyProperty
     def bookmarks(self):
         """
         Get's bookmarks for this repository
         """
         return self._get_bookmarks()
 
     def _get_bookmarks(self):
         if self._empty:
             return {}
 
         sortkey = lambda ctx: ctx[0]  # sort by name
         _bookmarks = [(safe_unicode(n), hex(h),) for n, h in
                  self._repo._bookmarks.items()]
         return OrderedDict(sorted(_bookmarks, key=sortkey, reverse=True))
 
     def _get_all_revisions(self):
 
         return map(lambda x: hex(x[7]), self._repo.changelog.index)[:-1]
 
     def get_diff(self, rev1, rev2, path='', ignore_whitespace=False,
                   context=3):
         """
         Returns (git like) *diff*, as plain text. Shows changes introduced by
         ``rev2`` since ``rev1``.
 
         :param rev1: Entry point from which diff is shown. Can be
           ``self.EMPTY_CHANGESET`` - in this case, patch showing all
           the changes since empty state of the repository until ``rev2``
         :param rev2: Until which revision changes should be shown.
         :param ignore_whitespace: If set to ``True``, would not show whitespace
           changes. Defaults to ``False``.
         :param context: How many lines before/after changed lines should be
           shown. Defaults to ``3``.
         """
         if hasattr(rev1, 'raw_id'):
             rev1 = getattr(rev1, 'raw_id')
 
         if hasattr(rev2, 'raw_id'):
             rev2 = getattr(rev2, 'raw_id')
 
         # Check if given revisions are present at repository (may raise
         # ChangesetDoesNotExistError)
         if rev1 != self.EMPTY_CHANGESET:
             self.get_changeset(rev1)
         self.get_changeset(rev2)
         if path:
             file_filter = match(self.path, '', [path])
         else:
             file_filter = None
 
         return ''.join(patch.diff(self._repo, rev1, rev2, match=file_filter,
                           opts=diffopts(git=True,
                                         ignorews=ignore_whitespace,
                                         context=context)))
 
     @classmethod
     def _check_url(cls, url):
         """
         Function will check given url and try to verify if it's a valid
         link. Sometimes it may happened that mercurial will issue basic
         auth request that can cause whole API to hang when used from python
         or other external calls.
 
         On failures it'll raise urllib2.HTTPError, return code 200 if url
         is valid or True if it's a local path
         """
 
         # check first if it's not an local url
         if os.path.isdir(url) or url.startswith('file:'):
             return True
 
         if('+' in url[:url.find('://')]):
             url = url[url.find('+') + 1:]
 
         handlers = []
         test_uri, authinfo = hg_url(url).authinfo()
 
         if authinfo:
             #create a password manager
             passmgr = urllib2.HTTPPasswordMgrWithDefaultRealm()
             passmgr.add_password(*authinfo)
 
             handlers.extend((httpbasicauthhandler(passmgr),
                              httpdigestauthhandler(passmgr)))
 
         o = urllib2.build_opener(*handlers)
         o.addheaders = [('Content-Type', 'application/mercurial-0.1'),
                         ('Accept', 'application/mercurial-0.1')]
 
         q = {"cmd": 'between'}
         q.update({'pairs': "%s-%s" % ('0' * 40, '0' * 40)})
         qs = '?%s' % urllib.urlencode(q)
         cu = "%s%s" % (test_uri, qs)
         req = urllib2.Request(cu, None, {})
 
         try:
             resp = o.open(req)
             return resp.code == 200
         except Exception, e:
             # means it cannot be cloned
             raise urllib2.URLError("[%s] %s" % (url, e))
 
     def _get_repo(self, create, src_url=None, update_after_clone=False):
         """
         Function will check for mercurial repository in given path and return
         a localrepo object. If there is no repository in that path it will
         raise an exception unless ``create`` parameter is set to True - in
         that case repository would be created and returned.
         If ``src_url`` is given, would try to clone repository from the
         location at given clone_point. Additionally it'll make update to
         working copy accordingly to ``update_after_clone`` flag
         """
 
         try:
             if src_url:
                 url = str(self._get_url(src_url))
                 opts = {}
                 if not update_after_clone:
                     opts.update({'noupdate': True})
                 try:
                     MercurialRepository._check_url(url)
                     clone(self.baseui, url, self.path, **opts)
 #                except urllib2.URLError:
 #                    raise Abort("Got HTTP 404 error")
                 except Exception:
                     raise
 
                 # Don't try to create if we've already cloned repo
                 create = False
             return localrepository(self.baseui, self.path, create=create)
         except (Abort, RepoError), err:
             if create:
                 msg = "Cannot create repository at %s. Original error was %s"\
                     % (self.path, err)
             else:
                 msg = "Not valid repository at %s. Original error was %s"\
                     % (self.path, err)
             raise RepositoryError(msg)
 
     @LazyProperty
     def in_memory_changeset(self):
         return MercurialInMemoryChangeset(self)
 
     @LazyProperty
     def description(self):
         undefined_description = u'unknown'
         return safe_unicode(self._repo.ui.config('web', 'description',
                                    undefined_description, untrusted=True))
 
     @LazyProperty
     def contact(self):
         undefined_contact = u'Unknown'
         return safe_unicode(get_contact(self._repo.ui.config)
                             or undefined_contact)
 
     @LazyProperty
     def last_change(self):
         """
         Returns last change made on this repository as datetime object
         """
         return date_fromtimestamp(self._get_mtime(), makedate()[1])
 
     def _get_mtime(self):
         try:
             return time.mktime(self.get_changeset().date.timetuple())
         except RepositoryError:
             #fallback to filesystem
             cl_path = os.path.join(self.path, '.hg', "00changelog.i")
             st_path = os.path.join(self.path, '.hg', "store")
             if os.path.exists(cl_path):
                 return os.stat(cl_path).st_mtime
             else:
                 return os.stat(st_path).st_mtime
 
     def _get_hidden(self):
         return self._repo.ui.configbool("web", "hidden", untrusted=True)
 
     def _get_revision(self, revision):
         """
         Get's an ID revision given as str. This will always return a fill
         40 char revision number
 
         :param revision: str or int or None
         """
 
         if self._empty:
             raise EmptyRepositoryError("There are no changesets yet")
 
         if revision in [-1, 'tip', None]:
             revision = 'tip'
 
         try:
             revision = hex(self._repo.lookup(revision))
         except (IndexError, ValueError, RepoLookupError, TypeError):
             raise ChangesetDoesNotExistError("Revision %s does not "
                                     "exist for this repository"
                                     % (revision))
         return revision
 
     def _get_archives(self, archive_name='tip'):
         allowed = self.baseui.configlist("web", "allow_archive",
                                          untrusted=True)
         for i in [('zip', '.zip'), ('gz', '.tar.gz'), ('bz2', '.tar.bz2')]:
             if i[0] in allowed or self._repo.ui.configbool("web",
                                                            "allow" + i[0],
                                                            untrusted=True):
                 yield {"type": i[0], "extension": i[1], "node": archive_name}
 
     def _get_url(self, url):
         """
         Returns normalized url. If schema is not given, would fall
         to filesystem
         (``file:///``) schema.
         """
         url = str(url)
         if url != 'default' and not '://' in url:
             url = "file:" + urllib.pathname2url(url)
         return url
 
     def get_hook_location(self):
         """
         returns absolute path to location where hooks are stored
         """
         return os.path.join(self.path, '.hg', '.hgrc')
 
     def get_changeset(self, revision=None):
         """
         Returns ``MercurialChangeset`` object representing repository's
         changeset at the given ``revision``.
         """
         revision = self._get_revision(revision)
         changeset = MercurialChangeset(repository=self, revision=revision)
         return changeset
 
     def get_changesets(self, start=None, end=None, start_date=None,
                        end_date=None, branch_name=None, reverse=False):
         """
         Returns iterator of ``MercurialChangeset`` objects from start to end
         (both are inclusive)
 
         :param start: None, str, int or mercurial lookup format
         :param end:  None, str, int or mercurial lookup format
         :param start_date:
         :param end_date:
         :param branch_name:
         :param reversed: return changesets in reversed order
         """
 
         start_raw_id = self._get_revision(start)
         start_pos = self.revisions.index(start_raw_id) if start else None
         end_raw_id = self._get_revision(end)
         end_pos = self.revisions.index(end_raw_id) if end else None
 
         if None not in [start, end] and start_pos > end_pos:
             raise RepositoryError("Start revision '%s' cannot be "
                                   "after end revision '%s'" % (start, end))
 
         if branch_name and branch_name not in self.allbranches.keys():
             raise BranchDoesNotExistError('Branch %s not found in'
                                   ' this repository' % branch_name)
         if end_pos is not None:
             end_pos += 1
         #filter branches
         filter_ = []
         if branch_name:
             filter_.append('branch("%s")' % (branch_name))
 
-        if start_date:
+        if start_date and not end_date:
             filter_.append('date(">%s")' % start_date)
-        if end_date:
+        if end_date and not start_date:
             filter_.append('date("<%s")' % end_date)
+        if start_date and end_date:
+            filter_.append('date(">%s") and date("<%s")' % (start_date, end_date))
         if filter_:
             revisions = scmutil.revrange(self._repo, filter_)
         else:
             revisions = self.revisions
 
         revs = revisions[start_pos:end_pos]
         if reverse:
             revs = reversed(revs)
 
         return CollectionGenerator(self, revs)
 
     def pull(self, url):
         """
         Tries to pull changes from external location.
         """
         url = self._get_url(url)
         try:
             pull(self.baseui, self._repo, url)
         except Abort, err:
             # Propagate error but with vcs's type
             raise RepositoryError(str(err))
 
     @LazyProperty
     def workdir(self):
         """
         Returns ``Workdir`` instance for this repository.
         """
         return MercurialWorkdir(self)
 
     def get_config_value(self, section, name=None, config_file=None):
         """
         Returns configuration value for a given [``section``] and ``name``.
 
         :param section: Section we want to retrieve value from
         :param name: Name of configuration we want to retrieve
         :param config_file: A path to file which should be used to retrieve
           configuration from (might also be a list of file paths)
         """
         if config_file is None:
             config_file = []
         elif isinstance(config_file, basestring):
             config_file = [config_file]
 
         config = self._repo.ui
         for path in config_file:
             config.readconfig(path)
         return config.config(section, name)
 
     def get_user_name(self, config_file=None):
         """
         Returns user's name from global configuration file.
 
         :param config_file: A path to file which should be used to retrieve
           configuration from (might also be a list of file paths)
         """
         username = self.get_config_value('ui', 'username')
         if username:
             return author_name(username)
         return None
 
     def get_user_email(self, config_file=None):
         """
         Returns user's email from global configuration file.
 
         :param config_file: A path to file which should be used to retrieve
           configuration from (might also be a list of file paths)
         """
         username = self.get_config_value('ui', 'username')
         if username:
             return author_email(username)
         return None

rhodecode/tests/vcs/test_changesets.py

@@ -1,352 +1,360 b''
 from __future__ import with_statement
 
 import datetime
 from rhodecode.lib import vcs
 from rhodecode.tests.vcs.base import BackendTestMixin
 from rhodecode.tests.vcs.conf import SCM_TESTS
 
 from rhodecode.lib.vcs.backends.base import BaseChangeset
 from rhodecode.lib.vcs.nodes import (
     FileNode, AddedFileNodesGenerator,
     ChangedFileNodesGenerator, RemovedFileNodesGenerator
 )
 from rhodecode.lib.vcs.exceptions import (
     BranchDoesNotExistError, ChangesetDoesNotExistError,
     RepositoryError
 )
 from rhodecode.lib.vcs.utils.compat import unittest
 
 
 class TestBaseChangeset(unittest.TestCase):
 
     def test_as_dict(self):
         changeset = BaseChangeset()
         changeset.id = 'ID'
         changeset.raw_id = 'RAW_ID'
         changeset.short_id = 'SHORT_ID'
         changeset.revision = 1009
         changeset.date = datetime.datetime(2011, 1, 30, 1, 45)
         changeset.message = 'Message of a commit'
         changeset.author = 'Joe Doe <[e-mail]>'
         changeset.added = [FileNode('foo/bar/baz'), FileNode('foobar')]
         changeset.changed = []
         changeset.removed = []
         self.assertEqual(changeset.as_dict(), {
             'id': 'ID',
             'raw_id': 'RAW_ID',
             'short_id': 'SHORT_ID',
             'revision': 1009,
             'date': datetime.datetime(2011, 1, 30, 1, 45),
             'message': 'Message of a commit',
             'author': {
                 'name': 'Joe Doe',
                 'email': '[e-mail]',
             },
             'added': ['foo/bar/baz', 'foobar'],
             'changed': [],
             'removed': [],
         })
 
 class ChangesetsWithCommitsTestCaseixin(BackendTestMixin):
     recreate_repo_per_test = True
 
     @classmethod
     def _get_commits(cls):
         start_date = datetime.datetime(2010, 1, 1, 20)
         for x in xrange(5):
             yield {
                 'message': 'Commit %d' % x,
                 'author': 'Joe Doe <[e-mail]>',
                 'date': start_date + datetime.timedelta(hours=12 * x),
                 'added': [
                     FileNode('file_%d.txt' % x, content='Foobar %d' % x),
                 ],
             }
 
     def test_new_branch(self):
         self.imc.add(vcs.nodes.FileNode('docs/index.txt',
             content='Documentation\n'))
         foobar_tip = self.imc.commit(
             message=u'New branch: foobar',
             author=u'joe',
             branch='foobar',
         )
         self.assertTrue('foobar' in self.repo.branches)
         self.assertEqual(foobar_tip.branch, 'foobar')
         # 'foobar' should be the only branch that contains the new commit
         self.assertNotEqual(*self.repo.branches.values())
 
     def test_new_head_in_default_branch(self):
         tip = self.repo.get_changeset()
         self.imc.add(vcs.nodes.FileNode('docs/index.txt',
             content='Documentation\n'))
         foobar_tip = self.imc.commit(
             message=u'New branch: foobar',
             author=u'joe',
             branch='foobar',
             parents=[tip],
         )
         self.imc.change(vcs.nodes.FileNode('docs/index.txt',
             content='Documentation\nand more...\n'))
         newtip = self.imc.commit(
             message=u'At default branch',
             author=u'joe',
             branch=foobar_tip.branch,
             parents=[foobar_tip],
         )
 
         newest_tip = self.imc.commit(
             message=u'Merged with %s' % foobar_tip.raw_id,
             author=u'joe',
             branch=self.backend_class.DEFAULT_BRANCH_NAME,
             parents=[newtip, foobar_tip],
         )
 
         self.assertEqual(newest_tip.branch,
             self.backend_class.DEFAULT_BRANCH_NAME)
 
     def test_get_changesets_respects_branch_name(self):
         tip = self.repo.get_changeset()
         self.imc.add(vcs.nodes.FileNode('docs/index.txt',
             content='Documentation\n'))
         doc_changeset = self.imc.commit(
             message=u'New branch: docs',
             author=u'joe',
             branch='docs',
         )
         self.imc.add(vcs.nodes.FileNode('newfile', content=''))
         self.imc.commit(
             message=u'Back in default branch',
             author=u'joe',
             parents=[tip],
         )
         default_branch_changesets = self.repo.get_changesets(
             branch_name=self.repo.DEFAULT_BRANCH_NAME)
         self.assertNotIn(doc_changeset, default_branch_changesets)
 
     def test_get_changeset_by_branch(self):
         for branch, sha in self.repo.branches.iteritems():
             self.assertEqual(sha, self.repo.get_changeset(branch).raw_id)
 
     def test_get_changeset_by_tag(self):
         for tag, sha in self.repo.tags.iteritems():
             self.assertEqual(sha, self.repo.get_changeset(tag).raw_id)
 
 
 class ChangesetsTestCaseMixin(BackendTestMixin):
     recreate_repo_per_test = False
 
     @classmethod
     def _get_commits(cls):
         start_date = datetime.datetime(2010, 1, 1, 20)
         for x in xrange(5):
             yield {
                 'message': u'Commit %d' % x,
                 'author': u'Joe Doe <[e-mail]>',
                 'date': start_date + datetime.timedelta(hours=12 * x),
                 'added': [
                     FileNode('file_%d.txt' % x, content='Foobar %d' % x),
                 ],
             }
 
     def test_simple(self):
         tip = self.repo.get_changeset()
         self.assertEqual(tip.date, datetime.datetime(2010, 1, 3, 20))
 
     def test_get_changesets_is_ordered_by_date(self):
         changesets = list(self.repo.get_changesets())
         ordered_by_date = sorted(changesets,
             key=lambda cs: cs.date)
         self.assertItemsEqual(changesets, ordered_by_date)
 
     def test_get_changesets_respects_start(self):
         second_id = self.repo.revisions[1]
         changesets = list(self.repo.get_changesets(start=second_id))
         self.assertEqual(len(changesets), 4)
 
     def test_get_changesets_numerical_id_respects_start(self):
         second_id = 1
         changesets = list(self.repo.get_changesets(start=second_id))
         self.assertEqual(len(changesets), 4)
 
     def test_get_changesets_includes_start_changeset(self):
         second_id = self.repo.revisions[1]
         changesets = list(self.repo.get_changesets(start=second_id))
         self.assertEqual(changesets[0].raw_id, second_id)
 
     def test_get_changesets_respects_end(self):
         second_id = self.repo.revisions[1]
         changesets = list(self.repo.get_changesets(end=second_id))
         self.assertEqual(changesets[-1].raw_id, second_id)
         self.assertEqual(len(changesets), 2)
 
     def test_get_changesets_numerical_id_respects_end(self):
         second_id = 1
         changesets = list(self.repo.get_changesets(end=second_id))
         self.assertEqual(changesets.index(changesets[-1]), second_id)
         self.assertEqual(len(changesets), 2)
 
     def test_get_changesets_respects_both_start_and_end(self):
         second_id = self.repo.revisions[1]
         third_id = self.repo.revisions[2]
         changesets = list(self.repo.get_changesets(start=second_id,
             end=third_id))
         self.assertEqual(len(changesets), 2)
 
     def test_get_changesets_numerical_id_respects_both_start_and_end(self):
         changesets = list(self.repo.get_changesets(start=2, end=3))
         self.assertEqual(len(changesets), 2)
 
     def test_get_changesets_includes_end_changeset(self):
         second_id = self.repo.revisions[1]
         changesets = list(self.repo.get_changesets(end=second_id))
         self.assertEqual(changesets[-1].raw_id, second_id)
 
     def test_get_changesets_respects_start_date(self):
         start_date = datetime.datetime(2010, 2, 1)
         for cs in self.repo.get_changesets(start_date=start_date):
             self.assertGreaterEqual(cs.date, start_date)
 
     def test_get_changesets_respects_end_date(self):
+        start_date = datetime.datetime(2010, 1, 1)
+        end_date = datetime.datetime(2010, 2, 1)
+        for cs in self.repo.get_changesets(start_date=start_date,
+                                           end_date=end_date):
+            self.assertGreaterEqual(cs.date, start_date)
+            self.assertLessEqual(cs.date, end_date)
+
+    def test_get_changesets_respects_start_date_and_end_date(self):
         end_date = datetime.datetime(2010, 2, 1)
         for cs in self.repo.get_changesets(end_date=end_date):
             self.assertLessEqual(cs.date, end_date)
 
     def test_get_changesets_respects_reverse(self):
         changesets_id_list = [cs.raw_id for cs in
             self.repo.get_changesets(reverse=True)]
         self.assertItemsEqual(changesets_id_list, reversed(self.repo.revisions))
 
     def test_get_filenodes_generator(self):
         tip = self.repo.get_changeset()
         filepaths = [node.path for node in tip.get_filenodes_generator()]
         self.assertItemsEqual(filepaths, ['file_%d.txt' % x for x in xrange(5)])
 
     def test_size(self):
         tip = self.repo.get_changeset()
         size = 5 * len('Foobar N') # Size of 5 files
         self.assertEqual(tip.size, size)
 
     def test_author(self):
         tip = self.repo.get_changeset()
         self.assertEqual(tip.author, u'Joe Doe <[e-mail]>')
 
     def test_author_name(self):
         tip = self.repo.get_changeset()
         self.assertEqual(tip.author_name, u'Joe Doe')
 
     def test_author_email(self):
         tip = self.repo.get_changeset()
         self.assertEqual(tip.author_email, u'[e-mail]')
 
     def test_get_changesets_raise_changesetdoesnotexist_for_wrong_start(self):
         with self.assertRaises(ChangesetDoesNotExistError):
             list(self.repo.get_changesets(start='foobar'))
 
     def test_get_changesets_raise_changesetdoesnotexist_for_wrong_end(self):
         with self.assertRaises(ChangesetDoesNotExistError):
             list(self.repo.get_changesets(end='foobar'))
 
     def test_get_changesets_raise_branchdoesnotexist_for_wrong_branch_name(self):
         with self.assertRaises(BranchDoesNotExistError):
             list(self.repo.get_changesets(branch_name='foobar'))
 
     def test_get_changesets_raise_repositoryerror_for_wrong_start_end(self):
         start = self.repo.revisions[-1]
         end = self.repo.revisions[0]
         with self.assertRaises(RepositoryError):
             list(self.repo.get_changesets(start=start, end=end))
 
     def test_get_changesets_numerical_id_reversed(self):
         with self.assertRaises(RepositoryError):
             [x for x in self.repo.get_changesets(start=3, end=2)]
 
     def test_get_changesets_numerical_id_respects_both_start_and_end_last(self):
         with self.assertRaises(RepositoryError):
             last = len(self.repo.revisions)
             list(self.repo.get_changesets(start=last-1, end=last-2))
 
     def test_get_changesets_numerical_id_last_zero_error(self):
         with self.assertRaises(RepositoryError):
             last = len(self.repo.revisions)
             list(self.repo.get_changesets(start=last-1, end=0))
 
 
 class ChangesetsChangesTestCaseMixin(BackendTestMixin):
     recreate_repo_per_test = False
 
     @classmethod
     def _get_commits(cls):
         return [
             {
                 'message': u'Initial',
                 'author': u'Joe Doe <[e-mail]>',
                 'date': datetime.datetime(2010, 1, 1, 20),
                 'added': [
                     FileNode('foo/bar', content='foo'),
                     FileNode('foobar', content='foo'),
                     FileNode('qwe', content='foo'),
                 ],
             },
             {
                 'message': u'Massive changes',
                 'author': u'Joe Doe <[e-mail]>',
                 'date': datetime.datetime(2010, 1, 1, 22),
                 'added': [FileNode('fallout', content='War never changes')],
                 'changed': [
                     FileNode('foo/bar', content='baz'),
                     FileNode('foobar', content='baz'),
                 ],
                 'removed': [FileNode('qwe')],
             },
         ]
 
     def test_initial_commit(self):
         changeset = self.repo.get_changeset(0)
         self.assertItemsEqual(changeset.added, [
             changeset.get_node('foo/bar'),
             changeset.get_node('foobar'),
             changeset.get_node('qwe'),
         ])
         self.assertItemsEqual(changeset.changed, [])
         self.assertItemsEqual(changeset.removed, [])
 
     def test_head_added(self):
         changeset = self.repo.get_changeset()
         self.assertTrue(isinstance(changeset.added, AddedFileNodesGenerator))
         self.assertItemsEqual(changeset.added, [
             changeset.get_node('fallout'),
         ])
         self.assertTrue(isinstance(changeset.changed, ChangedFileNodesGenerator))
         self.assertItemsEqual(changeset.changed, [
             changeset.get_node('foo/bar'),
             changeset.get_node('foobar'),
         ])
         self.assertTrue(isinstance(changeset.removed, RemovedFileNodesGenerator))
         self.assertEqual(len(changeset.removed), 1)
         self.assertEqual(list(changeset.removed)[0].path, 'qwe')
 
 
 # For each backend create test case class
 for alias in SCM_TESTS:
     attrs = {
         'backend_alias': alias,
     }
     # tests with additional commits
     cls_name = ''.join(('%s changesets with commits test' % alias).title().split())
     bases = (ChangesetsWithCommitsTestCaseixin, unittest.TestCase)
     globals()[cls_name] = type(cls_name, bases, attrs)
 
     # tests without additional commits
     cls_name = ''.join(('%s changesets test' % alias).title().split())
     bases = (ChangesetsTestCaseMixin, unittest.TestCase)
     globals()[cls_name] = type(cls_name, bases, attrs)
 
     # tests changes
     cls_name = ''.join(('%s changesets changes test' % alias).title().split())
     bases = (ChangesetsChangesTestCaseMixin, unittest.TestCase)
     globals()[cls_name] = type(cls_name, bases, attrs)
 
 
 if __name__ == '__main__':
     unittest.main()

rhodecode/tests/vcs/test_workdirs.py

@@ -1,90 +1,98 b''
 from __future__ import with_statement
 
 import datetime
 from rhodecode.lib.vcs.nodes import FileNode
 from rhodecode.lib.vcs.utils.compat import unittest
 from rhodecode.tests.vcs.base import BackendTestMixin
 from rhodecode.tests.vcs.conf import SCM_TESTS
 
 
 class WorkdirTestCaseMixin(BackendTestMixin):
 
     @classmethod
     def _get_commits(cls):
         commits = [
             {
                 'message': u'Initial commit',
                 'author': u'Joe Doe <[e-mail]>',
                 'date': datetime.datetime(2010, 1, 1, 20),
                 'added': [
                     FileNode('foobar', content='Foobar'),
                     FileNode('foobar2', content='Foobar II'),
                     FileNode('foo/bar/baz', content='baz here!'),
                 ],
             },
             {
                 'message': u'Changes...',
                 'author': u'Jane Doe <[e-mail]>',
                 'date': datetime.datetime(2010, 1, 1, 21),
                 'added': [
                     FileNode('some/new.txt', content='news...'),
                 ],
                 'changed': [
                     FileNode('foobar', 'Foobar I'),
                 ],
                 'removed': [],
             },
         ]
         return commits
 
     def test_get_branch_for_default_branch(self):
         self.assertEqual(self.repo.workdir.get_branch(),
             self.repo.DEFAULT_BRANCH_NAME)
 
     def test_get_branch_after_adding_one(self):
         self.imc.add(FileNode('docs/index.txt',
             content='Documentation\n'))
         self.imc.commit(
             message=u'New branch: foobar',
             author=u'joe',
             branch='foobar',
         )
+        self.assertEqual(self.repo.workdir.get_branch(), self.default_branch)
 
     def test_get_changeset(self):
+        old_head = self.repo.get_changeset()
         self.imc.add(FileNode('docs/index.txt',
             content='Documentation\n'))
         head = self.imc.commit(
             message=u'New branch: foobar',
             author=u'joe',
             branch='foobar',
         )
+        self.assertEqual(self.repo.workdir.get_branch(), self.default_branch)
+        self.repo.workdir.checkout_branch('foobar')
         self.assertEqual(self.repo.workdir.get_changeset(), head)
 
+        # Make sure that old head is still there after update to defualt branch
+        self.repo.workdir.checkout_branch(self.default_branch)
+        self.assertEqual(self.repo.workdir.get_changeset(), old_head)
+
     def test_checkout_branch(self):
         from rhodecode.lib.vcs.exceptions import BranchDoesNotExistError
         # first, 'foobranch' does not exist.
         self.assertRaises(BranchDoesNotExistError, self.repo.workdir.checkout_branch,
                           branch='foobranch')
         # create new branch 'foobranch'.
         self.imc.add(FileNode('file1', content='blah'))
         self.imc.commit(message=u'asd', author=u'john', branch='foobranch')
         # go back to the default branch
         self.repo.workdir.checkout_branch()
         self.assertEqual(self.repo.workdir.get_branch(), self.backend_class.DEFAULT_BRANCH_NAME)
         # checkout 'foobranch'
         self.repo.workdir.checkout_branch('foobranch')
         self.assertEqual(self.repo.workdir.get_branch(), 'foobranch')
 
 
 # For each backend create test case class
 for alias in SCM_TESTS:
     attrs = {
         'backend_alias': alias,
     }
     cls_name = ''.join(('%s branch test' % alias).title().split())
     bases = (WorkdirTestCaseMixin, unittest.TestCase)
     globals()[cls_name] = type(cls_name, bases, attrs)
 
 
 if __name__ == '__main__':
     unittest.main()